It is well-known that the linewidth of the electron spin resonance (ESR) of localized moments (LM) in metals can be described by the empirical relation A H = a + b T .(1)Here the thermal broadening b is determined by thermal fluctuations of the exchange interaction of LM and conduction electrons (the so-called Korringa relaxation), while the residual linewidth a a r i s e s from spin-spin interactions of LM and lattice defects; so it is quite obvious physically that a should be essentially positive.In the'present paper we report the results of ESR studies of Gd in LaGd and YGd, and Mn in CuMn in a wide range of compositions. We have observed that relation (1) is valid for all the alloys in the paramagnetic state. The most important peculiarity of the results obtained is a non-monotonous concentration dependence of the residual linewidth which is negative at high concentrations of a magnetic component (see Fig. 1). This behavior held for all the systems studied. A s an example, in Fig. 2 there is shown the temperature dependence of AH for two of the LaGd samples. Though similar observations have been published from time to time (see, for example, /l/), as far a s we know, yet there is no proper physical explanation of this negative a value puzzle. It was Barnes /2/ who, having taken into account the fact of finite lattice spacing, obtained the modified coupled Bloch equations. The solution ~ 1) 420008 Kazan, USSR. 7 physica (b)
The electron paramagnetic resonance (EPR) of rare-earth ions in noble metals has recently been the subject of an extensive study (see, for example, /l/). Most of the investigations were carried out on Er3+, Dy3+, Gd3+, and Yb3' ions in Ag and Au hosts. EPR in Cu was observed only on E r 3+ ions.In this paper we report what we believe to be the first observation of an EPR 3+ signal in a pure Cu sample doped with Dy ions. Using Cu of 99.996% purity a s starting material, a single crystal of dilute magnetic alloy was grown by the Bridgeman method. There arise considerable difficulties when determining small impurity concentrations in dilute metallic'alloys. The evaluation of the Dy concentration in our sample was carried out by comparison of the resonance signal intensities of Dy in Cu in accordance with their susceptibilities. Since the E r solubility in Cu i s known to be rather high / 2 , 3/, absolute measurements of i t s concentration are possible. The evaluation gave c(Dy) = 0.05 at%. EPR measurements were performed at Y = 9400 MHz in the temperature range 1.7 to 4.2 K.
3+ 3+ 'and E r Throughout the temperature range investigated an isotropic EPR signal was observed with a typical metallic line shape. Taking into account that r is the ground state of E r rare-eakh ions vary only slightly in the same metallic host, one may expect r to be the ground state also for Dy = 7.63 2 0.02 which i s reasonably close to the calculated value g = 7.55. At the lowest temperatures (T = 1.7 K), a hyperfine structure from the odd Dy isotopes has been observed (see Fig. 1). According to the Breit-Rabi expression, the hyperfine structure constant was determined giving A = (80 ? 3)-very close t o the respective value for the Dy163 isotope in the cubic crystal field in dielectrics. The temperature dependence of the EPR linewidth fitted well the formula AH = a + bT 7 3+ ions in Cu and assuming that the crystal field parameters for different 7 3+ ions in Cu. In fact, our measurements gave g =
The data on spin-phonon coupling (SPC) of localized magnetic states may be informative when treating a microscopic model of the crystalline field in metals.F o r dielectrics such information is usually obtained when studying the effect of an externally applied uniaxial s t r e s s on the E P R spectra /l/. In recent experiments on thin metallic films /2 to 4/ the SPC constants have been determined; however, these constants may appear to be s m a l l e r than they actually a r e due to the lack of deformation control of the sample. Up to now the possibility of performing such measurements on bulk metallic samples has seemed to b e rather doubtful due to plastic deformation effects, but our experiments provide evidence that these studies a r e not only possible but also very informative.We have chosen for the experiments a copper single crystal containing 0.04 at% E r . The [liO] direction of the sample approximately coincided with the tensile load axis and was perpendicular to the rotation plane of the external magnetic field. E P R measurements were carried out at = 9400 MHz and T = = 1 . 8 K. In an unloaded sample was observed an isotropic line from the E r ion with g-factor g = 6.810 + 0.005 and linewidth AH = (16 + 0.8)G. The gfactor shows that the ground state for Cu:Er is the doublet r,. Application of load to the sample altered the E P R spectrum in such a way that when rotating the sample the linewidth remained unchanged, while in the g-factor behaviour there a r o s e an anisotropy ( F i g . 1) which increased with load up to 1 . 2 kg/mm . The g-factor anisotropy observed may be well described by the empirical expression 3+ 0 --2 (1 1 2g = g + A + B c o s @ , 0 where 8 is the angle between the [ l l O ] direction and the orientation of the external magnetic field. The analysis of these experimental data leads to the con-1 ) Kazan 420029, USSR.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.