Distinguishing Types of Disorder in Diffuse Scattering: A Numerical Simulation Study
In crystallography, the central assumption of conventional studies is that the material is made up of an array of identical structural building-blocks, the unit cells of the crystal. In most experiments, we measure the sharp Bragg reflections, and these give us information about the long-range average structure. Bragg reflections obey Bragg’s Law, and occur at discrete points. However, many real systems show disorder. To study this, we use diffuse scattering, which is weak and delocalised and can occur throughout the diffraction pattern. If you have a single crystal then single crystal diffuse scattering can be performed, if not then a powder diffraction measurement. Now, a single crystal sample means you get a three-dimensional data set. One slice of such a dataset is shown in figure 1.
Figure 1: Single crystal diffuse scattering data collected from wüstite, Fe1-xO.
In the slice in figure 1, virtually all the features you can see are diffuse scattering features, and so are descriptive of local (nanoscale) ordering in the material, and are ignored in conventional analysis. If the sample is a powder, the three-dimensional data are collapsed down into a single dimension, giving a plot something like figure 2, where intensity is plotted against scattering vector.
Figure 2: Powder diffraction data from PZN (PbZn1/3Nb2/3O3), measured using neutrons.
When considering local order, the powder data are often transformed to give a pair distribution function, which describes the distribution of distances between pairs of atoms. This shows up information about local order.
We were curious as to how interpretable the two forms of data are, what aspects of disorder they are sensitive to, and whether some results in the literature might be over-interpreting their data. This is the essence of the new publication
‘Distinguishing Types of Disorder in Diffuse Scattering: A Numerical Simulation Study’ by D. J. Goossens and R. E. Whitfield, accepted for publication in Metallurgical and Materials Transactions A. (DOI:10.1007/s11661-013-1812-x) At the moment there is no full reference, but a DOI is good enough these days.