read data of nano fluid inside nanochannel

LAMMPS LAMMPS Mailing List Mirror
1

hi everyone
i simulate water-copper nanofluid and use the data file for initial geometry of molecules and atoms that visualizing by vmd…
i attached picture of it in # 1.
i don’t know where is my mistake beacause when i run my code with lammps the initial instruction fail and i visualize it too and attached in picture # 2.
can everyone give me some suggestions?
i modeled spc/e water molecules and use:

dimension 3
boundary p p p
atom_style full
read_data flow4s_autopsf.data
special_bonds charmm
neigh_modify delay 2 every 1

group hydrogen type 1
group oxygen type 2
group low type 3 units lattice
group up type 4 units lattice
group nanoparticle type 5 units lattice
set group hydrogen type 1
set group oxygen type 2
set group low type 3
set group up type 4
set group nanoparticle type 5
group wall union low up
group flow union hydrogen oxygen nanoparticle

pair_style lj/charmm/coul/long 2.5 3.2
bond_style harmonic
angle_style charmm
bond_coeff 1 529.581 1.00
angle_coeff 1 300.0 107.0 50.0 3.0
pair_coeff * * 1.0 1.0
pair_modify mix arithmetic
kspace_style pppm 1e-4

initial velocities

compute mamad flow temp/profile 1.1 0 0 y 20
velocity all create 0.0 12345678 dist uniform
fix 1 all nve
#fix 2 all shake 1e-6 500 0 m 1.0 a 1

fix pressup up aveforce 0.0 0.0 0.0
fix pressdown low aveforce 0.0 0.0 0.0

Couette flow

velocity up set 3.0 0.0 0.0 units box
velocity low set 0.0 0.0 0.0 units box
fix 3 wall setforce 0.0 0.0 0.0
fix 4 flow addforce 0.0 0.0 0.0
fix 5 wall spring/self 57.1

Run

timestep 0.0001
thermo 5000
thermo_style multi
thermo_modify temp mamad
minimize 1.0e-6 1.0e-6 1000 10000
dump 2 all xyz 1 dump.nanofsmooth.XYZ
run 10000
excuse me every one for Previous Incomplete email…

1.png

2.jpg

2

The particles are passing through each other. That should tell you
that there is a problem with your pair forces. Looking at your input
script, I see:

pair_style lj/charmm/coul/long 2.5 3.2
pair_coeff * * 1.0 1.0

These do not look like reasonable settings for water. The VDW radius
and epsilon parameter should depend on the atom type. For water,
these parameters are here:

http://lammps.sandia.gov/doc/Section_howto.html#howto_9
http://lammps.sandia.gov/doc/Section_howto.html#howto_7
http://www.moltemplate.org/examples/nanotube+water/spce.lt

Your timestep and units generally seem suspicious also.

If it helps, there is an example of how to set up water between two plates here:
http://www.moltemplate.org/
scroll-down to the "Carbon-Nanotube Capillary (all-atom, explicit water)"
This example has force-field parameters for water, timesteps,
fix-shake settings, etc. It also uses moltemplate, so many of these
settings are contained within write() blocks in an .LT file. However
the syntax of these commands is the same as they are in LAMMPS, so
hopefully it gives you an idea. There is another example for this
system ("capillary filling of a CNT with water ") here:
http://lammps.sandia.gov/scripts.html

----------- small suggestion -----------

Also, my own personal preference is to avoid using fix setforce, to
immobilize the wall atoms. Instead create a new group (example
"gpMobile", or "flow" in your case), and only integrate the equations
of motion for this group. Eg:

   use
fix 1 flow nve
   instead of
fix 1 all nve