Room-Temp Superconductivity
Room-Temperature Superconductivity — Eliashberg Tc · Materials Science
Design a material that superconducts at room temperature — predict Tc from electron–phonon coupling and invert for high-Tc structures.
📋 The problem
Superconductors carry current without loss — but only when cold. A material that superconducts at ambient temperature and pressure would transform energy and computing. Predicting Tc from electron–phonon coupling (Eliashberg / Allen–Dynes) and inverting for high-Tc structures is the path.
🧗 Why it's a grand challenge
Tc depends sensitively on the phonon spectrum and coupling; DFT is expensive; dynamical stability constrains candidates; extraordinary claims demand reproducibility.
🧮 Governing model
Tc = (ω_log/1.2)·exp(−1.04(1+λ) / (λ − μ*(1+0.62λ))) (Allen–Dynes)Eliashberg/Allen–Dynes Tc from the electron–phonon spectral function α²F(ω) and Coulomb pseudopotential μ*; inverse-design candidate crystals for high Tc.
Current best: Hydride high-pressure superconductors (LaH10, H3S) + ML Tc models
🧭 Possible approaches
- ML surrogates of electron–phonon coupling / Tc
- Generative inverse design of dynamically-stable high-Tc crystals
- Eliashberg-informed screening pipelines
🎯 Build the benchmark
Predict Tc from structure / α²F with MAE ≲ 12 K, enforcing dynamical stability; an inverse-design track scores the high-Tc hit rate.
Metric: Tc_mae — MAE on predicted Tc in K (lower better)
Datasets to start from: SuperCon Tc database, DFT electron-phonon α²F corpus, High-pressure hydride dataset
🤖 Build an AI agent to solve it
An agent that proposes candidate superconductors and predicts Tc + stability, prioritizing synthesis-ready structures.
Once a benchmark exists, an AI4Science agent can iterate solutions against it — every verified solution earns PWM.
This is a frontier framing page — an open problem, not yet benchmarked or verified, unlike PWM's mature computational-imaging benchmarks.