Skip to content

Qustion about NEVPT2 calculation for open shell systems to obtain excited states #286

@Yasto55

Description

@Yasto55

Hello!

I utilize NEVPT2 calculation to estimate the excited states of the open shell systems such as metal phthalocyanine. I am struggling with getting excited states by setting approprite active spaces composed of pi-orbitals and d-orbitals

When there is a half-occupied d-orbital derived from transitional metal atom, I think we must incorporate half-occupied orbital into active sptace to execute NEVPT2 calculations. It this correct? Is there some tequnique to avoid setting half occupied orbital as active space on executing NEVPT2 calculations? This is because the existence of half-occupied orbitals makes it very difficult to execute correct calculation with NEVPT2.

The trouble is that we can not get pi -> pi* excited configurations with NEVPT2.
On the other hand, we often obtain wrong excited state where excitations from double occupied orbitals to SOMO orbital often occurs.

Is it generally difficult to obtain excited states of open shell systems with NEVPT2?

If NEVPT2 is not good at calculating open shell systems, I will give up utilizing this method.

On preparing for the reference state , I usually execute UHF calculation. So I also think the refference state includes something wrong HF state where the pair relationship between alpha and beta electrons on each MO level is someting wrong.

Could you please give me some comments on utilizing NEVPT2 on the excited ststes of open shell systems?

Thanks in advqnce

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions