A computational approach to predict and enhance the sensitivity of X-ray resonant magnetic reflectometry to the magnetic behavior of deeply buried interfaces

Abstract

In the present work a computational approach is applied to model and predict the results of X-ray resonant magnetic reflectometry – a non-destructive synchrotron-based technique to probe chemical composition, crystallographic environment and magnetization in multilayer epitaxial heterostructures with nanoscale depth resolution. The discussed 2D mapping approach is a step forward with respect to conventional resonant X-ray reflectometry and consists of collecting a fine step array of reflected intensity as a function of grazing angle and photon energy across the absorption edge of a particular chemical element. With the use of circularly polarized photons the method can be extended to magnetic systems to produce a map of dichroic reflectance directly related to the magnetization profile of the heterostructure. Studying the magnetic field dependence of dichroic reflectance maps can provide valuable information on the magnetization reversal of individual sublayers of a multilayer heterostructure. In the present paper modeling is performed for a bilayer system mimicking the behavior of a 30 nm ɛ-Fe2O3 thin film that is known to exhibit a pronounced two-component magnetic hysteresis. A technique to find optimal energy/angle combinations in order to sense magnetization of individual sublayers is proposed. Also discussed is the advantage of heavy-element capping, which leads to a substantial increase of the dichroic intensity oscillation contrast in the pre-edge region where the sensitivity to the magnetic behavior of the deeply buried interfaces is most pronounced.