couette flow simulation with 2 type of atoms

Dear all,
I want to apply different pair coefficients in my simulation and take different velocity profiles by changing pair coeffs.

I have 3 regions but 2 type of atoms
(1) fluid atoms in the flow (fluid) region i.e:
region 2 block INF INF 3 18 INF INF units box

(2) solid wall atoms in the upper and lower wall regions i.e:
region 1 block INF INF 0 3 INF INF units box
region 3 block INF INF 18 21 INF INF units box

and I have 3 type of interactions between atoms i.e:

between fluid atoms
between fluid/wall atoms
between wall atoms

now my problem is that when I change
"pair_coeff 1 2 4.0 1.0 3" (in the input file below)
to
"pair_coeff 1 2 0.4 1.0 3"
the Velocity profile shows no differency.
that means changing interaction coeffs doesnot effect the simulation results or my simulation is wrong for this purpose??

here is the input file of my simulation.

couette flow simulation

units lj
dimension 2
boundary p s p
atom_style atomic

create geometry

lattice hex 0.81
region mybox block 0 30 0 21 -0.25 0.25
create_box 2 mybox
region 1 block INF INF 0 3 INF INF units box
region 2 block INF INF 3 18 INF INF units box
region 3 block INF INF 18 21 INF INF units box
create_atoms 1 region 1
create_atoms 1 region 3
create_atoms 2 region 2
mass 1 1.0
mass 2 1.0

LJ potentials

pair_style lj/cut 3
pair_coeff 1 1 0.0 0.0 0
pair_coeff 1 2 4.0 1.0 3
pair_coeff 2 2 1.0 1.0 3

define groups

group lower region 1
group upper region 3
group boundary union lower upper
group flow region 2
set group lower type 1
set group upper type 1
neighbor 0.3 bin
neigh_modify delay 5

initial velocities

compute mobile flow temp/profile 1 0 0 y 20
velocity flow create 1.0 482748 temp mobile
fix 1 all nve
fix 2 flow temp/rescale 200 1.0 1.0 0.02 1.0
fix_modify 2 temp mobile

Couette flow

velocity lower set -1.0 0.0 0.0 units box
velocity upper set 1.0 0.0 0.0 units box
fix 3 boundary setforce 0.0 0.0 0.0
fix 4 all enforce2d

Run

timestep 0.002
thermo 500
thermo_modify temp mobile
dump 1 all xyz 50 couette.xyz
fix vel_profile all ave/spatial 2 5000 10000 y center 0.05 vx units reduced file vel.profile
run 100000

Dear all,
I want to apply different pair coefficients in my simulation and take
different velocity profiles by changing pair coeffs.

I have 3 regions but 2 type of atoms
(1) fluid atoms in the flow (fluid) region i.e:
region 2 block INF INF 3 18 INF INF units box

(2) solid wall atoms in the upper and lower wall regions i.e:
region 1 block INF INF 0 3 INF INF units box
region 3 block INF INF 18 21 INF INF units box

and I have 3 type of interactions between atoms i.e:

between fluid atoms
between fluid/wall atoms
between wall atoms

now my problem is that when I change
"pair_coeff 1 2 4.0 1.0 3" (in the input file below)
to
"pair_coeff 1 2 0.4 1.0 3"
the Velocity profile shows no differency.
that means changing interaction coeffs doesnot effect the simulation
results or my simulation is wrong for this purpose??

have you thought about what would make
the velocity profile different?
are these conditions met?

have you visualized the simulation trajectory?

have you tried to increase sampling, e.g. by
running for more steps to reduce statistical noise?

axel.

Dear elena:
I guess you want to study the effect of wall-fluid interaction strength on the velocity profile in couette flow. Generally, the effect of wall-fluid interation strength is obvious at the boundary of fluids. There is many papers which show that the boundary slip on the fluid-solid interface is a function of the fluid-wall interaction strength. However, the interior velocity of fluid maybe insensitive to the fluid-wall interation because the characteristic length (sigma_wall_fluid) in you study is just 1.0. Maybe you can change it from 1.0 to 3.0, and compare the results.
I think you’d better attach the velocity profiles in the next E-mail. In this way, we can see clearly the different results and judge how to study it.

Best,
Jun Zhang

Dear elena:
Please check the attached input inscript and the corresponding results. I have obtained very encouraging results by using this input inscript. The velocity profiles are obviously different for different fluid-wall interaction strength. The larger interation strength causes “negative” velocity slip at the boundary. This phenomenon has been reported by many papers.
You’d better compare the attached input inscript and yours, and find out problems in yours.

Best,
Jun Zhang

in.flow.couette (1.59 KB)

U-y.jpg