We present results of ab initio calculations for the electronic structure and the magnetic moments of 4d (Ru, Rh, Pd) and 5d transition-metal bilayers (Ir, Pt) on the Ag(001) substrate. The bilayers of Rh, Pd, Ir, and Pt show ferromagnetism, but compared to the monolayers the moments are strongly reduced to less than 0.35${\mathrm{\mu }}_{\mathit{B}}$. This indicates that the magnetism in the 4d and 5d metal films depends critically on the film thickness. These results put a severe upper limit on the experimental imperfections permitted to verify 4d and 5d magnetism.

The pressure dependence of the metal-to-insulator (MI) transition in ${\mathrm{BaVS}}_3$ has been studied up to applied hydrostatic pressures of 16 kbar. There is a substantial decrease in the transition temperature, with 14.8 kbar suppressing ${\mathit{T}}_{\mathrm{MI}}$ from 69 K at ambient pressure to about 12 K. This behavior is discussed on the basis of the crystal structure of this quasi-one-dimensional compound. We argue that the ratio of the lattice constants parallel and perpendicular to the vanadium chains is the key parameter for understanding the pressure dependence of the MI transition. Thermal-expansion and magnetization measurements at ambient pressure show no additional phase transition below 69 K. Our results indicate that the metal-to-insulator transition may be a continuous isomorphic transition from the insulator with low susceptibility due to a gap in a band of extended states to a metal with a Curie-Weiss susceptibility due to correlations in the vanadium d band.

We use the real-space linear-muffin-tin-orbital approach, within the atomic-sphere approximation to calculate the electronic structure around vacancies in Al and Cu and divacancies in Al. A defect consisting of a Cd-monovacancy pair in Al is also considered. The electric-field gradient at the nucleus and the asymmetry parameter η for sites in the near neighborhood of the vacancy in these various situations were also obtained. The calculated values are in good agreement with experiment when available. Our results confirm the failure of the point-charge model when applied to obtaining the electric-field gradient at the nucleus in metallic systems and show that the defect with quadrupole frequency ${\mathrm{\omega }}_0$=64 Mrad/sec, created by low-temperature implantation of In in Al, can indeed be assigned to a Cd-monovacancy pair.

We have measured the real part ε’ and the imaginary part ε’’ of the dielectric susceptibility of the quasi-one-dimensional conductor ${\mathrm{TaS}}_3$ in its charge-density-wave (CDW) state in the wide frequency range f (${10}^{\mathrm{-}2}$–${10}^7$ Hz) in the low-temperature range (4.2–52 K). The frequency dependence of ε’(ω,T) as well as ε’’(ω,T) show the lnf scaling and a crossover between a behavior corresponding to a Gaussian distribution of logarithms of relaxation time τ at T≥30 K and a frozen-in state with very wide τ distribution at T≤24 K. The temperature dependences of ε’ measured at low frequencies (f≤${10}^5$ Hz) show maxima whose positions shift to low temperatures and whose amplitudes grow with decreasing frequency. The temperature dependences of τ are characterized by two branches. With decreasing temperature the long-time branch of τ diverges (α relaxation) while the shorter-time branch shows a monotonous behavior, corresponding probably to the tunneling mechanism of CDW local motion (β relaxation). A qualitative model is suggested that describes the transition of CDW into the glass state with decreasing temperature and the main properties of this state.

Ultraviolet and x-ray-photoemission spectra (XPS) as well as x-ray inverse-photoemission [bremsstrahlung isochromat spectroscopy (BIS)] and x-ray-absorption spectra of valence-fluctuating ${\mathrm{Sm}}_3$${\mathrm{Se}}_4$ are discussed to clarify its electronic structure. The constant initial state and constant final state spectra in the Sm 4d→4f excitation region are quantitatively interpreted in terms of the 4${\mathit{d}}^{10}$4${\mathit{f}}^5$→4${\mathit{d}}^9$4${\mathit{f}}^6$ and 4${\mathit{d}}^{10}$4${\mathit{f}}^6$→4${\mathit{d}}^9$4${\mathit{f}}^7$ multiplet structures. The Sm 3d absorption spectra support the bulk valence fluctuation. The ${\mathit{d}}^9$4${\mathit{f}}^5$ and ${\mathit{d}}^9$4${\mathit{f}}^6$ multiplet structures are clearly resolved in the 4d XPS spectra. The effective hybridization of the Sm 4f states with the conduction-electron states is found to be large and much different for the 3d photoemission, 4d photoemission, and 4f BIS final states, resulting in the different intensity ratio of the ${\mathrm{Sm}}^{2+}$ and ${\mathrm{Sm}}^{3+}$ signals in these spectra. In addition, the surface core-level shift of the Sm 4${\mathit{f}}^5$ final state is evaluated as ∼0.5 eV.

The density of states for a finite or an infinite cluster of scatterers in the case of both electrons and photons can be represented in a general form as the sum over all Krein-Friedel contributions of individual scatterers and a contribution due to the presence of multiple scatterers. The latter is given by the sum over all periodic orbits between different scatterers. General three-dimensional multiple-scattering theory for electromagnetic waves in the presence of scatterers of arbitrary shape is presented. Vector structure constants are calculated and general rules for obtaining them from known scalar structure constants are given. The analogue of the Korringa-Kohn-Rostocker equations for photons is explicitly written down.

A method is described that enables the absorption and injection of wave functions at the boundaries of a region in a numerical solution to the time-dependent Schrödinger equation. A number of results corresponding to one- and two-dimensional simulations are presented. Such boundary conditions enable the use of time-dependent simulations of geometries connected to contacts that correspond to sources and sinks of particles in thermal equilibrium. The approach presented has a number of attractive features from a numerical-implementation point of view.

We present the results of local-density-approximation-based Car-Parrinello calculations of orientationally ordered phases of ${\mathrm{K}}_3$${\mathrm{C}}_{60}$ and ${\mathrm{K}}_6$${\mathrm{C}}_{60}$. Our investigation focuses on the following points: the effects of alkali-mental intercalation on the structural and electronic properties of ${\mathrm{C}}_{60}$, the chemical bonding, the inhomogeneity of the electron distribution, and the degree of freedom of the potassium atoms in the interstitial region. The results are discussed in view of experimental data, such as x-ray and neutron-diffraction data as well as nuclear magnetic resonance spectroscopy.

The energy band structure of ${\mathrm{YTiO}}_3$ was examined using the full-potential linear augmented-plane-wave method with the local-density approximation. Our calculations show that experimentally observed ferromagnetism and lattice distortion are necessary conditions for ${\mathrm{YTiO}}_3$ to have a band gap at the Fermi level, although the cubic perovskite structure of ${\mathrm{TYiO}}_3$ shows no signs of its being an insulator.

Quantum beats in picosecond time-resolved resonance fluorescence are investigated in ${\mathrm{Cu}}_2$O for the yellow 1S quadrupole exciton polariton. In order to obtain information on the coherence of the resonant intermediate polariton states, we have studied the electric-dipole allowed light-scattering processes involving one and two ${\mathrm{\Gamma }}_3^{\mathrm{-}}$ phonons. A quantitative analysis provides exciton-polariton coherence times of about 1 ns at 2 K, decreasing with higher temperatures and larger wave vectors due to phonon scattering. From the magnetic-field dependence, low-field g values are determined with high accuracy. The measurements permit a clear distinction between Raman and hot-luminescence-like contributions to the scattering intensity. The strong effect of the spectral width of the excitation pulse on the time behavior of the scattering, as well as the temperature dependence of the coherence time, indicate that it is essential to take into account the polariton character.

First-principles calculations have been performed to study the chemical bonding trends and the phase stability of 3d transition-metal (TM) aluminides with equiatomic composition. The physical mechanisms for ordering in both the L$1_0$ and B2-type aluminides are presented. It is shown that the dominant factor for early TM aluminides is the directional bonding between the d orbitals of TM atoms, whereas for late TM aluminides, charge transfer and hybridization between Al s and p states, and TM d states play more important roles in the bonding mechanism and the properties of lattice defects. Our calculations also show the existence of a ferromagnetic phase for MnAl in the L$1_0$ structure, which is stabilized by the formation of a magnetic moment on the Mn sites (with a value of 2.0 ${\mathrm{\mu }}_{\mathit{B}}$). The chemical trends in the structural stability and heats of formation of 3d TM aluminides are found to correlate well with the densities of states at the Fermi level.

We examine the mathematical and physical basis for choosing the coefficients of an Ising model Hamiltonian, variably known as multisite interactions or effective cluster interactions, used in the study of phase stability of substitutionally disordered alloys. We show that concentration-independent interactions can be defined if and only if the system is assumed to be strictly finite, i.e., to contain a finite number of sites. If the system is infinite, only concentration-dependent interactions can be defined. This strict dichotomy has apparently been missed in previous discussions, and this oversight has led both to erroneous conclusions regarding the relationships between the two types of interactions, and to confusion about methods for determining them. Explicitly, no relationship between the two sets of interactions can exist since these two different types of interactions cannot be defined simultaneously. We also discuss the physical basis for the mapping of alloy configurational energies onto the Hamiltonian of an Ising model, and point out that the use of the model involves the generally unjustified assumption that the kinetics properties of an alloy will allow configurational transformations to occur according to the dictates of the equilibrium thermodynamics of a spin system. In a final section, we present the implications of our results to the study of alloy phase stability and transformations by means of the Ising model.

A metastable selenium-related complex defect in silicon has recently been reported to give rise to two different deep photoluminescence (PL) spectra, originating from a bound-exciton (BE) recombination at two different configurations (referred to as A and B) of the same center. In this paper, we report on studies of this defect by optically detected magnetic resonance (ODMR). The lowest BE state, in the A configuration, is shown to be a spin-triplet state and the symmetry of the defect in this configuration is determined to be monoclinic I. In this symmetry one of the principal spin axes (y) is along the [11¯0] direction, while the other two principal spin axes (z and x) lie in the (11¯0) plane. The z axis is in this case is 22° away from the [111] axis towards the [001] axis, and the x axis is orthogonal to the y and z axes. The broad ODMR linewidth is attributed to an unresolved hyperfine interaction, most likely due to a copper atom in the center. This is further supported by the experimentally observed shift of the ODMR resonance positions, induced by an anticrossing of the magnetic sublevels. The conversion between the two configurations, either by above band-gap light (${\mathrm{Se}}_{\mathit{A}}$→${\mathrm{Se}}_{\mathit{B}}$) or by thermal annealing at temperatures >60 K (${\mathrm{Se}}_{\mathit{B}}$→${\mathrm{Se}}_{\mathit{A}}$), can be followed in PL and also in ODMR.

The photobleaching of light-induced metastable dangling bonds (DB’s) in Si-rich a-${\mathrm{Si}}_{1\mathrm{-}\mathit{x}}$${\mathrm{N}}_{\mathit{x}}$:H alloys is carried out using subgap illumination at both room temperature (RT) and liquid-nitrogen temperature (77 K). The photobleaching is mainly aimed at the light-induced DB’s in the two different processes: the fast process and the slow process. Combining the results of photobleaching with those for RT annealing of the light-induced DB’s in both processes created at 77 K, we find that there are prominent differences between the natures of light-induced DB’s in the two processes: (1) The light-induced DB’s in the fast process can be bleached by subgap illumination, while those in the slow process cannot; (2) the light intensity needed for generating the light-induced DB’s in the slow process is stronger than that for the light-induced DB’s in the fast process. A comparison of the present results with those for a-Si:H shows that the light-induced DB’s in the fast process are similar to the light-induced DB’s created by low-intensity light at low temperature in a-Si:H, and the light-induced DB’s in the slow process are similar to the light-induced DB’s created by high-intensity light at RT in a-Si:H. The results suggest that the light-induced DB’s in the fast process are associated with charge trapping and the light-induced DB’s in the slow process with the creation of new defects.

Amorphous hydrogenated carbon films show a characteristic photoluminescence behavior that is of great interest for the study of their electronic structure. In the present paper we develop a model of the photoluminescence process in a polymerlike hydrocarbon on the basis of a mixed ${\mathit{sp}}^{2\mathrm{-}}$${\mathit{sp}}^3$ hybridization of the carbon atoms in this material. The ${\mathit{sp}}^2$ phase is assumed to be confined in clusters embedded within a ${\mathit{sp}}^3$ matrix. We assume excitation and recombination to take place within a single cluster. Thus the overall photoluminescence signal consists of the contribution of all single clusters, having different energy-gap values ${\mathit{E}}_{\mathit{g}}$, due to their various sizes and/or shapes. Applying the model to experimental data, we deduce a material-characteristic function f(${\mathit{E}}_{\mathit{g}}$) that turns out to be representative of the distribution of the different energy-gap values ${\mathit{E}}_{\mathit{g}}$ of the ${\mathit{sp}}^2$ clusters in the film. The good agreement with experimental data suggests that the classical Tauc gap ${\mathit{E}}_{\mathrm{Tauc}}$ in a-C:H films is a value averaging over all single-cluster gap values ${\mathit{E}}_{\mathit{g}}$. Moreover, our model reproduces all the peculiar features of the photoluminescence behavior of a-C:H films, such as the nonactivated dependence upon temperature.