singlet excitation

[lsikk]

Hi
Im trying to calculate the rotation barrier around a double bond in ground state and excited state, but I have some problems understanding the turbomole output.
I want to calculate the electronic energy difference between the ground and excited state. The ground state is ok, but in the excited state the electronic energy is about 5 kcal/mol higher than ground state (should be around 70 kcal/mol). To get the excited state enerrgy, should I add electronic energy to exitation energy, and then use this value to calculate the difference from the ground state?
Im so confused.... 
btw..  basis SV(P), RIDFT: BP86 and turbomole version 5.8

Lauri

[ederat]

Hello,

the difference of 70 kcal/mol you are referring is the gap between the singlet and triplet state of ethene. From a quick and dirty TM calculations:

$energy      SCF               SCFKIN            SCFPOT
     1   -78.51856422316    77.57399692559  -156.09256114875
     2   -78.23758493802    77.61698597217  -155.85457091018
     3   -78.40883976339    77.78000955365  -156.18884931704
$end

First line is ethene in single state, D2h geometry
Second line is ethene in triplet state, D2h geometry
Third line is ethene in triplet state,  relaxed geometry: the two CH2 are perpendicular. (C2v point group)
Same level as you
you can see the gap between 1 and 3 is 68.85 kcal/mol, close to the value you have in mind.

To reproduce this kind of calculation: rerun define and answer no at this question:
>>> DO YOU ACCEPT THIS OCCUPATION ?  DEFAULT=y
you will have the following choice after:
>>>  t         : CHOOSE UHF TRIPLET OCCUPATION

Hope this help,
Etienne

[lsikk]

Hi

My last post might have been a bit confusing.. so I try again 
I defined the structure as usual but in the last menu i chose ex -> rpas (and selected a'  1 )
optimized the geometry ( jobex -ri -ex -c 200) and the last line in the energy file looks like:
$energy      SCF               SCFKIN            SCFPOT            excitation
33 -645.3836643855      638.622016     -1284.00568     0.1064648732565

When I do the single point ground state ridft calculation for the same structure, I get the same SCF energy - so the energy between ex state and ground state is this exitation enery.
It is about 66 kcal/mol which is close to the numbers in this other article.
Is this correct or am i terribly wrong and should go to the zoo and feed the donkeys??? 

[christof.haettig]

Hi,

your are right. the total energy of the excited state is the sum of the 'SCF' and 'excitation' entries.
But better places to look up the total energies are the files 'job.last' (outputs from the last optimization
cycle, which contains both the total energy of the ground and of the excited state) and $grad, which
contains the total energies (all cycles) of the state for which the geometry is optimized.

Christof

[lsikk]

Hi

When doing these RPA singlet-excitation DFT calculations there is a problem:
egrad stops:
                       -------------------
                         excitation vector
                        -------------------


          IRREP   tensor space dimension   number of roots

           a              7987                   5

 dimension of super-tensorspace: 1

 Aufbau principle violation encountered. Check occupancies!

error in gradient step (1)

cant understand what is the problem, because the number of occupied orbitals is correct.