fcc Ni

Based on the VASP wiki example in this link

Task: For spin-polarized (collinear magnetism) fcc Ni. Do a lattice parameter optimization, calculate DOS and bandstructure.

System-specific instructions

Select instructions for the system you are using:

Instructions for use on the NAISS cluster Tetralith (NSC)

First, copy the example folder which contains some of the VASP input files

cp -r /software/sse/manual/vasp/training/ws2023/fcc_Ni .
cd fcc_Ni

and copy the latest POTCAR file for Ni

cp /software/sse/manual/vasp/POTCARs/PBE/2015-09-21/Ni/POTCAR .

Perform the calculations below in the same way as was done for the cd Si example.

Input files

POSCAR

fcc Ni
 3.53 
 0.5 0.5 0.0
 0.0 0.5 0.5
 0.5 0.0 0.5
Ni
   1
cartesian
0 0 0

INCAR

general:
  SYSTEM = fcc Ni
  ISTART = 0 ; ICHARG=2
  ENCUT  =    270
  ISMEAR =    1  ; SIGMA = 0.2
  LORBIT=11
spin:
  ISPIN=2
  MAGMOM = 1
  • ISTART=0, static calculation (default)

  • ICHARG=2, initial charge-density from overlapping atoms (default if ISTART=0)

  • ENCUT=270, default energy cutoff 270 eV

  • ISMEAR=1, Methfessel-Paxton smearing used for metal

  • SIGMA=0.2, default smearing

  • LORBIT=11, write DOSCAR and lm-decomposed PROCAR

  • ISPIN=2, gives a spin-polarized calculation and MAGMOM=1, an initial magnetic moment of 1 Bohr magnetons

KPOINTS

k-points
 0
Monkhorst Pack
 11 11 11
 0  0  0
  • Equally spaced k-mesh

  • Odd Monkhorst-Pack k-mesh > Gamma centered

1. Volume relaxation

Similar as for the cd Si example, check the total energy over a range of volumes, by using the tailored job script “run-vol.sh”. Submit the job (Tetralith)

sbatch run-vol.sh

or run it interactively (MeluxIna)

./run-vol.sh

after it finishes, check the total energy vs lattice constant e.g. using gnuplot

gnuplot 

and at the prompt type (the first two lines for creating an image, needed at MeluXina)

set term png
set output "SUMMARY.dia.png" 
plot "SUMMARY.dia" using ($1):($4) w lp

is the equilibrium lattice parameter close to a = 3.5 Å?

In the same way as was done in fcc Si and cd Si, investigate the output from Birch-Murnaghan EOS by using the “eqos.py” script.

  • What is the equilibrium lattice parameter in this case?

2. DOS

Create a new folder “dos”, copy the relevant files and go there

mkdir dos
cp INCAR POSCAR KPOINTS POTCAR run.sh dos
cd dos

As in the previous example for cd Si we compute DOS in a single step. Edit INCAR such that it looks like

general:
  SYSTEM = fcc Ni
  ISTART = 0 ; ICHARG=2
  ENCUT  =    270
  ISMEAR =   -5
  LORBIT=11
spin:
  ISPIN=2
  MAGMOM = 1

notice the change to ISMEAR=-5, the tetrahedron method with Blöchl corrections, suitable for DOS and total energies.

Finally, submit the job (Tetralith)

sbatch run.sh

or run it interactively (MeluXina)

srun --hint=nomultithread -n 8 vasp_std

To check the result after finish, use p4vasp or py4vasp as in the previous examples.

  • Does the DOS show the spin split?

3. Bandstructure

In the main “fcc_Ni” folder, create a new folder “band”, copy the relevant files and go there, note the treatment for the “KPOINTS” file

mkdir band
cp INCAR POSCAR POTCAR run.sh band
cp KPOINTS.band band/KPOINTS
cd band

the file KPOINTS.band, which was copied to KPOINTS, looks like

kpoints for bandstructure L-G-X-U K-G
 10
line
reciprocal
  0.50000  0.50000  0.50000    1
  0.00000  0.00000  0.00000    1

  0.00000  0.00000  0.00000    1
  0.00000  0.50000  0.50000    1

  0.00000  0.50000  0.50000    1
  0.25000  0.62500  0.62500    1

  0.37500  0.7500   0.37500    1
  0.00000  0.00000  0.00000    1

As in previous bandstructure examples, e.g. cd Si, we need a suitable CHGCAR file. First check that INCAR is correct, edit the file using e.g. vi such that

ICHARG=11

i.e. that CHGCAR is read. Copy CHGCAR from the previous DOS calculation

cp ../dos/CHGCAR .

thereafter submit the job (Tetralith)

sbatch run.sh

or run it interactively (MeluXina)

srun --hint=nomultithread -n 8 vasp_std

To check the results after finish, use p4vasp or py4vasp. Compare with the result on the VASP wiki.