Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

F(000)

  • To: Multiple recipients of list <coredmg@iucr.org>
  • Subject: F(000)
  • From: Brian McMahon <bm@iucr.org>
  • Date: Thu, 19 Feb 1998 09:55:49 GMT
I received some correspondence from Harry Powell on F(000), which I attach
below. I believe that the current CIF definition for F(000), which is cited
below, is an appropriate definition to maintain in the dictionary, so this
is intended just for your information. The definition does not exclude the
calculation of F(000) as simply the integer number of electrons present, but
I think the wording tends to discourage that, which should be a good thing.

It may be, however, that referees should be advised not to complain about
non-integer reported values. I think the current practice in Chester is to
query F(000) only when it deviates significantly from the value calculated
by PARST. (Syd: I don't see any check on F(000) listed in the current set of
data validation tests. Is it a test run by PLATON? In any case it's no
longer a required item.)

Regards
Brian

> Hello
> 
> I have a small problem with F(000)! I've refined a structure with
> "Crystals" which one of the referees for Acta C objects to (among other
> things which _are_ wrong, so I'm not complaining about those!) on the
> grounds that the calculated F(000) isn't an integer equal to the total
> number of electrons in the unit cell. 
> 
> I've looked into this, and I'm a little disturbed! I'm pretty sure that
> Crystals uses the correct calculation, and that neither CIFCHECK nor PARST
> do (nor does SHELXL-93 {vide infra}).  I've discussed it with David
> Watkin and with George Sheldrick, and their replies seem to agree with my
> conclusions. 
> 
> I've got the following definition from the IUCr;
> ****************************************************************************
> data_exptl_crystal_F_000
>     _name                      '_exptl_crystal_F_000'
>     _category                    exptl_crystal
>     _type                        numb
>     _list                        both
>     _list_reference            '_exptl_crystal_id'
>     _enumeration_range           0.0:
>     _definition
> ;              The effective number of electrons in the crystal unit cell
>                contributing to F(000). It may contain dispersion contributions,
>                and is calculated as 
> 
>                F(000) = [ sum (f~r~^2^ + f~i~^2^) ]^1/2^
> 
>                f~r~   = real part of the scattering factors at theta = 0
>                f~i~   = imaginary part of the scattering factors at theta = 0
> ;
> ***************************************************************************
> 
> I also noticed that SHELXL-93 (I don't have the latest version as I'm not 
> doing much structure solution and refinement at the moment) always gives 
> F(000) as an integer as it converts the individual contributions from 
> each atom type to an integer before summing them.
> 
> Reading further into it, Stout & Jensen (2nd ed, P235) say that "It can be
> shown that the proper value of F(000) is merely the total number of
> electrons in the cell (*but*) this is only true when the Fo's to be used
> with F(000) are on an absolute scale. (...) Small errors in the relative 
> scales of F(000) and the Fo's are comparatively unimportant, however."
> ***************************************************************************
> 
> This is a continuing problem that I have when submitting manuscripts to 
> Acta C! Normally I have less time to worry about it, and manually insert 
> what I know to be the incorrect value in order to keep the referees 
> happy. However, this is really beginning to bug me!
> 
> More to the point, it really is a big problem if the World's most widely 
> used program calculates a value incorrectly for a required parameter in a 
> CIF deposition.
>  
> Please can you advise?
> 
> Harry
> -- 
> Dr Harry Powell, Automation Office, University Library, West Road, Cambridge,
> CB3 9DR.              Telephone: +44 1223 333127         Fax: +44 1223 333160


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