In ferromagnetic thin films, broken inversion symmetry and spin-orbit coupling give rise to interfacial Dzyaloshinskii-Moriya interactions. Analytic expressions for spin-wave properties show that the interfacial Dzyaloshinskii-Moriya interaction leads to non-reciprocal spin-wave propagation, i.e. different properties for spin waves propagating in opposite directions. In favorable situations, it can increase the spin-wave attenuation length. Comparing measured spin wave properties in ferromagnet|normal metal bilayers and other artificial layered structures with these calculations can provide a useful characterization of the interfacial Dzyaloshinskii-Moriya interactions.

Ultrathin magnetic systems have properties qualitatively different from their thicker counterparts, implying that different physics governs their properties. We demonstrate that various such properties can be explained naturally by the Rashba spin-orbit coupling in ultrathin magnetic systems. This work will be valuable for the development of next generation spintronic devices based on ultrathin magnetic systems.Electric control of magnetic systems carries high potential towards device applications [1,2] such as magnetic memory and logic. Spin-transfer torque (STT) [3,4] is an efficient way to achieve the electric control of magnetic nanostructures. In view of device applications, magnetic nanostructures such MgO-based magnetic tunnel junctions are superior to silicon-based nanostructures in the simultaneous realization of nonvolatility and speed, but are estimated to require 100 times more energy [5] than silicon-based CMOS devices to write an information bit. This energy cost problem limits scope of device applications based on magnetic nanostructures. Since the writing energy decreases as a magnetic layer in magnetic nanostructures becomes thinner [2], properties of ultrathin magnetic layers are under intense investigation [6].While the magnetization switching for the information writing is conventionally achieved by a current perpendicular to a magnetic layer, a recent experiment [7] found that an in-plane current can also switch the uniform magnetization of an ultrathin (≈ 1 nm) magnetic layer (Co) sandwiched between a heavy metal layer (Pt) and an oxide layer (AlO x ) ( Fig. 1). Since the cross-sectional area (in yz-plane) for the in-plane current can be orders of magnitude smaller than the cross-sectional area (in xy-plane) for the perpendicular current, this alternative switching scheme may reduce the current required for the switching and the switching energy. It was also reported that the magnetic domain wall (DW) in the ultrathin magnetic layer moves as fast as 400 m/s [8] when in-plane current is supplied. This velocity is about 4 times higher than the highest velocity reported for thicker magnetic layers [9]. Thus the in-plane current effects on ultrathin magnetic systems open an attractive alternative path towards powerful spintronic devices.Ultrathin magnetic systems are interesting in view of fundamental science as well. Various features of ultrathin magnetic systems cannot be explained by existing theoretical knowledge learned from measurements on thicker counterparts; for example (i) magnetization switching by in-plane current [7] instead of perpendic-Oxid e Ferr oma gnet ic Meta l Hea vy Meta l Current x y z FIG. 1: (color online) Schematic structure of ultrathin magnetic nanostructure, where an ultrathin (≈ 1 nm) ferromagnetic layer is sandwiched between a heavy metallic layer and an insulating oxide layer. Examples include Pt/Co/AlOx and Ta/CoFeB/MgO. ular current, (ii) DW motion against the electron flow direction [8, 10] instead of along it, and (iii) anomalously high DW speeds [8]. These an...

We report a time-resolved propagating spin wave spectroscopy for Fe19Ni81 film. We show that the amplitude of the spin-wave packet depends on the direction of magnetization and that its phase can be controlled by the polarity of pulsed magnetic field for the excitation. The nonreciprocal emission of spin-wave packet can be utilized for the binary spin-wave input into the spin-wave logic circuit.

The current-induced modification of the attenuation of a propagating spin wave in a magnetic nanowire is studied theoretically and numerically. The attenuation length of spin wave can increase when the spin waves and electrons move in the same direction. It is directly affected by the nonadiabaticity of the spin-transfer torque and thus can be used to estimate the nonadiabaticity. When the nonadiabatic spin torque is sufficiently large, the attenuation length becomes negative, resulting in the amplification of spin waves.

The performance of spintronic devices critically depends on three material parameters, namely, the spin polarization in the current (P), the intrinsic Gilbert damping (α), and the coefficient of the nonadiabatic spin transfer torque (β). However, there has been no method to determine these crucial material parameters in a self-contained manner. Here we show that P, α, and β can be simultaneously determined by performing a single series of time-domain measurements of current-induced spin wave dynamics in a ferromagnetic film.

A magnetic vortex wall motion driven by propagating spin waves in a nanostrip is investigated by means of micromagnetic simulation. Propagating spin waves can drive a vortex wall into a stream motion in spite of its complex internal spin structure. Compared to the transverse wall, the vortex wall moves faster and its velocity is less sensitive to the spin wave frequency. The amplitude of spin waves changes when passing through the domain wall, closely related to the domain wall velocity. This domain-wall-type-specific study provides important information for developing the theory of the interaction between domain wall and spin waves.

Using micromagnetic simulations, the authors investigate the effects of the shape anisotropy of nanowires on the threshold current density (JC) in the adiabatic limit. It is observed that the type of domain wall significantly affects the dependence of JC on the shape anisotropy. For a transverse wall, JC is proportional to the shape anisotropy, but slightly deviates from the theoretical prediction [G. Tatara and H. Kohno, Phys. Rev. Lett. 92, 086601 (2004)] possibly due to antivortex formation. For a vortex wall, JC is almost independent of the shape anisotropy, as experimentally observed [A. Yamaguchi et al., Jpn. J. Appl. Phys. Part 1 45, 3850 (2006)]. Interestingly, JC and the velocity of the vortex wall at finite temperatures are in good agreement with the experimental values even when the nonadiabatic spin torque is not considered.

No abstract