TD-DFT Excited state geometry optimisations using wB97M-V - Analytical gradients

Questions
#1

Hello everyone,

I tried running a singlet excited state geometry optimisation in TD-DFT, using wB97M-V and the basis set def2-SVP. At the top of the output, after the references and the declaration of the environment where Q-Chem 5.4 is started, I get the following warning:

!!********************************************.
!!******************Warning*******************.
!!Desired Analytical derivatives not available.
!!Finite difference job might take a long time.
!!********************************************.
!!********************************************.

It seems to me that the diagnosis is quite simple: for wB97M-V, analytical gradients are not available, because by using a different functional (PBE). I want to understand if I’m doing anything wrong, since this is one of the very first excited state geometry optimisations I’m running. I am also curious if there are other modern range separated hybrid (RSH) functionals that are supported in Q-Chem for such kind of task.

Thank you for your time :slight_smile: I wish everybody a good day!

#2

Your diagnosis is correct - no excited-state gradients. It’s the -V (nonlocal VV10 correlation) in the functional that’s the issue; those gradients are under development but not implemented yet. RSH functionals such as wB97X-D, LRC-wPBE, etc (that do not use VV10) do have analytic gradients.

1 Like
#3

Hello John,

Thank you very much for the immediate reply. I will use one of the other RSH functionals that you suggested for my excited state geometry optimisation calculations :slight_smile:

Have a great day!