Elsevier

Superlattices and Microstructures

Volume 41, Issues 2–3, February–March 2007, Pages 104-108
Superlattices and Microstructures

Polarized neutron reflectometry study on a magnetic film with an ion beam imprinted stripe pattern

https://doi.org/10.1016/j.spmi.2007.02.003Get rights and content

Abstract

We used polarized neutron reflectometry for a quantitative study of the field dependent arrangement of the magnetization vector in the different regions of an ion beam imprinted stripe pattern in a magnetic film. For the magnetic patterning of the Co70Fe30 film we took advantage of the exchange bias to an antiferromagnetic Mn83Ir17 layer which was changed locally by He-ion bombardment. The exchange bias was set to be antiparallel in the two different striped regions. We found that after magnetization reversal of half of the stripes the magnetization in neighbouring regions is periodically canted with respect to the stripe axis so that the net magnetization of the ferromagnetic film turns almost perpendicular to the stripes. At the same time the projection of the magnetization vector onto the stripe axis has a periodically alternating sign.

Introduction

High package densities in magnetic recording media can be achieved by scaling the magnetic elements to smaller sizes of magnetic grains, clusters or structures patterned by lithographic methods. However, approaching very small feature sizes, the superparamagnetic limit may be reached where the long-term thermal stability of the magnetic elements is lost [1]. To avoid this problem patterning of the recording media into magnetically separated areas can be performed. It may be realized either by planar magnetic patterning, by using lithography methods [2], or by direct fabrication of topographically isolated, single-domain magnetic nanostructures [3]. We used ion bombardment induced magnetic patterning (IBMP) for changing the exchange-bias (EB) [4] between ferromagnetic (F) and antiferromagnetic (AF) layers in one part of the film, while keeping the properties of the untreated regions unchanged. For the resulting configuration of the magnetization in the sample both, the interfacial exchange bias properties but also intralayer exchange between the moments in the different regions have to be considered. For a detailed study of the magnetic interactions between neighbouring magnetic elements in a continuous film we designed a model system which allowed the efficient use of different experimental techniques for complementary analysis. Therefore, we did not approach very small length scales but designed our magnetic elements to be in the micrometer range and used methods as the magneto-optical Kerr effect (MOKE), Kerr microscopy (KM) and polarized neutron reflectometry (PNR). We have chosen a film with imprinted in-plane stripe-like magnetic domains. The stripes are expected to display alternating magnetization directions in the remanent demagnetized state.

Section snippets

Experimental

The sample studied is a EB Co70Fe30(28 nm)/Mn83Ir17(15 nm) F/AF bilayer with buffer layers Cu(28 nm)/SiO2(50 nm) on Si(111) and a TaO-Ta(9 nm) top layer prepared by magnetron sputtering. The initial EB direction was set by field cooling in a magnetic in-plane field of 1 kOe after annealing at 275 C which is above the blocking temperature. Subsequently the film was covered by a photoresist, which was patterned into equally spaced stripes with a width of d=2.5 μm and a periodicity of Λ=5 μm.

Results and discussion

KM results of the present sample are published recently [5]. It was shown that starting from saturation magnetization the reversal proceeds through an antiparallel alignment of magnetization. Here we present in Fig. 1 one KM image taken at a field at which the magnetization in neighboring regions is in this state. The longitudinal MOKE hysteresis loop along the EB axis together with results of fits to specular PNR data is displayed Fig. 1. A two-step reversal, corresponding to the two regions

Acknowledgements

We gratefully acknowledge financial support by the DFG via SFB 491 and BMBF O3ZA6BC1.

References (6)

  • A. Moser et al.

    J. Phys. D

    (2003)
  • C. Chappert et al.

    Science

    (1998)
  • M. Albrecht et al.

    Appl. Phys. Lett.

    (2002)
There are more references available in the full text version of this article.
View full text