Reaction Graph Networks for Inorganic Synthesis Condition Prediction of Solid State Materials
Abstract
The integration of advanced machine learning (ML) techniques with density functional theory (DFT) has significantly enhanced the optimization and prediction of stable material structures. However, translating these computational predictions into successful laboratory syntheses—whether by autonomous labs or human scientists remains time- and cost-intensive due to the complex optimization of solid-state reaction parameters. In this work, we present the first application of a Reaction Graph Network (RGN) to model precursor interactions in inorganic reactions and predict synthesis conditions based on solid state reactions. Our approach enables the efficient prediction of synthesis conditions and demonstrates improvements over previous methods. This streamlines the path from computational predictions to material synthesis and offers potential to accelerate materials discovery.
Cite
Text
Prein et al. "Reaction Graph Networks for Inorganic Synthesis Condition Prediction of Solid State Materials." NeurIPS 2024 Workshops: AI4Mat, 2024.Markdown
[Prein et al. "Reaction Graph Networks for Inorganic Synthesis Condition Prediction of Solid State Materials." NeurIPS 2024 Workshops: AI4Mat, 2024.](https://mlanthology.org/neuripsw/2024/prein2024neuripsw-reaction/)BibTeX
@inproceedings{prein2024neuripsw-reaction,
title = {{Reaction Graph Networks for Inorganic Synthesis Condition Prediction of Solid State Materials}},
author = {Prein, Thorben and Rahmanian, Fuzhan and Arul, Kesava Prasad and El-Wafi, Jasmin and Fotiadis, Menelaos Panagiotis and Heimann, Jan and Weinmann, Paul and Duan, Yifei and Pan, Elton and Olivetti, Elsa and Rupp, Jennifer L.M.},
booktitle = {NeurIPS 2024 Workshops: AI4Mat},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/prein2024neuripsw-reaction/}
}