Design Topological Materials by Reinforcement Fine-Tuned Generative Model

2025-04-17

AI Summary

The scarcity of full-bandgap topological insulators (TIs) and topological crystalline insulators (TCIs), coupled with the difficulty of conventional screening and generative methods in simultaneously satisfying topological nontriviality and thermodynamic stability, poses a major bottleneck in discovering practical topological materials. Method: We propose a novel graph neural network (GNN)-based generative paradigm leveraging reinforcement learning fine-tuning (ReFT)—the first application of ReFT to materials generation—explicitly optimizing dual objectives: topological invariants (Z₂ invariant, mirror Chern number, bandgap) and DFT-calculated formation energy. Our framework integrates a pre-trained GNN generator, high-throughput first-principles validation, and structural stability assessment. Results: We successfully design a series of new full-bandgap topological materials; notably, Ge₂Bi₂O₆ exhibits a bandgap of 0.26 eV—the largest reported to date—significantly expanding the library of experimentally viable topological materials.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Topological insulators (TIs) and topological crystalline insulators (TCIs) are materials with unconventional electronic properties, making their discovery highly valuable for practical applications. However, such materials, particularly those with a full band gap, remain scarce. Given the limitations of traditional approaches that scan known materials for candidates, we focus on the generation of new topological materials through a generative model. Specifically, we apply reinforcement fine-tuning (ReFT) to a pre-trained generative model, thereby aligning the model's objectives with our material design goals. We demonstrate that ReFT is effective in enhancing the model's ability to generate TIs and TCIs, with minimal compromise on the stability of the generated materials. Using the fine-tuned model, we successfully identify a large number of new topological materials, with Ge$_2$Bi$_2$O$_6$ serving as a representative example--a TI with a full band gap of 0.26 eV, ranking among the largest known in this category.

Problem

• Generate new topological materials with full band gaps
• Enhance generative model for TIs and TCIs using ReFT
• Discover stable topological materials like Ge2Bi2O6

Innovation

• Reinforcement fine-tuning for generative models
• Generating topological materials with full band gap
• Enhanced model alignment with design goals