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

Re: CoreCIF revision 2.3 ATOM_TYPE_SCAT



> COMMENTS ON #6
> --------------
> ATOM_TYPE_SCAT category



>           _atom_type_scat_factor_id
>           _atom_type_scat_factor_type_symbol
>           _atom_type_scat_factor_stol
>           _atom_type_scat_factor_scat_factor_real
>           _atom_type_scat_factor_scat_factor_imag
>           _atom_type_scat_factor_rad_type

   The names are cumbersome. 'factor' strikes me as being redundant and
'_scat_factor_scat_factor' overdoes it a bit. How about:

          _atom_type_scat_type_id
          _atom_type_scat_type_symbol
          _atom_type_scat_stol
          _atom_type_scat_real
          _atom_type_scat_imag
          _atom_type_scat_rad_type

   I'm also very suspicious because only the real and imaginary
contributions to the total (both resonant and non-resonant) scattering
factors are given names. In practice, for doing electron-density
calculations for X-ray diffraction ("Fourier maps") one really needs the
resonant (anomalous dispersion) and non-resonant scattering factors to
be separated. The non-resonant scattering factor is independent of
wavelength. The resonant scattering contributions are dependent on
wavelength but, to a very good approximation, independent of sin(theta).

> #   _atom_type_scat_dispersion_imag
> #   _atom_type_scat_dispersion_real
> #   _atom_type_scat_dispersion_source

   Since you are recategorizing them, and/or perhaps renaming them, the
modern tendency is to drop the term 'anomalous dispersion' [what's
anomalous about it? Dispersion categorizes a change in property with
wavelength but does not identify its physical origin.] and use 'resonant
scattering' which clearly identifies its well-understood physical
origin. So its worth thinking about:

#   _atom_type_scat_resonant_imag
#   _atom_type_scat_resonant_real
#   _atom_type_scat_resonant_source

These names are clear as they identify the resonant contribution to the
scattering process in contrast to _atom_type_scat_real and 
_atom_type_scat_imag which give the total real and imaginary
contributions.


BMc>     loop_
BMc>     _atom_type_symbol
BMc>     _atom_type_oxidation_number
BMc>     _atom_type_number_in_cell
BMc>     _atom_type_scat_dispersion_real
BMc>     _atom_type_scat_dispersion_imag
BMc>     _atom_type_scat_source
BMc>       C  0  72  .017  .009  International_Tables_Vol_IV_Table_2.2B
BMc>       H  0  100  0     0    International_Tables_Vol_IV_Table_2.2B
BMc>       O  0  12  .047  .032  International_Tables_Vol_IV_Table_2.2B
BMc>       N  0  4   .029  .018  International_Tables_Vol_IV_Table_2.2B
BMc> If the scattering factors are dependent on wavelength and
scattering
BMc> angle, what exactly do the numbers in the above example refer to
anyway?
   _atom_type_scat_dispersion_real and _atom_type_scat_dispersion_imag
refer to resonant scattering contributions only (anomalous dispersion)
which are assumed to be independent of sin(theta), and for a fixed,
unspecified or default, single wavelength. I understand that does not
answer Brian's question. What is saying is that the above loop
construction lacks a pointer or link to identify the wavelength for
which the resonant contributions are being specified.

  One of the worst case situations which this construction would
hopefully be able to deal with occurs when the absorption edge of the
element specified in _atom_type_symbol falls in between the alpha1 and
alpha2 wavelengths of the sealed-tube radiation source. eg Re with
Mokalpha.
   
> Is it the case that items like the Cromer-Mann coefficicients (which, as I 
> understand it, are used in an analytic approximation to scattering factors 
> in a function of lambda and theta)

  Cromer-Mann coefficients are analytic approximations to the
non-resonant contribution to scattering which is independent of
wavelength. The C-M coefficients have stol has independent variable.

BM> A more distinctive name for such a new category would
BM> be ATOM_TYPE_SCATTER; the distinction between ATOM_TYPE_SCAT and
BM> ATOM_TYPE_SCATT would be lost to the average user).

   I agree with you 5000% on that. User understanding is one big
problem. Also the typographic difference is tiny and in a rush that
final S will be overlooked.
   Nature has finally resolved the problem of my tearing my hair out
over REFLN and REFLNS.

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


Copyright © International Union of Crystallography

IUCr Webmaster