I want to put to zero any reaction between type 1 and 2.
Ive tried with this commands. but it wasn’t work and I think tersoff between two atom types remained .have you any idea?
pair_style hybrid lj/cut 50.0 tersoff
pair_coeff * * tersoff BNC.tersoff C C
pair_coeff 1 2 lj/cut 50.0 0.0 0.0
I want to put to zero any reaction between type 1 and 2.
Ive tried with this commands. but it wasn’t work and I think tersoff between two atom types remained .have you any idea?
please try this:
pair_style hybrid lj/cut 1.0 tersoff tersoff
pair_coeff * * tersoff 1 BNC.tersoff C NULL
pair_coeff * * tersoff 2 BNC.tersoff NULL C
pair_coeff 1 2 lj/cut 0.0 3.0
axel
Sorry, these commands didn’t make zero interactions between type 1 and 2. Is there any solution?
This is the solution. How did you determine that there are unwanted interactions?
I move one group of atoms through the velocity set command and I expect that the movement of this group has no effect on the other’s because of no interaction. On the other hand in another test as an other way for making the movement of two groups independent I freeze the first group by set_force command and move the other group. But in the both cases two groups move together and all commands to set interactions or positions doesn’t work. I don’t know the reason. Also I write the ensembles npt separately for both groups
I move one group of atoms through the velocity set command and I expect that the movement of this group has no effect on the other’s because of no interaction. On the other hand in another test as an other way for making the movement of two groups independent I freeze the first group by set_force command and move the other group. But in the both cases two groups move together and all commands to set interactions or positions doesn’t work. I don’t know the reason. Also I write the ensembles npt separately for both groups
there are multiple problems here. you have asked to block interactions between the two groups, but not within each group.
when you want to have a constant velocity motion of a group of particles and the group staying rigid, you have to use velocity set and fix setforce 0 0 0 or you can use fix move and not use any other time integration on those particles. fix move ignores forces completely.
when you use fix npt (in any variant), then you have to deal with the fact, that all particles will move, when the box is rescaled (unless you elect to rescale only some subset of atoms). but since the rescale is global, you MUST NOT have multiple box changing fixes (like two fix npt instances). those will conflict and lead to bogus results. besides, i don’t see why you would need to use fix npt (or any other box changing time integration) for the kind of system you are describing that you are looking at to simulate.
in summary, there seem to be some significant conceptual problems that you would have to resolve first by talking to your adviser/supervisor and more experienced colleagues and spending more time studying text books on MD simulation techniques. also, i would recommend practicing doing MD simulations with simpler systems, so that you are more confident to get the basics right before doing something more complex and then getting confused why things don’t work as expected. at the moment it looks that most of the things that don’t work as expected are because either your expectations are incorrect or how you implement them is not correct.
axel.