Project 1: Molecular dynamics simulation of Argon atoms

last updated: 27/02/2021

This is the first project for Computational Physics (AP3082) year 2020/2021, T-course of the master Applied Physics at the Delft University of Technology. The code is written for Python 3.8.6 using JupyterLab v2.2.9 web-based userinterface.

This program plots the relative distance, force and energy of 2 and 32 argon particles with 3 dimensions. The potential of the argon atoms is based on the Lennard-Jones potential and thus, so is the force. Simulation should take a maximum of 15 minutes.

---

To run the program: Import all files in the directory and run main.py
optional:
-edit parameters
allowed parameters with full functionality:
- sim.verlet
- sim.box_dim
- sim.num_atoms
- sim.rescaling
- sim.temp

run main.py (for supported versions of Python please see Installation)

---

functionality

• Non-dimensional SI units
  • All inputs and outputs are non-dimensional
• Periodic or non-periodic boundary conditions
• Initial random locations
  • Only inside predefined box size
• Initial 'random' velocities
  • Based, but not exact, on 3D equipartiton theorem, changing dimensions in parameters does not change random velocity algorithm
• Plots of total energy
• Plots of kinetic energy
• Plots of potential energy
• Choice of Euler method numerical approximation or Verlet algorithm
• Potential (and thus force) based on Lennard-Jones potential
• Rescaling of velocities based on target temperature

Installation

Usage of JupyterLab v2.2.9 is recommended, different versions might have reduced functionability/performance. For an installation manual please visit the Jupyterlab documentation. Since this code is written using Python 3.8.6, it is likely that most features will work for various IDE's running Python 3.8.6.

Acknowledgements

The following persons are the instructors of the course and supplied theoretical material:

• Dr. Michael Wimmer
• Dr. Anton Akhmerov

The following persons are the technical assistants of this course and are available for intermittent grading, feedback and questions:

• Helene Spring
• Marta Pita Vidal
• André Melo
• Anastasiia Varentcoval
• David van Driel

A public forum is available during this course where anyone related to this course can ask questions, and provide insights.
JupyterHub has been supplied by the Delft University of Technology to allow cloud computing, without the installment of any software.

Authors

• Abi Kanagaratnam
• Jim Koning