Alkylation of the compounds (PPN) $\sb2\lbrack$EFe$\rm\sb3(CO)\sb9\rbrack$ (E = S, Se, Te) was performed using methyl triflate and methyl iodide. The S-cluster yielded the novel compound (PPN) (Fe$\sb3$(CO)$\sb9$SMe), whereas the Se and Te-cluster alkylated at the Fe$\sb3$-base yielding (PPN) (MeFe$\sb3$(CO)$\sb9$E). For comparison, the clusters (PPN) $\sb2\lbrack$HE$\rm\{Fe(CO)\sb4\}\sb3\rbrack$ (E = Sb, As) were alkylated as well. Reaction of the Sb-cluster with MeI yielded (PPN) (MeSb(I)$\rm\{Fe(CO)\sb4\}\rbrack,$ whereas the reaction with EtI yielded (PPN) $\sb2\lbrack$ISb$\rm\{Fe(CO)\sb4\}\sb3\rbrack$ and ethane. The possibility of a radical chain reaction for the latter was ruled out by performing the reaction in the presence of a radical scavenger as well as in the dark. The compounds $\rm\lbrack Cat\rbrack\sb{2-x}\lbrack H\sb{x}M\sb3(CO)\sb9E\rbrack\ (cat=Et\sb4N\sp+,\ PPN\sp+;$ x = 0, 1; M = Fe, Ru; E = S, Se, Te) were shown to mediate the catalytic formation of methyl formate from methanol and CO. The reaction is pseudo first order in catalyst and the initial rate is independent of the pressure. NaAsO$\sb2$ reacts with Mo(CO)$\sb6$ in refluxing methanol or ethanol to form $\rm\lbrack Et\sb4N\rbrack\sb2\lbrack(OC)\sb5MoAsMo\sb3(CO)\sb9(\mu\sb3$-$\rm OR)\sb3Mo(CO)\sb3\rbrack$ (R=Me, Et). The compounds are electron rich, and extended Huckel calculations have shown that the extra electron pair resides in an a$\sb2$ orbital, equally delocalized over three molybdenum atoms. A $\sp{205}$Tl NMR study has been conducted on the following compounds with Tl-transition metal bonds: $\rm Tl\{CO(CO)\sb4\}\sb3,\ \lbrack BnMe\sb3N\rbrack\sb3\lbrack Tl\{Fe(CO)\sb4\}\sb3\rbrack,\ Tl\{M(CO)\sb3Cp\}\sb3$ (M = Cr, Mo, W), TlFp$\sb3,$ Fp = CpFe(CO)$\rm\sb2),\ \lbrack PPN\rbrack\sb2\lbrack Tl\sb2Fe\sb6(CO)\sb{24}\rbrack,\ \lbrack Et\sb4N\rbrack\sb2\lbrack Tl\sb2Fe\sb4(CO)\sb{16}\rbrack,\ \lbrack Et\sb4N\rbrack\lbrack LTl\{Fe(CO)\sb4\}\sb2\rbrack$ (L = bipy, en, phen, tmeda, dien), and $\rm\lbrack Et\sb4N\rbrack\sb4\lbrack Tl\sb4Fe\sb8(CO)\sb{30}\rbrack,$ as well as $\rm TlCo(CO)\sb4.$ The possibility of formation of carbonylate anion adducts was also investigated by $\sp{205}$Tl NMR. This technique was used to probe the dynamic behavior of the Tl-metal cluster complexes in solution, and it was shown that most larger Tl-Fe clusters dissociate into simpler fragments in solution.

The salt elimination reaction between Ph3BiBr 2 and a metal alkoxide MIOR (MI = Ag, Na; R = C6F5, C6Cl5, 2,6-C 6H3F2, OTf) yields the compounds Ph3Bi(OR)Br and Ph3Bi(OR)2 depending upon stoichiometry. The alcoholysis reaction between BiPh5 and HOR results in the formation of Ph 4Bi(OR). The structure in the solid state and in solution of those bismuth(V) alkoxide compounds are trigonal bipyramidal with phenyl groups occupying the equatorial positions. For compounds Ph3Bi(OR)Br an equilibrium in solution was observed: Ph3Bi(OR)2 + Ph3BiBr 2 &rlhar2; 2 Ph3BiBr(OR). The equilibrium constants for R = C 6F5, C6Cl5 have been determined. Mixing the bismuth(III) alkoxide [Bi(OC6F5) 3]2 with NaOC6F5 leads to the formation of mixed metal alkoxides, [NaBi(OC6F5)4THF] and Na4Bi2(O)(OC6F5)8. Compound NaBi4(O)2(OC6F5) 9 can be synthesized from BiCl3 and NaOC6F 5 in THF. These alkoxide compounds show interesting dynamics in solution which were investigated with variable temperature fluorine NMR spectroscopy. [NaBi(OC6F5)4THF] exists as a polymeric chain in the solid state. Dissolution of the bismuth(III) alkoxide, (Bi(OC6F5) 3]2 in organic solvents promotes rearrangement reactions of the bismuth alkoxide, involving ether elimination or microhydrolysis. The higher nuclearity bismuth oxo alkoxides Bi4(mu4-O)(mu 2-OC6F5)6{mu3- OBi(OC 6F5)3}2, Bi8(mu4 -O)2(mu3-O)2(OC6F 5)16 and Bi6(mu3-O)4(mu 3-OC6F5){mu3-OBi(OC6F 5)4}3 were isolated and structurally characterized. Common to all three bismuth oxo alkoxides is the presence of mu4-OBi 4 and mu3-OBi3 units. Variable temperature NMR studies suggest that the structure in solution resembles the structure in the solid state.

The first exhaustive purification and characterization of fullerenol samples, prepared by reaction of C60 in toluene with NaOH solution using tetrabutylammonium hydroxide as a phase transfer catalyst, have been performed. The resulting fullerenol was determined to have a molecular formula of Na+n[C60Ox(OH) y]n- (where n = 2--3, x = 7--9, and y = 12--15 for three different, but identical, preparations). The Na +-fullerenol was found to contain both hemiketal moieties (6--8 per fullerenol) and a carbonyl moiety (1 per fullerenol) by an X-ray photoelectron spectroscopy study. In addition, Na+-fullerenol was found to be anionic with an average negative charge of 2--3 and a molar conductance in water in the range of 174--297 cm-1mol -1O-1 at 25°C. Surprisingly, the Na+-fullerenol is also paramagnetic exhibiting mueffective values in aqueous solution of 1.9--2.1 B.M. at 0.5 T and 300 K and R1 proton relaxivities of 0.55--0.77 mM-1s-1 at 20 MHz and 40°C, values both slightly higher than those expected for a pure S = ½ spin system. In solid state SQUID measurements, the Na+-fullerenols had mueffective values of ca. 1.3 B.M. (300K). ESR studies of frozen aqueous solutions at 1.5 and 5.0 K established that Na+-fullerenol is mainly S = ½ with a minor, but significant, concentration of S = 1. Thus, this is the first report to characterize the widely-studied, water-soluble fullerenols as radical anion species. The Na+-fullerenol paramagnetism was partially quenched by the addition of acid, but was reversed upon addition of base. This acid/base chemistry can be used to partition a Na+-fullerenol sample into less paramagnetic and more paramagnetic fullerenol species, with the more paramagnetic species having more S = 1 character. It is hypothesized that the paramagnetic character of Na+-fullerenol results from the formation of cyclopentadienyl radical centers on the fullerene cage, formed by an oxidation reaction with atmospheric oxygen. Sodium ions in Na+-fullerenol can be ion exchanged for other cations (e.g. Ba2+, H+, tetrabutylammonium cation, and protonated meglumine) producing fullerenols with varying magnetic and spectroscopic properties. Finally, paramagnetic Na+-fullerenol (110 mg/kg) has been shown to serve in vivo as a magnetic resonance imaging (MRI) contrast agent in a rat model, although, the rat did not survive the experiment.

Carbon nanomaterials such as fullerenes and single-walled carbon nanotubes (SWNTs) have opened up an exciting new field of research in chemistry. Fullerene research thus far has focused on modification to the outside of the carbon cage to change specific chemical properties, such as solubility or reactivity, to suit a particular target application. Although fullerenes have been shown to undergo facile reduction and to readily react with nucleophiles, oxidation has presented a problem. We have employed a transition metal catalyst to epoxidize fullerenes and create open-caged fullerenes. We have investigated the reactivity of hydroxyfullerene with a variety of metal salts under ambient aqueous conditions as well as replicating possible environmental exposure. We have proven fullerenol reacts irreversibly across a wide pH range with a variety of metal salts to form insoluble cross-linked polymers. It is clear that the interaction of hydroxyfullerenes with metals is an important issue with regard to waste treatment, fullerene exposure in the environment, and fullerene-based pharmaceutical agents. There is much interest in the development of methods that allow for the synthesis of SWNTs with particular helicities, which determines the electronic properties of SWNTs. It would be desirable to use pre-formed SWNTs as seeds for the growth of longer SWNTs. In order to accomplish this goal, it is necessary that a catalyst particle be preferentially docked to the end of an individual SWNT. Purification and characterization of a suitable iron-molybdenum nanocluster and the methodology for coupling catalyst precursors to a pre-formed SWNT are discussed. We have discovered a new method of solubilizing SWNTs in organic solvents. This was accomplished by creating a SWNT salt with a crown ether encapsulated sodium counter ion. Lastly, the effect of carbon nanomaterials on mineral formation was investigated. Water-soluble fullerene and SWNT derivatives produced new and interesting morphologies of calcium carbonate and barium carbonate.

The formation of carboxylate substituted alumina nanoparticles, [Al(O) x(OH)y(O2CR)z]n, is an environmentally benign and inexpensive process. The alumoxane nanoparticles are synthesized from the mineral boehmite with carboxylic acids. The physical characteristics of carboxylate-alumoxanes, such as ceramic yield, particle size, pore volume, and pore diameter are dependent upon the carboxylate ligand that is substituted on the surface of the boehmite mineral. Of the synthesized alumoxanes, acetic acid-alumoxane (A-alumoxane), has the smallest particle size (11 nm), the highest ceramic yield (76%), and produces the smallest pore diameter (5 nm). These physical characteristics led to the investigation of the A-alumoxane ceramic precursor as a strengthening agent for porous alumina ceramics, and for the fabrication of ultra and nanofiltration ceramic membranes. Infiltration is one mechanism used to deposit strengthening agents, such as ceramic precursors, into porous alumina ceramics. Due to the particle size associated with the alumoxanes, they can be easily infiltrated into 94 nm pore sized alumina ceramics, and upon thermolysis forming a homophase, Al2O3, ceramic composite. When the alumoxanes are doped with metals forming MxAlyOz upon thermolysis, a heterophase ceramic composite forms. The additional ceramic material (either a homo or hetero-layer) into the interior of the porous ceramic decreases the material's porosity; a decrease in porosity presumably increases the ceramic's strength. The alumina derived from the thermolysis of A-alumoxane produces pore diameters between 5--17 nm, thus making the ceramic desirable for the use in ultrafiltration membranes. The small pore sizes associated with ultrafiltration membranes, hinders the flux through the resulting membrane. To improve the flux, the membrane layer should be as thin as possible without sacrificing the membrane's performance and rejection characteristics. Once the thinnest possible membrane is achieved, the flux can be further improved by altering the surface chemistry and/or the surface features of the membrane. Adjusting the surface pH by metal doping, or adding functional groups, and forming a hierarchical membrane were investigated to improve the membrane's performance.

The activity of a new catalyst precursor for the vapor-liquid-solid (VLS) growth of single walled carbon nanotubes (SWNTs), oxo-hexacarboxylate-tri-iron compounds, [Fe3(mu3-O)(mu-O2CR) 6(L)3], was investigated. Additionally, the pure form of the iron-molybdenum cluster, [HxPMo12O40⊂H 4Mo72Fe30(O2CMe)15O 254(H2O)98] (x = 1 or 2) (FeMoC), was confirmed to be a catalyst precursor for SWNT growth. In order to control the diameter and chirality of SWNTs, it has been envisioned that a preformed SWNT could be used as a template to seed the growth of that individual SWNT. Therefore, it is necessary to attach a catalyst precursor particle to the end of an individual SWNT. The functionalization of FeMoC with various coordinating ligands led to the use of an organic coupling reaction to attach pyridine groups to the end of SWNTs. The attachment of FeMoC to the end of SWNTs has resulted in the growth of SWNTs from these tube/catalyst seeds. The formation of highly epoxidized C60 species has been observed using a methyltrioxorhenium (MTO)/H2O2/4-bromopyrazole system. MTO, in combination with triphenylphosphine (PPh3), has been demonstrated to be a catalyst for the deoxygenation of fullerenes and SWNTs. The epoxide content of SWNT samples was quantitatively determined through 31P{1H} NMR. Results indicate that SWNTs contain a significant amount of epoxides on their sidewalls. Furthermore, the cutting of SWNTs is observed after the deoxygenation of ozonated SWNTs that had been irradiated by an electron beam. A variety of aluminum aryloxide and alkyl derivatives of pyrazine and bipyridines have synthesized and characterized. Initial studies suggested that the compounds [(tBu)2Al(OPh)]2(mu-L) would offer an ideal system for a quantitative study of the effect of binding two Lewis acid centers to different Lewis base sites on a single molecule (i.e., X3M-L-MX3). However, anomalous behavior was observed with regard to the relative position of the various equilibria as well as a distinct solvent effect on the equilibria. Finally, as part of a study on the reactivity of di-alkylaluminum phenoxides bidentate bridging Lewis bases, [Me2Al(mu-OPh)]2 was synthesized and structurally characterized. The discovery of the phenyl ring being coplanar with the Al2O2 core prompted a study of inter-molecular distances and an ab initio study to confirm the relative stability of the ring configuration.

The catalyzed etching of diamond by CO$\sb2$ in the presence of first-row transition metal oxides was studied and it was observed to occur in the presence of NiO, $\rm Fe\sb2O\sb3,\ Co\sb3O\sb4$ with a strong CO$\sb2$ dependence. The apparent activation energies were found to be similar for NiO, $\rm Fe\sb2O\sb3,\ Co\sb3O\sb4$ in the presence of CO$\sb2$ while they are different under an argon atmosphere. A three-step qualitative model was proposed involving the adsorption of CO$\sb2$ onto the metal oxide surface, the transfer of oxygen to the diamond surface, and the then desorption of CO from the diamond surface. The analysis of the model and the experimental facts indicates that the transfer of oxygen is most likely the rate determining step for the catalytic etching of diamond in the presence of NiO, $\rm Fe\sb2O\sb3,\ Co\sb3O\sb4.$ Several of the metal oxides were also observed to be able to etch diamond directly for which a similar model was proposed. The etching of diamond by CO$\sb2$ and H$\sb2$O in the presence of $\rm Cr\sb2O\sb3$ was studied. HCl, H$\sb2$O, H$\sb2,$ and $\rm CH\sb3Cl$ were found to be able to enhance the etching rate of diamond by CO$\sb2$ to different extents with only small amounts added. With $\rm Cr\sb2O\sb3$ present, water was observed to be able to etch diamond at a much lower temperature than water alone. A large difference was found between the apparent activation energies when HCl and H$\sb2$O were introduced into the reaction system respectively, the latter being similar to the activation energy for water etching alone. The formation of surface hydroxyl is believed to be essential for the reaction to happen.

Planar single-crystalline colloidal crystals are fabricated by exploiting the spontaneous crystallization of monodisperse silica spheres into close-packed arrays using a convective self-assembly method. Film thicknesses ranging from single monolayers to over 200 layers can be precisely controlled through varying solution concentration and colloid diameter. These high quality periodic arrays exhibit partial photonic band gaps, whose positions and band width depends on the sphere diameters and the number of layers. Their optical transmission is well described by the scalar wave approximation to Maxwell's equations. The thickness dependence of the photonic band gap has also been studied for the first time. Superlattice colloidal crystals comprised of alternating layers of different sphere sizes can also be formed by the convective self-assembly method. The resulting photonic crystal structures exhibit optical properties which resemble the superposition of the properties of each individual component, with additional structure that suggests the onset of superlattice-type miniband formation. These superlattice structures thus provide a new way to couple light into and out of photonic crystals. These planar colloidal films are then used as scaffolds to make macroporous materials with crystalline arrays of voids. Macroporous polymers are formed by filling the interstitial area with monomer which is subsequently polymerized. The silica templates can be removed by etching with hydrofluoric acid. The large voids defined by the silica colloids are not isolated, but rather interconnected by a network of monodisperse smaller pores whose sizes can be controlled by varying the polymerization temperature. These membranes exhibit striking optical properties and their photonic band gap behavior agrees well with theory. A seeded electroless deposition technique has also been developed for forming macroporous metal membranes. The gold particles attached to the thiol-coated silica colloidal crystals can catalyze the electroless deposition of metals (Ag, Au, Cu, Ni, Co, Pt) inside the arrays and lead to fully dense macroporous metallic films after silica removal. These samples are mechanically robust, electrically active, and possess unusual diffractive optical properties. The macroporous polymers are again used as hosts to grow a wide variety of complex and unusual colloidal structures. This modern "lost-wax" method effectively capitalizes on the perfection of the starting colloids and the resulting template voids to form monodisperse colloids and their colloidal crystals. A wide variety of highly monodisperse inorganic, polymeric and metallic solid and core-shell colloids, as well as hollow colloids with controllable shell thickness and their colloidal crystals can be made. The polymer template can be uniformly deformed to alter colloidal shape and elliptical particles with precisely controlled aspect ratios are formed for the first time. The hollow sphere titania colloidal crystals exhibit partial photonic band gaps, whose spectral position and width depend on the thickness of the shell and on the overlap between adjacent spheres, in a manner consistent with numerical simulations.

Reaction of 9-fluorenone with a Group 13 Lewis acid, MX3 (M = B, Al, Ga; X = Cl, Br), yields a Lewis acid-base adduct, MX3(9-fluorenone) n. Bond dissociation energies for MX3(9-fluorenone) in solution have been determined. BDE and other thermodynamic values are good indicators of the strength of a Lewis acid-base interaction. Correlating these values with various spectroscopic data should lead to a simple test of Lewis acidity. Hydrogen/deuterium exchange between naphthalene and C6D 6 is catalyzed by Hg(C6D6)2(GaCl 4)2 and occurs by electrophilic substitution. In this regard, we have studied the kinetics of H/D exchange, using 1H NMR. There are two distinct first order regimes, separated by a non-first order regime. In the first regime, Hg(C6D6)2(GaCl 4)2 deuterates free naphthalene. As the reaction progresses, a ligand exchange occurs between C6D6 and naphthalene, resulting in the protonation of C6D6 by Hg(C10 H8)2(GaCl4)2. This ligand exchange has been confirmed by 13C CPMAS NMR of Hg(C10 H8)2(GaCl4)2, as well as by DFT calculations, UV-visible spectroscopy, and MS. The rates of not only H/D exchange, but also the rate of ligand exchange has been determined. Reaction of M(tBu)3 with anthranilic, salicylic, and ortho-toluic acids yields [(tBu) 2M(mu-O2CC6H4-2-NH2)] 2, M = Al, Ga; [(tBu)2Ga(mu-O2CC 6H4-2-OH)]2, and [(tBu)2Ga(mu-O 2CC6H4-2-Me)]2, respectively. Reaction of anthranilic acid with two equivalents of Al(tBu)3 allows for the isolation of (tBu)2Al(mu-O 2CC6H4-2-NH2)Al(tBu) 3. Based on this compound's isolation and its conversion to [( tBu)2Al(mu-O2CC6H4-2NH 2)]2 during physical grinding, a proposal that the formation of [(tBu)2M(mu-O2CC6H 4-2-X)]2 occurs via a chelate species can be made. Additionally, intra-molecular hydrogen bonding, found in the anthranilate and salicylate ligands, remains present in the Group 13 compounds but is not enhanced.

Nanoscience is the area of science concerning materials on the level of nanometer scale. Currently much of the discussion of nanoscience is focused on carbon-based nanostructures. Tremendous studies have been carried out on fullerenes and carbon nanotubes in the past a couple of decade due to their unique chemical and physical properties. This thesis studies the chemistry of the other two novel nanostructures in the carbon family: nanodiamond and carbon nano-onions. Nanodiamond is relatively a new engineering material with particular applications for fabrication of wear-resistant surface coatings, lubricating films and prototypes field emission displays. These materials are also of interest for studies of chemical reactivity stemming from their nanoscale particle sizes. The surface fluorination of nanodiamond at various temperatures yields a fluoro-nanodiamond with up to 8.6 at. % fluorine content. The fluoro-nanodiamond was used as a precursor for preparation of the series of functionalized nanodiamonds by subsequent reactions with alkyllithium reagents, diamines, and amino acids. The fluoro-nanodiamond and all derivatives were characterized by SEM, TEM, XRD, TG-MS, FTIR, XPS, and Raman measurements. In comparison with the pristine nanodiamond, all functionalized nanodiamonds show an improved solubility in polar organic solvents, e.g., alcohols and THF, and a reduced particle agglomeration. The fluoro-nanodiamond powder was also used as a precursor for diamond coating on solid substrate surface e.g. glass. This novel approach is based on the wet chemistry process (solution reaction) occurring at low temperature and resulting in a covalent bonding of tiny nanodiamond crystals to substrate through a molecular linker, 3-aminopropyltriethoxysilane (APTES). SEM, AFM and XPS were used for evaluation of surface morphology and elemental analysis to confirm the presence of diamond particles on surface. Carbon nano-onion studied in this thesis is another new material synthesized by our collaborators. The layer-by-layer structures make it a potential candidate as lubricant materials. The nano-onions were fluorinated at various temperatures resulting in fluoro-onions with different fluorine content. The inside layered structures are damaged due to fluorination, which was characterized by Raman, XRD, TEM et al. Defluorination treatment was also performed. All the samples show improved lubricating properties according to the test carried out by our collaborators.