Symmetric Stretching CI Potentials
The plot to the right is particularly revealing as to the possible
sources of complexity in the vibronic energy levels of C3.
This is a plot of energy of the linear configuration against bond length,
with both bond lengths held equal. Perhaps the most unusual feature is
the near degeneracy (~300 cm–1) of the 1Σu-
and 1Δu states; these cross the A state at a vibrational
energy of 4500 cm-1. Given that bending will cause
all three of these to mix a complicated energy level pattern can be expected
as this crossing is approached. Calculations at bent configurations indicate
that this mixing is significant, so only a limited range of levels in
the A state will be free from perturbation. It also indicates why the
higher electronic states have not been analysed. Not only are the 1Σu-
and 1Δu states in very close proximity but in
addition to crossing with the lower A1Πu state
at 750 cm-1 they also cross a 1Πg
state at 2500 cm-1. These will all mix at bent configurations,
so the observed spectrum at shorter wavelengths will have contributions
from all four states.
These calculations are close to the best that can be achieved with current programs. They are multi-reference configuration interaction (MRCI) calculations including the Davidson correction using a cc-pVQZ basis set. Selected calculations (where some symmetry is present) were done with an active space including all the valence orbitals while the bulk of the calculations (at Cs symmetry) exclude the two highest energy orbitals. The figure above shows an A – X separation of ν00 = 24810 cm-1 (after correcting for zero point energy) in very good agreement with the experimental value (24676 cm-1) and the vibrational calculations described below also indicate the shapes are good.