I have a single crystal structure of a coordination polymer and by PXRD I can see, that I'm able to synthesize isotypical structures containing the same ligand and different metal atoms. This is confirmed by Pawley refinements. However, when I try to do Rietveld I am encountering some problems.
1. When I create a CIF file (using the Out_pdCIF macro), bonds within the ligand (implemented as rigid body) are added, which are unnecessary. E.g. bonds through an aromatic ring with 240 pm appear in the list of bonds. How can I prevent TOPAS from adding those?
2. I am also refining the x and y coordinate of the metal atom (which is located on a special position, wherefore I do not refine the z coordinate) and merge the occupancies by occ_merge. TOPAS adds a bond between the symmetry generated metal atoms with 0 pm. How can I prevent this?
3. Since the strucutre possesses solvent accessible voids, which are indeed occupied by solvent molecules, the calculated peak intensities do not match the recorded intensities (which is known to happen for guest molecules being present in pores for e.g. metal-organic frameworks). This significantly influences intensities of reflections at higher 2theta angles. I tried different approaches on solving this problem, including the correction of Variable_Divergence_Intensity leading to good results. But I'm not sure if this would be reasonable, since I did not use a variable divergence slit for constantly illuminated sample length during PXRD measurement. Or is this maybe fine to do, nonetheless?
For the first two challenges, I tried to change r_max, but this only influences the output of append_bond_lengths and not the CIF file. I also tried to implement some atomic_interactions, leading to no changes at all.
Unfortunately, since this concerns novel structures, I am not able to add some files or photos to this question. I am not complete new to TOPAS but I would definetly not consider myself an expert. Maybe someone can help me with or has some opinions on these problems.
Many thanks in advance!
Hello Marcel.
Normally to produce a CIF file for publication from TOPAS of a Rietveld refinement, it is done in a series of steps. First, the Out_CIF( name.cif ) macro is used. A simple CIF file was obtained with only cell parameters, coordinates and Biso parameter. This is done to avoid the mentioned problems (1 and 2).
To complement the rest of the information (correct bond distances and angles), as well as prepare a CIF file suitable for publication, a new CIF can be created with the Platon program (http://www.platonsoft.nl/platon/). Then, all the information is filled in or transferred to a template given by IUCr Journals (https://journals.iucr.org/services/cif/powder.html) or it could also use one of the CIF of my publications as a reference. Feel free to contact me if there is anything you don't understand about this way of generating the CIF file.
For the third problem it is necessary to review the data to be able to give you an appropriate opinion.
- Badges
- Science topic
- Similar topics
- Computer Science
- Program