Crystal Forms for Display
Previous  Top  Next

Dialog Box - called from: Input1 Menu

With this option you can display an external crystal shape which is not the same as the boundary faces for locating atoms. You can even show a crystal shape for structures in which the boundary option is No Boundaries or Translation Limits.

This dialog controls the generation of the faces, corners and edges which make up the display shape; the Crystal Edges dialog in the Input2 menu controls colors and other display aspects of the crystal edges - of course it is only the edges which are actually shown.

By default, the display shape is the same as the boundary shape for boundary options Default Unit Cell, Enter Forms and Slice, and there is no display shape for options No Boundaries, Translation Limits and Sphere. If you choose to change the display shape, or to show one where no crystal forms are used as boundaries, the input and revision thereof are similar to those for input of boundary crystal forms in the Enter Forms boundary option, through the Add/Revise Form dialog

For the 2D Drawing modes of ATOMS, the edges demarking the external crystal shape should be outside the atomic structure. As discussed in Drawing Crystal and Unit-cell Edges, an external crystal shape which lies inside or interpenetrates with the atomic structure will in many cases be drawn incorrectly. In the 3D Drawing Modes, interpenetration relations are drawn correctly.

Although this option is primarily intended for the Default Unit Cell and Enter Forms boundary options, it can also be used for molecules and polymers.

Note that you may have to change the Display radio buttons in the Crystal Edges dialog in the Input2 menu, to allow the crystal shape to be displayed.

Using display faces with Cartesian symmetry input. The symmetry option Cartesian matrices is intended primarily for molecules with non-crystallographic symmetry. However, it can be used with "crystal" faces with certain restrictions. You must use Cartesian face coefficients or inverse intercepts, and if some of these "indices" or the symmetry equivalents thereof are irrational they must be converted to large integers. For example, the face (100) in a trigonal or hexagonal crystal would also have coefficients 1,0,0 in a Cartesian system with y=b (or a2), but some of the symmetry equivalents of this face, such as (-110) and (0-10) would have irrational coefficients. If you enter 20000,0,0 as the "indices" of the form, then when the indices are multiplied by the Cartesian symmetry matrices the result will be large integral indices for the symmetry equivalents. If you enter 1,0,0, the results would be fractions, which would be incorrectly truncated or rounded off either to 0 or 1. As another example, an icosahedron may be drawn in either of the icosahedral groups I or Ih by giving the form indices 0,0,20000; the indices 0,0,1 would not work because many of the 19 faces equivalent to this one have irrational indices.

Using integers with up to four or five digits (99,999) gives adequate precision for the calculations involved in determining the external crystal shape.