Nanometer-Scale III-V Transistors: From THz to CMOS

Dr. Taewoo Kim

University of Ulsan, Korea

 

Abstract

Over the last 30 years, Si CMOS scaling has been the cornerstone of the microelectronic revolution. While it remains a matter of considerable debate, the semiconductor device technology that fuels this microelectronics revolution appears to be reaching the end of its use. The attention that propelled CMOS to experience the exponential growth in density, power, and speed that underpins Moore's law is quickly waning. This looming scenario has spurred interest in identifying alternative transistor technologies with performance potential that substantially improve upon Si CMOS. Among these, III-V and III-nitride compound semiconductors are strong candidates. With room temperature bulk electron mobilities that span from approximately 7,000 cm2/V-s for GaAs to about 30,000 cm2/V-s for InSb, these materials promise a significant enhancement in electron velocity that can no longer be obtained from Si. Among the III-Vs and III-nitride semiconductors, indium gallium arsenide (InGaAs) devices that contain compositions closely lattice-matched to InP appear rather unique. With an electron mobility that exceeds 10,000 cm2/V-s at 300 K and a rather mature processing technology, InGaAs transistors are based on InP and include heterostructure-bipolartransistors (HBTs) and high-electron-mobility-transistors (HEMTs); they have held the record frequency response for the highest-cut-off frequency (fT) transistors for nearly 20 years. In addition, GaN-based HEMTs have excellent breakdown characteristics in terms of DC and RF applications. HEMTs are suitable for high-power applications and have been extensively used to demonstrate > 100 GHz and > 100 Gb/s communication ICs with SSI-level complexity and reasonable reliability. Therefore, InGaAs and GaN represent the best balance between performance and maturity. This talk will summarize recent progress in our quest to map out the potential of III-V compound semiconductor for logic, identify issues of relevance to future III-V and III-Nitride transistors for high frequency and quantum computing applications, and propose innovative schemes to realize III-V/III-Nitride on Si by means of “Heterogeneous Integration”.

 

Speaker Biography

Dr. Kim majored in electrical engineering and computer science and then earned a PhD degree at Gwangju Institute of Science and Technology in South Korea in August 2008. After that, he worked as a postdoctoral research associate at the Massachusetts Institute of Technology (MIT) for 3 years. Next, he was with SEMATECH, INC., where he oversaw III-V CMOS development, including non-planar FinFET, TFET, and vertical nanowire MOSFETs for future high-speed and logic applications. In his next role, he worked at Samsung Austin Semiconductor, where he was the 14-nm FEOL module leader in the process integration group. Currently, he is employed at the University of Ulsan as an Associate Professor. His research accomplishments include 5 academic outstanding research awards/honors, 65 international presentations, 54 international publications, and 1 book chapter.