ricc2 - -> low lying complex roots

[mbeyhan]

Hi,

I'm trying to do an ricc2 calculation and I'm getting this in the output:

warning> found low lying complex root...
maximum overlap between eigenvectors: NaN

And excitation energies for the first two lowest states are not given but instead it says "Warning! Low lying complex roots".

Does anybody knows what could be the reason for this?

[christof.haettig]

It might be that the SCF reference wavefunction is not a good description of the ground state. Or you are in a region of the spectrum where double excitations play a considerable role. Another reason for troubles of this kind might be that one is close to a conical intersection or avoided crossing.  There is a small discussion on this topic in the manual.

Christof Haettig