A Generative Diffusion Model for Amorphous Materials
This work addresses the problem of designing and simulating amorphous materials for researchers in materials science, offering a significant speed-up and accuracy improvement over traditional methods.
The paper tackled the challenge of generating amorphous materials, which are complex structures not effectively handled by existing generative models, by presenting a diffusion model that produces these structures up to 1000 times faster than conventional simulations while accurately recovering their order and properties. The result includes validated applications to silica glass and metallic glasses, enabling new insights like ductile-to-brittle transitions and mesoporous structures.
Generative models show great promise for the inverse design of molecules and inorganic crystals, but remain largely ineffective within more complex structures such as amorphous materials. Here, we present a diffusion model that reliably generates amorphous structures up to 1000 times faster than conventional simulations across processing conditions, compositions, and data sources. Generated structures recovered the short- and medium-range order, sampling diversity, and macroscopic properties of silica glass, as validated by simulations and an information-theoretical strategy. Conditional generation allowed sampling large structures at low cooling rates of 10$^{-2}$ K/ps to uncover a ductile-to-brittle transition and mesoporous silica structures. Extension to metallic glassy systems accurately reproduced local structures and properties from both computational and experimental datasets, demonstrating how synthetic data can be generated from characterization results. Our methods provide a roadmap for the design and simulation of amorphous materials previously inaccessible to computational methods.