Nuclear-Spin-Lattice Relaxation in Dilute Magnetic Alloys Cu-Fe and Cu-Cr

Abstract

New data for impurity-induced relaxation of host nuclei in Cu-based alloys of Fe and Cr are presented. The impurity concentration was kept low to avoid complication arising from the impurity-impurity interaction. We used a pulsed method to measure ${T}_{1}$ at high fields with an accuracy of 2%. Impurity-induced relaxation in Cu-Fe (${T}_{K}\ensuremath{\approx}10$\ifmmode^\circ\else\textdegree\fi{}K) was investigated in the range 1.2-77\ifmmode^\circ\else\textdegree\fi{}K. Throughout this range a finite amount of the impurity-induced relaxation rate was observed for the 320-ppm sample, although the experimental error was too great to permit a quantitative argument for $T>{T}_{K}$. The accuracy was better in the range 1.2-4.2\ifmmode^\circ\else\textdegree\fi{}K, and therefore the functional dependence of impurity-induced relaxation on temperature and field was investigated in detail for 320-, 170-, and 92-ppm samples in this temperature range. We concluded that there was little temperature and field dependence for these alloys. A linear concentration dependence was observed in this range. We also measured ${T}_{1}$ in Cu-Cr (${T}_{K}\ensuremath{\approx}1$\ifmmode^\circ\else\textdegree\fi{}K) in the range 1.2-4.2\ifmmode^\circ\else\textdegree\fi{}K. The temperature and field dependence were small. The data for $T>{T}_{K}$ were compared with various theories proposed on the basis of the Ruderman-Kittel-Kasuya-Yosida interaction and the dipolar interaction. None of theories, however, could account completely for the magnitude and the functional dependence. For $T<{T}_{K}$ no theory is available to explain relaxation. The data in this range for Cu-Fe and Cu-Cr (the latter were taken by Gladstone) indicate a peculiar temperature dependence which may be characteristic of the anomaly associated with the magnetic ground state. The field dependence in these alloys can be explained on this model.