We present ab initio calculations for the Coulomb U parameter of all transition-metal impurities in Rb. The U parameter increases in each row by filling up the d shell and decreases with increasing row number. In addition we find a strong configuration dependence, with the Hubbard U of 3d impurities being about 30% larger for the divalent configuration than for the monovalent one.

Crystalline electric field (CEF) parameters and magnetic anisotropic energies for rare-earth atoms R in ${\mathit{R}}_2$${\mathrm{Fe}}_{17}$${\mathit{L}}_{\mathrm{\delta }}$ compounds are calculated based on first-principles electronic-structure calculations. The parameter ${\mathit{A}}_{20}$ of CEF is obtained by using the real charge distribution from the corresponding cluster-model calculation and by directly calculating the crystal CEF parameters from the expansion coefficients of the local potential in the embedded-cluster model. The magnetic anisotropic energy is studied by taking the spin-orbit coupling as a perturbation into account in the semirelativistic embedded-cluster approach. The obtained results are in agreement with the experiments.

We discuss the results of a study of the two-impurity Anderson model carried out by using a perturbative expansion in U, which is known to provide a power expansion of the exact solution in the one-impurity case. The results presented concern the uniform and staggered susceptibilities, the instantaneous spin-spin correlations 〈${\mathbf{S}}_1$⋅${\mathbf{S}}_2$〉, the impurity self-energy and spectral density, the linear coefficient of the impurity specific heat (γ), and the Wilson’s ratio. The calculations are performed up to second order in the interaction, and for the nearly ‘‘symmetric’’ case. Most of the quantities investigated display 2${\mathit{k}}_{\mathit{F}}$R oscillations as a function of the impurity separation R, with a series of regions of ferromagnetic and antiferromagnetic correlations between the spins, which are quite independent of the value of the interaction U. The uniform susceptibility and the linear coefficient of the specific heat are enhanced (suppressed) in correspondence with regions of ferromagnetic (antiferromagnetic) correlations. Moreover, we find that the oscillations in the static susceptibilities tend to increase considerably with increasing U, and conjecture that this increase might lead to large oscillations in the Kondo regime. These oscillations would be the hallmark of strong Ruderman-Kittel-Kasuya-Yosida interactions in a Fermi-liquid (Kondo) ground state. We also find that the Wilson ratio has a rich nonuniversal behavior, confirming previous investigations on the two-impurity Kondo model. The relevance of these results and the connection with the literature is discussed.

Using the techniques of spin thermodynamics in solids, we extended the theory of spin relaxation by slow bulk motion proposed by J. Jeener et al. [Adv. Magn. Reson. 14, 95 (1990)] to the more complex spectroscopic situation prevailing in gypsum in ‘‘2-lines’’ orientation. In order to test the calculations, we measured the additional rate of relaxation due to low-amplitude, low-frequency (1 KHz) rotational vibration of the whole crystal around various directions. In agreement with calculated predictions, we observed that the additional relaxation only affects the intermolecular spin-spin order at a rate which is temperature independent and proportional to the (average) square of the angular velocity of bulk rotation. The measured additional rate of relaxation also depends strongly on the orientation of the axis of rotation, in quantitative agreement with the calculated values.

We examine phases of the Shraiman-Siggia model of lightly doped, square lattice quantum antiferromagnets in a self-consistent, two-loop, interacting magnon analysis. We find magnetically ordered and quantum-disordered phases both with and without incommensurate spin correlations. The quantum-disordered phases have a pseudogap in the spin excitation spectrum. The quantum transition between the magnetically ordered and commensurate quantum-disordered phases is argued to have the dynamic critical exponent z=1 and the same leading critical behavior as the disordering transition in the pure O(3) σ model. The relationship to experiments on the doped cuprates is discussed.

In this paper, we continue our investigation of a one-dimensional, two-component, quantum many-body system in which like particles interact with a pair potential s(s+1)/${\sinh }^2$(r), while unlike particles interact with a pair potential -s(s+1)/${\cosh }^2$(r). For an equal number of particles of the two components, the ground state for s>0 corresponds to an antiferromagnet/insulator. Excitations consist of a gapless pair-hole–pair continuum, a two-particle continuum with gap and excitons with gap. For -1<s<0, the system has two gapless excitations—a particle-hole continuum and a two-spin-wave continuum. Using finite-size scaling methods of conformal field theory, we calculate the asymptotic expressions and critical exponents for correlation functions of these gapless excitations at zero temperature. The conformal structure is closely related to the Hubbard model with repulsive on-site interaction.

Recent experiments have shown that the Kondo effect is suppressed in thin films, a result which has so far eluded theoretical explanation. The previous experiments have been limited to the regime T≳${\mathit{T}}_{\mathit{K}}$, where ${\mathit{T}}_{\mathit{K}}$ is the Kondo temperature. In this paper we present new results for T≲${\mathit{T}}_{\mathit{K}}$ which demonstrate that even when the Kondo contribution to the resistivity, Δ${\mathrm{\rho }}_{\mathit{K}}$ is greatly suppressed, the Kondo temperature itself is not affected. We also present the results of experiments involving bilayer samples, i.e., Kondo films covered by nonmagnetic layers, which demonstrate the existence of a Kondo ‘‘proximity effect.’’

Several publications dealing with the infinite-range quantum XY and Heisenberg spin glasses in an external magnetic field K in the z direction share a common supposition: the correlation between the x and y components of the spins is zero. Using the Matsubara imaginary-time formalism we show that this assumption holds only for those models where K=0. With the help of the Trotter-Suzuki formulation we investigate the phase diagram of the XY model for spin S=1/2, and find that it is different from those previously reported because of the nonzero xy correlation. It is also argued that for a general spin S in the XY model at zero temperature the spin-glass phase persists up to ${\mathit{K}}_{\mathit{c}}$=2S.

Observations of the flux-line lattice in tilted superconducting specimens by standard methods of Lorentz microscopy can be understood in terms of a simple model which (i) first approximates the individual vortex line by a flux tube of negligible radius, thereby obtaining an analytical expression for the two-dimensional phase shift, and then (ii) takes into account the finite flux core structure by convoluting the former expression with the assumed model for the core. By taking the London model for the magnetic-field distribution of the core, thus having the vortex described by a single parameter, i.e., the London penetration depth, the most relevant contrast features observed in the experimental out-of-focus images of a Nb superconductor are interpreted.

Arovas and co-workers, in their well-known paper on the statistical mechanics of anyons [D. P. Arovas, R. Schrieffer, F. Wilczek, and A. Zee, Nucl. Phys. B251, 117 (1985)], found that the second virial coefficient of a two-dimensional gas of free anyons exhibits nonanalytic behavior as the statistical parameter is varied in the neighborhood of the Bose point. In this paper we consider the anyons to be interacting, via a two-body interaction U(‖${\mathbf{r}}_1$-${\mathbf{r}}_2$‖), where ${\mathit{r}}^2$U(r)→0 as r→0. Expanding the second virial coefficient in the vicinity of the Bose point, we show that the nonanalyticity is still present and independent of the particular form of U. Furthermore, we calculate explicitly, the second virial coefficient in the case of a short-range interaction.

We report measurements of the complex ac susceptibility χ(T) for both isotropic and anisotropic square superconducting wire networks as well as direct measurements of the magnetic normal-to-superconducting phase boundary ${\mathit{T}}_{\mathit{c}}$(H)${\mathrm{\Vert }}_{\mathrm{\chi }}$ for these systems. The χ(T) transition is substantially broader and exhibits greater depression in magnetic field than the resistive transitions R(T). Commensurability structures found in the ${\mathit{T}}_{\mathit{c}}$(H)${\mathrm{\Vert }}_{\mathrm{\chi }}$ measurements at low-order rational fields are greatly enhanced compared to those found in resistively measured phase boundaries ${\mathit{T}}_{\mathit{c}}$(H)${\mathrm{\Vert }}_{\mathit{R}}$. For square networks made anisotropic by different wire widths in the two perpendicular directions, the ${\mathit{T}}_{\mathit{c}}$(H)${\mathrm{\Vert }}_{\mathrm{\chi }}$ phase boundaries demonstrate that increasing the anisotropy greatly increases the depression of the susceptibility transition temperature at incommensurate applied magnetic fields. This indicates a weakening of the network’s ability to screen magnetic field with larger anisotropy despite the fact that anistropy is increased by adding material to one set of parallel wires. This result supports a picture of anisotropic localization of the order parameter and/or anisotropic vanishing of the flux pinning or helicity modulus in periodic systems in an irrational homogeneous field.

Zero-field and magnetic-field low-temperature specific-heat measurements have been carried out on single crystals of ${\mathrm{CuV}}_2$${\mathrm{S}}_4$. The electronic-specific-heat coefficient (in zero field) is found to be 62 mJ/mol ${\mathrm{K}}^2$ and the Debye temperature to be 205 K. A superconducting transition at 4.0 K has been reported in powder samples contaminated by an estimated 5 to 10 % sulvanite ${\mathrm{Cu}}_3$${\mathrm{VS}}_4$. Present experiments of the electrical resistivity and the specific heat, however, verify that high-quality single crystals do not exhibit a superconducting transition above 60 mK. Below 0.2 K, nuclear specific heat is observed. Applied magnetic fields increase the magnitude of the specific heat due to the nuclear Zeeman effect of V and Cu nuclei and the electric quadrupole interaction of Cu nuclei. The electric-field gradient at Cu nuclei is estimated to be eq=1.74×${10}^{22}$ V/${\mathrm{m}}^2$. Detailed analysis of the nuclear specific heat below 1.0 K is presented. A comparison of the theoretical analysis with the experimental result is given and are in excellent agreement.

We study the evolution of the ground state and the excitation spectrum of the two- and three-dimensional attractive (negative-U) Hubbard model as the system evolves from a Cooper-pair regime for U≪t, to a composite boson regime for U≫t. Our work is motivated by the observation that the high-temperature superconductors, with their short coherence lengths and unusual normal-state properties, may be in an intermediate coupling regime between these two limits. A mean-field analysis of pairing, suitably generalized to account for a shift in the chemical potential, is known to be able to describe the ground-state crossover as a function of U/t. We compute the collective-mode spectrum using a generalized random-phase-approximation analysis within the equations-of-motion formalism. We find a smooth evolution of the Anderson mode for weak coupling into the Bogoliubov sound mode for hard-core bosons. We then include a long-range Coulomb interaction and show that it leads to a plasmon which again evolves smoothly from weak to strong coupling.

We solve the Bogoliubov–de Gennes equations for a clean, one-dimensional superconductor in the presence of superfluid flow. The maximum electrical current occurs when the superfluid velocity ${\mathit{v}}_{\mathit{s}}$ equals the Landau depairing velocity Δ/${\mathit{p}}_{\mathit{F}}$, where Δ is the pairing potential of the superconductor and ${\mathit{p}}_{\mathit{F}}$ is the Fermi momentum. The resulting critical current is approximately 2/π times smaller than the value eΔ/ħ recently predicted for a superconducting point contact. The ‘‘discretized’’ critical current of (2/π)(eΔ/ħ) arises when all the conducting electrons are forced to drift at the Landau depairing velocity.

At low temperatures, the transport through a superconductor–normal-metal tunnel interface is due to tunneling of electrons in pairs. We show that the rate for this process is often determined by the interference of the electron waves on a space scale determined by the coherence length. Therefore, the subgap current strongly depends on the layout of the electrodes within this space scale. The approach developed allows us to evaluate the subgap current for different layouts of interest.

We analyze the effect that the gauge field has on the superconducting transition temperature in the t-J model. Mean-field theories of the t-J model tend to predict d-wave superconductivity at a very high temperature of the order of ${\mathit{T}}_{\mathit{c}}^0$≃0.15 J. We will show that this transition temperature is suppressed, if one takes a fluctuating gauge field into account. The underlying idea is that there is a significant reduction of free energy due to gauge-field fluctuations, which is partly lost when a superconducting gap opens up. This cost of energy prevents the system from going into a superconducting state. Superconductivity is only possible at an intermediate range of doping, when the superfluid density of holons is sufficiently large to stiffen the gauge field. These ideas are supported by a numerical analysis. We obtained a phase diagram in the doping-temperature plane, that shows that for t/J=3 the optimal ${\mathit{T}}_{\mathit{c}}$ occurs at a doping of x≃0.15. One consequence of our analysis is that in this model the spin-gap phase is completely destroyed by gauge-field fluctuations.

The influence of a logarithmic Van Hove singularity in the electronic density of states is studied in the framework of the BCS theory. We use a simplified model, neglecting effects such as interlayer coupling and pair breaking. Through analytic and numerical analyses it is found that although the model can give rise to high temperatures, there are other properties, such as the isotope effect and specific-heat jump at the transition temperature, which cannot be explained properly with this model, as recent experimental data in the high-${\mathit{T}}_{\mathit{c}}$ oxides show.

The momentum, energy, and temperature dependence of the dynamic spin susceptibility is calculated self-consistently for the two-dimensional Hubbard model with additional electron-phonon interaction. The spin-fluctuation spectral density considered as a function of momentum exhibits four incommensurate peaks and a local minimum at the zone edge of the Brillouin zone. The large self-energy corrections tend to smooth out the peak and gap structure of the spin fluctuation spectral density. In the presence of an electron-phonon interaction (${\lambda }_{\mathrm{ph}}^{\mathrm{eff}}$≊2.2,U/t=2.3) a transition temperature ${\mathit{T}}_{\mathit{c}}$=84 K is found for s-wave pairing. Above ${\mathit{T}}_{\mathit{c}}$ the incommensurate peak and gap structure seen in a momentum scan is strongly smoothed out, and for increasing temperature and energy, the two well-separated peaks evolve into a single broad peak at the zone edge. For lower energies the whole structure is rapidly suppressed as the temperature drops below ${\mathit{T}}_{\mathit{c}}$. The results for the resistivity, the nuclear-spin relaxation rate, and the neutron-scattering spectral weight are compared with experimental data on ${\mathrm{La}}_{2\mathrm{-}\mathit{x}}$${\mathrm{Sr}}_{\mathit{x}}$${\mathrm{CuO}}_4$ and ${\mathrm{YBa}}_2$${\mathrm{Cu}}_3$${\mathrm{O}}_{7\mathrm{-}\mathit{x}}$.

The Gutzwiller wave function for the three-band Hubbard model on a two-dimensional ${\mathrm{CuO}}_2$ plane is studied by using the variational Monte Carlo (VMC) method in the limit ${\mathit{U}}_{\mathit{d}}$→∞, where ${\mathit{U}}_{\mathit{d}}$ is the Coulomb repulsion between holes on a Cu site. The VMC results for the energy and the fraction of d holes are compared with those of the Rice-Ueda type Gutzwiller approximation (GA). The difference between the VMC and the GA results are most pronounced at half-filling, and away from half-filling the two results agree well in both the hole-doped and the electron-doped cases. The doping dependence of the momentum distribution function is also studied.

We have measured voltage-current characteristics for ${\mathrm{YBa}}_2$${\mathrm{Cu}}_3$${\mathrm{O}}_7$ thin films in magnetic fields from 5 T to ambient, including some in the millitesla range. In all cases, the resistivity–current density isotherms can be separated into two classes: those which exhibit upward curvature, and have constant resistivity at low currents, and those which exhibit downward curvature at all currents. These two classes of isotherms are separated by a field-dependent temperature ${\mathit{T}}_{\mathit{g}}$. For each field, the isotherms scale in a manner consistent with a three-dimensional vortex-liquid to vortex-glass phase transition. The region that can be scaled is several kelvin at 5 T and narrows with decreasing field. The low-current resistivity above ${\mathit{T}}_{\mathit{g}}$ varies as ‖1-T/${\mathit{T}}_{\mathit{g}}$${\mathrm{\Vert }}^{\mathit{s}}$ with s≊7.4 at all fields, but the critical scaling exponents z and ν become field dependent below about 10 mT.