about water-oil interface

LAMMPS LAMMPS Mailing List Mirror
1

hi all lammps users
I used NPT ensemble to simulate the water-oil interface, and found that the water-oil interface was constantly moving through the trajectory. What causes this phenomenon?
Looking forward to your reply, thanks very much!
Best wishes!
zhang

2

Zhang,

What do you mean when you say that the water-oil interface was constantly moving through the trajectory? Do you have any images to show us that explain what you mean? Or maybe you could describe it in more detail?

Will

3

hi all lammps users
I used NPT ensemble to simulate the water-oil interface, and found that the water-oil interface was constantly moving through the trajectory. What causes this phenomenon?

some residual center of mass momentum. this most often happens during equilibration. theoretically it should not happen if a system starts without any net momentum, but due to limitations of floating point math it can be that forces do not cancel completely when they should. the effect is often emphasized through the choice of thermostat.

in most cases, using the velocity command to rescale the velocities and remove the net momentum after equilibration will address this.
for pathological cases, there is also the fix momentum command, or you could use fix spring to tether the center of mass of one of the two phases to a fixed location.

axel.

4

Axel,

Thank you very much .

zhang

5

axel,

I found that because my system set the rigid body, but in the zero type of the velocity command, should I set the linear momentum of the rigid body to zero or the angular momentum to zero? I don’t know whether to choose linear or angular to cancel except the residual center of mass momentum you mentioned?

thanks very much

zhang

6

it doesn’t matter how you do it, but any combination of velocity commands that zeroes the total linear momentum should do.
so if you apply this first to all rigid bodies and then to the rest, the total center of mass momentum should be zero.
for a system as you describe, you cannot have a total rotation so zeroing the total angular momentum is not required.

axel.

7

thanks very much , axel

8

axel

I’m sorry to disturb you again.
After I add rigid nanoparticles to the system, the interface of the system is constantly moving as mentioned before.
Then I used “velocity all zero linear”, but it didn’t work and the interface was still moving. May I ask why?

zhang

9

sorry, but i am not a magician. i have given you an explanation of what is causing this effect and multiple options to counter it. the rest is up to you. i don’t have the time to do your research for you.

axel.

10

Thank you all the same , axel