[Wien] binding energies

Sat Dec 18 18:44:34 CET 2004

Careful :

it is true that feff uses a spherical approximation to the potential.
However, this is not what one commonly means by the term 'pseudopotential'.
Abinit and cpmd are pseudopotential codes.
Wien and feff are all electron codes.

Feff uses spherical potentials (at least in the current version available to the public; I expect this will change in the future) but FEFF IS AN ALL ELECTRON CODE.

Kevin Jorissen

EMAT - Electron Microscopy for Materials Science   (http://webhost.ua.ac.be/emat/)
Dept. of Physics

UA - Universiteit Antwerpen
Groenenborgerlaan 171
B-2020 Antwerpen
Belgium

tel  +32 3 2653249
fax + 32 3 2653257
e-mail kevin.jorissen at ua.ac.be

________________________________

Van: wien-admin at zeus.theochem.tuwien.ac.at namens Faisal M. Alamgir
Verzonden: vr 17-12-2004 20:47
Aan: wien at zeus.theochem.tuwien.ac.at
Onderwerp: Re: [Wien] binding energies

Hi Vicki and Gilles (long time no speak),

I would like to make a couple of minor points.

The energies excitation reported by FEFF8 are not reliable in and of
themselves.  Instead you are encouraged to play around with the Vr
parameter to add or subtract to the core potential until you get the
best fit with experimental near-edge spectra.  As Gilles pointed out,
FEFF8 does not calculate the full potential like WIEN does, but has a
soft pseudopotential.

Also, absolute final-state energies are difficult to calculate by WIEN,
but trends (eg.compositionally driven energy shifts) may be easier to
match with experiment.

A lot of things could be happening to give you wrong trends in the
binding energy, so I would suggest first finding out whether WIEN or
FEFF is doing a better job in simulating valence band structure
(assuming you have the XPS valence band data).  [In FEFF you could add
all the LDOS below Fermi energy to "simulate" the XPS valence band] Then
the core level shift trends can be  tweaked fine tuning the model.

I hope some of these points are useful.

Take care,

Faisal 

Gilles Hug wrote:

>Hi,
>I quite fully agree with Gerhard.
>Just a few more words for completeness.
>When I calculated excitation energy for a 1s electron (for X-rat
>absorption for example) I would make the difference in total energy
>between ground state and a "static" excited state. This would a kind of
>first order approximation. Then Wien gives a very good answer, say within
>5-10 eV. You mignt not be happy with it but considering it is a difference
>between two very large numbers, the accuracy is not so bad.
>On the other side FEFF is a kind of GW formalism which goes beyond ground
>state. However, although it has a built in fully relaxed core hole
>approximation, it is NOT a "full potential". It is meant to
>calculate the
>shape of the fine structure, their intensities and relative energy
>positions but NOT their absolute positions. The
>absolute position is calculated from a purely atomic calculation. Therfore
>it is usual to shift by a few eV or so the
>position of the spectral features. This is not harmfull and is documented
>in FEFF's UG (I think).
>So beware that you may have "accidental" good match. It will not be always
>the same.
>(Un?)fortunately, as Gerhard said, their is still some work ahead for
>theorists...
>Joyeux Noel
>Gilles
>
>On Thu, 16 Dec 2004, fecher wrote:
>
> 
>
>>The answer is simple, XPS does not measure the ground state binding energies.
>>
>>This is not just the problem of the N to N-1 electron transition alone (in the
>>sense of calculating just the two cases, ground and excited), but will also
>>depend , to say in not too physical words, on "secondary" effects some known
>>as "multiplett" effects, some as "shake up",  "shake down", etc. as found in
>>textbooks of corelevel photoemission (see.: e.g. reviewing articles by Chuck
>>S. Fadley and others). The results of calculations may strongly depend
>>whether one uses sudden approximation or adiabatic approximation.
>>Calculational results may also strongly depend how sorrowly the transition
>>matrix elements are calculated.
>>
>>With FEFF, the result will depend which potential you use and this is most
>>probably not the same like the one from Wien2k, so it is no wonder that the
>>energies and trends will be different.
>>
>>The problem of comparing calculated and measured photoemission spectra is, to
>>my knowledge, not completely attacked for the most general case, as this is a
>>very complicated many-body problem. However, FEFF seems to be a very advanced
>>one.
>>
>>
>>Ciao Gerhard
>>
>>
>>Am Donnerstag, 16. Dezember 2004 00:37 schrieb Vicki Keast:
>>   
>>
>>>Dear fellow Wien2k users,
>>>
>>>I am trying to compare experimental (XPS) binding energies to
>>>calculations for  S atoms in different Ni sulfides with different
>>>structures and co-ordination.
>>>
>>>The core state energies calculated by Wien are quite different from
>>>experiment and FEFF8 calculations. I assume this is because the zero in
>>>the Wien calculations is the average Coulomb potential not the the
>>>vacuum level. We are also seeing that the trends in the core level
>>>energies are different for Wien than experiment and FEFF.
>>>
>>>Does anyone have any advice on how to go about comparing calculated and
>>>experimental binding energies and also on how reliable we might expect
>>>them to be?
>>>
>>>Thanks and best wishes to you all,
>>>
>>>Regards
>>>Vicki
>>>     
>>>
>>_______________________________________________
>>Wien mailing list
>>Wien at zeus.theochem.tuwien.ac.at
>>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>   
>>
>
>_______________________________________________
>Wien mailing list
>Wien at zeus.theochem.tuwien.ac.at
>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> 
>

_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 9884 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20041218/8585e6d6/attachment.bin