# How to model an infinite tube which is full of fluid properly?

Dear lammps-users:
A system,an infinite tube which is filled with fluid at 150K and 50atm, makes me confused.I am trying to cool the fluid down to 40K to observe the process of nucleation,but some empty space appears because of the decreased density when the temperature is dropping.One more problem,the model is compressed in the axis of z in the way of “fix npt”. Apparently, this way does not work,so i want to fix a fluid reservoir(not gas reservoir) in the system at the requested condition,but i don’t know how to make it.Can lammps do this?Am i right in the way of thinking? Could anyone give me any suggestions?
Bests,
Su
The following is my input document that i used to run,but the result is incorrectly.

#in.relax for lammps

#relax for the bulk at 150k

units real
atom_style atomic
boundary f f p

lattice fcc 6.33949
region box block 0 16 0 16 0 20
create_box 2 box
create_atoms 1 box

mass 1 39.948
mass 2 39.948

region all block INF INF INF INF INF INF units lattice
region hole cylinder z 8 8 4 INF INF units lattice

group all region all
group hole region hole
group wall subtract all hole

set group wall type 2
set group hole type 1

compute holeT hole temp/region hole
#compute atomP hole stress/atom ke
#compute holeP hole reduce sum c_atomP[3]

velocity hole create 150.00 87287 temp holeT
velocity wall set 0 0 0

timestep 2.156

pair_style lj/cut 8.5125
pair_coeff * * 0.2381 3.405 8.5125
pair_coeff 1 2 0.7483 3.405 8.5125

neighbor 0.681 bin
neigh_modify every 10 delay 0 check yes exclude type 2 2

thermo_style custom step c_holeT pe ke etotal vol lx ly lz press
thermo 1000
#thermo_modify temp holeT

dump 1 wall cfg 1000 atom.*.cfg mass type xs ys zs

fix 1 hole npt temp 150 150 21560 z 50.0 50.0 21560 dilate hole
fix_modify 1 temp holeT
fix 2 wall setforce 0.0 0.0 0.0

run 50000
write_data in.relax

Dear lammps-users:
A system,an infinite tube which is filled with fluid at 150K and
50atm, makes me confused.I am trying to cool the fluid down to 40K to
observe the process of nucleation,but some empty space appears because of
the decreased density when the temperature is dropping.One more problem,the
model is compressed in the axis of z in the way of "fix npt". Apparently,
this way does not work,so i want to fix a fluid reservoir(not gas
reservoir) in the system at the requested condition,but i don't know how to
make it.Can lammps do this?Am i right in the way of thinking? Could anyone
give me any suggestions?

if you have a container material filled with a second​ material, there must
be inhomogeneities or strain whenever the two materials have a different
thermal expansion coefficient. if you are applying pressure, then
differences in compressibility will have an impact as well. now if you want
this to be a closed system, you either have to accept those facts and the
resulting effects on the system, or simulate an open system.

on top of that, nucleation on the atomic scale is a problem in itself.
there is plenty of published literature on that.

this all looks to me like you are trying to transfer concepts from
macroscopic experimentation and modeling to the atomic scale without paying
attention to how the differences in scale also result in differences of the
physics. remember, all this nanomaterials craze is for a reason: at the
nanoscale things are different!

axel.