Dear all,
When I use lammps to simulate the system consist of a liquid nanodroplet and a graphene substrate, some questions confuse me. Could someone help me?
I do separate equilibration for droplet and substrate, and then combine them to the system and start to simulate.
1. I use different fix nvt to droplet and substrate and find the equilibration is very quick, but the droplet continuously spreads along the substrate and finally is to be a liquid film. Could someone tell me why it is? Should I use one fix nvt for the whole system?
2. I find the droplet move along the substrate surface if I do not use the “fix MOMENTON ILs momentum 100 linear 1 1 1 angular rescale” command. I would like to know if there are other ways to solve this problem.
Best wishes
Yongji
Yongji Guan, Postdoc
Dear all,
When I use lammps to simulate the system consist of a liquid nanodroplet and a graphene substrate, some questions confuse me. Could someone help me?
I do separate equilibration for droplet and substrate, and then combine them to the system and start to simulate.
1. I use different fix nvt to droplet and substrate and find the equilibration is very quick, but the droplet continuously spreads along the substrate and finally is to be a liquid film. Could someone tell me why it is? Should I use one fix nvt for the whole system?
how should this be an issue of how you do time integration? the (rather obvious) first thing to look into would be the potential parameters describing the interaction between the droplet and the substrate.
2. I find the droplet move along the substrate surface if I do not use the “fix MOMENTON ILs momentum 100 linear 1 1 1 angular rescale” command. I would like to know if there are other ways to solve this problem.
this is difficult to discuss without knowing more details. your droplet may pick up some COM momentum from the initial velocity assignment and also from the shape of the droplet vs the geometry of the substrate. after your system is properly equilibrated you should be removing the COM momentum via the velocity command and then the droplet should be (mostly) in place. some “meandering” would not be unexpected. the smaller the droplet, the more likely that interactions with the substrate are not symmetrical and thus can result in a non-zero momentum. i don’t think that would be un-physical behavior that needs to be suppressed at all cost.
axel.