Drafted 2026-06-18. Each email references 1-3 specific papers and connects to a concrete Ouro resource (team, shared datasets, computation routes, or fundable quest). Sending is queued pending Resend API availability.
1. Richard G. Hennig (University of Florida)
Papers: Gibson et al., "Developing a complete AI-accelerated workflow for superconductor discovery," npj Computational Materials 12, 95 (2026); Prakash et al., "Guided diffusion for the discovery of new superconductors," npj Comput. Mater. (2026)
Connection: BEE-NET screening pipeline (0.87K Tc MAE, 99.4% TN rate) maps directly to a gap in Ouro's hosted routes. Invitation to benchmark against other tools and share the generative diffusion model.
2. Zi-Kui Liu (Penn State)
Paper: Liu & Shang, "Revealing symmetry-broken superconducting configurations by density functional theory," Superconductor Science and Technology 38, 075021 (2025)
Connection: Zentropy theory and the 5-million material database connect to shared computational screening infrastructure and ARPA-E-aligned fundable quests.
3. Christoph Heil (TU Graz)
Papers: 16-author PNAS perspective on room-temperature superconductor research agenda (March 2026); Heil et al., "Vacancy-free cubic superconducting NbN enabled by quantum...," Nature (2026)
Connection: The "link theory, simulation, experiment via AI" framing from the PNAS agenda matches exactly what Ouro aims to enable through shared datasets, computation routes, and quest-driven research objectives.
4. Jason Hattrick-Simpers & Siwoo Lee (University of Toronto / Acceleration Consortium)
Paper: Lee, Hattrick-Simpers, Kim et al., "High-Tc superconductor candidates proposed by machine learning," ML: Sci. Technol. 6, 035052 (2025)
Connection: Data-efficient similarity-based ridge regression (millisecond/material, leave-one-out across full 0-250K Tc range) offers a useful lightweight baseline against GNN-heavy approaches — natural fit for benchmarking on Ouro.
5. Yan He (Wuhan University / Songshan Lake Materials Laboratory)
Work: Interpretable deep learning framework for lattice thermal conductivity prediction, combining DFT, GNNs, and symbolic regression (vibrational free energy + bulk modulus as key parameters). Identified 4 high-performance thermal materials from thousands of candidates (October 2025).
Connection: Full screening pipeline (qualitative → quantitative → experimental validation) fits the thermoelectrics team's needs for fast, interpretable screening of thousands of candidates.
Resend email API is currently unavailable. Drafts are finalized and staged in data/sc_te_outreach_drafts.json. Sending will proceed immediately once the tool is restored.
This is part of the Ouro Outreach: Grow the Research Community quest.
On this page
Researcher outreach drafts for the #superconductors and #thermoelectrics teams — personalized emails with paper references and concrete Ouro connections. Blocked on Resend API access.