We use microscopic linear response theory to derive a set of equations that provide a complete description of coupled spin and charge diffusive transport in a two-dimensional electron gas (2DEG) with the Rashba spin-orbit (SO) interaction. These equations capture a number of interrelated effects including spin accumulation and diffusion, Dyakonov-Perel spin relaxation, magnetoelectric, and spin-galvanic effects. They can be used under very general circumstances to model transport experiments in 2DEG systems that involve either electrical or optical spin injection. We comment on the relationship between these equations and the exact spin and charge density operator equations of motion. As an example of the application of our equations, we consider a simple electrical spin injection experiment and show that a voltage will develop between two ferromagnetic contacts if a spin-polarized current is injected into a 2DEG, that depends on the relative magnetization orientation of the contacts. This voltage is present even when the separation between the contacts is larger than the spin diffusion length.

No abstract

We present a microscopic treatment of current-induced torques and thermal fluctuations in itinerant ferromagnets based on a functional formulation of the Keldysh formalism. We find that the nonequilibrium magnetization dynamics is governed by a stochastic Landau-Lifschitz-Gilbert equation with spin transfer torques. We calculate the Gilbert damping parameter α and the non-adiabatic spin transfer torque parameter β for a model ferromagnet. We find that β = α, in agreement with the results obtained using imaginary-time methods of Kohno, Tatara and Shibata [J. Phys. Soc. Japan 75, 113706 (2006)]. We comment on the relationship between s − d and isotropic-Stoner toy models of ferromagnetism and more realistic density-functional-theory models, and on the implications of these relationships for predictions of the β/α ratio which plays a central role in domain wall motion. Only for a single-parabolic-band isotropic-Stoner model with an exchange splitting that is small compared to the Fermi energy does β/α approach one. In addition, our microscopic formalism incorporates naturally the fluctuations needed in a nonzero-temperature description of the magnetization. We find that to first order in the applied electric field, the usual form of thermal fluctuations via a phenomenological stochastic magnetic field holds.

An electrical current can transfer spin angular momentum to a ferromagnet 1-3 . This novel physical phenomenon, called spin transfer, offers unprecedented spatial and temporal control over the magnetic state of a ferromagnet and has tremendous potential in a broad range of technologies, including magnetic memory and recording. Recently, it has been predicted 4 that spin transfer is not limited to ferromagnets, but can also occur in antiferromagnetic materials and even be stronger under some conditions. In this paper we demonstrate transfer of spin angular momentum across an interface between ferromagnetic and antiferromagnetic metals. The spin transfer is mediated by an electrical current of high density (~10 12 A/m 2 ) and revealed by variation in the exchange bias at the ferromagnet/antiferromagnet interface. We find that, depending on the polarity of the electrical current flowing across the interface, the strength of the exchange bias can either increase or decrease. This finding is explained by the theoretical prediction that a spin polarized current generates a torque on magnetic moments in the antiferromagnet. Current-mediated variation of exchange bias can be used to control the magnetic state of spin-valve devices, e.g., in magnetic memory applications.Spin valves 5 are now used in magnetic field sensors, in read heads for hard drives, in galvanic isolators, and in non-volatile random access memory devices. The simplest type of spin valve consists of two ferromagnetic layers separated by a thin nonmagnetic spacer. The spin-valve resistance is smallest when the magnetizations of the two ferromagnetic layers are parallel and largest when the magnetizations are antiparallel. The antiparallel alignment is achieved by making the two layers respond differently to an external magnetic field; an antiferromagnet in contact with one of the layers is used to effectively 'pin' the magnetization in this layer through an effect called 'exchange bias' [6][7][8] . The exceptional responsiveness of spin valves to magnetic fields has enabled very high areal packing densities in hard drives. In our experiments we study how exchange bias behaves when extremely high current densities are driven across these spin valve structures using point contacts.Point contacts were instrumental both for the original observation of spin transfer in ferromagnetic materials 3 and in probing the high-frequency manifestation of this phenomenon 9-11 . The extremely small size, less than a trillionth of a square cm, qualifies point contact as the smallest probe of spin transfer phenomena today and enables current densities up to 10 13 A/m 2 . Our point contacts were made with a standard system 3, 12 , using a sharpened Cu wire and a differential screw mechanism to move the Cu tip toward a FeMn/CoFe/Cu/CoFe spin valve structure. The spin-valve structures were sputtered onto Si substrates with individual layer thicknesses from 3-10 nm using techniques already

We discuss the influence of a uniform current j ជ on the magnetization dynamics of a ferromagnetic metal. We find that the magnon energy ⑀(q ជ ) has a current-induced contribution proportional to q ជ •J ជ , where J ជ is the spin current, and predict that collective dynamics will be more strongly damped at finite j ជ . We obtain similar results for models with and without local moment participation in the magnetic order. For transition metal ferromagnets, we estimate that the uniform magnetic state will be destabilized for jտ10 9 A cm Ϫ2 . We discuss the relationship of this effect to the spin-torque effects that alter magnetization dynamics in inhomogeneous magnetic systems.

We study the spin waves of the triangular skyrmion crystal that emerges in a two-dimensional spin lattice model as a result of the competition between Heisenberg exchange, Dzyalonshinkii-Moriya interactions, Zeeman coupling and uniaxial anisotropy. The calculated spin wave bands have a finite Berry curvature that, in some cases, leads to non-zero Chern numbers, making this system topologically distinct from conventional magnonic systems. We compute the edge spin-waves, expected from the bulk-boundary correspondence principle, and show that they are chiral, which makes them immune to elastic backscattering. Our results illustrate how topological phases can occur in self-generated emergent superlattices at the mesoscale.

A new MRI method is proposed for separately quantifying the two principal forms of tissue storage (nonheme) iron: ferritin iron, a dispersed, soluble fraction that can be rapidly mobilized, and hemosiderin iron, an aggregated, insoluble fraction that serves as a long-term reserve. The method utilizes multiple spin echo sequences, exploiting the fact that aggregated iron can induce nonmonoexponential signal decay for multiple spin echo sequences. The method is validated in vitro for agarose phantoms, simulating dispersed iron with manganese chloride, and aggregated iron with iron oxide microspheres. To demonstrate feasibility for human studies, preliminary in vivo data from two healthy controls and six patients with transfusional iron overload are presented. For both phantoms and human subjects, conventional R 2 and R 2 * relaxation rates are also measured in order to contrast the proposed method with established MRI iron quantification techniques. Quantification of dispersed (ferritin-like) iron may provide a new means of monitoring the risk of iron-induced toxicity in patients with iron overload and, together with quantification of aggregated (hemosiderin-like) iron, improve the accuracy of estimates for total storage iron.

Starting from the stochastic Landau-Lifschitz-Gilbert equation, we derive Langevin equations that describe the nonzero-temperature dynamics of a rigid domain wall. We derive an expression for the average drift velocity of the domain wall ṙ dw as a function of the applied current, and find qualitative agreement with recent magnetic semiconductor experiments. Our model implies that at any nonzero temperature ṙ dw initially varies linearly with current, even in the absence of non-adiabatic spin torques.