From chemistry-request@server.ccl.net Mon Feb 14 05:44:11 2000 Received: from netra.ksu.edu.sa ([212.138.39.67]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id FAA11206 for ; Mon, 14 Feb 2000 05:44:08 -0500 Received: from sun1.ksu.edu.sa (sun1.ksu.edu.sa [212.138.39.3]) by netra.ksu.edu.sa (8.9.0/8.9.0) with ESMTP id MAA05252 for ; Mon, 14 Feb 2000 12:41:31 -0300 (Etc/GMT) Received: from localhost (azhary@localhost) by sun1.ksu.edu.sa (8.9.0/8.7.3) with SMTP id MAA08097 for ; Mon, 14 Feb 2000 12:39:55 -0300 (GMT) Date: Mon, 14 Feb 2000 12:39:54 -0300 (GMT) From: "Adel Abbas Ali Elazhary Sci. College" X-Sender: azhary@sun1 To: chemistry@ccl.net Subject: basis set for metal complexes (summary) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by server.ccl.net id FAA11207 Dear ccl members: About two weeks ago a I sent a question about basis set for metals. I received more requests to summerize than answers. The following are my original question and the only answer I got. Best regards, Adel El-Azhary ======================= original question ====================== Dear ccl members" I am doing DFT calculations on metal complexes with organic molecules and do not know what basis set should I use. Is there any study about basis set effect on DFT calculated geometry or ir spectrum of metal complexes. It might even help if someone suggest a test molecule for which its vibrational spectrum (especially the bands corresponding to the metal vibrations) are known. Thanking you in advance. Best regards, ======================= only answer ! ============================ Calculating metal complexes there is not much choice of basis sets. Using Gaussian, 6-311G is defined for 3d metals, but not for 4d and 5d. 4d and 5d are best calculated with pseudo potentials, which do take care of relativistic effects. LanL2DZ is best known, but you can also choose the Stuttgart Potentials (which are said to be implemented in G98, but we tend to have problems with the internal basis sets and prefer to specify in the input line. You can get basis sets input files on the net (www.uni-stuttgart.de , than searching for the theochem group) For 3d metals you have the choice of 6-311 and LanL2DZ or SDD - my experience with copper complexes is, that 6-311 give slighty better geometries, but do have convergence problems. Generally I do use mixed basis sets - e.g. 6-31(d) for the ligand and LanL2DZ or 6-311G for the metal centre. Calculating frequencies - well, with pseudopotentials there is no direct way to calculate them. They are calculating by searching systematically the surface, which is a somewhat time-consuming procedure. Best wishes Achim Achim Lienke PostDoctoral Research Fellow Anorganisch Chemisches Institut Universität Heidelberg Im Neuenheimer Feld 270 D 69120 Heidelberg Phone: ++ 49 6221 54 86 53 Fax: ++ 49 6221 54 66 17 =============================== end of the only answer ! =============