Colloidal early transition metals are of great interest in both powder technology and catalysis. We report a new process for preparing organosols of zerovalent Ti, Zr, V, Nb and Mn

Metal particles are generated via the spontaneous reduction of Ag+ and AuCl4− ions by solvent molecules that takes place in air-saturated alcoholic solutions containing hydroxide ions. Changes in the plasmon band of the Ag particles are observed when the metal particles are in contact with Ag20 particles. The optical changes are explained in terms of surface effects of the metal particles. Larger shifts of the Au plasmon band and light scattering were observed at the initial stages of the particle formation process. These effects are explained in terms of formation of networks consisting of small metal particles. It is proposed that generation of small Au particles occurs mainly on silica surfaces, and that particle-networks are formed when small particles desorb from the surfaces.

A self consistent jellium approach to the chemisorption of molecular oxygen on copper clusters is investigated and compared with local density MO - LCAO calculations. The jellium model is found to be well suited for chemisorption studies and the results explain the main trends in the measured chemisorption properties of O2 on copper clusters.

Neutral, low-valent complexes of transition metals or an organometallic compound of Ge are prepared containing ancillary ligands which bear silicate ester or (alkoxy)silicon functional groups. Inclusion of these molecules as dopant species in a conventional sol-gel synthesis of silica xerogels affords silica xerogels in which dopant molecules have presumably been covalently incorporated into the xerogel matrix with uniform and high dispersion. Subsequent thermal treatment of these molecularly doped xerogels under reducing or oxidizing/reducing conditions gives nanocomposites containing nanoclusters of metals or semiconductor substances. By this procedure, nanocomposites containing nanoclusters of Ag, Cu, Pt, Os, CCo3, Fe2P, Ni2P, or Ge have been prepared. Preliminary evidence for the formation of a nanocomposite containing Pt-Sn nanoclusters derived from a bimetallic molecular precursor is also presented. Characterization data for the nanocomposite materials include TEM, electron diffraction, EDS, XRD, and selected use of micro-Raman spectroscopy. These results support the hypothesis that covalent incorporation of molecular precursors containing low-valent metals into a silica xerogel can afford nanocluster phases with high dispersion, relatively small particle size, and unusual elemental composition.

The design of molecular precursors to materials for catalysis will be described. Aluminum fluoride is an important catalytic material for fluorocarbon transformations which are key to the production of CFC alternatives. Catalytic activity is closely related to the crystalline phase of the bulk AIF3 and we show how this phase chemistry can be precisely controlled using molecular precursors to produce pure known phases and also to extend to previously unknown phases. A similar approach to vanadyl phosphate oxidation catalysts will be described where cluster chemistry can be invoked to generate isolated fragments of a catalyst's structure and then used either to explore mechanism or be used as precursors to functional catalyst materials. The underlying themes of controlled molecular precursor synthesis, resultant processability and eventual easy conversion to useful materials are emphasized throughout.

Diamond-like nanocomposite (DLN) and metal containing DLN (Me-DLN) films, synthesized in a combined process of deposition of carbon-silicon precursor and magnetron sputtering of a metal target, have been examined by Auger electron spectroscopy, Raman and IR spectroscopy, nanoindentation and internal stress measurements. The stability of the films under ion and electron irradiation and thermal annealing has been tested.

We have prepared GaP and GaAs nanoclusters from organometallic condensation reactions of E[Si(CH3)3]3 (E = P, As) and GaCl3. The size of the as synthesized clusters is 10 Å to 15 Å. Larger clusters of 20 Å to 30 Å size were obtained by thermal annealing of the as grown material. X-ray diffraction and transmission electron microscopy confirm the high crystalline quality. A lattice contraction of 6.7% could be seen for 10 Å sized GaAs clusters. The clusters are nearly spherical in shape. Optical absorption spectra show a distinct line which can be assigned to the fundamental transition of the quantum confined electronic state. The measured blue shift, with respect to the GaP bulk absorption edge is 0.53 eV. As the cluster is smaller than the exciton radius, we can calculate the cluster size from this blue shift and obtain 20.2 Å, consistent with the results from X-ray diffraction of 19.5 Å for the same sample.

The control of structure on the nanoscale relies on intermolecular interactions whose specificity and geometry can be treated on a predictive basis. DNA fulfills this criterion, and provides an extremely convenient construction medium: The sticky-ended association of DNA molecules occurs with high specificity, and it results in the formation of double helical DNA, whose structure is well known. The use of stable branched DNA molecules permits one to make stick-figures. We have used this strategy to construct in solution a covalently closed DNA molecule whose helix axes have the connectivity of a cube: The molecule has twelve double helical edges; every edge is two helical turns in length, resulting in a hexacatenane, each of whose strands corresponds to a face of the object. We have developed a solid-support-based synthetic methodology that is more effective than solution synthesis. The key features of the technique are control over the formation of each edge of the object, and the topological closure of each intermediate. The isolation of individual objects on the surface of the support eliminates cross-reactions between growing products. The solid-support-based methodology has been used to construct a molecule whose helix axes have the connectivity of a truncated octahedron. This figure has 14 faces, of which six are square and eight are hexagonal; this Archimedean polyhedron contains 24 vertices and 36 edges, and is built from a 14-catenane of DNA. Knotted molecules appear to be the route for cloning DNA objects. It is possible to construct three knotted topologies, as well as a simple cyclic molecule from a single precursor, by control of solution conditions. Control of both branching and braiding topology is strong in this system, but control of 3-D structure remains elusive. Our key aim is the formation of prespecified 2-D and 3-D periodic structures for use in diffraction experiments. Another application envisioned is scaffolding for the assembly of molecular electronic devices.

Molecular Self-assembly of amphiphilic phospholipid molecules (containing a hydrophobic acyl chain and a hydrophilic phosphate group attached to glycerol backbone) and other amphiphiles offers a versatile approach to form ordered structures. Stabilization of lipid microstructures by polymerization renders them useful for practical applications in the areas ranging from controlled release technology to template mediated synthesis of metals. Our efforts are focussed on the development and use of polymerizable diacetylenic phospholipids and their microstructures as template for chemical synthesis. The surface of vesicles and lipid microcylinders (0.5 μm dia.) is made reactive by chemically modifying the hydrophilic region of phospholipids. Lipids with chemically reactive sites were incorporated into lipid membranes predominantly formed from charge neutral lipids and used for binding metal ions and growing fine metal particles.

In order to investigate the influence of substrate functionalization on the subsequent selfassembly of multilayer films, multilayers composed of alternating hafnium and 1,10-decanediylbis(phosphonic) acid (DBPA) have been grown on three different substrates. Substrates studied include gold wafers functionalized with 4-mercaptobutylphosphonic acid, silicon wafers functionalized using a hafnium oxychloride solution, and silicon wafers coated with an octadecylphosphonate LB-template layer. The nature of these films is probed using ellipsometry and grazing angle x-ray diffraction. These studies indicate that the overall order and the individual layer thickness can vary substantially from sample to sample and depend strongly on the initial surface functionalization prior to multilayer growth.

Single-phase nanocomposites containing montmorillonite, MoS2, MoO3 or TiS2 with poly(ethylene oxide) are obtained by the exfoliation of the layered solid, adsorption of polymer, and subsequent precipitation of solid product. Aqueous solutions can be employed for all syntheses except PEO/TiS2, which is prepared from lithiated TiS2 in an N-methyl formamide (NMF) solution. X-ray diffraction indicates that the resulting solids increase in basal-plane repeat by approximately 4 or 8 Å, consistent with the incorporation of a single or double layer of polymer between the inorganic layers. Reaction stoichiometries and elemental analyses provide compositions for the single-phase products.

Nano-composites have been prepared from Na+ and Ca++ montmorillonite (MMT) in a polystyrene matrix via chemical intercalation. Vinyl monomer-g-MMT was prepared by exchanging the mineral cation by vinylbenzyl trimethylammonium chloride, thus rendering the mineral organophilic and forming polymerizable moieties directly bonded to the lamellar surface of the mineral. Styrene was added and polymerized by free radical in selected solvents. The ratio of mineral to the bound polymer ranged from 0.3 to 1.25 (by weight) depending on the initial mineral concentration in the feed and on the solvent used. The mineral domains in the composite, measured from suspension cast film fall in the range of 150 nm to 400 nm. Measured from compression molded samples, the domains were ca. 50 nm which is much smaller than the mineral aggregate and comparable to that of the primary particle of the mineral. WAXD disclosed that the d (001) spacing of MMT in the composite ranged from 1.7 to 2.5 nm suggesting that the mineral aggregates (ca. 10 μm) were dissociated into individual layers then reassembled into lamellar nanoclusters.

The Langmuir Blodgett (LB) Process has been shown to be an appropriate method for use in mimicking of biological processes for producing engineering materials such as bioceramics. The main advantages of this approach are that the layers form at low temperatures, that they are fully dense and that the process of densification is by infiltration rather than by sintering. Moreover, biological hard tissues are self-assembled to perform certain functions; the architecture being controlled by an epitaxial organic matrix. Clearly, if this process can be understood in detail then it is possible that LB films may be used to replicate this architecture for engineering purposes.

Atomic Force Microscopy (AFM) and X-ray diffraction (XRD) have been used to study and characterise LB films of calcium stearate obtained by the repeated dipping into and withdrawal of a (001) Si wafer from a subphase containing calcium ions and using stearic acid as the surfactant. Contact-mode AFM images of the film surface have been used to measure the thickness of the LB layers and to reveal the nature and distribution of defects in the film. The measured thickness of the calcium stearate layers is about 2.5 nm; a value consistent with that obtained by XRD, but smaller than the length of an individual calcium stearate molecules.

The properties of bone, as a polymer reinforced with nanometer-sized ribbon-shaped crystals of mineral, are compared with the properties of synthetic polymer composites. Bone does show some superiority to existing composites. The improvements can be attributed to the microstructure. Methods for reproducing this structure in a synthetic material are discussed.

Intrigued by the deceptive simplicity and beauty of macromolecular self-assembly, our laboratory began studying models of self-assembly using solids, glasses, and colloidal substrates. These studies have defined a fundamental new colloidal material for supporting members of a biochemically reactive pair.

The technology, a molecular transportation assembly, is based on preformed carbon ceramic nanoparticles and self assembled calcium-phosphate dihydrate particles to which glassy carbohydrates are then applied as a nanometer thick surface coating. This carbohydrate coated core functions as a dehydroprotectant and stabilizes surface immobilized members of a biochemically reactive pair. The final product, therefore, consists of three layers. The core is comprised of the ceramic, the second layer is the dehydroprotectant carbohydrate adhesive, and the surface layer is the biochemically reactive molecule for which delivery is desired.

We have characterized many of the physical properties of this system and have evaluated the utility of this delivery technology in vitro and in animal models. Physical characterization has included standard and high resolution transmission electron microscopy, electron and x-ray diffraction and ζ potential analysis. Functional assays of the ability of the system to act as a nanoscale dehydroprotecting delivery vehicle have been performed on viral antigens, hemoglobin, and insulin. By all measures at present, the favorable physical properties and biological behavior of the molecular transportation assembly point to an exciting new interdisciplinary area of technology development in materials science, chemistry and biology.

An attractive and challenging approach to the construction of robust, thin film materials with large second-order optical nonlinearities is the covalent self-assembly of aligned arrays of high-β molecular chromophores into multilayer superlattices. In this paper, we describe the dispersion of second harmonic generation (SHG) in a self-assembled (SA) monolayer containing a stilbazolium chromophore. The frequency-dependent measurements were performed on 25 Å thick monolayers on glass using a tunable (0.4–2 μm) light source based on optical parametric amplification (OPA). The SHG spectrum contains a clear two-photon resonance at hω = 1.3eV. The maximum in the second-order susceptibility coincides with a low energy chromophore-centered charge-transfer excitation at 480 nm. The experimental SHG dispersion values compare favorably with theoretical results computed using a sum-over-states (SOS) formalism. However, the measured values exhibit a somewhat broader band response than the theoretical curve, and the origin of this behavior is discussed.

We studied the optical properties of silicon nanocrystals incorporated into colloidal and solgel matrices. The silicon nanocrystals are produced by ultrasonic dispersion of porous silicon layers. We report results on the dependence of the photoluminescence (PL) spectra with excitation intensity. The PL shows a blue peak (at ∼ 415-460 nm.) and a red peak (at ∼ 680 nm). This PL spectrum shows a remarkable dependence on the excitation intensity. As the intensity is increased, the blue peak grows at the expense of the red. A model is suggested for this behavior. We also report on the excitation intensity dependence and the emission wavelength dependence of the PL decay at low (1 kHz) and high (82 MHz) repetition rates of optical excitation. When low repetition rate excitation is used, the PL decay times are all exponential, short (ns), and appear to vary little with emission wavelength. This sharply contrasts with what is observed in porous silicon. With high repetition rate excitation, both red and blue peaks show long (100's ns) and short (ps-ns) lifetime components. We contrast the different optical properties of these silicon nanocrystals with that observed in porous silicon.

This paper describes the processing of rare-earth/organic dye composites fabricated via three different sol-gel routes. In the first approach, low hydroxyl ormosil matrices were fabricated via reaction of a methyl-modified silicon halide with tertiary alcohol and subsequently doped with erbium iodide and a near-infrared dye. In the second approach, gels were made from tetramethoxysilane and doped with erbium complexes and dyes. In the third approach, a hybrid siloxane method was used. No Er3+ luminescence at 1.55 mm was observed in any of the three cases, mainly due to the strong absorption of the matrices centered around 1.4 mm. Fluorescence of Er3+ in the visible was observed in the first matrice, but no dye luminescence was detected. Dye luminescence was observed in the second type matrix, along with some reabsorption of the dye luminescence by Er3+. In the third approach, neodymium exhibited optical activity in the near infrared, as well as the dye.

The previously developed method to prepare highly dispersed metals in SiO2 by sol-gel processing of metal complexes containing alkoxysilyl-substituted ligands was extended to the preparation (i) of bimetallic particles in SiO2 and (ii) of highly dispersed metals in TiO2.

Sponge type deposits composed of nanometer sized Si whiskers were formed on Davisil (porous SiO2) gel substrates by chemical vapor deposition (CVD) using methyltrichlorosilane (MTS) as a precursor and H2 as carrier gas. The diameters of the fine whiskers are estimated to be 100 nm or smaller from SEM observations. XRD analyses revealed that the coatings formed at 800 and 900°C contained microcrystalline Si and the Auger analysis indicated the existence of free carbon. The surface area of the coated Davisil was approximately 189 ± 5 m2/g, as obtained from BET measurements. The coatings were also applied on several other substrates, such as AIN and low surface area (< 224 m2/g) silica gel, with the same coating conditions. On these substrates, highly porous Si has also been observed on AIN substrates, and nonporous, half sphere shaped coatings were observed on low surface area silica gel substrates.