The plane-wave pseudopotential (PWPP) method has been used to calculate structural and electronic properties of wurtzite and zinc-blende GaN. In contrast to previous studies using the PWPP method, the gallium 3d electrons were treated as valence electrons. This yields larger lattice constants and smaller energy gaps as compared with previous PWPP results. For wurtzite, the three structural parameters were found to be a=3.162 Å, c=5.142 Å, and u=0.377. For zinc blende, the cubic lattice constant was found to be ${\mathit{a}}_0$=4.460 Å. The lattice constants are about 1% smaller than measured values and the cubic lattice constant is within about 0.2% of results from full-potential all-electron calculations. This study demonstrates that it is both possible and practical to treat gallium 3d electrons explicitly using the PWPP method.

Electron scattering by a spatially correlated system of DX charges has been described using the formalsim of composition waves. The matrix element for the scattering rate is given through an I(Q) interference function (Q=scattering vector) containing pair correlation functions ε(p) defined at the p lattice vectors. The ε(p) are able to describe long-range as well as short-range order and they are simply related to short-range order correlation functions given in the literature. The method is developed for scattering centers having equal charges; the case of positively and negatively charged impurities present together is briefly discussed. A comparison between the two extreme cases of randomly distributed scattering centers and of centers arranged in a superlattice suggests, for intermediate cases, an I(Q) given by an array of Gaussian shaped functions with common dispersion σ, centered on the reciprocal nodes of a virtual superlattice. On this basis, experimental mobility data for Si-doped ${\mathrm{Al}}_{0.25}$${\mathrm{Ga}}_{0.75}$As samples prepared by molecular beam epitaxy have been analyzed and discussed. Data refer to isothermal electron capture transients into DX centers, as well as to steady-state measurements taken for different free electron densities under a persistent photoconductivity regime. It has been confirmed that the initial stage of the capture process takes place together with increasing order in the scattering center distribution (decreasing σ), whereas the contrary happens during the final stage (increasing σ).

Radioactive Li ions were implanted into high-resistivity p-type Si at temperatures between 40 and 530 K and their lattice sites determined by measuring the channeling and blocking effects of the emitted alpha particles. Implantations below room temperature lead to the occupation of the well-known tetrahedral interstitial sites, whereas for implantations above room temperature bond-center sites could be unambiguously identified. As cause of the corresponding lattice site changes, an interaction of diffusing Li with double vacancies is suggested.

The dielectric function of cubic and hexagonal CdSe has been measured by spectroscopic ellipsometry in the 2–25-eV range. The results are compared with similar data for CdS and other II-VI compounds and also with band-structure calculations. Relevant critical points are listed. Transitions originating in the Cd 4d levels are identified and an estimate of the corresponding exciton binding energies is made.

The electronic properties of native monovacancy defects were investigated using techniques based on positron annihilation in n-type Si-doped GaAs. The positron lifetime measurements as a function of temperature have shown the thermal ionization of native As vacancies from negative to neutral charge state: ${\mathit{V}}_{\mathrm{As}}^{\mathrm{-}}$→${\mathit{V}}_{\mathrm{As}}^0$. By combining the results of positron lifetime and two-dimensional angular-correlation-of-annihilation-radiation (2D-ACAR) experiments, the 2D-ACAR distributions of neutral and negative arsenic vacancies were separated from the partial bulk contribution. We found that 2D-ACAR distributions of the As vacancies are different from those of the bulk, and we also observed differences between the distributions of ${\mathit{V}}_{\mathrm{As}}^0$ and ${\mathit{V}}_{\mathrm{As}}^{\mathrm{-}}$. When compared to the negative charge state ${\mathit{V}}_{\mathrm{As}}^{\mathrm{-}}$, the neutral charge state ${\mathit{V}}_{\mathrm{As}}^0$ is characterized by a narrower momentum distribution from valence electron annihilations and a lower high-momentum tail from core electrons. These findings lead us to conclude that the electronic and ionic structure of the As vacancies depend strongly on their ionization. We also infer that ${\mathit{V}}_{\mathrm{As}}^{\mathrm{-}}$ has a smaller open volume than ${\mathit{V}}_{\mathrm{As}}^0$. This conclusion is in agreement with previous positron lifetime results and molecular-dynamics calculations.

An analytical expression of the Verdet constant of semimagnetic semiconductors is derived from a microscopic analysis of the transverse susceptibility responsible for the Faraday effect. This expression is obtained by integrating, in k space, the wave-vector-dependent transverse polarizability. The latter is obtained within the density-matrix formalism using a Lorentzian shape for the wave-vector dependence of the exchange interaction. Using only two parameters, this model reproduces very well the experimental Faraday-rotation spectra available for both ${\mathrm{Cd}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Te and ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Te for various manganese concentrations and sample temperatures. The normalization parameter is independent of the manganese concentration and sample temperature for a given type of semimagnetic semiconductor, and the same exchange parameter is used for all manganese concentrations at a given temperature. Using the experimental data obtained in ${\mathrm{Cd}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Te samples of precisely known manganese concentrations, we are able to extract the value q=0.022 at 77 K and q=0.018 at 290 K for the exchange interaction parameter with a relative precision of 5%, while for ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Te a smaller value (q=0.003 at room temperature) is obtained from previous studies.

The rate at which athermal, nearly monoenergetic free-carrier distributions initially scatter out of optically injected states in bulk GaAs is studied experimentally and theoretically. Pump-probe absorption bleaching measurements done with 2-eV 75-fs pulses are reported for several injected densities ranging from <1×${10}^{16}$ ${\mathrm{cm}}^{\mathrm{-}3}$ to >5×${10}^{18}$ ${\mathrm{cm}}^{\mathrm{-}3}$. By using a technique in which the amplitude of the peak bleaching is measured as a function of the injected carrier density at a fixed (∼30 fs) pump-probe delay, the density-dependent carrier-carrier scattering (CCS) component of the signal is revealed. Calculations of the CCS rate, which include the effects of dynamical screening within Gaussian distributions are also reported, and compared with the results obtained using the static-screening approximation. In both cases the CCS rate is found to depend weakly on the density, N; approximately as ${\mathit{N}}^{0.35\pm 0.05}$. A heuristic model is described that provides insight as to why CCS rates within athermal distributions are relatively insensitive to the density and dynamic screening effects.

Photoconduction spectra of GaAs samples at photon energies around the Ga and As K edges were measured. Extended x-ray-absorption fine-structure oscillation was observed in these spectra. It is found that the spectrum shape of the photocurrent depends on the sample thickness. A microscopic theory is presented to determine the photoconduction response to x-ray absorption. The theory is in good agreement with experiment.

We present the calculation of the full phonon spectrum for silicon and germanium with the pseudopotential method and the local-density approximation without using linear-response theory. The interplanar-force constants for three high-symmetry orientations [(100), (110), and (111)] are evaluated by supercell calculations using the Hellmann-Feynman theorem. By considering the symmetry of the crystal, three-dimensional interatomic-force-constant matrices are determined by a least-squares fit. Interactions up to the eighth nearest neighbors are included. The dynamical matrix, which is the Fourier transform of the force constant matrix, is hence constructed and diagonalized for any arbitrary wave vector in the Brillouin zone, yielding the phonon dispersion. In this paper we will present the calculation details and discuss various aspects of convergence. Phonon dispersions of Si and Ge calculated are in excellent agreement with experiments.

We find that Slater’s averaged Fock exchange potential, unlike the local-density approximation or Hartree-Fock approximation, results in Ge and GaAs energy gaps in good agreement with experiment. Its cohesive energies disagree with experiment by amounts that are larger than, and of opposite sign to, those of the local-density approximation.

The branch conductance as well as the conductance sum rule for the hierarchical edge channel currents of a fractional-quantum-Hall-effect state are derived analytically within the Haldane-Halperin hierarchy scheme. The comparison of our result and those of previous works is discussed. We provide also an intuitive interpretation for the heirarchical drift velocities of the edge excitations which are different for different edge channels.

Laser-induced suppression of electron tunneling in a biased asymmetric double-quantum-well structure is numerically studied based on the Markoff master equation. A pump pulse is used to prepare initially the electronic wave packet which is irradiated by a cw laser. It is shown that if the bias voltage is adjusted to align the upper two electronic states in the narrow well and the wide well, suppression can occur before dephasing is complete. An analytical expression for the time evolution of the wave packet is derived by a degenerate-two-level model to give a qualitative interpretation to the origin of the electron localization.

We present a technique for obtaining the current and charge densities arising from delocalized current-carrying states in two-dimensional quantum systems. The key to the method is the development of a computationally efficient transfer-matrix solution of the two-dimensional Schrödinger equation. Because of its efficiency, this method is ideal for use in self-consistent solutions of the coupled Schrödinger and Poisson equations in two-dimensional open systems where the states form a continuum. We demonstrate the utility of the method in calculating current and charge densities in an example system.

The 2×1 and c(4×2) surface reconstructions on Ge(100) are investigated by the ab initio, all-electron, molecular-cluster method, which solves the local-density-functional equations and provides analytical energy gradients. We use finite-size clusters (up to 71 atoms including 39 Ge atoms) to model the Ge(100) surfaces. Atomic-force calculations are extensively used to obtain the minimum-energy geometry for the different structures investigated. We determine and compare the binding energy and geometry up to the fourth layer of the symmetric (2×1), buckled (2×1), as well as the higher-order c(4×2) reconstruction. Important energetic and structural differences are found compared to the corresponding Si(100)2×1 surface. The asymmetric dimer model is found to be 0.34 eV/dimer lower than the symmetric one with the up-dimer atom being 0.19 Å above the plane of the unreconstructed surface and a dimer tilt of 15°. The buckled 2×1 and c(4×2) reconstructions are found to be close in energy, which suggests that both could be present on the surface at room temperature. These results are in excellent agreement with scanning-tunneling-microscopy experiments and previous theoretical studies using a slab geometry. This energy is well below the energy of the symmetric dimer reconstruction indicating that dimer flipping recently suggested for the Si(100)2×1 surface is unlikely to occur in the case of the Ge(100) surface. In significant contrast to the Si(100) surface, we found that the Ge-Ge dimer is weaker with bond lengths that are slightly above the bulk value of 2.44 Å, at 2.48 and 2.50 Å for the asymmetric 2×1 and c(4×2) reconstructions, respectively. It suggests that the Ge(100) surface might show some different behavior towards adsorption.

We present an extensive study of the current response of isolated mesoscopic rings (noninteracting electrons in the diffusive regime) to a small ac flux superimposed on a dc Aharonov-Bohm flux. This response can be very different from the conductance of the same ring connected to a voltage source. We emphasize the importance of the inelastic rate γ compared to the level spacing Δ and the driving frequency ω. We essentially focus on the discrete-spectrum limit γ≪Δ. The conductivity has an imaginary component which, at low frequency, is related to the flux derivative of the persistent current through the ring; as previously shown this quantity presents an ensemble average which is zero in the grand canonical case but which is finite in the canonical statistical ensemble. At frequencies larger than γ we show evidence of an extra contribution to the imaginary conductivity which is finite and temperature independent in the grand canonical ensemble, and increases with temperature in the canonical case. The real part of the average conductivity exhibits ${\mathrm{\Phi }}_0$/2-periodic flux oscillations which are also very sensitive to the statistical ensemble: for the canonical ensemble, these oscillations can assume either sign compared to the Altshuler, Aharonov, and Spivak oscillations observed in long cylinders and connected arrays of rings. We indeed identify different contributions giving rise to flux oscillations of opposite sign. An off-diagonal contribution related to interlevel transitions which is connected with the flux dependence of the level statistics and gives rise to a negative low-field magnetoconductance, and a diagonal contribution related to the flux dependence of the occupation of the different levels. This past quantity is proportional to the average square of the single-level persistent current and to the inelastic scattering time, and gives rise to a positive low-field magnetoconductance. It can be much larger than the Drude conductance. Our results rely on numerical simulations; most of them are also justified analytically.

We derive a formula that describes interdiffusion profiles of quantum wells and shows how the formula accurately models interdiffusion in quantum wells of lattice-matched ${\mathrm{In}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$${\mathrm{As}}_{\mathit{y}}$${\mathrm{P}}_{1\mathrm{-}\mathit{y}}$ and ${\mathrm{Al}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$As alloy semiconductors. Our formula includes the different interdiffusion coefficients between layers and interfacial discontinuity of interdiffused species. Our formula explains how quantum energy shifts due to interdiffusion vary with annealing time and annealing temperature in various wide well layers of both ${\mathrm{In}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$${\mathrm{As}}_{\mathit{y}}$${\mathrm{P}}_{1\mathrm{-}\mathit{y}}$/InP and GaAs/${\mathrm{Al}}_{\mathit{x}}$${\mathrm{Ga}}_{1\mathrm{-}\mathit{x}}$As quantum wells. We also show the quantitative difference between interdiffusion profiles of these two materials.

We investigate the influence of the dielectric mismatch between barrier material and well material on the properties of excitonic polaritons in quantum wells. A quantum theory of the interaction between quantum-well excitons and dielectric waveguide eigenmodes is presented and polariton dispersion, radiative decay rate, and polariton eigenstates are obtained. The modifications as compared to previous theories are mainly a photonlike dispersion of the upper polariton branch for energies above the excitonic resonance as well as a finite imaginary part of the exciton self-energy for all values of ${\mathit{Q}}_{\mathrm{\parallel }}$. This theory can easily be extended to a description of structures with a more complicated profile of the dielectric constant.

The electronic properties of shallow band-edge surface traps in nanometer-size CdSe quantum dots are probed using fluorescence-line-narrowing spectroscopy. We find large changes in electron-hole-pair radiative lifetimes and couplings to LO phonons as the temperature is changed from 1.75 to 10 K. We attribute these changes to the localization of the photogenerated hole at the surface of the dots, accompanied by thermally activated motion between these surface localized states. A simple model based on the observed exciton–LO-phonon couplings is constructed to estimate the extent of hole localization in the luminescing state. A size-dependent study (20–80 Å diameter) indicates that surface effects diminish rapidly with increasing size.

We have studied optical properties of one-dimensional oligothiophenes and thiophene-based oligomers. The thiophene-based oligomers have the well-barrier-well structures in quasi-one-dimensional backbone chains, where the well parts are the oligothiophenes with different conjugation lengths and the barrier parts are nonvisibly luminescent materials with larger energy gaps. The chain-length dependence of the optical properties of oligothiophenes shows that excitons are localized within two thiophene rings and the luminescence decay is controlled by the diffusion of localized excitons. Furthermore, luminescence properties of thiophene-based oligomers such as luminescence yield, luminescence lifetime, and luminescence wavelength can be controlled by changing the conjugation length of the well part and the barrier structure. Spectroscopic analysis and molecular-orbital calculations show that the exciton localization and radiative recombination processes are affected by the rotational motion of thiophene rings.

The creation of optical gain through excitonic mechanisms in quantum wires has been theoretically investigated by taking into consideration the longitudinal-optical-phonon-assisted exciton transitions. The gain values obtained are comparable to those obtained by the conventional electron-hole recombination process. Moreover, the excitonic gain is found to set in at a much lower electron-hole pair density than the gain due to the conventional process. These results strongly suggest the involvement of excitons in creating optical gain in actual quantum wires.