DFT-D2 on VASP 5.2.11

[abruix]

Dear VASP people,

I am trying to use the DFT-D2 as implemented in vasp 5.2.11. I have succesfully run calculations with only C atoms and have been able to introduce different parameters for vdw_c6 and vdw_r0. However, when I put Au on my system vasp does not read the parameters I introduce. It does not work with only Au eihter.

The part of the OUTCAR regarding dispersion forces looks like this:


Grimme's potential added to compensate for missing vdW interactions
-------------------------------------------------------------------

Parameters for Grimme's potential:
C6(Jnm^6/mol) R0(A)
-----------------------------
^@^@ 0.000 0.000
C 1.750 1.452

Cutoff radius: 12.000 A

Global scaling factor: 0.750 A

Parameter d: 20.000 A


And the INCAR was:


LVDW = .TRUE.
VDW_RADIUS = 12
VDW_SCALING = 0.75
VDW_D = 20

VDW_C6 = 40.62 1.75
VDW_R0 = 1.772 1.452


With this setup, if I change the parameters for C it still reads them but never for Au.

Thank you for your valuable help,

Albert.

[clinton]

As far as I know, VASP can set the VDW_C6 and VDW_R0 itself according to your POSCAR. But not all elements are included, and Au is not included in the code. So you can hardly do DFT-D2 calculation when you have Au in your system; at least the interaction with Au is not included.

Btw, how do you know the VDW_C6 and VDW_R0 for Au is 40.62 and 1.772 respectively?

[abruix]

It says in the manual that the default values are included for the first five rows of the periodic table and that parameters for the rest of the elements should be explicitly included.

With respect to the values, you can derive them yourself or you can look at the literature to see if anyone has derived the values for the systems you are interested in. So far, I am just making some tests to see how this method behaves.

By the way, why do you think an interlayer separation of 3.2 is unacceptable?

[santana]

...
<span class='smallblacktext'>[ Edited Wed Apr 18 2012, 06:14PM ]</span>