Two kagome superconductors predicted by machine learning and confirmed in the lab
A screening pipeline that narrowed over 1.3 million candidate structures picked out YRu3B2 and LuRu3B2, which were then synthesized and measured to superconduct at 0.81 K and 0.95 K.
- Model
- Unknown
- Field
- Materials science
- Date
- 2026-06-17
- Human collaborators
- Rose Albu Mustaf, Päivi Törmä, Emilia Morosan, B. Andrei Bernevig, Miguel A. L. Marques
What was found
The SuperC consortium used machine-learning pre-screening to cut a very large chemical space down to a tractable candidate list, then ran first-principles calculations on what survived, arriving at 741 dynamically and thermodynamically stable compounds with DFT-predicted Tc above 5 K. Morosan's group at Rice synthesized two of them and confirmed bulk superconductivity by magnetization and specific heat measurement. Both compounds carry a kagome lattice, in which electrons form flat bands.
Novelty check
Machine-learning screening of superconductor candidates is an established and crowded field, and the JHU APL work on novel superconductors predates this. What is claimed as new is the end-to-end path: candidates identified from scratch by a machine-learning-guided pipeline, then synthesized and experimentally confirmed, rather than screening validated only against already-known superconductors. YRu3B2 and LuRu3B2 were not previously reported as superconductors. Whether this is the first such end-to-end confirmation is the consortium's framing and is not independently adjudicated here.
Caveats and known objections
The gap between prediction and measurement is the story the coverage mostly buries: the screen selected for DFT-predicted Tc above 5 K, and the two compounds actually measured at 0.81 K and 0.95 K, roughly a factor of five to six low. These are sub-1-kelvin superconductors, nowhere near room temperature, and the consortium's stated goal of a room-temperature superconductor by 2033 is a research programme, not a result. Two confirmed compounds out of 741 stable candidates is also not a hit rate the paper establishes as generalizable. The specific machine-learning method is not named in the coverage consulted, so model is recorded as Unknown.
Nobody outside the lab has checked this yet.
Reading the primary source closely enough to say whether it supports the claim counts as a check, and you are credited on the entry.
Or on GitHub: submit a check challenge the grade send a correction or send a pull request
Flag this for triage
Signals order the review queue and nothing else. They are never published, and they never move a grade: that takes a citation.
Entry history (1 event)
- AddedEntered the registry graded Peer reviewed and Search scaffold.
Entries are never deleted. A grade that does not hold up is downgraded on the record, with the reason beside it.
Graded peer reviewed for verification and search scaffold for autonomy. What these mean.
Cite this entry
whataifound.org. (2026). Two kagome superconductors predicted by machine learning and confirmed in the lab. whataifound.org: A Registry of AI Scientific and Mathematical Discoveries. https://whataifound.org/finding/2026-06-17-kagome-superconductorsBibTeX
@misc{whataifound-aaltouniversity-2026-superconductors,
title = {Two kagome superconductors predicted by machine learning and confirmed in the lab},
author = {{whataifound.org}},
year = {2026},
howpublished = {whataifound.org: A Registry of AI Scientific and Mathematical Discoveries},
note = {Result by Aalto University / Rice University. Verification: Peer reviewed. Autonomy: Search scaffold.},
url = {https://whataifound.org/finding/2026-06-17-kagome-superconductors}
}