Page 1 of 1

Atomic relaxations in cartesian

Posted: Thu Aug 26, 2010 7:04 pm
by sergio
Hello everyone,

I have a question about the atomic relaxation in VASP. I want to relax atoms along one cartesian direction (say the z-direction). I choose selective dynamics and add the tag F F T next to the atomic coordinates. However, the lattice vectors all have components in two or three different directions and since VASP relaxes along these lattice vectors, it will not relax only along the z direction. Is there a way I can relax just along the z direction while still keeping the same lattice vectors?

Atomic relaxations in cartesian

Posted: Fri Aug 27, 2010 2:53 am
by metosa
you need to write some bash scripts. for example suppose that your POSCAR is like this:

poscar
1.00
10 0 0
0 11 0
0 0 12
C H
1 1
0 0 0
0.5 0.5 0

some simple bash lines such as:

for i in `seq -w 0.990 0.002 1.010`
do
mkdir $i
cp INCAR KPOINTS POTCAR $i
cd $i
cat > POSCAR <<EOF
poscar
1.00
10 0 0
0 11 0
0 0 `echo 12*$i | bc -l`
C H
1 1
0 0 0
0.5 0.5 0
EOF

$vasp

cd ..

done


will relax your lattice in z-direction only...

Atomic relaxations in cartesian

Posted: Fri Aug 27, 2010 2:53 am
by metosa
you need to write some bash scripts. for example suppose that your POSCAR is like this:

poscar
1.00
10 0 0
0 11 0
0 0 12
C H
1 1
0 0 0
0.5 0.5 0

some simple bash lines such as:

for i in `seq -w 0.990 0.002 1.010`
do
mkdir $i
cp INCAR KPOINTS POTCAR $i
cd $i
cat > POSCAR <<EOF
poscar
1.00
10 0 0
0 11 0
0 0 `echo 12*$i | bc -l`
C H
1 1
0 0 0
0.5 0.5 0
EOF

$vasp

cd ..

done


will relax your lattice in z-direction only...