We study the quantum ground state of the Frenkel-Kontorova model in the ‘‘strongly nonlinear’’ regime in which in the corresponding classical limit the coordinates of the atoms are distributed on ‘‘Cantori.’’ We identify (many) quasidegenerate configurations that contribute to the quantum ground state. When the characteristic quantum and classical energy scales are roughly equal (the ‘‘intermediate’’ quantum regime), we find, consistent with earlier numerical studies, that the ‘‘standard map’’ determining the coordinates in the classical ground state is renormalized to an effective ‘‘sawtooth’’ map, which determines the expectation values of the coordinates in the quantum ground state. We also discuss the dynamics of the model and estimate the characteristic time for various quantum tunneling effects.

We study the transition between sharp and smooth density distributions at the edges of quantum Hall liquids in the presence of interactions. We find that, for strong confining potentials, the edge of a ν=1 liquid is described by the ${\mathit{Z}}_{\mathit{F}}$=1 Fermi-liquid theory, even in the presence of interactions, a consequence of the chiral nature of the system. When the edge confining potential is decreased beyond a point, the edge undergoes a reconstruction and electrons start to deposit a distance ∼2 magnetic lengths away from the initial quantum Hall liquid. Within the Hartree-Fock approximation, a new pair of branches of gapless edge excitations is generated after the transition. We show that the transition is controlled by the balance between a long-ranged repulsive Hartree term and a short-ranged attractive exchange term. Such a transition also occurs for quantum dots in the quantum Hall regime and should be observable in resonant tunneling experiments. We find that the edge theory for sharp edges also applies to smooth edges (i.e., reconstructed edges) once the additional pairs of edge branches are included. Electron tunneling into the reconstructed edge is also discussed.

An approach is presented for calculating the binding energy of neutral donors in semiconductor heterostructures. The technique avoids the need for analytical solutions. A generalization of the method to include central cell corrections, effective mass, and dielectric mismatch at heterojunctions, together with band nonparabolicity, is discussed. The utility of the method is demonstrated via a study of the spin-flip Raman spectrum of double quantum wells. Finally it is shown how spin-flip Raman spectroscopy could be used, in conjunction with δ doping of donors, as a probe of the possible enhancements to the paramagnetism of a dilute magnetic semiconductor at a heterojunction with a nonmagnetic material.

We present an experimental and theoretical study of the dephasing of heavy-hole excitons in a 70-Å GaAs/${\mathrm{Al}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$As multiple-quantum-well stucture. We measure the time-integrated three-pulse four-wave-mixing signal of heavy-hole excitons in the quantum-well sample. We model the third-order nonlinear optical response incorporating the effects of a stochastic modulation of the transition frequency. We find that the dephasing of the nonlinear polarization is mainly the result of a spectral-migration process of the excitons and, to a lesser extent, due to a homogeneous dephasing process. The time dependence of the four-wave-mixing signal is very well described by theory if the spectral-migration process is assumed to be energy dependent.

We calculate the persistent current of one-dimensional rings of spinless fermions with short-range interactions on a lattice with up to 20 sites, and in the presence of disorder, for various band fillings. We find that both disorder and interactions always decrease the persistent current by localizing the electrons. Away from half-filling, the interaction has a much stronger influence in the presence of disorder than in the pure case.

We present a theory of resonant Raman scattering by acoustic phonons in semiconductor quantum wells and superlattices for the case when translational invariance along as well as perpendicular to the growth direction is perturbed due to layer thickness fluctuations and interface roughness, both with and without an applied magnetic field. The breakdown of crystal momentum conservation along the growth direction (${\mathit{q}}_{\mathit{z}}$) leads to scattering from individual layers and folded acoustic phonons from the whole mini-Brillouin zone contribute to a continuous emission background of geminate recombination. The Raman intensity is evaluated from the phonon displacement field. It is proportional to the difference in the amplitudes of pairs of counterpropagating waves constituting each mode. This dependence causes characteristic intensity variations (peaks and dips) superimposed on the emission continuum at energies of gaps of the folded phonon dispersion. Interface roughness also allows for elastic scattering of photoexcited intermediate electronic states in the Raman process into states with nonzero crystal momentum (${\mathit{q}}_{\mathrm{\parallel }}$) perpendicular to the growth direction. This momentum can be compensated by acoustic phonons in higher-order Raman processes. Due to this effect internal gaps of the phonon dispersion, e.g., at anticrossings of longitudinal and transverse acoustic branches for ${\mathit{q}}_{\mathrm{\parallel }}$≠0, acquire enough strength to appear in the spectrum as additional structure.

We study the transport properties of one-dimensional (1D) interacting fermions through a barrier by numerically calculating the Kohn charge stiffness constant and the relative Drude weight. We find that the transport properties of the 1D Hubbard model are quite different from those of the 1D spinless fermion model. For example, the presence of the attractive interaction between electrons in the 1D Hubbard model actually suppresses the dc conductance, while a small repulsive interaction enhances the dc conductance. These results show that the spin degree of freedom plays an important role in the transport properties of the 1D interacting fermion systems.

We have developed a variational multicomponent hypernetted-chain theory for calculation of the elementary collective excitations in semiconductor superlattices. The infinite lattices are mapped onto multicomponent systems with a Bloch-sum type of transformation. The finite multilayer systems can also be treated as a straightforward extension of the multicomponent fluid theory. Numerical results are presented for both finite and infinite type-I lattices. Some simple analytic results that go beyond the random-phase approximation and self-consistent-field approximation are also presented for type-II and polytype systems. Numerical results for pressure and bulk modulus in electron-electron and electron-hole two-layer systems primarily indicate the region of stability of these systems with respect to electron density and interlayer separation.

We have used scanning tunneling microscopy (STM) to investigate (2×3)N-induced structures arising from the bombardment of activated nitrogen onto the Cu{110} surface followed by annealing. We have examined these structures for a variety of nitrogen doses, beam energies, and annealing temperatures. Increasing the nitrogen dose resulted in the gradual appearance of an additional N-induced structure also possessing a (2×3) periodicity. Studies were performed on various desorption mechanisms (thermal and tip induced). Images of the surface after heating to temperatures at which partial desorption of the (2×3)N reconstruction occurred show that thermal desorption results in islands of clean Cu centered on step edges. Pulsing the voltage between the sample and tunneling tip was also found to induce desorption. We have formed codomains of the (2×1)O and (2×3)N reconstructions. Images of the codomains yield the registry of features within the (2×3)N reconstructions with respect to the substrate and suggest a double added row mechanism for the formation of the (2×3)N reconstruction. Models are presented for both the (2×3)N-induced structures.

From a time-dependent calculation and by assuming an exponential decay law for the resonances formed, we have computed lifetimes of selective-adsorption resonances (SAR’s) in the elastic scattering of He atoms from the Cu(110) and Cu(117) surfaces. In previous works, where SAR’s were studied within the time-independent close-coupling method, we found that in addition to the energy width obtained by analyzing plots of diffraction intensities versus the energy of the incident beam, another internal energy width (with a very different numerical value) can be defined. Here we show that it is this internal width that can be directly related to the lifetime of the resonance. Also, we have performed a simulation of the half collision or ‘‘selective-desorption’’ process. This approach is shown to yield accurate values for lifetimes of SAR’s.

We discuss the role of stress in destabilizing flat surfaces of thin solid films. The presence of this instability in stressed ${}_{}{}^4{}_{}{}^{}\mathrm{He}$ and several heteroepitaxial films is widely acknowledged and well documented. We pay special attention to the study of possible in-plane morphologies of the islands which appear as a result of the stress-driven destabilization of flat traction-free interfaces. We also discuss the critical thickness at which these morphologies change.

Surface-enhanced second-harmonic diffraction from a corrugated silver surface is studied as a function of the Fourier decomposition of the surface profile. Numerical results are obtained using the reduced Rayleigh equations for the linear and second-harmonic fields. Scattering of the surface-plasmon-polariton (SPP) -enhanced, evanescent nonlinear polarization wave into radiative channels is shown to provide a sensitive probe of spatial-harmonic content. The presence of a specific higher harmonic in the surface profile allows preferential scattering of the enhanced nonlinear polarization into a certain diffraction order where a higher-order scattering mechanism might otherwise be operative. The propagating orders can thereby be selectively enhanced, in some cases, by many orders of magnitude. Calculations are presented for symmetric profiles and a range of grating periods. The nature of these selective enhancements suggests that optimized profiles for second-harmonic diffraction into a particular order can be formed by a superposition of two appropriately selected Fourier components. To explore this possibility, gratings with groove densities of 1200 and 1290 grooves/mm were studied by first determining the enhancement of each of the propagating orders at their respective optimum groove depths, assuming a purely sinusoidal profile. A search for the maximum enhancement of each order was then performed by varying the amplitudes of the grating fundamental and relevant order-enhancing higher harmonic. For the two periods considered, optimized profiles were found. The degree of coupling to the SPP at the second-harmonic frequency is shown to be important in determining the optimized profile as demonstrated by the substantially different enhancing properties of these similar groove-density gratings.

Using high-resolution low-energy electron diffraction, we have observed nonconventional dynamic scaling during the molecular-beam-epitaxy growth of Pb/Pb(110). The growth front becomes increasingly rough as more Pb atoms are deposited. At a low growth rate, the interface width w changes with deposition time t in a scaling form w∼${\mathit{t}}^{\mathrm{\beta }}$ with β=0.77±0.05. The other scaling hypothesis involving the lateral correlation length ξ∼${\mathit{t}}^{\mathrm{\beta }/\mathrm{\alpha }}$ is no longer valid and the local roughness increases dramatically. However, the short-range height-height correlation function H(r) still scales with r, the horizontal distance between two points in the surface, in the form of H(r,t)∼f(t)${\mathit{r}}^{2\mathrm{\alpha }}$ with α=1.33±0.05. At the later stage of growth, the rough surface eventually induces an anistropic {111} faceting. Faceting is found to occur earlier when the deposition rate is high.

It has been shown previously that antimony evaporated onto a Au(111) surface grows in a layer-by-layer mode. The present work compares the initial, fast-oxidation stage of these films with that on bulk polycrystalline antimony. Measurements were carried out in an ultrahigh-vacuum environment using Auger electron spectroscopy, low-energy electron diffraction, electron-energy-loss spectroscopy, and work-function change measurements. At 300 K and for oxygen pressures between 1×${10}^{\mathrm{-}5}$ and 2×${10}^{\mathrm{-}3}$ Torr, oxidation follows first-order Langmuir kinetics, with an initial sticking probability of about ${10}^{\mathrm{-}5}$. Oxidation essentially saturates by ${10}^7$ L (1 L=${10}^{\mathrm{-}6}$ Torr s). A layer of oxide about 8.5 Å thick is formed in this initial stage, both for Sb films and for bulk antimony. An initial antimony film on Au(111) of about 1.7 monolayers, or 1.6×${10}^{15}$ atoms/${\mathrm{cm}}^2$, is required to complete this oxide layer, for saturation oxidation. The chemical composition of the oxidized films and the oxidized surface of bulk antimony is ${\mathrm{Sb}}_2$${\mathrm{O}}_3$. These results are compared with earlier work on Pb, Bi, and Sn.

We describe an efficient Monte Carlo simulation of profile decay via surface diffusion on a (1+1)D square-lattice system with suppressed roughening. Our algorithm takes advantage of analytical results for a one-dimensional random walker and is valid in the low-temperature limit of noninteracting activated walkers. We find that an initially sinusoidal profile decays nonexponentially with a characteristic decay time which increases with the initial amplitude. At a fixed ratio of initial amplitude to wavelength, the decay time τ increases with wavelength λ as τ∼${\lambda }^{3.5\pm 0.1}$. In this model, profile decay arises from irreversible decay of the peak and valley terraces rather than from step-step interactions.

Low-energy-electron-diffraction analyses of Fe-, Mn-, and Rh-stabilized Au{001}1×1 surfaces find the atomic structure to be a relaxed bulk termination. The first and the second interlayer spacings are contracted by about (2±2.8)% and (2.8±2.8)%, respectively, of the bulk spacing along 〈001〉. Photoemission experiments with synchrotron radiation on reconstructed Au{001}5×20 find a surface-resonance band above the bulk d bands which exhibits dispersion with photon energies between 14 and 24 eV, in accordance with the buckled character of the hexagonal overlayer responsible for the surface reconstruction. All remaining features in the photoemission spectra from Au{001}5×20 can be explained on the basis of the self-consistent relativistic calculation of Eckardt, Fritsche, and Noffke to within 0.4 eV. Photoemission from Rh-stabilized Au{001}1×1 is similar to that from the reconstructed surface except for the surface resonance, which is markedly reduced on the 1×1 surface, and a shift of one band which points toward a transfer of s-like charge from the Rh to the Au atoms in the surface region.

Elastic scattering and excitation of the intermolecular and intramolecular vibrational modes by electron impact on solid films of ${\mathrm{N}}_2$ is reported for various incident energies (1–19 eV). The scattered electron intensity as a function of the incident energy for the excitation of the vibrational modes exhibits several maxima that can be ascribed to various electron shape resonances found in the gas phase or to variations in the magnitude of the multiple electron scattering on phonons (i.e., intermolecular vibrational modes) which may reflect the conduction-band density of states of the solid. In contrast, the intensity of the elastic peak exhibits an energy dependence that cannot be correlated with any electron resonances. However, the lowest electron shape resonance of ${\mathrm{N}}_2$, which has the ${}_{}{}^2{}_{}{}^{}\mathrm{\Pi }_{\mathit{g}}^{}{}_{}{}^{}$ symmetry, is observed to decay into multiphonon excitations of about 30 meV within the incident electron energy range 1–3 eV. The presence of the multiphonon excitations in conjunction with the suppression of resonant elastic scattering via the ${}_{}{}^2{}_{}{}^{}\mathrm{\Pi }_{\mathit{g}}^{}{}_{}{}^{}$ state of ${\mathrm{N}}_2^{\mathrm{-}}$, which has been initially proposed by Fano et al., is discussed in terms of a semiclassical model developed recently by Mills.

We present a study of the electronic and CO chemisorption properties of ultrathin Pd films on Au(111). Using ultraviolet photoemission spectroscopy (UPS) we examined the valence-band electronic structure of Pd films for different thicknesses. The UPS spectrum of a Pd monolayer on Au(111) shows a narrow Pd 4d density of states at 1.6 eV below the Fermi level and a reduced density of states near the Fermi level. The valence-band spectra of a Pd monolayer annealed to increasing temperatures was also examined. As a result of Pd-Au interdiffusion, the Pd density of states is seen to sharpen and shift closer to the Fermi level. In the valence spectra of a Pd monolayer annealed above 300 K we identified a feature near 3.5 eV which we attribute to a Pd-Au hybrid state. Using UPS, temperature programmed desorption (TPD), and high-resolution electron-energy-loss spectroscopy (HREELS), we also examined the CO chemisorption properties of ultrathin Pd films on Au(111). From CO TPD experiments we estimate that the activation energy for CO desorption from a Pd monolayer on Au(111) is 32 kcal/mol. HREELS spectra of CO adsorbed on a Pd monolayer on Au(111) show that CO occupies threefold bridge sites at low CO coverage and both the twofold bridge and atop sites at saturation CO coverage. From these results we conclude that the CO chemisorption properties of a Pd monolayer on Au(111) are very similar to those of Pd(111) surfaces, which is striking in view of the large differences between the valence-band electronic structure of these two surfaces.

The theoretical formulas for universal conductance fluctuations (UCF’s) are numerically evaluated in quasi-two-dimensional systems. The dependence of the UCF on the two characteristic lengths, the thermal length ${\mathit{L}}_{\mathit{T}}$ and the inelastic length ${\mathit{L}}_{\mathit{i}}$, is particularly emphasized. While the impurity-averaged UCF is essentially proportional to ${\mathit{L}}_{\mathit{T}}^2$ with a logarithmic correction, the single-impurity effect is proportional to [${\mathit{L}}_{\mathit{T}}$${\mathit{L}}_{\mathit{i}}$${]}^2$. The magnetic-field dependence is calculated for the UCF and the single-impurity effect. The half width of the magnetic correlation function is determined in both cases as a function of ${\mathit{r}}_{\mathit{i}}$=[${\mathit{L}}_{\mathit{T}}$/${\mathit{L}}_{\mathit{i}}$${]}^2$. The half-width field depends on ${\mathit{r}}_{\mathit{i}}^{\mathit{p}}$ where p is about 0.5 for the UCF and 1 for the single-impurity effect. That means the magnetic area corresponding to ${\mathit{B}}_{1/2}$ is essentially ${\mathit{L}}_{\mathit{T}}$${\mathit{L}}_{\mathit{i}}$ for the UCF and ${\mathit{L}}_{\mathit{i}}^2$ for the single-impurity effect.

We introduce a theoretical analysis of the temporal and spatial evolution of the surface topography of solids following interference between incident and scattered pulsed laser beams. The essential role played by the nonlinear delayed feedback in the laser-radiation–surface system is considered. We show that it finally determines the surface topography evolution from pulse to pulse. In order to complete the analysis, numerical calculations have been conducted under the hypothesis of strong attenuation of laser radiation into the sample and of a limited heat diffusion during the action of a laser pulse. We predict an evolution from very simple to complex (chaotic) structures under multiple-pulse-laser irradiation of solid surfaces. This evolution is determined by some key irradiation parameters; initial surface microrelief, incident laser intensity, and the number of applied laser pulses. Experiments were performed in order to check the main predictions of the theoretical analysis. The system of transversal excited atmospheric pressure–${\mathrm{CO}}_2$ laser radiation (λ=10.6 μm)—interacting with fused silica was chosen as appropriate for performing test experiments. Optical microscopy studies of laser-treated zones evidenced special modifications of the surface topography in good accordance with the conclusions following from the theoretical analysis. The theoretical analysis is also in good agreement with some available data from the literature, at the same time providing a coherent interpretation of previously unexplained behaviors.