This repository contains the codes necessary to perform a chemical-motif characterization of short-range order, as described in our Quantifying chemical short-range order in metallic alloys paper and our Chemical-motif characterization of short-range order using E(3)-equivariant graph neural networks paper.
This framework allows for correlating any per-atom property to their local chemical motif. It also allows for the determination of predictive short-range chemical fluctuations length scale. It is based on E(3)-equivariant graph neural networks. Our framework has 100% accuracy in the identification of any motif that could ever be found in an fcc, bcc, or hcp solid solution with up to 5 chemical elements.
# To install the latest PyPi release
pip install --upgrade chemicalmotifidentifier
# To install the latest git commit
pip install --upgrade git+https://github.com/killiansheriff/ChemicalMotifIdentifier.git
You will also need to install torch
, torch_scatter
and torch_geometric
.
A jupyter notebook presenting a few test cases can be found in the examples/ folder.
If you use this repository in your work, please cite:
@article{sheriff2023quantifying,
title={Quantifying chemical short-range order in metallic alloys},
author={Sheriff, Killian and Cao, Yifan and Smidt, Tess and Freitas, Rodrigo},
journal={arXiv},
year={2023},
doi={10.48550/arXiv.2311.01545}
}
and
@article{sheriff2024chemicalmotif,
title={Chemical-motif characterization of short-range order with E(3)-equivariant graph neural networks},
author={Killian Sheriff and Yifan Cao and Rodrigo Freitas},
journal={arXiv},
year={2024},
doi={10.48550/arXiv.2405.08628}