Author Topic: plotting density of open shell  (Read 5964 times)

cavallo

  • Newbie
  • *
  • Posts: 4
  • Karma: +0/-0
plotting density of open shell
« on: August 28, 2008, 05:15:55 pm »
Hi,

working with an open-shell C(S) symmetric system, I would like to plot the total (alpha+beta) density of the a' and a'' irriducible representations in the xz-plane separately.  From the manual it seems I've to uste TM-5.10, and put some lines into the control file.

It is unclear what lines... what I've understood so far is something like :

$pointval geo=plane
  grid1 vector 1 0 0 range -10,10 points 501
  grid2 vector 0 0 1 range -10,10 points 501
  origin 0  0  0   

however, I think this way I get the total (a'+a'') density into the td.plt file.  Should I modify the first line something like:

$pointval geo=plane mo 1a' 2a' ... up to the last occupied a' MO

to get the total a' (alpha-beta) density ? Smells wrong to me, but this is the best guess I can imagine.
TIA
Luigi

Arnim

  • Developers
  • Sr. Member
  • *
  • Posts: 218
  • Karma: +0/-0
Re: plotting density of open shell
« Reply #1 on: September 01, 2008, 02:25:04 pm »
Hi,

I think, there is no automatic way to do that.
If you use the option
$pointval mo
then each MO is written on seperated files, ordered by symmetry and alpha-beta.
If you want to visualize the density of all the MOs belonging to
the same symmetry, you would have to sum them up somehow.

With
$pointval geo=plane mo 1-10 fmt=xyz
the informations are written as plane 3d coordinates.
Then you have to sum up the values in 4th column, probably using a script.

I hope this helps a bit
Arnim


cavallo

  • Newbie
  • *
  • Posts: 4
  • Karma: +0/-0
Re: plotting density of open shell
« Reply #2 on: September 02, 2008, 02:32:28 pm »

Oh, it helps a lot.  It seems this is the way to go.
Summing up density from the various MO is easy.
Thanks,
Luigi

mpjohans

  • Full Member
  • ***
  • Posts: 26
  • Karma: +0/-0
    • .
Re: plotting density of open shell
« Reply #3 on: September 04, 2008, 10:02:22 am »
Hello!

Another way that should work just fine is to edit the control file and comment out first the a' and then the a" orbital occupations, and run "dscf/ridft -proper". This way, only the density of the a'/a" orbitals will be considered in the analysis part.

Have a nice day,
    Mikael J.
    http://www.iki.fi/~mpjohans

resofidentity

  • Full Member
  • ***
  • Posts: 65
  • Karma: +0/-0
Re: plotting density of open shell
« Reply #4 on: February 15, 2013, 02:51:29 pm »
It appears to me, that you can actually plot any density based on arbitrary subsets (sums) of orbitals by changeing the occupation number and the orbital-number you want in control.
But I think then it is important to set
$sciterlimit 0
!