From owner-chemistry@ccl.net Wed Jan  4 13:44:00 2012
From: "Barbara Fresch bfresch++ulg.ac.be" <owner-chemistry : server.ccl.net>
To: CCL
Subject: CCL:G: starting guess from unrestricted DFT for CAS-SCF in G09
Message-Id: <-46111-120104123908-11514-WH5Ot1LEgx23bcA5RJ0LNw : server.ccl.net>
X-Original-From: "Barbara  Fresch" <bfresch%%ulg.ac.be>
Date: Wed, 4 Jan 2012 12:39:03 -0500


Sent to CCL by: "Barbara  Fresch" [bfresch^-^ulg.ac.be]
hello everybody,

i'm trying to obtain the CASSCF wavefunction of the ground state of a metal cluster with G09. 
I want to select the active space from the orbitals obtained by an unrestricted B3LYP optimization. I can also use those orbitals as starting guess for the CASSCF (I just use Guess=Read). 

Here where I'm confused: since from the unrestricted-DFT one obtains two set of orbitals (the alpha and the beta orbitals) while as far as i understood the CASSCF wavefunction is a combination of restricted determinants, how the initial guess is adapted? Does Gaussian use only e.g. the alpha set of orbitals? 
  
thanks in advance for your help.
Barbara Fresch
e-mail: bfresch\a/ulg.ac.be