rydberg basis with GENANO and CASSCF (DELETE and ALTER)


[ Molcas user's WWWBoard ]

Posted by Claire Loison on December 10, 2009 at 13:14:13:

Dear molcas users,

For the first time I am willing to calculate some electronic excitation energies and oscillator strengths using a basis including rydberg-type functions.

The system is a "simple" conjugated aromatic planar molecule studied in Cs group (naphthol). Following some published papers (eg Rubio et al, Chem. Phys. 179 (1994), Roos et al. in Ad. Chem. Phys Vol 93 (1996) ), I first constructed a basis set using a cation CASSCF calculation and GENANO, and then I add this contracted rydberg basis to an ano-s basis. Now I have done some CASSCF calculation and found a rydberg MO in the CAS. so far so good. In a second step I want to use the keyword DELETE to put it away but i am not sure how it works.

Here are my questions :

1) how to estimate the validity of the ANO
contraction, and the result of genano ?
how to interpret the fig file ?

here is my input for GENANO after a casscf
calculation . Is it the right way to
use the keyword RYDBERG ?
------------------
&GENANO &END
Title
naphthol cation for rydberg basis set
sets
1
Center
X
RYDBERG
End of input
--------------------


2) after the first casscf, the MO 8 in symmetry 2 is the rydberg MO i want to get rid of in the cas.
there are 55 MO in symmetry 2.
here is the casscf input i use:
------------
&RASSCF &END
Title
Naphthol
nActEl
10 0 0
Inactive
32 1
Ras2
0 13
LUMORB
SYMMETRY
1
SPIN
1
CIROOT
7 7 1
ITER
30 100
ALTER
1
2 8 55
DELETE
0 1
End of input
-------------------

after the calculation i get a new rasorb and jobiph file. Is the deleted rydberg MO now labeled 8 or 55 ?
=> i supposed it is 55..
should i delete it again for next calculations ?
=> i supposed yes.
When i use the new rasorb or jobiph for a next calculation, should i alter the order again or not ?
=> i supposed not

I did not and used as a second input for the next casscf the following input :
-------------------------
&RASSCF &END
Title
Naphthol
nActEl
10 0 0
Inactive
32 1
Ras2
0 13
JOBIPH
SYMMETRY
1
SPIN
1
CIROOT
7 7 1
ITER
30 100
DELETE
0 1
End of input
-------------------------

What bothers me is that I get this part in the output :
-----------------------------
Molecular orbitals for symmetry species 2: a"

Orbital 11 12 13 14 15 16 55
Energy .0000 .0000 .0000 .0000 .0375 .0408 .0000
Occ. No. .9201 .1327 .2183 .0645 .0000 .0000 .0000

.... further down

53 X 2pz -.0035 -.1572 -.0682 -.0013 .0282 .0526 -.8534
54 X 3d1- .0067 -.0004 -.0035 -.0064 .1203 1.0508 -.0206
55 X 3d1+ -.0026 .0010 .0033 -.0004 1.0507 -.1249 .0413

---------------------------
so the orbital 55 is at energy 0, as if it would be part of the CAS !
do delete orbitals have no energy ?

it seems that the rydberg MO are the number
15,16,55, all of them are out of the cas.

So the rydberg MO of the cas was rotated against
a valence MO.

Should I first force now the other rydberg MO (15 and 16) to be part of the cas (either by increasing the cas, or by altering the order of MO and deleting them afterwards ) ? what would you choose ?

Thanks for reading this email to the end !
Claire Loison





Follow Ups:



Post a Followup

Name:
E-Mail:

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:


[ Follow Ups ] [ Post Followup ] [ Molcas user's WWWBoard ]