CCL:G: Help with Geometry Opt for first singlet excited state usingTDDFT

 Sent to CCL by: Mariusz Sterzel [m.sterzel^^^]
I disagree, it is possible to optimize excited states geometry with gaussian but it is computationaly very very expensive. I agree that TURBOMOLE is much better choice since it does have analytical gradients. This also allows you to calculate harmonic frequencies.
 Anyway going back to gaussian, this question was asked several times.
 The answer I got some time ago from
 1) build _GOOD_ z-matrix for your molecule
 2) optimize ground state with opt(z-matrix)
 3) calculate frequencies at optimized geometry
 4) use following input for say 3rd excited state opimization
 TD(NStates=5,Singlets,Root=3) Opt(ReadFC,Z-matrix,EnOnly)
in addition to your usual gaussian settings. You can also use forces from CIS calculation.
Gradients are numerical so use as high symmetry as possible for optimization. And you cannot calculate frequencies so I'd rather use TURBOMOLE (which also have analytical gradients for excited states at CC level).
 Mariusz Sterzel
 Email: m.sterzel/^at^\
 On Thu, 9 Mar 2006, Jens Spanget-Larsen wrote:
 Sent to CCL by: Jens Spanget-Larsen [spanget[A]]
 Dear CY,
 TD-DFT geometry-optimization of an excited singlet state is not possible
 with Gaussian03. Turbomole has this facility.
 Jens >--<
   JENS SPANGET-LARSEN         Office:      +45 4674 2710
   Department of Chemistry     Fax:         +45 4674 3011
   Roskilde University (RUC)   Mobile:      +45 2320 6246
   P.O.Box 260                 E-Mail:     spanget+*
   DK-4000 Roskilde, Denmark
 Changyong Qin changyong(-) wrote:
 Sent to CCL by: "Changyong  Qin" [changyong|-|]
 Dear All,
 Does anybody have experience in using TDDFT in Gaussian 03 to optimize the first
 singlet excited state? If yes, could you please share the input file ? Thanks.