No Arabic abstract
As a non-magnetic heavy metal is attached to a ferromagnet, a vertically flowing heat-driven spin current is converted to a transverse electric voltage, which is known as the longitudinal spin Seebeck effect (SSE). If the ferromagnet is a metal, this voltage is also accompanied by voltages from two other sources, i.e. the anomalous Nernst effect in both the ferromagnet and the proximity-induced ferromagnetic boundary layer. By properly identifying and carefully separating those different effects, we find that in this pure spin current circuit the additional spin current drawn by the heavy metal generates another significant voltage by the ferromagnetic metal itself which should be present in all relevant experiments.
We present a semiclassical theory of spin-diffusion in a ferromagnetic metal subject to a temperature gradient. Spin-flip scattering can generate pure thermal spin currents by short-circuiting spin channels while suppressing spin accumulations. A thermally induced spin density is locally generated when the energy dependence of the density of states is spin polarized.
We present a theoretical analysis of current driven ferromagnetic resonance in a ferromagnet/normal-metal/ferromagnet tri-layer. This method of driving ferromagnetic resonance was recently realized experimentally by Tulapurkar et al. [Nature 438, 339 (2005)] and Sankey et al. [Phys. Rev. Lett. 96, 227601 (2006)]. The precessing magnetization rectifies the alternating current applied to drive the ferromagnetic resonance and leads to the generation of a dc voltage. Our analysis shows that a second mechanism to generate a dc voltage, rectification of spin currents emitted by the precessing magnetization, has a contribution to the dc voltage that is of approximately equal size for the thin ferromagnetic films used in the experiment.
We investigate electron cooling based on a clean normal-metal/spin-filter/superconductor junction. Due to the suppression of the Andreev reflection by the spin-filter effect, the cooling power of the system is found to be extremely higher than that for conventional normal-metal/nonmagnetic-insulator/superconductor coolers. Therefore we can extract large amount of heat from normal metals. Our results strongly indicate the practical usefulness of the spin-filter effect for cooling detectors, sensors, and quantum bits.
Efficient spin/charge interconversion is desired to develop innovative spin-based devices. So far, the interconversion has been performed by using heavy atomic elements, strong spin-orbit interaction of which realizes the interconversion through the spin Hall effect and the Edelstein effect. We demonstrate highly efficient charge-to-spin conversion in a ferromagnetic metal/Cu/Al2O3 trilayers, which do not contain any heavy element. The resulting spin torque efficiency is higher than those of conventional spin Hall and Rashba systems consisting of heavy elements such as Pt and Bi. Our experimental results qualitatively deviate from typical behaviors arising from spin transport. However, they are surprisingly consistent with the behaviors arising from the orbital transport. Our results thus demonstrate a new direction for efficient charge-to-spin conversion through the orbital transport.
Studies of thermally induced transport in nanostructures provide access to an exciting regime where fluctuations are relevant, enabling the investigation of fundamental thermodynamic concepts and the realization of thermal energy harvesters. We study a serial double quantum dot formed in an InAs/InP nanowire coupled to two electron reservoirs. By means of a specially designed local metallic joule-heater, the temperature of the phonon bath in the vicinity of the double quantum dot can be enhanced. This results in phonon-assisted transport, enabling the conversion of local heat into electrical power in a nano-sized heat engine. Simultaneously, the electron temperatures of the reservoirs are affected, resulting in conventional thermoelectric transport. By detailed modelling and experimentally tuning the interdot coupling we disentangle both effects. Furthermore, we show that phonon-assisted transport gives access to the energy of excited states. Our findings demonstrate the versatility of our design to study fluctuations and fundamental nanothermodynamics.