Influence of the Fe content on the Gd magnetic ordering temperature in Ni${}_{1-x}$Fe${}_x$/Gd multilayers

We explore the influence of Fe content on the interfacial magnetic properties of Ni${}_{1-x}$Fe${}_x$(50 Å)/Fe ($t$ Å)/Gd(50 Å)/Fe ($t$ Å)/Ni${}_{1-x}$Fe${}_x$(50 Å) ($x$ $=$ 0.2 and 0.6, and $t$ $=$ 0 and 10 Å) multilayers by means of element-specific x-ray magnetic circular dichroism measurements at Gd M${}_5$ and (Ni, Fe) L${}_{2,3}$ absorption edges, superconducting quantum interference device magnetometry, and specular reflectivity measurements. Increasing Fe content in the Ni-Fe layer enhances the Gd ordering temperature in the interfacial regions. Addition of a 1-nm Fe spacer results in the largest enhancement in Gd ordering temperature in both interior and interfacial Gd regions, even for high Ni content in the Ni-Fe layer. This indicates that the Fe spacers act as effective barriers (albeit imperfect) in reducing Gd-Ni contact interactions responsible for limiting the interfacial Gd magnetic ordering temperature to values larger than bulk Gd, but significantly lower than observed at Fe/Gd interfaces.

Figure 1

Measured and fitted reflectivity curves at six incident x-ray energies for (a) Ni${}_8$Fe${}_2$ and (b) Ni${}_4$Fe${}_6$. Depth profile of the real part of the scattering length density (SLD) at 7224 eV for (c) Ni${}_8$Fe${}_2$ and (d) Ni${}_4$Fe${}_6$. The lines in black are the fitted SLD profiles before roughness effects are accounted for, and the lines in purple/medium gray are the roughness-convoluted SLD profile.

Figure 2

X-ray absorption near-edge structure spectra at the (a) and (b) (Ni, Fe) K edges and (c) Gd L${}_3$ edge. The black line represents the Ni${}_8$Fe${}_2$ sample, while the red/dark gray one is used for the Ni${}_8$Fe${}_2$/Fe sample.

Figure 3

(a) X-ray magnetic circular dichroism hysteresis loops in fluorescence mode measured at the Gd M${}_5$ absorption edge (1181.8 eV) at $T$ $=$ 30 K for the samples Ni${}_8$Fe${}_2$ (Δ), Ni${}_8$Fe${}_2$/Fe (▴), Ni${}_4$Fe${}_6$ (○), and Ni${}_4$Fe${}_6$/Fe (•). The loops are shifted vertically for clarity. (b) Temperature-dependent SQUID magnetometry data for the samples Ni${}_8$Fe${}_2$ (Δ), Ni${}_8$Fe${}_2$/Fe (▴), Ni${}_4$Fe${}_6$ (○), and Ni${}_4$Fe${}_6$/Fe (•). We have included as an example the $T_M$ achieved for the sample Ni${}_8$Fe${}_2$.

Figure 4

X-ray magnetic circular dichroism spectra across Gd M${}_5$, and (Ni, Fe) L${}_{2,3}$ edges at different temperatures. The asymmetry ratios are plotted as a function of the incident x-ray energy for (a) Ni${}_8$Fe${}_2$, (b) Ni${}_8$Fe${}_2$/Fe, (c) Ni${}_4$Fe${}_6$, and (d) Ni${}_4$Fe${}_6$/Fe. The first column reflects the XMCD spectra across the Gd edge M${}_5$ edge, the second column across the Ni L${}_{2,3}$ and the third column across the Fe L${}_{2,3}$ edge.

Figure 5

(a)–(d) Gd asymmetry ratio as a function of temperature for all samples. The data have been normalized to $T$ $=$ 30 K in order to fit the data to a two-layer model (Ref. 5). We have included in each graph the parameters inferred from the model, $T_C$${}_1$, $T_C$${}_2$, and $d_2$.

Figure 6

Gd XMCD data at 300 K. (a) Gd XMCD spectra. (b) Areas under the Gd XMCD spectra for samples with different Fe content.