cell volume

[chemical]

Hi everybody,

I’m a new vasp user and I’m trying to optimize my lattice parameters. I'm working hexagonal structure (a=b #c and ß=120). I’wd like to use the cell volume(203.098) in my POSCAR file instead a,b andc. But i found that the KPOINTS in my OUTCAR is very large 384. other thing, in my COUNTCAR i found that my cell volume decruse untel 0.95.

POSCAR:
System=surface1
-203.098
2.48 0.0 0.0
0.0 2.48 0.0
0.0 0.0 5.86
1 2
DIRECT
0.0 0.0 0.0
0.66 0.33 0.33
0.33 0.66 0.66

can some one please show me where I’m wrong and wath is formula that you used to calculate cell volume. Thanks par advance






regards

[forsdan]

The number of k-points will depend on your symmetry and your settings.

Do you get an initial volume of 203.098 (check the OUTCAR file)? If the volume changes, you must have allowed for it in your INCAR by the use of ISIF. I find it hard to believe that you after relaxation get 0.95 Ã…^3 in total volume for three atoms. That would be very unlikely. Are you sure you are reading the CONTCAR correctly?

Can you give us your INCAR, CONTCAR and KPOINTS so we can give you a more detailed answer.

Best regards,
/Dan Fors

<span class='smallblacktext'>[ Edited Tue Jul 15 2008, 04:14PM ]</span>

[chemical]

thank you for your reply, that's my files:

INCAR
system=MgCl2 hp
PREC=ACCURATE
GGA=91
NUBDOWN=2
ENMAX=380

NSW=150
IBRION=2

CONTCAR:
system=MgCl2 hp
0.95303
2.48000001 0.000000 0.0000000000
0.00000000 2.480001 0.0000000000
0.00000000 0.000000 5.8599999999
1 2
DIRECT
-0.000013656 0.00003376 0.0024465
0.660650076 0.32940439 0.3077803
0.329531567 0.66025795 0.6797730


KPOINTS:
11x1X11
0
MONKHORST-PACK
11 11 11
0 0 0

[forsdan]

I copied your input-files and executed them,

1. Regarding the high number of k-points. According to the symmetry analysis in the OUTCAR file, there are very few space point operations so there is next to nothing that reduce the number of k-points you specify. First check that the geometry is correct, compare with experiment or other published work. Otherwise you have to reduce your mesh to get fewer k-points.

I changed the k-points to "5 5 5" (I didn't check if this was converged however) and ran the calculation.

2. The volume in the CONTCAR file you gave is 0.95303^3 * 2.48^2 * 5.86 = 31.20 Ã…^3. I don't see how you can get this. In my run the initial volume is 203.098 and it doesn't change. Which vasp version do you use?

NOTE: There is a typo in your INCAR, "NUBDOWN" should read "NUPDOWN". Also the cutoff energy is specified with ENCUT.

Best regards,
/Dan Fors

[chemical]

I'm using vasp version 4.6. I'm going to try again. thank you so much

regards