Magnetic-susceptibility (ac and dc) and heat-capacity measurements have been carried out on the compound PrRhSb. These measurements reveal two magnetic transitions in this compound—one at about 18 K and the other around 6 K. In the dc susceptibility the 18-K transition is evident as the temperature below which a magnetic correlation sets in and the susceptibility is found to be field dependent. The lower transition manifests as a peak in the susceptibility of zero-field-cooled samples which were measured in low applied fields. The electronic-specific-heat coefficient, γ, is found to be 33 mJ/mol ${\mathrm{K}}^2$ between 40 and 70 K after correcting for the lattice contribution taken to be the same as in its La analog. The γ value is fairly large for a Pr compound and may be indicative of moderately heavy quasiparticles. A Kondo-type interaction of the Pr 4f electrons with the conduction electrons may be responsible for high-magnetic-ordering temperatures and the moderately large γ value in this compound.

Within the general cluster framework of Kandel, Ben-Av, and Domany, we develop a cluster algorithm for Monte Carlo simulations of the antiferromagnetic Ising model on a triangular lattice. The algorithm does not suffer from problems of metastability and is extremely efficient even at T=0, which allows us to extract the static exponent η=0.5 as well as the effective dynamical critical exponent of the algorithm z=0.64±0.02.

The magnetic properties of granular, annealed, melt-spun ribbons of the "giant" magnetoresistors, ${\mathrm{Cu}}_{80}$${\mathrm{Co}}_{20}$ and ${\mathrm{Cu}}_{80}$${\mathrm{Co}}_{15}$${\mathrm{Fe}}_5$, have been studied by a variety of techniques. These include x-ray dfiffraction, electron microscopy, ferromagnetic resonance, SQUID magnetometry, Mössbauer-effect spectroscopy, and magnetoresistance. We utilize each of these measurements to reveal different aspects of the particle size distribution as a function of annealing temperatures. These melt-spun alloys require large magnetic fields for magnetic saturation, impairing their utility as magnetic sensors. However, the properties of melt-spun ribbons provide an understanding of why all granular magnetic materials are difficult to saturate. The magnetoresistance ratio of these alloys is maximized by a 500°C anneal with $\frac{\Delta \rho }{\rho }\cong 14\%$ at 4.2 K. The paramagnetic fraction determined by SQUID magnetometry at 4.2 K is 33% for this annealing temperature. The paramagnetic fraction determined by Mössbauer spectroscopy is 14% for samples annealed by 500°C, and vanishes when the sample is annealed at 900°C. The discrepancy between the two measurements of the paramagnetic fraction is due to the vastly different averaging times of the two techniques.

Finite-temperature phase boundaries of ultrathin magnetic films are obtained from Monte Carlo simulations as a function of the coupling strengths due to exchange (J), dipolar interaction (g), and z-axis anisotropy (K). We found three phases, corresponding to the appearance of the in-plane magnetization, the disappearance of the z magnetization, and the eventual disappearance of the in-plane magnetization. At small K-g, the transition temperature for the disappearance of the z magnetization can be well approximated by ${\mathit{T}}_{\mathit{c}}^{\mathrm{-}1}$=A+B ln[(K-g/0.120 83)/J] for constants A and B. For g=0, the magnetization disappears above the transition temperature. The constants A and B are changed only slightly (10%) as g becomes nonzero. The inverse temperature for the disappearance of the in-plane magnetization also depends linearly on ln(g/J) and only weakly on K. For a large enough K/g the magnetization never rotates onto the plane.

${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_{2\mathrm{-}}$${\mathrm{FeCl}}_3$ graphite bi-intercalation compounds (GBIC’s) with the stacking sequence -G-${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$-G-${\mathrm{FeCl}}_3$-G- along the c axis have been prepared by a method of sequential intercalation. The intraplanar exchange interaction in ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ layers is ferromagnetic, while the intraplanar exchange interaction in ${\mathrm{FeCl}}_3$ layers is antiferromagnetic. The magnetic properties of ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_{2\mathrm{-}}$${\mathrm{FeCl}}_3$ GBIC’s have been studied by using dc and ac magnetic susceptibility, and low- and high-field SQUID magnetization measurements. The ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ layer undergoes a magnetic phase transition at the critical temperature ${\mathit{T}}_{\mathit{c}}$ which changes from 19.48 K at c=0 to 9.10 K at c=1. A cluster glass phase appears below ${\mathit{T}}_{\mathit{c}}$ where the spin directions of ferromagnetic clusters are frozen because of frustrated interisland interactions. Due to the intervening ${\mathrm{FeCl}}_3$ layer the critical behavior of ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$-${\mathrm{FeCl}}_3$ GBIC’s at ${\mathit{T}}_{\mathit{c}}$ is three dimensional rather than two dimensional.

Below ${\mathit{T}}_{\mathit{N}}$ (≊4 K) the cluster glass phase may coexist with an antiferromagnetic long-range order occurring in the ${\mathrm{FeCl}}_3$ layers. The effect of antiferromagnetic interplanar interaction between the ${\mathrm{FeCl}}_3$ layer and the ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ layer is clearly seen in the magnetization of ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$-${\mathrm{FeCl}}_3$ GBIC’s. The nature of the coexisting phase is complicated by the spin frustration effect arising from the competition between the interplanar interaction between ${\mathrm{FeCl}}_3$ and ${\mathrm{Co}}_{\mathit{c}}$${\mathrm{Ni}}_{1\mathrm{-}\mathit{c}}$${\mathrm{Cl}}_2$ layers and the antiferromagnetic intraplanar interaction in the ${\mathrm{FeCl}}_3$ layer.

Au-sandwiched bcc (001) Fe wedges were produced by the molecular beam epitaxy method (0–20 Å). The polar Kerr rotation and Kerr ellipticity spectra (1.5–5 eV) were measured as a function of the Fe thickness d. In both kinds of spectra new peaks appear. The complex Kerr rotation oscillates as a function of the film thickness. More than four oscillations could be clearly observed in the measured thickness range. Magneto-optical transitions from the ${\mathrm{\Delta }}_5$ band to quantum-well states (QWS’s) in the ${\mathrm{\Delta }}_1$ majority-spin band change the intrinsic dielectric properties. The oscillation period and transition energies are in agreement with theory. Thickness-dependent structures below 2.5 eV strongly suggest the occurrence of partially confined QWS’s around ${\mathit{E}}_{\mathit{F}}$ or QWS’s in the ${\mathrm{\Delta }}_5$ band.

We have studied the local magnetic behavior of isolated Rh impurities in ${\mathrm{Au}}_{1\mathrm{-}\mathit{x}}$${\mathrm{Fe}}_{\mathit{x}}$ alloys for compositions x=0.01, 0.03, 0.10, and 0.18, using the time-differential perturbed-angular-correlation (TDPAC) method and by electronic structure calculations within the local spin-density (LSD) approximation. From the measured Curie-Weiss local susceptibility of the ${}_{}{}^{100}{}_{}{}^{}\mathrm{Rh}$ probe as well as the LSD calculations we find large stable magnetic moments 0.5-1.0${\mathrm{\mu }}_{\mathit{B}}$ on Rh for compositions x≥0.10. In low Fe concentration alloys the observed weak local susceptibility can be reconciled with the calculated large local moments, both positive and negative, on Rh in terms of a typical Ruderman-Kittel-Kasuya-Yosida (RKKY)-type mechanism between Fe and Rh. In the reentrant spin-glass regime, T≤60 K for x=0.18, the hyperfine field ${\mathit{B}}_{\mathrm{hf}}$ shows a two-step temperature dependence, increasing sharply below 60 K, such that its extrapolated value at 0 K is enhanced by 30% over its saturated value in the ferromagnetic region.

We study, within the dynamic-pair approximation, the behavior of the antiferromagnetic Ising model in contact with a heat bath and subject to an external source of energy. The contact with the heat bath is simulated by the Glauber process while the continuous flux of energy into the system is simulated by the Kawasaki process. We find the phase diagram of this model, and we show that, conversely to what happens for the ferromagnetic Ising model, the antiferromagnetic Ising model does not show the phenomenon of self-organization within the dynamic-pair approximation.

Spectral density for the elementary excitation of the multichannel Kondo model is investigated at finite temperatures with the use of the Bethe-ansatz method. Our formulation is based on the method of Yang and Yang developed in one-dimensional interacting boson systems. In the orbital singlet case, the narrow peak structure develops in the low-energy region as the temperature is decreased, whereas in the underscreened case, the excitation spectrum exhibits a divergence property at zero excitation energy, irrespective of the temperature. In the overscreened case, in which the non-Fermi-liquid fixed point is stable in the ground state, the peak structure develops in the low-energy region and the weight at zero excitation energy increases as the temperature is decreased. We discuss characteristic properties of the excitation spectrum in connection with bulk properties.

We report the results of an x-ray-scattering study of the b-axis lattice constant of the quasi-one-dimensional spin-1/2 antiferromagnetic chain system ${\mathrm{CuGeO}}_3$ as a function of temperature and magnetic field. A spontaneous thermal contraction Δb along the b axis perpendicular to the chain direction is observed below the spin-Peierls transition temperature ${\mathit{T}}_{\mathit{c}}$ near 14 K. The contraction, Δb, is well described by a simple power law, Δb∝(1-T/${\mathit{T}}_{\mathit{c}}$${)}^{\mathit{x}}$, where the exponent x is found to be close to 0.5. Below the transition temperature, Δb scales with the intensity of the superlattice reflections with indices (h/2,k,l/2)(h,l: odd, k: nonzero integer) measured by neutron scattering. The shift of the transition temperature, Δ${\mathit{T}}_{\mathit{c}}$≡${\mathit{T}}_{\mathit{c}}$(0)-${\mathit{T}}_{\mathit{c}}$(H), is found to scale as ${\mathit{H}}^2$ in quantitative agreement with the results of magnetic susceptibility measurements and with theory. A small increase in the a-axis lattice constant is observed below ${\mathit{T}}_{\mathit{c}}$.

We calculate the temperature dependence of the specific heat, susceptibility, and resistivity for an isolated cerium ion in crystal-field environments appropriate for the cerium sites in ${\mathrm{CeCu}}_6$ and ${\mathrm{CeAl}}_3$. We use the self-consistent ladder approximation for intermediate and high temperatures and match to exact results to obtain low-temperature behavior. We find unusual temperature dependence for all the calculated quantities as a result of the competition between crystal-field effects and the Kondo effect. Our results for specific heat and susceptibility agree well with the data, except that the independent ion approximation predicts somewhat smaller Wilson ratios than observed, perhaps implying weak ferromagnetic correlations. The resistivity is not well reproduced.

The magnetic and magnetoelastic properties (third-order magnetic susceptibility and parastriction) of ${\mathrm{TbPO}}_4$, which has the zircon-type tetragonal structure, are analyzed in the paramagnetic phase. The susceptibility formalism is then used to describe all the symmetry-lowering modes in a rare-earth insulator. The dominant magnetoelastic coupling is associated with the δ symmetry and is, at least partly, responsible for the transition occuring at 2.15 K in the antiferromagnetic phase and clearly dominates the γ-symmetry mode. However, none of the different symmetry-lowering modes can be completely neglected and we determine parameters for the α1 and α2 tetragonal modes and the ɛ monoclinic mode. Owing to the close vicinity of the first excited singlet to the ground-state doublet, the α quadrupolar interactions play an important role in determining the temperature dependence of the level spacing at low temperature, in particular in the ordered phases. The determination of all the magnetoelastic modes of ${\mathrm{TbPO}}_4$ is an essential step towards the understanding of the complex magnetic properties in the ordered phases, where numerous interactions coexist.

The electrical resistivity, ρ, of β-${\mathrm{HoH}}_{2+\mathit{x}}$ has been measured in the interval 1.3≤T≤330 K for the entire x range within the pure β phase, 0≤x≤0.14. It is shown that the excess hydrogens, x, on octahedral sites interact below ∼200 K with each other to form short-range-ordered (SRO) structures for x≳0.05, transforming into long-range-ordered (LRO) configurations for x≳0.1. Disorder introduced into these structures by quench from room temperature recovers in two stages, centered near ${\mathit{T}}_{\mathit{p}}^{\mathrm{}\left(1\right)\mathrm{}}$∼150 K (for SRO) and near ${\mathit{T}}_{\mathit{p}}^{\mathrm{}\left(2\right)\mathrm{}}$∼180 K (for LRO), with activation energies of ${\mathit{E}}_{\mathit{m}}^{\mathrm{}\left(1\right)\mathrm{}}$∼0.15 eV and ${\mathit{E}}_{\mathit{m}}^{\mathrm{}\left(2\right)\mathrm{}}$=0.26–0.27 eV, respectively. Several magnetic transitions are observed below ∼10 K (three for low x values, 0≤x≤0.03) attributed to commensurate and incommensurate antiferromagnetism; they decrease in T with increasing x due to weakening Ruderman-Kittel-Kasuya-Yosida interaction. For higher x (≳0.05), additional magnetic transitions are induced by the ordered hydrogen configurations acting upon the crystal-field symmetry. A tentative magnetic phase diagram is constructed from the present ρ results and published susceptibility, neutron-diffraction, and specific-heat data. Metal-semiconductor transitions are observed near 250–270 K (MS), attributed to the breakdown of a delocalized band when crossing the x ordering interval, and near 100–130 K (SM) assigned to carrier localization due to atomic disorder.

The dynamics of Ising spin glasses above the freezing temperature ${\mathit{T}}_{\mathit{g}}$ have been studied numerically. A simple method is given which predicts the active-bond-percolation critical temperature ${\mathit{T}}_{\mathit{p}}$ in Ising spin glasses. The physical significance of this temperature and the associated damage-spreading temperature is discussed, together with the form of the autocorrelation-function relaxation between this temperature and ${\mathit{T}}_{\mathit{g}}$.

Magnetic systems consisting of a layer with finite thickness coupled to a semi-infinite bulk are studied with Ginzburg-Landau (GL) mean-field theory. The critical point of the system is determined in terms of the coefficients in the GL free-energy functional and the thickness of the thin layer. From the correlation functions within the semi-infinite bulk, an effective extrapolation length is able to be defined, although only the continuity of magnetization is assumed as the interfacial condition. This extrapolation length describes the coupling between the two subsystems and diverges with a new exponent ν̃=1. The critical behavior of spontaneous magnetization according to the variance of the layer thickness is derived as ${\mathit{m}}_{\mathit{s}}$∼ √L/${\mathit{L}}_{\mathit{c}}$-1 with the critical thickness ${\mathit{L}}_{\mathit{c}}$ and a new critical exponent ${\mathrm{\beta }}_{\mathit{L}}$=1/2.

The magnetic phase diagram of the pseudoternary (${\mathrm{Dy}}_{\mathit{x}}$${\mathrm{Gd}}_{1\mathrm{-}\mathit{x}}$)${\mathrm{Fe}}_{11}$Ti compounds has been experimentally and theoretically investigated. The thermal dependence of the low-field ac susceptibility on powdered samples exhibits a sharp peak, due to an axial-conical spin reorientation (SR) transition. The SR temperature (${\mathit{T}}_{\mathrm{SR}}$) decreases with increasing Gd content. The angular dependence of magnetization on magnetically aligned powdered samples has been measured in a steady applied magnetic field, in the 4.2–300 K temperature range. From these measurements we have determined the angle θ between the spontaneous magnetization and the tetragonal c axis. Onset temperatures for magnetization canting agree with the ${\mathit{T}}_{\mathrm{SR}}$ obtained from susceptibility data within 10 K. Both the ${\mathit{T}}_{\mathrm{SR}}$ vs x and the θ vs T dependences have been calculated using a single-ion, mean-field model for the rare-earth and transition-metal interactions. Very good agreement between theory and experiment was obtained.

Mössbauer spectra, magnetization measurements, and self-consistent spin-polarized electronic structures of ${\mathrm{YFe}}_{12\mathrm{-}\mathit{x}}$${\mathrm{Mo}}_{\mathit{x}}$, where x=0.5, 1.0, 2.0, 3.0, and 4.0, are reported. The ternary compounds ${\mathrm{YFe}}_{12\mathrm{-}\mathit{x}}$${\mathrm{Mo}}_{\mathit{x}}$ have the crystalline tetragonal ${\mathrm{ThMn}}_{12}$ structure. Analyses of the Mössbauer spectra show that Mo atoms occupy the 8i Fe sites of the ${\mathrm{ThMn}}_{12}$ structure, in agreement with previous observations. Room-temperature magnetic and Mössbauer measurements show that the compounds with x≤2.0 are ferromagnetic and with x≥3.0 are paramagnetic. Measurements at 25 K show that all the samples are magnetically ordered. The magnetic hyperfine field is found to decrease with increasing Mo concentration, which is in qualitative agreement with the calculated magnetic moments. The calculated magnetization decreases less rapidly with increasing x than the experimental data. In general the data suggest that with increasing Mo concentration there is an increase of antiferromagnetic coupling among the Fe moments, which leads to cluster-glass or spin-glass-like phenomena. The measured isomer shift relative to α-iron is found to decrease linearly with x.

X-ray magnetic circular dichroism (XMCD) in the ${\mathit{L}}_{2,3}$ absorption spectra of transition metals is a relatively new experimental technique for the investigation of magnetism. XMCD is element specific, giving it obvious advantages over methods which measure averaged magnetic properties. Another potential strength of XMCD is the separate quantitative measurement of 〈${\mathit{L}}_{\mathit{Z}}$〉 and 〈${\mathit{S}}_{\mathit{Z}}$〉 through application of dichroism sum rules. In this paper we present the results from a set of experiments designed to test the applicability of the dichroism sum rules to XMCD spectra measured by total electron yield. We find that with proper experimental control of angle of incidence, degree of polarization and film thickness, the total electron yield measurement of absorption spectra is accurate to within 5%. The uncertainties involved in applying the dichroism sum rules are discussed in detail, including possible effects of diffuse magnetic moments which are known to exist in Fe, Co, and Ni. The magnitude of these uncertainties makes it difficult to obtain absolute values of 〈${\mathit{L}}_{\mathit{Z}}$〉 and 〈${\mathit{S}}_{\mathit{Z}}$〉, however, the dichroism sum rules are shown to be qualitatively accurate and therefore capable of measuring changes in 〈${\mathit{L}}_{\mathit{Z}}$〉 and 〈${\mathit{S}}_{\mathit{Z}}$〉.

Exact expressions for the spin polarization due to a point field have been obtained for various inhomogeneous situations. The presence of boundaries or potentials modifies the nodes and amplitudes of the polarization, which determines the coupling between ion spins or nuclear magnetic moments in metals. For the half-space both in three and one dimensions, the analytic solution contains the usual Ruderman-Kittel-Kasuya-Yosida polarization around the source plus a polarization wave originating from the mirror point, although with the same phase at the border; there both are canceled by a third term. The case of a slab in three and one dimensions has been treated. A one-dimensional model with a reflectionsless localized potential with one bound state proved to be tractable. It is shown that the integrated polarization is given by the probability densities of the electron states at the position of the field summed over the Fermi surface.

We present results of numerical studies of the two-dimensional XY model with four- and eightfold symmetry-breaking fields. This model has recently been shown to describe hydrogen-induced reconstruction on the W(100) surface. Based on mean-field and renormalization-group arguments, we first show how the interplay between the anisotropy fields can give rise to different phase transitions in the model. When the fields are compatible with each other there is a continuous phase transition when the fourth-order field is varied from negative to positive values. This transition becomes discontinuous at low temperatures. These two regimes are separated by a multicritical point. In the case of competing four- and eightfold fields, the first-order transition at low temperatures opens up into two Ising transitions. We then use numerical methods to accurately locate the position of the multicritical point, and to verify the nature of the transitions. The different techniques used include Monte Carlo histogram methods combined with finite-size scaling analysis, the real-space Monte Carlo renormalization-group method, and the Monte Carlo transfer-matrix method. Our numerical results are in good agreement with the theoretical arguments.