Stefan Grimme prompted me= to exemplify the statement below, and share the results with the CCL, so h= ere it is.

<= /p>

Take ethane at the HF/6-3= 1G* level. The optimized C-H bond lengths in the staggered and eclipsed for= ms are 1.0853 and 1.0846 =C5, respectively<= /o:p>

Following the lowest normal mode in Cartesian coordinates gives th= e following set of connected torsional angles and C-H bond distances:<= /o:p>

60 … 1.085

37 … 1.105

26 … 1.128

16 … 1.160

In Cartesian coordinates the rotation is therefore coupled to C-H = bond changes, and the coupling is in the wrong direction. Since the C-H bon= ds are stiff, this will dominate the energy profile along that normal mode.

Alternatively, if the vibrational analysis is done in internal coo= rdinates, the lowest vibration is a pure torsional mode, and thus completel= y uncoupled from C-H bond stretches and bending, i.e. it corresponds to a rotation with frozen bond distances = and angles. This is of course also incorrect, but I would argue that it is = a better zero’th order approximation to the ‘true’ rotati= onal motion than the mode in Cartesian coordinates.

Alternatively stated, the size of the harmonic region depends on t= he choice of coordinates, in internal coordinates the mode can be followed = further out before the deviation in energy between the ‘mode following’-curve and the ‘true&= #8217;-adiabatic curve becomes significant.

Frank

<= /p>

Frank Jensen
Assoc. Prof.
Dept. of Chemistry
Aarhus University
http://old.chem.au= .dk/~frj

<= /p>

From: owner-ch= emistry+frj=3D=3Dchem.au.dk],[ccl.net [mailto:owner-chemistry+frj=3D= =3Dchem.au.dk],[ccl.net] On Behalf Of Frank Jensen frj.:.chem.au.dk
Sent: 18. april 2013 09:44
To: Frank Jensen
Subject: CCL:G: SV: CCL:G: Scanning potential energy along a normal = mode

The harmonic frequencies = are independent of the coordinates used for the analysis, but any finite di= splacement along the corresponding normal modes depends strongly on the coordinates used for the analysis. Essentially all standar= d programs use Cartesian coordinates, but these are usually poor for repres= enting the actual vibrational motion for a non-trivial displacement. Calcul= ating the energy along a Cartesian normal mode displacement is unlikely to give an accurate picture of the vi= brational potential along that normal mode. If you doubt that, try followin= g the Cartesian version of the normal mode describing the rotation around t= he C –C bond in ethane, and you will rapidly discover that it significantly alters the C-H bond distances = as well. Performing the vibrational analysis in e.g. internal coordinates w= ill give a much better representation for large displacements, but few prog= rams have that capability.

<= /p>

Frank

<= /p>

Frank Jensen
Assoc. Prof.
Dept. of Chemistry
Aarhus University
http://old.chem.au= .dk/~frj

<= /p>

From: owner-ch= emistry+frj=3D=3Dchem.au.dk]=3Dcl.net [mailto:owner-chemistry+frj= =3D=3Dchem.au.dk]=3Dcl.net] On Behalf Of Jens Spanget-Larsen spanget**ruc.dk
Sent: 17. april 2013 19:14
To: Frank Jensen
Subject: CCL:G: SV: CCL:G: Scanning potential energy along a normal = mode

Dear Devandra,

I do not think Gaussian has a procedure to do this. Some time ago = I performed analyses of the so-called bell-clapper vibrational mode of 1,6,= 6a-trithiapentalene by scanning along the normal coordinate. The energy was calculated for a series of points cor= responding to displacement of the nuclei along the coordinate. You can= find the results and some relationships in: Spanget-Larsen & Ande= rsen, PCCP 3, 908-16 (2001).

Jens >--<

------------------------------------------------------

JENS SPANGET-LARSEN &nbs= p; Office: +45 4674 2710

Dept. of Science (18.1) Fax: &= nbsp; +45 4674 3011

Roskilde University &nbs= p; Mobile: +45 2320 6246

P.O.Box 260 = E-Mail: &= nbsp; spanget++ruc.dk

DK-4000 Roskilde, Denmark http://www.ruc.dk/~spanget

------------------------------------------------------

Fra: owner-chemistry+spanget= =3D=3Druc.dk++ccl.net [owner-chemistry+spanget=3D=3Druc.dk++ccl.net] p=E5 vegne af D= evendra Mani devdmani10%a%gmail.com [owner-chemistry++ccl.net]
Sendt: 17. april 2013 15:32
Til: Jens Spanget-Larsen
Emne: CCL:G: Scanning potential energy along a normal mode

Dear CCLers, =

I want to s= can potential energy along a particular normal mode of vibration . Is there= anyway to do it using Gaussian or any other software ?

Any kind of suggestions would be highly appreciated.

Thanks,

Devendra