Rotation of a quadruple set of spherical copper nanoparticles on a graphene nanosheet

LAMMPS LAMMPS Beginners
,
1

Hello everyone, I am a relatively amateur user of Lamps software, I am producing copper/graphene composite with Lamps software, the problem is that during the process, the set of copper nanoparticles (4 upper particles together and four lower particles) (together) constantly move on the graphene sheet and the more I increase the relaxation time, it has no effect, my code is as follows =

dimension 3
boundary p p p
units metal
atom_style atomic
#NanoParticles
read_data data1.txt
lattice fcc 3.615
region box1 sphere 23.85599 25.339 -29.7 25 units box
create_atoms 2 region box1
region box2 sphere 73.85599 25.339 -29.7 25 units box
create_atoms 2 region box2
region box3 sphere 123.85599 25.339 -29.7 25 units box
create_atoms 2 region box3
region box4 sphere 173.85599 25.339 -29.7 25 units box
create_atoms 2 region box4
region box5 sphere 23.85599 75.339 -29.7 25 units box
create_atoms 2 region box5
region box6 sphere 73.85599 75.339 -29.7 25 units box
create_atoms 2 region box6
region box7 sphere 123.85599 75.339 -29.7 25 units box
create_atoms 2 region box7
region box8 sphere 173.85599 75.339 -29.7 25 units box
create_atoms 2 region box8
region box9 sphere 23.85599 25.339 29.7 25 units box
create_atoms 2 region box9
region box10 sphere 73.85599 25.339 29.7 25 units box
create_atoms 2 region box10
region box11 sphere 123.85599 25.339 29.7 25 units box
create_atoms 2 region box11
region box12 sphere 173.85599 25.339 29.7 25 units box
create_atoms 2 region box12
region box13 sphere 23.85599 75.339 29.7 25 units box
create_atoms 2 region box13
region box14 sphere 73.85599 75.339 29.7 25 units box
create_atoms 2 region box14
region box15 sphere 123.85599 75.339 29.7 25 units box
create_atoms 2 region box15
region box16 sphere 173.85599 75.339 29.7 25 units box
create_atoms 2 region box16
region left1 block -1.22802 0.22802 0 100.678 -1.7 1.7 units box
group left1 region left1
region right1 block 197.94 198.94 0 100.678 -1.7 1.7 units box
group right1 region right1
region left2 block 0.22802 197.94 0 0.5 -1.7 1.7 units box
group left2 region left2
region right2 block 0.22802 197.94 100.378 100.678 -1.7 1.7 units box
group right2 region right2

eam_potential

pair_style hybrid airebo 3 0 0 eam lj/cut 8.0625
pair_coeff * * airebo CH.airebo C NULL
pair_coeff 2 2 eam Cu_u3.eam
pair_coeff 1 2 lj/cut 0.01996 3.225
timestep 0.001
velocity all create 300.0 12345678 mom yes rot yes dist gaussian
velocity left1 create 0 12345678 mom yes rot yes dist gaussian
velocity left2 create 0 12345678 mom yes rot yes dist gaussian
velocity right1 create 0 12345678 mom yes rot yes dist gaussian
velocity right2 create 0 12345678 mom yes rot yes dist gaussian
fix 4 left1 setforce 0 0 0
fix 5 left2 setforce 0 0 0
fix 6 right1 setforce 0 0 0
fix 7 right2 setforce 0 0 0
thermo 100
thermo_style custom step etotal temp density pe ke
dump 1 all xyz 100 MDr.xyz
#process
fix 1 all nvt temp 300.0 300.0 0.1
run 200000
unfix 1
fix 2 all nvt temp 300.0 950 0.1
run 200000
unfix 2
fix 3 all nvt temp 950 950 0.1
run 200000
unfix 3

The fix 4,5,6,7 parts are placed in order to fix the four sides of the graphene with a very small width, because otherwise the graphene will stick to the nanoparticles and make the situation more difficult.

The code read_data data1.txt is related to the coordinates of carbon atoms and the dimensions of the simulation box, I have included a part of the data1 file below:

Carbon Positions in CNT Column1

7872	atoms			
2	atom	types		
				
-15 240	xlo	xhi		
-15 120	ylo	yhi		
-75 75 zlo zhi

Masses

1	12.01					
2	63.55

Atoms

1	1	0	0	0
2	1	-1.22802	0.709	0
3	1	-1.22802	2.127	0
4	1	0	2.836	0

…
1
2

Does anyone have any suggestions to stop the copper nanoparticles from moving on the graphene?
Thank you for your attention dear ones

2

Something must be wrong with your input beyond what you are showing here to have the copper atoms moving.

The input is needlessly complex and convoluted. If you don’t want the copper atoms to move, simply don’t use fix nvt on group all, but on a group with only the graphene sheet. That will also result in correct thermostatting. Fix nvt will add a temperature compute for the fix group. If that group is all, and the copper atoms are immobile, then the carbon atoms must be accelerated even more to generate the desired total temperature. Since copper is quite heavy compared to carbon, your nanosheet will have to become very hot to have the total temperature at your desired value.

3

Thank you, dear akohlmey, for reading my problem. In fact, my goal is to make a copper/graphene nanocomposite by using the sintering process, which is why I have used fix nvt because the temperature cycle is the basis of the work.
The complete cycle is as follows:
#relaxing
fix 1 all nvt temp 300 300 0.1
run 200000
unfix 1

#heating
fix 2 all nvt temp 300 950 0.1
run 200000
unfix 2

#holding
fix 3 all nvt temp 950 950 0.1
run 200000
unfix 3

#cooling
fix 4 all nvt temp 300 300 0.1
run 200000
unfix 4

Therefore, I have to control not only the location of the copper atoms, but also their temperature.
The problem is that the set of eight copper nanoparticles rotates both from above and below in all stages (relaxing,heating,holding,cooling ).
If I understand correctly, you said to use nve thermostat for the copper nanoparticles, which are heavier, and nvt thermostat for the graphene nanosheet. Now, given that I need to control the temperature of the copper nanoparticles, how do I do this?

4

First and foremost, most of your problems are not LAMMPS problems but problems of how to design and implement meaningful simulations. That is independent from the simulation software and a topic for discussion with people that care about your research like your adviser, your tutor(s), your colleague(s) and so on. In the LAMMPS category, we prefer to focus on LAMMPS issues.

What you are saying and what you are providing in your example input fragments is contradictory. I have no interest to fill in for your adviser and I don’t know anything about the objectives of your research, so I cannot give you specific instructions, just a few general remarks.

  • You have to decide whether you want to do science or computer animations. In the former case, you have to follow what the model and thus the physics of your simulations determine, for the latter case, you probably want to use a different software from LAMMPS where you have more control over how things are moving around.
  • It is not clear from the image of the model, what is the mechanism of heating. I don’t think that a Nose-Hoover thermostat is a good idea here. Its algorithm is designed to maintain a given temperature by coupling it to a fictitious heat bath and thus induce fluctuations that are close to those of a large homogeneous bulk system. It is not well suited to move large amounts of kinetic energy into or out of a system or to promote equipartitioning. For that you have to use a dissipative thermostat like fix langevin. But then again, you have an isolated system, so what is it coupling to to get heated?
  • You are not showing what kind of rotation you are seeing and how that would affect the validity of your simulation. Are you sure this is unphysical behavior?

No. You do not understand correctly.