Option of adding bonds in simulation

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,
I am using lammps version 31Mar2017. I am very confused about bond creation commends.

I am trying to simulate a tribology system using lammps. My system involves three types of material, a) metal oxide layers; b) protection layer (mono layer of organic molecules); c) centre lubricant layers (hydrocarbon molecules). I would like to bond the metal oxide layers with the protection layer. I have tried bond/create commend and create_bonds commend. Create_bonds seems trying to unset the entire bonding condition of the system and recreate the new bonding for every atom in the system. Is my recognition correct here? For bond/create, I manage to successfully run the simulation. However, the bond distance between the metal layer and protection layer does not change much and it looks the bonds are not created at all. It would be very appreciated someone could point to me a correct and clear direction of how should I make this work. By the way, I am using OPLS-AA force field.

Here is my input for bond/create commend:

fix hia hi-bindfe bond/create 1 210 930 1.9333 387 iparam 2 931

fix hib hi-bindfe bond/create 1 211 930 1.4710 388 iparam 3 931

fix loa lo-bindfe bond/create 1 210 930 1.9333 387 iparam 2 931

fix lob lo-bindfe bond/create 1 211 930 1.4710 388 iparam 3 931

Here is my input for trying create_bonds:

pair_coeff 930 930 0.34 2.32

pair_coeff 931 931 0.17 2.96

pair_coeff 211 930 0.26 2.66

pair_coeff 210 930 0.28 2.64

create_bonds many hi-bindfe hi-bindOa 387 1.5 2.0

create_bonds many hi-bindfe hi-bindOb 388 0.1 1.6

create_bonds many lo-bindfe lo-bindOa 387 1.5 2.0

create_bonds many lo-bindfe lo-bindOb 388 0.1 1.6

bond_style harmonic

bond_coeff 387 harmonic 75.0 1.9333 # bondID : Fe-O = 1.91Ang

bond_coeff 388 harmonic 75.0 1.4710 # bondID : Fe-O = 2.03Ang

kspace_style none

neighbor 0.03 bin

Thanks,
Jing Yang

2

what you are writing is confusing and thus doesn’t make much sense to me. also, there is not enough information here to easily reproduce what you are claiming, so it is impossible to give any meaningful specific advice.

since you suspect, that there may be something wrong with LAMMPS, you should first update to the very latest patch (development) version (2 Aug 2018 at the time of this writing) and test again. nobody wants to investigate a possible bug that may already have been fixed.

beyond, that i suggest you create a tiny (doesn’t have to make physically sense) but runnable input deck and explain how you determine, that LAMMPS is behaving unexpectedly.

axel.

3

First of all, thank you for your reply.

I don’t think this is a bug in LAMMPS, but I will test it with newer version and simple system. I am asking for help on how to correctly use the bond-creation-related commends. Please excuse my confusion from the previous post. My questions are:

  1. Is it feasible to use bond-creation-related commends given that I am using OPLS forcefield?
  2. If so, which bond-creation commends work better, bond/create or create_bonds? I think create_bonds is not trivia to use, because it lacks of instructions and examples on both the documentation online and manual. I am looking for some detailed explanations or experiences from people that have been successfully used this commend before.

Thanks,
Jing

4

First of all, thank you for your reply.

I don’t think this is a bug in LAMMPS, but I will test it with newer version and simple system. I am asking for help on how to correctly use the bond-creation-related commends. Please excuse my confusion from the previous post. My questions are:

  1. Is it feasible to use bond-creation-related commends given that I am using OPLS forcefield?

fix bond/create doesn’t care what your force field is. whether what it does is sufficient for your system and your force field is something that you have to determine based on the documentation. it was designed with simpler force fields in mind. for more complex operations, you may have to have a look at something more complex like the recently added fix bond/react.

  1. If so, which bond-creation commends work better, bond/create or create_bonds? I think create_bonds is not trivia to use, because it lacks of instructions and examples on both the documentation online and manual. I am looking for some detailed explanations or experiences from people that have been successfully used this commend before.

i personally don’t like topology building from inside LAMMPS at all. since LAMMPS is a parallel program, these operations are complicated and error prone. there are some applications, where they are required, but your use case doesn’t sound like it. i would use some external tool and scripting to do what needs to be done.
with scripting, you have much more control, and you don’t have to trick some complex tool into doing what you want, but rather just do what needs to be done straight away.

axel.

5

First of all, thank you for your reply.

I don’t think this is a bug in LAMMPS, but I will test it with newer version and simple system. I am asking for help on how to correctly use the bond-creation-related commends. Please excuse my confusion from the previous post. My questions are:

  1. Is it feasible to use bond-creation-related commends given that I am using OPLS forcefield?

are you looking to also correctly update surrounding angles, dihedrals, etc? if so, I would be happy to help get bond/react working for your system