[Wien] TELNES broadening

Wed Jul 5 22:33:01 CEST 2006

Dear WIEN2k users,

here's an update on  BROADENING EELS SPECTRA FROM THE TELNES2 PROGRAM.

The broadening works in 3 steps :
* Lorentzian core state lifetime broadening
* Lorentzian final state lifetime broadening
* Gaussian instrumental broadening

We are here concerned with the second step, of final state lifetime broadening.  In principle, this lifetime should be calculated.  However, we don't get that data out of the WIEN2k code, and therefore we use a model.  I have just sent a slightly updated version of the broadening program to P. Blaha, which I hope will appear in the WIEN2k updates one of these days.  The program has four models :
0       final state broadening set to tiny constant (i.e., effectively switched off) ;
1       final state broadening is linear in energy above threshold
2       final state broadening is quadratic in energy above threshold
3       final state broadening depends somewhat more cleverly on energy above threshold (Moreau et al.)

Model 3 is new and the reference for this model is Moreau et al, Phys. Rev. B 73 195111 (2006).  I'm interested in people's experiences with this model.  I also encourage all TELNES2 users to read a recent paper by Cecile Hebert, a review paper on EELS calculations with the WIEN2k code, which is now in print in the journal Micron.  This paper also contains a section on broadening with an instructive illustration and references.

Now on a more technical note :
To calculate an EELS spectrum, you first run the TELNES2 program.
To add broadening to this spectrum, you then run the BROADENING program.  This program takes input from the file case.inb .  The TELNES2 writes a case.inb file for you with parameters set correctly for Model 1, which the authors ( K. Jorissen, J. Luitz, C. Hebert) feel is a good compromise (although Model 3 may become default if it proves better).
Here is such an example case.inb (which I'm borrowing from the e-mail I'm replying to ;-)) :
1>Graphite C K edge of first atom.
2> ELNES
3> 1 1 0
4> 0.0000000   1.00000000   0.00000000
5> 8.676E-002   8.676E-002
6> 1     0.00000
7> 1.000000

(The line numbers are of course added by me and shouldn't be there in a real file.)
The first number on line 6 is the index of the Model used.
If you want to try the Moreau model, just replace the index by 3.  Ready to go.
If you want Model 0 (no valence broadening), just put 0 there.  Ready to go.
If you want Model 2, it's a little trickier.  First, put "2" on line 6.  Now, this model needs additional parameters that you have to specify.  Add three new lines :
8> E0
9> E1
10> E2
E0 is the energy of the bottom of the valence band.
E1 is the plasmon energy.
E2 is a dummy variable that is in the file only for compatibility with xspec files, and will be ignored for your calculation.  Put 0.00 there and don't worry about it.  The other two values ought to be given in eV.
You can look at the Muller paper (PRB 57 8181) for more details about this scheme and its parameters.
Personally, I don't like this quadratic scheme too much.

If there are any problems with the TELNES, TELNES2 or BROADENING programs, you're welcome to contact me and explain your situation to me.  (Please do this over the WIEN2k mailing list so that other users can learn from your case, too ; but send files to my private account, as the ML blocks everything over 50K or so.)

Best wishes,

Kevin Jorissen

Kevin Jorissen
University of Washington
Dept. of Physics
Box 351560
Seattle, WA-98195-1560
U.S.A.

phone  +1 206 543 3904
fax      +1 206 685 0635

e-mail kevin.jorissen at ua.ac.be
web page http://fraangelico.phys.washington.edu/~jorissen

________________________________

From: wien-bounces at zeus.theochem.tuwien.ac.at on behalf of Andrew Scott
Sent: Wed 6/14/2006 9:58
To: A Mailing list for WIEN2k users
Subject: [Wien] Re: TELNES broadening

I am interested in applying energy dependent final state lifetime (Muller: E2) broadening to my energy loss spectra.  The *.inb file is shown below.

Line 5: core state broadening (constant)

Line 7: spectrometer broadening (constant)

Line 6: if I input '2' I get the error: "Not enough broadening parameters specified!"

Is this option working for absorption spectra?

If so, what values do I need to input in lines 6, 8, 9 and 10?

*.inb file:

Graphite C K edge of first atom.                                                

ELNES

  1  1  0

   0.000000000        1.000000000        0.000000000    

   8.6762718359103701E-002   8.6762718359103701E-002

           2    0.0000000000000     

   1.000000000000000     

   0.000000000    

   0.000000000    

   0.000000000    

I am using the latest version of the code 

Best wishes, Andrew.

********************************

Dr Andrew Scott

Academic Research Fellow

Institute for Materials Research

University of Leeds

Leeds LS2 9JT

United Kingdom

e-mail: a.j.scott at leeds.ac.uk

tel:    (44) (0)113 343 2573

fax:    (44) (0)113 343 2384

********************************

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 11229 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20060705/a60d9957/attachment.bin