About Orbital reordering

[woozino]

Dear all,

I am trying to reorder the orbitals in C1 symmetry.

For example, I want to exchange 5th orbital and 7th orbital in C1 symmetry.

In manual, I couldn't understand how to use the '.reorder' option.

Can anyone explain more details of reorder option?

Thanks in advance.

[taylor]

It would help to have some idea of what you want to do after reordering the orbitals. One situation where this might be needed is in optimizing core-hole or core-excited states to ensure you have the right active space. Another is when the actual symmetry is higher than the symmetry Dalton uses, for example for linear molecules. Reordering can be used to avoid getting orbitals of the wrong higher symmetry in the active space.

You say your calculation is run in C₁ which suggests the molecule has no symmetry --- are you therefore planning to calculate a core hole or similar state? In any event, to exchange the fifth and seventh MOs:

Code: Select all

.REORDER
  2
  5  7
  7  5

You are telling the program that you want to reorder two MOs, and the new MO 5 is the old MO 7, and the new MO 7 is the old MO 5.

If there were symmetry present, you have to indicate how many reorderings there are in each symmetry, and then specify those reorderings.

Best regards
Pete

[woozino]

Dear Pete,

I appreciate for your reply. Now I have one more question about symmetry.

For example, in case of Cs symmetry, if i want to exchange 5th orbital and 7th orbital of A symmetry, 'and' 3rd orbital and 5th orbital of A' symmetry, can you explain how to specify in input?

Best regards
Jin

[taylor]

Yes. You would specify

Code: Select all

.REORDER
  2  2
  5  7
  7  5
  3  5
  5  3

I think you can also put the reordered orbitals from the same symmetry on one line, like

Code: Select all

.REORDER
  2  2
  5  7  7  5
  3  5  5  3

You still give no reason for why you want to reorder orbitals? It is not a very common thing to do (except perhaps for core holes).

Best regards
Pete

[hjaaj]

Yes, you can specify all reordering for a given symmetry in one line when non-zero number of reordered orbitals in that symmetry, and you can also use multiple lines. If you do not have any orbitals to reorder in a symmetry, then you must not write an empty line for reordering specification for that symmetry.

[woozino]

Dear Pete,

Thank you so much for explaining this to me. I understand how to use it.
Actually, i'm trying to calculate benzene without symmetry (C1) using CASSCF(6,6) only with pi orbitals.
You probably know that I need to reorder the ordering to use only pi orbitals.

Thank you again,
Jin

[woozino]

Dear Hans Jorgen

Thank you for explaining to me. It helped me a lot.

Best,
Jin