Magnetic Circular Dichroism in Photoemission from Rare-Earth Materials: Basic Concepts and Applications

Abstract

Magneto-optical Kerr effect and Faraday effect provide the basis of established methods for studying the magnetic properties of matter by polarized light in the visible spectral range. It was only quite recently that an analogous effect in the x-ray region, magnetic circular dichroism in x-ray absorption, was first observed by Gisela Schütz et al. for the near-edge fine structure at the K edge of ferromagnetic iron.1 Later on, magnetic circular x-ray dichroism (MCXD) was also observed at the LII,III thresholds of rare-earths2 and 3d transition metals,3 opening up the possibility for element-specific analyses of magnetic moments in compound magnets and multilayers. Today MCXD is mainly used as a tool at the LII,III x-ray absorption thresholds of 3d transition metals, where relatively large MCD asymmetries in the white lines upon reversal of either sample magnetization or circular polarization (photon spin) of the absorbed light are observed. MCXD can be understood in the simplest way in a one-electron picture by taking the spin polarization of the excited electron due to the inner-shell spin-orbit coupling (Fano effect4) into account as well as the spin-split density of final states at and above the Fermi level.5 More rigorous theoretical treatments have been given,6,7,8 which allow to recognize the three important ingredients for magnetic circular dichroism: (i) Exchange interaction as the driving force for long-range spin order; (ii) use of circularly polarized light with preferential propagation along the magnetic quantization axis; (iii) spin-orbit interaction providing the mechanism for an effective coupling between the angular momentum of the circularly polarized photon and the magnetically ordered electron spins.KeywordsCircular PolarizationMagnetic Circular DichroismExchange SplittingAntiparallel OrientationRare EarthThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.