Researchers at Cornell University have built a working field-effect transistor from an oxide-semiconductor family with an exceptionally wide electronic bandgap. The team grew silicon-doped aluminum gallium oxide films with measured bandgaps exceeding 7 electron volts (eV), then demonstrated both a Schottky diode and an experimental transistor called the AlphaFET. Cornell announced the development on November 3, 2026, following the study’s October 7 publication in Nature. The result marks a laboratory materials-and-device milestone, not a commercial high-voltage chip ready for installation. Cornell’s research announcement explains the experiment.
Why the bandgap matters
A bandgap is the energy separating electronic states occupied in a material from states that carry electrical current. Wider-gap semiconductors can theoretically tolerate stronger electrical fields before breakdown, a key property for efficient power switching. Silicon’s gap is about 1.1 eV; familiar wide-bandgap alternatives include silicon carbide and gallium nitride. Cornell’s peer-reviewed study pushes well into the colossal bandgap range above 6 eV. Our lesson on how electronic bands form covers the underlying physics.
EDS at Cornell seminar by Arizona State University’s Nidhin Kalarickal, April 2025, discussing gallium-oxide power electronics. Archival technical background, not a recording of the 2026 AlphaFET experiment. Original YouTube seminar.
Making a near-insulator conduct electricity
The obstacle is that very wide-bandgap materials are often extremely difficult to dope into useful conductors. Cornell doctoral researcher Jacob Steele used suboxide molecular-beam epitaxy, a controlled crystal-growth technique, to deposit aluminum gallium oxide on sapphire. Silicon atoms added during growth supplied mobile electrons. In samples above the 6-eV threshold, the researchers measured room-temperature conductivities more than 100 million times higher than earlier results in that specialized material category. The comparison is conductivity against prior colossal-gap reports, not 100 million times greater computer speed or energy efficiency.

The team fabricated a rectifying Schottky diode and a gate-controlled transistor. The AlphaFET showed more than four orders of magnitude of current modulation, meaning an on/off change exceeding a factor of 10,000 in the preliminary measurements. The transistor proves electrostatic switching in an extremely wide-gap channel. However, the highest reported film bandgap and every demonstrated device metric should not be assumed to describe the identical sample. A 2026 IEEE-listed technical seminar discusses devices around the 6.7-eV regime.
A candidate for future electrical infrastructure
Silicon-carbide and gallium-nitride devices already compete to reduce losses in electric vehicles, renewable-energy inverters and dense computing. Cornell’s material could eventually widen the design choices for demanding switching environments, but mature products need reproducible wafers, contacts, high-current operation, manageable heat flow, reliability data and manufacturing scale. Compare the research with Infineon’s working 300-mm silicon-carbide wafers, larger aluminum-nitride substrates and GaN proposals for AI-rack power.
EDS at Cornell technical seminar by Bennett Cromer, February 2025, on realistic performance limits for ultrawide-bandgap power devices. Earlier educational material, not an AlphaFET field demonstration. Watch the source video.
Sapphire substrates offer a possible route toward larger research wafers, but Cornell also reported substantial crystal defects. The researchers expect further optimization before meaningful commercial power-device comparisons. The credible conclusion is narrower than a claim that silicon has been replaced: engineers demonstrated electronic control in a previously difficult bandgap regime. The broader materials landscape is discussed in our semiconductor technology outlook.
Original sources and media
- Cornell Chronicle, November 3: study, methods and device results
- Nature, October 7: A 7-eV bandgap semiconductor based on silicon-doped aluminum gallium oxide
- PubMed: research abstract and publication record
- Cornell technology licensing coverage, November 17
- EDS at Cornell: gallium oxide power electronics seminar, 2025
- EDS at Cornell: ultrawide-bandgap device limits seminar, 2025
Visual and testing disclosure: The 1200 × 630 colored-pencil featured cover is an original conceptual depiction of laboratory probing, not the actual AlphaFET chip. The approved BitcoinVersus.tech green-on-black watermark appears once at bottom-left without redesign. The Cornell wafer photograph is archival. Both seminars predate the research paper, and actual embedded-video playback requires a separate browser test.
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. If you would like to support to help further secure the integrity of our research initiative, please scan the QR code below.


Leave a Reply