Future Challenges in High Power Technologies for Microwave and Millemeter-Wave Applications
Greg Baker has worked in the RF and microwave industry for 29 years in various roles including IC design, marketing, general management, and sales.
Mr. Baker was Senior Vice President of Strategy for RF and Microwave at MACOM before departing at the end of 2017. He also served there as Senior Vice President and General Manager of RF and Microwave Products and Vice President of International Sales.
Mr. Baker joined MACOM through their acquisition of Nitronex in 2014, a GaN on Si RF amplifier company, where he was CEO. Prior to Nitronex, Mr. Baker lived in the Netherlands working for NXP where he was General Manager of RF Small Signal Products.
Prior to NXP, Mr. Baker served in various engineering, general management, and sales roles at Mimix Broadband, Sirezna Microdevices, Fujitsu, and Motorola. Mr. Baker holds a BSEE from Texas A&M University, an MSEE from Georgia Tech, and an MS in Global Finance from HKUST/NYU Stern School of Business.
GaN-based HEMTs for RF Power Applications
Dr.Nandita DasGupta received her B.E. degree in Electronics and Telecommunication Engineering from Jadavpur University, Kolkata, India in 1982, M.Tech and Ph.D from I.I.T. Madras in 1984 and 1988 respectively.
She was awarded Alexander von Humboldt Fellowship in 1991 and spent one year in Germany doing Post-doctoral research in the area of Surface Passivation of Compound Semiconductors. She has been a Faculty member in the Department of Electrical Engineering, I.I.T. Madras since 1993 and is currently a Professor. Her research interest is in the area of Silicon and Compound Semiconductor Technology and Modelling as well as MEMS. She has nearly one hundred research publications in International Journals and Proceedings of International Conferences and has co-authored a book on Semiconductor Devices – Modelling & Technology.
With the fast growing communication technologies, the demand for RF and microwave power amplifiers is increasing. Several technologies like strained SiGe, Si Lateral Diffused MOS (LDMOS), MESFETs, HEMTs and HBTs based on GaAs and InP, SiC MESFETs and GaN based HEMTs have been developed to meet these requirements. Silicon and GaAs have lower breakdown field strength compared to GaN and hence large devices are needed to produce higher currents. However, larger devices have larger parasitic capacitance which limits their maximum operating frequency. GaN based devices are potentially free from these drawbacks due to their higher breakdown field strength and higher electron saturation velocity. The Johnson’s figure of merit which qualifies a material for RF performance is about 15 times larger for GaN than for GaAs.
GaN based HEMTs are very attractive for both high power and high frequency applications due to the favourable material properties of GaN, like large band gap, high breakdown field strength and high electron saturation velocity. While AlGaN/GaN is the most studied GaN-based material system and already has a mature growth technology on sapphire, SiC and Si substrates, lattice-matched Al0.83In0.27N/GaN is also gaining popularity due to the higher spontaneous polarization in AlInN/GaN compared to AlGaN/GaN, leading to a higher 2DEG density at the hetero-interface, even for much lower barrier thickness. This results in higher drain current, transconductance (gm) and fT. Also, as Al0.83In0.27N is lattice matched to GaN, there is no piezoelectric polarization component and hence AlInN/GaN devices do not suffer from inverse piezoelectric effect. These properties of AlInN/GaN HEMT make it more attractive than conventional AlGaN/GaN HEMT.
Despite all these advantages, there are two major challenges associated with GaN-based HEMTs : high gate leakage current (IG) due to Schottky gate contact and normally-on (Depletion-mode) device operation due to the presence of 2-DEG even at zero gate bias. To mitigate the problem of high IG in HEMTs, a gate insulator layer under the gate metal is generally added to form the metal-insulator-semiconductor HEMT (MIS-HEMT). However, commonly reported gate dielectrics used for MIS-HEMTs cause a negative shift in threshold voltage (VTh) compared to that of HEMT. For safety and reliable operation, it is preferred to have normally-off or Enhancement-mode type of device. Several techniques have been reported in literature to obtain enhancement-mode devices, such as thinning of AlGaN barrier, fluorine plasma treatment, p-GaN gate etc. However, all these techniques suffer from drawbacks like reduced 2-DEG, increased on-resistance as well as reliability issues.
We have recently reported a positive shift in the threshold voltage of AlGaN/GaN as well as AlInN/GaN MISHEMTs using reactive-ion-sputtered Al2O3 as gate dielectric. This positive shift is achieved by engineering the charge at the interface of the gate-dielectric and semiconductor. Using this technique along with reduction in barrier-layer thickness by recess etching, we have successfully demonstrated normally-off AlInN/GaN MIS-HEMTs. In addition, significant reduction in IG, higher gm and Ion/Ioff ratio have been achieved for these devices. A comprehensive analytical model has also been developed to explain this positive shift in the threshold voltage.
Overview of GaN Technologies at III-V Lab & UMS addressing Microwave and Millimeter-Wave Applications
Stéphane Piotrowicz, in charge of the GaN HEMT RF program at III-V Lab. A joint lab of Nokia Bell Labs France, Thales Research and Technology & CEA Leti. Stéphane PIOTROWICZ received the PhD. Degree in Electronics from the University of Lille in 1999 at Institute for Electronics Microelectronics and Nanotechnology (IEMN).
In 2000, he joined the Thales Research Center and worked on the design of hybrid and MMIC power amplifiers on InGaP/GaAs HBT technology for Radar and Space Applications. Since 2004, he has contributed to develop GaN technologies at III-V Lab (a joint lab of Nokia Bell Labs France, Thales Research and Technology & CEA Leti) from L-Band to E-band.
He is currently in charge of the GaN HEMT for RF applications program. His background concerns design, modeling and RF characterization at transistor and circuit levels, for power switches, power amplifiers and low noise amplifiers.
This article will show the current technological developments at III-V lab and UMS addressing microwave and millimetre waves applications.
III–N semiconductors are expected to be the best candidates in the field of high-power electronics for high frequency operations. Since few years, GaN research and industry has been moving toward higher frequency range of operation, pushed by the need of high power at millimetre-wave frequencies, mainly for telecommunication market and future 5G requirements. At technology level, for applications above 30 GHz, it becomes necessary to reduce short channel effects and obtain high transconductance to save microwave gain. To achieve these goals, as we reduce the gate length of microwave devices, it is also necessary to decrease the barrier thickness of the HEMT structure to keep a good control of the 2D gas. Several materials are possible for the barrier composition: AlGaN is the most convention one but many researches and developments are on-going worldwide on alternative HEMT structures with different barrier layers such as InAlN, InAlGaN or AlN. Preliminary studies done at III-V Lab on InAlN-based lattice matched on GaN heterostructures showed promising results up to 18 GHz, reaching 12.5 W/mm. Quaternary barrier InAlGaN HEMTs introduce some strain in comparison to In17Al83N but show higher mobilities than ternary In17Al83N, while keeping similar electron gas density. With a dedicated epitaxial heterostructure including a patented AlGaN back-barrier buffer layer associated with an optimized process especially on the passivation step, we present here an overview of the achieved performances on 0.15µm gate length technology covering the full chain of fabrication from the epitaxy to robustness assessment, confirming that InAlGaN-based technology can be very attractive for millimetre waves applications.
Nonlinear RF signal processing : old and new challenges
Raymond Quéré (F’09) received the Ph.D. degree in electrical engineering from the University of Limoges, Limoges, France, in 1989, where he was appointed as a Full Professor in 1992.
From 1998 to 2013, he led the Department of High Frequency Devices, Circuits, Signals and Systems, XLIM Laboratory, CNRS, University of Limoges, where he has been the Deputy Director of the XLIM Laboratory since 2013. His research interests include modeling and design of nonlinear circuits for telecommunications and radar systems. He has authored or co-authored over 300 publications in international journals, international, and national conferences. He was appointed as a General Chairman of the European Microwave Week in 2005. He is the holder of the Chair “Design of future integrated smart RF transceivers (DEFIS-RF)” funded by Thales Alenia Space, Thales Corporation and the French Research Agency