Room temperature magnetic semiconductors discovery

This is a research write-up by Vals AI describing how a team of Claude Opus 5.5 agents, working with a human researcher, searched for materials useful for spintronic memory. The target was a Luttinger-compensated antiferromagnet: a semiconductor with zero net magnetisation, but with spin-up and spin-down states separated by energy at the band edges, as in a ferromagnet.

The agents ran density functional theory simulations at two levels, PBE+U for speed and HSE06 for greater accuracy. The reported band gaps and spin windows come from HSE06. The first candidate, YBaMnFeO₅, was designed by the agents. According to the post, it has not been made or proposed as this kind of magnet before. It is predicted to have a 2.35 eV band gap, with spin windows of about 1.0 eV for holes and 1.4 eV for electrons. These are compared against room-temperature thermal energy of about 26 meV. A second candidate is a compound first synthesised in 1999 that the calculations predict has the desired properties.

The results are computational predictions only and have not been confirmed by experiment, as the post and creator coverage note. The post also includes calculations, code and caveats. It is a blog report, not a product or model release, so there is nothing to subscribe to or download from the provided material.