The magnetic domain structure of CoCr layers studied by neutron depolarisation

Abstract

Abstract. The domain structure of relatively thick RF-sputtered CoCr (81 : 19 at 9%) layers with a columnar morphology in the range of thickness 1-5 pm was investigated using a polarised neutron beam. On the basis of the results a model of straight domains running through the entire film thickness has to be rejected. Arguments are given for a serpentinelike domain structure with a domain width about the same as the column diameter (0.1-0.3 pm), resulting in a 'chain of columns' model. Sputtered CoCr layers have already been applied as a perpendicular recording medium with extremely high density (Iwasaki 1980). The morphology and crystal structure and therefore the magnetic properties of the film depend strongly on the deposition and growth conditions.Our films consist of conically shaped crystals or columns with c-axis textured hcpstructure (Lodder et al 1983). The crystal or column diameter increases almost as the square root of the film thickness. In order to understand the magnetisation process, domain studies are of great importance. If the exchange force is restricted to the columns the domain size coincides with the crystal diameter. The columns will act as single domain particles and a rotational switching mechanism is to be expected for the magnetisation process. In this case the domain structure is determined by a minimum of the magnetostatic energy and will appear as a kind of checkerboard pattern formed by the individual columns. In the case of a continuous model, where the exchange forces extend over the crystal boundaries, the magnetisation process is due to domain wall motion which is more or less hindered by the column boundaries.The domain dimensions of both particle and continuous models are too small to be observed by Kerr microscopy or the Bitter technique. The most suitable method for observing small domains is Lorentz transmission electron microscopy. A 1 MeV microscope has to be used for observing CoCr layers thicker than 0.1 pm, up to a maximum of 0.3 pm Iwasaki 1982, Grundy andAli 1983). A specially designed sample stage to exclude an external perpendicular field from the specimen during domain observations is essential Iwasaki 1982, Ohkoshi andKusuda 1983).Different domain configurations with in-plane magnetisation are observed in thin (<60 nm) CoCr layers (Grundy and Ali 1983, Ohkoshi and Kusuda 1983, Wielinga 1983. This can be explained by variations in the preparation methods. None of these films have a well oriented hcp c-axis. For a good textured layer (Co77Cr23), with a

Sans experiments on thin CoCr films with perpendicular domain structure

“…Magnetic scattering is perpendicular to the magnetization, therefore scattering in the y direction from these "closure" domains is expected. This picture is consistent with Kerr observations [4] and neutron depolarization results which also showed the existence of domain walls parallel to the xy-plane and to the magnetic field in the y-direction of the film plane [4,5].…”

“…The differences in maximum value of S(Q) for the samples do not agree with the ratio of the squares of the domain height as predicted by formula (4).…”

“…The resolution of this method for measuring the domain size in the plane of the film is about 0.2 pm. More recently the neutron depolarization technique (NDT) [4,5] has been introduced which enables one to determine the domain height together with the domain width in two orthogonal directions. The limitations of this last method are in the film thicknesses which can be studied, unless a large number of films in one packet is used.…”

Neutron depolarisation in CoCr layers to study the perpendicular magnetic domain structure

“…The upper part of table I which was published earlier [4] contains data from films of 4 thicknesses. The values for ~b and cos~4 ~b are obained from (M~L) using eq.…”

Initial Slope of the Hysteresis Curve

“…For the thinner films T, and T, are very similar, while for thicker films this is not the case. In [3,5, 61 there are indications for the possible presence of small reversed domains at the surface, known as spikes, (proposed by Privorotskii [7]). At a certain depth inside the thin film each domain begins to split.…”