Workflow for simulating HEAs using LAMMPS Python API.
It is suggested to create a virtual environment before starting to run and analyze MD simulations. It is highly suggested to use mamba for this purpose.
The file environment.yml contains an example of a list of modules that you may require. Feel free to modify it, at your own risk.
Create a virtual environment by running:
mamba env create --name workflow-hea --file environment.yml
activate it by running:
mamba activate workflow-hea
and deactivate it by running:
mamba deactivate
I haven't found a way to install Ovito from the .yml file that doesn't break MACE. If you want to use Ovito, run the following line after activating the environment:
mamba install --strict-channel-priority -c https://conda.ovito.org -c conda-forge ovito=3.11.1
The simulation is described in water-graphene/SIMULATION-README.txt. You can use the ´stub´ notebook water-graphene/workflow.ipynb to initialize and run a very short simulation, as the full simulation requires either substantial compute resources or a lot of time.
The simulation is essentially based on an example by S. Gravelle: https://github.com/simongravelle/lammps-input-files/tree/main/inputs/melting-gold. You can use the notebook workflow-hea/workflow.ipynb to initialize and simulate a cube of melting gold using the EAM potential.