Calculations on Gallium

Nearly a year ago, I posted the following question to the CCL.
 Unfortunately, the project went in a completely different direction very
 soon after posting, and I never got a chance to try any of the
 suggestions.  An inquiry this morning reminded me that I also did not
 summarize to the list despite the fact that I explicitly promised to in
 the body of my question.  Shame!  Anyway, here is the original post and
 the responses - a belated thanks to those who replied!
 Original Post:
 I'm sorry for the bandwidth of the non-contributing type, but I would
 appreciate hearing experiences from anyone who has run calculations on
 Ga(III) or In(III) complexes with organic ligands.  Particularly, what
 basis sets/levels of theory/ECP's did you find useful?  A search of the
 CCL archives and and ACS journals through their website turned up
 nothing of interest, perhaps due to incompetent searching on my part...
 Plain old HF and B3LYP/3-21g* and PM3(tm) calculations gave non-useful
 results, and before I slide down the slope of simply ramping up the
 level of calculation on reasonably large systems until I get something I
 like, I'd like the input of more experienced users in the field.
 As usual, I'll be thrilled to advertise your research to the list if you
 reply usefully...
 Dave Giesen
 ============================
 Well, don't worry about quoting me, but there seem to be two obvious
 answers:
 (1) B3LYP/LANL2DZ (Hay-Wadt), with d-functions on the Gallium.  I've
 done
 well with that for some mixed R-Ga-M(CO)n complexes.
 (2) B3PW91/6-31G*, which is what F. Albert Cotton, and Feng used.
 This presumes you're using Gaussian 94/98.  My natural inclination would
 be to use MP2/SBK* (Stevens-Basch-Krauss ECPs, with associated
 polarization functions) in GAMESS-US.  In principle, they're in Gaussian
 98, but I've had all sorts of integral problems with them in that
 program.
 Of course, you could wander off the ranch, and use ADF with the
 appropriate Slater basis sets, or Spartan and the DN* (double numeric +
 polarization).  Both have worked, but require additional software.
                                                         0fred
 "No science has ever made                 Frederick P. Arnold, Jr.
  more rapid progress in a                 A&HPRC, U. of Chicago
  shorter time than Chemistry."            5640 S. Ellis Ave
         -Martin Heinrich Kloproth, 1791   Chicago, IL 60637
 ==============================
 Dear Dr. Giesen!
 I got usefull results for Ga(III) and In(III) with:
 - PM3 (the original PM3, as implemented in AMPAC/MOPAC/VAMP)
 - B3LPY/LANL2DZ (adding extra polarisation can be necessary)
 Greetings
 Ralph Puchta
 ==============================
 While water is not strictly an organic ligand, I have done calculations
 on
 the hexaaqua complexes of the said metals using
 (HF/MP2)/(6-31G*,6-31+G*),
 including frequency calculations for some of these, and also
 HF/(6-31G*,6-31+G*)on the octadecaaqua complexes [6+12], to model the
 vibrational frequencies of
 the GaO6 cluster. The paper is being written up. For the Ga/In basis
 set,
 we took the best Huzinaga minimal basis set that we could find, split
 the
 valence region and added polarization funxctions, to give:
 Ga: (43321/4321/41*)
 In: (433321/43321/431*)
 We took the basis sets from Huzinaga's book.
 We have done this successfully for other atoms (Sc,Zn,Cd) in similar
 systems
 CC Pye et al, JPC A, 1998, 102, 9933-9943 - MgAq6 2+
 WW Rudolph et al, JPC B, 1998, 102, 3564-3573 - CdAq6 2+
 CC Pye et al, JPC, 1996, 100, 601-605 - LiAq4+
 WW Rudolph et al, JPC, 1995, 99, 3793-3797 - LiAq4+
 and for Ge,Sn,As,Sb,Se,Te,Br,I in organic systems.
 RA Poirier et al, JOC, 1995, 60, 2328-2329
 JD Xidos et al, JOC, 1998, 63, 105-112
 >
 > Plain old HF and B3LYP/3-21g* and PM3(tm) calculations gave non-useful
 > results, and before I slide down the slope of simply ramping up the
 > level of calculation on reasonably large systems until I get something I
 > like, I'd like the input of more experienced users in the field.
 I believe that at least a HF/DZ + polarization + diffuse is necessary
 for
 binding energies to metals, and MP2 can improve things. One thing to
 watch out
 for is the nearly isoenergetic nd/(n+1)s orbitals, especially if frozen
 cores
 are involved.
    *************
  *****************  !  Dr. Cory C. Pye
 ***   **    **  **  !  Postdoctoral Fellow
 **   *  ****        !  Theoretical and Computational Chemistry
 **      *  *        !  cory "at@at" ucalgary.ca
 **      *  *        !  http://www.cobalt.chem.ucalgary.ca/cory
 ***     *  *    **  !
  *****************  !  Les Hartree-Focks
    *************    !  (Apologies to Montreal Canadien Fans)
 --
 Dr. David J. Giesen
 Eastman Kodak Company                           david.giesen "at@at" kodak.com
 2/83/RL MC 02216                                (ph) 1-716-58(8-0480)
 Rochester, NY 14650                             (fax)1-716-722-2327