MC-exchange Python Package

The MC-exchange Python package has been developed to support the dynamic bond exchange reactions (BERs) during a molecular dynamics (MD) simulation run using LAMMPS (Large-scale Atomic/Molecular Massively Parallel Simulator) software. Using the LAMMPS Python package, a MD simulation can be run directly from a Python script, allowing an interface between Python and LAMMPS. The MC-exchange Python package uses this to gather the necessary data from the LAMMPS MD simulation, analyze the data, determine which bonds to exchanged, and finally communicate the altered bond data back to LAMMPS, where the actual bond exchange can take place.

These simulations are useful for studying the effects of dynamic bonds of associative covalently adaptive polymer networks.

The package is written with the ease of use in mind. However, users must be warned that a thorough understanding of the LAMMPS Python package is necessary to be able to gather the necessary bits of data prior to performing bond exchange reactions using the MC-exchange Python library.

General Use

The package can be installed from PyPI distribution archive using:

pip install MC-exchange

After successful installation of the MC-exchange Python package, it can be imported into a script as follows:

import MC_exchange as mc

The above statement imports all user-necessary functionalities of the package.

Package organization

The MC-exchange Python package is organized into modules. Currently, there are three modules dedicated to the base functionalities, and one module where these functionalities are combined to perform bond exchange reactions. The three base modules are calculations.py, data.py and neigh_list.py. All functions necessary to perform dynamic bond exchange reactions are imported direcly, thus the user does not need to search these modules for the required functionalities. Nevertheless, we provide a brief overview of the different modules, so that the user can better understand the structure of the library.

Base modules

Currently, three base modules are responsible for the primary functionalities of the package. data.py is the module responsible for gathering data from the LAMMPS simulations. calculations.py contains calculations that use the data obtained by the functions of data.py to compute properties that can be necessary during bond exchange reactions. Lastly, neigh_list.py contain the functionalities required to construct spacial neighbor lists using atom data.

Functionalities

Currently, in version 1.1.0, there are two main functions used to perform associative bond exchange reactions. Firstly, the evaluate_bond_exchange() function is used to perform two types of bond exchange reactions between stickers of any kind. The two types of bond exchange reactions allowed are bond shifts and bond swaps. The first is an 3-body exchange reaction between a bonded sticker pair and a 'free' sticker where a single bond migrates, while the latter is a 4-body exchange between two pairs of bonded stickers requiring the rearrangement of two bonds. In either case, the number of total bonds in the system does not change.

Secondly, the complementary_bond_exchange function, as the name suggests, allows reactions to occur between stickers that are of different types (i.e., A-B crosslinks). In the function, these types can be defined according to LAMMPS' numerical type convention under arguments sticker_types_A and sticker_types_B.

Both of the above mentioned bond exchange functions allow bond shifts and bond swaps to be turned on and off according to the user's requirements.

The software is designed such that it works when LAMMPS simulations are run parallel (i.e., on multiple processors). Communication between processes is handled by the mpi4py Python package.

Example script

In order to run the following script...

Users are encouraged to clone the repository of this project to access all source files and example code.

To run the example script, make sure that the Python environment into which MC-exchange was installed is activated and you are located in the src/example/ directory. To run the example script simply run the following command in the terminal:

mpiexec -np {n_procs} python3 loop.py

where {n_procs} is the number of parallel processes the user wishes to launch.

Note, this requires that the user has installed OpenMPI or MPICH to the local machine.

Developer

Balint Magyari: PhD Student at University of Naples Federico II

For any questions please contact balint.magyari@unina.it.