Transition Metal d Orbital Splitting

Transition Metal d Orbital Splitting

Postby wtransue » Sun Feb 14, 2016 8:30 pm

I would like to use NBOs to probe d-orbital splitting of transition metal centers in various geometries, and I'm running into results I did not expect. I'm trying to find out if I'm asking NBO6 to perform the wrong subroutines or if there's a flaw in my understanding.

I've set up a calculation on Ni(CO)4 in its standard tetrahedral geometry, and run NBO6 with the keywords NPA NBO NLMO PLOT AONBO=W77. From the natural population analysis, nickel's valence natural atomic orbitals are as follows:
Code: Select all
   4   Ni  1  s      Val( 4s)     0.46760       0.20127
  22   Ni  1  dxy    Val( 3d)     1.80924      -0.30860
  25   Ni  1  dxz    Val( 3d)     1.80845      -0.30876
  28   Ni  1  dyz    Val( 3d)     1.83888      -0.30491
  31   Ni  1  dx2y2  Val( 3d)     1.81572      -0.30789
  34   Ni  1  dz2    Val( 3d)     1.80095      -0.30976

The d orbital energies don't work out nicely in a low energy e set (z^2 and x^2-y^2) and a higher energy t2 set (xz, yz, xy). Following the NBO subroutine (total non-Lewis electrons: 1.94857 or 2.320% of 94), the nickel's valence NBOs are as follows:
Code: Select all
   18. LP ( 1)Ni  1             1.84780    -0.30479
   19. LP ( 2)Ni  1             1.84739    -0.30493
   20. LP ( 3)Ni  1             1.79608    -0.30935
   21. LP ( 4)Ni  1             1.79557    -0.30938
   22. LP ( 5)Ni  1             1.79458    -0.30942
   43. LV ( 1)Ni  1             0.46762     0.20182

The NBOs do seem to appear in an e set (18 & 19 correspond to z^2 and x^2-y^2, respectively) and a t2 set (20, 21, & 22 correspond to yz, xz, and xy, respectively); however, the t2 set is lower in energy than the e set, which goes against expectation from crystal field theory. Am I missing something?
wtransue
 
Posts: 1
Joined: Sun Feb 14, 2016 8:00 pm

Return to General NBO Discussion

Who is online

Users browsing this forum: No registered users and 1 guest