Monte Carlo simulation is used to calculate the composition profiles of surface segregation of Cu-Ni alloys. The method of Bozzolo, Ferrante, and Smith [Phys. Rev. B 45, 493 (1992)] is used to compute the energetics of these systems as a function of temperature, crystal face, and bulk concentration. The predictions are compared with other theoretical and experimental results.

We have utilized static minimum-potential-energy calculations to determine microscopic diffusion mechanisms of the simplest molecular adsorbate, namely a dimer, on a periodic substrate. We consider a rigid homonuclear dimer physically adsorbed on a SC(100) crystal surface by means of a Lennard-Jones potential. We show that, by varying the dimer length relative to the lattice constant, a wide range of rich and varied diffusion behavior can be achieved. The model has experimental analogs and implications, which will be discussed.

A recently developed formalism for calculating ion-scattering yields from simulated crystal surfaces has been extended to the case of bimetallic surfaces. MeV ${\mathrm{He}}^{+}$ scattering yields are calculated for thin Ni films reacting with Al(110) surfaces. Interatomic potentials based on the embedded-atom method are used in a Monte Carlo approach to simulate the evolution of the Ni-Al interface as a function of Ni coverage. The calculated ion-scattering yields are in good quantitative agreement with the measured yields, showing two distinct stages of reaction and overlayer growth. Limitations of the simulation and measurement methods are discussed.

Comparison of valence-photoelectron to Auger-electron spectra has been carried out for three phases of ${\mathrm{K}}_{\mathit{x}}$${\mathrm{C}}_{60}$, x=0, 3, 6, yielding Hubbard correlation energies, U, of 1.4 eV (1.5 eV) for x=0 (6), consistent with the results of Lof et al. [Phys. Rev. Lett. 68, 3924 (1992)]. At x=3 the results are best described by intermolecular charge-transfer screening of the C 1s core hole in the Auger initial state, with U of the same magnitude as for x=0 and 6.

Electronic-structure calculations have been carried out on the ordered delafossite-type superoxides ${\mathrm{YCuO}}_{2+\mathrm{\delta }}$ (0.5≤δ≤0.7) that Cava et al. have synthesized recently by inserting interstitial O into the triangular Cu planes of ${\mathrm{YCuO}}_2$. The results exhibit a unique feature where each interstitial O inserts two half-filled, cluster-derived ${\mathrm{Cu}}_3$${\mathrm{O}}_7$ ‘‘impurity levels’’ into the ∼2.7-eV semiconductor gap of ${\mathrm{YCuO}}_2$. These levels evolve into half-filled ‘‘impurity bands’’ as adjacent clusters are corner-linked to form continuous one-dimensional (1D) linear (δ≊1/2) and 2D planar (δ≊2/3) networks that may exhibit interesting transport and magnetic properties.

The equilibrium volumes and bulk moduli of nonmagnetic 3d, 4d, and 5d transition metals have been calculated using a full-potential linear-muffin-tin-orbitals (FP-LMTO) method and the generalized gradient approximation (PW91) of Perdew and Wang [in Electronic Structure of Solids 1991, edited by P. Ziesche and H. Eschrig (Akademie Verlag, Berlin, 1991), Vol. 11]. The gradient-corrected FP-LMTO results are found to be superior to the corresponding local-density approximation (LDA) results in all transition-metal series. Furthermore, the atomic-spheres approximation (ASA) is found to yield larger equilibrium volumes and smaller bulk moduli than the FP method. The inaccuracies introduced by the ASA are of the same magnitude as the gradient corrections, suggesting that a FP treatment is important for gradient-corrected calculations. Excellent agreement with experiment is found for the c/a ratios of the hcp metals using both the LDA and the PW91, while for the fcc-bcc structural energy differences of 4d metals disagreement with semiempirical estimates remains.

The diffusivity of positively charged hydrogen (${\mathrm{H}}^{+}$) in GaAs has been directly determined from capacitance transient measurements. The hydrogen was released from donor-hydrogen complexes in the space charge layer of Schottky diodes on n-type GaAs by pulsed-laser injection of minority carriers. Capacitance-voltage measurements reveal the recovery of donor dopants after the injection, which demonstrates minority-carrier-enhanced dissociation of a donor-hydrogen complex in GaAs. Analysis of capacitance transients recorded during migration of ${\mathrm{H}}^{+}$ were analyzed to obtain diffusivities near room temperature; at 320 K the diffusivity is ∼1×${10}^{\mathrm{-}12}$ ${\mathrm{cm}}^2$/sec, within a factor of 2. An Arrhenius analysis of the migration time constant yields an activation energy for ${\mathrm{H}}^{+}$ diffusion of ∼0.66 eV.

A suppression of the exciton dispersion by an incompressible quantum liquid strongly affects the recombination emission spectrum. It comprises two parts; one of them originates from a direct recombination of k=0 excitons, and the other from indirect, single magnetoroton, transitions from an extensive area in k space, kl≊1, l is the magnetic length. Intensities of the peaks show opposite temperature dependencies. We show that the minimal frequency of magnetorotons may be found from emission spectra by a proper treatment of experimental data.

We calculate how noise generated by a finite environmental impedance limits the ultimate performance of Coulomb-blockade devices and possible current or charge standards. We have expressed the environmental theory of the Coulomb blockade in terms of the spectral density of the voltage noise arising from the environment, and have calculated the resulting single-junction tunneling and two-junction cotunneling rates. These rates are used to predict the tunneling rate of electrons through a five-junction pump and may explain the anomalously large rates that are observed experimentally.

When an electron tunnels into a quantum dot, it may be stuck in a state higher than the lowest available singe-particle level. Equilibration with a lead is greatly slowed down by the Coulomb blockade in the dot and Pauli exclusion in the lead at low temperatures. This may produce heating of an electron turnstile or pump, if the intradot relaxation rate is slow compared with the pumping frequency. These ideas are demonstrated by simple models with realistic parameters, and the effect of phonon-induced intradot transitions is estimated in some detail.

BEEM spectroscopy has been used to characterize hot-carrier transport through AlAs/GaAs heterostructures. The dependence of electron transmission on AlAs thickness has been directly measured, and the position of the AlAs ${\mathit{L}}_1$ minima, which has been subject to some uncertainty in the past, has been determined. First-principles transmission calculations, based on a tight-binding formalism, are compared to the results of BEEM spectroscopy.

We have performed a theoretical (first-principles) and experimental (contactless electroreflectance) investigation of the temperature dependence (17<T<351 K) of the electric-field-modulated interband transitions in a pseudomorphic ${\mathrm{Ga}}_{0.81}$${\mathrm{Al}}_{0.19}$As/${\mathrm{In}}_{0.20}$${\mathrm{Ga}}_{0.80}$As/GaAs modulation-doped quantum well that contains a two-dimensional electron gas. We find that many-body effects are a factor at all temperatures although they are most pronounced for T<=152 K. Because of the derivativelike nature of our approach, both theoretical and experimental, the weak Fermi-edge transition is still visible up to at least 152 K, as a complement to the results of previous theoretical and experimental investigations.

The electronic transitions responsible for the formation of F centers at the surfaces of epitaxial ${\mathrm{CaF}}_2$ films on Si(111) have been identified via photon-stimulated desorption spectroscopy. Results for films ranging from submonolayer to bulk ${\mathrm{CaF}}_2$ indicate that excitations of the Ca 3p core level result in ${\mathrm{F}}^{+}$ desorption, and hence, in F-center formation, while excitations of the F 2s do not. It is proposed that this difference is due to inward nuclear motion that occurs prior to the deexcitation of the core hole.

The energy loss of a hot electron to a thermalized electron-hole plasma has been studied using the optical recombination of a hot electron at a neutral acceptor. The dependence of the energy-loss rate on plasma density was measured in both bulk and quantum-well samples. In bulk, the energy-loss rate is significantly sublinear as a result of screening. In quantum wells, the rate is nearly linear in plasma density, showing that screening is less important. Measurements in an asymmetric coupled-well sample show that the interaction between a hot electron in one well and a plasma in an adjacent well is much weaker than if they are in the same well.

We calculate conductance versus gate voltage for single-electron tunneling through a lateral semiconductor quantum dot in the Coulomb blockade regime using the calculated self-consistent electronic structure of the device. We show that variations in the level spacings result in an experimentally observed but previously unexplained envelope modulation of peak amplitudes. We present a formula for the activated component of peak conductance as a function of level spacings and tunneling coefficients.

We report on the martensitic fcc-bcc phase transformation for Fe films grown at room temperature on Cu(100) in a thickness range from 10 to 20 monolayers. bcc(110) grains are highly elongated along 〈011${\mathrm{〉}}_{\mathrm{fcc}}$∥[11¯1${]}_{\mathrm{bcc}}$ and exhibit characteristic island growth with steps along [001${]}_{\mathrm{bcc}}$. The martensitic (i.e., sudden, collective, nondiffusive) transformation in the bulk of the film is demonstrated by STM topography, including the behavior of steps at grain boundaries and the presence of tilted surfaces arising from fcc-bcc interfaces.

Diamond (111) and (100) surfaces have been investigated by low-energy electron diffraction and high-resolution electron-energy-loss spectroscopy (HREELS). Samples were boron-doped semiconducting diamond surfaces grown homoepitaxially on natural diamond by microwave-plasma-assisted chemical-vapor deposition. A dipole-active C-H stretching vibration was observed at 352 meV (2839 ${\mathrm{cm}}^{\mathrm{-}1}$) on the (111) surface and 363 meV (2928 ${\mathrm{cm}}^{\mathrm{-}1}$) on the (100) surface. In the off-specular EELS spectra, two C-H stretching modes (352 and 361 meV) were clearly resolved on the (111) surface, while only one C-H stretching vibration was detected on the (100) surface. The two modes on the (111) surface are assigned to symmetric and asymmetric vibrations of ${\mathrm{CH}}_3$ species, respectively. This study clarifies hydrogen termination of the epitaxially grown diamond (111) and (100) surfaces in the forms of ${\mathrm{CH}}_3$ (methyl) and CH (monohydride) groups, respectively.

The melting temperatures of uranium have been measured to 45 GPa with an inert Ar pressure medium in a diamond-anvil cell using a spatially resolved laser-heating system. The temperature was measured within 1.4 μm at the center of laser-heating spot using reflecting optics, and the melting was visually observed as the material flows forming a microchannel and/or by a shape change in the edge of the sample. Theoretically, the melting curve has also been calculated to 100 GPa via γ(bcc) solid and liquid free energies derived from generalized pseudopotential theory. The calculated melting temperatures agree well with the experimental measurements. It is also found that the theoretical melting curve obeys a Lindemann scaling law for solid densities up to about 25.5 g/${\mathrm{cm}}^{\mathrm{-}3}$, corresponding to 100 GPa in pressure.

It was found that the decay time of the Auger-electron-free luminescence of ${\mathrm{BaF}}_2$ decreases with increasing linear energy transfer (LET) or excitation density. The luminescence decayed exponentially at the initial and final stages, in contrast with a single exponential for photoirradiation. The decay time of a fast component was reduced with increasing LET, while that of a slow component was invariant. A model that Auger-free transitions compete with nonradiative recombinations between core holes and quasifree electrons can explain these experimental results. The recombination rate attained a constant value near the LET of 1.5 MeV/amu Kr ion. The reciprocal of the recombination rate at this LET gives the lifetime of the core exciton.

Raman scattering studies were carried out to investigate the effects of ion implantation on the structure of diamond, graphite, and polymers. Carbon phases produced by chemical vapor deposition (CVD diamond) and rf discharge (diamondlike carbon or DLC) were also analyzed. Two types of amorphous carbon phases were distinguished with relevance to hardness. In general, amorphous carbon phases produced by electron beam evaporation and sputtering are soft (hardness ≪1 GPa), while DLC and some ion-beam-modified polymers are much harder. In all cases, the characteristic Raman bands of the starting material were lost upon ion implantation, and for the lowest fluences the one-phonon bands near 1360 ${\mathrm{cm}}^{\mathrm{-}1}$ (D line) and 1580 ${\mathrm{cm}}^{\mathrm{-}1}$ (G line) of disordered polycrystalline graphite appeared. With increasing fluence these bands coalesced into a broad, asymmetric peak with the D line shifting to higher wave number and the G line shifting to lower wave number. This trend was clearly distinguishable from the finite crystallite size effect seen in graphite, where, in addition to the appearance of the D line, the G line shifts to higher wave number with decreasing crystallite sizes. Raman scattering could not distinguish between soft and hard amorphous carbon. There was also no indication that the hardness of DLC films and ion-beam-modified polymers was due to diamondlike ${\mathit{sp}}^3$ bonds. Instead, hardness in these materials is related to the three-dimensional interconnectivity of chemical bonds. Experimental results suggest that the amorphous carbons examined in this study are composed of random networks of distorted sp, ${\mathit{sp}}^2$, and ${\mathit{sp}}^3$ bonded atoms, sometimes in a hydrogenated state. The hard carbons such as DLC films and ion beam modified polymers have long-range chemical connectivity while the soft carbons such as damaged graphite, and carbon films prepared by sputter deposition lack such connectivity.