Building-Block Aware Generative Modeling for 3D Crystals of Metal Organic Frameworks

2025-05-13

AI Summary

To address the challenges of repetitive building-block usage, small-unit-cell constraints, and limited chemical diversity in MOF 3D structure generation, this work introduces the first fully atomistic SE(3)-equivariant diffusion generative model tailored for MOFs. The model explicitly decouples topological networks from metal nodes and organic linkers—eliminating reliance on predefined building blocks or fixed unit-cell dimensions—and enables end-to-end generation of geometrically valid crystals with >1,000 atoms. Innovatively integrating crystallographic topology encoding and CoRE-MOF database–driven training, it achieves controllable generation of experimentally unreported MOFs with unprecedented novelty and structural diversity. A designed prototype, [Zn(1,4-TDC)(EtOH)₂], was validated via XRD, TGA, and N₂ adsorption measurements, confirming structural fidelity. Generated structures exhibit >95% geometric validity at the kilo-atom scale, with significantly higher novelty and database coverage than state-of-the-art methods.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Metal-organic frameworks (MOFs) marry inorganic nodes, organic edges, and topological nets into programmable porous crystals, yet their astronomical design space defies brute-force synthesis. Generative modeling holds ultimate promise, but existing models either recycle known building blocks or are restricted to small unit cells. We introduce Building-Block-Aware MOF Diffusion (BBA MOF Diffusion), an SE(3)-equivariant diffusion model that learns 3D all-atom representations of individual building blocks, encoding crystallographic topological nets explicitly. Trained on the CoRE-MOF database, BBA MOF Diffusion readily samples MOFs with unit cells containing 1000 atoms with great geometric validity, novelty, and diversity mirroring experimental databases. Its native building-block representation produces unprecedented metal nodes and organic edges, expanding accessible chemical space by orders of magnitude. One high-scoring [Zn(1,4-TDC)(EtOH)2] MOF predicted by the model was synthesized, where powder X-ray diffraction, thermogravimetric analysis, and N2 sorption confirm its structural fidelity. BBA-Diff thus furnishes a practical pathway to synthesizable and high-performing MOFs.

Problem

• Generates diverse and novel 3D MOF structures efficiently
• Overcomes limitations of small unit cells and known building blocks
• Expands accessible chemical space with unprecedented metal-organic components

Innovation

• SE(3)-equivariant diffusion model for 3D MOFs
• Generates novel metal nodes and organic edges
• Samples large unit cells with 1000 atoms