• Journal of the Korea Institute of Military Science and Technology
• /
• v.12 no.4
• /
• pp.483-490
• /
• 2009

In this paper, a novel low-damage silicon nitride passivation for 100nm InAlAs/InGaAs MHEMTs has been developed using remote ICPCVD. The silicon nitride deposited by ICPCVD showed higher quality, higher density, and lower hydrogen concentration than those of silicon nitride deposited by PECVD. In particular, we successfully minimized the plasma damage by separating the silicon nitride deposition region remotely from ICP generation region, typically with distance of 34cm. The silicon nitride passivation with remote ICPCVD has been successfully demonstrated on GaAs MHEMTs with minimized damage. The passivated devices showed considerable improvement in DC characteristics and also exhibited excellent RF characteristics($f_T$of 200GHz).The devices with remote ICPCVD passivation of 50nm silicon nitride exhibited 22% improvement(535mS/mm to 654mS/mm) of a maximum extrinsic transconductance($g_{m.max}$) and 20% improvement(551mA/mm to 662mA/mm) of a maximum saturation drain current ($I_{DS.max}$) compared to those of unpassivated ones, respectively. The results achieved in this work demonstrate that remote ICPCVD is a suitable candidate for the next-generation MHEMT passivation technique.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.295-298
• /
• 1999

AlGaAs/InGaAs power P-HEMTS (Pseudo-morphic High Electron Mobility Transistors) with 1.0-${\mu}{\textrm}{m}$ gate length for PCS applications have been fabricated. We adopted single heterojunction P-HEMT structure with two Si-delta doped layer to obtain higher current density. It exhibits a maximum current density of 512㎃/mm, an extrinsic transconductance of 259mS/mm, and a gate to drain breakdown voltage of 12.0V, respectively. The device exhibits a power density of 657㎽/mm, a maximum power added efficiency of 42.1%, a linear power gain of 9.85㏈ respectively at a drain bias of 6.0V, gate bias of 0.6V and an operation frequency of 1.765㎓.

• ETRI Journal
• /
• v.24 no.3
• /
• pp.190-196
• /
• 2002

This paper presents millimeter wave monolithic microwave integrated circuit (MMIC) low noise amplifiers using a $0.15{\mu}m$ commercial pHEMT process. After carefully investigating design considerations for millimeter-wave applications, with emphasis on the active device model and electomagnetic (EM) simulation, we designed two single-ended low noise amplifiers, one for Q-band and one for V-band. The Q-band two stage amplifier showed an average noise figure of 2.2 dB with an 18.3 dB average gain at 44 GHz. The V-band two stage amplifier showed an average noise figure of 2.9 dB with a 14.7 dB average gain at 65 GHz. Our design technique and model demonstrates good agreement between measured and predicted results. Compared with the published data, this work also presents state-of-the-art performance in terms of the gain and noise figure.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.33 no.2
• /
• pp.99-104
• /
• 2020

In this study, we fabricated a metamorphic high-electron-mobility transistor (mHEMT) device with a T-type gate structure for the implementation of W-band monolithic microwave integrated circuits (MMICs) and investigated its characteristics. To fabricate the mHEMT device, a recess process for etching of its Schottky layer was applied before gate metal deposition, and an e-beam lithography using a triple photoresist film for the T-gate structure was employed. We measured DC and RF characteristics of the fabricated device to verify the characteristics that can be used in W-band MMIC design. The mHEMT device exhibited DC characteristics such as a drain current density of 747 mA/mm, maximum transconductance of 1.354 S/mm, and pinch-off voltage of -0.42 V. Concerning the frequency characteristics, the device showed a cutoff frequency of 215 GHz and maximum oscillation frequency of 260 GHz, which provide sufficient performance for W-band MMIC design and fabrication. In addition, active and passive modeling was performed and its accuracy was evaluated by comparing the measured results. The developed mHEMT and device models could be used for the fabrication of W-band MMICs.

• Journal of Sensor Science and Technology
• /
• v.30 no.6
• /
• pp.441-445
• /
• 2021

The components affecting the extrinsic transconductance (g_{m_ext}) in In_0.7Ga_0.3As quantum-well (QW) high-electron-mobility transistors (HEMTs) on an InP substrate were investigated. First, comprehensive modeling, which only requires physical parameters, was used to explain both the intrinsic transconductance (g_{m_int}) and the g_{m_ext} of the devices. Two types of In_0.7Ga_0.3As QW HEMT were fabricated with gate lengths ranging from 10 ㎛ to sub-100 nm. These measured results were correlated with the modeling to describe the device behavior using analytical expressions. To study the effects of the components affecting g_{m_int}, the proposed approach was extended to projection by changing the values of physical parameters, such as series resistances (R_S and R_D), apparent mobility (𝜇_{n_app}), and saturation velocity (𝜈_sat).

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.15 no.5
• /
• pp.490-496
• /
• 2015

We have grown AlN/GaN heterostructure which is a promising candidate for mm-wave applications. For the growth of the high quality very thin AlN barrier, indium was introduced as a surfactant at the growth temperature varied from 750 to $1070^{\circ}C$, which results in improving electrical properties of two-dimensional electron gas (2DEG). The heterostructure with barrier thickness of 7 nm grown at of $800^{\circ}C$ exhibited best Hall measurement results; such as sheet resistance of $215{\Omega}/{\Box}$electron mobility of $1430cm^2/V{\cdot}s$, and two-dimensional electron gas (2DEG) density of $2.04{\times}10^{13}/cm^2$. The high electron mobility transistor (HEMT) was fabricated on the grown heterostructure. The device with gate length of $0.2{\mu}m$ exhibited excellent DC and RF performances; such as maximum drain current of 937 mA/mm, maximum transconductance of 269 mS/mm, current gain cut-off frequency of 40 GHz, and maximum oscillation frequency of 80 GHz.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.6
• /
• pp.195-200
• /
• 2018

A W-band frequency octupler is implemented on 100-nm GaAs pHEMT process. The fabricated octupler can be used as a local oscillator or a signal source of W-band transceivers. Three common-source doublers are connected in cascade to multiply an input signal of 10.75 GHz to 83 GHz. A common-source amplifier is followed for each doubler to improve the conversion gain and suppress the unwanted harmonics. The fabricated octupler showes high output of more than 6 dBm in the 80 - 84 GHz band and achieved excellent spurious suppression performance over 20 dBc.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.48 no.11
• /
• pp.1-8
• /
• 2011

One of the most important parameters that limit maximum output power of transistor is breakdown. InAlAs/InGaAs/GaAs Metamorphic HEMTs (MHEMTs) have some advantages, especially for cost, compared with InP-based ones. However, GaAs-based MHEMTs and InP-based HEMTs are limited by lower breakdown voltage for output power even though they have good microwave and millimeter-wave frequency performance with lower minimum noise figure. In this paper, InAlAs/$In_xGa_{1-x}As$/GaAs MHEMTs are simulated and analyzed for breakdown. The parameters affecting breakdown are investigated in the fabricated 0.1-${\mu}m$ ${\Gamma}$-gate MHEMT device having the modulation-doped $In_{0.52}Al_{0.48}As/In_{0.53}Ga_{0.47}As$ heterostructure on the GaAs wafer using the hydrodynamic transport model of a 2D commercial device simulator. The impact ionization and gate field effect in the fabricated device including deep-level traps are analyzed for breakdown. In addition, Indium mole-fraction-dependent impact ionization rates are proposed empirically for $In_{0.52}Al_{0.48}As/In_xGa_{1-x}As$/GaAs MHEMTs.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.5
• /
• pp.336-343
• /
• 2018

We herein present the design and fabrication of a Rx core chip operating in the X-band (10.5~13 GHz) using Win's commercial $0.25-{\mu}m$ GaAs pHEMT process technology. The X-band core chip comprises a low-noise amplifier, a four-bit phase shifter, and a serial-to-parallel data converter. The size is $1.75mm{\times}1.75mm$, which is the state-of-the-art size. The gain and noise figure are more than 10 dB but less than 2 dB, and both the input and output return losses are less than 10 dB. The RMS phase error is less than $5^{\circ}$, and the P1dB is 2 dBm at 12.5 GHz, the performance of which is equivalent to other GaAs core chips. The fabricated core chip was packaged in a QFN package type with a size of $3mm{\times}3mm$ for the convenience of assembly. We confirmed that the performance of the packaged core chip was almost the same as that of the chip itself.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.10A
• /
• pp.1579-1587
• /
• 1999

Bias-dependent noise models of $0.2\mu\textrm{m}$ gate length P-HEMT's which are scalable with gate width are proposed. To predict S-parameters of the P-HEMT's the intrinsic parameters except for $\tau$ subtracted the offsets introduced in this paper are normalized to the gate width and then scaled. The small-signal model parameters are expressed as fitting functions of the drain current to $\textrm{I}_{dss}$ ratio and gate width. In addition, to estimate accurately noise parameters the noise temperature $\textrm{T}_{g}$ of the intrinsic resistance, the equivalent noise conductance $\textrm{G}_{ni}$ of the gate current noise source, and the equivalent noise conductance $\textrm{G}_{no}$ of the drain current noise source are adopted as the noise model parameters. The extracted values of $\textrm{T}_{g}$ are nearly independent of drain current and gate width and their average is around the ambient temperature. The extracted values of $\textrm{G}_{ni}$ are small enough to be neglected to the circuit characteristics. From the comparison of the noise model with only $\textrm{G}_{no}$ and that having $\textrm{T}_{g}$, $\textrm{G}_{ni}$ and $\textrm{G}_{no}$ to the measured data it is fund that even the former model is in good agreement with the measured noise parameters. Thus, from a practical point of view the noise model having only the drain current noise source is confirmed as a scalable bias-dependent model.