The mechanisms of the magnetization switching of magnetic multilayers driven by a current are studied by including exchange interaction between local moments and spin accumulation of conduction electrons. It is found that this exchange interaction leads to two additional terms in the Landau-Lifshitz-Gilbert equation: an effective field and a spin torque. Both terms are proportional to the transverse spin accumulation and have comparable magnitudes.
Domain walls in ferromagnetic metals are known to be a source of resistance since the early experiments on iron whiskers. Recently it has been possible to identify this contribution from data on cobalt and nickel films which display stripe domains in which the current is driven normal to the domain walls. With the same Hamiltonian as used to explain giant magnetoresistance in structures with collinear magnetic alignments we have determined the spin flip, as well as nonflip, scattering present in domain walls. We calculate the resistivity in zero field, i.e., in the presence of striped domains, and at saturation to show the amount of magnetoresistance that is attributable to domain wall scattering.[S0031-9007(97)04820-5] PACS numbers: 73.40.Cg, 73.50.Bk, Since the early experiments on iron whiskers [1], it was recognized that walls between domains in ferromagnets are a source of electrical resistance in addition to that present in the domains. By applying magnetic fields to saturate magnetization, and thereby erase domains in an otherwise multidomain ferromagnet, the resistance of iron was found to drop significantly. Cabrera and Falicov [2] considered two mechanisms by which the walls produce additional scattering: paramagnetic effects from the reflection of incoming electron waves from the ferromagnetically ordered domains as they enter the twisted spin structure of a wall, and diamagnetic effects "due to the zigzagging character of the electron orbits when going between up and down regions of the magnetization." In metals with few impurities, such as iron whiskers, the relaxation times are sufficiently long, and Cabrera and Falicov concluded that these diamagnetic effects can account for the observed negative magnetoresistance (MR) [3]. Indeed, both Cabrera and Falicov, and Berger [4], found that the reflection for an electron at the Fermi surface from a 180 ± wall which is 10 nm thick is negligible; unless the walls are so abrupt to be of monolayer thickness, electron scattering from reflections do not contribute to the electrical resistance.While the resistivity of iron whiskers at low temperatures (about 4.2 K) is small, of the order of 0.01 mV cm, most of the ferromagnetic layers that are used in magnetoresistive elements, e.g., spin valves, have a considerably higher resistivity due to inherent defects such as grain boundaries and roughness at the boundaries of these thin layers. Recently Gregg et al. [5] obtained the first direct observation of ferromagnetic domain wall scattering by passing a current that is nominally perpendicular to a striped domain structure in thin films of cobalt (1000 A). The resistivity of their films was about 10 mV cm, so that the diamagnetic effects that are putatively responsible for the negative MR in pure iron whisker cannot be responsible. Also, as the thickness of the walls is estimated to be 15 nm, reflections of impinging electrons with the Fermi wavelength cannot account for the additional resis-tivity due to domain walls which was estimated to be about 0.52 mV cm. Greg...
Zhang et al. Reply: The preceding Comment [1] uses an incorrect set of equilibrium states to analyze transport and arrives at a wrong interpretation of the spinor dispersion relation used in our Letter [2].The spinor energy-momentum dispersion introduced in our Letter [see Eq. (1)], or Eq. (2) of the Comment,
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.