[Ifeffit] Small clusters in Feff

[Bruce Ravel]

On Thursday 06 October 2005 10:21, Kelly, Shelly D. wrote:
> All feff paths in the calculation were used in the model fits.  No
> mention was made in the manuscript of the muffin tin radii used in the
> Feff8.2 calculation based on this small cluster, hence I assume that
> these authors were unaware of the potential pit falls in doing so.
>
>
>
> Using feff7, I have run a comparison of the calculated signal from a
> bounded and unbounded shell of atoms at the same distance.   I found
> that the unbounded shell of atoms has a muffin tin radii that is really
> big because it is not bounded by any other atoms giving rise to an error
> in the calculated signal from these atoms.  Has anyone run this
> experiment with Feff8.2?

I think that Feff8 will have the same problem with muffin tin size in
the case of a small unbounded cluster as Feff6 always has.  The effect
might be mitigated somewhat by self-consistency.  Or not.  Knowing
that would require testing that I certanly have not done myself.

In any case, I would agree with you that it is probably better
practice to use a cluster in the feff input file that extends at least
one shell beyond the furthest shell you intend to analyze.


> I believe that the non-expert users of Feff, find it difficult to
> remember not to use the exact model within Feff because of this
> "feature".  Is there any discussion of adding a hydration sphere option
> into Feff?

I discussed this several times with John back in grad school, but
don't know if anyting formal has been done in Feff8 with respect to an
outer sphere -- but I might just be ignorant.  That's certainly been
known to happen ;-)

Perhaps John can shed some more light on this....

B


-- 
 Bruce Ravel  ---------------------------------------------- bravel at anl.gov

 Molecular Environmental Science Group, Building 203, Room E-165
 MRCAT, Sector 10, Advance Photon Source, Building 433, Room B007

 Argonne National Laboratory         phone and voice mail: (1) 630 252 5033
 Argonne IL 60439, USA                                fax: (1) 630 252 9793

 My homepage:    http://cars9.uchicago.edu/~ravel 
 EXAFS software: http://cars9.uchicago.edu/~ravel/software/exafs/