Nanofabrications of T shape gates for high electron mobility transistors in microwaves and THz waves, a review

High electron mobility transistors (HEMTs) are the basic building block in microwave monolithic integration circuits (MMICs) for broad applications in micrometer (0.3–100 GHz), millimeter (100–300 GHz) and tera [1] hertz (300 GHz–10 THz) wave. The key part in HEMT is the nanoscale T shape gates whose foot-width characterizes the frequency regime of the device, for which modern nanolithography techniques are required. Since the first T shape gate was proposed in early 1980s, which has been proved to be the best configuration among others, a big variety of fabrication processes have been developed to meet various kinds of needs by the applications. For pattern generation, various lithography techniques such as optical lithography, interference lithography, X-ray lithography, nanoimprint lithography, and electron beam lithography have been applied. For pattern transfer, film deposition and dry etch are involved. For the geometry configuration of the gates, both symmetry and asymmetry T shape gates, Y shape gates, T shape gates with broad heads and field plate gates have been proposed. In the layer stack of resists, single layer, bilayer, trilayer and multilayer, exposed by one step, two step and multistep with high precision registration, have been developed. It can be seen that the whole history of technical development for T shape gates actually reflects the advance of nanofabrication technique as a whole in the past four decades. This paper is dedicated to give an overview of the techniques established for T shape gates, aiming at categorizing various processes for different gates and giving comments on their advantages and limitations for particular application. Based on the technical foundation established so far for T shape gates applied in III-V HEMTs, the further development has been briefly discussed.In the preparation for this reviewing paper, every effort has been made to cover as many processes as possible. But, it is still impossible to address every corner of this vast nanofabrication area. For example, the current on-going work for T shape gates in GaN HEMTs is not included in this review and it is expected a better one could come to the light shortly