[Wien] Au 4f XPS spectra

[Peter Blaha]

You are misunderstanding something.

"Open-core" calculations means to put a fixed number of valence 4f 
states of the rare earth elements into the core and remove them from 
valence. This is done because we know that DFT gives qualitatively wrong 
results for them and pins them at EF.

The Au 4f states are anyway very good core states. The outputst file shows:
   5S      -7.943656     -7.943499  1.00  1.00    0.9997  T
   5P*     -5.276740     -5.275931  1.00  1.00    0.9988  F
   5P      -4.066685     -4.064492  2.00  2.00    0.9963  F
   4F*     -6.100071     -6.099001  3.00  3.00    1.0000  T
   4F      -5.819979     -5.818895  4.00  4.00    1.0000  F
   5D*     -0.583391     -0.563904  2.00  2.00    0.9026  F
   5D      -0.472643     -0.450391  3.00  3.00    0.8765  F
   6S      -0.448308     -0.395939  1.00  0.00    0.4222  F

They are around -6 Ry in the atom and FULLY localized inside the sphere.

I don't know what you have done, but with open-core calculations you 
cannot calculate 4f states as valence.

The question is: What do you want to simulate ?

The valence-XPS (PES-program) is a possible solution for the shape and 
intensity of valence XPS  spectra, and their dependency on the photon 
energy. It can be used for states within 20-30 eV below EF.

For the 4f states in Au I guess the XPS gives you basically 2 sharp 
lines from the spin-orbit splitted 4f core levels.
Their binding energy cannot be obtained from the plain DFT eigenvalues 
(epsilon_4f - EF), because DFT eigenvalues are not excitation energies.

A more "reasonable" XPS binding energy can be obtained with "Slaters 
transition state theory". Create a supercell and on one of the Au atoms 
remove 0.5 4f electrons from case.inc. After scf, these eigenvalues 
should be better approximations to XPS binding energies.



Am 12.10.2025 um 05:49 schrieb Francisco Garcia:
> Dear Prof. Blaha and Users,
> 
> I am interested in calculating the 4f XPS spectra of Au. I want to know 
> if the sequence of steps below is correct.
> 
> Step 1: Treat the Au 4f states as valence electrons using the open-core 
> method. This requires running a regular SCF calculation and making the 
> necessary and careful adjustments to case.inc, case.in1 and case.in2.
> 
> Step 2: With the modified case.inc, case.in1 and case.in2 files, re- 
> initialize dstart and continue the calculations.
> 
> Step 3: Since the 4f states have been forced into the valence region, 
> the recently developed Bagheri-Blaha valence band XPS method can be used 
> to compute the 4f XPS spectra.
> 
> Thanks and let me know your thoughts.
> 
> FG
> 
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

-- 
-----------------------------------------------------------------------
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Email: peter.blaha at tuwien.ac.at
WWW:   http://www.imc.tuwien.ac.at      WIEN2k: http://www.wien2k.at
-------------------------------------------------------------------------