Hi all,
I am trying to relax the Carbon-nanotube for energy minimization (geometrical optimization) using Tersoff potential in LAMMPS. I have few questions regarding this approach:
1) Since, I am using the tersoff potential, so I do not need to define the bond-type. But suppose, my CNT is distorted such that there are some bonded and some non-bonded carbon atoms in my structure, then how should I modify the script to take into account this factor?
2) For defining the pair_coefficient, the tutorial says that we must write all the atoms in the pair_coefficient command. For example: pair_coefficient * * Si.tersoff C C C C, if my structure has 4 atoms. But when my structure has, lets say, 400 atoms, then how do we write the pair_coefficient command? I'm assuming that we do not need to write C 400 times.
3) Lastly, when I run the script attached with this email, I get the error: ERROR: Incorrect args for pair coefficients. Please suggest what should I modify in the script to get past this error.
Thank you.
alpha21
in.gcg3 (999 Bytes)
data.gcg4 (16.8 KB)
Hi all,
I am trying to relax the Carbon-nanotube for energy minimization
(geometrical optimization) using Tersoff potential in LAMMPS. I have
few questions regarding this approach:
1) Since, I am using the tersoff potential, so I do not need to define
the bond-type. But suppose, my CNT is distorted such that there are
some bonded and some non-bonded carbon atoms in my structure, then how
should I modify the script to take into account this factor?
please read the publication describing the tersoff potential
and its uses. this should clarify why this kind of many-body
potential can handle different "bond orders".
2) For defining the pair_coefficient, the tutorial says that we must
write all the atoms in the pair_coefficient command. For example:
it says atom _types_.
if your system is all carbon nanotube there should
be only one atom type.
pair_coefficient * * Si.tersoff C C C C, if my structure has 4 atoms.
But when my structure has, lets say, 400 atoms, then how do we write
the pair_coefficient command? I'm assuming that we do not need to
write C 400 times.
only if you insist on assigning a new atom type for each atom,
which would probably also have a bad impact on performance.
there is no need for it.
3) Lastly, when I run the script attached with this email, I get the
error: ERROR: Incorrect args for pair coefficients. Please suggest
what should I modify in the script to get past this error.
before i am looking at the script. please consider the
aforementioned suggestions and see if you can fix the
problem by yourself. it should be no rocket science.
cheers,
axel.