Does van der Waals interaction description work well on VASP DFT?
Posted: Wed Mar 09, 2011 12:51 pm
I use VASP.5.2.11 which supports van der Waals interaction calculation. When doing bulk graphite relaxing, I set the related parameters as follows:
LVDW=.TRUE.
VDW_RADIUS=30
VDW_SCALING=0.75
VDW_D=20.0
VDW_C6=1.75
VDW_R0=1.452
The inter-layer separation I got was nearly 3.2A, which is unacceptable.
In the VASP manual(on web), the dispersion energy is defined as E=-(s/2)*∑∑C*f/r^6.
But as I know, dispersion energy is defined as E=-s*∑∑C*f/r^6,which is discribed in Grimme's paper.
Maybe I made a mistake, but what is the truth?
Thanks in advance!
LVDW=.TRUE.
VDW_RADIUS=30
VDW_SCALING=0.75
VDW_D=20.0
VDW_C6=1.75
VDW_R0=1.452
The inter-layer separation I got was nearly 3.2A, which is unacceptable.
In the VASP manual(on web), the dispersion energy is defined as E=-(s/2)*∑∑C*f/r^6.
But as I know, dispersion energy is defined as E=-s*∑∑C*f/r^6,which is discribed in Grimme's paper.
Maybe I made a mistake, but what is the truth?
Thanks in advance!