Characterisation of Antiferromagnets

Abstract

The characterisation of antiferromagnets (AFs) for almost all applications is very challenging because intrinsically an AF gives little or no signal. Above the Neel temperature (TN) AF materials exhibit paramagnetic behaviour in a similar manner to a ferromagnet above its Curie point. This does allow for the determination of the Neel temperature but reveals little about the behaviour of the material once the AF order is established. Historically the structure of AFs was determined using techniques such as neutron scattering [1], [2] and more recently some studies have been undertaken using X-ray techniques such as XMCD [3]. However such techniques and particularly neutron scattering require large samples with dimensions of the order of millimetres. For all technological applications AF materials are used in thin film form and hence such techniques are not available for use due to the very long counting times that would be required. A second problem with AF materials is that their behaviour has not been well established until recently. For example Neel predicted the existence of AF domains but there are very few reports of their observation [4], [5]. Their behaviour is not well understood because conventional domain theory is based around the existence of magnetostatic energy in a conventional ferromagnet which is not present in an AF. Other critical factors associated with the structure of thin films are also poorly understood. Principal amongst these is that in polycrystalline films a granular structure will exist but the question then arises as to at what critical grain size will single domain behaviour be observed? In single crystal or large grain thin films where presumably AF domains will form, what is the nature and consequence of domain wall pinning? In this tutorial lecture the fundamental nature of antiferromagnets will be discussed addressing metals, alloys and oxides. However the focus will be on those materials which have, or are most likely to find application in spintronic devices. Techniques that allow the behaviour of AF thin films at least to be inferred will also be discussed. These are principally associated with the measurement of exchange bias systems where a ferromagnetic layer is used as an indicator of the structure of an underlying AF. However in exchange bias systems it is also the case that the exchange field from the ferromagnet causes a reaction and possibly a change of order in the AF. This concept is the basis of the so-called York Protocols which allow for the measurement of the behaviour of granular, generally sputtered, AF films. The underlying York Model of Exchange Bias developed in 2008 will then be discussed in detail as it allows for the full characterisation of granular films and in particular the determination of the anisotropy constant of AFs. This model has been used by all the major manufacturers of hard drive read heads to design improved AF layers in their stacks. For the case of large grain or single crystal thin films a brief review of complex large scale computer models of possible domain structures will be presented with the emphasis placed on a simple strong domain wall pinning model which has been found to replicate qualitatively the observed behaviour in such structures. Finally the effect of reduced dimensions on the behaviour of AFs will be presented as it is well established that small lithographically defined structures containing AF layers behave quite differently to bulk films.