5
$\begingroup$

I have noticed that for certain Hessian calculations that I run with GAMESS(US), the IR intensities are not produced in the log file. Below I have an example input card (without the molecule $DATA group) of a Hessian calculation I ran that didn't produce the IR intensities only the vibrational frequencies (also given below).

Why is this the case and how can I resolve it?

Input Card

$CONTRL SCFTYP=UHF MULT=1 NPRINT=0 COORD=UNIQUE 
RUNTYP=HESSIAN ICUT=12 ITOL=25 DFTTYP=BLYP      
MAXIT=200 QMTTOL=1E-7 ICHARG=0 ISPHER=1 $END    
$SYSTEM MWORDS=800 MEMDDI=800 $END              
$STATPT OPTTOL=1E-6 NSTEP=200 $END              
$FORCE METHOD=FULLNUM NVIB=2 PROJCT=.TRUE. $END 
$SCF DIRSCF=.TRUE. FDIFF=.FALSE. CONV=1d-7 $END 
$DFT JANS=2 $END                                
$BASIS GBASIS=ACCQ $END                         

Frequency Output

FREQUENCIES IN CM**-1, IR INTENSITIES IN DEBYE**2/AMU-ANGSTROM**2,
REDUCED MASSES IN AMU.

                       1           2           3           4           5
   FREQUENCY:         0.00        0.00        0.00        0.00        0.00  
    SYMMETRY:         B2          B2          B2          B2          B2  
REDUCED MASS:      1.45984     5.02837    27.97693     1.31228     1.81766

Note, that normally the output would also include the IR intensities for each frequency.

$\endgroup$
5
  • 2
    $\begingroup$ Which version of GAMESS are you using? I tried the same input file on my laptop with an H2 molecule (I modified the $system section to change memory), and I am getting vibrational frequencies without any issue. $\endgroup$
    – S R Maiti
    Feb 5, 2021 at 21:13
  • 2
    $\begingroup$ Also, may I ask why you are using fully numeric differentiation? U-BLYP has analytic hessian programmed into the code. Apart from that, you can try adding vibanl=.true. in the \$force section. Another thing would be to look into the punched .dat file, see if it contains the \$hess group. $\endgroup$
    – S R Maiti
    Feb 5, 2021 at 21:14
  • 3
    $\begingroup$ @ShoubhikRMaiti Those two comments can fit into one. In the future if you add a comment within 5 minutes of the previous comment, please just edit the first commend with the second comment attached (so both comments combined in one). You can't do it now because you can only edit comments within the first 5 minutes, but if you make too many comments then the system will try to get the comment chain moved to chat, and we want to avoid this: meta.stackexchange.com/q/353643/391772 $\endgroup$ Feb 5, 2021 at 21:36
  • 1
    $\begingroup$ @ShoubhikRMaiti I am using version 30 JUN 2019 R1 Patch 1. The issue isn't not getting vibration frequencies, it is not getting IR intensities. It works for some molecules but for some it does not, maybe it is the numerical differentiation that causes the issue, I will check that. Thank you. $\endgroup$
    – Cavenfish
    Feb 5, 2021 at 22:00
  • 2
    $\begingroup$ @Cavenfish Sorry, I misread your question. I am getting IR intensities too, without any problem. But I am using Septemeber 30 2020 R2 version, so that might be causing the difference. Also, if you have the \$hess and \$dipdr group from a calculation, GAMESS should be able to read that and print the IR intensities, although I have never tried that before. $\endgroup$
    – S R Maiti
    Feb 6, 2021 at 10:16

1 Answer 1

5
$\begingroup$

I was able to figure out what was going wrong, but forgot to check back here and leave an answer if none were give.

@ShoubhikRMaiti should be credited with helping me figure this out. From this comment

Also, may I ask why you are using fully numeric differentiation? U-BLYP has analytic hessian programmed into the code.

I was curious to try removing this from the input card to see if that resolved the issue. That is exactly what was needed. I can't explain it, but if fully numerical differentiation is used for the Hessian it will not print out infrared intensities along with the vibrational frequencies.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .