Molcas Forum

xyin

Member

Registered: 2022-05-12

Posts: 6

How to alter an n orbitals to active space?

Hi, I want to ask you about how to add an n orbital to the active space.
In my system, I chose 14 active orbtials and 14 active electrons, which already includes all pairs of PI orbitals. However, the Reference weightof my CASPT2 result is only 0.52124. I think it is because I did not add n orbital. However, when I tried to use Alter keyword to exchange orbit, I found that I could not successfully add n orbit into the active space, it still all pi and pi* orbtials. I felt very confused. How can I solve this problem?

Here is my input file with 14 active orbtials and 14 active electrons:

&RASSCF
Title = 1414-def-S0
FILEORB = 36-deftzvp_uhf_uno_asrot.INPORB
Charge = 0
Spin = 1
RAS2 = 14
nActE1 = 14 0 0
CiRoot = 5 5 1
Rlxroot= 1

&CASPT2
Multistate = 5 1 2 3 4 5
Imaginary Shift = 0.2
IPEA=0

The result:

Reference energy:        -931.9507889785
E2 (Non-variational):      -2.9238728169
Shift correction:          -0.0141821523
E2 (Variational):          -2.9380549692
Total energy:            -934.8888439476
Residual norm:              0.0000003092
Reference weight:           0.52124

Contributions to the CASPT2 correlation energy
Active & Virtual Only:         -0.0767940125
One Inactive Excited:          -0.6962928736
Two Inactive Excited:          -2.1507859307

Here is the file that I alter an n orbital:

&RASSCF
Title = 1413-def-S0
FILEORB = 36-deftzvp_uhf_uno_asrot.INPORB
Charge = 0
Spin = 1
RAS2 = 13
nActE1 = 14 0 0
Alter = 1; 1 67 68
CiRoot = 5 5 1
Rlxroot= 1

&CASPT2
Multistate = 5 1 2 3 4 5
Imaginary Shift = 0.2
IPEA=0

Last edited by xyin (2022-05-12 08:57:29)

lijingbai2009

Member

Registered: 2019-08-28

Posts: 29

Re: How to alter an n orbitals to active space?

In your second input, the number of active orbitals reduces to 13. Does it mean you remove a pair of pi and pi* and then add the n-orbital? Nevertheless, if the n orbital keeps rotating out, it suggests there are small contributions from the n->pi* configurations in the lowest 5 states. You may want to increase the state-averaging to include at least one excited state dominated by the n->pi* configuration.

xyin

Member

Registered: 2022-05-12

Posts: 6

Re: How to alter an n orbitals to active space?

In your second input, the number of active orbitals reduces to 13. Does it mean you remove a pair of pi and pi* and then add the n-orbital? Nevertheless, if the n orbital keeps rotating out, it suggests there are small contributions from the n->pi* configurations in the lowest 5 states. You may want to increase the state-averaging to include at least one excited state dominated by the n->pi* configuration.

Yes, i remove a pair of pi and pi* to alter the n orbtial. I will try to increase the state-averaging. Thanks a lot.

xyin

Member

Registered: 2022-05-12

Posts: 6

Re: How to alter an n orbitals to active space?

In your second input, the number of active orbitals reduces to 13. Does it mean you remove a pair of pi and pi* and then add the n-orbital? Nevertheless, if the n orbital keeps rotating out, it suggests there are small contributions from the n->pi* configurations in the lowest 5 states. You may want to increase the state-averaging to include at least one excited state dominated by the n->pi* configuration.

Hi, I have increased the state-averaging for CiRoot = 7 7 1, but the calculation results show that the Reference weight is still very small (about 0.51572), and  the exchange of n orbitals into active space is still not successful.  Should I continue to increase state ?
Looking forward to your reply, I will appreciate it very much.

lijingbai2009

Member

Registered: 2019-08-28

Posts: 29

Re: How to alter an n orbitals to active space?

In your second input, the number of active orbitals reduces to 13. Does it mean you remove a pair of pi and pi* and then add the n-orbital? Nevertheless, if the n orbital keeps rotating out, it suggests there are small contributions from the n->pi* configurations in the lowest 5 states. You may want to increase the state-averaging to include at least one excited state dominated by the n->pi* configuration.

Hi, I have increased the state-averaging for CiRoot = 7 7 1, but the calculation results show that the Reference weight is still very small (about 0.51572), and  the exchange of n orbitals into active space is still not successful.  Should I continue to increase state ?
Looking forward to your reply, I will appreciate it very much.

I don't have a simple answer for this. You can try to increase the state-averaging to test. But I would also do some TDDFT(e.g. wB97XD) to check the contributing orbitals in the excited states. That usually gave a reasonable starting point to select active orbitals and state-averaging. It might tell you where the n->pi* state could be. Maybe you have to include all pi and pi* to keep the n-orbital in your active space, if I guess.