CCL: Making symmetry equivalence atomic orbitals

 Sent to CCL by: "Wirawan  Purwanto" [wirawan0{}yahoo.com]
 Hi,
 Can someone give me a pointer to how to make "symmetry equivalenced"
 orbitals? I am not that familiar with this terminology, and in some papers this
 is mentioned without any citation.
 This is my limited understanding of it: in the context of Hartree-Fock
 calculation of an isolated atom (say, ROHF calculation of Co atom), the five 3d
 orbitals ended up not having the same shape and energy because of the
 symmetry-broken configuration (3d7 4s2). The symmetry equivalenced 3d orbitals
 would then be obtained by somehow restoring the degenerate nature of these
 orbitals (which would raise the variational energy by a little bit). I think
 this can also be done for e.g. diatomic molecule where the pix-piy degeneracy
 could be broken due to open shell nature of the HF solution.
 Here are my questions:
 1. Is this understanding correct?
 2. What are the step-by-step way to make the symmetry equivalenced orbitals from
 the usual ROHF or UHF orbitals? Any reference or program that can do this from a
 MO output given by a quantum chemistry program (e.g. GAMESS, molpro...whatever
 you like), assuming that I know the composition of the underlying AO basis (i.e.
 which function is 1s, ...2px,... 3dxy, and so on).
 Thanks,
 Wirawan