XAS at the 2p edge

[Johannes Niskanen]

Hi,

I am trying to calculate 2p XAS with CP2K. I use all-electron basis set for 
the atom to be excited, whereas other atoms have pseudopotentials, which 
makes identification of orbitals a bit easier, as I can always count 1 to 5 
from snapshot to snapshot.

However I was wondering whether I get the matrix elements as an average 
over all the 3 states or only for one of them. My input reads
   &XAS T 
      XAS_METHOD TP_HH 
      EPS_DIIS 1.0
      MAXSTEP 100
      STATE_TYPE 2P # Type of the orbitas that are excited for the xas 
spectra calculation (default = 1s) 
      ATOMS_LIST  1 # Indexes of the atoms to be excited. 
      STATE_SEARCH 5 # of states where to look for the one to be excited 
(default = 1)
      ADDED_MOS 100 # Number of additional MOS added spin up only (default 
= 1)
      &LOCALIZE T 
        LIST  1 2 3 4 5
      &END LOCALIZE
    &END XAS

If I get the amplitudes only for one of the nearly degenerate 2p orbitals, 
how would I be able to do calculations all of them? Putting 3 or 4 in state 
search does not work, tried that. 

Naturally I will add spin orbit splitting ad hoc afterwards, but I would 
first need to know a working input for amplitudes for ALL 3 of the 2p 
levels.

Thanks for help in advance,

Johannes Niskanen, PhD
Department of Physics
University of Helsinki
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20140423/f19af5bd/attachment.htm>