Hello all,
I am using ReaxFF potential to equilibrate epon 862 crosslinked with detda. During equilibration, it is breaking into several pieces. I checked the total energy curve, which seems to be well converged.
What could be the possible reason behind disintegration, how can I mend it?
I used following code:
REAX potential for CHON system
…
units real
atom_style charge
boundary p p p
read_data data.EPON
pair_style reax/c lmp_control lgvdw yes
pair_coeff * * ffield.reax.EPON H C N O
neighbor 2 bin
neigh_modify every 10 delay 0 check no
fix 1 all nve
fix 2 all qeq/reax 1 0.0 10.0 1e-6 param.qeq
fix 3 all temp/berendsen 500.0 500.0 100.0
thermo 100
thermo_style custom step temp pe etotal press pxx pyy pzz
thermo_modify lost ignore flush yes
timestep 0.25
dump 1 all custom 50 reaxEPON.xyz id type x y z
run 15000
Thanks!
Hello all,
I am using ReaxFF potential to equilibrate epon 862 crosslinked with
detda. During equilibration, it is breaking into several pieces. I checked
the total energy curve, which seems to be well converged.
what "total energy curve" are you talking about here and how did you check
its "convergence"?
i have not heard these terms used in this form in the context of molecular
dynamics simulations.
What could be the possible reason behind disintegration, how can I mend it?
hard to say with the limited information you provide. most likely is an
initial geometry with high (local) potential energy, i.e. unrelaxed
positions or bad choice of box dimensions causing close contacts etc. some
of that can often be cured by running a minimization before starting MD,
but others need a careful review of the initial geometry and perhaps
adjustments to the process of building it.
axel.
Hello Axel,
Thanks for your insight.
what “total energy curve” are you talking about here and how did you check its “convergence”?
i have not heard these terms used in this form in the context of molecular dynamics simulations.
I am talking about curve that I draw (Total energy vs No of Time Steps) in excel from the data that I got from MD in Log.lammps file.
hard to say with the limited information you provide. most likely is an initial geometry with high (local) potential energy, i.e. unrelaxed positions or bad choice of box dimensions causing close contacts etc. some of that can often be cured by running a minimization before starting MD, but others need a careful review of the initial geometry and perhaps adjustments to the process of building it.
Can I know how to perform those minimization steps?
Thanks!,
Rajni Chahal
Hello Axel,
Thanks for your insight.
*what "total energy curve" are you talking about here and how did you
check its "convergence"?*
*i have not heard these terms used in this form in the context of
molecular dynamics simulations.*
I am talking about curve that I draw (Total energy vs No of Time Steps) in
excel from the data that I got from MD in Log.lammps file.
to study the progress of equilibration, it is usually better to monitor
the potential energy, but that shouldn't make too much of a difference in
your case. but i would talk about "convergence", but rather whether the
potential energy would "plateau". however, the energy itself is rarely the
only property to look at.
what is really worrying is your use of:
thermo_modify lost ignore
that will make LAMMPS silently ignore very problematic conditions in your
simulation.
hard to say with the limited information you provide. most likely is an
initial geometry with high (local) potential energy, i.e. unrelaxed
positions or bad choice of box dimensions causing close contacts etc. some
of that can often be cured by running a minimization before starting MD,
but others need a careful review of the initial geometry and perhaps
adjustments to the process of building it.
Can I know how to perform those minimization steps?
study MD text books, study the LAMMPS documentation, previous discussions
on this mailing list. there is a gazillion places with useful information,
as this is not LAMMPS specific but a question about general MD simulation
workflows.
axel.