Author Topic: quadrupole moments  (Read 6918 times)

a.karl

  • Newbie
  • *
  • Posts: 1
  • Karma: +0/-0
quadrupole moments
« on: February 13, 2008, 10:17:51 am »
Hi to all.
For my study, I want to calculate the dipole, quadrupole and octupole moments of my complexes.
As a test case,  I tried the H2 molecule. The molecule is oriented along the z axis and the center of mass is located at (0,0,0).
According to literature, due to symmetry reasons the components of the quadrupole moments should be:  Θxx = Θyy = -Θzz/2 .
However, when  i run turbomole (DFT or MP2) , I get completely different results.
Below, I paste the results from  a riMP2/def2-TZVPP optimization.
I also get the same results, if I run a properties calculation with the  $moments keyword.

              nuc          SCF,e       -> SCF,tot          MP2       ->  total
 ------------------------------------------------------------------------------
                     quadrupole moment
 ------------------------------------------------------------------------------
  xx       0.000000      -1.550279      -1.550279       0.021371      -1.528908
  yy       0.000000      -1.550279      -1.550279       0.021371      -1.528908
  zz       0.969885      -2.060752      -1.090867      -0.001570      -1.092437
  xy       0.000000       0.000000       0.000000       0.000000       0.000000
  xz       0.000000       0.000000       0.000000       0.000000       0.000000
  yz       0.000000       0.000000       0.000000       0.000000       0.000000

     1/3  trace=      -1.383418
     anisotropy=       0.436472
------------------------------------------------------------------------------

As you can see, the program gives no Θxx = Θyy = -Θzz/2  relation.

Other qc programs, like Gaussian, Gamess and Molpro give the expected results.
Do you have any suggestions?
Do I have to specify a special keyword in the control file?

Thanks in advance,
Antonio.

christof.haettig

  • Global Moderator
  • Sr. Member
  • *****
  • Posts: 265
  • Karma: +0/-0
    • Hattig's Group at the RUB
Re: quadrupole moments
« Reply #1 on: March 04, 2008, 10:32:28 am »
Hi,

there exist different conventions for quadrupole moments in the literature.
What is printed by default in the output of SCF and MP2 (gradient) calcualtions
is the quadrupole moment with trace (also known as the second-moment of charge).
You refer to the traceless quadrupole moment which you obtain by subtracting
the trace (i.e. (xx+yy+zz)/3) from the three diagonal elements. The traceless quadrupole
moments (for MP2) are also found in the output of ricc2, if you require in the input
the calculation of quadrupole moments with fop keyword in $response).

Christof