simulation of Reactive Cross-Linking between two Polymers on LAMMPS

Dear all,
Currently, I am doing simulation of Reactive Cross-Linking between two Polymers (bisphenol F and DETDA) as described in the article “Accelerated ReaxFF Simulations for Describing the Reactive Cross-Linking of Polymers” (Aniruddh Vashisth, Chowdhury Ashraf, Weiwei Zhang, Charles E. Bakis and Adri C. T. van Duin).

I want to implement the method that described in the article and developed within the ReaxFF reactive force field framework to accelerate the cross-linking reaction by providing a sufficient amount of energy, based on the distances and orientations of the reactants to help the reactants overcome the reaction barrier.
If a suitable initial configuration is recognized, it then provides additional energy to the system by stretching or compressing bonds at a predefined rate such that this additional energy can help achieve the energy to cross the energy barrier of the polymerization.
The additional energy provided to the system goes to zero after 10,000 time steps (of 0.25 fs in this study), and the system reverts back to a simple NVT simulation until the spatial requirements for accelerated simulations are satisfied again. This is consistent with the “bond-boost” approach.

For that, I should check if a suitable configuration is recognized and create a pair style cpp class for the added potential.

I have some questions:

  1. Does this functionality already exist?
  2. Is it a good idea to do this in LAMMPS?
  3. If it is, Should I implement the suitable configuration check on the pair style cpp class of the potential or as a separated fix code?

Sincerely,
Ofek Barazani
Student of Software Engineering
Azrieli College of Engineering Jerusalem, Israel

Dear all,
Currently, I am doing simulation of Reactive Cross-Linking between two Polymers (bisphenol F and DETDA) as described in the article "Accelerated ReaxFF Simulations for Describing the Reactive Cross-Linking of Polymers" (Aniruddh Vashisth, Chowdhury Ashraf, Weiwei Zhang, Charles E. Bakis and Adri C. T. van Duin).

I want to implement the method that described in the article and developed within the ReaxFF reactive force field framework to accelerate the cross-linking reaction by providing a sufficient amount of energy, based on the distances and orientations of the reactants to help the reactants overcome the reaction barrier.
If a suitable initial configuration is recognized, it then provides additional energy to the system by stretching or compressing bonds at a predefined rate such that this additional energy can help achieve the energy to cross the energy barrier of the polymerization.
The additional energy provided to the system goes to zero after 10,000 time steps (of 0.25 fs in this study), and the system reverts back to a simple NVT simulation until the spatial requirements for accelerated simulations are satisfied again. This is consistent with the “bond-boost” approach.

For that, I should check if a suitable configuration is recognized and create a pair style cpp class for the added potential.

I have some questions:
1. Does this functionality already exist?

not in the official LAMMPS distribution, if this is what you are
asking. all functionality in the LAMMPS distribution is documented in
the manual.

2. Is it a good idea to do this in LAMMPS?

i don't see anything (yet) that would make it a bad idea or that would
be particularly difficult to do in LAMMPS. how meaningful this would
be in general is difficult to judge, and how difficult this is to do,
depends on the particular details.

3. If it is, Should I implement the suitable configuration check on the pair style cpp class of the potential or as a separated fix code?

in the interest of code maintainability, this would be best
implemented as a separate fix. there are already multiple fixes in the
USER-REAXC package, that access internal data of the reax/c pair
style, so you can learn from them.

HTH,
     axel.