High-electron-mobility transistor

The High Electron Mobility Transistor (HEMT) emerged in July 1979 when Takashi Mimura at Fujitsu Laboratories in Atsugi, Japan, conceived a new way to make field-effect transistors faster. The breakthrough came from an unexpected source: a 1978 Applied Physics Letters article on heterojunction superlattices developed at Bell Labs. Mimura recognized that modulation doping—spatially separating conduction electrons from their parent donor impurity atoms—could dramatically increase electron mobility and thus transistor speed.

The adjacent possible for HEMT had opened through advances in compound semiconductor materials and understanding of electron behavior at material interfaces. Since 1977, Mimura had been researching gallium arsenide (GaAs) MOSFETs as alternatives to silicon devices. The fastest transistor before HEMT was the GaAs Metal-Semiconductor Field Effect Transistor (MESFET), invented in 1966. In MESFETs, impurities added to supply electrons also scattered those electrons, limiting mobility. Mimura's insight was that the electrons and the impurities didn't need to be in the same place.

The HEMT was the first transistor to incorporate an interface between two semiconductor materials with different energy gaps—in the original design, an n-AlGaAs/GaAs heterostructure. Electrons accumulated in a 'two-dimensional electron gas' at this interface, free from the ionized impurities that would normally slow them down. In Mimura's 1980 demonstration paper, the electron mobility at 77K was 5.5 times higher than MESFET, and transconductance was 3 times higher.

Convergent emergence characterized HEMT development. Independently, Daniel Delagebeaudeuf and Tranc Linh Nuyen at Thomson-CSF in France filed a patent for a similar device in March 1979. Both groups built on Ray Dingle's 1978 Bell Labs work on modulation doping, demonstrating how theoretical advances at one laboratory could enable practical inventions at multiple others nearly simultaneously.

The cascade from HEMT transformed telecommunications infrastructure. HEMTs proved superior for high-speed, high-frequency applications: radio telescopes, satellite broadcasting receivers, cellular base stations, and wireless communication systems. In 1990, Mimura and crystal-growth collaborator Satoshi Hiyamizu won the IEEE Morris N. Liebmann Memorial Award. The technology became fundamental infrastructure for the wireless communication revolution.

By 2024, HEMT derivatives power everything from 5G base stations to automotive radar systems. The original insight—that keeping electrons away from their donor atoms could make them faster—became a cornerstone of high-frequency electronics. Mimura's 1979 conception, building on Bell Labs theory and enabled by Fujitsu's compound semiconductor capabilities, opened an adjacent possible that shaped how humanity communicates wirelessly.