Showing papers by "Albert Fert published in 1985"

Hall effect in Ce 1 − x Y x Pd 3 mixed-valence alloys

Abstract: Mixed-valence and Kondo lattice systems exhibit large anomalous Hall coefficients with a striking change of sign at low temperature in several systems (${\mathrm{CePd}}_{3}$, ${\mathrm{CeCu}}_{6}$,..., etc.). We have studied the Hall effect of ${\mathrm{Ce}}_{1\mathrm{\ensuremath{-}}\mathrm{x}}$${\mathrm{Y}}_{\mathrm{x}}$${\mathrm{Pd}}_{3}$, in which the substitution of small amounts of Y for Ce prevents the development of coherence at low temperature. We find that the Hall coefficient does not change its sign at low temperature and can be well understood in the one-impurity model of Ramakrishnan, Coleman, and Anderson. We infer that the change of sign observed in ${\mathrm{CePd}}_{3}$ is an effect of coherence.

Magnetic ordering in YGd alloys

Abstract: We present magnetization and resistivity measurements on single crystals of yttrium doped with small concentrations of gadolinium (c=1, 2, and 3 at. %). The low‐field susceptibility exhibits a spin‐glass‐like sharp cusp for the magnetic field along the c axis. However, there is no irreversible behavior below the temperature of the cusp, which rules out the existence of a spin‐glass state. Moreover, our magnetization measurements show up a spin‐flop transition for H in the basal plane. The existence of a long‐range ordering is confirmed by our resistivity measurements. The resistivity parallel to the c axis exhibits a maximum just below the ordering temperature, which is reminiscent of the superzone effects observed in many rare‐earth metals. The antiferro‐ magneticlike properties of the YGd alloys contrast with the spin‐glass behavior observed in other Y–rare‐earth alloys at similar concentrations.

Random anisotropy and phase diagram of spin glasses

Abstract: Torque measurements on CuMn, CuMnAu, and AuFe alloys confirm the phase diagram predicted by Kotliar and Sompolinsky for Heisenberg spin glasses with random anisotropy in the presence of an applied field. At low fields the transition is Ising‐like. High enough fields restore the behavior expected for Heisenberg glasses without random anisotropy.

Anisotropy in metallic spin glasses arising from gold impurities.

Abstract: Recent studies of CuMn alloys have shown that gold impurities introduce a large anisotropic coupling between the Mn ions. This coupling comes from the scattering of conduction electrons by the spin-orbit forces at the impurity sites. Since plane waves are scattered weakly by the spin-orbit interaction, the observation of strong anisotropy indicates that the orbital angular momentum of the conduction electrons in the vicinity of the gold impurity has been enhanced. The assumption that it is the p component which has been enhanced when the conduction electron wave function is orthogonalized to the 5p core states of gold cannot explain the data. We have concluded that the 5d shell of gold is not filled when gold is embedded in a copper matrix. A simple scaling argument shows that a vacancy of approximately (1/6) of an electron in the 5d shell accounts for the data. Our conclusion is supported by chemical valence data and provides an answer to an old mystery of the relative values of the residual resistivities of CuAg, CuAu, and AgAu.

Crystal field of Gd 3 + in hexagonal metals

Abstract: A second-rank effective-crystal-field Hamiltonian is calculated for ${\mathrm{Gd}}^{3+}$ in hexagonal metals We assume a 5d-like resonance (virtual-bound-state-like) near the gadolinium impurity The spin-orbit coupling of the 5d states transmits information about the crystal field acting on it to the spin part of the wave function This information is conveyed to the 4f shell through the isotropic 4f-5d atomic exchange interaction By using the point-charge model to estimate the crystal field acting on the 5d state, we find a 4f crystal-field splitting for Gd in magnesium (divalent) close to the one observed For the trivalent host metals, such as scandium, it is necessary to include higher-multipole contributions to the crystal field acting on the 5d virtual bound state in order to obtain the correct sign and magnitude of the splitting for Gd in these systems

Anisotropic magnetoelastic coupling in CeNi intermediate valence compound

Abstract: We present magnetostriction and thermal expansion measurements on the orthorhombic intermediate valence compound CeNi, at low temperature along the three symmetry axes. The field effects are clearly the sum of a main contribution arising from the intermediate valency character, which is temperature independent, and a smaller one whose temperature and field dependence is characteristic of Ce 3+ impurities. The low temperature thermal expansion and the main contribution to the magnetostriction are large and strongly anisotropic, the b -axis showing a particularly different behaviour. These results can be accounted for in a model based on an electronic free energy F e = − kTNf ( T / T f ), where the fluctuation temperature T f (ϵ a ,ϵ b ,ϵ c ) depends on the strains ϵ a ,ϵ b ,ϵ c in a different way. In this model the anisotropies of the thermal expansion and the magnetostriction are related to strong differences between the 3 Gruneisen parameters γ λ = − ∂ In T f /∂ϵ λ . Finally, we discuss the microscopic mechanisms wchic can give rise to this behaviour.