regarding cross-linking in polymer

Hi all,
I’m using lammps to simulate cross-linked polymer by using pcff forcefield. I’m trying to create a situation of shock compression by utilizing msst methodology implemented under lammps framework. However, as my intension is calculate degree of cross-linking for a certain cutoff distance, so can it be possible within lammps to compute how many atoms are in a cross linked environment ?

Is it possible to calculate change of bond energy due to cross-linking by using “ebond” ? Should this “ebond” encompass those contribution of crosslinked bond energy with the predefined class 2 bond ?

It will be a great help if somebody having such experience, share and make suggestion.

Thanks in advance for your attention!

Best,
Wakof

Hi all,
I'm using lammps to simulate cross-linked polymer by using pcff forcefield.
I'm trying to create a situation of shock compression by utilizing msst
methodology implemented under lammps framework. However, as my intension is
calculate degree of cross-linking for a certain cutoff distance, so can it
be possible within lammps to compute how many atoms are in a cross linked
environment ?

Just a quick response for now.
This is a vague question.
Can you clarify (perhaps with pictures) exactly what atoms you are
trying to count?

Generally, LAMMPS users are responsible for writing short programs or
scripts which can answer their questions. With the help of pizza.py
(or dump2data.py -raw), you can extract the coordinates from a DUMP
file (trajectory file) and calculate whatever quantities you are
interested in. It is up to you.

Is it possible to calculate change of bond energy due to cross-linking by
using "ebond" ? Should this "ebond" encompass those contribution of
crosslinked bond energy with the predefined class 2 bond ?

By default "ebond" will include all contributions to the energy from
any pair of atoms which are bonded together.
http://lammps.sandia.gov/threads/msg48542.html

If you use "bond_style class2", then it will be

ebond = sum_over_ij K2*(rij-R0)*2 + K3*(rij-R0)^3 + K4*(rij-R0)^4

summed over all pairs of bonded atoms i,j
(where K2,K3,K4, and R0 may depend on ij)
http://lammps.sandia.gov/doc/bond_class2.html

Are you trying to calculate the energy of a single bond, or all of the
bonds. You can limit the pairs of atoms in the summation by defining
a group containing only some of the atoms, and passing it to the
compute.
http://lammps.sandia.gov/doc/compute.html
http://lammps.sandia.gov/doc/group.html
Also see:
http://lammps.sandia.gov/doc/compute_property_local.html

and more generally:
http://lammps.sandia.gov/doc/Section_howto.html#howto_15
http://lammps.sandia.gov/doc/thermo.html
http://lammps.sandia.gov/doc/thermo_style.html
http://lammps.sandia.gov/doc/compute_modify.html
http://lammps.sandia.gov/doc/dump.html
http://lammps.sandia.gov/doc/region.html

Andrew

This is probably irrelevant, however after running your simulation,
you can retroactively calculate the contribution to the energy due to
a particular type of bond by deleting it from the data file, and then
and use the "rerun" command to read the old dump file, calculate the
energy with the bond deleted, and compare with the old energy (with
the bond present).
http://lammps.sandia.gov/doc/rerun.html

It will be a great help if somebody having such experience, share and make
suggestion.

I'm not sure if my reply helped.

Nothing in your post indicates you are creating/destroying
bonds with your model. So all the “cross-linking” is set initially
when you create the bond topology for your system. In which
case you can pre/post compute it yourself, outside of LAMMPS.

Steve

1 Like

Yes, I have understand the point what Steve has raised, now I’m trying to use fix bond/create/break for a reactive site within a predefined threshold cutoff. But at the end I got one error that, ERROR: Cannot yet use fix bond/break with this improper style (…/fix_bond_break.cpp:165).

I’m using PCFF+

atom_style full

pair_style lj/class2/coul/long 9.5
kspace_style pppm 0.00001
pair_modify tail yes
bond_style class2
angle_style class2
dihedral_style class2
improper_style class2
special_bonds lj/coul 0.0 0.0 1.0

read_data structure.dat

pair_coeff 1 1 0.054 4.01 # c1 - c1
pair_coeff 2 2 0.054 4.01 # c2 - c2
pair_coeff 3 3 0.054 4.01 # c3 - c3
pair_coeff 4 4 0.2247 3.92 # cl - cl
pair_coeff 5 5 0.02 2.995 # hc - hc

So, I just wanted to know that is there any restriction for using fix bond/create/break with PCFF+(improper style class 2) ?

Lammps version: LAMMPS (30 Oct 2014), MC package is installed.

As there is a lot of literature of crosslinking of polymer using pcff and lammps, so I wonder if really there is any such restriction ?

Yes, I have understand the point what Steve has raised, now I'm trying to
use fix bond/create/break for a reactive site within a predefined threshold
cutoff. But at the end I got one error that, ERROR: Cannot yet use fix
bond/break with this improper style (../fix_bond_break.cpp:165).

I'm using PCFF+

atom_style full

pair_style lj/class2/coul/long 9.5
kspace_style pppm 0.00001
pair_modify tail yes
bond_style class2
angle_style class2
dihedral_style class2
improper_style class2
special_bonds lj/coul 0.0 0.0 1.0

read_data structure.dat

pair_coeff 1 1 0.054 4.01 # c1 - c1
pair_coeff 2 2 0.054 4.01 # c2 - c2
pair_coeff 3 3 0.054 4.01 # c3 - c3
pair_coeff 4 4 0.2247 3.92 # cl - cl
pair_coeff 5 5 0.02 2.995 # hc - hc

So, I just wanted to know that is there any restriction for using fix
bond/create/break with PCFF+(improper style class 2) ?

yes, there is. class2 topologies are different from all other
topologies and require multiple multi-particle entries to be created
within angles, dihedrals and impropers. none of the bond
creation/breaking fixes knows how to do this, because it is
non-trivial and requires additional knowledge of the force field
parameterization.

Lammps version: LAMMPS (30 Oct 2014), MC package is installed.

As there is a lot of literature of crosslinking of polymer using pcff and
lammps, so I wonder if really there is any such restriction ?

there is. those simulations were very likely done with fixed
topologies that were crosslinked offline in the way steve indicated.

axel.