CCL: vinyl group twist

Detlev--
 I think you are correct about the triplet state being the lowest electronic
 state in twisted vinyl-X compounds, but would the reaction really proceed
 through *two* intersystem crossings (singlet to triplet then back to
 singlet)?
 --David Shobe
 On Mon, Aug 29, 2022, 4:09 PM Detlev Conrad Mielczarek detlevcm{:}
 googlemail.com <owner-chemistry..ccl.net> wrote:
 >
 > Sent to CCL by: Detlev Conrad Mielczarek [detlevcm%googlemail.com]
 > If you are twisting through a double bond, I would expect at least one of
 > the bonds to break (forming. Asingle bond) and see the transition progress
 > from a singlet through a triplet state before reforming the double bond and
 > becoming s singlet again.
 > (The ORCA manual contains an example of twisting an ethene molecule.)
 >
 > In this case, you would need to ideally treat this with a multi-reference
 > method such as CASSCF.
 >
 > Running a transition state search in a single multiplicity will invariably
 > lead to incorrect results in such cases, even if some methods seem to work
 > through them.
 >
 > You may be able to use two separate calculations to obtain a valid energy
 > value for the transition state.
 >
 > Detlev
 >
 >
 >
 >
 >           Original Message
 >
 >
 > > From: owner-chemistry++ccl.net
 > Sent: 29 August 2022 21:21
 > To: detlevcm++googlemail.com
 > Reply to: chemistry++ccl.net
 > Subject: CCL: vinyl group twist
 >
 >
 > CCL group--
 >
 >
 > I am trying to find transition states for twisting the double bonds in
 > vinyl-X compounds (X is any group, not necessarily halogen), with the
 > torsion angle going from 0° to 180°, through a transition state
 that should
 > have a torsion angle of approximately 90°. Of course, I expect these
 to be
 > high-energy transition states. I have tried both opt=ts with a few values
 > of X, and once tried opt=qst3.
 >
 >
 >
 > The opt=qst3 method generated something weird, in which the vinyl group
 > had a linear H-C=C angle (instead of 120°). The other distal H on the
 > vinyl's methylene group had a H-C=C angle of 96°  It is a transition
 state
 > (1 imag freq), but definitely not the one I am looking for.
 >
 >
 > Using opt=(usually along with noeigentest), I invariably end up with a
 > much-lower-energy transition state involving torsion of the vinyl group as
 > a whole in the vinyl-X molecule, or with a torsion of some subgroup within
 > X.
 >
 >
 > Is it normal for attempts to calculate a high-energy transition state to
 > end up with a different, lower-energy transition state? More importantly,
 > is there any way to avoid this phenomenon and get the desired transition
 > state?
 >
 >
 > --David Shobe
 >
 >
 >
 > -= This is automatically added to each message by the mailing script =->
 >
 >