AI-Assisted Rapid Crystal Structure Generation Towards a Target Local Environment

2025-06-09

AI Summary

To address key bottlenecks in crystal structure prediction—namely low efficiency, limited system scale, and inadequate symmetry modeling—this work proposes LEGO-xtal, a symmetry-aware local geometric generation framework. LEGO-xtal innovatively jointly encodes space-group symmetry and local coordination geometry via a symmetry-constrained graph neural network, bypassing traditional energy-based optimization. Instead, it employs machine-learned structural descriptors to guide lightweight post-generation refinement and supports the generation of supercells containing hundreds of atoms. Trained on small-scale datasets augmented with symmetry-aware data expansion, the model discovers over 1,700 novel structures—within 0.5 eV/atom accuracy—from 25 known sp²-carbon allotropes. Moreover, it demonstrates strong generalizability to modular material systems, including metal–organic frameworks (MOFs) and solid-state battery materials.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

In the field of material design, traditional crystal structure prediction approaches require extensive structural sampling through computationally expensive energy minimization methods using either force fields or quantum mechanical simulations. While emerging artificial intelligence (AI) generative models have shown great promise in generating realistic crystal structures more rapidly, most existing models fail to account for the unique symmetries and periodicity of crystalline materials, and they are limited to handling structures with only a few tens of atoms per unit cell. Here, we present a symmetry-informed AI generative approach called Local Environment Geometry-Oriented Crystal Generator (LEGO-xtal) that overcomes these limitations. Our method generates initial structures using AI models trained on an augmented small dataset, and then optimizes them using machine learning structure descriptors rather than traditional energy-based optimization. We demonstrate the effectiveness of LEGO-xtal by expanding from 25 known low-energy sp2 carbon allotropes to over 1,700, all within 0.5 eV/atom of the ground-state energy of graphite. This framework offers a generalizable strategy for the targeted design of materials with modular building blocks, such as metal-organic frameworks and next-generation battery materials.

Problem

• Overcoming limitations in AI-generated crystal structures for materials design
• Addressing lack of symmetry and periodicity in existing generative models
• Enabling rapid generation of large-unit-cell crystal structures efficiently

Innovation

• Symmetry-informed AI generative approach
• Machine learning structure descriptors optimization
• Generates structures from augmented small dataset