Author Topic: negative MP2 (spin) densities for ROHF spin-triplets?  (Read 5321 times)

Jonas

  • Jr. Member
  • **
  • Posts: 13
  • Karma: +0/-0
negative MP2 (spin) densities for ROHF spin-triplets?
« on: July 13, 2010, 06:35:05 PM »
Dear all,

I am currently trying to plot -- and understand -- TURBOMOLE's Hartree--Fock and MP2 spin densities.
I did ROHF calculations on a small spin-triplet system, and RI-MP2 calculations on top on that. I use TURBOMOLE 5.10, obtained td.plt, sd.plt, mp2d.plt, and mp2sd.plt files via the "$pointval dens" flag in control, and plot with gOpenMol 3.00.

I am somewhat confused that the MP2 density and spin density is actually not positive definite: gOpenMol gives min and max values of -0.010 and 0.014 for mp2d.plt, and -0.340 and 0.874 for mp2sd.plt, respectively.
As e.g. the Hartree--Fock density has min and max values of 0.000 and 42.173, I assume that mp2d.plt and mp2sd.plt are only the MP2 contributions to the total densities, and what I seek is the sum of, e.g., td.plt and mp2d.plt?
Is this correct?
The manual does not state anything about that, but simply that mp2d.plt is "the MP2 density" ...?

Thanks in advance,
Regards, Jonas