We observed the conduction electron spin resonance (CESR) in fine powders of MgB2 both in the superconducting and normal states. The Pauli susceptibility is chi(s) = 2.0 x 10(-5) emu/mole in the temperature range of 450 to 600 K. The spin relaxation rate has an anomalous temperature dependence. The CESR measured below T(c) at several frequencies suggests that MgB2 is a strongly anisotropic superconductor with the upper critical field, H(c2), ranging between 2 and 16 T. The high-field reversible magnetization data of a randomly oriented powder sample are well described assuming that MgB2 is an anisotropic superconductor with H(ab)(c2)/H(c)(c2) approximately 6-9.

A combination of x-ray diffraction, magnetization, and 75 As nuclear magnetic resonance (NMR) experiments were performed on single-crystal EuFe 1.9 Co 0.1 As 2. The strength of the hyperfine interaction between the 75 As nuclei and the Eu 2+ 4f states suggests a strong coupling between the Eu 2+ moments and the Fe 1.9 Co 0.1 As 2 layers. Such a strong interlayer coupling may be due to an indirect exchange interaction between the localized Eu 2+ 4f moments, mediated by the Fe 3d conduction electrons. Magnetic susceptibility as well as 75 As-NMR measurements reveal a decrease of the SDW transition temperature to T SDW = 120 K as a result of Co doping. A change of the slope in the temperature dependence of the NMR frequency of the 75 As lower-satellite line was observed at 225 K. At the same temperature also a change of the satellite line shape was found. These changes of the NMR spectra may be caused by the formation of a nematic phase below 225 K in EuFe 1.9 Co 0.1 As 2 .

The temperature dependence of the electron spin relaxation time in MgB2 is anomalous as it does not follow the temperature dependence of the resistivity above 150 K, it has a maximum around 400 K, and it decreases for higher temperatures. This violates the well established Elliot-Yafet theory of electron spin relaxation in metals. We show that the anomaly occurs when the quasiparticle scattering rate (in energy units) becomes comparable to the energy difference between the conduction-and a neighboring band. We find that the anomalous behavior is related to the unique band structure of MgB2 and the large electron-phonon coupling. The saturating spin-lattice relaxation can be regarded as the spin transport analogue of the Ioffe-Regel criterion of electron transport.PACS numbers: 74.70. Ad, 74.25.Nf, 76.30.Pk, 74.25.Ha Knowledge of the electron spin-lattice relaxation time, T 1 , of conduction electrons plays a central role in assessing the applicability of metals for information processing using electron spins, spintronics [1]. T 1 is the time it takes for the conduction electron spin ensemble to relax to its thermal equilibrium magnetization after a non-equilibrium magnetization has been induced e.g. by conduction electron-spin resonance (CESR) excitation [2] or by a spin-polarized current [1]. The Elliott-Yafet (EY) theory of T 1 in metals [3,4] has been well established in the past 50 years on various systems such as elemental metals [5], strongly correlated one-dimensional [6], and some of the alkali fulleride salt [7] metals. It is based on the fact that the spin part of the conduction electron wave functions is not a pure Zeeman state but is an admixture of the spin up and down states due to spin-orbit (SO) coupling. As a result, momentum scattering due to phonons or impurities induces electron spin-flip, which leads to spin relaxation. Typically every millionth momentum scattering is accompanied by the electron spinflip due to the relative weakness of the SO coupling. Thus, T 1 ≫ τ (τ being the momentum relaxation time) which explains the motivation behind the efforts devoted to the spintronics applications of metals.A consequence of the EY theory is the so-called Elliottrelation, i.e. a proportionality between T 1 and τ [3]:Here α is a band structure dependent constant and for most elemental metals α ≈ 1..10 (Ref.[5]). L is the SO splitting for spin up and down electrons in a valence (or unoccupied) band near the conduction band with an energy separation of ∆E. E.g. in sodium, the conduction band is 3s derived and the relevant SO state is the 2p with ∆E = 30.6 eV andThe Elliott-relation shows that the temperature dependent resistivity and CESR line-width are proportional, the two being proportional to the inverse of τ and T 1 , respectively. This enabled to test experimentally its validity for the above mentioned range of metals. Much as the Elliott-relation has been confirmed, it is violated in MgB 2 as therein the CESR line-width and the resistivity are not proportional above 150 K [8].Here, we study ...

We present a novel method to determine the resonant frequency and quality factor of microwave resonators which is faster, more stable, and conceptually simpler than the yet existing techniques. The microwave resonator is irradiated at a frequency away from its resonance. It then emits an exponentially decaying radiation at its eigen-frequency when the excitation is rapidly switched off. The emission is down-converted with a microwave mixer, digitized and its Fourier transformation (FT) directly yields the resonance curve in a single shot. Being an FT based method, this technique possesses the Fellgett (multiplex) and Connes (accuracy) advantages and it conceptually mimics that of pulsed nuclear magnetic resonance. We also establish a novel benchmark to compare accuracy of the different approaches of microwave resonator measurements. This shows that the present method have similar accuracy to the existing ones.