Observation of Nagaoka's Bound State for Conduction Electrons in Dilute Magnetic Alloys

Abstract

a shift nor a broadening of the emitted line with respect to the incident line has been observed within the limits of error of ±3 A given by the resolution of the monochromator. Furthermore, the intensity of emission obtained by irradiating the foil with light of a narrow spectral width has been found to be the same as that obtained at the same frequency by irradiating the foil with a large frequency band (>1000 A). These results are in good agreement with the emission experiments made on silver. 4 Obviously, the plasma behaves like a resonator.The half-width of the emitted line is 400 A or 460 meV in good agreement with the energy-loss measurements. (However, these two values are not directly to be compared, since the width of the emitted line is dependent on the observation angle 6. It needs to be extrapolated to 9 = 0 in order to be compared with the half-width of the electron energy loss. For this purpose, the half-width has to be measured Recent theories differ as to whether a negative s-d exchange interaction between a localized-impurity moment in a dilute alloy and the conduction-electron spins will result in the formation below some critical temperature T c of a quasibound state analogous to the Cooper pair in the theory of superconductivity, 1 "" 3 or will merely cause a strongly temperaturedependent scattering of conduction electrons at low temperatures. 4 Electrical resistivity measurements over the three-decade temperature range from 40 mdeg to 40°K reported here for dilute alloys of iron in copper permit a clear choice between the two predictions. Preliminary susceptibility measurements on the same material also tend to support the bound-state model. Quantitative agreement below T c between Nagaoka's theory and the resistivity data permit T c and hence Kondo's exchange coupling constant J to be determined. Above T c serious disagreement is found with the predictions of pertur-as a function of 0. These experiments are in preparation. This correction, however, is not of great importance.)It may be added that the maximum of the emission can be found at longer wavelengths than 3300 A, depending on the conditions of evaporation; the minimum of the transmission (plasma-resonance absorption) is displaced in the same way.These experiments show that the plasma of free electrons, excited by light, also shows the phenomenon of plasma-resonance emission.\J. Brambring and H. . bation theory, heretofore considered by most authors to be valid in the region sufficiently far above this temperature. The high value of T c (16°K) and the observed absence of impurity-impurity interactions make dilute Cu-Fe a nearly ideal system for studying the quantitative features of the low-temperature state.The alloy samples were vacuum cast in alumina-coated spherical graphite molds 11 mm in diameter, using high-purity copper combined with a suitable quantity of a Cu-0.5%Fe master alloy to create the very dilute samples needed. The larger susceptibility sample was prepared from the same master in a similar mold of different shape. A...

Temperature-Dependent Internal Field Distribution and Magnetic Susceptibility of a Dilute Ising Spin System

“…3. Further possible evidence that the properties of Cu-Mn are determined by internal field effects may be found in the radically different resistivity of Cu-Mn and Cu-Fe at low temperatures, as was found by Steyert et al, 12 that Cu-Fe does not exhibit a lowtemperature resistivity maximum whereas Cu-Mn does. 10 This maximum has been discussed as arising from internal field effects.…”

Role of phonon dimensionality on electron-phonon scattering rates

“…Since the Kondo contribution to the resistivity is dependent only on the electron distribution at the Fermi surface [14], the electrical resistance will yield the effective electron temperature [15]. The Kondo resistance anomaly of a CuCr film can be as large as a few percent of the residual resistivity for moderate Cr concentration [16], allowing the electron temperature to be determined accurately as a function of heating power. The electron gas temperature is defined in equilibrium where the electron energy has a Fermi distribu- In the nonequilibrium situation of a heating experiment, the electron distribution will not necessarily resemble a Fermi distribution at an elevated temperature.…”

Observation of Strong Electron Dephasing in Highly DisorderedCu93Ge4Au3Thin Films

“…This figure clearly indicates that, both in B 0 and 9 T, the R ᮀ reveals a lnT behavior all the way down to 50 mK. There is not even a sign of a crossover to a saturation characterizing a Kondo system in the unitary limit [13,27]. In fact, if there were c m * 200 ppm in the films, we should have been in a spin-glass state at low T < T K and a marked decrease in R ᮀ should be observed as the local moments freeze into a collective state [28].…”