DiffSyn: a generative diffusion approach to materials synthesis planning.

Pan, Elton; Kwon, Soonhyoung; Liu, Sulin; Xie, Mingrou; Hoffman, Alexander J; Duan, Yifei; Prein, Thorben; Sheriff, Killian et al. · Nat Comput Sci · 2026

basic_science · Level V

Where this comes from

Abstract

The synthesis of crystalline materials, such as zeolites, remains a notable challenge owing to a high-dimensional synthesis space, intricate structure-synthesis relationships and time-consuming experiments. Here, considering the 'one-to-many' relationship between structure and synthesis, we propose DiffSyn, a generative diffusion model trained on over 23,000 synthesis recipes that span 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/Al<sub>ICP</sub> of 19.0, which is expected to improve thermal stability.