The aims of this thesis are to investigate the structure and magnetic interactions in multilayer systems and how these affect their magnetic properties, to understand the mechanisms at work in nanoscale structures such as thin films and how these affect bulk properties, and to contrast the different properties of iron as a magnetic constituent in intermetallics, multilayers and oxides.
The main investigatory technique used in this thesis is Fe Mössbauer spectroscopy. Mössbauer spectroscopy is an isotope specific technique which can provide information about magnetic ordering via the hyperfine field, moment orientation, charge states and the local environment of the Mössbauer isotope. Atoms in sites which differ in their electronic or magnetic environment can be distinguished.
Chapter 2 covers the theory of the Mössbauer effect, its physical basis and how it can be utilised as a spectroscopic technique. The various hyperfine interactions which affect the shape of the resultant spectrum are explained and illustrated. Most Mössbauer data in this thesis have been obtained using Conversion Electron Mössbauer Spectroscopy, a surface sensitive technique. The differences between this and the more common Transmission Mössbauer Spectroscopy technique and CEMS' depth dependence are explained.
SQUID magnetometry has been used as the primary technique for examining the Ce/Fe multilayers. The physics of SQUIDs is given in Chapter 3. This is combined with an overview of magnetometry in general. Although magnetometry data were only used for a small part of the data presented in this thesis they are included to provide a resource for the basic principles of magnetism which are used in the explanation of much of the Mössbauer data as well.
Chapter 4 gives information about the actual equipment used to facilitate the theoretical techniques outlined in the above chapters, along with an overview of the analysis routines used for analysing the Mössbauer results.
As magnetic interactions, such as magnetic anisotropy, play an important role in the behaviour of many of the systems studied, Chapter 5 comprises a brief account of these topics.
Chapter 6 is the first chapter of experimental results. The crystal structure of Rare Earth/Iron cubic Laves Phase systems is discussed with respect to bulk systems. The effects, such as strain, of growing these systems as thin films with Molecular Beam Epitaxy, are then applied to the bulk system. Results and analysis of the CEMS data obtained from thin film samples are presented first. These are then extended to multilayer samples of DyFe and a non-magnetic Dy spacer layer, or DyFe
/YFe
multilayers of varying layer thicknesses. Applied field measurements were made on all samples apart from the DyFe
/Dy system, to give information on the magnetic anisotropy.
Chapter 7 presents results from two contrasting multilayer systems; Lanthanide/Iron and Actinide/Iron. SQUID magnetometry results from Ce/Fe multilayers are presented complementary to Mössbauer Spectroscopy results obtained by S. Case[3]. The variation in the coupling type, and strength, of the iron layers across the cerium layers is examined. CEMS data and analysis from a series of U/Fe multilayers with different uranium and iron thicknesses are then presented. The moment orientations and crystalline nature of the iron layers is discussed. The data are compared to x-ray reflectivity data taken by A. Herring of the Liverpool University Condensed Matter Physics group.
Chapter 8 is an investigation of iron oxide systems, with particular reference to magnetite. The crystalline and magnetic structure of magnetite is outlined. A commercial study for Laporte is given, where the ratios of iron oxides was determined in printer toner powders. The behaviour of magnetite in bulk systems is then compared to thin film systems, with a Mössbauer study of magnetite thin films on different substrates. The crystalline growth and composition of the layers is discussed.
Finally, Chapter 9 presents the conclusions produced from the analysis and suggestions for future work on the samples studied.
Dr John Bland, 15/03/2003