In this tutorial, we will take the example of Aluminum Al. We need the param file and a script to create the cell file with different values of cutoff energy.

Al.param

task     SinglePoint          ! The TASK keyword instructs CASTEP what to do
xc_functional   LDA           ! Which exchange-correlation functional to use.
cut_off_energy  700       !
opt_strategy    speed         ! Choose algorithms for best speed


kpoint-conv.sh

#!/bin/sh

NAME=Al

rm -f ${NAME}.castep
rm -f ${NAME}_KPOINT.dat

for K in  3 4 5 6 7 8 9 10 12 14 16

do

cat > ${NAME}.cell << EOF
%BLOCK LATTICE_ABC
2.86 2.86 2.86
60 60 60
%ENDBLOCK LATTICE_ABC

%BLOCK POSITIONS_ABS
Al     0   0   0
%ENDBLOCK POSITIONS_ABS

kpoint_mp_grid $K $K $K 

symmetry_generate
EOF


echo ${NAME}_${K}_${K}_${K}
mpirun -np 4 castep.mpi ${NAME}

done

#analyse the runs and extract the values we need into a single results file
grep 'MP grid size for SCF calculation is ' ${NAME}.castep | awk '{print $8}' > K
grep 'Final energy' ${NAME}.castep | awk '{print $5}' > E
paste K E >> ${NAME}_KPOINT.dat

rm K E
echo ' '
echo 'finished with results in '${NAME}_KPOINT.dat

Execution 

$ sh kpoint-conv.sh

Al_3_3_3
Al_4_4_4
Al_5_5_5
Al_6_6_6
Al_7_7_7
Al_8_8_8
Al_9_9_9
Al_10_10_10
Al_12_12_12
Al_14_14_14
Al_16_16_16
 
finished with results in Al_KPOINT.dat

finished with results in Al_KPOINT.dat

 

The content of the file Al_KPOINT.dat is :

3	-109.1831853078
4	-109.7184971323
5	-109.7164492004
6	-109.7672338408
7	-109.8214800225
8	-109.7791609360
9	-109.7447828790
10	-109.7778457064
12	-109.7731871378
14	-109.7726274274
16	-109.7748461704

 

Plotting

To plot the results we need to use the following python script:

kpoint-conv.py

import matplotlib.pyplot as plt
from matplotlib import rcParamsDefault
import numpy as np

plt.rcParams["figure.dpi"]=150
plt.rcParams["figure.facecolor"]="white"

x, y = np.loadtxt('Al_KPOINT.dat', delimiter='	', unpack=True)
plt.plot(x, y, "o-", markersize=5, label='Etot vs kpoints')
plt.xlabel('# kpoints')
plt.ylabel('Etot (eV)')
plt.legend(frameon=False)
plt.show()


Before running the script we need to install matplotlib tool for plotting

$ sudo apt-get install python3-matplotlib 
$ python kpoint-conv.py

 

We get the following picture