Spin relaxation in mesoscopic superconducting Al wires

Abstract

We studied the diffusion and the relaxation of the polarized quasiparticle spins in superconductors. To that end, quasiparticles of polarized spins were injected through an interface of a mesoscopic superconducting Al wire in proximity contact with an overlaid ferromagnetic Co wire in the single-domain state. The superconductivity was observed to be suppressed near the spin-injecting interface, as evidenced by the occurrence of a finite voltage for a bias current below the onset of the superconducting transition. The spin diffusion length, estimated from finite voltages over a certain length of Al wire near the interface, was almost temperature independent in the temperature range sufficiently below the superconducting transition but grew as the transition temperature was approached. This temperature dependence suggests that the relaxation of the spin polarization in the superconducting state is governed by the condensation of quasiparticles to the paired state. The spin relaxation in the superconducting state turned out to be more effective than in the normal state. Recently the spin-dependent electron transport has been the subject of intensive studies. The key element of the phenomenon is to inject a current of spin-polarized conduction electrons into a mesoscopic or nano-scale non-magnetic metal or semiconductor, control, and detect the resulting spin state. Spin-polarized electron can be injected from a ferromagnet (F) into the system under study. 1,2,3,4,5,6,7,8 To realize the spin-dependent electronic conductance or "spintronics" it is essential to obtain the accurate information on the characteristic spin-relaxation time or length of the injected electrons in the metallic or semiconducting system in the presence of spin-relaxing scattering. 1,2,3,4,5,6,7,8 The spin-relaxation originates from both scattering by magnetic impurities and spin-orbit scattering of conduction electrons, but the relaxation due to spin-orbit scattering is dominant without magnetic impurities. A number of studies on the spin relaxation in metals have been done using nonlocal spin injection, 1,8,9 conduction electron spin resonance, 10,11,12 weak localization, 13,14 and superconducting tunneling spectroscopy. 15,16,17,18 Observed spin relaxation rate using different techniques at room temperature, where the electron-phonon interaction predominates the spin-orbit scattering, reveals reasonable consistency, but it shows a wider spread at low temperatures around liquid helium temperature. It has been pointed out that 9 , as the impurity scattering predominates the spin-orbit scattering at low temperatures, the measured spin relaxation rates may depend on different measurement techniques which are sensitive to different impurity-induced spin-orbit scattering.Recently, the spin relaxation in a superconductor (S), both conventional 19,20,21,22,23,24 and high-T c cuprate, 25,26,27,28 has attracted much research interest in relation with the recombination mechanism of the spin-polarized quasiparticles into the singlet Cooper-paired state. ...

Revealing Charge Density Wave Formation in theLaTe2System by Angle Resolved Photoemission Spectroscopy

“…Therefore, a natural question is whether there exists any proven instance of a 2D CDW system where the CDW phase is driven by FS nesting, yet non-metallic. The 2D CDW rare earth ditellluride system LaTe 2 [18]- [20], formed by square tiled Te layers separated by RTe slabs (R = rare earth), is ideal to address this question, having been previously shown to be non-metallic [20,21]. The CDW phase is well established, first by transmission electron microscopy (TEM) measurements which identified a modulation wave vector 0.5a * [18], which we will refer to as q 1 , and later by single crystal X-ray diffraction [19], which proposed a 2 × 2 × 1 superstructure.…”

“…The CDW phase is well established, first by transmission electron microscopy (TEM) measurements which identified a modulation wave vector 0.5a * [18], which we will refer to as q 1 , and later by single crystal X-ray diffraction [19], which proposed a 2 × 2 × 1 superstructure. Recently TEM measurements have reported another CDW vector with q=.6a * +.2b * [20] which we will refer to as q 2 . While ARPES has successfully studied other tellurides such as RTe 3 , showing its CDW phase to be FS driven, only recently has it been used for the LaTe 2 system [20], although a complete study of the CDW phase is missing.…”

“…Recently TEM measurements have reported another CDW vector with q=.6a * +.2b * [20] which we will refer to as q 2 . While ARPES has successfully studied other tellurides such as RTe 3 , showing its CDW phase to be FS driven, only recently has it been used for the LaTe 2 system [20], although a complete study of the CDW phase is missing.…”

“…The experimental FS fits remarkably well with independent LDA calculation [25] (yellow lines). Along with the outer FS (dashed lines, "outer contour"), of particular interest is a small square contour centered around the Γ point (dotted lines, 'inner diamond'), predicted by LDA but never resolved within this energy window [20]. Using photon energies closer to the Te 4p orbital binding energy (≈103eV) appears to be crucial to resolving this feature.…”

Pressure Dependence of the Charge-Density-Wave Gap in Rare-Earth Tritellurides

“…First of all, from the β-value β = 2π 5 k B ( kB vs ) 3 of the phononic part of the specific heat in LaTe 3 [9,15] one achieves the sound velocity v s = 1923 m/s [16]. Knowing that B 0 = ρv 2 s , ρ=6837 kg/m 3 being the density, one gets B 0 = 25 GPa.…”