The error about Ge core-level calculation

[qiangxu]

Dear Professor,

I calculated the core-level of Ge in dimond structure using PAW_GGA potential and taking its 3d electrons as core electrons, I got the following information:

running on 2 nodes
distr: one band on 1 nodes, 2 groups
vasp.4.6.34 5Dec07 complex
POSCAR found : 1 types and 2 ions
LDA part: xc-table for Ceperly-Alder, standard interpolation
POSCAR, INCAR and KPOINTS ok, starting setup
FFT: planning ... 1
reading WAVECAR
entering main loop
N E dE d eps ncg rms rms(c)
DAV: 1 0.828317131923E+01 0.82832E+01 -0.22505E+03 32 0.450E+02
DAV: 2 -0.861321340914E+01 -0.16896E+02 -0.15740E+02 48 0.669E+01
DAV: 3 -0.884075443635E+01 -0.22754E+00 -0.22754E+00 32 0.113E+01
DAV: 4 -0.884101571810E+01 -0.26128E-03 -0.26128E-03 48 0.461E-01
DAV: 5 -0.884101574098E+01 -0.22879E-07 -0.22879E-07 32 0.445E-03 0.250E+00
DAV: 6 -0.874354197280E+01 0.97474E-01 -0.52278E-02 32 0.129E+00 0.151E+00
DAV: 7 -0.870029788842E+01 0.43244E-01 -0.98983E-02 32 0.186E+00 0.140E-01
DAV: 8 -0.870158932040E+01 -0.12914E-02 -0.21516E-03 32 0.299E-01 0.238E-02
DAV: 9 -0.870168865698E+01 -0.99337E-04 -0.92027E-05 40 0.695E-02
TOO MANY TRIES REQUIRED
0.5605E-44 0.0000E+00 0.0000E+00 -0.2079E+01 -0.1000E-04

TOO MANY TRIES REQUIRED
mesh R V*r A ,6X, B

TOO MANY TRIES REQUIRED
0.5605E-44 0.0000E+00 0.0000E+00 -0.2079E+01 -0.1000E-04

TOO MANY TRIES REQUIRED
mesh R V*r A ,6X, B

However, when I take its 3d electrons as valence electrons, I can get the correct results. I past my input files in the follwing, can you help me to fix my problems. Thanks a lot.

INCAR:

System = diamond Ge
ISTART = 0 ; ICHARG=2
ENCUT = 240
PREC = Accurate
ISMEAR = 0; SIGMA = 0.1;

ICORELEVEL = 1
GGA=91

KPOINTS:

K-Points
0
Monkhorst Pack
2 2 2
0 0 0

POSCAR:

Ge
5.5836
0.00000000 0.50000000 0.50000000
0.50000000 0.00000000 0.50000000
0.50000000 0.50000000 0.00000000
2
Direct
0.00000000 0.00000000 0.00000000
0.25000000 0.25000000 0.25000000


Regards

Qiang Xu
20/Mar/2009

[admin]

corelevelshifts are not officially supported for vasp.4.6, because testing was done mainly in vasp.5.
however, you can increase the loop parameter NTIMES (in cl_shift.F) from 200 and recompile vasp. This worked for your case. The difference of the 3d-level engeries are negligible compared to vasp.5 (vasp.4.6: -20.1208, vasp.5.:-20.1203 eV)
however, if you need core level energies for your work in future, I strongly recommend to upgrade to vasp.5 which will be released very soon.

[qiangxu]

Thank you very much!