From chemistry-request@ccl.net Sat Jan 25 11:40:24 1992
Date: 25 Jan 92   10:57 LCL
From: PA13808%UTKVM1.BITNET@OHSTVMA.ACS.OHIO-STATE.EDU
To: CHEMISTRY@ccl.net
Subject: BITNET mail follows
Status: R

 TRANS MET COMPLEXES GEOM OPT.
   IT is hard to give you useful advice without more details.Your
 problem could be associated with an incorrect starting config
 accompanied by the program changing the config ie a change in what
orbitals are occupied.as the calculation proceeds. You can control
the starting occupations with the GUESS =ALTER command page 81 of the
 manual. You will then probably have to invoke the VSHIFT=N command
 on p19. This is designed to stop the occ and unocc orbitals from
  interchanging. The starting geometry is also imp since it must give
 the correct occupied orbitals. My experience is that one needs a lot of
 of knowledge, patience and computer time to get a correct answer.
      One topic I would like any comments on is why I have never been
 able to converge onto any of the Geometries quoted by Dobbs and HEHRE
J.COMPCHem vol8 p861(1987) eg for CUCL or CrO4 2-. I get close on
geom and energy but never what they quote eg in TABLE X using their basis
  John E. Bloor(PA13808 at UTKVM1)