• Journal of electromagnetic engineering and science
• /
• v.1 no.1
• /
• pp.73-77
• /
• 2001

We studied the fabrication of GaAs-based pseudomorphic high electron mobility transistors(PHEMT`s) for the purpose of millimeter- wave applications. To fabricate the high performance GaAs-based PHEMT`s, we performed the simulation to analyze the designed epitaxial-structures. Each unit processes, such as 0.1 m$\mu$$\Gamma$-gate lithography, silicon nitride passivation and air-bridge process were developed to achieve high performance device characteristics. The DC characteristics of the PHEMT`s were measured at a 70 $\mu$m unit gate width of 2 gate fingers, and showed a good pinch-off property ($V_p$= -1.75 V) and a drain-source saturation current density ($I_{dss}$) of 450 mA/mm. Maximum extrinsic transconductance $(g_m)$ was 363.6 mS/mm at $V_{gs}$ = -0.7 V, $V_{ds}$ = 1.5 V, and $I_{ds}$ =0.5 $I_{dss}$. The RF measurements were performed in the frequency range of 1.0~50 GHz. For this measurement, the drain and gate voltage were 1.5 V and -0.7 V, respectively. At 50 GHz, 9.2 dB of maximum stable gain (MSG) and 3.2 dB of $S_{21}$ gain were obtained, respectively. A current gain cut-off frequency $(f_T)$ of 106 GHz and a maximum frequency of oscillation $(f_{max})$ of 160 GHz were achieved from the fabricated PHEMT\\`s of 0.1 m$\mu$ gate length.h.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.7
• /
• pp.1657-1662
• /
• 2014

In this paper, the impact of substrate bias($V_{BS}$) on the zero capacitor RAM(Z-RAM) in n-channel junctionless multiple gate MOSFET(MuGFET) has been analyzed experimentally. Junctionless transistors with fin width of 50nm and 1 fin exhibits a memory window of 0.34V and a sensing margin of $1.8{\times}10^4$ at $V_{DS}=3.5V$ and $V_{BS}=0V$. As the positive $V_{BS}$ is applied, the memory window and sensing margin were improved due to an increase of impact ionization. When $V_{BS}$ is increased from 0V to 10V, not only the memory window is increased from 0.34V to 0.96V but also sensing margin is increased slightly. The sensitivity of memory window with different $V_{BS}$ in junctionless transistor was larger than that of inversion-mode transistor. A retention time of junctionless transistor is better than that of inversion-mode transistor due to low Gate Induced Drain Leakage(GIDL) current. To evaluate the device reliability of Z-RAM, the shifts in the Set/Reset voltages and current were measured.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.451-454
• /
• 2015

This paper introduces 3 type of OTAs with $0.35-{\mu}m$ standard CMOS technology for Low-Power, Low-Voltage. The first type is a two-stage OTA designed to operate with a 1-V VDD and it has $1.774{\mu}W$ low power consumption. All transistors are operating in strong inversion. It takes Gm-Enhancement techniques to compensate gm, which is lowered by Bulk-Driven technique and has an Wide swing current mirror for low voltage operation and a Class-A output. The second type is a Two-stage OTA designed to operate with a 0.8-V VDD and It has 52nW low power consumption and 112dB high gain. The current mirror uses Composite Transistor binding Gates of two MOSFET to raise Rout which is similar with cascode structure. The third type is a Two-stage OTA designed to operate with a 0.6-V VDD and It has 160nW low power consumption and 72dB high gain. It takes Level Shift technique by Common Gate structure to amplify signals without additional bias voltage at second stage.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.458-461
• /
• 2015

In this paper 2 types of voltage references and a current reference suitable for low-voltage, low-power circuits are proposed and designed with $0.35{\mu}m\;CMOS$ process. MOS transistors operating in weak inversion and bulk-driven technique are utilized to achieve low-voltage and low-power features. The first voltage reference consumes 1.43uA from a supply voltage of 1.2V while it has a reference voltage of 585mV and a TC(Temperature Coefficient) of $6ppm/^{\circ}C$. The second voltage reference consumes 48pW from a supply voltage of 0.3V while having a reference voltage of 172mV and a TC of $26ppm/^{\circ}C$. The current reference consumes 246nA from a supply voltage of 0.75V with a reference current of 32.6nA and a TC of $262ppm/^{\circ}C$. The performances of the designed references have been verified through simulations.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2008.06a
• /
• pp.133-133
• /
• 2008

Recently, as the down-scailing of field-effect transistor devices continues, Schottky-barrier field-effect transistors (SB-FETs) have attracted much attention as an alternative to conventional MOSFETs. SB-FETs have advantages over conventional devices, such as low parasitic source/drain resistance due to their metallic characteristics, low temperature processing for source/drain formation and physical scalability to the sub-10nm regime. The good scalability of SB-FETs is due to their metallic characteristics of source/drain, which leads to the low resistance and the atomically abrupt junctions at metal (silicide)-silicon interface. Nevertheless, some reports show that SB-FETs suffer from short channel effect (SCE) that would cause severe problems in the sub 20nm regime.[Ouyang et al. IEEE Trans. Electron Devices 53, 8, 1732 (2007)] Because source/drain barriers induce a depletion region, it is possible that the barriers are overlapped in short channel SB-FETs. In order to analyze the SCE of SB-FETs, we carried out systematic studies on the Schottky barrier overlapping in short channel SB-FETs using a SILVACO ATLAS numerical simulator. We have investigated the variation of surface channel band profiles depending on the doping, barrier height and the effective channel length using 2D simulation. Because the source/drain depletion regions start to be overlapped each other in the condition of the $L_{ch}$~80nm with $N_D{\sim}1\times10^{18}cm^{-3}$ and $\phi_{Bn}$ $\approx$ 0.6eV, the band profile varies as the decrease of effective channel length $L_{ch}$. With the $L_{ch}$~80nm as a starting point, the built-in potential of source/drain schottky contacts gradually decreases as the decrease of $L_{ch}$, then the conduction and valence band edges are consequently flattened at $L_{ch}$~5nm. These results may allow us to understand the performance related interdependent parameters in nanoscale SB-FETs such as channel length, the barrier height and channel doping.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.1
• /
• pp.37-44
• /
• 2015

In this paper, we have investigated a selective assembly method of single-walled carbon nanotubes (SWCNTs) on a silicon substrate using only photolithographic process and then proposed a fabrication method of field effect transistors (FETs) using SWCNT-based patterns. The aminopropylethoxysilane (APTES) patterns, which are formed for positively charged surface molecular patterns, are utilized to assemble and align millions of SWCNTs and we can more effectively assemble on a silicon (Si) surface using this method than assembly processes using only the 1-octadecyltrichlorosilane (OTS). We investigated a selective assembly method of SWCNTs on a Si surface using surface-programmed APTES and OTS patterns and then a fabrication method of FETs. photoresist(PR) patterns were made using photolithographic process on the silicon dioxide (SiO2) grown Si substrate and the substrate was placed in the OTS solution (1:500 v/v in anhydrous hexane) to cover the bare SiO2 regions. After removing the PR, the substrate was placed in APTES solution to backfill the remaining SiO2 area. This surface-programmed substrate was placed into a SWCNT solution dispersed in dichlorobenzene. SWCNTs were attracted toward the positively charged molecular regions, and aligned along the APTES patterns. On the contrary, SWCNT were not assembled on the OTS patterns. In this process, positively charged surface molecular patterns are utilized to direct the assembly of negatively charged SWCNT on SiO2. As a result, the selectively assembled SWCNT channels can be obtained between two electrodes(source and drain electrodes). Finally, we can successfully fabricate SWCNT-based multi-channel FETs by using our self-assembled monolayer method.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2009.10a
• /
• pp.421-427
• /
• 2009

We successfully fabricated 4.7-inch organic thin film transistors array with $640{\times}480$ pixels on flexible substrate. All the processes were done by photolithography, spin coating and ink-jet printing. The OTFT-Electrophoretic (EP) pixel structure, based on a top gate OTFT, was fabricated. The mobility, ON/OFF ratio, subthreshold swing and threshold voltage of OTFT on flexible substrate are: 0.01 ^2/V-s, 1.3 V/dec, 10E5 and -3.5 V. After laminated the EP media on OTFT array, a panel of 4.7-inch $640{\times}480$ OTFT-EPD was fabricated. All of process temperature in OTFT-EPD is lower than $150^{\circ}C$. The pixel size in our panel is $150{\mu}m{\times}150{\mu}m$, and the aperture ratio is 50 %. The OTFT channel length and width is 20 um and 200um, respectively. We also used OTFT to drive EP media successfully. The operation voltages that are used on the gate bias are -30 V during the row data selection and the gate bias are 0 V during the row data hold time. The data voltages that are used on the source bias are -20 V, 0 V, and 20 V during display media operation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.6A
• /
• pp.634-639
• /
• 2006

In this paper an MMIC 2-stage power amplifier is designed and fabricated for 5GHz wireless LAN applications using $0.5{\mu}m$ gate length PHEMT transistors. The PHEMT gate width is optimized in order to meet the linearity and efficiency of the MMIC power amplifier. The $0.5{\mu}m\times600{\mu}m$ PHEMT for the drive stage and $0.5{\mu}m\times3000{\mu}m$ PHEMT for the amplification stage are the optimized sizes to achieve more than 25dBc of third order IMD at the power level of 3dB back-off from the input P1dB and more than 22dBm output power under a supply voltage of 3.3V. The two-stage MMIC power amplifier is designed to be used for the both of HIPERLAN/2 and IEEE 802.11a because of its broadband characteristics. The fabricated PHEMT MMIC power amplifier exhibits a 20.1dB linear power gain, a maximum 22dBm output power, a 24% power added efficiency under 3.3V supply voltage. The input and output on-chip matching circuits are included on a chip of $1400\times1200{\mu}m^2$.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.3
• /
• pp.60-68
• /
• 2008

This work describes a re-configurable 10MS/s to 100MS/s, low-power 10b two-step pipeline ADC operating at a power supply from 0.5V to 1.2V. MOS transistors with a low-threshold voltage are employed partially in the input sampling switches and differential pair of the SHA and MDAC for a proper signal swing margin at a 0.5V supply. The integrated adjustable current reference optimizes the static and dynamic performance of amplifiers at 10b accuracy with a wide range of supply voltages. A signal-isolated layout improves the capacitor mismatch of the MDAC while a switched-bias power-reduction technique reduces the power dissipation of comparators in the flash ADCs. The prototype ADC in a 0.13um CMOS process demonstrates the measured DNL and INL within 0.35LSB and 0.49LSB. The ADC with an active die area of $0.98mm^2$ shows a maximum SNDR and SFDR of 56.0dB and 69.6dB, respectively, and a power consumption of 19.2mW at a nominal condition of 0.8V and 60MS/s.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.44 no.9
• /
• pp.71-80
• /
• 2007

This paper presents a new push-push VCO technique that extracts a second harmonic output signal from a capacitive commonnode in a negativegm oscillator topology. The generation of the $2^{nd}$ harmonics is accounted for by the nonlinear current-voltage characteristic of the emitter-base junction diode causing; 1) significant voltage clipping and 2) different rising and falling time during the switching operation of core transistors. Comparative investigations show the technique is more power efficient in the high-frequency region that a conventional push-push technique using an emitter common node. Prototype 12GHz and 17GHz MMIC VCO were realized in GaInP/GaAs HBT technology. They have shown nominal output power of -4.3dBm and -5dBm, phase noise of -108 dBc/Hz and -110.4 dBc/Hz at 1MHz offset, respectively. The phase noise results are also equivalent to a VCO figure-of-merit of -175.8 dBc/Hz and -184.3 dBc/Hz, while dissipate 25.68mW(10.7mA/2.4V) and 13.14mW(4.38mA/3.0V), respectively.