questions regarding FEFF from Removic

[Bruce Ravel <ravel@phys.washington.edu>]

Katarina,

Firstly,  sorry it has taken me a few days to respond to your email.  I
was at the APS last week running an experiment.

Secondly, you will notice that I took the liberty of sending this
message to the FEFFUSERS mailing list.  You should join it as it is
the most appropriate place to send questions of the sort that you ask.
Point your browser at

  http://leonardo.phys.washington.edu/feff/Notes_comments/feffusers.html

and follow the instructions.


Now, on to your questions:


 KR> 1)I need polarization card in the program but I don't know how to
 KR> utilize this card. Actually I don't know what is the meaning of
 KR> x,y,z components in this card.

These three numbers define a direction in space (specifically, a space
that is 3-dimensional and Cartesian) for the electric vector of the
photons incident on your sample.  I think an example would be useful.

Suppose your feff.inp file was for tetragonal PbTiO3.  Here is the
atoms.inp file:

     title PbTiO3 25C
     title Glazer and Mabud, Acta Cryst. B34, 1065-1070 (1978)
     core=ti    space  P 4 m m    
     a=3.905    c=4.156
     rmax=3.6
     atom
     ! At.type  x        y       z      tag
	Pb     0.0      0.0     0.0     
	Ti     0.5      0.5     0.539  
	O      0.5      0.5     0.1138  axial
	O      0.0      0.5     0.6169  planar

If you wanted to do a calculation with the incident polarization along
the z-axis, you would need to put this line in the feff.inp file:

     POLARIZATION 0 0 1

Thus the direction of polarization is defined by the orientation of
the cluster in the feff.inp file.  In this example, the tetragonal
axis is the c-direction and the c-direction is along the z-axis in
space.  Thus, the polarization must be in the (001) direction.


 KR> 2) is it important what is the direction of the E vector in the
 KR> XY plane i.e. do I should get the same results with (1,0,0) and
 KR> (1,1,0) polarization card. Some people said to me that it
 KR> shouldn't be difference but it isn't true according to the FEFF6
 KR> calculation.

Well, this depends on what the XY plane looks like.  If there the
symmetry in the plane is square and four-fold (as it is in the example
above), you can convince yourself with simple geometric arguments that
it does not matter which direction you choose.  You could also just
run feff for a few different directions in the plane and convince
yourself that the chi.dat file that comes out is the same (within
numerical precision) for each.

However, if you do not have square symmetry in the plane, then you
may need to be more clever.  If your sample is perfectly textured
single crystal, then you can, in fact, orient the beam along a
particular direction.  In that case, you should choose the
POLARIZATION vector appropriate to the experiment.  However, it is
common for a sample to be well oriented along one direction (probably
the direction parallel to the substrate surface normal), but
unoriented in the plane.  In that case, the plane is randomly oriented
in two dimensions relative to any incident beam direction.

A surface which is randomized in the plane and which has linearly
polarized incident radiation is, in a practical sense, identical to a
single crystal with circularly polarized incident radiation.
Fortunately FEFF can compute the latter using these keywords in the
feff.inp file:

  POLARIZATION   1 1 0
  ELLIPTICITY  1 0 0 1

That says to consider a vector in the plane ( (110) is in the XY
plane), but which is rotating about the 001 axis with unit
ellipticity (a circle has unit ellipticity and ellipticity approaches
0 as the eccentricity diverges, i.e. approaches linear).  So using
those two cards will do a good simulation of linearly polarized light
incident on a sample randomly oriented in the plane of the substrate.

If your sample is actually textured but not fully textured in the
plane, then the suggestion above is not valid and you may need to do
something even more clever, but that will probably depend on the exact
details of your experiment.

 KR> 3)how to take in consideration the L2 and L3 edges in the same
 KR> time.  I will be grateful to You if You have some answer for me.

You cannot.  You have to run them separately and add them together by
hand.  Drat.



As a final note, you may find the materials at

  http://feff.phys.washington.edu/~ravel/course/

to be useful.  The issue of polarization is addressed to some extent
in chapter 3 of the notes.pdf file.


Hope that helps and don't hesitate to submit these queries to the
FEFFUSERS list in the future.

Regards,
Bruce


-- 
 Bruce Ravel  ----------------------------------- ravel@phys.washington.edu
 U.S. Naval Research Laboratory, Code 6134          phone: (1) 202 767 5947
 Washington DC 20375, USA                             fax: (1) 202 767 1697

 NRL Synchrotron Radiation Consortium (NRL-SRC)
 Beamlines X11a, X11b, X23b, X24c, U4b
 National Synchrotron Light Source
 Brookhaven National Laboratory, Upton, NY 11973

 My homepage:    http://feff.phys.washington.edu/~ravel 
 EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs.html