fix langevin friction

Dear all,

I use lammps to study the friction behavior between a tip and a four-layer graphene substrate. The topmost layers of the tip are treated as rigid. The bottommost layer of the substrate is fixed. To control the temperature of the system, fix langevin is applied to three atomic layers in the middle of the tip and one atomic layer in the middle of the substrate. The damp factor in fix langevin command is 0.05ps. The timestep of the simulation is 0.0005ps. A constant normal force and a constant sliding velocity Vx=10 m/s are applied to the rigid layer of the tip. Then I have the following questions:

  1. The damp factor in fix langevin command is appropriate?

2.When I use the following command to control the temperature, I found the mean force in x direction of the tip is larger than that of the substrate. I think langevin thermostat lead to such results. Which command did other people use to eliminate its effects?

dimension 3

boundary p p s

units metal

compute Ttipm tipm temp

compute Tgram1 gram1 temp

compute 1 gra property/atom fx fy fz

compute 1x gra reduce sum c_1[1]

compute 1y gra reduce sum c_1[2]

compute 1z gra reduce sum c_1[3]

compute 6 tip property/atom fx fy fz

compute 6x tip reduce sum c_6[1]

compute 6y tip reduce sum c_6[2]

compute 6z tip reduce sum c_6[3]

fix 3 gram1 langevin 300.0 300.0 0.05 68954256 zero yes

fix 4 tipm langevin 300.0 300.0 0.05 68954256 zero yes

fix_modify 3 temp Tgram1

fix_modify 4 temp Ttipm

  1. When I use commands as follows,I found the force in the x-direction of the tip and the substrate is a pair of interacting forces. Is the result appropriate?

compute Ttipm tipm temp/com

compute Tgram1 gram1 temp/com

compute 1 gra property/atom fx fy fz

compute 1x gra reduce sum c_1[1]

compute 1y gra reduce sum c_1[2]

compute 1z gra reduce sum c_1[3]

compute 6 tip property/atom fx fy fz

compute 6x tip reduce sum c_6[1]

compute 6y tip reduce sum c_6[2]

compute 6z tip reduce sum c_6[3]

fix 3 gram1 langevin 300.0 300.0 0.05 68954256 zero yes

fix 4 tipm langevin 300.0 300.0 0.05 68954256 zero yes

fix_modify 3 temp Tgram1

fix_modify 4 temp Ttipm

Step Temp 1x 1y 1z 6x 6y 6z

110000 297.702 -1.67969 0.81919 -222.65799 1.67969 -0.81919 4.20516

110500 301.11943 -0.13162 2.90778 -221.06796 0.13162 -2.90778 2.61513

Thanks in advance,

Lingfei

please note, that this is a mailing list to help people with and
discuss features of the LAMMPS MD simulation software. what you are
asking about is to get approval for your choices. that is not going to
work. as a researcher, it is *your* responsibility and part of your
work, to find and follow ways to justify your choices. ultimately,
you'll have to publish your results as your work and you cannot write
in a paper "some random dude on the LAMMPS mailing list said it was ok
to do it this way". you can study the published literature, and then
try to reproduce such work, learn in the process which factors impact
your results how and then apply what you've learned to your research.

axel.