The progress being made in the field of magnetic recording is extremely fast. The need to keep this progress going, leads to new types of recording materials which require advanced measurement systems and measurement procedures. Furthermore, the existing measurement methods need to be reviewed as due to the differing morphology of novel materials the conventional way to interpretate the measurement results might have become inadequate. These three tasks (development of measurement systems, development of measurement procedures and a review of existing methods) were the main subject of the research described in this thesis. Design specifications and choice of the measurement system. With the decreasing bit-size, the behaviour on micro- or even nano-scale becomes increasingly important. At the same time macroscopical measurements can assist the knowledge and understanding of the new media, especially if combined with observations on the micro-scale. The macroscopical measurements will then present information on the average behaviour of millions or billions of entities where the micro-investigations should ideally detect the behaviour of the individual entities, which could be either grains, particles or domains. In chapter one we have seen that macro-scopical measurements should be used for determining the macroscopical parameters like coercivity and saturation magnetisation. Furthermore, they should be used for angular, vectorial investigations and if possible for obtaining on the average interaction in the media as it is believed that interaction on the scale of the written bits will eventually determine the performance of recording systems. An extrapolation into the future of recording materials, leads to the following desired specifications of a future macroscopical measurement system: · A sensitivity of 10 nAm2 or better (when 1cm2 samples are used, otherwise higher). · High field (³ 2500 kA/m). · The possibility to measure at zero and other static fields with a field resolution of at least 0.1 kA/m. · A vector detection system and rotatable field over at least ± 180°. · Flexible control software that allows the addition and quick implementation of new measurements. · Fast, easy to operate and maintain. In the second chapter several existing macroscopical measurement systems have been compared, leading to the conclusion that the vibrating sample magnetometer (VSM) is the most suitable instrument for this task. The most competitive other systems are either extremely difficult to handle and slow (SQUIDs) or not suitable for measurements at completely static fields (AGFM). Design of the VSM and its detection system. The third and fourth chapters describe the design of the VSM and its detection coil system (DCS) respectively. Especially the design of the VSM itself appeared to be a difficult and multi-disciplinary task that involved as much mechanical as electronic aspects. Eventually a system was created with specifications comparable to or better than any known other system. Some other systems might be better on details but, to our knowledge, none combines these better specifications with a comparable vectorial detection system, reproducibility and position independency. Systems with a better sensitivity, are generally designed for the use of smaller samples and therefore do not perform better in practice as they are restricted to small, low signal samples. It has been shown that the noise of a VSM can be brought down to the level of the Johnson noise, which is the theoretical limit. In the VSM described in this thesis, this means that the peak to peak noise can be as low as 10 nAm2 with a time constant of 3 seconds. With extensive averaging, the noise can be brought down further if necessary, at the cost of the measuring time. The background signal is determined by the diamagnetism of the sampleholder and the vibrations in the system. Using a symmetrical sample holder design, the diamagnetic signal can be minimised. Extensive vibration damping and vibration compensation by means of an anti-phase vibrator, can reduce the system vibrations sufficiently. The total background signal at present is approximately 20 nAm2/T, possibly this figure can be reduced further by tuning the anti-phase vibrator in such a way that the vibrational background signal and the diamagnetic background signal compensate each other. Another approach is the addition of small pieces of paramagnetic material to exactly compensate the diamagnetic signal. This has already been tested on another scale, in that case leading to a 10 fold reduction of the background signal. Almost all vectorial detection coil systems show extreme difficulties with angular calibration. The DCS used here has prevented these problems by avoiding relative rotations between sample and DCS. The main disadvantage of the DCS described here is its increased sensitivity towards vibrations and an angle dependent image signal. The image effect has been reduced down to ± 1% and the vibrational problems have been solved as described above. Therefor, in this VSM no angular calibration is necessary, for a system with specifications comparable to the best conventional systems. In conventional DCSs of the Malinson type where the ratio of sample-coil distance and sample size becomes too small, the signal in the x-direction coils is influenced by the y-direction magnetisation and vice versa. This also means that uni-axial systems of this type are inadequate as soon as the magnetisation vector is not aligned with the field, even when the only magnetisation component of interest is the projection on the field direction. As standard calibration methods neglect this effect, measurements where the magnetisation has an angle with the applied field can be up to 30% in error. A calibration method has been described that corrects for these effects. The software for system and measurement control. The software that has been created for the control of measurements on the VSM has been written in such a way that it can can control practically all measurement systems where a (set of) signals is measured as a function of a (set of) quasi-statically changing parameters. Furthermore a single measurement procedure has been developed that can control any type of measurement procedure. One of the major advantages is that maintenance has become restricted to only one routine. Measurement procedures and the interpretation of their results. Vectorial detection of the magnetisation leads to an increased insight in the magnetic behaviour of materials. The standard method of looking only to the component of the magnetisation in the field direction is not necessarily the best and can in some cases lead to a misinterpretation of the magnetic behaviour. For measurements performed at an angle with the easy axis of the medium, the coercive field can not be defined as the field where the magnetisation becomes zero, as there is no such field, a better definition would be the field where the magnetisation in the field direction becomes zero. However, when the field is applied under an angle with the easy axis, the field where the majority of the magnetisation switches is not the coercive field but the field where the total magnetisation vector reaches a minimum length. A plot of the defined average switching field as a function of the field angle, is more useful than the angular coercivity plot and more easy to interpret in terms of the anisotropy and demagnetisation. A VSM measures the average behaviour of billions of particles or domains. Therefore it is impossible to determine whether observed behaviour is representative for the complete material. This means that if measurements indicate a particulate reversal behaviour, there might very well be areas in the material which show a domain-wall-motion reversal behaviour. Therefore it is impossible to derive conclusive evidence for the suitability of a material for magnetic recording. The opposite is possible in many cases however. Torque measurements are possible with a bi-axial VSM and lead to results that are comparable to results obtained with a torque magnetometer. The latter instrument has a much higher sensitivity however. An advantage of the VSM is that as the magnetisation vector is known rather than the torque, extrapolation to infinite field is not necessary anymore. Furthermore, the demagnetisation compensation becomes more exact as the exact magnetisation component perpendicular to the sample is known and no assumptions on the alignment of the magnetisation with the field are necessary. Angular Remanence Measurements (ARM) are very useful for the fast and easy determination of the easy axis of the material and the spread in easy axis directions. From ARMs on double-layers it is possible to determine the easy axes of the individual layers. If demagnetisation compensation is used, the ARM gives results that are in agreement with morphological observations. There are indications that the magnetisation in the individual layers of a double layer ME tape mutually influence each other. From ARMs, the resulting easy axes can be determined, and possibly also the direction of the magnetisation in the individual layers resulting from the influence of the other layer.
|Award date||1 Oct 1995|
|Place of Publication||Enschede|
|Publication status||Published - Oct 1995|
- SMI-TST: From 2006 in EWI-TST
- SMI-EXP: EXPERIMENTAL TECHNIQUES