Molecular Dynamic simulation of Lennard-Jones gas engine.

Visualization for 961 particles distibuted on the plane:

Red particle represents Brownian motion.

3D visualization using Ovito:

Build

Requirements:

• C++17 compatible compiler
• cmake 3.1+
• OpenMP
• Eigen
• nlohmann/json
• fmt
• rapigcsv

Note: use git --recurse-submodules to download submodules (see extlibs/)

Building

• Create and change working directory to build directory: mkdir build && cd build.
• Build on Linux: cmake .. && cmake --build .. It will build all required dependencies and the project itself.

Default simulation parameters, Lennard-Jones properties, etc. can be modified in data/conf.json. All executable files will be placed in build/ directory.

Running and Usage

• Run simulation: ./calc.exe. It will save all simulation output data in data/ directory.
• Run output formatter: ./output_format.exe. You will be asked to specify filenames to format (e.g. position_data.csv, absolute_position_data.csv). It will format all output data in data/ directory and save it with .lammps extension.
• Use OVITO to get 3D visualization: File -> Load File -> <Auto-detect file format>(*) and select formatted output file with extension .lammps.

Configuration:

You can specify simulation (e.g. gas density, number of threads, Lennard-Jones constants, etc.) parameters in data/conf.json file.

Acknowledgements:

This project contains

• Lennard-Jones potential
• Velocity Verlet algorithm
• OpenMP

Results

See report_RU.pdf for more details.

References:

• Transport coefficients of the Lennard-Jones model fluid. II Self-diffusion
• Self‐Diffusion in Argon
• Молекулярно-динамическое моделирование процесса установления термодинамического равновесия нагретого никеля (RU)