English

DiffSyn: A Generative Diffusion Approach to Materials Synthesis Planning

Materials Science 2025-09-26 v2 Artificial Intelligence Machine Learning

Abstract

The synthesis of crystalline materials, such as zeolites, remains a significant challenge due to a high-dimensional synthesis space, intricate structure-synthesis relationships and time-consuming experiments. Considering the one-to-many relationship between structure and synthesis, we propose DiffSyn, a generative diffusion model trained on over 23,000 synthesis recipes spanning 50 years of literature. DiffSyn generates probable synthesis routes conditioned on a desired zeolite structure and an organic template. DiffSyn achieves state-of-the-art performance by capturing the multi-modal nature of structure-synthesis relationships. We apply DiffSyn to differentiate among competing phases and generate optimal synthesis routes. As a proof of concept, we synthesize a UFI material using DiffSyn-generated synthesis routes. These routes, rationalized by density functional theory binding energies, resulted in the successful synthesis of a UFI material with a high Si/AlICP_{\text{ICP}} of 19.0, which is expected to improve thermal stability and is higher than that of any previously recorded.

Keywords

Cite

@article{arxiv.2509.17094,
  title  = {DiffSyn: A Generative Diffusion Approach to Materials Synthesis Planning},
  author = {Elton Pan and Soonhyoung Kwon and Sulin Liu and Mingrou Xie and Alexander J. Hoffman and Yifei Duan and Thorben Prein and Killian Sheriff and Yuriy Roman-Leshkov and Manuel Moliner and Rafael Gomez-Bombarelli and Elsa Olivetti},
  journal= {arXiv preprint arXiv:2509.17094},
  year   = {2025}
}
R2 v1 2026-07-01T05:48:18.153Z