Hydrogenated amorphous silicon (a-Si:H) films have been thermally annealed at temperatures in the range 220–270 °C for 24–48 h either under intense visible light illumination (4–16 W/${\mathrm{cm}}^2$) or in the dark. After each annealing, the hydrogen-concentration profile was measured with Rutherford-backscattering-spectrometry and elastic-recoil-detection-analysis ion-beam analysis methods. A model is proposed which shows that, in good agreement with our results, the hydrogen-diffusion constant ${\mathit{D}}_{\mathrm{H}}$ is proportional to the power of illumination and also proportional to the loosely bonded hydrogen concentration. Other consequences of the model are discussed.

We study, both theoretically and experimentally, the variations of the absorption coefficient of bulk GaAs in the presence of a nonthermalized electron-hole plasma. We investigate the changes of the Coulomb enhancement factor (CEF) due to nonthermal carrier distributions. For comparison with experimental data, time-dependent nonlinear absorption spectra are calculated using carrier distributions obtained from the numerical solution of Boltzmann’s equations. Experimental absorption spectra are derived from differential transmission and reflectivity measurements of GaAs thin films in the subpicosecond regime. Both experimental and theoretical spectra show a redshift of the transient-hole-burning peak that can be attributed to the spectral dependence of the CEF changes. An induced absorption above the spectral hole is predicted for narrow distributions. This effect, which critically depends on the width of the carrier distributions, is strongly suppressed by fast collisional broadening in accordance with the experimental results.

The results of electrical resistivity and Hall measurements of n-type GaAs under uniaxial stress in the [100] and [111] directions performed at low temperature are presented. After the transformation of the EL2 defect into its metastable configuration, a stress-induced decrease of both electron Hall concentration and mobility was observed. This effect was related to the capture of electrons by the acceptor state of the metastable EL2 ([EL$2^{\mathrm{*}}$${]}^{\mathrm{-}/0\mathrm{}}$). It was found that this level split under [111] stress into two orientationally degenerate components: the triplet and the singlet. The dependence of Hall concentration and mobility upon [111] stress obtained after the EL2 photoquenching performed under a high stress was different from that measured after EL2 photoquenching performed without an external stress. This difference was due to ordering of metastable EL2 centers after photoquenching performed under a high stress and it is consistent with the isolated ${\mathrm{As}}_{\mathrm{Ga}}$ model of the EL2 defect. The stress coefficients for [EL$2^{\mathrm{*}}$${]}^{\mathrm{-}/0\mathrm{}}$ sublevels split under [111] stress are found to be ${\mathit{dE}}_{\mathit{T}}$/dχ[111]=-17 meV/GPa and ${\mathit{dE}}_{\mathit{S}}$/dχ[111]=-41 meV/GPa for triplet and singlet, respectively. The investigated level did not split under [100] stress.

We report results for photoluminescence studies of an impurity-related defect system in 450 °C annealed Czochralski-grown silicon resulting in optical emission with a wavelength near 1.7 μm. The photoluminescence emission is found to have a significant intensity that persists to room temperature. The external quantum efficiency at 300 K is measured to be approximately 2.5×${10}^{\mathrm{-}5}$.

The first-neighbor distance r, its deviation Δσ, and bond-angle deviation Δθ of a-Si and a-Si:H films are determined by extended x-ray-absorption fine-structure and Raman-scattering methods. The r of these materials is longer than that of crystalline Si. This distance increase results from the Δσ increment via anharmonicity of the two-body correlation. The values of r and Δσ for a-Si:H are smaller than those for a-Si. These reductions of the structural parameters are caused by a hydrogen bonding effect on network relaxation. The Δσ is proportional to Δθ, which implies equal division of storage energy between deviations of bond length and bond angle. The relationship between Δσ and Δθ is in good agreement with calculated network models.

Single crystals of ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Cd}}_{\mathit{x}}$${\mathrm{Ga}}_2$${\mathrm{Se}}_4$ and ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Cd}}_{\mathit{x}}$${\mathrm{Ga}}_2$${\mathrm{Se}}_4$:${\mathrm{Co}}^{2+}$ were grown by a chemical-transport reaction method using the ingots synthesized with 3 at. % excess of selenium as starting materials. It has been found that these single crystals have a defect chalcopyrite structure, and the tetragonal distortion factor (2-c/a) decreases due to the introduction of an excess of selenium. It has also been shown that these compounds have a direct optical-energy gap, and the direct optical-energy gap decreases when a composition X increases and when a cobalt impurity is introduced. The temperature dependence of the optical-energy gap is well given by the Varshni equation. Impurity optical absorption of the ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Cd}}_{\mathit{x}}$${\mathrm{Ga}}_2$${\mathrm{Se}}_4$:${\mathrm{Co}}^{2+}$ single crystals has been investigated in the temperatures 289 and 80 K and in the wavelength region 250–2500 nm. It is concluded that a cobalt atom in the ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Cd}}_{\mathit{x}}$${\mathrm{Ga}}_2$${\mathrm{Se}}_4$:${\mathrm{Co}}^{2+}$ single crystals with a small tetragonal distortion occupies nearly a cubic ${\mathit{T}}_{\mathit{d}}$ symmetry site of the host lattice as a ${\mathrm{Co}}^{2+}$ ion; then the observed impurity optical-absorption peaks are attributed to the electronic transition between the energy levels of a ${\mathrm{Co}}^{2+}$ ion in a crystal field of a cubic ${\mathit{T}}_{\mathit{d}}$ symmetry.

Exchange interactions in the diluted magnetic semiconductors ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Co}}_{\mathit{x}}$S, ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Se, and ${\mathrm{Cd}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Se were studied by electronic paramagnetic resonance in the far-infrared region in a high magnetic field up to 40 T. For x≊5%, transmission spectra exhibit very sharp satellite lines on both sides of the single magnetic-ion resonance. These features are attributed to the exchange of coupled spin pairs. A calculation of the energy levels of the pairs, including Dzyaloshinskii-Moriya (DM) effects and symmetrical exchange terms of the interaction tensor, allows us to extract from the measurements the values of the anisotropic exchange constants: ${\mathit{scrD}}_{\mathit{e}}$=0.014 K and ${\mathit{D}}_{\mathrm{DM}}$=0.18 K in ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Se, ${\mathit{scrD}}_{\mathit{e}}$=0.009 K and ${\mathit{D}}_{\mathrm{DM}}$=0.12 K in ${\mathrm{Cd}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Mn}}_{\mathit{x}}$Se, and ${\mathit{scrD}}_{\mathit{e}}$=0.06 K in ${\mathrm{Zn}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Co}}_{\mathit{x}}$S. The values of the DM constants compare well with predictions by Larson and Ehrenreich. Moreover, in the case of cobalt, the occurrence of an absorption peak at half-field is interpreted by taking the symmetric anisotropic terms into account.

We have observed broad Raman scattering around 1500 ${\mathrm{cm}}^{\mathrm{-}1}$ in undoped, n-type (${\mathit{N}}_{\mathit{e}}$=${10}^{17}$ ${\mathrm{cm}}^{\mathrm{-}3}$) and p-type (${\mathit{N}}_{\mathit{h}}$=${10}^{17}$ ${\mathrm{cm}}^{\mathrm{-}3}$) germanium crystals at room temperature when stimulated by the ∼1.5-eV laser light. The integrated intensity of the Raman scattering has a square-law power dependence at high excitation densities above 0.5 kW/${\mathrm{cm}}^2$. This power dependence suggests that the origin of the scattering should be related to some photoexcited product. Since the power dependence becomes linear below 0.5 kW/${\mathrm{cm}}^2$ only in the p-type sample, the origin of the Raman scattering is attributed to holes in the valence band. The intensity of the Raman scattering is found to decrease rapidly at low temperatures. The peak position and the temperature dependence of the 1500-${\mathrm{cm}}^{\mathrm{-}1}$ component are successfully explained by a model based on k-conserving resonant inter-valence-band Raman scattering of photoexcited holes.

We have used hydrostatic pressure as a means for studying a resonant Raman mode observed at 47 ${\mathrm{cm}}^{\mathrm{-}1}$ in highly disordered, ion implanted, unannealed GaAs. The mode shifts weakly (-0.07±0.15 ${\mathrm{cm}}^{\mathrm{-}1}$/GPa), supporting an identification of this band-resonant vibration as stemming from the breathing mode of the gallium vacancies, which are expected to be in high concentration. We measure a pressure coefficient of the longitudinal-optic phonon in these (5.5 nm) nanocrystals of GaAs to be 3.6±0.1 ${\mathrm{cm}}^{\mathrm{-}1}$/GPa. The good agreement between our value and the pressure shift of this phonon in bulk GaAs implies that the bulk modulus is independent of size at least down to this size crystallite.

The thermalization and energy relaxation of highly photoexcited carriers in p-type and intrinsic GaAs is investigated at room temperature using an ensemble Monte Carlo approach. We present an alternative model for treating carrier-carrier scattering in Monte Carlo simulations. The model takes into account the wave vector and frequency dependence of the dielectric function by using a screened Coulomb potential in q space due to a charge moving at the velocity of the center of mass of the interacting particles. In comparison to the statically screened model, the calculated effective carrier temperatures for an excitation density of 2×${10}^{18}$ ${\mathrm{cm}}^{\mathrm{-}3}$ are in good agreement with the experimental results for both p-type and intrinsic GaAs. At the lower doping and excitation density of 2.5×${10}^{16}$ ${\mathrm{cm}}^{\mathrm{-}3}$, the dynamic and the static screening models result in a comparable time evolution of the carrier energies and band populations.

We demonstrate the feasibility of ab initio studies of piezoelectricity within an all-electron scheme. The focus of our analysis is on wurtzite ZnO; for comparison, some results are also presented for the related materials BeO and ZnS. The comparative study is performed in order to understand the microscopic origin of the peculiar behavior of ZnO, whose piezoelectric response is the strongest among the tetrahedrally bonded semiconductors. In all such materials, the piezeoelectric effect results from two different terms of opposite sign: these are usually referred to as the ‘‘clamped-ion’’ and the ‘‘internal-strain’’ contributions. Cancellation among them is least effective in ZnO, where the dominant effect is due to a rigid-ion-like mechanism. Furthermore, we compute the spontaneous polarization of ZnO and we discuss the puzzling agreement between our calculated value and a very indirect experimental estimate of the same quantity.

A systematic study of the vacuum uv reflectivity (4–10 eV) for an electric field perpendicular to the c-axis is presented for SiC polytypes. Experimental results for the 4H, 15R, 6H, and 3C SiC polytypes are compared to theoretical calculations within the local-density approximation in conjunction with a muffin-tin-orbital basis set for the 2H, 4H, 6H, and 3C polytypes. Agreement to ∼0.2 eV is obtained between theoretical and experimental peak positions assuming a single, k-point-, energy-, and polytype-independent gap correction, the value of which agrees well with recent calculations based on Hedin’s GW approximation. The main features and their trends with degree of hexagonality of the polytype are interpreted in terms of interband transitions involving extended regions in the Brillouin zone where valence and conduction bands are nearly parallel.

A method for calculating the total energy of Si systems, which is based on the effective-medium-theory concept of a reference system, is presented. Instead of calculating the energy of an atom in the system of interest, a reference system is introduced where the local surroundings are similar. The energy of the reference system can be calculated self-consistently once and for all while the energy difference to the reference system can be obtained approximately. We propose to calculate it using the tight-binding linear-muffin-tin-orbital scheme with the atomic-sphere approximation (ASA) for the potential, and by using the ASA with charge-conserving spheres we are able to treat open systems without introducing empty spheres. All steps in the calculational method are ab initio in the sense that all quantities entering are calculated from first principles without any fitting to experiment. A complete and detailed description of the method is given together with test calculations of the energies of phonons, elastic constants, different structures, surfaces, and surface reconstructions. We compare the results to calculations using an empirical tight-binding scheme.

Full-potential linear muffin-tin orbital calculations are reported for the bulk properties, density of states, and electron density in A2 Mo, A15 ${\mathrm{Mo}}_3$Si, D$8_{\mathit{m}}$ ${\mathrm{Mo}}_5$${\mathrm{Si}}_3$, C${11}_{\mathit{b}}$ ${\mathrm{MoSi}}_2$, and A4 Si. Both the magnitude and the directional anisotropy of the valence charge density associated with bonds in the three silicides suggest a hierarchy in bond strength, ordered from strongest to weakest: Mo-Mo, Mo-Si, and Si-Si. A combined Mulliken and tight-binding analysis shows this behavior to coincide with variations in occupation of the respective types of bonding states, which is ultimately traced to Mo(4d) site energies that are nearly degenerate with the Fermi level, in contrast to Si(3p) site energies lying about 5 eV higher. These characteristics point to the dominance of Mo-Si bonding in C${11}_{\mathit{b}}$ ${\mathrm{MoSi}}_2$, consistent with the near pd hybridization gap in this material, which appears to explain the unique position of the disilicide in systematics of the molybdenum silicide bulk properties. Finally, having identified the critical elements of the electronic structure in the molybdenum silicides, this work provides a physically motivated basis upon which to improve models used in the study of large-scale Mo-Si interface problems.

Ab initio total-energy studies are used to determine the lattice constants and the atomic positions within the unit cells for 3C-, 6H-, 4H-, and 2H-SiC. The electronic structures are calculated for the atomic geometries obtained theoretically within the density-functional theory (DFT) and the local-density approximation (LDA). We state more precisely the ordering of the conduction-band minima and derive effective masses. By adding quasiparticle corrections to the DFT-LDA band structures we find indirect fundamental energy gaps in agreement with the experiment. A physical explanation of the empirical Choyke-Hamilton-Patrick relation is given. Band discontinuities, bandwidths, crystal-field splittings, and ionic gaps are discussed versus hexagonality.

The decay of the photoluminescence of the conjugated polymers, poly(phenylenevinylene), poly(phenylphenylenevinylene) (PPV), and PPPV blended with polycarbonate is measured as a function of the position of the spectral detection window employing frequency-up-conversion and streak-camera techniques. Upon probing at the high-energy side of the inhomogeneously broadened ${\mathit{S}}_1$←${\mathit{S}}_0$ 0-0 transition, an initial falloff characterized by a (1/e) decay time of about 300 fs is observed. This decay slows down by three orders of magnitude upon shifting the detection window towards lower energies. We present a quantitative analysis of these experimental data by performing simulations of dipole-dipole excitation transfer on a random lattice chain with energetic site disorder. They give evidence for the notion that photon absorption generates neutral excitation undergoing a random walk among segments of the polymer chain with different excitation energies.

Quasiparticle band structures of six II-VI compounds (ZnS, ZnSe, ZnTe, CdS, CdSe, and CdTe in the zinc-blende and wurtzite structures) are calculated using the GW approximation. Results for band structures, calculated using the ab initio pseudopotential method within the local-density approximation (LDA), are given along with the energies of the quasiparticle excitations at symmetry points. Comparisons are made with measured values and trends in the GW corrections to the LDA band structures are examined.

We obtain the field-theory representations of several network models that are relevant to two-dimensional transport in high magnetic fields. Among them, the simplest one, which is relevant to the plateau transition in the quantum Hall effect, is equivalent to a particular representation of an antiferromagnetic SU(2N) (N→0) spin chain. Since the latter can be mapped onto a θ≠0, U(2N)/U(N)×U(N) σ model, and since recent numerical analyses of the corresponding network give a delocalization transition with ν≊2.3, we conclude that the same exponent is applicable to the σ model.

Extending the Hedin-Lundqvist approximation, originally developed for the three-dimensional electron gas, we derive the correlation energy ${\mathit{E}}_{\mathit{c}}$ for the two-dimensional electron gas as a function of electron density ρ. A simple two-parameter function ${\mathit{E}}_{\mathit{c}}$(ρ) reproduces very well the first-principles numerical results in the whole density range. We also present a simple and accurate form of the correlation part of the chemical potential, which can be used, e.g., as a correlation part of the effective potential in the local-density approximation for two-dimensional electron systems.

We have employed a semiempirical molecular-orbital method to simulate the adsorption of chlorine (Cl) atoms on GaAs(110) for three cases: monatomic Cl, molecular Cl, and 0.16-monolayer (ML) coverage of Cl. For the single Cl atom, the equilibrium bonding site has the Cl atom at the on-top site of the surface As atom. The ${\mathrm{Cl}}_2$ molecule on GaAs(110) displays dissociative adsorption with both Cl-Ga and Cl-As bonding configurations. For the third case, Cl adsorption on GaAs(110) leads to enhanced relaxation beyond that of the already relaxed surface. In this geometry, the surface As atoms gain charge at the expense of the surface Ga atoms and are pushed further away from the surface plane while the Ga atoms relax inward. The microscopic structure of 0.16-ML coverage of Cl displays both Ga-Cl and As-Cl bonding configurations. For the Ga-Cl bonding configuration, the Cl atom is adsorbed at the surface Ga dangling-bond site while for As-Cl, the Cl atom is sited atop the surface As atom, offset slightly toward the dangling-bond direction. The Cl adsorbed at the Ga site is lower in height than Cl atop the surface As atom relative to the substrate and the Cl-Cl nearest-neighbor distance calculated is 3.60 Å, quite close to the value of 3.46 Å observed experimentally with the scanning tunneling microscope (STM). These results compare favorably with the findings of Stepniak, Rioux, and Weaver from recent photoelectron spectroscopy experiments, as well as the structural information obtained from STM.