Medium Resolution Results

In an attempt to conclusively establish the apparent anomalous change in intensity in the vicinity of 140-120K which was suggested by the low-resolution results, the experiment was undertaken a second time using the medium-resolution germanium optics. The immediate consequence of the higher resolution, however, is an order of magnitude reduction in the scattered intensity. More critically, it was also found that the very precise alignment required with higher resolution is affected much more significantly by the random fluctuations in sample position. Although also encountered in the previous measurements, they only resulted in low random fluctuations limiting the accuracy of the measurements. In this case, where the observations are no longer averaging over a distribution of grains within the mosaic of the crystal, but are selecting a single grain, the thermal changes between temperatures were sufficient to completely destroy the alignment. This made realignment necessary after every temperature change, and consequently made reliable and repeatable measurements of the kind required for this study impossible.

The problem is illustrated in Figure 6.6(a) where profiles of the (0 0.21 19) satellite reflection are shown at a succession of temperatures. It is impossible to make any comparison when such random fluctuations in the reflections are present, and these measurements were only possible after careful re-alignment of the sample. The problem was nearly as pronounced in the fundamental reflections as well. The attempt to make a quantitative study of the intensity behaviour at medium-resolution could not therefore be completed.

However, the measurements do produce one interesting and unexpected result which was not apparent in the low-resolution work. The profiles in Figure 6.6(b) show the ${\bf c}^*$ splitting of the satellites which was first investigated at room temperature in Chapter 3. The changes with temperature appear to be very marked, and to involve a transfer of intensity from one of the ${\bf c}^*$ splinters to another. Although it is difficult to disentangle these changes from those random fluctuations of the fundamental reflection, the [0 1 0] profiles do show that in some cases there is an almost complete reduction in intensity of the primary satellite at the commensurate ${\bf c}^*$ position. Due to the difficulties discussed, it proved impossible to establish whether there was any systematic process at work in these changes with temperature. Perhaps the simplest explanation would be if the modulation corresponding to the non-commensurate ${\bf c}^*$ resided exclusively in single crystal grains whilst others possessed exclusively the commensurate version. The only effect of temperature is then to see-saw between the different grains with changes in temperature. Therefore, no structural change is required to explain the observations, but it does then indicate that the two modulation structures do not co-exist within the same crystal areas.

Figure 6.6: Profiles of the satellite (0 0.21 19) in (a) the ${\bf b}^*$ direction and in (b) the ${\bf c}^*$ direction; measured at various temperatures using medium-resolution.