I ran into an issue trying to run dftb.org parameter files on G09 and found questions asked on this but not answers. I thought I would post the solution here to help anyone else who may run into this problem.

  • $\begingroup$ (Remember to note these changes in your method description.) $\endgroup$ – Aesin Jun 19 '13 at 20:18

First, at least for C.01, dftb=readparameters, in the route section, does not work dftb=read does but is not necessary.

Skip a line after molecular descritption and include your parameter files there. Make sure you include all pair interaction files, not just bonded atoms. It should look something like this:

@C:/G09W/C-C_mio.skf @C:/G09W/C-H_mio.skf @C:/G09W/H-C_mio.skf @C:/G09W/H-H_mio.skf

(I included the mio because we are running more than one parameter set, not a requirement just a name)

The real glitch comes from my final hint. The DFTB parameters generally have more than 500 points, but the first line in the code is 500 (not 519, or 520, etc). G09 Rev. D.01 doesn't care about this, B.01 and C.01 do. You simply need to change 500 ro reflect the correct amount of points and add in the correct order for your atomic numbers (this part is explained on Gauss' website) and you are good to go.

Hopefully this helps anyone trying to run DFTB on G09 for the first time, and helps save some time.


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