From chemistry-request@ccl.net Sun Jun  8 01:42:20 2003
Received: from sungoddess.ccs.yorku.ca (mailrelay.yorku.ca [130.63.236.144])
	by server.ccl.net (8.12.8/8.12.8) with ESMTP id h585gKkr018637
	for <chemistry*at*ccl.net>; Sun, 8 Jun 2003 01:42:20 -0400
Received: from curl.gkcl.yorku.ca (curl.gkcl.yorku.ca [130.63.232.224])
	by sungoddess.ccs.yorku.ca (8.12.9/8.12.8) with ESMTP id h585gJ0O022793
	for <chemistry*at*ccl.net>; Sun, 8 Jun 2003 01:42:19 -0400 (EDT)
Received: (from chan@localhost)
	by curl.gkcl.yorku.ca (8.10.0/8.10.0) id h585gIL04695
	for chemistry*at*ccl.net; Sun, 8 Jun 2003 01:42:18 -0400 (EDT)
From: Wai-To Chan <chan*at*curl.gkcl.yorku.ca>
Message-Id: <200306080542.h585gIL04695*at*curl.gkcl.yorku.ca>
Subject: On basis sets for calculating thermochemistry properties
To: chemistry*at*ccl.net
Date: Sun, 8 Jun 2003 01:42:18 -0400 (EDT)
X-Mailer: ELM [version 2.5 PL3]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Dear all,

I am calculating thermochemistry properties by Gaussian DFT method. Can 
anyone give me some suggestions on basis sets for the calculation? 
Which is the minimum basis set I can use in this calculation is 
reasonable? With small basis sets (sto-3g, 3-21g), which one can get 
better results for thermochemistry properties by Gaussian DFT method?

Thanks to you all,

Holly
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

I used the xalpha/3-21g method for geometry optimization
of LARGE molecules in the past and got some decent results. But 
sto-3g and 3-21g won't be a good choice for reliable thermochemistry 
calculation with Hybrid DFT method like B3LYP. Use a basis set at least 
as large as 6-31+G(d)/6-31+G(d,p). The inclusion of '+' can make a difference. 
You may refer to 

   J. Phys. Chem. A, (2003) vol 107 pg 1384-1388 by Lynch, Zhao and Truhlar

for an assessment of the accuracy of the MPW1PW91/6-31+G(d,p) method. 

Wai-To Chan