Data Items to be corrected in CIF-files Laboratory for Chemical Crystallography, University of Basel

Please read the following page carefully before archiving your CIF-File!

CIF-files produced automatically by crystallographic programs usually need some corrections in order to be complete. In the following page the most common corrections will be listed and commented. CIF datanames are written in like this (Courier font, red).

The datablock entry

CRYSTALS starts a datablock always with data_CRYSTALS_cif. It is useful to substitute the standard structure name CRYSTALS_cif by some name which contains more information about the compound and permits to identify the structure once the file is isolated from the context of structure solution and refinement.

Entries concerning electronic publication in Acta Cryst. C

Most of the entries in the beginning of the file are requested only if it is intended to publish the structure in one of the journals accepting electronic submission using the CIF as only media of submission. This is mainly Acta Cryst. C for the moment. As the vast majority of the structures done here will be published elsewhere it makes no sense to keep all these entries containing no information concerning the structure. In standard CIF files all entries between (but not including) the entries _audit_creation_method and _chemical_name_systematic can be deleted.

The entries concerning the composition of the compound

The values concerning the composition are derived from the model actually stored in the binary datafile of CRYSTALS (the file usually called crfilev2.dsc) and may therefore be different from the composition expected from the point of view of chemistry. This is usually caused by missing H-atoms that could not be localized or by disordered solvent molecules. Please note that if you correct the composition _chemical_formula_sum, you have to take care to keep the following values consistent: _chemical_formula_moiety, _chemical_formula_weight, _exptl_crystal_density_diffrn, _exptl_crystal_F_000 and _exptl_absorpt_coefficient_mu. It is best to use the tool provided on this site to calculate all these values again starting from _chemical_formula_moiety, the volume of the unit cell and the radiation used. This tool - sorry - is still in preparation yet. Please visit us soon again!

The value _exptl_crystal_density_meas is usually not reported as the determination of the density by flotation is tedious and the use of halogenated solvents causes trouble.

In the CIF-files produced by CRYSTALS the value given for _exptl_crystal_F_000 corresponds to the electron count calculated from the fourier map and is therefore a good measure for the quality of the structure if you compare it to the value calculated from the composition of the structure. The value should be fairly close to the theoretical value, in most cases the error is less than 0.5. It is a good idea to add the calculated value as a comment.