[Ifeffit] Sorbed complex in Atoms

[Matt Newville]

Gnu,

First, we prefer people to use their real names here, because we are all
adults discussing science.

On Thu, Jun 9, 2016 at 11:12 AM, Gnu script <iam.gnu.py at gmail.com> wrote:

> Hi Bruce,
> Thanks for your reply. I am familiar with outer-sphere complexation and
> how it works but I am much less familiar with Artemis and how such question
> can be approached. In the past we used Excurve and DL-Excurve where you can
> simply name the backscatter with initial estimates of N and R but without
> any need for STRUCTURE. Now, having moved away from these tools into
> Demeter and Artemis I am trying to learn rather than asking people to do
> work for me!
>
> So, what I have done so far was to replace Ti in the Atoms file by Ni but
> it is not clear to me what I need to change in order to test outer-sphere
> vs inner-sphere complexes of Ni.
>
>
> space = P b c n
> a     =   8.71280    b    =   5.23270    c     =  14.48700
> alpha =  90.00000    beta =  90.00000    gamma =  90.00000
> rmax  =   7.84905    core  = Ti
> # polarization = 0.0  0.0  0.0
> shift =     0.00000    0.00000    0.00000
> atoms
> # el.     x           y           z        tag
>   Na     0.06430     0.64500     0.15370   Na
>   Ti     0.15130     0.13220     0.33090   Ti
>   Si     0.34210     0.29610     0.52670   Si
>   O      0.00000     0.00600     0.25000   O1
>   O      0.18170     0.44070     0.27580   O2
>   O      0.33360     0.29670     0.41520   O3
>   O      0.00890     0.24280     0.42800   O4
>   O      0.23570     0.05980     0.56700   O5
>
>
> Thanks,
>  G
>
>
>
Atoms represents a crystallographic structure.  That's not exactly what is
need to simulate EXAFS - for that you want a cluster of atoms, as
represented in the feff.inp file.   The Atoms program (or Atoms step of
Artemis) conveniently converts a crystallographic representation into a
cluster of atoms.     Atoms (rightly) refuses to do partial occupancy
because for EXAFS it matters greatly how random any substitution would be,
whereas for simple analysis of diffraction data, it often doesn't.

For what you want to do (and for many other needs), you want to **edit the
cluster of atoms in feff.inp**.     To do this, you would probably start
with your NaTi silicate.    If you want to represent a surface, you could
(for example) remove all atoms with Z > 0.   You can move atoms around by
changing their X,Y,Z coordinates.    If you want to substitute a Ti atom
with Ni, you could do this in two steps:

   1. add a Ni "potential".  If Ni is the absorbing atom, Make "IPOT=0" Ni
by changing the atomic number for ipot 0.
   2. change the central atom and any scatterers as desired.

Again, the crystallographic data for Atoms is a convenient starting place,
but actually as very little to do with the simulation of EXAFS.

Hope that helps,

--Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20160609/ed47d563/attachment-0001.html>