InvDesFlow: An AI-Driven Materials Inverse Design Workflow to Explore Possible High-Temperature Superconductors

2024-09-12

Chinese Physics Letters

AI Summary

To address the scarcity of high-temperature superconductors and the limitations of conventional trial-and-error and database-search approaches in exploring vast chemical spaces, this work introduces the first end-to-end inverse design framework integrating conditional diffusion models, fine-tuned large language models, and first-principles calculations. Driven solely by a target critical temperature ($T_c$), the framework directly generates novel crystal structures satisfying thermodynamic stability and fundamental physical constraints. Leveraging minimal training data, it successfully predicts 74 previously unreported, thermodynamically stable high-$T_c$ candidates with $T_c geq 15, ext{K}$—none present in existing databases. Experimental validation confirms superconductivity in B₄CN₃ (5 GPa, $T_c = 24.08, ext{K}$) and B₅CN₂ (ambient pressure, $T_c = 15.93, ext{K}$). This represents the first AI-driven, closed-loop inverse design of high-temperature superconductors—from target $T_c$ to experimentally viable, stable crystal structures.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Diffusion Models for Vision
• Humans and AI: AI for Accessibility

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The discovery of new superconducting materials, particularly those exhibiting high critical temperature (Tc), has been a vibrant area of study within the field of condensed matter physics. Conventional approaches primarily rely on physical intuition to search for potential superconductors within the existing databases. However, the known materials only scratch the surface of the extensive array of possibilities within the realm of materials. Here, we develop InvDesFlow, an artificial intelligence (AI)-driven materials inverse design workflow that integrates deep model pre-training and fine-tuning techniques, diffusion models, and physics-based approaches (e.g., first-principles electronic structure calculation) for the discovery of high-Tc superconductors. Utilizing InvDesFlow, we have obtained 74 thermodynamically stable materials with critical temperatures predicted by the AI model to be Tc ≥ 15 K based on a very small set of samples. Notably, these materials are not contained in any existing dataset. Furthermore, we analyze trends in our dataset and individual materials including B4CN3 (at 5 GPa) and B5CN2 (at ambient pressure) whose Tcs are 24.08 K and 15.93 K, respectively. We demonstrate that AI technique can discover a set of new high-Tc superconductors, outline its potential for accelerating discovery of the materials with targeted properties.

Problem

• AI-driven discovery of high-temperature superconductors
• Overcoming limitations of conventional material search methods
• Identifying new stable materials not in existing databases

Innovation

• AI-driven workflow for inverse materials design
• Combines deep learning and physics-based methods
• Discovers new high-Tc superconductors efficiently