A 1000Å YFe sample was also studied in the the same way as that outlined for the DyFe sample. The zero applied field spectrum is compared to other thin film samples in Figure 6.3.
The Isomer Shifts of each site are fixed to be equal but free to vary as a whole, as are the relative line intensities. The Isomer Shift values are the same as for DyFe which is expected as rare earths are chemically very similar.
The results from applying a in plane magnetic field are shown in comparison to zero field in Table 6.4 and Figure 6.5. The most noticeable difference is in the angle of the iron moments to the incident gamma ray (from the relative line intensities), the moments have been forced more in plane by to lie at an angle of to the sample plane. There is no error on the change in angle as the spectra were recorded sequentially.
As yttrium does not carry a magnetic moment in this system there is no single ion magnetocrystalline anisotropy, as in the DyFe sample, strongly locking the iron moments to a particular crystalline direction. The applied field of is of the order necessary to overcome the exchange anisotropy between iron atoms, displacing the domain walls in the untrained sample producing a single domain sample. In this case the relative line intensities give a unique angle for the direction of the iron moments, recorded as to the sample plane. This lies in between either the or directions, which make angles of and respectively to the sample plane for a sample normal direction of (110). Although the observed angle is much closer to the direction it is also closer to the sample plane. With no magnetocrystalline anisotropy to overcome through coupling with the rare earth it would be expected that the iron moments would be more free to rotate in the direction of the applied field, which is into the plane of the sample. The angle observed is one that balances the magneto-elastic and external field energies in this sample and so is unlikely to be truly indicative of the magnetic easy axis in zero field. Thus this result cannot be taken as absolute evidence of a direction without other corroborating data.
The iron moments would be expected to lie in plane, even under zero applied field; the shape anisotropy favours in plane alignment, as does the applied field. As this isn't the case the strain in the sample plane must be causing a magnetostrictive effect through the iron atom's dipolar interactions but of a much smaller magnitude than for the DyFe system as evidenced by the much larger change in moment alignment under applied field. Magnetic rare earths display much stronger magnetostriction than transition elements.
The change in the average magnetic hyperfine field is . This is slightly larger than the applied field and significantly larger than the change recorded in the DyFe sample. This is explained by the iron moments lying more parallel to the applied field and the increased magnetisation of the sample as the separate domains rotate with the field, increasing the magnetic field within the sample. In the DyFe sample the domains were still pinned by the magnetic anisotropy of the dysprosium atoms and thus the change in magnetisation was negligible.
Dr John Bland, 15/03/2003