Knowledge of the surface stress anisotropy is a prerequisite for the description of the mesoscopic structure of stepped Si(001) surfaces. However, it is known that results from first-principles calculations for the (1×2) reconstructed surface surpass the experimental value by about a factor of 2. We demonstrate that this discrepancy is primarily not an extrinsic effect due to surface defects but a consequence of the elastic interaction between the dimers on the defect free surface.

Quantum dots with high lateral quantization energies are created on InSb by means of a perforated Schottky gate. With a gate voltage the lateral quantization energies can be tuned and approach values up to ħ${\mathrm{\omega }}_0$≊25 meV, which is comparable to the subband separations connected with the vertical motion in the corresponding quasi-two-dimensional electron system. By far-infrared spectroscopy, we observe the two fundamental modes ${\mathrm{\omega }}_{\pm }$ in a magnetic field and verify the predicted polarization selection rules. The measured transition energies are described by employing a realistic three-dimensional model potential and taking into account the conduction band nonparabolicity of InSb and the resulting coupling of the vertical subbands.

We have calculated analytically the potential generated in a two-dimensional electron gas by a lateral superlattice on a GaAs/${\mathrm{Al}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$As heterostructure, formed by an array of parallel metal striped gates on the surface. The shape of the electrostatic potential produced by a bias on the gates depends on the boundary condition on the exposed surface of GaAs in the gaps between the gates. The usual assumption of a pinned GaAs surface gives a potential containing only odd harmonics. This is not in agreement with a recent experimental measurement of the potential from commensurability oscillations in the magnetoresistance. An alternative model where the surface charge is frozen in the gaps gives an asymmetry between the gates and the gaps which is closer to experiment. A quite different origin of the potential is a result of elastic strain resulting from differential contraction between the metal gates and the GaAs as the sample is cooled, coupling through the deformation potential. This gives a good fit to experiment and shows that the strain from mismatched gates cannot be neglected in electronic transport phenomena that are sensitive to weak potentials.

The phase transition of Si(001) has been studied by investigating the temperature dependence of streak patterns in low-energy electron diffraction above the transition temperature. The streak pattern remains up to well above the transition temperature of c(4×2) to 2×1. The temperature dependence of the width and the length of the streak cannot be described by a simple two-dimensional Ising system. The result is discussed in terms of effects of a strong anisotropic coupling between adjacent asymmetric dimers, dimer vacancies, and a small amount of the p(2×2) configuration.

When current is forced through a quantum Hall bar it is understood that it flows primarily through edge states. These represent extra charges that in turn produce a more widely distributed current density that falls off rather slowly with distance and then drops rapidly as a result of screening by particles beyond the depletion region. We have solved numerically the self consistent Schro¨dinger equation in the Hartree approximation for ideal samples with small widths and sharp boundaries, and combined these results with the Wiener-Hopf technique to obtain an accurate picture of the situation for macroscopically wide samples. Our results indicate that the redistribution of states in the bulk of a quantum Hall bar is a very important effect and that the bulk states contribute to a significant fraction of the total current.

We have determined the band structure for the surface-localized states on a clean and hydrogen-covered Rh(111) surface using angle-resolved photoemission. Several of our observed bands are strikingly similar to those observed previously on Pt(111) if the Fermi level is shifted appropriately given the fewer valence electrons in rhodium. A deviation from this rigid-band behavior is observed for surface bands in the vicinity of the Fermi level near the K¯ point of the surface Brillouin zone. We also compare our results to existing calculations. We find that these generally provide a good match to our results, although more surface bands are observed than calculated.

We have considered the effects of electron energy losses on experimentally measured inverse-photoemission spectra. The approximations needed to cast the equations for inverse photoemission into a form such that the measured inverse-photoemission spectrum can be expressed as the convolution of two functions—the true spectrum and a response function—are discussed qualitatively. The true spectrum is the inverse-photoemission spectrum that would result if there were no inelastic background present. The response function is the energy spectrum of the inelastically scattered incident electrons inside the sample; it is modeled by an electron-energy-loss spectrum. The true spectrum is found by deconvoluting the response function from the measured spectrum. This is the first time, to our knowledge, that an experimentally measured response function has been used for background subtraction in ultraviolet inverse photoemission. Results of this background subtraction method are presented for inverse-photoemission spectra of polycrystalline silver and UHV-cleaved ${\mathrm{V}}_2$${\mathrm{O}}_5$(001).

Two previously unappreciated features in photoemission spectra from alkali atoms adsorbed on W(110), namely, the sign of the alkali-induced surface-atom core-level shift of the substrate at low coverage and the very large alkali shallow core-hole lifetime width at all coverages, show that the alkali-substrate interaction is not well described by a transfer of alkali charge. Instead, both features point to the formation of a charge cloud between the alkali adatom and substrate that is derived largely from alkali valence states.

The surface structure of Ni{001} and the adsorption systems p(2×2) O/Ni{001} and c(2×2) O/Ni{001} have been studied by shadow-cone-enhanced secondary-ion mass spectrometry. The secondary ${\mathrm{Ni}}^{+}$-ion intensity has been measured as a function of the incidence angle of the primary ${\mathrm{Ar}}^{+}$-ion beam. The enhanced intensity features in the spectra are compared with results from a two-body-interaction calculation which uses the Moliére approximation to the Thomas-Fermi potential. For the clean Ni{001} surface the analysis indicates that the spacing between the first and second layer decreases from the bulk value of 1.76 to 1.68±0.06 Å. Moreover, the spacing between the second layers and the third layer remains nearly bulklike at 1.74±0.16 Å. For the oxygenated surfaces the O-Ni bond length is determined to be 1.96±0.05 Å, which corresponds to a height for the oxygen of 0.85 Å above the Ni substrate. The presence of oxygen is found to cause buckling in the second Ni layer in accord with previous low-energy-electron-diffraction observations. The magnitude of the buckling is found to be 0.26±0.12 Å and 0.20±0.10 Å for the p(2×2) and c(2×2) surfaces, respectively. These results suggest that the O-Ni bonding is highly localized and only modestly dependent upon coverage.

A systematic study of surface structures of Li-covered Cu(110) was performed by means of low-energy electron-diffraction (LEED) and Auger electron spectroscopy techniques. The experiments were carried out with increasing deposition of Li both at low temperature (90 K) and room temperature (300 K). At 90 K a series of LEED patterns, (3×1), (2×1), (3×1), and successive complicated structures were observed with increasing Li coverage. The sequence of the LEED patterns was interpreted as simple overlayer structures of Li in terms of a continuous increase of the packing density of deposited Li atoms sitting in the trough along the [11¯0] direction of the Cu(110) surface. Marked differences in the sequence of the LEED structures were observed for observations at 300 K; the sequence of the structures was (1×2), (1×1), (4×1), (5×1), and (n×1) where n<8 with an increase of Li coverage. The first (1×2) structure was assigned to a missing-row-type restructuring of the substrate Cu atoms. The structure models for the (4×1), (5×1), and so on were discussed in terms of the formation of Li-Cu surface alloys proposed previously by Tochihara and Mizuno for the Li/Cu(100) system.

The diffuse scattering of conduction electrons at a rough surface has a considerable contribution to the resistivity of thin films, which is well known as the classical size effect. We have separated this contribution from other scattering mechanisms such as the phonon and bulk defect scattering by the following procedure: Silver was deposited at low temperatures (≊130 K) onto a well-annealed relatively thick silver base layer (≊24 nm), which has been deposited onto a Si(111)-7×7 substrate. During deposition the resistivity of the thin film was measured in situ, and afterwards the surface roughness was determined quantitatively with profile analysis of low-energy electron diffraction. An increase of both resistivity and surface roughness was observed. Such an increase of the measured resistivity must be due to the increase of surface roughness because there was no change in temperature or concentration of bulk defects. With the measured surface roughness we are able to evaluate the additional resistivity without any free parameters. The experiment reveals excellent agreement with the theoretical predictions.

The problem of coupling of the shear oscillations of a quartz crystal bounded by a rough surface with damped waves in a liquid has been investigated. The shift of the resonance frequency of a quartz crystal due to either random roughness or periodical corrugation has been considered. A relation between the shift and a correlation function of roughness has been found. It has been shown that the roughness effect is most pronounced in the low viscosity limit. It has been demonstrated that the results of quartz-crystal microbalance measurements can provide valuable information on the average height and the correlation length of surface roughness.

A Monte Carlo study of the ordering of In atoms embedded in the top layer of a Cu(100) surface is presented. The interaction energies between the In and Cu atoms were derived from atom-embedding calculations, with Finnis-Sinclair potentials. It was found that the interaction between In atoms in the close-packed 〈110〉 rows is repulsive, while it is attractive in the 〈100〉 direction. The strengths of the repulsive and attractive interactions were found to be such that it is energetically favorable for In atoms to form one-dimensional rows in the 〈100〉 direction instead of a two-dimensional c(2×2) structure. The ordering of In atoms was studied for In coverages between 5% and 25% of a monolayer and for temperatures between 50 and 2000 K. Below a critical temperature, one of the two 〈100〉 diagonal directions prevails. Above this critical temperature, the system consists of two types of domains in each of which one diagonal direction dominates. Close to the critical temperature these domains are large and become smaller with increasing coverage. From a statistical analysis of the Monte Carlo simulation results, the entropy, internal energy, and average In-row length were calculated as a function of temperature and In coverage. The results were compared with the results of perturbed-angular-correlation experiments. The main trends observed in the experiments were reproduced by our calculations.

The surface scattering of kilovolt-energy ions and atoms at glancing-incidence angles has contributed substantially to knowledge of elementary electronic excitations in the interface region. We consider the influence of inelastic exchange of mechanical energy, due to recoil of the target atoms, on the distribution of scattered projectile atoms. Although with a glancing-incident-angle geometry the relative energy loss by this mechanism is small, it can have a substantial effect on the angular distribution of the scattered particles. These conclusions are supported by numerical calculations of energy losses and scattered particle distributions.

Reflection high-energy electron diffraction with 200-ps time resolution is used to study the structural behavior of the three low-index faces of Pb. The surfaces are subjected to heating and cooling rates on the order of ${10}^{11}$ K/s. The open Pb(110) surface is seen to reversibly disorder below the bulk melting temperature, ${\mathit{T}}_{\mathit{m}}$. In contrast to Pb(110), the close-packed Pb(111) surface sustains superheating to ${\mathit{T}}_{\mathit{m}}$+120 K. Pb(100), a surface with atomic packing intermediate to that of Pb(110) and Pb(111), shows evidence of residual order at a temperature above the bulk melting point.

The work function and photoemission from p-type CdTe(100) single-crystal phosphorus doped to ∼${10}^{15}$ ${\mathrm{cm}}^{\mathrm{-}3}$ and crystals that were subsequently highly doped to ∼${10}^{19}$ ${\mathrm{cm}}^{\mathrm{-}3}$ with Li and covered with submonolayers of Cs were studied. In some instances the cesium-coated surfaces were modified by deposition of oxygen. Large work-function changes were found in CdTe at the initial deposition stages of Cs indicating the formation of large dipole moments on the CdTe(100) substrate. Band-gap-limited (at 1.56 eV) long-wavelength photoemission was observed from the above samples. The total spectral photoemission for visible and near-infrared regions was determined.

X-ray-reflectivity measurements have been carried out on silver films which were vapor deposited onto silicon substrates, to investigate the thickness evolution of the film’s surface roughness. The growth exponent was found to be β=0.26±0.05, and the roughness exponenet was found to be H=0.70±0.10.

High-energy ion-scattering spectroscopy, x-ray-photoemission spectroscopy, and low-energy electron diffraction (LEED) were used to study the growth of very thin Ti films on Al(110) surfaces at room temperature. The Al surface peak areas in the backscattering spectra of 0.96-MeV ${\mathrm{He}}^{+}$ ions, incident along the [1¯ 1¯0] direction of Al(110), decreased sharply during the deposition of the first five monolayers of Ti. This suggests a growth model in which a pseudomorphic Ti film with a fcc structure forms, shadowing the Al surface atoms. This model is supported by LEED measurements where sharp diffraction spots persisted with the rectangular symmetry of the substrate lattice structure. Throughout this coverage regime, the attenuation of the Al photopeak intensities as a function of Ti coverage agrees with the attenuation rate calculated using a laminar growth mode. With additional Ti deposition the Al surface peak intensities increase while the Al photopeak intensities remain unchanged. A discussion of possible growth models for the high-coverage regime is presented.

Assuming a model tip (${\mathrm{Si}}_4$${\mathrm{O}}_{10}$${\mathrm{H}}_{10}$) as a reasonable representation of the surface of a ${\mathrm{Si}}_3$${\mathrm{N}}_4$ cantilever stylus having a hydrogen-terminated asperity and a broader load-bearing base, we investigate the interaction of an atomic force microscope (AFM) with an alkali halide crystal by quantum chemical methods. Structural relaxation of the sample during engagement is allowed, and defect formation is investigated. Force curves above cation and anion positions are calculated, determining maximum sustainable loads and indicating a basis for atomic contrast. Experiments using a ${\mathrm{Si}}_3$${\mathrm{N}}_4$ cantilever for AFM imaging of 12 alkali halide and alkaline earth fluoride crystals in air and desiccated helium are reported, in the widest AFM survey of such materials to date. Adsorbed water is shown to significantly enhance the observation of atomic periodicity on ionic halide samples, and rapid surface diffusion on alkali halide crystals is illustrated as it affects prospects for defect investigations. Observations of step edges and point-defect candidates at atomic scale are reported. The theoretical and experimental results are discussed together in the effort to provide a quantum-mechanical model for observations of alkali halide samples at atomic resolution, and to examine a possible basis for atomic resolution in the presence of long-range attractive forces.

We have carried out high-sensitivity measurements of the emission of Na atoms induced by irradiation with μs electron pulses of unirradiated NaCl (100) surfaces and with ns laser pulses of the NaCl (100) surfaces preirradiated with electrons. Similar to the results of ion bombardment, irradiation with an electron pulse is found to cause spontaneous emission of Na atoms with observable delays below 100 °C. It is found also that irradiation with a ns laser pulse of the preirradiated surface causes emission of Na atoms. The excitation spectrum for the laser-induced emission shows a peak near the F-band maximum. The yields of both the initial electron-pulse-induced emission and laser-induced emission on preirradiated surface are found to be reduced as the surface is damaged.