|Direct Imaging of Asymmetric Magnetization Reversal|
The phenomenon of exchange bias has transformed how data is read on magnetic hard disks and created an explosion in their information storage density. However, it remains poorly understood, and even the fundamental mechanism of magnetic reversal for exchange-biased systems in changing magnetic fields is unclear. By using x-ray photoemission electron microscopy at the ALS to directly image the magnetic structure of an exchange-biased film, a team from the University of Washington and the Stanford Synchrotron Radiation Laboratory has identified separate magnetic-reversal mechanisms in the two branches of a hysteresis loop. This advance in fundamental understanding will provide new insights for developing the next generation of information storage and sensing devices where exchange bias is expected to play a critical role.
There are two basic energies involved in the manipulation and control of the magnetic properties of materials. Exchange controls magnetic order, and anisotropy controls magnetic orientation. A soft ferromagnet such as iron has a large exchange parameter but a small anisotropy, making ferromagnetic order stable at higher temperatures but with an unpredictable orientation of the magnetization, especially in structures of nanoscale dimensions. On the other hand, many antiferromagnets have weak exchange interactions (low ordering or Néel temperatures) but large anisotropies that result in very stable orientations.
Exchange bias arises when a thin ferromagnetic film is grown on an antiferromagnet and the resulting heterostructure is cooled in a magnetic field through the Néel temperature of the antiferromagnet. As a result of exchange coupling between the layers, the ferromagnet both retains a stable order and gains a higher anisotropy at room temperature. Moreover, the unidirectional character of the anisotropy results in a shifted hysteresis loop that is now centered on a non-zero magnetic field. This exchange bias makes the ferromagnet an excellent magnetic reference layer in modern nanolayer magnetic devices because it is very difficult to demagnetize it.
More than fifty years of research has provided varying insight into the exchange-bias phenomenon but not yet a comprehensive description of all its salient features. To gain more insight, the Washington–Stanford team resorted to x-ray photoemission electron microscopy (PEEM) imaging of high-quality single-crystal ferromagnetic iron epitaxially grown on antiferromagnetic MnPd (all on an MgO substrate), samples that had been previously well-characterized magnetically and structurally.
At an iron absorption resonance, absorption of circularly polarized x rays at ALS Beamline 184.108.40.206 is sensitive to the angle between the magnetization within a ferromagnetic domain and the polarization vector. With the PEEM-2 microscope, this x-ray magnetic circular dichroism (XMCD) effect allows an exact determination of the direction of the local domain magnetization at the surface of ferromagnets with a spatial resolution of 50 nm or less.
By means of XMCD measurements taken at points in hysteresis loops with the applied field in different crystallographic directions of the iron ferromagnet, the team has accumulated the first direct imaging evidence for an asymmetry in the magnetic-reversal mechanism in exchange-biased systems, evidence that until now has only been inferred indirectly by measurements such as neutron scattering.
Normally, magnetic reversal in ferromagnets occurs either by coherent rotation of magnetic moments in the domain or by nucleation and growth of reverse domains. Generally, the mechanism is determined by the material microstructure and is symmetric with respect to the applied field, i.e., it is the same in both branches of the hysteresis loop. However, the team found that in exchanged-biased ferromagnetic iron, the magnetization reversal occurs by moment rotation for decreasing fields, while it proceeds by domain nucleation and growth for increasing fields. The observed domains are also consistent with the crystallography of the bilayers and favor a configuration that minimizes the overall magnetostatic energy of the ferromagnetic layer.
Research conducted by P. Blomqvist and K.M. Krishnan (University of Washington) and H. Ohldag (Stanford Synchrotron Radiation Laboratory).
Research funding: U.S. Department of Energy, Office of Basic Energy Sciences (BES). Operation of the ALS is supported by BES.
Publication about this research: P. Blomqvist, K.M. Krishnan, and H. Ohldag, "Direct imaging of asymmetric magnetization reversal in exchange-biased Fe/MnPd bilayers by x-ray photoemission electron microscopy," Phys. Rev. Lett. 94, 107203 (2005).