set up for DFT-D3 method

[jiabo]

Hi everyone

When I am using the DFT-D3 method for my calculation, I found a new feather that I did not use before ( CALCULATE_C9_TERM ).  This term is calculating the three-body effect, I am not very sure if it is necessary to use this term when employing DFT-D3 because I did not turn it on for my pervious calculation. 

The vdw section is shown below, CALCULATE_C9_TERM is commented out now. Should I turn it on for all DFT-D3 calculation or depends on my test calculation?

      &VDW_POTENTIAL

         DISPERSION_FUNCTIONAL PAIR_POTENTIAL

         &PAIR_POTENTIAL

            TYPE DFTD3

            PARAMETER_FILE_NAME ./dftd3.dat

            REFERENCE_FUNCTIONAL PBE

#            CALCULATE_C9_TERM  TRUE

#           REFERENCE_C9_TERM  TRUE

         &END PAIR_POTENTIAL

      &END VDW_POTENTIAL

Bests,

Jiabo

[S Ling]

Hi

According to our own experience (on metal-organic frameworks), the three-body term gives better description to energetics in many cases, and the structures with and without the three-body term are quite similar. There is a recent paper from Prof Stefan Grimme's group on the effect of the three-body term on large molecular complexes, see http://dx.doi.org/10.1021/ct301081n. If your system size is big, I believe the additional computational cost due to the three-body term can be neglected.

SL

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at https://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.