Hi,
I’m trying to run a lammps simulation: a isotherm of nitrogen (77K - pressure 0-1 bar) in an empty box.
In lammps mail list, I didn’t find a solution to prepare an input file.
How I could fix it?
I attach the file for the nitrogen probe
Regards.
Alberto
N2_probe_pcff.data (733 Bytes)
Hi,
I'm trying to run a lammps simulation: a isotherm of nitrogen (77K -
pressure 0-1 bar) in an empty box.
you'll make your life massively easier, if you start with a box that
has already one molecule in it.
In lammps mail list, I didn't find a solution to prepare an input file.
How I could fix it?
axel.
Dear Axel,
thanks a lot you answer.
I’m trying with this input, because I want try the keyword ‘mol’ but I obtain the error
’ Unknown section in molecule file’
Al
units real
atom_style full
dimension 3
boundary p p p
newton on on
pair_style lj/class2/coul/long 33.0
pair_modify shift yes mix sixthpower
kspace_style ewald 1.0e-8
bond_style class2
angle_style class2
dihedral_style class2
improper_style class2
special_bonds lj 0 0 1 coul 0 0 1
#gcmc setting
molecule 1 N2_probe_pcff.data
fix 2 all gcmc 10 1000 1000 1 29494 77.0 -0.5 0.01 mol N2_probe_pcff
write_data isotherm_N2_77K.data
Dear Axel,
thanks a lot you answer.
I'm trying with this input, because I want try the keyword 'mol' but I
obtain the error' Unknown section in molecule file'
because you are feeding LAMMPS a data file for a molecule file. that
will not work. read the manual!
axel.
I receive again always the same error message
A.
I receive again always the same error message
the documentation for the molecule command explains the required file
format IN DETAIL. follow it carefully.
if you get the error 'Unknown section in molecule file', then your
file does not have the correct format or syntax.
it is as simple as that.
axel.
...and one more recommendation. i already mentioned that you should
not start with an empty system, but a system already containing one
molecule.
try to get that to work first (w/o gcmc and the molecule file). there
are some obvious (and thus educational) barriers to overcome there as
well.
i assume you have studied the literature on suitable potentials that
can reproduce the properties you are after with sufficient accuracy?
axel.
with a system including a molecule of N2
I obtain an error
‘ERROR: Fix gcmc cannot exchange individual atoms belonging to a molecule’
and this seems a code bug reading any discussions on mailing list
A.
with a system including a molecule of N2
I obtain an error'ERROR: Fix gcmc cannot exchange individual atoms belonging to a molecule'
and this seems a code bug reading any discussions on mailing list
first of all, you didn't follow my recommendation. i do not have time
to help people that do not care to pay attention to what i suggest.
second, if you are using the latest patch level of LAMMPS, all known
bugs/problems outside of those listed at the following URL are
addressed: Issues · lammps/lammps · GitHub
axel.
I’m sorry if I’m not so skilled, but I’m not e computational chemist.
I’m a experimental chemist.
For these reason, I ask an input file, as example, so I can study the syntax code.
regards
A.
I'm sorry if I'm not so skilled, but I'm not e computational chemist.
I'm a experimental chemist.
how is that an excuse for not doing your due diligence and learning
the tools that you want to use?
what would you say, if a theoretical chemist with no wet lab training
at all came to you asking to use your equipment for an ambitious,
complex synthesis??
For these reason, I ask an input file, as example, so I can study the syntax
code.
as already mentioned, the documentation is quite detailed and there
are examples bundled with LAMMPS. however, what you are trying to do
is using a quite advanced technique and thus by no means suitable for
an unskilled, untrained beginner. MD simulations follow the GI-GO
principle (garbage in, garbage out) and nobody has yet found a way to
implement a G-t-G mode (i.e. garbage to gold).
in order to judge whether you are using suitable settings and
parameters, and are obtaining accurate results, you *will* have to
study a significant chunk of statistical thermodynamics and learn
basic and some advance MD simulation techniques. otherwise, you will
not be able to tell whether your results are garbage or gold. ... and
trust me, the chances to get things wrong are much larger than to get
it right by chance, input example or not.
my recommendation is to find a suitably trained collaborator to work
on solving your problem with you.
axel.
thank you a lot for your patient
Al