The dependence of the spin pumping efficiency and the spin mixing conductance on the surface processing of yttrium iron garnet (YIG) before the platinum (Pt) deposition has been investigated quantitatively. The ferromagnetic resonance driven spin pumping injects a spin polarized current into the Pt layer, which is transformed into an electromotive force by the inverse spin Hall effect. Our experiments show that the spin pumping effect indeed strongly depends on the YIG/Pt interface condition. We measure an enhancement of the inverse spin Hall voltage and the spin mixing conductance by more than two orders of magnitude with improved sample preparation.

The dependence of the spin-pumping effect on the yttrium iron garnet (Y 3 Fe 5 O 12 , YIG) thickness detected by the inverse spin Hall effect (ISHE) has been investigated quantitatively. Due to the spin-pumping effect driven by the magnetization precession in the ferrimagnetic insulator Y 3 Fe 5 O 12 film a spin-polarized electron current is injected into the Pt layer. This spin current is transformed into electrical charge current by means of the ISHE. An increase of the ISHE voltage with increasing film thickness is observed and compared to the theoretically expected behavior. The effective damping parameter of the YIG/Pt samples is found to be enhanced with decreasing Y 3 Fe 5 O 12 film thickness. The investigated samples exhibit a spin mixing conductance of g ↑↓ eff = (3.87 ± 0.21) × 10 18 m −2 and a spin Hall angle between θ ISHE = 0.013 ± 0.001 and 0.045 ± 0.004 depending on the used spin-diffusion length. Furthermore, the influence of nonlinear effects on the generated voltage and on the Gilbert damping parameter at high excitation powers is revealed. It is shown that for small YIG film thicknesses a broadening of the linewidth due to nonlinear effects at high excitation powers is suppressed because of a lack of nonlinear multimagnon scattering channels. We have found that the variation of the spin-pumping efficiency for thick YIG samples exhibiting pronounced nonlinear effects is much smaller than the nonlinear enhancement of the damping.

No abstract

The fundamental phenomenon of Bose-Einstein Condensation (BEC) has been observed in different systems of real and quasi-particles. The condensation of real particles is achieved through a major reduction in temperature while for quasi-particles a mechanism of external injection of bosons by irradiation is required. Here, we present a novel and universal approach to enable BEC of quasi-particles and to corroborate it experimentally by using magnons as the Bose-particle model system. The critical point to this approach is the introduction of a disequilibrium of magnons with the phonon bath. After heating to an elevated temperature, a sudden decrease in the temperature of the phonons, which is approximately instant on the time scales of the magnon system, results in a large excess of incoherent magnons. The consequent spectral redistribution of these magnons triggers the Bose-Einstein condensation.Bosons are particles of integer spin that allow for the fundamental quantum effect of Bose-Einstein Condensation (BEC), which manifests itself in the formation of a macroscopic coherent state in an otherwise incoherent, thermalized many-particle system. The phenomenon of BEC was originally predicted for an ideal gas by Albert Einstein in 1924 based on the theory developed by Satyendra Nath Bose. Nowadays, Bose-Einstein condensates are investigated experimentally in a variety of different systems which includes real particles such as ultra-cold gases (1, 2) as well as quasi-particles with the likes of exciton-polaritons (3, 4), photons (5, 6) or magnons (7-9). The phenomenon can be reached by a major decrease in the system temperature or by an increase in the particle density. In order to condensate atomic gases, extremely low temperatures on the order of mK are required since the density of such gases must be very low to prevent their cohesion. In contrast, the quasi-stationary cooling of a quasi-particle system is accompanied by a decrease in its population and prevents BEC. Thus, an artificial injection of bosons is required to reach the threshold for BEC. Since quasi-particle systems allow for high

We carried out a concerted effort to determine the absolute sign of the inverse spin Hall effect voltage generated by spin currents injected into a normal metal. We focus on yttrium iron garnet (YIG)|platinum bilayers at room temperature, generating spin currents by microwaves and temperature gradients. We find consistent results for different samples and measurement setups that agree with theory. We suggest a right-hand-rule to define a positive spin Hall angle corresponding to the voltage expected for the simple case of scattering of free electrons from repulsive Coulomb charges.The bon mot that the sign is the most difficult concept in physics since there are no approximate methods to determine it has been ascribed to Wolfgang Pauli. Indeed, the struggle to obtain correct signs permeates all of physics. While the negative sign of the electron charge is just a convention, that of derived properties, such as the (conventional) Hall voltage, has real physical meaning. Often it is much easier and sufficient to determine sign differences between related quantities. However, a complete understanding requires not only the relative but also the absolute sign. Here we address the sign of the (inverse) spin Hall effect [(I)SHE] [1-10] and related phenomena. The characteristic parameter is the spin Hall angle, defined as the ratio θ SH ∝ J s /J c of the transverse spin current J s caused by an applied charge current J c (a more precise definition is given below). The sign of θ SH may differ for different materials. Since the spin Hall angle for Pt is generally taken to be positive, θ SH of Mo [11], Ta [12], and W [13] must be negative.The sign of θ SH governs the direction of the spin transfer torque on a magnetic contact relative to that caused by the Oersted magnetic field induced by the same current J c [12,13]. It also determines the sign of the induced transverse voltage in experiments in which the ISHE is used to detect spin currents [7]. This technique is now widely used to study spin current injection by a magnetic contact, through "spin pumping" induced by ferromagnetic resonance (FMR) [11,[14][15][16][17][18] or by temperature differences [19][20][21][22][23] ("spin Seebeck effect", SSE). * michael.schreier@wmi.badw.de However, the pitfalls that can affect the determination of the sign of the θ SH , such as the sign of the spin currents [24] and magnetic field direction are often glossed over in experimental and theoretical papers. Moreover, a mechanism for a sign reversal of the longitudinal spin Seebeck effect has recently been proposed [25]. A careful analysis of experimental results with respect to the signs of of FMR and thermal spin pumping voltages generated by the inverse spin Hall effect is therefore overdue.In this letter we present the results of a concerted action to resolve the sign issue by comparing experiments on microwave-induced spin pumping and spin Seebeck effect for a bilayer of the magnetic insulator yttrium iron garnet (YIG) and platinum (Pt) at room temperature. Samples grown by differe...