Dumbbell-like Au-Fe(3)O(4) nanoparticles are synthesized using decomposition of Fe(CO)(5) on the surface of the Au nanoparticles followed by oxidation in 1-octadecene solvent. The size of the particles is tuned from 2 to 8 nm for Au and 4 nm to 20 nm for Fe(3)O(4). The particles show the characteristic surface plasmon absorption of Au and the magnetic properties of Fe(3)O(4) that are affected by the interactions between Au and Fe(3)O(4). The dumbbell is formed through epitaxial growth of iron oxide on the Au seeds, and the growth can be affected by the polarity of the solvent, as the use of diphenyl ether results in flower-like Au-Fe(3)O(4) nanoparticles.

Soluble CdSe quantum wires are prepared by the solution-liquid-solid mechanism, using monodisperse bismith nanoparticles to catalyze wire growth. The quantum wires have micrometer lengths, diameters in the range of 5-20 nm, and diameter distributions of +/-10-20%. Spectroscopically determined wire band gaps compare closely to those calculated by the semiemipirical pseudopotential method, confirming 2D quantum confinement. The diameter dependence of the quantum wire band gaps is compared to that of CdSe quantum dots and rods. Quantum rod band gaps are shown to be delimited by the band gaps of dots and wires of like diameter, for short and long rods, respectively. The experimental data suggest that a length of ca. 30 nm is required for the third dimension of quantum confinement to fully vanish in CdSe rods. That length is about six times the bulk CdSe exciton Bohr radius.

The size dependence of the bandgap is the most identifiable aspect of quantum confinement in semiconductors; the bandgap increases as the nanostructure size decreases. The bandgaps in one-dimensional (1D)-confined wells, 2D-confined wires, and 3D-confined dots should evolve differently with size as a result of the differing dimensionality of confinement. However, no systematic experimental comparisons of analogous 1D, 2D or 3D confinement systems have been made. Here we report growth of indium phosphide (InP) quantum wires having diameters in the strong-confinement regime, and a comparison of their bandgaps with those previously reported for InP quantum dots. We provide theoretical evidence to establish that the quantum confinement observed in the InP wires is weakened to the expected extent, relative to that in InP dots, by the loss of one confinement dimension. Quantum wires sometimes behave as strings of quantum dots, and we propose an analysis to generally distinguish quantum-wire from quantum-dot behaviour.

The serendipitously discovered solution-liquid-solid (SLS) mechanism has been refined into a nearly general synthetic method for semiconductor nanowires. Purposeful control of diameters and diameter distributions is achieved. The synthesis proceeds by a solution-based catalyzed-growth mechanism in which nanometer-scale metallic droplets catalyze the decomposition of metallo-organic precursors and crystalline nanowire growth. Related growth methods proceeding by the analogous vapor-liquid-solid (VLS) and supercritical fluid-liquid-solid (SFLS) mechanisms are known, and the relative attributes of the methods are compared. In short, the VLS method is most general and appears to afford nanowires of the best crystalline quality. The SLS method appears to be advantageous for producing the smallest nanowire diameters and for variation and control of surface ligation. The SFLS method may represent an ideal compromise. Recent results for SLS growth are summarized.

No abstract