My Community

Hi all,

I'm doing some slabcalculations, and I noticed that only in some cases the PARCHG-file gives a correct zero(or almost zero, 7 orders of magnitude smaller than bulk) charge while in other situations the chargedensity in vacuum becomes of the same order of that in the bulk.
The calculations are almost identical, the only difference is the kpoints used.
My basic calculation has 6 irreducible kpoints, and this calculation generates a zero chargedensity in the vacuum region, when I do an LPARD=true calculation with
KPUSE= 1 2 3 4 5 6
EINT =-1.5
NBMOD= -3
I get a charge in the vacuum region of the same order of magnitude as the bulk(and it this is 5 to 6 orders of maginitude larger than the vacuum level in the basic calculation.)

When I change KPUSE = 1 2 4 5 then again I get a zero charge in my vacuum region.
Using KPUSE = 1 2 4 gives a non-zero charge.

Anyone any idea what is going on and how this could be solved?

thx,
Danny

Problem solved.

It seems that you have to tell VASP explicitly ISTART=1 otherwise it will generate random wavefunctions, even if this is the only goal of the job.

Maybe it might be interesting if contradictory input from the incar file was also reported on output(the same way you now are given hints to use eg an other option for LREAL)

Danny

how should we set the value of EINT?

thanks for the reply

as far as i do the calculations for charge density, i use the E int value from the DOS plot.

the regioun or the peaks u want to find the charge density is what u have to provide as the range of E int

EINT = -1.25 -1.00 something like this....


hih