Graphene temperature modeled with AIREBO potential

Dear Lammps users,

I am equilibrating graphene sheets using AIREBO potential under NVT simulation at 250.0K in metal units. In my simulation setup I have also water molecules modeled by TIP4P and I converted all the energy (epsilon) values to eV as the default unit in metal.

Graphene temperature is initalized at 250K as I set as thermal velocity, then decreased to 134K then fluctuated around this value. I don’t know what the reason of this would be. The structure of graphene sheets looks fine during the simulation but the temperature is not fluctuating around its set point. Any help would be appreciated. Below, I provided the pair-style I used in this simulation.

pair_style hybrid airebo 3.4 lj/cut/tip4p/long 2 1 1 1 0.1577 12.0 8.5 lj/cut 10.2
pair_coeff * * airebo CH.airebo NULL NULL C C C C C C C C C C C C C C
special_bonds lj/coul 1.0 1.0 1.0

fix 1 ihice nvt temp 250.0 250.0 20.0
fix 2 graph nvt temp 250.0 250.0 20.0

compute myTempW ihice temp
compute myTempG graph temp

Please let me know if any more information would be needed.

Thank you,

I presume you are using metal units, since that matches the provided AIREBO potential file.
LAMMPS time is ps for metal units, so that means 20.0 in your fix nvt command is 20 ps.
Which is likely ~20,000 timesteps. Which means your Temp will
equilibrate on that timescale, which is very long for normal simulations.
Try a value of 0.1 (100 steps) instead and see if the temp stays at 250K.