[Ifeffit] appropriatness of feff model when central atom is not "embedded" (or only 1 layer embedded)

[Andrea L Foster]

Hello All,  can someone please help with this:

(1) feff requires that the adsorbing atom be "embedded"  (that is, 
surrounded by) other atoms in order to calculate potentials correctly, but 
how deeply does the atom need to be embedded? 

For example, if  I construct a model of an adsorbate on a mineral surface, 
where that adsorbate has a full complement of 1st shell coordinating OH, 
O, and H2O groups as appropriate, but if I do not include all the 
uncoordinated f water atoms that would be nearby (so that I essentially 
have a surface in vacuum), will the potentials for my central atom be 
calculated correctly (or correctly enough for use in IFEFFIT)?

a more specific example: the AsO4 tetrahedron bonded to the goethite 
surface, which is made of octahedrally-coordinated iron.  The arsenate 
bonds to the goethite  by sharing 1 edge of the tetrahedron with the 
apices of 2 adjacent Fe octahedra.

a related question that comes to mind: if I just wanted to see the signal 
from the 2 adjacent Fe atoms, could I create a perfectly "useable" FEFF 
model (for IFEFFIT) by limiting my calculations to the small cluster of 1 
arsenate tetrahedron and 2 Fe octahedra?

thanks, andrea

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Andrea L. Foster, Ph.D.
U.S.G.S. WMESC
345 Middlfield Rd., MS 901
Menlo Park, CA 94025
T: 650-329-5437
F: 650-329-5491

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20120307/9bc2f39b/attachment.html>