From owner-chemistry%!at!%ccl.net Tue Oct 17 23:43:00 2006
From: "Rajendran Arivazhagan raji],[anal.chem.tohoku.ac.jp" <owner-chemistry!^!server.ccl.net>
To: CCL
Subject: CCL:G: Effect of basis set on frequency calculations with g03
Message-Id: <-32826-061017234146-25369-FTtaKQ2a6xfrUo7w0B13iA!^!server.ccl.net>
X-Original-From: "Rajendran  Arivazhagan" <raji(-)anal.chem.tohoku.ac.jp>
Date: Tue, 17 Oct 2006 23:41:46 -0400


Sent to CCL by: "Rajendran  Arivazhagan" [raji[a]anal.chem.tohoku.ac.jp]
Dear CCL Members, 
     I am doing frequency calculation for protonated organic molecule using g03. My molecule have two protonation sites, say for example site1 and site2. Freq calculation with B3LYP/6-311+G(d) gave an imaginary/negative frequency for site1 and all real frequencies were observed for site2. My input file looks like: 

B3LYP/6-311+G(d) SCF=Tight test Opt Freq Pop=full gfprint SCRF=(CPCM, Solvent=Water)

But the results are completely opposite with B3LYP/3-21G and here negative frequency was observed for site2 and all real frequencies were observed for site1. Input file looks like:

B3LYP/3-21G SCF=Tight test Opt Freq Pop=full gfprint SCRF=(CPCM, Solvent=Water)

So which result I can believe and where is the problem? Basis set with polarization and diffuse functions are not suitable for frequency calculations? Any suggestion is most welcome. Thank you very much.
A.Rajendran.