8
$\begingroup$

In my bachelor thesis, I am trying to use the VEDA program to analyze the vibrational energy distribution to associate different vibrations to the calculated modes from gaussian. I am stuck now because the VEDA program seems to provide unrealistic output. I calculated methane as a simple way to show my problems.

The input file to gaussian:

%chk=methan.chk
#p opt freq=SaveNormalmodes B3LYP/6-311G(d,p) nosymm

 methan

0 1
6         -2.69391        0.67007        0.00000
1         -1.62391        0.67007        0.00000
1         -3.05057       -0.15258       -0.58389
1         -3.05058        1.58706       -0.42049
1         -3.05058        0.57573        1.00438

I use the log and fchk file from this calculation as input for VEDA. VEDA creates the .fmv file from this input.

#p opt freq=SaveNormalmodes B3LYP/6-311G(d,p) nosymm        

Atomic orientation and masses
  1   6   -2.69391033    0.67006991    0.00000005     12.00000000
  2   1   -1.60343431    0.67007262    0.00000174      1.00782504
  3   1   -3.05739820   -0.16832644   -0.59506760      1.00782504
  4   1   -3.05740398    1.60460983   -0.42853870      1.00782504
  5   1   -3.05740406    0.57392428    1.02360451      1.00782504

Force constants in Cartesian coordinates
 5.33572579E+00  3.66810060E-02  3.66778578E-02  3.66800417E-02  3.04938267E-02
 4.78002374E-02  9.34030448E-02 -1.23337201E-02 -7.49599366E-02 -8.44033441E-02
 5.90658770E-02 -4.86412743E-02  6.40541790E-02  1.12695453E-01  0.00000000E+00
 3.66810060E-02  5.33567095E+00  3.66758568E-02  3.66792176E-02  3.05027864E-02
-1.23355319E-02 -7.49650905E-02  4.77820938E-02  9.34122780E-02 -8.43965515E-02
 5.90646100E-02 -4.86546903E-02 -1.12696475E-01 -6.40417842E-02  0.00000000E+00
 3.66778578E-02  3.66758568E-02  5.33575627E+00  3.66785235E-02 -1.91501878E-02
 5.31942717E-02 -8.29810618E-02  5.32006592E-02 -8.29790123E-02  7.87153470E-02
 4.16992065E-02  1.25083636E-01  6.25372507E-02 -6.25463857E-02  0.00000000E+00
 3.66800417E-02  3.66792176E-02  3.66785235E-02  5.33573582E+00 -4.18424802E-02
-8.86563336E-02  6.45291179E-02 -8.86563832E-02  6.45448771E-02  9.00810935E-02
-1.59827744E-01 -2.77950854E-02 -1.39023999E-02  1.38926855E-02  0.00000000E+00
 3.04938267E-02  3.05027864E-02 -1.91501878E-02 -4.18424802E-02  3.03668368E-01
-1.34786428E-01 -2.01270285E-01 -1.34786926E-01 -2.01266979E-01  3.68441953E-01
 2.06837032E-01 -1.25413382E-01 -6.27074976E-02  6.27058841E-02  7.78459520E+01
 4.78002374E-02 -1.23355319E-02  5.31942717E-02 -8.86563336E-02 -1.34786428E-01
 5.72465784E-01 -3.47913849E-01 -3.19509681E-02  3.82124777E-01 -4.39938925E-01
 1.65917376E-01  1.44129337E-01  2.18665399E-01  7.45360625E-02 -6.80511800E+00
 9.34030448E-02 -7.49650905E-02 -8.29810618E-02  6.45291179E-02 -2.01270285E-01
-3.47913849E-01  1.11839029E+00  3.82124891E-01 -5.42349371E-01 -4.08981216E-01
-2.25554975E-01 -1.06247502E-01  8.31603827E-03  1.14563541E-01  1.70127922E+00
-1.23337201E-02  4.77820938E-02  5.32006592E-02 -8.86563832E-02 -1.34786926E-01
-3.19509681E-02  3.82124891E-01  5.72467245E-01 -3.47916055E-01 -4.39937727E-01
 1.65917460E-01  1.44130988E-01 -7.45350363E-02 -2.18666024E-01  1.70128008E+00
-7.49599366E-02  9.34122780E-02 -8.29790123E-02  6.45448771E-02 -2.01266979E-01
 3.82124777E-01 -5.42349371E-01 -3.47916055E-01  1.11839152E+00 -4.08983943E-01
-2.25552099E-01 -1.06251760E-01 -1.14564521E-01 -8.31276143E-03  1.70127939E+00
-8.44033441E-02 -8.43965515E-02  7.87153470E-02  9.00810935E-02  3.68441953E-01
-4.39938925E-01 -4.08981216E-01 -4.39937727E-01 -4.08983943E-01  1.32939890E+00
-8.75646378E-02  4.96524369E-02  2.48257890E-02 -2.48266478E-02  1.70127933E+00
 5.90658770E-02  5.90646100E-02  4.16992065E-02 -1.59827744E-01  2.06837032E-01
 1.65917376E-01 -2.25554975E-01  1.65917460E-01 -2.25552099E-01 -8.75646378E-02
 4.56529024E-01  2.29232226E-02  1.14607952E-02 -1.14624275E-02  0.00000000E+00
-4.86412743E-02 -4.86546903E-02  1.25083636E-01 -2.77950854E-02 -1.25413382E-01
 1.44129337E-01 -1.06247502E-01  1.44130988E-01 -1.06251760E-01  4.96524369E-02
 2.29232226E-02  2.83728369E-01  1.41864449E-01 -1.41863920E-01  0.00000000E+00
 6.40541790E-02 -1.12696475E-01  6.25372507E-02 -1.39023999E-02 -6.27074976E-02
 2.18665399E-01  8.31603827E-03 -7.45350363E-02 -1.14564521E-01  2.48257890E-02
 1.14607952E-02  1.41864449E-01  2.88104389E-01  1.46239939E-01  0.00000000E+00
 1.12695453E-01 -6.40417842E-02 -6.25463857E-02  1.38926855E-02  6.27058841E-02
 7.45360625E-02  1.14563541E-01 -2.18666024E-01 -8.31276143E-03 -2.48266478E-02
-1.14624275E-02 -1.41863920E-01  1.46239939E-01  2.88103859E-01  3.11383808E+01
 0.00000000E+00  0.00000000E+00  0.00000000E+00  0.00000000E+00  7.78459520E+01
-6.80511800E+00  1.70127922E+00  1.70128008E+00  1.70127939E+00  1.70127933E+00
 0.00000000E+00  0.00000000E+00  0.00000000E+00  3.11383808E+01  0.00000000E+00

Harmonic frequencies
F =3134.24   IR =  27.91   S = A
   3  -7   5    -31  -1   1      4   6   7    -28  72 -32     15   2 -40   
F =3134.23   IR =  27.91   S = A   
  -7   1   5     69   0   1     -5  -8  -4      1  -7   5     22   6 -66   
F =3134.22   IR =  27.91   S = A   
  -4  -6  -5     41  -1  -1     28  66  46     -7  15  -9    -10  -4  24   
F =3028.36   IR =   0.00   S = A
   0   0   0     50   0   0    -17 -38 -27    -17  43 -20    -17  -4  47   
F =1560.38   IR =   0.00   S = A   
   0   0   0      0 -33 -38     46 -20   0    -14  16  46    -32  37  -8   
F =1560.38   IR =   0.00   S = A   
   0   0   0      0  38 -33    -11 -25  42     45  21   6    -34 -33 -15   
F =1341.02   IR =  16.94   S = A
   0 -12  -4      0  57  19    -22   9 -15     15  16  38      4  59   5   
F =1341.02   IR =  16.94   S = A   
  -7   3 -10     -5 -15  47     24 -38  31     18  27  28     47 -12  10   
F =1341.02   IR =  16.94   S = A   
 -10  -3   7     -8  12 -32     47   5 -37     52   6 -29     31   7  20   

To receive the potential energy distribution I create the .dd2 file with VEDA and use the standard optimization options in Veda. I optimize the file until there is no more improvement in the Emp value calculated by VEDA.

methan.vdf

abbreviation of: e:\bachelorarbeit\veda\methan\methan.ved
IR spectrum from file: e:\bachelorarbeit\veda\methan\methan.log
diagonality factor = 51.86   <EPm> = 53.85
    IR      CM-1
  27.91  3134.24  *8201.25   s2 48   s5 10   s9 29
  27.91  3134.23  *2652.43   s3 92
  27.91  3134.22  *1441.36   s2 -10   s5 -16   s9 57
   0.00  3028.36  *1434.71   s1 27   s7 -27   s8 37
   0.00  1560.38  *1382.52   s4 -15   s5 14   s7 31   s8 22
   0.00  1560.38  *1081.74   s4 55   s7 17   s8 16
  16.94  1341.02  * 885.94   s1 51   s5 22
  16.94  1341.02  * 447.29   s4 17   s6 76
  16.94  1341.02  *-8129.88   s2 -40   s5 39   s8 12
with (*) frequencies calculated by Veda

methan.dd2

Average max. Potential Energy <EPm> = 53.852
TED Above 100 Factor TAF=0.000
Average coordinate population 1.333
Most complex coordinate No. 4 , population = 2 
s 1     1.00   STRE    1    4   CH    1.090476  f1435 27  f886 51
s 2     1.00   STRE    1    5   CH    1.090476  f8201 48  f1441 10  f8130 40
s 3     1.00   STRE    1    2   CH    1.090476  f2652 92
s 4     1.00   STRE    1    4   CH    1.090476  f1383 15  f1082 55  f447 17
        1.00           1    3   CH    1.090476
s 5     1.00   BEND    2    1    3   HCH   109.47  f8201 10  f1441 16  f1383 14  f886 22  f8130 39
        1.00           3    1    4   HCH   109.47
s 6     1.00   BEND    3    1    4   HCH   109.47  f447 76
s 7     1.00   BEND    4    1    5   HCH   109.47  f1435 27  f1383 31  f1082 17
        1.00           3    1    4   HCH   109.47
s 8     1.00   OUT     1    4    2    5   CHHH    19.47  f1435 37  f1383 22  f1082 16  f8130 12
s 9     1.00   OUT     1    3    2    4   CHHH    19.47  f8201 29  f1441 57

Why differ the calculated modes from methan.vdf file to this extent in comparison to the calculated modes from gaussian? Is there an error in my input, which results in that difference? The calculations of methane should result in CH-stretching vibrations at 3000 1/cm. The assignment by VEDA for the mode at 3134.23 1/cm from the gaussian log file or 2652.43 1/cm calculated by VEDA seems right. The other modes over 3000 1/cm are partly assigned to the bending-vibrations s5 and out-of-plane-vibrations s9 which are described in the dd2 file. I do not understand why the VEDA program assigns these vibrations to a mode at 3000 1/cm. In a vibration visualization by VEDA and other programs, the modes calculated over 3000 1/cm are clearly identified as CH-stretching vibrations.

$\endgroup$
5
  • $\begingroup$ Where do you see a bending assignment to the high-frequency modes? - It would probably make sense if you cut down all your output to just the relevant parts. $\endgroup$
    – Antimon
    Oct 14 at 18:42
  • $\begingroup$ Thanks for the edit, it's more clear now. I was able to reproduce your results using your fmv input. However, I also ran your initial Gaussian input, formchk'd the checkpoint file, renamed it to fmu, and plugged it into VEDA, and it worked fine. So I think your problem must be somewhere further upstream. For reference, here are the fmv file and the vdf file that VEDA produced from my formatted checkpoint file. You can see that the force constants are noticeably different. $\endgroup$
    – Antimon
    Oct 18 at 18:44
  • $\begingroup$ Thank you for the help. The problem seems to be my gaussian calculation. Do you have an idea what I did wrong? Why do we get different results from the same input file? I tried to vary the functional and basis set and also tried calculations with and without keywords like nosymm. What option did you use to convert the chk file to the fchk file? I use the formchk -3 option. $\endgroup$
    – Till
    Oct 20 at 9:15
  • $\begingroup$ Let's continue this in the chat. $\endgroup$
    – Antimon
    Oct 20 at 18:54
  • 1
    $\begingroup$ @Antimon not a huge issue, but we have generally been trying to limit the creation of new chat rooms by sorting things into topic rooms, see here. We think it preserves discussions better and avoids info being buried in old frozen chatrooms. In the future, check to see if a topic is already covered by a room and mods or even other users can transfer comments to that room. If no room covers it, we can discuss making a new room or altering the topic of an underused room. $\endgroup$
    – Tyberius
    Oct 23 at 13:04
5
$\begingroup$

After some experimentation, I was able to track down the problem.

Your output files tell me that you are using Gaussian 16. Upon inspection, the formchk utility in that version seems to put an "Internal Force Constants" section into the formatted checkpoint file in addition to the "Cartesian Force Constants" that VEDA is expecting. Thus, VEDA gets tripped up and reads junk from the file. You can confirm this by deleting that "Internal" section from the formchk file, or even just mangling the word "Force" in the title line so that VEDA doesn't pick it up anymore. Then the results are reasonable again.

I did my tests with Gaussian 09, where the formchk utility doesn't print those problematic internal forces. Thus I was not able to reproduce it at first. However, if I use the formchk utility from Gaussian 16 on your chk file, then I get the same issue.

VEDA seems to work fine if you just rename the Gaussian 16 output file to fmt and read it in. That would probably be the easier way to go than messing with your fchk files.

$\endgroup$
1
  • 1
    $\begingroup$ Thank you for your help! The problem can also be solved by using the formchk -2 option in gaussian 16 to create a gaussian 09 fchk file. After being renamed to fmu and put into VEDA it creates the same results. $\endgroup$
    – Till
    Oct 21 at 14:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.