Mirrors are a relatively new feature for laboratory X-ray diffractometers, though they have been used for many years for neutrons and synchrotron X-rays. The simplest X-ray mirrors exploit the physics of refraction to bend the path of the beam off a metal surface. Since the refractive index is just below unity, total internal reflection occurs, but only at very low angles well below 1°.
Typically, the simplest mirrors are coated with a single layer of refracting material. However, the use of several layers of alternating material, each several nanometres thick, improves the reflectivity by an interference effect leading to diffraction of the X-ray beam off the mirror. Again, since the multilayer spacing is relatively large, the diffraction angle is also relatively small as for the simple case of refraction.
It is possible to vary the thickness of the multilayers along the length of a curved mirror so that a divergent incident beam emerges as a parallel one as shown in the diagram below:
Although the "reflectivity" of the mirror is a function of wavelength (since both refraction and diffraction are functions of wavelength), they cannot be used to monochromate an incident beam from a sealed-tube source to give a pure Kα1 wavelength.
However, the parallel beam that is produced by a parabolic diffraction mirror has some advantages over the standard divergent beam, particularly for powder diffraction studies for flat samples using Bragg-Brentano geometry (as will be discussed shortly). Employing a mirror with such a geometry has the advantage that the surface of the sample need not be perfectly flat: this situation often arises for samples that are heated inside a furnace to high temperature due to either sample expansion or decomposition.
The disadvantage is the presence of several wavelengths. Since mirrors function more efficiently with the longer wavelengths, the presence of Kβ is significantly reduced relative to Kα, but a Ni filter, for example, may still be required if a very high level of Kβ suppression is required.
In principle, it is possible to combine a mirror with an incident or diffracted beam monochromator, but this combination is rarely seen in practice.
|© Copyright 1997-2006. Birkbeck College, University of London.||Author(s): Jeremy Karl Cockcroft|