Temperature explodes with Reaxff potential

I am trying to simulate the oxidation of graphite with the reaxff potential. My aim is to do it at high temperatures but for now, I am just building my input file and am trying simple things as a check.
I am following the example given in the reaxff potential by LAMMPS, in the CHO folder. When I am trying to run some steps in the nve ensemble using the berendsen thermostat, the temperature just goes up very quickly and I cannot control it… I have tried smaller time steps (between 0.1 and 0.25 fm) and still have the same pb… any idea? my simulation cell is pretty big (22000 particles). Is there any recommendation to know about the reaxff potential with big cell, about charge equilibration etc? Thanks a lot

My input file is:

units real
atom_style charge

read_data oxygen.lmp
read_data graphite.lmp add append shift 0 0 51
read_data oxygen.lmp add append shift 0 0 130

pair_style reax/c lmp_control
pair_coeff * * ffield.reax.cho H C O

neighbor 2 bin
neigh_modify every 10 delay 0 check no

fix 1 all nve
fix 2 all qeq/reax 1 0.0 10.0 1e-6 param.qeq
fix 3 all temp/berendsen 500.0 500.0 100.0

timestep 0.25
thermo 20

dump 1 all custom 100 test.*.dump q x y z

run 1000

The ReaxFF potential file has to be chosen not so much for the size of the problem but the composition of the specific system. There are multiple parameterizations that support hydrogen, carbon and oxygen. Some are for studying combustion reactions, others for polymers. You thus have to carefully check the publication associated with a particular ReaxFF parameter file for which kind of systems and under which conditions those have been tested on and are trained for. Using a combustion parameter set for a polymer, for example, will produce bogus results.

1 Like

Ok thanks I will definitely check that!