[molpro-user] selection of CAS

[A.Zech at stud.uni-heidelberg.de]

Thanks for your reply. I'm fairly new to Molpro so please excuse if i  
missed an obvious solution to this problem.
Yes, i do know which orbitals to swap (6.2 and 7.2, respectively).
Including the 6.2 orbital into CAS (i.e. 1.2 to 6.2) results in a  
lower energy than moving that orbital out of the active space (i.e.  
CAS(6,6) = 1.2 to 5.2, plus 7.2).Thus the orbital in question always  
gets rotated back during wavefunction optimization. With a slightly  
distorted geometry (Cs, one angle changed by 0.001 degrees), i tried  
the DONT directive to turn off the optimization, but i only get as  
close as 0.1eV to the corresponding energy in C2v. I'm convinced that  
the ROTATE directive is the best strategy here. Is there a way to keep  
the CAS orbital selection (after ROTATE) constant during the  
wavefunction optimization?

Best regards,

Alexander Zech

--------------------------------------------------------
Provided you know which orbitals you want to move in/out of the active  
space, this can be achieved with the ROTATE directive - see  
http://www.molpro.net/info/2012.1/doc/manual/node240.html
Peter

On 23 Aug 2013, at 14:38, A.Zech at stud.uni-heidelberg.de wrote:

> Hello all,
>
> i would like to exclude a single orbital from a certain CAS. In my  
> case i used
> the following B1 and A2 orbitals (molecule has C2v symmetry, lies in  
> yz-plane):
> 1.2
> 2.2
> 3.2
> 4.2
> ---
> 1.4
> 2.4
> For some normal modes the symmetry gets reduced to Cs. My problem  
> now is that i can't use the same CAS anymore since the order of  
> orbitals gets mixed up.
> The order in A'' is then:
> 1.2  [former 1.2]
> 2.2  [former 1.4]
> 3.2  [former 2.2]
> 4.2  [former 2.4]
> 5.2  [former 3.2]
> 6.2  [former 3.4]*
> 7.2  [former 4.2]
> Following the molpro syntax for OCC and CLOSED, it's not possible to  
> get the same CAS without including 6.2 [formerly 3.4], which i  
> marked with an asterisk.
> Is there any possibility to exclude that orbital from the active  
> space, maybe with the ORBITAL directive?
>
> Best regards,
>
> Alexander Zech
>
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user

--
Prof. Peter J. Knowles
School of Chemistry, Cardiff University, Main Building, Park Place,  
Cardiff CF10 3AT, UK
Telephone +44 29 208 79182 Email KnowlesPJ at Cardiff.ac.uk
WWW  
http://www.cardiff.ac.uk/chemy/contactsandpeople/academicstaff/knowles.html