[molpro-user] No imaginary frequency but SURF cannot run

[Guntram Rauhut]

Dear MJ Wu,

most likely the problem you face is due to the extremely low harmonic 
frequency of 7.5 cm-1. In the harmonic frequency program Molpro has the 
peculiarity to separate low frequencies from the rest. Once this happens 
only a subset of harmonic frequencies will be passed to the SURF 
program, which then may cause strange effects. You may avoid this by 
providing the option "low=5.0" to the harmonic frequency calculation. 
Besides this, a frequency of just 7.5 cm-1 may cause you severe trouble 
within the generation of the PES. You need also to be aware that your 
current input most likely will perform more than 100.000 MCSCF 
single-point calculations. Once you know the CPU time needed for one 
point you can easily estimate the time for generating the entire PES.

Best wishes,

     Guntram


On 08/14/2013 07:06 PM, Kenny W wrote:
> Dear Molpro Experts,
>
> I tried to perform a VSCF/VCI calculation for a nine-atom anion in the 
> doublet state, after the geometry optimization followed analytic 
> frequency at MCSCF level. The frequencies reported were all positive 
> values, however, the subsequent 1D SURF calculation complained about 
> imaginary frequency and could not continue.
>
> First, the MCSCF OPTG and FREQUENCY calculation keywords are shown as 
> follows:
>
> *NOSYM;*
> *orient,mass*
> *mass,iso*
> *basis={default=aug-cc-pwCVDZ}*
> *
> *
> *{multi,energy=1.0e-8,gradient=1.0e-5,step=1.0e-5*
> *maxiter,40*
> *occ,50;closed,37;        !CAS(13,13)*
> *wf,87,1,1;*
> *natorb,print=20,ci,state=1.1}*
> *
> *
> *{optg,maxit=100,gradient=1.0e-5,energy=1.0e-8,step=1.0e-5}*
> *pop;*
> *
> *
> *{multi,energy=1.0e-8,gradient=1.0e-5,step=1.0e-5;cpmcscf,hess,accu=1.d-5*
> *maxiter,40*
> *occ,50;closed,37; *
> *wf,87,1,1;*
> *natorb,print=10,ci,state=1.1}*
> *
> *
> *{frequencies,analytical*
> *print,hessian,low}*
> *
> *
> *put,molden,cas1313_awcdz_nosym_optfrq.molden;*
> *---*
>
> and the reported frequencies (all positive) are:
>
> *_FREQUENCIES(1:27)   =      0.00001845     0.00001629     0.00000707 
>     0.00000225     0.00001146     0.00001581*
> *                             7.52426274    58.84544984    64.37850876 
>    70.83457530   103.14123353   126.11691779*
> *                           137.75677617   181.48189951   234.48735837 
>   270.42956902   345.45378008   427.77056976*
> *                           461.65694834   524.95228098   531.51838006 
>   575.23162819   586.87961440  1633.71618449*
> *                          1637.44993389  1999.82375492  2062.79045692*
>
> The second step was to calculate VSCF and VCI by restarting from the 
> files 1 and 2 of the previous calculation:
>
> * restart;*
> * file,1,vscf.int*
> * file,2,vscf.wfu*
> * file,3,vscf.res,new*
> *NOSYM;*
> *orient,mass*
> *mass,iso*
> *basis={default=aug-cc-pwCVDZ}*
> *
> *
> *label1*
> *{multi,energy=1.0e-8,gradient=1.0e-7,step=1.0e-5*
> *maxiter,40*
> *occ,50;closed,37;*
> *wf,87,1,1;*
> *state,1*
> *natorb,2141.3,state=1.1*
> *cpmcscf,grad,1.1,accu=1.d-9}*
> *
> *
> *{surf,start1D=label1,ndim=3,dellog=1,sym=auto    !(3) generate PES*
> *scalnm,auto=on*
> *intensity,dipole=2}*
> *
> *
> *vscf *
> *vci,pmp=3,version=1 *
> *vci,pmp=3,version=2*
> *---*
>
> The job bump out at SURF program, with the final dying message:
>
> * You have an imaginary vibration. Please check the FREQ calculation.*
>
> However in the previous job all of the harmonic frequencies were found 
> positive. May I know what was wrong with my SURF input? Thank you very 
> much for your kind attentions.
>
> ==========
> MJ Wu
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user

-- 
********************************************************************************

  Prof. Dr. Guntram Rauhut
  Institut f. Theoretische Chemie
  Universitaet Stuttgart
  Pfaffenwaldring 55
  D-70569 Stuttgart
  Germany

  Tel. :   +49/(0)711/685-64405
  FAX :    +49/(0)711/685-64442
  E-Mail : rauhut at theochem.uni-stuttgart.de
  HTTP :   www.theochem.uni-stuttgart.de/~rauhut/

********************************************************************************

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20130815/15f38b6f/attachment.html>