Widely reported broadening of a bipolaron formation region in two dimensions should be revised in view of a concrete mechanism of electron confinement to a two-dimensional layer.

The p-${\mathrm{H}}_2$ vibrons hae been studied by high-resolution Raman scattering in mixed crystals of (o-${\mathrm{H}}_2$${)}_{\mathit{x}}$(p-${\mathrm{H}}_2$${)}_{1\mathrm{-}\mathit{x}}$, (HD${)}_{\mathit{x}}$(p-${\mathrm{H}}_2$${)}_{1\mathrm{-}\mathit{x}}$, and (o-${\mathrm{D}}_2$${)}_{0.069}$(p-${\mathrm{H}}_2$${)}_{0.931}$. For increasing o-${\mathrm{H}}_2$ concentration, the line shape broadens and becomes asymmetric with a tail towards higher frequencies. This effect is less pronounced in the HD- and the o-${\mathrm{D}}_2$-doped crystals, where the energy difference between the p-${\mathrm{H}}_2$ vibron and the stretching vibration of the impurity is much larger. Isolated impurities still have a symmetrical line shape, showing that the asymmetry is not due to inhomogeneities. Calculated line shapes by the coherent-potential-approximation technique are in very good qualitative agreement with the experimental results. The theory also accounts for previously published experimental data on the variations of the relative positions and intensities of the ortho and para peaks in (o-${\mathrm{H}}_2$${)}_{\mathit{x}}$(p-${\mathrm{H}}_2$${)}_{1\mathrm{-}\mathit{x}}$.

The ${\mathrm{D}}_2$ vibrons have been studied by high-resolution Raman scattering in mixed crystals of (p-${\mathrm{D}}_2$)x(o-${\mathrm{D}}_2$)1-x. The o-${\mathrm{D}}_2$ and p-${\mathrm{D}}_2$ vibrational transitions are found to overlap. Line shapes calculated by the coherent potential approximation are in good qualitative agreement with the experimental data. The linewidth of the o-${\mathrm{D}}_2$ transition is much larger than the linewidth of the p-${\mathrm{H}}_2$ transition in (o-${\mathrm{H}}_2$${)}_{\mathit{x}}$(p-${\mathrm{H}}_2$${)}_{1\mathrm{-}\mathit{x}}$ mixed crystals with comparable J=1 concentrations. Possible explanations for this difference are discussed.

Magnetization measurements, reported for ${\mathrm{Fe}}_{3\mathrm{-}\mathit{x}}$${\mathrm{Ti}}_{\mathit{x}}$${\mathrm{O}}_4$ in the compositional range 0≤x<0.04, were used to determine magnetocrystalline anisotropy constants as a function of x and temperature. Magnetic measurements were performed along the principal crystallographic directions on single-crystalline specimens that had been annealed so as to preserve ideal oxygen stoichiometry. In contrast with the differences in the anisotropy of magnetite with Ti substitution, and those in nonstoichiometric magnetite above ${\mathit{T}}_{\mathit{V}}$, a remarkable similarity is observed in the anisotropy constants of ${\mathrm{Fe}}_{3\mathrm{-}\mathit{x}}$${\mathrm{Ti}}_{\mathit{x}}$${\mathrm{O}}_4$ and ${\mathrm{Fe}}_{3\left(1\mathrm{-}\mathrm{\delta }\right)}$${\mathrm{O}}_4$ in their dependence on x and 3δ at T=4.2 K. At this temperature, all anisotropy constants approach zero at the Ti concentration x≃0.037, which coincides with the simultaneous disappearance of the Verwey transition. Analysis of the results indicates that for magnetite both ${\mathrm{Fe}}^{2+}$ and ${\mathrm{Fe}}^{3+}$ ions contribute to the total anisotropy energy. The unusual anisotropy constant variation in the temperature range below ${\mathit{T}}_{\mathit{V}}$, observed in the second-order samples, leads to the conclusion, that either some kind of incomplete long-range order is present or that long-range order is entirely missing for x>0.012.

Microwave-magnetic-envelope (MME) solitons generated from nonlinear magnetostatic-backward-volume wave packets have been observed in magnetic thin films. The MME signals were excited by 5–50-ns wide microwave pulses at 5.8 GHz in a 15-mm-long by 2.5-mm-wide, 7.2-μm-thick single-crystal yttrium iron garnet (YIG) film strip magnetized in plane and parallel to the long side of the strip. The wave packets were propagated parallel to the static field. The wave packets were launched and the propagating MME pulse signals were detected with planar microstrip transducers 4 mm apart. Envelope soliton behavior was evident from the time-resolved wave forms observed for various input power and pulse width combinations. At low power levels, one sees a relatively broad output pulse which scales with the width of the input pulse and a peak power which increases linearly with the input power. As the input power is increased above some threshold in the 0.5–1-W range, output pulses show a narrowing and steepening which is characteristic of microwave-magnetic-envelope solitons. Further increases in input power produce multiple-peak profiles, characteristic of multiple soliton generation. The experimental results are consistent with the various characteristic times for linear and nonlinear MME pulse propagation and soliton formation. However, numerical modeling based on the magnetic form of the nonlinear Schrödinger equation with initial conditions and parameters which match the experiments yields calculated profiles which show soliton effects but do not quantitatively match the experimental results.

We study an anisotropic nearest-neighbor Ising antiferromagnet on the triangular lattice, using Monte-Carlo simulations and random-phase (or related) approximations. Our main interest concerns the high-temperature magnetic scattering. Exact results are available at the disorder line (Stephenson; Welberry and Galbraith). Of greater experimental significance is the Lifshitz line where (i) splitting of the scattering peaks occurs and (ii) dislocationlike defects appear in the (instantaneous) spin configurations. A good analytical fit of the Lifshitz line is given by the Bethe-Peierls approximation. These results could be relevant to other systems with disorder lines, e.g., the ANNNI model and microemulsions.

The Hubbard tight-binding Hamiltonian is used to derive the local magnetic moments of vicinal Cr surfaces and interfaces. For Cr vicinal surfaces, the magnetic arrangement is a double-cell configuration which consists of monoatomic steps antiferromagnetically coupled to nearest-neighbor monoatomic steps. These results allow us to explain the apparently contradictory observations obtained by spin-resolved and angle-and-energy-resolved photoemission. For the Fe monolayer on vicinal Cr surfaces, a magnetic configuration with a periodicity of two steps is also obtained. As a consequence, the sign of the magnetization at the Fe overlayer oscillates from step to step. This result is in agreement with recent experiments using scanning-electron-microscopy with polarization analysis and magneto-optic Kerr effect.

Local-spin-density-approximation calculations of the magnetic and electronic properties of CoH are reported. In its ground state CoH is found to be a strong ferromagnet with a moment of approximately 1.16${\mathrm{\mu }}_{\mathit{B}}$ and a magnetic energy of 10.1 mRy/f.u. The magnetic properties and their relationship to those of Co are discussed in terms of the electronic structure. The large moment in CoH is explained in terms of proton-induced hybridization of the electronic states.

We develop an extended self-consistent mean-field approach to study a spin-1 antiferromagnetic (AF) chain with the Hamiltonian H=${\mathit{tsum}}_{\mathit{i}}$ [${\mathit{S}}_{\mathit{i}}^{\mathit{x}}$${\mathit{S}}_{\mathit{i}+1\mathrm{}}^{\mathit{x}}$+${\mathit{S}}_{\mathit{i}}^{\mathit{y}}$${\mathit{S}}_{\mathit{i}+1\mathrm{}}^{\mathit{y}}$ +λ${\mathit{S}}_{\mathit{i}}^{\mathit{z}}$${\mathit{S}}_{\mathit{i}+1\mathrm{}}^{\mathit{z}}$+D(${\mathit{S}}_{\mathit{i}}^{\mathit{z}}$${)}^2$] based on Anderson’s spin-wave theory. The ground-state energy changes little compared with the traditional spin-wave theory and the excitation spectrum shows a gap which lends support to Haldane’s conjecture. At the Heisenberg point (λ=1,D=0), the ground-state energy and the gap are -1.3505 and 0.3914, respectively. In the Heisenberg-Ising case (λ>0,D=0), the gap decreases when λ goes down from 1 and at a critical ${\lambda }_{\mathit{c}}$>0 the gap goes to 0, while for the single-ion anisotropy (λ=1,D>0) the gap decreases with the increasing of D and at a critical value ${\mathit{D}}_{\mathit{c}}$ the gap goes to 0. These results agree well with those of other theoretical methods and experiments. We study the one-magnon excitation spectrum carefully and suggest an explanation for its crossing with the two-magnon excitation spectrum at low momentum. We estimate the crossing occurring at k≃0.3π. The results agree well with the very recent theoretical and experimental results. The picture of the Haldane gap is discussed from our extended self-consistent mean-field ground state and many interesting properties of the spin-1 AF chain are discussed.

Electron-paramagnetic-resonance (EPR) spectra recorded for ${\mathrm{Mn}}^{2+}$-doped ${\mathrm{ZnZrF}}_6$⋅${6\mathrm{H}}_2$O and ${\mathrm{CoZrF}}_6$⋅${6\mathrm{H}}_2$O show the presence of ‘‘forbidden transitions’’ Δ${\mathit{m}}_{\mathit{I}}$=±1,±2 for the central fine-structure transition and Δ${\mathit{m}}_{\mathit{I}}$=±1 for the remaining fine-structure transitions in the case of the zinc lattice and Δ${\mathit{m}}_{\mathit{I}}$=±1 again for the central fine transition for the cobalt lattice. At elevated temperatures, a phase transition indicated by differential-scanning-calorimetry does not appear in the EPR spectra, possibly due to a change in the space group but not the point symmetry at the ${\mathrm{Mn}}^{2+}$ ion site. The doublet separations for forbidden transitions are accounted for by using third-order perturbation theory invoking a quadrupole interaction, Q’=0.5 G. The ${\mathrm{Mn}}^{2+}$ spectra in the cobalt lattice are characterized by (a) large linewidths that decrease with increasing temperature and broaden out at low temperatures, (b) an anomalous intensity distribution among the fine-structure lines, and (c) a marked positive g shift, all of which are attributed to the Co-Mn interaction.

We wish to describe antiferromagnetic insulators at T=0 using a one-electron approach. To do this, we propose to overcome the inadequacies of standard local-spin-density theory by adding a spin-dependent magnetic pseudopotential to the Kohn-Sham equations. We then investigate in a general way using a simple two-dimensional model the physical behavior to be expected as band occupancy and interaction strength are varied. We find three phases: nonmagnetic metal, antiferromagnetic metal, and antiferromagnetic insulator. We emphasize the transition between large and small Fermi surfaces which occurs as the band occupancy changes.

Numerical simulation on resistance and giant magnetoresistance in magnetic granular alloys is performed for finite-size systems by making use of the real-space Green’s-function method based on the Kubo formalism. Spin-dependent scattering causes the giant magnetoresistance in the magnetic granular alloys as in the magnetic multilayers. It is shown by using the microscopic theory that the magnetoresistance increases with decreasing the size of magnetic grains, in agreement with experiment. The theoretical results indicate that scattering at the surfaces of magnetic grains governs the resistance and magnetoresistance in granular alloys. Difference between the giant magnetoresistances in the granular alloys and multilayers is discussed.

The spin-dependent interface resistance due to electron reflections and refractions at an interface potential step between ferromagnetic and nonmagnetic metals and due to impurities located at the interface is calculated within the Landauer approach for a current perpendicular to the interface. We show that experimental values of the interface resistance in Co/Ag and Co/Cu multilayers can be accounted for by assuming reasonable values of the potential steps and impurity concentrations.

We consider a classical spin chain characterized by axial anisotropic couplings between nearest neighbors. In particular the antisymmetrical part of exchange is taken into account by means of a Dzialoshinski coupling characterized by a regular or an alternate direction from pair to pair. We expand each operator exp(-β${\mathit{H}}_{\mathit{i}}$) on the infinite basis of spheroidal functions. We show that the zero-field partition function ${\mathit{Z}}_{\mathit{N}}$(0) has a closed-form expression. The spin-spin correlations can be derived as well as the susceptibilities ${\mathrm{\chi }}_{\mathit{u}\mathit{u}}$ (with u=x or y) and ${\mathrm{\chi }}_{\mathit{z}\mathit{z}}$. We achieve a thorough investigation of the chain low-temperature behaviors. It appears that there are predictable and original behaviors. In particular, when the z axis is favored, the lowest-energy configuration corresponds to a ferromagnetic or an antiferromagnetic arrangement, with the spins lying along this axis. When the X-Y plane is favored, the spins lie within this plane and exhibit a ferromagnetic, an antiferromagnetic, a helical, or a canted structure. Finally, we point out interesting crossover phenomena, but they are not discussed in detail.

The dynamics of the impurity spin in the Kondo problem has been studied in detail in the context of specific spectroscopic tools. The bosonized form of the Kondo Hamiltonian has been used to derive complete line-shape expressions relevant for interpreting NMR, Mössbauer, and ESR experiments. The calculation is equivalent to the so-called dilute-bounce-gas approximation in the treatment of the recently introduced spin-boson Hamiltonian. Analytical results are derived for longitudinal and transverse relaxation rates, NMR Knight shift, and the ESR g shift. From these, several known perturbative results, including the Abrikosov law, are recovered as special cases. The more general results derived here have been used to analyze a wide variety of spectroscopic data and the fits are found to be very satisfactory.

Fully relativistic spin-polarized calculations of U, Np, and Pu impurities in bcc iron have been carried out and used for the study of the electronic structure and magnetic hyperfine interactions of U, Np, and Pu impurity atoms in bcc Fe. A form of orbital-polarization correction within a one-electron approach is investigated. The results obtained are compared with available experimental data.

We study the magnetization of the ground states of Ising spin glasses as a function of the magnetic field. The grids are two-dimensional with Gaussian bond distribution. Finding a ground state in this case is a difficult combinatorial problem. We present a method that guarantees finding a true ground state. Although it is unlikely that one can have an algorithm with a polynomial bound for the running time, in all cases we tried, our method found a solution in a reasonable amount of time. We used this to obtain estimations of the critical exponents.

In this paper we investigate the influence of a surface anisotropy and the dipole interaction on the magnetization of thin films. The surface anisotropy favors a perpendicular orientation while the dipole interaction favors an in-plane magnetization. It is shown that these competing interactions for certain values of the parameters may result in a temperature-driven switching transition from an out-of-plane to an in-plane ordered state.

The mean-field theory of static and dynamic properties of the Dzyaloshinskii-Moriya (DM) interaction in Heisenberg spin glasses is developed. The main idea of the paper is based on the property of microscopic DM vectors ${\mathbf{D}}_{\mathit{i}\mathit{j}}$ describing this interaction: each vector ${\mathbf{D}}_{\mathit{i}\mathit{j}}$ is defined up to a sign. It is a reason for the existence of additional degrees of freedom connected with ordering of vectors ${\mathbf{D}}_{\mathit{i}\mathit{j}}$ on orientations below the freezing temperature. The theory is formulated on the basis of the eigenmodes of the exchange integrals matrix ${\mathit{J}}_{\mathit{i}\mathit{j}}$. There are two additional important assumptions to define the model. The first one is the independence of the sum ${\mathit{tsum}}_{\mathit{i}}$${\mathit{J}}_{\mathit{i}\mathit{j}}$ from the index j that restricts possible distributions of random exchange integrals. However, due to this assumption the total magnetization appears to be a hard eigenmode of the exchange part of the Hamiltonian and it is possible to a great extent to escape a detailed consideration of the spin-glass transition as well as to separate terms containing magnetization vectors in an equation for the thermodynamic potential in explicit form. The description of ordering of DM vectors ${\mathbf{D}}_{\mathit{i}\mathit{j}}$ is carried out in terms of some effective paramagnetic pseudospin, of which the state depends on the thermodynamic path of the spin-glass sample. This property displays itself in the distinction between the field and zero-field-cooling susceptibility and in an appearance of the macroscopical unidirectional anisotropy field (the DM field). The second assumption is to consider the DM field as a slow dynamical variable with a broad spectrum of relaxation times which, however, can behave as a quenched quantity at low enough temperatures. Motion equations of a dissipative character for pseudospins are examined. Pseudospins corresponding to localized eigenmodes with characterizing size ξ∼${10}^2$–${10}^3$ Å effectively contribute to observable slow relaxation. Using the dynamic susceptibility calculated the frequency dependence of freezing temperature and the magnetization decay law are found.

We study the phase diagram of the three-state Potts model on a triangular lattice with general interactions (ferro- and/or antiferromagnetic) between nearest-neighbor spins. When the interactions along two lattice-vector directions are antiferromagnetic and infinitely strong, this model becomes equivalent to a six-vertex model and exhibits a first-order potassium dihydrogen phosphate (KDP) transition from an ordered phase into a critical phase. Comparing the excitations that occurred by relaxing the restriction of infinite-strength interactions and those in the eight-vertex model, we analytically obtain the critical index for those excitations and demonstrate the existence of a critical phase for the case of finite antiferromagnetic interactions in two directions and ferromagnetic interactions in the other direction. When the interactions are antiferromagnetic in all three directions, Monte Carlo simulations show that a first-order line emerges from the KDP point and separates completely an ordered phase and a disordered phase. Along the special line where all three antiferromagnetic interactions have the same strength, the cell-spin analysis reveals that the symmetry of the ground states is dual to the symmetry of the n=3 ferromagnetic cubic model which is known to exhibit a first-order phase transition.