From owner-chemistry@ccl.net Sat Dec 7 16:37:00 2013 From: "William F. Polik polik#,#hope.edu" To: CCL Subject: CCL: scaling factor in frequency calculation Message-Id: <-49412-131207160135-2844-f5w63eZgEUoeVFL89I6Q4Q .. server.ccl.net> X-Original-From: "William F. Polik" Content-Type: multipart/alternative; boundary="------------020202020406050400060001" Date: Sat, 07 Dec 2013 16:01:26 -0500 MIME-Version: 1.0 Sent to CCL by: "William F. Polik" [polik:-:hope.edu] This is a multi-part message in MIME format. --------------020202020406050400060001 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit A few years ago I heard a wonderful talk by Leo Radom from Australia National University in which he described their work on deriving scaling factors for various methods and basis sets (and the associated complications). In paticular, I recall that low frequency vibrations might need to be treated differently. Let me simply refer you to their review at: Jeffrey P. Merrick, Damian Moran, and Leo Radom, "An Evaluation of Harmonic Vibrational Frequency Scale Factors", J. Phys. Chem. A 2007, 111, 11683-11700 Will Polik On 12/2/13 12:02 PM, partha kundu partha1kundu{:}gmail.com wrote: > Dear All, > I am trying to see the effect of Cu2+ binding on a ligand by Raman > spectroscopy and trying to match those results with DFT. For the > ligand, I got a very good match with the experimental spectrum after > applying a scaling factor. However for the complex if I use the same > scaling factor, I see some modes in the 1500-1700 cm-1 regions are > underestimated. If I do not use any scaling factor in that the match > is better. I am using lanl2dz ECP and basis set combination for Cu. Is > it okay to use dual scaling (0.961 for 1000-1500 and 1 for the > 1500-1700 region) for matching the experimental data. Then I can not > directly compare my theoretical results with the experiment. Any > suggestion would be of great help. > Partha -- ------------------------------------------------------------------------ Dr. William F. Polik Hofma Professor of Chemistry and Chairperson Department of Chemistry Schaap Science Center 2122 Hope College 35 East 12th Street Holland, MI 49422-9000 USA polik:hope.edu http://www.chem.hope.edu/~polik tel: (616) 395-7639 fax: (616) 395-7118 ------------------------------------------------------------------------ --------------020202020406050400060001 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit A few years ago I heard a wonderful talk by Leo Radom from Australia National University in which he described their work on deriving scaling factors for various methods and basis sets (and the associated complications). In paticular, I recall that low frequency vibrations might need to be treated differently. Let me simply refer you to their review at:

Jeffrey P. Merrick, Damian Moran, and Leo Radom, "An Evaluation of Harmonic Vibrational Frequency Scale Factors", J. Phys. Chem. A 2007, 111, 11683-11700

Will Polik

On 12/2/13 12:02 PM, partha kundu partha1kundu{:}gmail.com wrote:

Dear All,

I am trying to see the effect of Cu2+ binding on a ligand by Raman spectroscopy and trying to match those results with DFT. For the ligand, I got a very good match with the experimental spectrum after applying a scaling factor. However for the complex if I use the same scaling factor, I see some modes in the 1500-1700 cm-1 regions are underestimated. If I do not use any scaling factor in that the match is better. I am using lanl2dz ECP and basis set combination for Cu. Is it okay to use dual scaling (0.961 for 1000-1500 and 1 for the 1500-1700 region) for matching the experimental data. Then I can not directly compare my theoretical results with the experiment. Any suggestion would be of great help.

Partha

Dr. William F. Polik
Hofma Professor of Chemistry and Chairperson

Department of Chemistry
Schaap Science Center 2122
Hope College
35 East 12th Street
Holland, MI 49422-9000
USA

polik:hope.edu
http://www.chem.hope.edu/~polik
tel: (616) 395-7639
fax: (616) 395-7118

--------------020202020406050400060001--