[molpro-user] FCI calculation

[Peter Knowles]

The occ directive in FCI defines the numbers of orbitals in each  
symmetry that are included in the FCI calculations. In your inputs you  
have specified this as just the valence orbitals. With CASSCF orbitals  
you should then get just the CASSCF energy, and with other orbitals,  
something higher. If you want the full calculation including the  
complete basis, simply omit occ.
Peter
On 29 Jan 2008, at 11:40, rama chandran wrote:

> Dear Molpro users,
> I did one molpro calculation for a triatomic linear
> system containing 3 electron  HeH2+ linear.
> Earlier the energy for this system was reported to be
>
>
>        MRCI            CASSCF          HF-SCF
>     -3.51818587     -3.48901389     -3.47036561
>
> using th input
>
>
> ***,HeH2plus
>  Memory,100,M
>  geometry={H;
>  H1,H,r1;
>  He,H,r2,H1,a1}
>   r1=2.075338
>   r2=1.931059
>   a1= 180.0
>  basis=v5Z
> {HF
> occ,2,0,0,0
> wf,charge=1,symmetry=1,spin=1
> }
> casscf
> mrci
>
>
> I tried to redo the problem using  Full-CI
> calculations
> I tried it in 3 different ways as I have not done any
> full CI calculations before...
> when I tried with the keyword fci alone without
> specifying anything else it terminated with the error
> saying that orbital record not found.
> Then I specified the wavefunction details as above,
> still the same error.
> Then I did the fci followed taking  previous hf
> wavefunction   and also another calculation in which
> fci taking casscf wavefunction as follows,
>
> ***,HeH2+intermediate-example-FCI
> memory,50,m
> geometry={H;
>  H1,H,r1;
>  He,H,r2,H1,a1}
>   r1=2.075338
>   r2=1.931059
>   a1= 180.0
> basis,H=v5z
> !define basis
> hf
> {fci
> occ,2,0,0,0
> wf,charge=1,symmetry=1,spin=1
> }
>
>
>
> and
>
>
> ***,HeH2+intermediate-example-FCI
> memory,50,m
> geometry={H;
>  H1,H,r1;
>  He,H,r2,H1,a1}
>   r1=2.075338
>   r2=1.931059
>   a1= 180.0
> basis,H=v5z
>
> {casscf
> occ,2,0,0,0
> wf,charge=1,symmetry=1,spin=1
> }
> {fci
> }
>
> both gives the same energy as -3.50605528 Hartrees
>
> As you see this energy is very much different from the
> values they reported through MRCI calculations....
> I was expecting a lower energy through full CI
> calculations.  Is something wrong....??
>
> Can anyone advice me with some details?
>
>
>      Download prohibited? No problem. CHAT from any browser, without  
> download. Go to http://in.messenger.yahoo.com/webmessengerpromo.php/

--
Prof. Peter J. Knowles
School of Chemistry, Cardiff University, Main Building, Park Place,  
Cardiff CF10 3AT, UK
Telephone: +44 29208 79182 Fax: +44 2920874030 Email KnowlesPJ at Cardiff.ac.uk
WWW http://www.cardiff.ac.uk/chemy/contactsandpeople/academicstaff/knowles.html