How to define a polymer box?

Dear all,

I am interested in modeling how a polymer (polycarbonate) gets into a cavity due to a pressure inlet until it freezes due to heat transfer with the walls.

My first aim is to define a 50nmx50nmx20nm box with melted PC. I have been able to define such box with individual atoms, but not with PC units. I guess I should use coarse grained model though. Could someone give me a hint about how to define this?

Thanks for your help.

Dear all,

I am interested in modeling how a polymer (polycarbonate) gets into a cavity
due to a pressure inlet until it freezes due to heat transfer with the
walls.

My first aim is to define a 50nmx50nmx20nm box with melted PC. I have been
able to define such box with individual atoms, but not with PC units. I
guess I should use coarse grained model though. Could someone give me a hint
about how to define this?

study the literature, find papers describing simulations of such a
system. there should be plenty of simulations with a variety of
potentials/models. it is good practice to try to reproduce published
data before starting new simulations. this is how you can practice how
to find good simulation settings for your system and how to set up
inputs. you can set up multiple systems with different potentials and
geometries (also polymerization properties like average chain length
and cross-linking) and compare and test, which would be the most
suitable for your needs.

for running molecular systems, it is usually easiest to start from a
data file. this needs to be generated. there are some tools, but for
polymers, it is often need to write some kind of script or program to
generate the exact system you are interested in. LAMMPS contains a
very simply fortran code in tools/chain.f to build simple bead-spring
polymers.
such automatic generated structures often have overlaps and close
contacts, and thus a strategy like in the micelle example using the
"soft" potential and fix adapt can be used to gently move a random
polymer structure to a more reasonable one.

tools like fix bond/create can be used for cross-linking.

axel.

Dear Dr. Axel Kohlmeyer ,

I have followed your suggestion and I have read some papers, like the one of Feng Jie et al “Molecular dynamics simulation of injection of polyethylene fluid in a variable cross-section nano-channel” from 2011

I am currently solving the data file generation with the polymer chains using Polymermodeler tool of www.nanohub.org, which works great.

However, I am facing now the following problem. From a macro scale simulation carried out using Solidworks Plastics, I have a pressure profile (P with time). I want to use this pressure profile as a boundary condition in my LAMMPS simulation box but, as far as I know, it is not possible to define a pressure inlet boundary condition in molecular dynamics. So, how could I couple my pressure profile extracted from Solidworks with LAMMPS?

Thanks in advance.

Dear Dr. Axel Kohlmeyer ,

I have followed your suggestion and I have read some papers, like the one of
Feng Jie et al “Molecular dynamics simulation of injection of polyethylene
fluid in a variable cross-section nano-channel” from 2011

I am currently solving the data file generation with the polymer chains
using Polymermodeler tool of www.nanohub.org, which works great.

However, I am facing now the following problem. From a macro scale
simulation carried out using Solidworks Plastics, I have a pressure profile
(P with time). I want to use this pressure profile as a boundary condition
in my LAMMPS simulation box but, as far as I know, it is not possible to
define a pressure inlet boundary condition in molecular dynamics. So, how
could I couple my pressure profile extracted from Solidworks with LAMMPS?

this is a completely new question and is unrelated to the subject line.
please post it as a new email to the mailing list with a suitable subject line.

axel.