cubic diamond Si

Based on the VASP wiki example in this link

Task: Do a volume relaxation as well as calculate the DOS and bandstructure for cubic diamond (cd) structure Si.

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

cp -r /software/sse/manual/vasp/training/ws2020/cd_Si .
cd cd_Si

and copy the latest POTCAR file for Si

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

Input files

POSCAR

cubic diamond Si
   5.5
 0.0    0.5     0.5
 0.5    0.0     0.5
 0.5    0.5     0.0
Si
  2
Direct
 -0.125 -0.125 -0.125
  0.125  0.125  0.125

• cubic diamond Si with lattice constant 5.5 Å
• fcc cell
• 2 atoms in the cell

INCAR

System = diamond Si

ISTART = 0 
ICHARG=2
ENCUT  =    240
ISMEAR = 0 
SIGMA = 0.1

KPOINTS

k-points
 0
Monkhorst Pack
 11 11 11
 0  0  0

1. Volume relaxation

Here, investigate the total energy as a function of volume in a similar way as for the fcc Si example. There’s a job script which can be used “run-vol.sh”

#!/bin/bash
#SBATCH -A snic2020-13-76  
#SBATCH -t 0:30:00
#SBATCH -n 4 
#SBATCH -J vaspjob

module load VASP/6.1.0.28012020-nsc1-intel-2018a-eb

for i in 5.1 5.2 5.3 5.4 5.5 5.6 5.7 ; do
mkdir -p $i
cd $i
cp /software/sse/manual/vasp/POTCARs/PBE/2015-09-21/Si/POTCAR .
cp /software/sse/manual/vasp/training/ws2020/cd_Si/INCAR .
cp /software/sse/manual/vasp/training/ws2020/cd_Si/KPOINTS .
cat >POSCAR <<!
cubic diamond Si
   $i 
 0.0    0.5     0.5
 0.5    0.0     0.5
 0.5    0.5     0.0
Si
  2
Direct
 -0.125 -0.125 -0.125
  0.125  0.125  0.125
!
mpprun vasp_std
E=`awk '/F=/ {print $0}' OSZICAR` ; echo $i $E  >>../SUMMARY.dia
cd ..
done

submit it with

sbatch run-vol.sh

after some short time, the folders 5.1 - 5.7 with the output should appear, the “slurm-JOBID.out” and a file “SUMMARY.dia”:

5.1 1 F= -.10233838E+02 E0= -.10233747E+02 d E =-.180914E-03
5.2 1 F= -.10528200E+02 E0= -.10528187E+02 d E =-.271766E-04
5.3 1 F= -.10713336E+02 E0= -.10713334E+02 d E =-.215112E-05
5.4 1 F= -.10806746E+02 E0= -.10806746E+02 d E =-.111754E-06
5.5 1 F= -.10823085E+02 E0= -.10823085E+02 d E =-.431458E-08
5.6 1 F= -.10775151E+02 E0= -.10775151E+02 d E =-.205580E-09
5.7 1 F= -.10673630E+02 E0= -.10673630E+02 d E =-.114038E-10

check it e.g. using gnuplot

gnuplot

and at the prompt type

plot "SUMMARY.dia" using ($1):($4) w lp

From the plot one can see that the equilibrium lattice constant is close to a = 5.5 Å.

As in the previous example compare with an equation of state method, first extract the volume and total energies by running

./get-vol-etot.sh

thereafter, use the ASE script “eqos.py”

python eqos.py

the output might look like

Equation of state:
-------------------
E0: -10.825495 eV
V0: 40.987525 A^3
B: 87.612808 GPa

using the Birch-Murnaghan EOS. Similar as before, we quickly compute the lattice constant a using python

python

and input at the “»>” prompt

(4*40.987525) ** (1./3.)

• If a cubic diamond unit cell contains 8 atoms, why multiply by 4 when computing the lattice constant above (compare with fcc Si)? Hint: “grep volume OUTCAR”

2. Density of states (DOS)

Calculate DOS in a new folder “dos”

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

For a single step calculation, edit INCAR e.g. with gedit or your text editor of choice, such that it looks like

System = diamond Si

ISTART = 0 
ICHARG=2
ENCUT  =    240
ISMEAR = -5 
#SIGMA = 0.1
LORBIT = 11

thereafter submit the job

sbatch run.sh

After it finished, check that the output looks fine, e.g.

cat slurm*out    

Now use p4vasp to check DOS, similar as was described for fcc Si DOS

p4v &

3. Bandstructure

Return to the main example folder “cd_Si” and prepare for a bandstructure calculation

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

and edit INCAR such that it looks like

System = diamond Si

ISTART = 0 
ICHARG=11
ENCUT  =    240
ISMEAR = 0 
SIGMA = 0.1
LORBIT = 11

assuring that CHGCAR will be read (ICHARG=11) and same as for DOS, setting LORBIT=11. Submit the job with

sbatch run.sh

and wait for it to finish. Check that the output looks fine and once again use p4vasp, following the previous example of fcc Si bandstructure.

In addition, test to plot the orbital characters of the bands following the examples (see bottom figures) in the VASP wiki.

Extra tasks

• Test with other methods for EOS by editing the “eqos.py” script

• Perform the volume relaxation in 1. with a higher ENCUT in INCAR