[molpro-user] Subject: oscillatory strength calculation by eom-ccsd method

Tatiana Korona tania at tiger.chem.uw.edu.pl
Sat Aug 10 16:07:04 BST 2013


Dear Arpita,

If you look at the output, you will see that the calculation of the Lambda 
vector did not converge in this case:

Solve a set of linear equations for Lambda
  <Lambda|[e-T H eT,exc.] = -<0|[e-T H eT,exc.]

  Iteration   Residuum      time/iteration   time/total

     1       0.11E+01           29.7           29.7
     2       0.59E+00           29.7           59.5
     3       0.50E+00           29.9           89.4 liis
     4       0.56E+00           29.8          119.2 liis
     5       0.67E+00           29.8          148.9 liis
     6       0.88E+00           29.9          178.9 liis
     7       0.12E+01           29.9          208.8 liis
     8       0.16E+01           29.8          238.6 liis
     9       0.22E+01           30.1          268.7 liis
    10       0.30E+01           30.3          299.0 liis
    11       0.42E+01           30.0          328.9 liis
    12       0.59E+01           29.9          358.9 liis
    13       0.82E+01           29.7          388.5 liis
    14       0.11E+02           29.8          418.3 liis
(...)
Convergence not achieved in  50 iterations

so the results for transition moments, where the Lambda vector is used, are all 
nonsense. For some reason the program did not stop after this unsuccessful part, 
as it should.

Try to repeat your calculations after the following change in the last line

{ccsd;dm;eom,-5.5,trans=1,startt1=32}
!                        ^^^^^^^^^^^

This change forces another starting guess for the set of linear equations for 
Lambda.

Independently of the convergence problems, you should be aware that the 
reliability of the CCSD ground-state calculations in this case is not high: note 
that a value of the D1 diagnostics is much higher than the upper limit for the 
"safe region" of D1.

Best wishes,

Tatiana

On Sat, 10 Aug 2013, Arpita Ghosh wrote:

> Dear Molpro users and developers
> I would like to calculate the excited states and oscillator strength of C17
> molecule by eom-ccsd method. But I am getting *****  for oscillator
> strength instead of the numerical value.
>
> Here is my input file :
>
> ***,C17 molecule in xyz coordinates
> memory,300,m
> basis=cc-pvdz
> geomtype=xyz
> geometry={17
> optimization of C17 molecule
> c   0.000000    0.000000   10.419026
> c   0.000000    0.000000    9.102347
> c   0.000000    0.000000    7.788613
> c   0.000000    0.000000    6.495247
> c   0.000000    0.000000    5.193215
> c   0.000000    0.000000    3.896865
> c   0.000000    0.000000    2.596988
> c   0.000000    0.000000    1.298906
> c   0.000000    0.000000    0.000000
> c   0.000000    0.000000   -1.298906
> c   0.000000    0.000000   -2.596988
> c   0.000000    0.000000   -3.896865
> c   0.000000    0.000000   -5.193215
> c   0.000000    0.000000   -6.495247
> c   0.000000    0.000000   -7.788613
> c   0.000000    0.000000   -9.102347
> c   0.000000    0.000000  -10.419026
> end}
> {rhf;wf,102,1,0}
> {ccsd;dm;eom,-5.5,trans=1}
> exit
> The output file is looking like this :
>
> State    Exc. Energy (eV)            X              Y              Z
>   2.5          2.890
>
> Right transition moment           0.00000000     0.00000000***************
> Left  transition moment           0.00000000     0.00000000   -15.87124227
> Dipole strength              ***************
> Oscillator strength          ***************
> Dipole moment                     0.00000000     0.00000000
> 0.00000000   Norm:     0.00000000
>
>
> Thank you.
>

Dr. Tatiana Korona http://tiger.chem.uw.edu.pl/staff/tania/index.html
Quantum Chemistry Laboratory
University of Warsaw
Pasteura 1, PL-02-093 Warsaw, POLAND


`The man who makes no mistakes does not usually make anything.'
                                        Edward John Phelps (1822-1900)



More information about the Molpro-user mailing list