Magnetization reversal in (0 0 1)Fe thin films studied by combining domain images and MOKE hysteresis loops

https://doi.org/10.1016/S0304-8853(99)00761-1Get rights and content

Abstract

The magnetization reversal of epitaxial single-crystal Fe films has been studied by combining domain images and hysteresis loops. The reversal is quantitatively described by combining the coherent rotation model and the domain wall displacement model. The pinning energy exerted on the domain walls and the domain wall angle at the switching fields are obtained by fitting this model to experimental hysteresis loops. The field-dependent pinning energy and the domain wall angle in the reversal process, and the contributions of second-order magneto-optic effect to hysteresis loops, are revealed to be two important features of single-crystal Fe films.

Introduction

Magnetization reversal in thin magnetic films is of continued interest, because of its importance for the study of most magnetic phenomena. Take for example interlayer coupling and its identification. It can be studied qualitatively and quantitatively via its influence on the remagnetization process. Hence, a good understanding of the remagnetization and the way it reveals itsself in the experiment in single films is a prerequisite to understand the reversal in double and multilayers where coupling can occur in addition. From the viewpoints of practical application, the importance of magnetization reversal is obvious because both giant magnetoresistance in magnetic multilayers and tunneling magnetoresistance in trilayer junctions are related to magnetization reversal. For fundamental studies, in order to get as clear conditions as possible, one would prefer homogeneous single-crystalline films with few pinning centers, i.e. small coercivity BC=μ0HC. For Fe films, BC values below 2 mT can be considered as small, but we will be dealing in the following even with BC≈0.2 mT which is achieved by strain free, single-crystalline growth.

In the past few years much attention was paid to the magnetization reversal process in epitaxial Fe thin films [1], [2], [3], [4], [5] and different magnetization reversal models were proposed [1], [2], [3]. Because the fitting based on the coherent rotation model could not reconstruct simultaneously the experimental values of the switching fields and the saturation field, Florczak [1] suggested that the magnetization reversal in single-crystal Fe films occurs through the nucleation and/or unpinning of 90° domain walls at two distinct transition fields followed by coherent rotation. Further studies indicated that the magnetization reversal process of the `two jumps’ and even `three jumps’ in the small field range could be explained by the domain wall unpinning model [2], [3]. But this model could not explain the switching behavior at high fields.

Up to now most experiments are based on domain images to study the mechanism of the magnetization reversal in detail [4], [5] and magneto-optic Kerr effect (MOKE) hysteresis loops to study the entire reversal [1], [2], [3], [4], [5]. Due to the technical difficulties associated with measuring weak magnetic contrast in finite external fields, only qualitative information can be obtained from the domain patterns. On the other hand, MOKE hysteresis loops themselves are often complicated and difficult to explain owing to a significant second-order contribution of the magneto-optic effect [6], [7]. In this paper the magnetization reversal process is studied by combining domain observations and MOKE hysteresis loops. Taking into account the second-order magneto-optic response and combining the coherent rotation model and the domain wall displacement model, we are able to reproduce the experimental MOKE hysteresis loops with high precision. As a result the field-dependent pinning energy of the domain walls and the domain wall angle are obtained.

Section snippets

Experiments

The epitaxial BCC Fe(0 0 1) films with various thicknesses from 5 to 50 nm were deposited at 200°C by MBE on a (0 0 1)Au(150 nm)/Fe(1 nm)/GaAs substrate–buffer system. On top a protective and anti-reflective 50 nm ZnS layer was deposited at room temperature.

The domain pictures were obtained by digitally enhanced magneto-optic microscope [8]. MOKE hysteresis loops were measured in the longitudinal configuration using a photodiode system. The incident light (λ=750 nm) was polarized perpendicular to the

Experimental results and theoretical model

For BCC Fe(0 0 1) films epitaxially grown on Au/Fe/GaAs substrate–buffer system, only the in-plane four-fold cubic crystal anisotropy was found by MOKE. Maybe this is due to the good matching between (0 0 1) plane of FCC Au and the (0 0 1) plane of BCC Fe after a 45° in-plane rotation. So in the following paragraphs, we only show the experimental results for the applied field in the range of 45°⩽φH⩽90° (φH is the angle between the positive direction of the external field and the [1 0 0] easy axis

Discussions

In our theoretical model of magnetization reversal, we take into account the pinning energy of the domain walls, instead of the specific properties of the defects. We also neglect the nucleation process of the domains. Even so, our theoretical model is accurate enough because the above simplification is based on the experimental results. From the domain pictures (Fig. 2) observed by magneto-optic microscope, we saw that the new domains always quickly sweep through the whole sample as soon as

Conclusions

The magnetization reversal in single-crystal Fe thin films is quantitatively described by combining the coherent rotation model and the domain wall displacement model. The pinning energy exerted to the domain walls and the domain wall angle at the switching fields are first revealed to depend on the external field. The contribution of the second-order magneto-optic effect to MOKE hysteresis loops is taken into account as a key factor in fitting.

Acknowledgements

The authors would like to thank O. de Haas, A. Bürke, D. Olligs, F. Voges, P. Röttländer, and M. Buchmeier for their help. Shi-shen Yan is pleased to thank the Alexander von Humboldt Foundation for support.

References (9)

  • K. Postava et al.

    J. Magn. Magn. Mater.

    (1997)
  • J.M. Florczak et al.

    Phys. Rev. B

    (1991)
  • R.P. Cowburn et al.

    J. Appl. Phys.

    (1995)
  • R.P. Cowburn et al.

    Phys. Rev. Lett.

    (1997)
There are more references available in the full text version of this article.

Cited by (41)

  • Anisotropy of ferromagnetism in Co-implanted rutile

    2005, Journal of Physics Condensed Matter
View all citing articles on Scopus
View full text