Infinite Graphene with explicit bonds causing compressive stresses during straining

LAMMPS LAMMPS Mailing List Mirror
1

Dear LAMMPS Users,

In my research project I have a graphene sheet in the middle of a box on xy plane which is infinite only along x direction and polymer inside the box with periodic boundary conditions in all directions.
In my lammps data file I have defined the bonds, angles, and dihedrals explicitly and for the infinite direction of the graphene sheet because of the potential with explicit bonds.
I am equilibrating the system in NPT with allowing only z dimension relax to reach the correct density at low temperature. (Y dimension of the box is larger than graphene width in this direction and kept fixed along with the X dimension). Lammps doesn’t give any warnings during the equilibration.

After it has reached to the correct temperature, I apply uniaxial tensile straining along y direction of the box. (Using change_box command).

However when I calculate the Stress components, the stress in the straining direction (y direction) of the graphene sheets shows a compression in the middle. This is only for some of the x lengths of the graphene sheet (In the infinite direction). And for some x lengths of the graphene sheet, this stress shows tension in the middle of the sheet which is the expected. It doesn’t look like a problem due to polymer equilibration because of the visual observations and the polymer density calculations show correct data.

At this moment, I have rechecked the code which I have been using to generate the input structures for possible errors and there is none that I could see.
Please let me know if there is anyone who encountered a similar problem or any suggestions to find out the reason behind this or anything to do with the lammps distribution I am using. (lammps-15May15)

Thank you!

Best Regards,
- Thushara

2

After it has reached to the correct temperature, I apply uniaxial tensile straining along y direction of the box. (Using change_box command).
Why do you use change_box? This applies a large amount of strain all in one go, which is not very realistic. Better to use fix deform. Also,
you are probably applying strain to the graphene sheet (can’t say for sure without seeing script), which is unphysical, since it does not span
the box periodically in y. You should prevent graphene from being strained in your change_box or fix deform commands.

However when I calculate the Stress components, the stress in the straining direction (y direction) of the graphene sheets shows a compression in the middle.
The most likely explanation is that it was already in compression before you applied the strain. You did not mention what the pre-strain stress field looked like. Other possibilities include short-lived thermal fluctuations.