problems on single Cu and H atom calculation

[fegg7502]

hallo vaspers:

I have two problems need your help, thank you very much for your patience and help in advance.

first, I clculated neutral single Cu and H atom in a 15 A box, with the INCAR below,

SYSTEM = single atom
ENCUT = 350
ISTART = 0 ; ICHARG = 2
ISMEAR = 0
ISPIN = 2
LREAL = Auto
dynamic:
INIWAV = 1
ALGO = Fast
IBRION = 2
NSW = 400
POTIM = 0.3
ISIF = 2
GGA = 91
VOSKOWN = 1
PREC = Accurate
EDIFF = 1.E-04
EDIFFG = -0.05

the results show that energies of Cu and H atoms are -0.253 and -1.117 eV respectively.,,
Could anyone tell me H atom is more stable than Cu?


second, I also tried to calculate H+ , precisely a proton.
with the INCAR as below:

SYSTEM = H
ENCUT = 350
ISTART = 0 ; ICHARG = 2
ISMEAR = 0
ISPIN = 2
LREAL = Auto
dynamic:
INIWAV = 1
ALGO = Fast
IBRION = 2
NSW = 400
POTIM = 0.3
ISIF = 2
GGA = 91
VOSKOWN = 1
PREC = Accurate
NELECT = 0
EDIFF = 1.E-04
EDIFFG = -0.05
IDIPOL = 4
DIPOL = 0.5 0.5 0.5
LDIPOL=.TRUE.

the problem is that the calculation is still running as for neutral H atom.


thank you very much!

[forsdan]

Regarding the first question:

The energies given in VASP are with respect to the non-polarized atom(s). So when you perform spin-polarized calculations for an atom you get the energy difference with respect to the non-polarized reference atom.
Therefore, the energies you have obtained only reflect the effect of including spin-polarization for each respective atom.

The 2nd question I leave to someone else since I haven't done that kind of calculations myself. But I guess it has something to do with that you'll get a background charge compensating the lack of the electron, since the supercell itself must be neutral.

Cheers,
/Dan
<span class='smallblacktext'>[ Edited Mon Aug 03 2009, 10:45PM ]</span>

[fegg7502]

help...

[alex]

Adding/clarifying to forsdan:
The total energy of the atoms comes alog with the POTCAR file. So what you see is, what forsdan answers, the difference between your actual atom and th eatom in the POTCAR you've used.
Btw. for atoms it could be very useful to leave cubic boxes behind. You'll destroy some of the symmetry and end up with the right electronic configuration. Check e.g. your DOS for that.

2nd: The total energy of a system without an electron is exactly zero (0.) ... Include the energies of the reference atoms (meaning POTCAR) to that. Try with H3O+ and water ...

Cheers

Alex

[fegg7502]

THANK YOU VERY MUCH!