Plasma and transit-time mechanisms of the terahertz radiation detection in high-electron-mobility transistors

A. Satou, J. Khmyrova, V. Ryzhii, M. S. Shur

Research output: Contribution to journalArticlepeer-review

67 Citations (Scopus)


We develop a device model for a high-electron-mobility transistor (HEMT) affected by the incoming terahertz radiation. The model takes into account the electron plasma oscillations in the HEMT channel, tunnelling of electrons from the channel into the gate layer and electron transit-time effects in this layer. It is shown that the excitation of plasma oscillations accompanied by the delay in the electron propagation across the gate layer and a strong nonlinearity of the tunnelling current can result in significant features of the HEMT high-frequency linear and nonlinear characteristics. We derive a formula for the HEMT gate-to-source/drain admittance. We also calculate the variation of the dc current induced by the terahertz radiation and the HEMT detection responsivity. It is found that the detection responsivity exhibits sharp resonant peaks corresponding to the frequencies of plasma oscillations. The resonant plasma frequencies and the positions of the admittance and detection responsivity peaks depend on the gate length and the lengths of the contact regions (source-to-gate and gate-to-drain spacings) and can be tuned by the gate voltage. The coincidence of the plasma and transit-time resonances can lead to a marked sharpening of the responsivity peaks.

Original languageEnglish
Pages (from-to)460-469
Number of pages10
JournalSemiconductor Science and Technology
Issue number6
Publication statusPublished - 2003 Jun 1
Externally publishedYes

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Electrical and Electronic Engineering
  • Materials Chemistry


Dive into the research topics of 'Plasma and transit-time mechanisms of the terahertz radiation detection in high-electron-mobility transistors'. Together they form a unique fingerprint.

Cite this