DIFRAC

DIFRAC System design

Table 1 presents the list of items that we considered essential to include in the converted single-crystal diffractometer output file. This information is intended to be complete and independent of the particular instrument used. It serves for the calculations of structure determination and refinement and contains the items considered necessary for machine submission of documents for publication or data bases. Some of the items we have chosen to include or exclude from our output list require special comment:
  1. The chemical formula has been included in the converted output file although in some cases this might have to be revised as a result of the crystal structure analysis. The formula is useful in identifying the compound, and the element names and composition are necessary for running structure-solution and refinement software. Moreover, the subroutine (see below) which undertakes the interpretation of the chemical formula was already available to us.
  2. On the other hand, the space group of the crystal has not been included. For a crystal of unknown structure, the exact space group will not be determined reliably until after structure refinement. Further it was thought that flexible software to test the integrity of the input of space group information in any form (e.g. a symbol, generators, or individual symmetry operations) would have greatly lengthened the time necessary to produce this version of DIFRAC.
  3. The crystal orientation matrix (or matrices in the case of resetting) for the diffractometer (often called the UB matrix) and the diffractometer setting angles (omega, two-theta, chi, phi or kappa) for each reflection have not been included in the converted output file. All orientation information has been condensed into a crystal-based azimuthal angle psi for each reflection measurement. This psi is independent of the instrument geometry and axial definitions of the instrument. Full details are to be found in Schwarzenbach and Flack, (1989, 1992).
  4. An index or code is included to identify the reference reflections. In our view it is not possible to correctly identify a reference reflection from its Miller indices alone. The inclusion of an index eases the work and programming of data-treatment and refinement software analysing the variations in the reference reflections.
  5. The file includes a fairly detailed description of the incident-beam characteristics. This has been done to allow the calculation of polarization and beam-inhomogeneity corrections (Harkema, Dam, Van Hummel and Reuvers, 1980; Markov, Fetisov and Zhukov, 1990).

A choice had to be made concerning the format of the converted output file. The necessity of being able to transmit the file over networks or by electronic mail imposes a clear-text (ASCII) form with no more than 80 characters per line. Additionally we required that the content and format of the file be clearly defined and relatively compact. The CIF - Crystallographic Information File (Hall, Allen and Brown, 1991) is available for just this purpose and we have it.

The information to be included in the converted output file is obtained from the following sources: (a) fixed geometrical details of the single-crystal diffractometer, (b) information in the diffractometer parameter file, (c) information in the diffractometer data file and (d) interactive questions and answers. An underlying design principle in this arrangement is that information from the diffractometer files should be used, if available, otherwise a question will be asked just once only. This is most important to avoid confusion and frustration on the part of the user. It further explains our preference for installing the executable programs on the computer controlling the diffractometer so that each installation may be customized for its instrument and have access to those files which are not specific to the user.