Temperature vs Magnetisation Scans

The variation with temperature of a sample's magnetisation can give information upon the type and strength of magnetic ordering in the sample. All samples were field cooled from room temperature to $ 1.8\ensuremath{\unskip\,\mathrm{K}}$ under an applied field of $ 1000\ensuremath{\unskip\,\mathrm{Oe}}$. The sample magnetisation was then measured under the same applied field from $ 1.8\ensuremath{\unskip\,\mathrm{K}}$ to $ 400\ensuremath{\unskip\,\mathrm{K}}$.

As the sample size is not known for all of the samples the magnetisation cannot be compared quantitatively for these scans. As information can be obtained from the shape of the scans rather than the actual magnetisation the magnetisation vs temperature data has been normalised for each sample to its own magnetisation at $ 1.8\ensuremath{\unskip\,\mathrm{K}}$. This allows groups of samples to be compared visually.

Most of the datasets were obtained whilst the SQUID unit was experiencing considerable electronic noise problems. This has been reduced significantly by smoothing the resultant data with a Gaussian convolution. Whilst the main shape of the scan becomes more visible the smoothing cannot totally remove large noise artifacts when they are over a large number of data points and so any fine structure in the scans obtained from the sample cannot be distinguished from that produced by the noise.

The room temperature CEMS results obtained by G.S. Case[3] show the iron to be composed of ordered bcc iron and an amount of paramagnetic amorphous iron which increases in proportion as the iron layer thickness decreases. In samples with iron layer thicknesses $ <20\ensuremath{\unskip\,\mathrm{\AA{}}}$ the amorphous/bcc iron ratio increases and Perpendicular Magnetic Anisotropy increases, tipping the spins out of the plane of the sample. For the discussion of the magnetometry results the iron layers are taken to be ferromagnetically ordered in the iron layers, whilst the interlayer coupling can be antiferromagnetic, ferromagnetic or uncoupled depending upon the cerium layer thickness and the magnetisation of the iron layers. The results are plotted in sets for comparison, where either the cerium or iron layer thickness has been kept constant. The data is plotted as magnetisation, but as the field, $ H$, is constant the y-axis also corresponds to $ \chi=\nicefrac{M}{H}$.

Dr John Bland, 15/03/2003