Previously I posted about Pauling points in Quantum chemistry: as one increases the size or sophistication of a calculation [e.g. through a larger basis set, a higher order or perturbation theory, or a larger active space] one does not necessarily get closer to the correct answer. Sometimes the quality of results degrade with increasing the number of degrees of freedom.
My colleague, Seth Olsen has just published a nice paper in J. Phys. Chem. A, A Four-Electron, Three-Orbital Model for the Low-Energy Electronic Structure of Cationic Diarylmethanes: Notes on a ‘Pauling Point’. For the specific case of the dye Michler's Hydrol Blue he systematically increases the size of the active space and observes the effect on the two lowest lying excited singlet states. One has to increase the active space from 3 orbitals to 15 to before one recovers the same transition difference densities. This means increasing the number of Slater determinants by 6 orders of magnitude!