[molpro-user] Core/valence energy in RS3

[Terry Frankcombe]

Hi folks

I'm doing RS3 calculations on CNN using 2006.1.

For two nearby geometries, full valence CASSCF gives energies of
-146.73652 and -146.73690 Eh.  When I subsequently do RS3 calculations
the energies are different by 0.07 Eh (with the second one being clearly
wrong).

Before the RS3 iteration commences, the following energies are printed:

 Nuclear energy:                      33.69036790
 Core energy:                       -139.17974579
 Zeroth-order valence energy:         -8.18276272
 Zeroth-order total energy:         -113.67214062
 First-order energy:                 -33.06438373
----
 Nuclear energy:                      33.42831725
 Core energy:                       -108.09004074
 Zeroth-order valence energy:        -31.66283490
 Zeroth-order total energy:         -106.32455839
 First-order energy:                 -40.41234781

The second set are clearly wrong, with 31 Eh being shuffled out of the
'Core energy'.  Presumably this is why the RS3 energy converges to the
incorrect value.

What causes this incorrect result?  How can I work around it?

Regards
Terry


-- 
Dr. Terry Frankcombe
Research School of Chemistry, Australian National University
Ph: (+61) 0417 163 509    Skype: terry.frankcombe