[lammps-users] Using fix viscous to drain out the kinetic energy


I would like to use fix viscous to freeze the atoms after MD run, but after running for several picoseconds, the energy doesn’t change. Could you enlighten me what is going on here? The related commands I am using are listed here:

velocity all create 300.0 4928459 temp new
fix 3 all nvt 300.0 300.0 0.01
thermo 50
thermo_modify temp new
neigh_modify delay 10 every 1 check yes
timestep 0.0005
run 10000
unfix 3
fix 4 all viscous 80
run 5000

My unit is metal, from the log file I can see the kinetic energy is around 200eV, And I have checked the force and velocity units in metal, and the unit for Gamma should be ev/ps. So I suppose the kinetic energy should be drained out in 2.5 ps considering the gamma is set to be 80 ev/ps. But it doesn’t change. Could anybody tell me where I am getting wrong here?


Fix viscous doesn't do time integration. You need a fix nve
or the like, as well.