[lammps-users] Solving NPT+NVT together

In his Macro paper, Ahmed mentioned that he uses NPT to form the
equilibrium PDMS melt structure, and then 'insert' water molecule. I
don't understand how can one 'insert' molecules later on. I am also
trying to develop similar kind of simple LJ system for two different
chains. I start with two chain system, using repulsive LJ between
chains and attractive for same chain. That separates the chains out,
and then use NPT for one to form the melt of that chain. But that
causes problem as the melt can be positioned anywhere not where I
would like them to be. Also, separating the chains are not always
successful.

It will be great if Ahmed sheds some light on this!

Thanks,
Monojoy

In his Macro paper, Ahmed mentioned that he uses NPT to form the
equilibrium PDMS melt structure, and then ‘insert’ water molecule. I
don’t understand how can one ‘insert’ molecules later on.

I don’t think there are too many other Ahmeds on the list, especially with Macromolecules papers on PDMS, so I’ll assume you’re referring to me here. :slight_smile:

The addition of the water molecule was not done within LAMMPS. I took the end configuration of a the LAMMPS NPT run as the “starting” configuration. I then used a script to test random locations within the system box until it located a void space in the melt large enough to accommodate the water molecule(s). I then placed the molecule in the box with a random orientation at the chosen location.

I am also
trying to develop similar kind of simple LJ system for two different
chains. I start with two chain system, using repulsive LJ between
chains and attractive for same chain. That separates the chains out,
and then use NPT for one to form the melt of that chain. But that
causes problem as the melt can be positioned anywhere not where I
would like them to be. Also, separating the chains are not always
successful.

I have no idea what you mean to do here. We didn’t create single-chain melts, so I don’t have a good feel for what you’re trying to do. (Why start with the chains, and then separate them, and then condense them into a melt?)

–AEI

Yes, you could very well guess, I was referring you :slight_smile:

Thanks for the response, let me try to use the same.

Well, our system is a 2 (different) chain bead-spring system. 1 chain
form droplet (PDMS droplet kind) 2nd chain then adsorb on the surface
of the droplet. Both the chains are hydrophobic, but with different
interaction strengths. The way I am doing right now is, start with
both the chains, then use 1-2 strong repulsion whereas 1-1 and 2-2
strong attraction, that way they phase separate. Then, I used 1-2
attraction to adsorb chain 1 on the surface of the 2 drop. But, I
think your way of doing is what is needed because of a length-scale
issue.

Thanks,
Monojoy