Abstract
The exchange interaction and the resulting exchange bias between ferromagnet and antiferromagnet layers are fundamental to the operation of magnetic data storage devices. Despite considerable work, and the development of many theoretical models, quantitative agreement between theory and experiment is still poor. Here we report a systematic comparison between experiments on Ni81Fe19epitaxial Cr2O3 bilayers and a phenomenological model of the effect which can fit experimental data and enables key ingredients of previous rigorous models to be applied to experimental data. Our results show unambiguously that the uniaxial anisotropy which leads to enhanced coercivity in most exchange-coupled systems and the exchange bias have fundamentally independent origins and that only a very restricted subset of possible crystal orientations can result in a large exchange bias. These factors are masked in experiments on polycrystalline systems and on rough interfaces in single-crystalline systems.
Original language | English |
---|---|
Article number | 224421 |
Journal | Physical Review B - Condensed Matter and Materials Physics |
Volume | 72 |
Issue number | 22 |
DOIs | |
State | Published - 1 Dec 2005 |