I am trying to perform nucleation simulations with molecular dynamics. To prepare the metastable fluid, I am equilibrating at temperatures about 30 K above the melting temperature, and then "quenching" down to a sub cooled liquid state. For those familiar with lammps, the method I am using to perform the quench is
fix 1 all nve fix 2 all viscous 0.01
This just adds a drag force proportional to the velocity to suck heat out of the system.
In the nucleation simulation literature, I can't think of a time I have come across a paper where the author mentions an actual quench rate, or discusses whether the quench rate can have implications for the final results. Is anyone familiar with what an appropriate quench rate to prepare a metastable liquid, or whether I can produce spurious physical artifacts depending not quench rate? Currently for my water simulation, I am quenching from 300 K to 190 K in 50 steps with a 2 fs time step.