Calculation of atoms

The following files are needed for the calculation of atoms

Before using a pseudopotential intensively it is not only necessary to check it in different bulk phases but the pseudopotential should also reproduce exactly the eigenvalues and the total energy of the free atom for which it was created. If energy cutoff and cell size are sufficient, the agreement between the atomic reference calculation and a calculation using VASP is normally better than 1 meV. Calculations for an atom are relatively fast and unproblematic in most cases.

For the calculation only the $\Gamma$ point should be used i.e. the KPOINTS file should have the following contents:

Monkhorst Pack
0
Monkhorst Pack
 1  1  1
 0  0  0

The size of the cell depends on the element in question. Some values for reliable results are compiled in Tab. 1.

*Tab. 1: Typical convenient settings for the cellsize for a calculation of atoms.*
	Cellsize
Aluminum	14 $\AA$
Potassium	12 $\AA$
Copper, Rhodium, Palldium ...	10 $\AA$
Nitrogen	7 $\AA$
Carbon	8 $\AA$

A simple cubic cell can be often enough but it is usually recommended to use nearly cubic cells with minimal orthorombic distortion. This can be important for some atoms due to higher degrees of freedom for relaxation. An example POSCAR file should look like:

atom
1
     10.00000    .00000    .00000
       .00000  10.00100    .00000
       .00000    .00000  10.00200
   1
cart
 0    0    0

Due to the large cell NGX, NGY, NGZ and NGXF, NGYF, NGZF need to be increased. This can be most easily achieved by setting PREC=Accurate or PREC=High.

The INCAR file can be very simple:

SYSTEM = Pd: atom
#Electronic minimisation
  Template:TAGBK  = 200.00 eV  #energy cut-off for the calculation
  EDIFF  =   1E-04    #stopping-criterion for ELM
  NELMDL =  5

#Ionic parameters
  POMASS = 106.42     #mass (not really used in this case)
  ZVAL   =  10.00     #valence

#DOS related values
  ISMEAR =    0
  [[{{{1}}}|{{{1}}}]]    #use smearing method

The only difference to the bulk calculation is that Gaussian smearing must be used now. You might set SIGMA to a very small value; this is necessary if atomic orbitals are almost degenerated.

Mind: Look at the right value for the energy. It is not $F=E+ \sigma S$ , which contains a "meaningless" entropy term, but the energy $E$ . If SIGMA is very small both values are the same, but for extremely small SIGMA values VASP might have difficulties to converge to the correct atomic ground state.