SRD solvent in a nanopore

LAMMPS LAMMPS Mailing List Mirror
1

Hi LAMMPS community,

I am trying to use SRD solvent in a nanopore membrane system.

I begin with a very simple system, an impermeable atomistic wall in SRD solvent.

The problem is SRD particles are moving through the wall. I have been through the full mailing list related to SRD.

I played around these …

  1. Increased the number density of the wall (to a very high value)

  2. Increase the diameter of the wall atoms.

  3. Make sure the wall atoms are overlapping and there are no holes. The lattice constant is smaller than the diameter of the wall atoms.

  4. Increase the ATOMPERBIN (fix_srd.cpp) value from 30 to 100 to have a very high dense wall.

With all these, the number of SRD particles crossing the wall is decreasing, but still few particles are crossing the wall.

I have attached my lammps config file, which takes only a few seconds to run on a single processor. Simulation can be visualized with the file - all.xyz.

Any idea is appreciated …

Thank you very much.

Cheers !!!

Sridhar Kumar Kannam :slight_smile:

in.srd (1.37 KB)

2

Have you visualized your system. Can you track a
single particle that escapes and see where it
moves with respect to the wall particles?

Steve

3

Hi Steve,

Yes, I have visualized the system, and thats how I find the srd particles passing through the wall.

In the attached file, I have defined a rectangular simulation box with a wall and srd fluid fluid only one side of the wall.

Simulations takes only a few seconds and can by visualized by using all.xyz.

At frame number around 50, you will see a srd particle passing through the wall.

Note, to run the simulation, increase the ATOMPERBIN (fix_srd.cpp) value from 30 to 100 to have a very high dense wall.

Thank you very much.

4

I don’t know. You have a periodic system - what keeps SRD particles
from going around the box to the other side of the wall?

What is the big-big separation of particles in the wall? You
should turn off the radius 0.88 option in fix srd, as it effectively
shrinks the size of your big particles, which is not useful when
they are supposed to be a wall.

If you are 100% convinced there are no holes in your wall,
then I suppose you will have to identify the SRD particle
that escapes and put some print statements in fix_srd.cpp
to see why it is not interacting correctly with the nearby
big particles.

Steve

5

Hi Steve,

Yes, its perfectly fine for the particles to go around the box. But the problem is few srd particles are passing through the wall.

When I remove the option radius 0.88, it complained about higher number of bigger particles in SRD bin, so I increased the ATOMPERBIN value from 100 to 200.

The separation between wall atoms is 0.22 and the diameter is 2.0. There are more than 100 wall atoms per SRD bin. So I think, there are no holes in the wall.
But still some of the srd particles are passing through the wall.

Could you please look into it.

Thank you very much.

6

Hi Steve,

I played quite a bit with srd, but couldn’t figure out why the srd particles are passing through the wall.
With increasing the wall number density, the srd particles are finding it hard to cross the wall, but even with ATOMPERBIN value set to 500 (very unrealistic) and wall density 200 (lattice fcc 200), once in a while the srd particles are crossing the wall. The attached lammps files can run on a single processor with in a few seconds. I went through the whole lammps list and earlier someone also complained the same issue, which is not fixed yet. Could you please have a look.

Thank you very much.