# Argon Flow over Carbon Nanotube

Hi,
I am a beginner and would like to simulate uniform liquid argon flow over a carbon nanotube and calculate the drag force. My methodology is similar to that mentioned in the page 26 of this link:
My results for the time-averaged drag force are convergent. However, they are almost 3 times as great and the calculated temperatures of fix temp/rescale are way too big! I couldn’t find the problem. Please help me. I have provided the code scripts below:

# Argon Flow Over A Carbon Nanotube

dimension 3
boundary p p p

units metal
atom_style atomic

region box block -122.24 183.36 -127.35 127.35 -1.433 19.867 side in units box
region nanotube cylinder z 0 0 4.690706 -1.433 19.867 side out units box
region intersect intersect 2 box nanotube side in units box
region control block -122.24 -113.072 -127.35 127.35 -1.433 19.867 side in units box
region flow block -122.24 -113.072 -127.35 127.35 -1.433 19.867 side out units box
region intersect2 intersect 2 intersect flow side in units box

lattice sc 3.4
create_atoms 2 random 33670 123456 intersect

group nanotube type 1
group flow region intersect2
group control region control
group argon type 2
lattice sc 3.4

velocity nanotube set 0 0 0 units box
fix 1 nanotube setforce 0 0 0

pair_style lj/cut 7.65
pair_coeff 1 1 0.014541 3.746
pair_coeff 1 2 0.012262 3.573
pair_coeff 2 2 0.01034 3.4

velocity argon create 95 123456 mom yes rot yes dist uniform units box
velocity argon set 1.58 0 0 sum yes units box
#Should I write 1.58 0 0 or 1.58 NULL NULL to superimpose 1.58 on the x component of velocity without changing its other components?
fix 3 argon nve
fix 2 argon temp/rescale 1 95 95 1.5 1

compute 1 nanotube group/group argon
thermo 50
thermo_style custom step temp ke etotal c_1[1] c_1[2] f_2
timestep 0.00215
#dump 1 all xyz 50 dump.xyz

restart 20000 restart.*
minimize 0 0 1000 1000

run 1000000 every 50 “velocity control create 95 654321 mom yes rot yes dist uniform sum no units box” &
“velocity control set 1.58 0 0 sum yes units box”

#Should I write 1.58 0 0 or 1.58 NULL NULL to superimpose 1.58 on the x component of velocity without changing its other components?