[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Double core-hole

Dear feffusers
I've been trying to simulate a multi-electron excitation in Li (LiCl) by
changing the subroutine getorb as described in the previous e-mails (see below).
However, during the procedure I found a couple of serious errors.Here they are: 

#1st error messagge#
 
Feff 8.20
  XANES:
 LiCl
 Calculating potentials ...
    free atom potential and density for atom type    0
check number of electrons in getorb.f

This problem can be fixed by changing the 0.001 value to 1.001 in the 
following lines (it is not very elegant but it works): 

-------------------------------------------------------------------
       call getorb (nz, ihole, xionin, iunf, norb, norbsc, iorb,
     1             iholep, nq, kap, xnel, xnval, xmag)
      xk=0
      do 411 i=1,norb
 411  xk=xk+xnel(i)
      if ( abs(nz-xionin-xk) .gt. 1.001) then 			<--------- (default was 0.001)
         call par_stop('check number of electrons in getorb.f')
c        stop
      endif
-------------------------------------------------------------------

However, after a new tentative the calculation crashed with a 2nd error message:
WARNING: fatal error in subroutine corval. Try
  to reduce ca1 in SCF card. If does not help,
SEND bug report to AUTHORS
CORVAL-1
STOP:

I been trying to reduce the ca1 value but it doesn't seem to help.  Is there any other way to overcome such a problem?

Sincerely,
Maurizio



John J. Rehr wrote:
Dear Maurizio,

 Your changes to getorb.f are consistent with my suggestions on how to treat
a double core hole (e.g., a multi-electron excitation).

 However, doing this for Li metal to get a 1s0 2s1 configuration
 is a worst case senario, since then you have two holes in the 1s orbital
and only a single electron in the valence band! Although the procedure I
described will usually work, I won't be surprised if FEFF complains, since
it will try to screen the two holes.  Please let us know how it works out.

  Sincerely,
  John Rehr

  
I have a question concerning the simulation of the 1s double core-hole 
in Li (for example). According to the last John Rehr's
e-mail I tried to edit the subroutine getorb and change the following lines

data (iocc( 3,i),i=1,29) /2,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0, <----- Li 1s2 2s1
1 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0/
data (ival( 3,i),i=1,29) /0,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0,
1 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0/
data (ispn( 3,i),i=1,29) /0,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0,
1 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0/

into

data (iocc( 3,i),i=1,29) /1,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0, <----- Li 1s1 2s1
1 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0/
data (ival( 3,i),i=1,29) /0,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0,
1 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0/
data (ispn( 3,i),i=1,29) /0,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0,
1 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0/
    


  

-- 
Maurizio Mattesini, Ph.D. (room 14204)
Condensed Matter Theory Group
Department of Physics, Uppsala University
Box 530 SE-751 21 (Sweden)
Tel.    +46 18 471 58 58
Mobile  +46 73 970 08 66
Fax     +46 18 471 35 24
http://www.fysik4.fysik.uu.se/