Acta Cryst. (1991). A47, 655-685

Atomic site identification

The identification and labelling of atomic sites is vital in crystallography, both for the structure determination process, and for the description and interpretation of the final results. The atom labelling scheme adopted in the CIF embraces a very wide range of the common styles and conventions in current use. The scheme is fully detailed in Section 4 of the CIF Dictionary (Appendix I), but short introductory notes are also given here.

The labelling of atoms in most crystallographic studies serves two distinct purposes: (a) to identify a site in the crystal, and (b) to identify the chemical element that occupies that site. The former is used, for example, to link the site to the corresponding set of scattering-factor coefficients, whilst the latter is used in descriptions of the structure which are related to its chemistry, e.g. lists of geometrical data. The CIF Dictionary makes this distinction clear by defining _atom_site_ and _atom_type_ as separate data categories. The connection between the two is made through the identical data items _atom_site_type_symbol and _atom_type_symbol. However, normal crystallographic practice is to use a single label to define both the site and the chemical species occupying it. The CIF definition also makes provision for this. Thus, the _atom_site_label has a structure (described in detail in Section 4 of Appendix I) that can start with an element symbol (optionally followed by an oxidation state) followed by an identifier. Together these two parts uniquely define each site, with the first part, termed component 0, identically matching one of the _atom_type_symbol codes.

The _atom_site_label may contain up to six further components, in addition to component 0 described above, to provide for full and explicit site identification. Component 1, the atom number code, would normally be present and is simply concatenated with component 0. Other components, e.g. the residue code, the sequence code etc., permit the construction of specialist _atom_site_label codes. These additional components are linked to previous ones by the underline character.

The link between the _atom_site_ and _atom_type_ data items can be given in two ways: (a) if the _atom_site_type_symbol is given, then it must match identically one of the _atom_type_symbol values, or (b) if it is not given, then the first part (component 0) of the _atom_site_label must match one of the _atom_type_symbol codes. In this way, the common practice of identifying atom sites as e.g. C1, Hg7, H1C1 etc. is preserved.

This site labelling scheme can also be used for sites that are occupationally disordered, either by defining _atom_type_ items with the appropriate average properties, or by defining two atom sites that have identical coordinates, but which point to different atom types.

Back to title page

On to Creating a CIF

Copyright © 1991 International Union of Crystallography