Thermal Characterisation of GaN HEMTs Grown on SiC Substrate for High Frequency Applications

Abstract

The DC, RF, noise and nonlinearity performance of the AlGaN/GaN high electron mobility transistor (HEMT) with a 150 nm and 250 nm gate length on a SiC substrate over a temperature range of - 40 to 150°C including experimental characterisation, modelling and analysis for high frequency application have been investigated. All the DC, small signal parameter and noise parameter variations with ambient temperature on the same set of devices have been reported for the first time using on-wafer measurements up to 50 GHz. The temperature coefficient of all the DC, small signal parameters as well as ft and fmax including noise parameters are reported. Some of the extracted equivalent circuit parameters with the theoretical data of the evolution of electrical parameters and the relevant physical equations involved have been compared using the same biasing condition for further accuracy. The theoretical results are shown to be consistent with the extracted data. Some results are also experimentally verified with previous works. Anomaly and defects characterisation such as kink effect, threshold voltage shifts with temperature are also presented using I-V and the published data. To examine the performance of the nonlinearity in terms of multi biasing, input power, frequency, and temperature of these devices, the analysis of two-tone intermodulation distortion (IMD) is employed. A temperature-dependent analytical model taking into account of multi bias, input power and frequency for the nonlinearity of the device has been developed. The modelled data are consistent with the two-tone IMD measurements providing an effective means for analyzing these devices. The overall results are important in meeting the circuit design specifications‟ challenges over a wide range of temperature and frequency.

Date of Award	31 Dec 2016
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Zhipeng Wu (Supervisor) & Ali Rezazadeh (Supervisor)