Proper sample preparation is one of the most important requirements in the analysis of powder samples by X-ray diffraction . This statement is especially true for soils and clays that contain finely divided colloids, which are poor reflectors of x-rays, as well as other types of materials such as iron oxide coatings and organic materials that make characterization by XRD more difficult. Sample preparation includes not only the right sample treatments to remove undesirable substances, but also appropriate techniques to obtain desirable particle size, orientation, thickness, etc. Several excellent books are available that deal with appropriate sample preparation techniques for clays and soils (Jackson, 1979, Brindley and Brown, 1980, Moore and Reynolds, 1989, Bish, 1992, Iyengar, 1997).
Analysis of powders by XRD requires that they be extremely fine grained to achieve good signal-to noise ratio (and avoid fluctuation in intensity), avoid spottiness and minimize preferred orientation. Reduction of powders to fine particles also ensures enough particle participation in the diffraction process. The recommended size range is around 1-5µm (Klug and Alexander, 1974, Cullity, 1978, Brindley, 1980), especially if quantification of various phases is desired. For routine qualitative evaluation of mineral components, the samples are usually ground to pass through a 325 mesh sieve (45 µm). Grinding is accomplished either through hand grinding or in a mechanical grinder. The effects of excessive grinding include lattice distortion and possible formation of an amorphous layer (Beilby layer) outside the grains.
There are two types of mounts normally employed depending on the nature of crystallite orientation required. Random mounts are preferred when identification of phases in a specimen is required. In this type of mount, particles ground to 1-5µm are packed to a flat surface onto a sample holder to assume different orientations and ensure reflections from various hkl planes. Oriented mounts are used when analysing clay minerals which rarely show strong diffraction effects from Bragg planes other than the (00l). In general, these are prepared by making a slurry of the sample with distilled water. The water is then allowed to evaporate until the slurry is smeared into a sample holder (could be a glass slide or ceramic tile).
Below is a flowchart which details the steps in preparing a sample of many different types using various techniques.
Important factors in sample preparation
- Sample properties also influence the quality of a powder pattern by either reducing intensities or distorting intensities.
Preferred orientation or texture: By texture we mean that the powder particles do not have an arbitrary shape but a strongly regular anisotropic shape, typically platelets or needles. On preparation these are then preferably oriented along the sample surface massively changing the peak intensities. Several techniques may be employed to minimize this effect.
- The most efficient way is to form a slurry in a highly viscous liquid such as nail varnish. In such a liquid, the random orientation is retained on drying.
- Alternatively, the anisotropic particle shape can be reduced by grinding in a ball mill. This should be done with great care as excessive grinding can easily break down the particle size to the nanometer size and lead to amorphisation. It is recommended to try the effect of subsequent 5 minute grinding intervals to optimize the process on respective samples. In the case of coatings or thin films preferred orientation is often a desired effect. In this case Rietveld refinement can be used to determine the degree of texture.
- Crystallite size and strain: the broadness of a diffraction peak corresponds to the mean crystallite size in a reciprocal manner. The smaller the average crystallite size, the broader the reflections and the lower the absolute intensities. This effect becomes visible below and average crystallite size of less than 200 nm. Related to crystallite size broadening is strain broadening. Strain broadening occurs due to the presence of defects in crystals. Such strain can be introduced via substitution of constituting atoms but also via special thermal treatment. It is possible to distinguish strain broadening from size broadening as the angle dependence is much larger than in the latter case.
- Sample preparation height: Rotating sample holders improve the measurement statistics and thus provide the best results. However, they are not available for all machines. The most severe error during sample preparation is to fill the sample holder too high or too low. Both result in a significant shift of peak positions which can make the interpretation difficult.