[molpro-user] ccsd(t) frequencies

[Kadir Diri]

Hello!
Just as a note for (1) below. If you use symmetry in the calculation, 
many programs may be stuck at some stationary point that is not a 
minimum, simply because they cannot break the symmetry. I don`t know how 
Molpro handles these, but maybe worth checking.
If you don`t have symmetry, I would say your problem is probabbly (2), 
because it is a typical case in molecular clusters. (I assume you`re 
optimizing one of those, by looking at your "edimer=energy" line in the 
input)
Cheers!
kadir



Li, Jun wrote:

>I am not sure if I fully understand your question, but in general one
>has two possibilities:
>
>(1) your geometry optimization locates a stationary point, which is not
>a minimum.  Then you will really have some imaginary frequencies.
>
>(2) your molecule is indeed optimized to its (local or global) minimum,
>but the PES is so flat that you have very small REAL frequencies.  These
>small frequencies are difficult to be calculated accurately and will
>actually show as imaginary unless your geometry optimization reached
>very tight convergence criteria (which then requires high accuracy in
>the integrals and the SCF convergence).  
>
>Jun Li
>
>-----Original Message-----
>From: owner-molpro-user at molpro.chem.cf.ac.uk
>[mailto:owner-molpro-user at molpro.chem.cf.ac.uk] On Behalf Of Omololu
>Sent: Monday, August 22, 2005 3:22 PM
>To: molpro-user at molpro.net
>Subject: [molpro-user] ccsd(t) frequencies
>
>Dear All,
>
>I did a ccsd(t) geometry optimization and located the minimum. However,
>when i try to calculate the ccsd(t) frequencies, i get a couple of
>imaginary frequencies in addition to the other ones.
>Any suggestions would be appreciated.
>
>o.
>PS:
>The relevant part of my input is:
>
> hf;                                                  
>                          
> ccsd(t);                                             
>                          
>edimer=energy                                         
>                         
> optg,rstep=0.0002,maxit=100,grad=3.5d-8;coord,3n
> frequencies
>---
>
>and the relevant part of my output:
>
> Imaginary Vibration  Wavenumber
>        Nr             [1/cm]
>        1               36.17
>        2               18.21
>
>   Low Vibration      Wavenumber
>        Nr             [1/cm]
>        1                7.36
>        2               10.90
>        3               33.85
>        4               44.84
>
>     Vibration        Wavenumber
>        Nr             [1/cm]
>        1             1645.08
>        2             3810.40
>        3             3919.50
>
> FREQUENCIES * CALCULATION OF NORMAL MODES FOR CCSD(T)
>
> Gradient norm at reference geometry: 0.37750D-04
>
>***************** SEEK GOD! *******************
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around
>http://mail.yahoo.com 
>  
>