Ce($x$Å)/Fe(15Å)

Figure: Normalised magnetisation vs temperature scans for Ce($x$Å)/Fe(15Å) multilayers.

Figure 7.4 shows the scans for Ce($x$Å)/Fe(15Å) multilayers. This set of scans has a fixed iron layer thickness of $15\ensuremath{\unskip\,\mathrm{\AA{}}}$.

The 20/15 sample shows an antiferromagnetic coupling, with a $T_{N} = 157\pm5\ensuremath{\unskip\,\mathrm{K}}$. The region below $T_{N}$ is relatively flat (within equipment error), indicating moments perpendicular to the applied field.

The 27/15 sample also shows antiferromagnetic coupling, with a $T_{N}=125\pm5\ensuremath{\unskip\,\mathrm{K}}$, but with the temperature dependence of a mixed state of perpendicular and parallel moments. The $T_{N}$ for this sample is lower than that of the 20/15 sample with the more perpendicularly aligned moments.

Mössbauer measurements show both samples have moments lying $32^{\circ}$ out of plane, but the 27/15 sample has a lower hyperfine field compared to bcc iron than the 20/15 sample. The hyperfine fields are reduced by $20\%$ and $12\%$ for the 27/15 and 20/15 samples respectively. This difference in hyperfine fields and perpendicular or parallel moment alignment does not coincide for all samples, however.

It cannot be determined from these data alone whether the variation in moment alignment is due to the cerium layer thickness or merely a coincidence of the sample construction. Indeed the following Ce($x$Å)/Fe(20Å) measurements show the opposite arrangement for cerium layer thicknesses of $20\ensuremath{\unskip\,\mathrm{\AA{}}}$ and $27\ensuremath{\unskip\,\mathrm{\AA{}}}$.