We report on room temperature terahertz generation by a submicron size AlGaN/GaN-based high electron mobility transistors. The emission peak is found to be tunable by the gate voltage between 0.75 and 2.1 THz. Radiation frequencies correspond to the lowest fundamental plasma mode in the gated region of the transistor channel. Emission appears at a certain drain bias in a thresholdlike manner. Observed emission is interpreted as a result of Dyakonov-Shur plasma wave instability in the gated two-dimensional electron gas.

Room-temperature generation of terahertz radiation in nanometer gate length InAlAs∕InGaAs and AlGaN∕GaN high-mobility transistors is reported. A well-defined source-drain voltage threshold for the emission exists, which depends on the gate bias. Spectral analysis of the emitted radiation is presented. The highest emission power emitted from a single device reached 0.1μW.

In order to assess possible mechanisms of gate reverse-bias leakage current in AlInN/GaN high electron mobility transistors (HEMTs) grown by metalorganic chemical-vapor deposition on SiC substrates, temperature-dependent current-voltage measurements combined with Fourier transform current deep level transient spectroscopy (FT-CDLTS) are performed in the temperature range of 200–400 K. In this range of temperature reverse-bias leakage current flow is found to be dominated by Poole–Frenkel emission. Based on CDLTS measurements, a model of leakage current transport via a trap state located at the AlInN/metal interface with an activation energy of 0.37 eV is suggested. The trap nature is shown to be an extended trap, most probably associated with dislocations in the AlInN barrier layer.

No abstract

Mg + ions were implanted at room temperature in n-type hexagonal GaN for the device isolation purposes. The implantation dose varied from 7.5 × 10 12 to 10 16 ions cm −2 . We performed resonance Raman spectroscopy and DC electrical measurements in order to monitor the structural and electrical changes of non-annealed and annealed implanted GaN samples. Annealing was carried out at 900 • C for 30 s, these conditions being used to achieve good Ohmic contacts. The aim was to determine, on the one hand, the influence of ion doses on the device isolation and, on the other, to establish the order of the technological steps which should be made between ion implantation and Ohmic contact annealing. On increasing the implantation dose from 7.5 × 10 12 to 2 × 10 14 ions cm −2 , an increase in the electrical isolation and a decrease in the photoluminescence (PL) were observed. For the highest dose, the implanted layer became conductive owing to a hopping mechanism and only the first-order phonon lines remained observable. After annealing, the implanted samples became conductive and the PL reappeared or increased compared with the nonannealed samples at same implantation doses, except for the sample implanted at the highest dose, which became insulating. Then, it is possible to achieve device electrical isolation by using a lower ion dose without thermal annealing or using a higher ion dose with thermal annealing.