I am wondering whether it is possible to run only one Q-Chem calculation that would produce gradients of several TDDFT states. It looks like a TDDFT calculation with
JOBTYPE = FORCE
requests CIS_STATE_DERIV, which can be specified only to a specific state.
I can see a selector of several states only for $derivative_coupling section, but not for gradients.
ORCA, for example, has such gradient selectors (SGRADLIST and TGRADLIST).
Not that I’m aware of but you can always set it up as multiple jobs in the same input file, use SCF_GUESS = READ in the subsequent jobs. Shouldn’t be much extra effort since the CPSCF equations would need to be solved for each excited state anyway.
Thank you! This idea will work if the electronic states are not degenerate. However, for the degenerate electronic states this may not work because the subsequent diagonalizations may deliver different eigenvectors spanning the same degenerate subspace.