The electric resistance of ferromagnet/normal-metal/ferromagnet perpendicular spin valves depends on the relative angle between the magnetization directions. In contrast to common wisdom, this angular magnetoresistance is found to be not necessarily a monotone function of the angle.The parameter dependence of the global resistance minimum at finite angles is studied and the conditions for experimental observation are specified.
The current induced magnetization reversal in nanoscale spin valves is a potential alternative to magnetic field switching in magnetic memories. We show that the critical switching current can be decreased by an order of magnitude by strategically distributing the resistances in the magnetically active region of the spin valve. In addition, we simulate full switching curves and predict a new The prediction that a spin-polarized current can excite and reverse a magnetization 1,2 has been amply confirmed by recent experiments. 3,4 The current-induced magnetization dynamics is interesting as an efficient mechanism to write information into magnetic random access memories as well as to generate microwaves. 5 Unfortunately, the critical currents for magnetization reversal are still unattractively high. 6 In this letter, we apply a previously developed microscopic formalism 7 to understand the critical current in spin valves quantitatively and propose a strategy to reduce it by up to an order of magnitude. We also solve the micromagnetic equations with accurate angle-dependent magnetization torque and spin-pumping 8 terms and predict switching to a precessional state.We will first consider a generic F͑erromagnetic͉͒N͑ormal͉͒F spin valve biased by a voltage difference V. The two ferromagnetic reservoirs are assumed to be monodomain; the magnetizations differ by an angle . Charge and spin currents excited by an applied bias can be calculated accurately by magneto-electronic circuit theory 7 with parameters determined from first-principle calculations 9 that agree well with experimental data. 10 To this end we dissect the pillar into three nodes (the reservoirs and the normal metal) connected by two, not necessarily identical resistive elements G L and G R . Each of them is characterized by the conductance g = g ↑↑ + g ↓↓ , the polarization p = ͑g ↑↑ − g ↓↓ ͒ / g and the normalized mixing conductance =2g ↑↓ / g. g ↑↑ and g ↓↓ are, respectively, the conductances for electrons with spin parallel and antiparallel to the magnetization and g ↑↓ is the material parameter that governs the magnetization torque. The magnetically active region includes layers of thickness up to the spin-flip diffusion lengths from the interfaces. Any resistance outside this region is parasitic and not considered here. The conductances are effective parameters determined by the resistance of the ferromagnetic and normal metal bulk, that of the interfaces to the normal metal and the resistance of an eventual outer normal metal that fits into the magnetically active region. For simplicity we disregard the bulk resistance of the normal metal island and the imaginary part of g ↑↓ (for metallic interfaces smaller than 10% of the real part 9,11 ).The transverse component of the spin current is absorbed in the ferromagnet 11 and the associated spin-transfer torque can excite the magnetization. 1,2 Circuit theory has been used to derive analytic expressions for the torques in a symmetrical spin valve as function of the angle between the magnetization direc...
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.