Temperature dependence of ferromagnet-antiferromagnet spin alignment and coercivity in epitaxial micromagnet bilayers

Soft x-ray photoemission electron microscopy with an in situ magnetic field has been used to study the relationship between ferromagnetic and antiferromagnetic spin alignment and the switching/reversal field of epitaxial micromagnetic structures. We investigated a model system consisting of a bilayer of ferromagnetic $\mathrm{L}{\mathrm{a}}_{0.7}\mathrm{S}{\mathrm{r}}_{0.3}\mathrm{Mn}{\mathrm{O}}_3$ and antiferromagnetic $\mathrm{LaFe}{\mathrm{O}}_3$ where the spin axes in each layer can be driven from mutually perpendicular (spin-flop) to parallel alignment by varying the temperature between 30 and 300 K. Results show that not only does this spin alignment noticeably influence the bilayer micromagnet coercivity compared to $\mathrm{L}{\mathrm{a}}_{0.7}\mathrm{S}{\mathrm{r}}_{0.3}\mathrm{Mn}{\mathrm{O}}_3$ single-layer micromagnets, but the coercivity within this materials system can be tuned over a wide range by careful balance of material properties.

Figure 1

(Left column) XMLD images of the LFO layer in the zigzag wires at temperatures between 30 K, where the AFM spin axis is perpendicular to the wire edge (spin-flop), through frustrated alignments at 100 and 200 K, to 300 K, where the AFM spin axis becomes parallel to the wire edge. (Right column) Crystallographic legend, XMCD image of the LSMO zigzag wire at 100 K with FM domains oriented parallel to edges of the wire, and a schematic of nanowire dimensions. The incident x rays are along the $\left[1\overline{1}0\right]$ direction for all X-PEEM images.

Figure 2

(a) XMCD image of a partially switched $(\langle 110\rangle$-oriented, LSMO single layer at 105 K after applied field of 49 Oe) micromagnet array combined with schematic of the X-PEEM experimental setup depicting the incident x-ray angle, micromagnet orientation, and applied field direction. (b) Angles for applied field $H_{\mathrm{ext}}$, FM spin axis ${\phi }_{\mathrm{FM}}$, and AFM spin axis ${\phi }_{\mathrm{AFM}}$ defined relative to the long axis of the individual micromagnets. (c) Percentage of micromagnet array switched vs applied field for the LSMO single-layer (left) and LFO/LSMO bilayer (right) samples at temperatures between 30 and 200 K. Solid lines are normal cumulative distribution function fits, and the intersection with the horizontal dashed line corresponds to the field where 50% of micromagnets have switched.

Figure 3

(a) $\mathrm{MuMa}{\mathrm{x}}^3$ simulations of saturated, remanent state for both $\langle 100\rangle$ (top) and $\langle 110\rangle$ (bottom) oriented micromagnets showing the presence and absence of a vortex flux-closure domain in the $\langle 100\rangle$ and $\langle 110\rangle$ structures, respectively. (b) XMCD image of $\langle 100\rangle$-oriented micromagnets. (c) XMCD image of $\langle 110\rangle$-oriented micromagnets at 100 K. Both (b) and (c) show only two $2\mu \mathrm{m}\times 0.5\mu \mathrm{m}$ rectangular structures from an array of $\sim 100$.

Figure 4

50% switching field for $2\mu \mathrm{m}\times 0.5\mu \mathrm{m}$ rectangular micromagnet arrays plotted as a function of temperature. Blue corresponds to single-layer LSMO and red to LFO/LSMO bilayer samples. Data collected in this study on micromagnets with long edges parallel to $\langle 110\rangle$ are represented by circles. The $\langle 100\rangle$ data included from [8] are denoted by diamond symbols.