iOpenShell » Electronic Structure and Quantum Chemistry » Basic questions concerning TD-DFT...need help!

Basic questions concerning TD-DFT...need help!

Moderators: kadir, krylov, piotr.

Page: 1

Author Post
Registered: Dec 2012
Posts: 3
I am mainly an experimental chemist, and use theory to compliment experimental results (so for the guru's out there, go easy on me, haha!). Over the years, I have been trying to teach myself the basics of the theory I need in order to do the calculations I need, which has been very frustrating at times and very fruitful in others!

Anyways, the latest calculations I am trying to do is a TD-DFT calculation to calculate the transition energies and oscillator strengths in a Cu(I) transition-metal dimer complex that has a halogen bridge between the two moieties of the dimer (not the easiest molecule to begin with, I know!!). The command line for my input file is as shown below:

#p TD=(NStates=10,50-50,Direct)/gen Density=Current GFINPUT Pop=(NBO,FULL) nosymm

my questions are these:

1) where in the calculation input do I indicate, if I even do or can, the functional I used for the previous geometry optimization? does this sort of TD-DFT calculation even need/account for specific functionals or does it only require basis set inputs?

2) are TD-DFT transition energies, etc...meaningless if calculated with a different basis set than the previously calculated geometry optimization whose coordinates are used as the input for the TD-DFT calculation?

3) when calculating molecular orbitals (with the Pop=(NBO,FULL) command) are these orbitals more "trustworhty" when done in the TD-DFT calculation as above, or as part of the ground state geometry optimization?

Thanks in advance for any insight and help anyone can give into these issues :)
Registered: Aug 2008
Posts: 43
1) think of td as coming after the functional and basis set declaration, so you could put td any place in the #p line, whereas the functional and basis set need to be coupled by the "/".

eg #P tda(50-50,nstates=40) pbe1pbe/genecp 5D 7F scf(tight) Int(acc2e=14,Grid=SuperFine)

2) No, but some care should be taken to ensure that geometry would not change much if the larger basis set were used. Optimizing the geometry with some combination of 6-31G(d), 6-31G(d,p), 6-31+(d), 6-311G(d,p), etc. should give you enough of an idea.

3) Sorry, I can't help you with this one.

Page: 1

iOpenShell » Electronic Structure and Quantum Chemistry » Basic questions concerning TD-DFT...need help!