[Pw_forum] LDA norm-conserving pseudopotential for Fe

helen at fh.huji.ac.il helen at fh.huji.ac.il
Thu Sep 22 16:53:09 CEST 2011


Thank you for your help. My problem is the fhi2upf code doesn't  
include values for principle number and assumes a s-p-d-f ordering,  
which doesn't make sense for Fe which doesn't have any occupied f  
orbitals.

   DO i=1, nwfs
      PRINT '("Wavefunction # ",i1,": label, occupancy > ",$)', i
      READ (5,*) els(i), oc(i)
      nns (i)  = 0
      lchi(i)  = i-1
      rcut(i)  = 0.0d0
      rcutus(i)= 0.0d0
      epseu(i) = 0.0d0
   ENDDO



Could they have used 3s3p3d3f and ignored 4s completely in the fhi  
file? Is this the way the fhi2upf code is written because that is how  
fhi files are organized (only s-p-d-f ordering)? Or should I manually  
change the UPF file afterwards to have a more likely 3s3p3d4s ordering?
Thank you again,
Helen

Dr. Helen Eisenberg
Fritz-Haber, Hebrew University, Israel

On Thu, 2011-09-22 at 11:29 +0300, helen at fh.huji.ac.il wrote:
My problem is that it asks me for the label and occupancy of the
   wavefunctions, and I don't know what was used to generate the
fhi file on the abinit site.

guess something sensible, then verify if it works. The wavefunction
label is used only for the printout of projected dos. The l value
is used, but I think that the code always assume s-p-d-f ordering.

P.




----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.



More information about the users mailing list