Pseudo-tetragonal twinning

4/6: Find true space group symmetry using Platon

To check for missed symmetry, we'll use ADDSYM in Platon. For this, Platon needs a cif and a 'LIST 4' format fcf file, which are written by SHELXL if both 'ACTA' and 'LIST 4' are present in the ins file. These files:

monI.cif
monI.fcf

On reading the cif  into Platon and clicking 'ADDSYM' (top of the SYMMETRY column), we get the following:

ADDSYM suggests a primitive monoclinic cell with a quarter the volume of the I-centered cell. [Note: At the time of writing (Nov-Dec 2020), the latest available version(s) of Platon might require you to click on one of the options on the right hand column (e.g. NonFitPerc) before it makes the transformation to monoclinic-P ]. It also (on the lower right) gives the option to write out a new SHELXL res file with the structure transformed for P2/n. Go ahead and create the file as we'll use it later - it will have '_pl.res' appended to the file name stem. A copy is available here:

monI_pl.res

ADDSYM also supplies a matrix to transform from the I-centered cell to the P2/n cell:


The above transformation generates many non-integer reflection indices. Normally, that is not a problem as they'd be simply ignored or deleted; non-integer indices do not represent actual Bragg peaks. The twinning, however, causes such 'impossible' reflections to coincide with Bragg maxima from other twin domains, so we cannot not simply discard them.

The relationship between the I-centered tetragonal pseudo-cell and the true monoclinic-P cell of each individual is shown below:


There is no way to take account of the four individuals for this structure using the TWIN command in SHELXL while maintaining the correct space group symmetry, even with an unconventional setting. The way forward is to make a data file in 'HKLF 5' format that preserves all the information. At the time of writing, I'm not aware of a way to do this in Platon without first transforming to P2/n. Unfortunately, for the reason mentioned above, that would eliminate about half the observed intensities. In section 5, 'Generate 'HKLF 5' format dataset', we'll manually prepare an 'HKLF 5' format file. It's a bit tedious, but not difficult.