Search

Silicon clusters in the size range from 5 to 66 atoms were generated by laser vaporization in a supersonic nozzle and injected into the ion trap of a specially‐designed Fourier transform ion cyclotron resonance apparatus. On the positively charged clusters ammonia chemisorption reaction rates were found to vary by over three orders of magnitude as a function of cluster size, with clusters of 21, 25, 33, 39, and 45 atoms being particularly unreactive, and cluster 43 being the most reactive. For the negative cluster ions, 43 was the only cluster found to react substantially. Although the reaction behavior of many clusters clearly indicated that several structural isomers were present with different reaction rates, the strikingly low net reactivity of such clusters as 39 and 45 provides evidence that they have effectively crystallized into a single specially stable form.

The recent report of electrical conductivity in the alkali metal fullerides and the discovery of superconductivity at 18 K for K x C 60 has raised fundamental questions about the electronic states on either side of the Fermi level, their occupancy with K intercalation, and the mechanism of superconductivity. Direct photoemission evidence is presented of filling of bands derived from the lowest unoccupied molecular orbital as a function of K incorporation for the metallic and insulating phases. This filling is not rigid band-like, and it reflects disorder in the K sites. Theoretical analysis indicates that K x C 60 is a strong coupling superconductor, and we suggest that the enhanced electron-phonon interaction is related to the unique hybridization of the C sp -derived states.

Two potassium fulleride phases, metallic K3C60 and nonmetallic K6C60, are formed when potassium is incorporated into thin C60 films under ultrahigh vacuum conditions. Phase separation is observed for intermediate stoichiometries. Results obtained for the C60-K3C60 heterostructure demonstrate that it is stable against potassium migration from the K3C60 phase. In contrast, the C60-K6C60 interface is not stable and K3C60 is formed.

The electronic spectra of supersonically cooled C60 and C70 have been observed in a molecular beam by two‐photon ionization spectroscopy in the regions 375 to 415 nm and 595 to 640 nm. Clear spectral features are observed in both regions for C60 and in the long wavelength region for C70. Neither molecule is responsible for the diffuse interstellar bands.

The electronic structure of small Ga_xAs_y clusters (x+y≤10) are calculated using the local density method. The calculation shows that even‐numbered clusters tend to be singlets, as opposed to odd‐numbered clusters which are open shell systems. This is in agreement with the experimental observations of even/odd alternations of the electron affinity and ionization potential. In the larger clusters, the atoms prefer an alternating bond arrangement; charge transfers are observed from Ga sites to As sites. This observation is also in agreement with recent chemisorption studies of ammonia on GaAs clusters. The close agreement between theoretical calculations and experimental results, together with the rich variation of electronic properties of GaAs clusters with composition makes GaAs clusters an ideal prototype system for the study of how electronic structure influences chemical reactivity.

Consideration of the factors involved in the production of fullerene nanotubes in carbon arcs leads to the notion that a high electric field may be the critical factor that causes the tubes to grow. This thought then leads to a suggestion that it may be possible to grow continuous fullerene fibers many centimeters in length by an electric-field-and-laser induced pyrolysis of gas phase hydrocarbons of fullerenes on the tip of the growing fiber as it extends out from its place of attachment on a high voltage needle. Use of metal- or boron-doped fullerenes (dopyballs) in such an apparatus may lead to the production of doped fullerene fiber nanowires of high strength and conductivity.

Recent experiments have demonstrated that C28 is the smallest fullerene cage that successfully traps elements in its inside. In this work, we have studied the electronic structures, equilibrium geometries, and binding energies of the title molecules at the self‐consistent field (SCF) Hartree–Fock level of theory employing basis sets of double‐zeta quality. The empty C28 fullerene is found to have a 5A2 open‐shell ground state and behaves as a sort of hollow superatom with an effective valence of 4, both toward the outside and inside of the carbon cage. The theoretical evidence suggests that C28H4 and C28F4 should be stable molecules. The possibility of simultaneous bonding from the inside and outside of the C28 shell, as in (Ti@C28)H4, is also explored. Our calculations show that the binding energy of the M@C28 species is a good indicator of the success in experimentally trapping the metal atoms (M) inside the fullerene cage. Based on these results, we propose that elements with electronegativities smaller than 1.54 should form endohedral fullerenes larger than a minimum size which depends on the ionic radius of the trapped atom. This qualitative model, correctly reproduces the available experimental evidence on endohedral fullerenes.

The electronic structure of recently discovered hollow‐cage molecule Ti8C12 has been calculated and compared to similarly shaped hypothetical molecules M8X12 (M=metal and X=B, C, or N). The cohesion of the cage is related to the σ‐bonding between M–X arising from d–sp hybridization. It is also shown that the cage can become electronically closed shell with all the valence electrons paired when the carbon is replaced by boron or nitrogen, e.g., in the formula of Sc8B12.

Carbon nanotubes produced in arcs have been found to have the form of multiwalled fullerenes, at least over short lengths. Sintering of the tubes to each other is the predominant source of defects that limit the utility of these otherwise perfect fullerene structures. The use of a water-cooled copper cathode minimized such defects, permitting nanotubes longer than 40 micrometers to be attached to macroscopic electrodes and extracted from the bulk deposit. A detailed mechanism that features the high electric field at (and field-emission from) open nanotube tips exposed to the arc plasma, and consequent positive feedback effects from the neutral gas and plasma, is proposed for tube growth in such arcs.

Electrochemistry has proven to be very useful for the study of guest-host systems, particularly, carbon intercalation compounds. Not only does electrochemistry provide essential information about the thermodynamics and kinetics of these systems, but it also offers accurate control of guest stoichiometry which is difficult to achieve by other doping methods. Therefore, electrochemical doping has been used extensively to study the properties of carbon guest-host systems. In situ X-ray diffraction and electrochemical doping were used to study the phase diagram of Li xC6 graphite, 1 phase transitions in Li-doped polyacetylene 2 and the structure of Li-doped solid C 60. 3 In situ resistivity measurements were used to study the electronic transport properties of K- and Na-doped polyacetylene. 4,5 In this work, electrochemistry was used to study a new carbon guest-host system: Li/carbon nanotubes. Two types of carbon nanotubes can be distinguished according to their structural properties: multiwall (MWNT) and single wall (SWNT). 6 MWNT consist of graphitic sheets rolled into closed concentric cylinders, with a structure similar to that of Russian dolls. The concentric tubes are separated by Van der Waals gaps of ,3.4 A, a typical interlayer spacing in turbostratically disordered graphite. External diameters can be as large as 50 nm, and lengths are of micrometer scale. SWNT can be envisioned as a single graphene sheet rolled into a cylinder, with diameters in the range 1-2 nm and lengths of several micrometer. SWNT of nearly uniform diameters self-organize into long crystalline “ropes” in which parallel nanotubes are bound by Van der Waals forces. 7 The diameter of a rope is typically 10-50 nm corresponding to 30-600 tubes per rope. Ropes containing as few as 2-3 tubes or as many as several thousand are occasionally found. Figure 1 presents a high resolution transmission electron microscope (HRTEM) image of purified and annealed SWNT, in which several entangled ropes with different diameters can be observed. The parallel fringes within each rope are due to the constructive scattering from the parallel planes of SWNT. The fact that the fringe spacings differ among ropes does not arise from a wide distribution in nanotube diameters, but rather from the different orientation of each rope zone axis with respect to the electron beam. Figure 2 shows an X-ray profile from purified and annealed SWNT. The well