Temperature drops to zero in NVT with ReaxFF

Dear All,

I am trying to run a simple NVT test for bulk fcc Ni with a reaxff potential and I see that the temperature somwhow drops to zero, from the initial 900K, after less than 50 time steps. My input file and the potential file are attached. The version of LAMMPS I’m using is that of April 24, 2013.

Here is the output that I get

in.Ni_perf (804 Bytes)

ffield.reax (25.1 KB)

This run does not seem to have good dynamics. Can you run a NVE test? If it fails, try with reduced timestep size.

Ray

Thank you for the quick reply.

I will try to do an NVE now. Could you please elaborate on the “run not having good dynamics”. I do not quite understand what you mean.

Regards,

Hand

Meaning dynamics resulted from too big timesteps, bad force field description, bad structure, bad damping, etc.

Ray

Are you seggesting that I use a smaller timestep? I do not think there is anything wrong with the structure: I use fcc lattice and 6x6x6 simulation box. I’m not sure I understand the damping part though.

Regards,

Hand