Absence of spin scattering of in-plane spring domain walls

José L. Prieto, Bas B. van Aken, José I. Martín, A. Pérez-Junquera, Gavin Burnell, Neil D. Mathur, and Mark G. Blamire

Phys. Rev. B 71, 214428 – Published 29 June 2005

Abstract

We have performed current perpendicular-to-plane measurements in Gd-transition metal multilayers and amorphous ${Co}_{x} {Gd}_{1 - x} ∕ {Co}_{y} {Gd}_{1 - y}$ bilayers. Both systems create in-plane exchange spring domain walls (SWs) whose width can be controlled by an external magnetic field. The results can be explained fully by Lorentz magnetoresistance (MR) plus the effect of the interface. They show no intrinsic scattering from the SWs, suggesting that the “mistracking” of the spin to the local magnetization is not the main cause of MR of domain walls but the change of spin populations across a thin barrier as in GMR structures.

Received 16 February 2005

DOI:https://doi.org/10.1103/PhysRevB.71.214428

Authors & Affiliations

José L. Prieto^1,*, Bas B. van Aken¹, José I. Martín², A. Pérez-Junquera², Gavin Burnell¹, Neil D. Mathur¹, and Mark G. Blamire¹

¹Department of Materials Science and Metallurgy, University of Cambridge, Pembroke Street, Cambridge CB2 3QZ, United Kingdom
²Departamento Física, Facultad de Ciencias, Universidad de Oviedo, c/ Calvo Sotelo s/n, 33007 Oviedo, Spain

^*Electronic address: joseluis.prieto@upm.es

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 71, Iss. 21 — 1 June 2005

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Transport measurements for the samples with Gd and different transition metals. (a) Measurements in which the interface is $Ge ∕ Fe$ . Inset in (a) shows a schematic drawing of the formation of DWs in both Gd and Fe layers when a magnetic field is applied. (b) Samples without a Fe interlayer and therefore a slightly poorer interface. Inset in (b) shows the measurement in pure Fe in order to emphasize the similarity with the samples Gd30Ni30 and Gd30Co30.Reuse & Permissions
Figure 2
The transport measurements of the sample Gd30F30 for different temperatures. Inset: resistance vs field normalized to the resistance at zero fields (extrapolated from the parabolic part of the curves) for different temperatures: (●) 10 K, (∎) 20 K, (▴) 30 K, (◻) 50 K (엯) 70 K.Reuse & Permissions
Figure 3
Behavior of the magnetization around the interface for the samples Gd30Fe30 (◆), Gd30Fe2Ni26Fe2 (▴), and Gd30Fe2Co26Fe2 (∎). The total rotation of the spring wall is the difference between the angle at the interface (in either Gd of TM depending on the case) and the angle at the center of that layer.Reuse & Permissions
Figure 4
Estimated MR obtained with the available theories and the magnetic configuration obtained from the micromagnetic simulation. (a) represents the estimated MR exclusively from the behavior of the interface with field [see Fig. 6e]. (b) represents the estimated MR exclusively from the spring DWs in the Gd and the TM layers. Gd30Fe30 (◆), Gd30Fe2Ni26Fe2 (▴), and Gd30Fe2Co26Fe2 (∎).Reuse & Permissions
Figure 5
(a) MOKE hysteresis loop of ${Gd}_{0.12} {Co}_{0.88} ∕ {Gd}_{0.29} {Co}_{0.71}$ bilayer (sample A) at 20 K; the magnetic configurations for each step in the magnetization reversal process are sketched, indicating the presence of the DW at the interface at higher fields. (b) Transport measurements for the samples ${Gd}_{x} {Co}_{1 - x} ∕ {Gd}_{y} {Co}_{1 - y}$ measured at 20 K; the total resistances at zero field are 37.4, 22.7, and $42.5 Ω$ for samples A, B, and C respectively. Inset: a visualization of the angular barrier created by the spring domain wall, separating two parallel domains with the same magnetization direction.Reuse & Permissions
Figure 6
(a) Inset shows the MOKE signal of the bottom layer of one of the GdCo samples. The samples are grown in parallel with the ones used for the transport measurements. The hysteresis loop is complementary to the one shown in Fig. 5a because MOKE is only sensitive to Co moments. The main panel is a magnification (indicated in the inset) of the saturation part where the formation of the spring wall is shown. (b) Field dependence of the width (d) of the SW deduced from the curve in (a) using the model developed in Ref. 28 for the MOKE signal, that presents a $H^{- 1 ∕ 2}$ behavior in agreement with Ref. 10. An approximate value of 4 nm for 4 T can be extrapolated.Reuse & Permissions

Physical Review B

covering condensed matter and materials physics