[ase-users] Extracting only the asymmetric unit from a cif file?

[Ask Hjorth Larsen]

Dear Armin Römer,

Am Mo., 26. Aug. 2019 um 21:34 Uhr schrieb Armin Römer via ase-users
<ase-users at listserv.fysik.dtu.dk>:
>
> Dear *,
>
> my name is Armin Römer from the TUM and I am currently encountering a problem with ase, for which I can’t find a solution in documentations, threads and the source code.
>
> My situation:
> I have hundreds of cif- files and I am trying to set up a data set consisting of their structures. However, I need to geometry-optimize everything in vacuum first and for this I need to extract the asymmetric units in the cif-files. The ase-function ase.io.read, however, always builds up the entire unit cell and the function’s arguments don't exhibit any way of preventing this.

This is a reasonable think to want, but as you also say, I think we
don't have an option for it.

I think it would be good if we had a CIF object which allowed
immediate access to the asymmetric cell, plus the primitive cell, plus
the conventional one along with any other information specific to CIF
files.  That way you would read once, and have an object which does
whatever job you want.  In case someone familiar with the CIF format
reads this ...

Best regards
Ask

>
> I didn’t find any helpful information in the documentation and in threads.
>
> I took a look at the source code and tried to locate the point, where the unit cell is constructed. I then tried to copy the source code and modify it so that it leaves out the unit cell construction, but I didn’t succeed and got all the more confused.
>
> I also tried to extract the asymmetric unit from the unit cell with the unique_sites() function of the Spacegroup class, but I also only get the entire unit cell.
>
> Have you ever encountered a similar problem or do you have any idea that could help me overcome this problem?
>
> Thank you in advance and kind regards
>
> Armin Römer
> _______________________________________________
> ase-users mailing list
> ase-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/ase-users