[molpro-user] problem with MRCI breaking degeneracy of 1-Delta state

[Irikura, Karl K. Dr. (Fed)]

Dear Prof. Werner,

Thank you!  Very helpful and so fast! 

Best regards,
Karl 

-----Original Message-----
From: Hans-Joachim Werner [mailto:werner at theochem.uni-stuttgart.de] 
Sent: Wednesday, May 23, 2018 9:15 AM
To: Irikura, Karl K. Dr. (Fed) <karl.irikura at nist.gov>
Subject: Re: [molpro-user] problem with MRCI breaking degeneracy of 1-Delta state

This is not an error but a feature of the internally contracted CI. If you do 2 states in sym 1 together you get a larger configuration space and therefore a lower energy. In the current case you can avoid this by doing the 2 states separately:

{ci; wf,16,1,0; state,1; save,4010.1; }
mrci(3) = energy
{ci; wf,16,1,0; state,1,2; save,4020.1; }  ! 1-state calculation for second root
mrci(4) = energy

which yields
 Relative energies in cm-1

        CAS      MRCI
        0.0       0.0
     6116.3    7908.5
     6116.3    7908.5
    11589.0   13342.7

In cases where calculations with state,1,2 (or for higher states) do not converge, try adding the directive
option,maxdav=20
to the mrci input. This often helps. Otherwise you can use nosym, as you found out yourself. This should give the best possible results, but is of course more expensive.

Joachim Werner
---
Prof. Dr. Hans-Joachim Werner
Institut für Theoretische Chemie
Universität Stuttgart
Pfaffenwaldring 55
70569 Stuttgart, Germany
e-mail: werner at theochem.uni-stuttgart.de




> Am 22.05.2018 um 23:10 schrieb Irikura, Karl K. Dr. (Fed) <karl.irikura at nist.gov>:
> 
> Dear MOLPRO experts,
>  
> For the 1-Delta state of oxygen (O2), I get nice degeneracy with CASSCF but an (incorrect) energy splitting from MRCI.  There is no problem if I use NOSYM, but it will take long in a bigger system.  What is my mistake?
>  
> --------problematic case (D2h)---------- ***, O2 singlet states ! 
> casscf(6,4), state-averaged orbitals !
> memory,1000,m
> basis=6-311G*
>  
> angstrom
> geometry={
> O
> O,1,1.2156
> }
>  
> {rhf
>     occ,3,1,1,0,2,1,1,0;
>     closed,3,1,1,0,2,0,0,0;
>     wf,16,4,2;
> }
>  
> {multi
>     occ,   3,1,1,0, 2,1,1,0;
>     frozen,1,0,0,0, 1,0,0,0;
>     closed,3,0,0,0, 2,0,0,0;
>     wf,16,4,2; state,1; lquant,0;
>     wf,16,4,0; state,1; lquant,2;
>     wf,16,1,0; state,2; lquant,2,0;
>     expec2,LZZ;
> }
>  
> cas = energy
>  
> ! MRCISD calculations
> {ci; wf,16,4,2; save,4042.1; }
> mrci = energy
> {ci; wf,16,4,0; save,4040.1; }
> mrci(2) = energy
> {ci; wf,16,1,0; state,2; save,4010.1; }
> mrci(3) = energy
>  
> ! convert absolute hartree to relative wavenumber do I=4,1,-1
>     cas(I) = (cas(I) - cas(1)) * TOCM
>     mrci(I) = (mrci(I) - mrci(1)) * TOCM enddo
>  
> ! print table of relative energies
> table, cas, mrci
> title, Relative energies in cm-1
> ftyp,f,f
> digits,1,1
>  
> output:
> Relative energies in cm-1
>         CAS      MRCI
>         0.0       0.0
>      6056.6    7751.0
>      6056.6    7620.8
>     11466.1   12890.0
>  
> -------no problem (C1)-------------
> ***, O2 singlet states
> ! casscf(6,4), state-averaged orbitals !
> memory,1000,m
> basis=6-311G*
>  
> NOSYM
> angstrom
> geometry={
> O
> O,1,1.2156
> }
>  
> {rhf
>     occ,9;
>     closed,7;
>     wf,16,1,2;
> }
>  
> {multi
>     occ,9;
>     frozen,2;
>     closed,5;
>     wf,16,1,2; state,1; lquant,0;
>     wf,16,1,0; state,3; lquant,2,2,0;
>     expec2,LZZ;
> }
>  
> cas = energy
>  
> ! MRCISD calculations
> {ci; wf,16,1,2; save,4012.1; }
> mrci = energy
> {ci; wf,16,1,0; state,3; save,4010.1; }
> mrci(2) = energy
>  
> ! convert absolute hartree to relative wavenumber do I=4,1,-1
>     cas(I) = (cas(I) - cas(1)) * TOCM
>     mrci(I) = (mrci(I) - mrci(1)) * TOCM enddo
>  
> ! print table of relative energies
> table, cas, mrci
> title, Relative energies in cm-1
> ftyp,f,f
> digits,1,1
>  
> output:
> Relative energies in cm-1
>         CAS      MRCI
>         0.0       0.0
>      6056.6    7564.9
>      6056.6    7564.9
>     11466.1   12822.4
>  
> ------------------------
>  
> Thanks for any help!
> Karl
>  
>  
> -------------------------------------------------------------------
> Dr. Karl K. Irikura
> National Institute of Standards and Technology
> 100 Bureau Drive, mail stop 8320
> Gaithersburg, MD 20899-8320
> voice:  301-975-2510   fax:  301-975-3670
> email:  karl.irikura at nist.gov
> http://www.nist.gov/mml/csd/informatics_research/index.cfm
> -------------------------------------------------------------------
>  
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.mo
> lpro.net%2Fmailman%2Flistinfo%2Fmolpro-user&data=02%7C01%7Ckarl.irikur
> a%40nist.gov%7C0f4945b972cb46917fd908d5c0af2711%7C2ab5d82fd8fa4797a93e
> 054655c61dec%7C1%7C1%7C636626780872568250&sdata=OLxPW5jETNo8BUsg%2BWfY
> pw64EabY5vey%2B47eSa778zs%3D&reserved=0