This directory contains the crystallographic data for Silver Bromide
along with a transmission scan of a solid solution of AgBr and AgCl
taken at 20, 50, and 80 Kelvin at beamline X11A at the NSLS.  These
data were taken from Bruce's master class course using the old
versions of the codes at 
     http://leonardo.phys.washington.edu/~ravel/course/

This is sufficient information to begin a fitting project for solid
solution sample.  In this readme file, I'll write steps to take to
work through this example.  You should not consider this to be a
recipe -- at each step you should play around with the setting in both
Athena and Artemis to understand fully these data.

These data are mixture of AgBr and AgCl.  Thus we expect to find both
Br and Cl atoms in the first coordination shell.  We will, therefore
need to include information from Feff for both kinds of scatterers in
the fit.

1.  Fire up Athena.  Import all the data.  Find good background
    removal and Fourier transform parameters for each data set.
    Align the data.  Make merged spectra at each temperature.

2.  Save the project and export the chi(k) files (if necessary)
    a. If you are using a version of Artemis that can read the Athena
       project file, then it is unnecessary to save the chi(k) files.
    b. If you are using a version of Artemis which cannot read Athena
       project files, then you will need to save the chi(k) by
       selecting the correct group in the Athena groups list and then
       selecting "Save as chi(k)" form the File menu.

3.  Fire up Artemis.

4.  Import the chi(k) data for the 20 Kelvin data.
    a. If you are using a version of Artemis that can read the Athena
       project file, verify that you are still happy with the Fourier
       transform parameters.
    b. If you are using a version of Artemis which cannot read Athena
       project files, set the Fourier transform and fit range
       parameters to sensible values.

5.  Import the crystallographic data contained in the file AgBr.inp.
    Run Atoms.  Verify that the Feff input data is reasonable, then
    run Feff.

6.  When the path selection dialog comes up, choose to import only the
    first path.

7.  Import the crystallographic data contained in the file AgBr.inp
    for a second time.  This will create a second Feff branch in the
    Data and Paths List.

8.  Click on the first Feff calculation in the Data and Paths List.
    Select "Rename this Feff calculation" from the Theory menu.  At
    the prompt enter "AgBr".  Do the same for the second Feff entry,
    this time calling it "AgCl".  This step is not necessary, but
    makes your project much easier to read.

9.  Edit the atoms data, replacing the Br atom with a Cl atoms.  If
    you know the AgCl lattice constant, change that as well.  If you
    do not know the AgCl lattice constant, leave it untouched.  That
    will result in Feff input data that is incorrect (the AgCl lattice
    constant is considerably smaller than the AgBr constant).
    However, the Feff calculation should be close enough that we can
    begin to analyze the data.  It is common in doing EXAFS analysis
    that we do not know best structure to use in Feff a priori.  It is
    necessary then to start somewhere reasonable and see where the fit
    takes us.  It is in that spirit that will use the AgBr lattice
    constant for the Cl part of the fit.

10. Run Atoms.  Verify that the Feff input data is reasonable, then
    run Feff.

11. When the path selection dialog comes up, choose to import only the
    first path.

12. When the second Feff calculation was started, a second set of
    automatically generated parameters were inserted into the GDS
    page.  In this fit, we will use a single amplitude term and a
    single E0 parameter for both kinds of scatterer.  So discard the
    parameters amp_1 and enot_1 from the GDS page.  Edit the Cl
    scatterer path to use the same S02 and E0 parameters as the Br
    path.

13. Since the distance to the Br and Cl atoms might be different, we
    should have separate parameters fir delta R for the two paths.
    You can use the ones that were automatically generated or you
    might edit them to have more mnemonic names, such as "dr_br" and
    "dr_cl".

14. Similarly we should have separate sigma^2 parameters for the two
    kinds of scatterers.  You might again choose mnemonic names, such
    as "ss_br".

15. Now we need to consider the fact that the first coordination shell
    contains a mixture of Br and Cl.  This sample was, ostensibly a
    50/50 mixture.  On the GDS page create a new parameter called "x",
    assign it a value of 0.5, and make it a set parameter.  Now click
    on the Br scatterer.  Edit its S02 path parameter to read
    "amp*x".  (That assumes that the amplitude parameter is called
    "amp".)  Next edit the S02 path parameter for the Cl scatterer so
    that it reads "amp * (1-x)".  Do you see how this accomplishes the
    chore of populating the first shell with equal quantities of Br
    and Cl?

16. Run the fit.  How does it look?  Are the parameters reasonable?
    You should see that the delta R for Br is slightly negative.  That
    is reasonable given that these data are at 20K and the
    crystallographic data were for room temperature.  The delta R for
    the Cl scatterer should be considerably more negative.  This too
    is reasonable since we used the AgBr lattice constant in the AbCl
    Feff calculation, which we knew to be too large.

17. How could you use this fitting model to measure the populations in
    the first shell?  The answer should involve only one mouse click
    before hitting the big green button.


This strategy of using a mixing parameter as a set or guess parameter
is a common one for situations that require two or more Feff
calculations.  The mixing parameter is a convenient way of correctly
mixing the contributions of different scatterers in a heterogeneous
environment or in a situation involving multiple crystallographic
sites for the absorber.
