wB97M-V for excited states

Questions
#1

Dear all,
is wB97M-V functional implemented also to compute excited states and General Excited-State Analysis ?
I’m having trouble with the convergence of 20 excited states for a molecule of 20 atoms.

Thank you in advance

#2

Definitely yes. If you have trouble converging, try bigger xc_grid, nl_grid, and more cis_n_roots.

#3

Thanks a lot. I will do some tests. I was already using the grid (99,590)

Best regards

#4

We have never experienced problems with the default grid, which is SG-2 and will be much faster than EML(99,590). More roots with often help by increasing the size of the Davidson subspace.

#5

Technically speaking it should definitey work (assuming that your molecule has enough pi and pi* orbitals for 20 low energy states).

Practically speaking, it is often the case that the wB97 functionals strongly overestimate the energies of CT states, see for example here: The role of excited-state character, structural relaxation, and symmetry breaking in enabling delayed fluorescence activity in push–pull chromophores - Physical Chemistry Chemical Physics (RSC Publishing) . In that case it is probably better to use a functional with less HF-exchange or an optimally tuned RSH.