[IUCr Home Page] [CIF Home Page]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Query re: cif_core 2.1 definitions]

Dear Brian,

I received the question that you referred to several days ago together
with a copy of a CIF file. The full text of my reply follows, I would
have thought that it is clear enough, taking into account that the
person who asked the question is an experienced crystallographer! 

I see no need to change anything, though possibly to avoid such
confusion in the future a footnote could be added to both definitions to
say that some refinment programs (e.g. SHELX) combine these two
logically distinct quantities into one number in their input files
because this is required for the structure factor calculation anyway,
but that these programs output the two numbers separately in their CIF
output files as required by the CIF definitions. 

I get many such trivial questions each day, but try to be diplomatic in
my replies.

Best wishes, George

Full text of my reply:

Dear Richard,

If you look up the CIF core definitions you will see that your
CIF file is entirely correct. When I wrote SHELX in the late
1960's, it seemed a good idea save memory by combining occupancy
and site multiplicity into a single number. Many years later 
the authors of CIF decided to quote them separately. Both 
conventions have advantages and disadvantages.

Best wishes, George


Brian McMahon wrote:
> Dear Colleagues
> I have received the following inquiry from a user. Do others see there to be
> a real problem here? Have you any suggestions for a clean resolution?
> Regards
> Brian
> ----- Forwarded message -----
> I'd like to clarify two definitions that are in the 2.1 dictionary. The two
> datanames are _atom_site_occupancy and _atom_site_multiplicity. There is no
> mention of any interaction between the two nor what interpretation should
> occur if the two values are at odds, e.g. _atom_site_occupancy is 1 and
> _atom_site_multiplicity is 2. I mentioned this to George Sheldrick (since it
> is from a CIF written by SHELXL that came to my notice) and he stated that he
> is correct. My view differs. Since the refinement programs explicitly have an
> occupancy set to 0.5 for a atom site of 2-fold symmetry should not the
> _atom_site_occupancy value be *required* to be 0.5 if that is what the
> program is set to (or refined to)? Should this not be independent of the
> _atom_site_multiplicity value which describes the nature of that position
> regardless of how much of an atom is put there?
> ----- End forwarded message -----

Prof. George M. Sheldrick
Dept. Structural Chemistry
Tammannstr. 4
D37077 Goettingen, Germany
Tel. +49-551-393021
Fax. +49-551-392582

[Send comment to list secretary]
[Reply to list (subscribers only)]

Copyright © International Union of Crystallography

IUCr Webmaster