DFT-D2 parameters for tungsten

[ashish dabral]

Hi,

Can someone tell me how to obtain DFT-D2 parameters for tungsten?

I am trying Vanderwal's correction for a tungsten atom on top of a substrate using 

&VDW_POTENTIAL

        POTENTIAL_TYPE PAIR_POTENTIAL

        &PAIR_POTENTIAL

          TYPE DFTD2

          REFERENCE_FUNCTIONAL PBE

          R_CUTOFF  10.0

          SCALING 0.0

        &END PAIR_POTENTIAL

&END VDW_POTENTIAL

unit cell height is 20 angstrom for the monolayer substrate.

I am getting an error that the dispersion parameters are not found for tungsten. I use DZVP-MOLOPT-SR-GTH basis and PBE potential.

Thanks

[Andreas Funk]

Why do you want to use the D2 parameters? The D3 set of parameters is much more reliable and the D2 set should no longer be used. If you want to obtain system-specific coefficients, I recommend to read Ehrlich et al. (2011) System-Dependent Dispersion Coefficients for the DFT-D3 Treatment of Adsorption Processes on Ionic Surfaces. ChemPhysChem 12(17): 3414-3420. doi: 10.1002/cphc.201100521

[ashish dabral]

Thanks for your reply Andreas.

I know D3 parameters are more reliable and would like to use them. I used the D2 keyword as mentioned in my post to do simulations for 

oxygen on substrate and it worked. Same with Nickel, but with tungsten it throws an error that the parameters are not found. I assume the 

parameters are defined in the package and in case of tungsten they are not found. 

Are D3 parameters also defined in the package? Some posts say that they are not and in that case I will need an external file.

I just wanted to test a couple of things and would move to D3 when more reliability is required and I have the D3 parameters. 

[Andreas Funk]

D3 parameters for tungsten are available in the package. Those found in cp2k/tests/QS/dftd3.dat seem to be correct.

[ashish dabral]

Thanks a lot Andreas! I will use them.