Temperature

Statics_Eng · March 17, 2017, 10:28am

Hi all
I have a question about how imposing the temperature of the environment in my simulation.
If I want to simulate a functional CNT at 600K (the temperature of the environment) and a random velocity is adopted to reach the room temperature and the simulation commences by relaxing the CNT at room temperature for 5 ps,
Is it true to consider 600K as room temperature?
Indeed, I use this approach but interestingly, I see the strange behavior in my simulation. Some functional group is dissociated from CNT’s wall! (I use AIREBO potential and CH2=C as functional group)

Thanks.

akohlmey · March 17, 2017, 12:51pm

Hi all
I have a question about how imposing the temperature of the environment in
my simulation.
If I want to simulate a functional CNT at 600K (the temperature of the
environment) and a random velocity is adopted to reach the room temperature
and the simulation commences by relaxing the CNT at room temperature for 5
ps,

not sure what you are saying here. it would be better, if you provide your
input and explain what you are referring to with which statement.

Is it true to consider 600K as room temperature?

https://en.wikipedia.org/wiki/Standard_conditions_for_temperature_and_pressure

Indeed, I use this approach but interestingly, I see the strange behavior
in my simulation. Some functional group is dissociated from CNT's wall! (I
use AIREBO potential and CH2=C as functional group)

doesn't seem very interesting to me. you could just have an unrelaxed or
unstable initial geometry. also, at elevated temperatures, bonds will break
more easily., especially, when they are already strained in some way.

axel.

Statics_Eng · March 17, 2017, 6:46pm

This is my input file:

---------- Initialize Simulation ---------------------

units metal
dimension 3
atom_style atomic
boundary s s s
atom_modify map array
timestep 1e-3

---------- Simulation Box -----------------------------

read_data data.DAT

---------- Define Interatomic Potential --------------

pair_style airebo 3.0 1 1
pair_coeff * * CH.airebo C H

---------- Define Settings ----------------------------

…

#Initial velocities------------------------------------------------

velocity all create 300 511124 rot yes mom yes

Relaxing the Imported structure in the desired temperature------------------

fix relax all nvt temp 300 300 0.01
compute mystress middle_atoms stress/atom myTemp
compute reducedstress middle_atoms reduce sum c_mystress[3]

thermo_style custom c_reducedstress
thermo 1000
dump relax_position all atom 5000 dump2.atom #For having video!
run 5000
undump relax_position
unfix relax

Zeroing Stress----------------------------------------------------

thermo_style custom fmax fnorm
min_style cg
minimize 0 0 1000 1000000

Defining fixes-------------------------------------------------------

variable stepnum equal 1000
variable Nstep loop ${stepnum}
variable disp_upper_atoms equal 0.01*v_Nstep

fix fixing_Left_side lower_atoms setforce 0 0 0
fix fixing_Right_side upper_atoms setforce 0 0 0

…

label looplable

displace_atoms upper_atoms move 0 0 -0.01
fix fix1 middle_atoms nvt temp 300 300 0.01
thermo 5000
thermo_style custom step temp pe v_sigma
run 5000

next Nstep
jump in.CNT looplable

Is it true to change my input file to:

#Initial velocities------------------------------------------------

velocity all create 300 511124 rot yes mom yes =====> velocity all create 600 511124 rot yes mom yes

Relaxing the Imported structure in the desired temperature------------------

fix relax all nvt temp 300 300 0.01 ======> fix relax all nvt temp 600 600 0.01

…

Defining fixes-------------------------------------------------------

…

label looplable

displace_atoms upper_atoms move 0 0 -0.01
fix fix1 middle_atoms nvt temp 300 300 0.01 ======> fix fix1 middle_atoms nvt temp 600 600 0.01
thermo 5000
thermo_style custom step temp pe v_sigma
run 5000

next Nstep
jump in.CNT looplable

in order to simulate my functional CNT in the thermal environment? (600K is the temperature of the environment. Indeed, I want to simulate the functional CNT in thermal environment)
I hope I can explain my purpose truly.
Thanks.

akohlmey · March 17, 2017, 7:41pm

Is it true to change my input file to:

#Initial velocities------------------------------------------------

velocity all create 300 511124 rot yes mom yes =====>
velocity all create 600 511124 rot yes mom yes

# Relaxing the Imported structure in the desired
temperature------------------

fix relax all nvt temp 300 300 0.01 ======> fix relax
all nvt temp 600 600 0.01

...

# Defining fixes-------------------------------------------------------
....

label looplable

displace_atoms upper_atoms move 0 0 -0.01
fix fix1 middle_atoms nvt temp 300 300 0.01 ======>
fix fix1 middle_atoms nvt temp 600 600 0.01
thermo 5000
thermo_style custom step temp pe v_sigma
run 5000

next Nstep
jump in.CNT looplable

in order to simulate my functional CNT in the thermal environment? (600K
is the temperature of the environment. Indeed, I want to simulate the
functional CNT in thermal environment)
I hope I can explain my purpose truly.

whether this does what you want or not, is not something that you should
ask the mailing list for approval about, but rather, you should a) read the
documentation, b) try it out and observe and c) discuss with an experienced
colleague or tutor or supervisor what your findings are and whether the
simulation acts to your expectations.

none of this is really a question about LAMMPS, but it is a question about
how to do simulations in general and that is beyond the scope of this
mailing list, especially considering the rather fundamental and similarly
trivial topic of your inquiry.

axel.