Dear Sue
It is certainly true that this mixing of the LUMO and
LUMO+1 is the problem causing the mixing of the two transitions. Usually such
mixings occur at transition states. Is it a transition state or just a local
minima?
The systems we observed
such interchanges are _not_
symmetric either. Thus, we forced them to be. Maybe you are able to do something
similar.
The _calcall_ calculates the forces at each
point in the optimization. Thus, the minimization algorithm can easily step into
the direction of the largest force contribution. This speeds up convergence (or
makes it possible for shallow potentials), but strongly slows down your
calculations. So use it just in case you have convergence
problems.
Cheers
Markus
Von: owner-chemistry[A]ccl.net [mailto:owner-chemistry[A]ccl.net]
Gesendet: Dienstag, 20.
November 2007 04:04
An: Thut, Markus
Betreff: CCL:G: AW: CIS Optimization for the First Singlet
Excited State
Thank you very much for your suggestion.
The compound that I have studied do not have any symmetry. In my
case, the LUMO and LUMO+1 is mixed. That might caused the problem for the
interchange of the excited states.
The basis set that I used is 6-31g* for all atoms. Is
the optimization more easiler to get converged with Calcall
keyword?
"Markus
Thut markus.thut[-]iac.unibe.ch"
<owner-chemistry|-|ccl.net>
wrote:
Sent to CCL by: "Markus Thut"
[markus.thut+/-iac.unibe.ch]
Dear Sue
We usedd to encounter this
problem too. What we found is that
i) upon interchange of the excited
states, the electronic character of the
two states interchange. Thus, a
"major" change in geometry occurs upon this
change. Please check the
evolution of your optimized molecule.
ii) Usually,
such interchanges occur due to symmetry-breaking optimization
steps.
Sometimes, you can prevent interchanges by restricting the
optimization to
e.g. Cs instead of C1, even though the cluster/molecule you
are optimizing is
presumably C1.
iii) if you have a symmetry wich allows optimizations for
a' or a"
transitions, try to optimize for the desired transition (you can do
this in
TURBOMOLE, I'm not sure about Gaussian03). In this case, you should
get a
minima for the a' and on for the a" transition. Comparing these
two
geometries, you should find the structural change responsible for
the
interchange of the two states. We found, for example thath
for
hydroxyquinolines, such an interchage is induced by a significant change
of
the angle of the quinone-N.
iv) there might be some problems with
your basis set (inducing structural
changes and so on), maybe you first
choose a standard basis set?
v) try the CalcAll parameter for
the opt command.
Cheers,
Markus
>
-----Ursprüngliche Nachricht-----
> Von: owner-chemistry .. ccl.net
[mailto:owner-chemistry .. ccl.net]
> Gesendet: Samstag, 17. November 2007
03:05
> An: Thut, Markus
> Betreff: CCL: CIS Optimization for the
First Singlet Excited State
>
>
> Sent to CCL by: "Sue L"
[chsue2004 ~~ yahoo.com]
> Hi all,
>
> I have performed a CIS
optimization for the first singlet excited state
> with the keywords #n
cis(root=1,direct) gen opt
>
> However, the optimization cannot be
converged. I found that in the middle
> of the optimization, the first-
and second- excited state swapped, as
> indicated in the section of
Excitation energies and oscillator strength
> (as below).
>
>
---------
> Excited State 1: Singlet-A 3.7320 eV 332.22 nm
f=0.1634
> 93 -> 94 0.67286
> This state for optimization
and/or second-order correction.
> Copying the excited state density for
this state as the 1-particle RhoCI
> density.
>
> Excited
State 2: Singlet-A 5.5654 eV 222.78 nm f=0.2648
> 91 -> 94
-0.12140
> 91 -> 96 -0.14274
> 92 -> 94 0.46889
> 92
-> 95 -0.14756
> 93 -> 95 -0.35249
> ---------
>
> ---------
> Excited State 1: Singlet-A 5.5227 eV 224.50 nm
f=0.2035
> 91 -> 94 -0.11890
> 91 -> 95 -0.11553
> 91
-> 96 -0.13341
> 92 -> 94 0.46589
> 92 -> 95
-0.17450
> 93 -> 95 -0.35340
> This state for optimization and/or
second-order correction.
> Copying the excited state density for this
state as the 1-particle RhoCI
> density.
>
> Excited State 2:
Singlet-A 4.0937 eV 302.86 nm f=0.1943
> 92 -> 95 -0.11059
> 93
-> 94 0.66385
> --------
>
> Could anyone know what the
problem is?
>
> Thank you very much.
>
> Sue
>
>
>
>
>> > > > > >
>
>
> > >
> =>
>
<BR<BR<BR<BR
<BR <BR
Subscribe/Unsubscribe: <BR
Job: http://www.ccl.net/jobs<BR
<BR
<BR
<BR
<BR
?