Dear All,
can anybody please advise me how to use correctly the GD2 empirical dispersion correction for the B3LYP and CAM-B3LYP functionals? I added the keyword EmpiricalDispersion=GD2 to the command line, e.g.:
#B3LYP/6-31G* scf=xqc guess=read EmpiricalDispersion=GD2 opt=(readfc,maxcycle=400,cartesian)
and received the "Syntax error" message. Thus, I am feeling confused, what might be wrong here. I am using the Gaussian 09 version B.01 .
Thanks a lot for your help!
Aleksey
Aleksey E. Kuznetsov Dear professor,
the keyword enabling D2 version of Grimme dispersion (EmpiricalDispersion=GD2) is correctly added, here. Can you please share the full error message.
I think the error is related to another keyword. For instance, you are using "guess=read", this allows the initial guess to be read from the checkpoint file. I'm not expert, but I believe this keyword (guess=read) is often specified along with Geom=Checkpoint. My second remark is : You are employing B3LYP and CAM-B3LYP ( long range corrected version of B3LYP functional) , and you want to add a Grimme dispersion correction. I strongly suggest the use of the wB97XD functional which by default includes simultaneously, empirical dispersion and long range corrections.
Dear professor, Aleksey
I think this error may be due to using an older version of Gaussin-09. EmpiricalDispersion=GD2 function is correctly working in Gaussian-09 D.01. Version. I tried it and its working for me correctly in the D.01 version
Dear Melek and Adnan,
thank you very much for your responses and advices!
With kind regards,
Aleksey.
- Badges
- Science topic
- Similar topics
- Correction