[molpro-user] CI singles from SCF reference - spin symmetry issue

[Dan Haxton]

Hello,

I am trying to perform a very simple calculation
on NO_2^+ with version 2002.6 in which I include all
single excitations from a (closed-shell) SCF
reference configuration.  I would like to do this
for all spin and spatial symmetries, not only those
of the SCF reference configuration.  Here is my
input, done in cs symmetry with two irreps.




geomtyp=xyz

geometry
 3
 myno2
 n    0.0        0.0     0.0
 o    0.0     1.12  0.01
 o    0.0     1.14  0.01
end

basis=cc-pvtz
gprint,orbital,civector;

rhf;wf,22;occ,9,2
ci;occ,9,2;nopair;wf,22,1,0;state,1
ci;occ,9,2;nopair;wf,22,1,2;state,1
ci;occ,9,2;nopair;wf,22,2,0;state,1
ci;occ,9,2;nopair;wf,22,2,2;state,1




( I will be calculating excited states once I get
this working. )  The program stops on the second
ci calculation with the statement "SPIN SYMMETRY
NOT POSSIBLE."  I may not get around this error by
instead defining the entire orbital space as a
reference space and using restrict, i.e.

ci;occ,60,30;wf,22,1,2;restrict,0,1,10.1,11.1,12.1 ......

because of the hard-wired restriction upon the
number of active orbitals.  ("TOO MANY ACTIVE
ORBITALS")  I also may not use the REF card, which,
as far as I can tell, only allows one to input
additional spatial symmetries, not spin symmetries.

There must be a way to do this calculation, and I
apologize if I have missed something obvious, but I
cannot find any options in the molpro manual that
seem like they would get this to work.  Any help
would be very much appreciated!

Dan Haxton
Postdoctoral research associate, JILA, CU-Boulder