CrystalDiffract lets you simulate the effects of non-random crystallite orientations in your powder sample. Such "preferred orientation" can result in some peaks appearing much stronger than expected, whilst other peaks may be weaker - or entire absent from the diffraction pattern.

CrystalDiffract can simulate preferred orientation from one of two common scenarios:

The degree of alignment is represented by a fraction, with zero representing a completely-random sample, and 1 being a fully-ordered sample (e.g., all plates exactly parallel). In between the two, there is a statistical distribution of orientations.

This can best be visualized by thinking of plate normals (or needle axes) plotted on a stereogram or "pole figure". As the degree of alignment increases, the randomly-aligned poles begin to clump around the centre of the pole figure. The actual distribution of angles varies from -90° to +90°, but its standard deviation decreases as the alignment increases.

CrystalDiffract assumes a normal (Gaussian) distribution of orientations, as first proposed by March in 1932 (Z. Kristallogr. 81:285-297). (Note that CrystalDiffract's's alignment parameter is equal to: 1 - R, where R is the March coefficient which measures the degree of randomness.

To correct for preferred orientation Show

Do the following:

  1. Locate the diffraction patterns you wish to modify in the Patterns List, and select them.

  2. Locate the Preferred Orientation group in the Parameters Inspector.

  3. Enable Preferred Orientation by using the checkbox.

  4. Choose your sample geometry: plate or needle.

  5. Specify the preferred-orientation direction as either a plane normal or a lattice vector (use the small popup menu on the right-hand side of the text fields to switch from hkl to UVW).

  6. Use the slider control to adjust the degree of preferred orientation, from 0 (completely-random powder) to 1 (fully-aligned crystallites).

See also

Peak geometry and instrumental broadening

Particle size and strain