A schematic diagram of the experimental geometry used in the SW/wedge (swedge) method, showing the ML mirror that is used as a SW generator, on top of which has been grown a Cr wedge, a Fe layer of constant thickness, and finally a protective alumina cap. The SW modulation of the electric field is pinned in phase to the top of the ML as the position of the x-ray beam is scanned in . Due to the wedge profile of the Cr, this scan in also scans the SW through the layers and interfaces above the wedge. The Fe is magnetized in the direction, and depth relative to the surface is defined as . In addition to rocking curve measurements, is varied over the first-order Bragg angle , with the angle between incident light and outgoing detected x rays being fixed.
XMCD of the Fe region obtained by total-electron-yield detection: (a) The individual intensities of left and right circularly polarized (LCP and RCP) light, as well as (b) the XMCD curve derived from them. The weakness of high-energy satellites in the spectra in (a) suggests a low degree of Fe oxidation.
A typical broadscan XES/RIXS spectrum taken with an position corresponding to a thickness of the Cr wedge. LCP radiation at was used for excitation. The Fe and Cr features of principal interest are highlighted.
The Cr intensity ratio as derived via a rocking curve scan over the Bragg angle at a Cr thickness of . Both experiment and a best-fit XRO calculation are shown.
[(a) and (b)] The individual Cr and Fe intensities (as summed over LCP and RCP), as well as (c) the ratio of these intensities, Cr , as derived by a SW scan with sample motion along and fixed at the Bragg angle of 12.9°. The position has been converted to Cr wedge thickness intensities as a function of Cr thickness at the Bragg angle.
Fe x-ray RIXS magnetic circular dichroism taken at Cr thickness and an excitation energy of at the absorption resonance. Dichroism is plotted as a percentage of the peak heights.
SW scans in (or Cr thickness) of Fe x-ray emission magnetic circular dichroism: (a) total Fe intensity as measured by , (b) the quantity that is proportional to Fe magnetization, and (c) the final Fe MCD as defined in the text.
Top and bottom interface structures as derived by best fits of XRO calculations to the data in Figs. 4, 5, and 7. Interface concentration profiles are assumed to be linear. Atom-specific magnetization profiles are assumed to be Gaussian. The best-fit parameters are also indicated. The functional form of the concentration and the magnetization are shown, with each function normalized to the value in the center of the Fe layer in which there is no interdiffusion/roughness and the magnetization is constant.
XRO calculations of Fe total intensity and RIXS MCD in which parameters have been systematically varied around the optimum fit to the experimental data: (a) only the thickness of the interface is varied, with the ratio also constant; (b) only the thickness of the interface is varied, with the ratio also constant. All parameters are given in angstroms.
X-ray emission MCD measurements as a function of Cr thickness are compared to theoretical calculations for various choices of parameters, including the best fit as well as several other calculations used to illustrate the sensitivity of the fits. The calculations include upward and downward variations in the thickness of the top and bottom interfaces, as well as extreme limits in which the atom-specific magnetization is constant through both interfaces (e.g., with and ) and the MCD becomes constant in the SW scan.
Article metrics loading...
Full text loading...