# Developing a complete AI-accelerated workflow for superconductor discovery

Article · Jason Gibson, Ajinkya C. Hire, Pawan Prakash, Philip M. Dee, Benjamin Geisler, Jung Soo Kim, Z. Li, J. J. Hamlin, Gregory R. Stewart, P. J. Hirschfeld, Richard G. Hennig

- Archive id: arxiv:2503.20005
- Published: 2026-01-27
- Venue: npj Computational Materials
- DOI: https://doi.org/10.1038/s41524-026-01964-8
- arXiv: https://arxiv.org/abs/2503.20005
- Citations: 3
- Source URL: https://doi.org/10.1038/s41524-026-01964-8
- PDF: https://www.nature.com/articles/s41524-026-01964-8_reference.pdf
- Provenance: openalex+arxiv

## Abstract

The quest to identify new superconducting materials with enhanced properties is hindered by the prohibitive cost of computing electron-phonon spectral functions, severely limiting the materials space that can be explored. Here, we introduce a Bootstrapped Ensemble of Equivariant Graph Neural Networks (BEE-NET), a machine-learning model trained to predict the Eliashberg spectral function and superconducting critical temperature with a mean-absolute-error of 0.87 K relative to DFT-based Allen-Dynes calculations. Intriguingly, BEE-NET achieves a true-negative-rate of 99.4\\%, enabling highly efficient screening for the rare property of superconductivity. Integrated into a multi-stage, AI-accelerated discovery pipeline that incorporates elemental-substitution strategies and machine-learned interatomic potentials, our workflow reduced over 1.3 million candidate structures to 741 dynamically and thermodynamically stable compounds with DFT-confirmed $T\_{\\mathrm{c}} \> 5$ K. We report the successful synthesis and experimental confirmation of superconductivity in two of these previously unreported compounds. This study establishes a data-driven framework that integrates machine learning, quantum calculations, and experiments to systematically accelerate superconductor discovery.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [Computational & AI-Driven Superconductor Discovery](https://space-kitty.com/topics/superconductor-discovery)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.