The conduction band electronic structure in periodically Si-doped GaAs was studied by Shubnikov–de Haas measurements in tilted fields of 0–14 T. The samples had a doping period of 140 Å and 220 Å, with a sheet carrier density per superlattice period of 2.8×${10}^{12}$ ${\mathrm{cm}}^{\mathrm{-}2}$ and 2.2×${10}^{12}$ ${\mathrm{cm}}^{\mathrm{-}2}$, respectively. The periodicity and the intensity of the oscillations in the magnetoresistance, as a function of the tilt angle, follow accurately the theoretical values obtained from a self-consistent calculation of the Fermi surface. An extra period appears in the spectra when the magnetic-field direction is tilted away from the axis of the superlattice, which is seen as a peak in the Fourier transform of the oscillations. The angular dependence of the position and of the intensity of this peak allows us to establish that it arises because of magnetic breakdown.

The band structure, total energies, and relaxed geometries are calculated for the C(111), C(100), and C(110) surfaces using a parametrized tight-binding model for carbon. The method and addition of C-H parameters to the model are described in detail. Results for the bare and hydrogenated C(111) surfaces are used to compare the accuracy of the method with ab initio techniques. A stable hydrogenated (2×1) π reconstructed surface is found, which resembles the C(110) surface. Removal of one H atom from the dihydride C(100) results in a 3/2 hydride surface, where the odd hydrogen is bonded equally to two surface carbons. Although the fully H-covered C(100)(2×1) surface has a clean gap, the partially covered surface has a half-filled state, consistent with photoemission data. The geometries and H vibrations are also presented for the C(110) surface. The surface chains on the bare C(110) have bond lengths close to graphite and dimerize from a Peierls distortion. Addition of H to this surface restores the bond lengths to approximately that of bulk diamond. Comparison of the band structures and H vibrations with experiment helps identify the nature of the hydrogen coverage on the surfaces.

Confined and interface optical phonons are observed in InAs/InP single quantum wells. Under InAs ${\mathit{E}}_1$-gap resonant excitation, confined longitudinal-optical phonons of ${\mathit{A}}_1$ and ${\mathit{B}}_2$ symmetries are observed in both z(x,x)z¯ and z(x,y)z¯ polarization configurations. This relaxation of the selection rules is analyzed in terms of impurity-induced scattering. The enhancement of the intensity of the confined modes under resonant excitation indicates that the scattering involves ${\mathit{E}}_1$-gap excitons strongly confined in the InAs well. Both InAs-like and InP-like interface phonons are observed. Their energy positions are well explained by an electrostatic continuum model, assuming that two distinct mechanisms contribute to the presence of nonzero wave-vector modes. The first involves impurity-induced scattering and the second is attributed to a modulation of the InAs well thickness.

The importance of intraband Auger processes in determining the ionization balance in quantum dots is reported. The numerically inexpensive binary-encounter model for a Coulomb collision between identical particles is found to be a good estimator of the intraband Auger rates out of a quantum dot. Intraband and the conventional interband Auger processes differ in that the former involve only intraband transitions whereas the latter always involve a radiationless interband transition. As such, intraband Auger rates do not involve the evaluation of the very small overlap integral of a conduction band with a valence band Bloch wave function and are thus much larger than interband Auger rates, especially for large-band-gap semiconductors like GaAs. Though intraband Auger processes are not strong enough to establish a quasiequilibrium within the entire conduction band at the room-temperature free-carrier concentrations (${10}^{16}$ ${\mathrm{cm}}^{\mathrm{-}3}$) and bound energy separations (greater than an LO phonon energy) commonly assumed in the quantum-dot literature, they are capable of placing almost as many bound carriers in states near the band edge as would be predicted erroneously by a quasiequilibrium Fermi-Dirac distribution.

Such large bound state occupations are important for quantum-dot laser design. A sufficient condition for a quasiequilibrium to exist within all of an energy (conduction or valence) band is found to be the existence of many inverse Auger processes faster than interband spontaneous emission, which occurs for total (bound plus free) electron concentrations greater than 5×${10}^{17}$ ${\mathrm{cm}}^{\mathrm{-}3}$ at room temperature in 100 Å radius GaAs/${\mathrm{Al}}_{0.3}$${\mathrm{Ga}}_{0.7}$As quantum dots whose centers are separated by 400 Å. The nonlocal thermodynamic equilibrium populations in quantum dots can be understood from a simple model in which states connected by fast Auger or phonon processes are in Saha-Boltzmann equilibrium. All other states have occupation factors which are determined by the ratio of intraband collisional to interband radiative lifetimes, as described by a Fokker-Planck equation modeling diffusion in energy of bound particles.

We have calculated the surface energy and work function of the light actinides Fr, Ra, Ac, Th, Pa, U, Np, and Pu by means of a Green’s-function technique based on the linear-muffin-tin-orbitals method within the tight-binding representation. In these calculations we apply an energy functional which combines the kinetic energy calculated within the atomic-sphere approximation with Coulomb- and exchange-correlation-energy terms calculated by means of the complete nonspherically symmetric charge density derived from the atomic-sphere potential within nonoverlapping and space-filling cells. The calculated surface energies and work functions are in good agreement with the limited experimental data.

Angle-resolved photoemission spectra of the Bi(111) surface are presented. The results confirm the existence of a surface state in a spin-orbit gap. In addition, a bulk band is found to cross the Fermi energy. These features may be involved in the superconducting properties of granular systems built from Bi clusters. Moreover, the Bi 5d core level is found to be shifted at the surface by 0.22±0.02 eV towards lower binding energy, which is in excellent agreement with theoretical prediction.

We introduce a model to calculate the equilibrium crystal configuration of a monolayer lattice of large, planar organic molecules bonded to a substrate by van der Waals (vdW) forces. The model significantly simplifies analysis by replacing the conventional atom-atom vdW potential summation with a single ellipsoidal potential centered in the molecular plane. Our results indicate that recent observations of crystalline quasiepitaxial vacuum growth of incommensurate lattices of these planar molecular films result from the relatively large intralayer stiffness as compared to the interlayer shear stress. Good agreement between calculated and observed structures is achieved using no adjustable parameters. The model is used to predict molecular structures which are likely to form quasiepitaxial layers. Comparison with previous models describing physisorption of incommensurate layers of atomic vdW systems on graphite substrates is also made. In the subsequent paper (paper II) we present experimental data comparing the grown structures of several model compounds with the theoretical predictions made here.

We study the growth of the archetypal planar molecular compounds; 3,4,9,10 perylenetetracarboxylic dianhydride 3,4,7,9 naphthalenetetracarboxylic dianhydride, and copper phthalocyanine on a variety of substrates by the ultrahigh-vacuum process of organic molecular-beam deposition. The thin-film structures are grown as monolayers, multilayers, alternating thin-film stacks (or multiple quantum wells), and bulk thin films with thicknesses extending to 4000 Å. Techniques used to investigate these structures include scanning tunneling microscopy, in situ reflection high-energy electron diffraction, electron mi- croscopy, x-ray diffraction, and optical and electronic characterization. We find that highly ordered crystalline thin-film growth (i.e., quasiepitaxy) can often be achieved independent of the commensurability of the grown layer and the substrate. Such structural ordering is shown to strongly influence both the optical and electronic properties of the films. The structures experimentally determined for these film systems are consistent with the theoretical predictions made in paper I, where we analyze films consisting of large planar organic molecules bonded by van der Waals forces. These results provide evidence for the generality of quasiepitaxial growth of such molecular thin-film systems.

Coherent transient effects in CdSSe microcrystallites are observed by a pump-probe measurement. The probe transmission shows oscillatory structures in the temporal and spectral domains when the probe reaches the sample before the pump. This oscillation is attributed to the photon echo propagating in the direction of the probe. The A-B splitting of the valence sublevel in CdSSe microcrystallites hidden in the wide inhomogeneous broadening can be resolved using the polarization dependence of the transitions from A and B valence sublevels to the conduction sublevel.

Co films of 1000-Å thickness were grown by molecular-beam epitaxy on substrates of mica, oxidized silicon, and glass at temperatures between -33 °C and 800 °C. X-ray diffraction and nuclear magnetic resonance were used to study the structural quality and phase composition as a function of growth temperature and type of substrate. The magnetic anisotropy of the films was measured using vibrating-sample magnetometry. The film quality is observed to improve sharply for growth temperatures above 200 °C. Below this temperature the films have a strong polycrystalline character. Above this temperature the films on mica and oxidized silicon show well defined (111)/(00.2) texture and coherence lengths on the order of the film thickness. The structural phase composition depends on both substrate and growth temperature. For growth on mica at about 500 °C an almost single phased hcp Co film was obtained. Above this temperature the fcc phase dominates the structure, whereas below this temperature stacking faults dominate. For the films grown on oxidized silicon the structure is dominated by stacking faults and the fcc phase over the whole growth temperature range. The overall magnetic anisotropy of the films strongly depends on the structural phase composition. Its magnitude can be satisfactorily described by a compositional average of the respective bulk anisotropies.

Two sources of elastic light scattering, usually associated with scattering by particles, by bulk, polycrystalline samples of partly ionic compounds are demonstrated. In the visible part of the spectrum where the material is transparent, subsurface inhomogeneities cause backscattering which contributes to the total reflectance of the surface. This effect is modeled with a combination of Mie scattering by isolated spherical voids filled with air, and Monte Carlo calculations to account for multiple-scattering events. In the infrared spectral range, these compounds typically exhibit one or more lattice resonances. Such a resonance causes strong dispersion in the optical constants, resulting in a wavelength at which the conditions for the lowest surface-mode resonance are satisfied. It is shown that small inhomogeneities on the surface are sufficient to give a measurable contribution and cause a substructure in the reststrahlen band which has been observed previously but not fully understood. Light-scattering measurements over the wavelength range 0.3–20 μm for bulk samples of polycrystalline beryllium oxide and 2–20 μm for silicon carbide powder are reported to verify the interpretation. A coating of polyethylene was applied which shifted the surface resonance to longer wavelengths in agreement with the theory for isolated-particle scattering.

The near-field optical interaction between a pointed detector and dielectric or metallic substrates can be exploited to perform nanometer-scale surface spectroscopy. This paper presents a general framework for a realistic description of the optical near field with application to local spectroscopy of metallic aggregates deposited on a transparent sample. The treatment is based on the field-susceptibility method associated with perturbation theory. The practical solution of this model by discretization in real space is obtained by a self-consistent procedure which takes all multiple-scattering effects into account. Numerical results illustrate the evolution of direct space images when scanning metallic aggregates at various frequencies. Our simulations clearly show the interest of such a local spectroscopy for morphologic studies and for the characterization of surface mesoscopic structures.

Auger photoelectron coincidence spectroscopy (APECS) confirms that a substantial part of the low-energy tail of the ${\mathit{L}}_{23}$VV Auger lines of the 3d transition metals is due to intrinsic initial- and final-state shake-up/shake-off processes and not just secondary losses. Data were collected for the ${\mathit{L}}_{23}$VV Auger lines of pure Cu, Co, and Ni. APECS enables a direct experimental measurement of the initial-state shake-up/shake-off satellites in solid-state Auger spectroscopy. The importance of these effects in Auger spectroscopy as a qualitative and quantitative surface-analysis technique is discussed.

A process based on the Kratschmer-Huffman carbon arc method of preparing fullerenes has been used to generate carbon-coated cobalt and cobalt carbide nanocrystallites. Magnetic nanocrystallites are extracted from the soot with a gradient field technique. For Co/C composites, structural characterization by x-ray diffraction and high-resolution transmission electron microscopy reveals the presence of a fcc Co phase, graphite, and a minority ${\mathrm{Co}}_2$C phase. The majority of Co nanocrystals exists as nominally spherical particles, 0.5–5 nm in radius. Hysteretic and temperature-dependent magnetic response, in randomly and magnetically aligned powder samples frozen in epoxy reveals fine-particle magnetism associated with monodomain Co particles. The magnetization exhibits a unique functional dependence on H/T, and hysteresis below a blocking temperature, ${\mathit{T}}_{\mathit{B}}$≃160 K. Below ${\mathit{T}}_{\mathit{B}}$, the temperature dependence of the coercivity is given by ${\mathit{H}}_{\mathit{c}}$=${\mathit{H}}_{\mathit{c}\mathit{i}}$[1-(T/${\mathit{T}}_{\mathit{B}}$${)}^{1/2}$], with ${\mathit{H}}_{\mathit{c}\mathit{i}}$≃450 Oe.

Several models for the elementary processes causing the emission of alkali atoms by electronic excitation of NaCl (100) surfaces have been investigated theoretically. First, the desorption of a Na atom neighboring an electronically excited F center on the surface is simulated using a quantum-mechanical embedded-cluster technique. It is shown that emission of a Na atom is energetically favorable. The kinetics of this process is shown to be controlled by the probability of a nonradiative transition between the two states: the excited state of the F center and that corresponding to a Na atom desorbing from the surface. The potential barrier for desorption of an excited Na atom from the excited F-center state is found to be 2.1 eV. It is also found that the energy for emission of a Na atom from a cluster of F centers (the ${\mathit{F}}_3$ center) is considerably reduced (for a certain configuration of the defect) with respect to the similar energy for a single F center. The energy barrier for emission of a Na atom neighboring an F’ center on the surface is calculated to be 1 eV. It is shown that the electronic excitation of kinklike sites, with a Na atom at the edge, can lead to a barrierless emission of a Na atom, leaving a ${\mathit{V}}_{\mathit{k}}$-type defect behind. The results of calculations are discussed critically on the basis of existing experimental data.

Classical trajectory calculations were employed to study oxygenated (100) diamond surfaces. Atomic forces were computed from two types of potential energy functions, a quantum mechanical potential based on the semiempirical PM3 (parametric method number 3) Hamiltonian which was used to describe the central code core of the surface model and an empirically parametrized potential that was developed to extend the size of the model surface. The results indicate that the energetically most favorable structure of an oxygenated surface is a bridge oxygen monolayer. However, due to the effect of entropy, an on-top oxygen monolayer dominates at high temperatures. Surfaces saturated with oxygen molecules were found to be unstable even at room temperature and to convert to the on-top atomic oxygen structures. Reconstruction of the surface from a (1×1) to a (2×1) configuration was observed to take place when half of the CO was removed from the on-top oxygen monolayer.

We have calculated the electronic structure and segregation profiles of the (001) surface of random Cu-Ni alloys with varying bulk concentrations by means of the coherent potential approximation and the linear muffin-tin-orbitals method. Exchange and correlation were included within the local-density approximation. Temperature effects were accounted for by means of the cluster-variation method and, for comparison, by mean-field theory. The necessary interaction parameters were calculated by the Connolly-Williams method generalized to the case of a surface of a random alloy. We find the segregation profiles to be oscillatory with a strong preference for Cu to segregate towards the surface of the alloy.

In recent calculations of optical response at a metal surface, an unusual set of oscillations have been found in the induced charge density. These oscillations only appear when a crystal potential is included in the calculation. They propagate deep into the bulk, but are incommensurate with the lattice. We illustrate the occurrence of such oscillations in simple systems, and discuss their origin. For more general cases, we find that the amplitude of the oscillations is sensitive to the amount of spatial variation in directions parallel to the surface of either the crystal potential or the probing field. We also discuss the challenge of detecting this phenomenon experimentally.

We present a treatment of the surface of liquid alkali metals based on a perturbative expansion to second order in the electron-ion pseudopotential. We make a simple direct approximation for the inhomogeneous electronic response function, in contrast to previous work which approximates the effective pair interaction. It is found that the surface tension, obtained by using a nonmonotonic parametrization of the ionic profile, is in good agreement with the experimental value and improves the results of the previous approaches.

Collisions of ${\mathrm{C}}_{60}$ with hydrogen-terminated diamond-(111) surfaces were studied by molecular-dynamics simulations based on a semiempirical density-functional approach. The dominating factor determining the energy partition in the ${\mathrm{C}}_{60}$-diamond-(111)-surface collision at constant impact energy is the orientation of the molecules relative to the initial impact points on the surface. In agreement with the experimental results using velocity-selective time-of-flight mass spectroscopy the center-of-mass kinetic energies after the collisions are distributed around a mean value which is only slightly affected by the initial impact energy. The final energy content of the ${\mathrm{C}}_{60}$’s increases with the incident kinetic energy, depends strongly on the impact scattering topology, and yields a narrow distribution with increasing mean values in correlation with lower center-of-mass kinetic energies.