Cellular Monte Carlo study of RF short-channel effects, effective gate length, and aspect ratio in GaN and InGaAs HEMTs

Abstract

An investigation of RF short-channel effects in state-of-the-art GaN and InGaAs HEMTs, in relation to effective gate length and aspect ratio, is performed through our full band Cellular Monte Carlo simulator. In particular, the short-circuit current gain cut-off frequency, f <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</sub> , is extracted using two different methods for several gate lengths. The first method relates f <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</sub> to the electron transit time in the gate region, and from the electron velocity profile allows a direct estimation of f <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</sub> , the effective gate length L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">eff</sub> , and the investigation of the nananoscale carrier dynamics in the channel. The second extraction methods derives f <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</sub> through small-signal analysis. Our results indicates that the increasing difference between effective gate length and metallurgical gate length, as the device is scaled, plays a major role in limiting the RF performance. Moreover, maintaining a minimum aspect ratio of 5 for InGaAs HEMTs, and 10 for GaN devices, helps mitigating the short-channel effects.