CCL: question on molecular orbitals in CO and NO

 Sent to CCL by: Brian Salter-Duke [b_duke^^^octa4.net.au]
 On Thu, Nov 10, 2005 at 06:17:07PM -0500, Eric Scerri scerri__chem.ucla.edu
 wrote:
 > Sent to CCL by: Eric Scerri [scerri!^!chem.ucla.edu]
 >
 >
 >
 > In teaching qualitative molecular orbital theory one encounters the
 > question of the crossing of the sigma 2p and pi 2p bonding orbitals
 > in homonuclear diatomics.
 >
 >
 > This is such that N2 has the pi orbitals of lower energy than sigma
 > whereas for O2 the energies are reversed.
 This is an interesting point. The statement about N2 is false. At all
 basis sets greater than minimum basis the sigma level lies lower than
 the pi level. This means that a Koopmans Theorem prediction of the
 ordering of the lowest PES ionisation levels is wrong. The lowest
 ionisation is from the sigma orbital, but this does NOT mean that the
 sigma orbital is the highest occupied. To predict the correct PES
 ordering one has to go beyond Hartree-Fock level. Even the delta-SCF
 method where the total energy of the N2 molecule and the N2+ ion are
 determinated at Hartree-Fock level gives the wrong ordering.
 Introduction of correlation at MP2 or beyond is required to predict the
 correct order of the ion states. I discussed this with Brian O'Leary
 in:-
 Non-Koopmans' Molecules. B. J. Duke and B. O'Leary. Journal of Chemical
 Education, 72, 501 - 504, 1995.
 For an earlier discussion of the orbital energy ordering see "Atoms and
 Molecules" by Martin Karplus and Richard Porter. The diagram is on page
 349 in my edition (1970). I do not know of a later text that looks at
 this in anything like this way.
 At all levels of teaching we should not draw conclusions about the
 ordering of molecular orbitals from experimental PES data. Theory should
 predict experiment and be modified or improved if the predictions are
 wrong. Experiment should not predict theory.
 Eric, I'm happy to run any of these diatomics if you want, but be rather
 clearer about what sort of calculation you want.
 Brian.
 > But what about heteronuclear diatomics where each of the atoms ?
 >
 >
 > The ordering given in general chemistry textbooks shows a wide
 > variation.  See for example Oxtoby, Zumdahl or Olmsted and Williams,
 > all three of which contradict each other.
 >
 > Herzberg's, Spectra of Diatomic Molecules specifically states the
 > configurations of CO and NO.
 >
 > For CO the ordering is as in the case of N2 whereas for NO the
 > ordering is as in O2.
 >
 > However Herzberg's book is now a little out of date (1950).
 >
 > Do more recent calculations on these two molecules suggest anything
 > different?  Would anyone be prepared to run these calculations and
 > tell me the results?
 >
 > How about experimental evidence such as PES on CO and NO?
 >
 >
 >
 > regards,
 > eric scerri
 --
          Brian Salter-Duke (Brian Duke) b_duke * octa4.net.au
        Post: 626 Melbourne Rd, Spotswood, VIC, 3015, Australia
     Phone 03-93992847. http://members.iinet.net.au/~linden1/brian/
 Honorary Researcher, Chem., Melbourne Univ. & Med. Chem., Monash Univ.