Comparison of Ultra-Thin InAs and InGaAs Quantum Wells and Ultra-Thin-Body Surface-Channel MOSFETs

Similar documents
Performance Analysis of Ultra-Scaled InAs HEMTs

III-V CMOS: What have we learned from HEMTs? J. A. del Alamo, D.-H. Kim 1, T.-W. Kim, D. Jin, and D. A. Antoniadis

Ultra-Scaled InAs HEMTs

The Prospects for III-Vs

Effect of the High-k Dielectric/Semiconductor Interface on Electronic Properties in Ultra-thin Channels

OMEN an atomistic and full-band quantum transport simulator for post-cmos nanodevices

Technology Development & Design for 22 nm InGaAs/InP-channel MOSFETs

Technology Development for InGaAs/InP-channel MOSFETs

Microsystems Technology Laboratories, MIT. Teledyne Scientific Company (TSC)

High Mobility Materials and Novel Device Structures for High Performance Nanoscale MOSFETs

InGaAs Double-Gate Fin-Sidewall MOSFET

Carbon Nanotube Electronics

ALD high-k and higher-k integration on GaAs

The Pennsylvania State University. Kurt J. Lesker Company. North Carolina State University. Taiwan Semiconductor Manufacturing Company 1

Journal of Electron Devices, Vol. 18, 2013, pp JED [ISSN: ]

Low Frequency Noise in MoS 2 Negative Capacitance Field-effect Transistor

Quantum-size effects in sub-10 nm fin width InGaAs finfets

Performance Analysis of 60-nm Gate-Length III-V InGaAs HEMTs: Simulations Versus Experiments

Simple Theory of the Ballistic Nanotransistor

EECS130 Integrated Circuit Devices

Performance Comparisons of III-V and strained-si in Planar FETs and Non-planar FinFETs at Ultrashort Gate Length (12nm)

Tri-Gate Fully-Depleted CMOS Transistors: Fabrication, Design and Layout

ECE 305: Fall MOSFET Energy Bands

Enhanced Mobility CMOS

A final review session will be offered on Thursday, May 10 from 10AM to 12noon in 521 Cory (the Hogan Room).

30 nm In 0.7 Ga 0.3 As Inverted-type HEMT with Reduced Gate Leakage Current for Logic Applications

EE410 vs. Advanced CMOS Structures

Scaling Issues in Planar FET: Dual Gate FET and FinFETs

Components Research, TMG Intel Corporation *QinetiQ. Contact:

A Multi-Gate CMOS Compact Model BSIMMG

Study of Carrier Transport in Strained and Unstrained SOI Tri-gate and Omega-gate Si Nanowire MOSFETs

A Numerical Study of Scaling Issues for Schottky Barrier Carbon Nanotube Transistors

ECE-305: Fall 2017 MOS Capacitors and Transistors

Prospects for Ge MOSFETs

The Critical Role of Quantum Capacitance in Compact Modeling of Nano-Scaled and Nanoelectronic Devices

III-V Nanowire TFETs

Self-Aligned InGaAs FinFETs with 5-nm Fin-Width and 5-nm Gate-Contact Separation

Indium arsenide quantum wire trigate metal oxide semiconductor field effect transistor

arxiv: v2 [cond-mat.mes-hall] 27 Dec 2017

Multiple Gate CMOS and Beyond

CHAPTER 5 EFFECT OF GATE ELECTRODE WORK FUNCTION VARIATION ON DC AND AC PARAMETERS IN CONVENTIONAL AND JUNCTIONLESS FINFETS

Tunnel-FET: bridging the gap between prediction and experiment through calibration

ECE 305 Exam 5 SOLUTIONS: Spring 2015 April 17, 2015 Mark Lundstrom Purdue University

Lecture 18 Field-Effect Transistors 3

High-Performance Complementary III-V Tunnel FETs with Strain Engineering

Lecture #27. The Short Channel Effect (SCE)

Beyond Si: Opportunities and Challenges for CMOS Technology Based on High-Mobility Channel Materials T.P. Ma Yale University

High aspect-ratio InGaAs FinFETs with sub-20 nm fin width

Physics an performance of III-V nanowire heterojunction TFETs including phonon and impurity band tails:

Ultimately Scaled CMOS: DG FinFETs?

Lecture 3: Transistor as an thermonic switch

Lecture 5: CMOS Transistor Theory

Quantum and Non-local Transport Models in Crosslight Device Simulators. Copyright 2008 Crosslight Software Inc.

A Theoretical Investigation of Surface Roughness Scattering in Silicon Nanowire Transistors

Subthreshold and scaling of PtSi Schottky barrier MOSFETs

Quantum Mechanical Simulation for Ultra-thin High-k Gate Dielectrics Metal Oxide Semiconductor Field Effect Transistors

Physics-based compact model for ultimate FinFETs

Nanoscale CMOS Design Issues

Quantum Phenomena & Nanotechnology (4B5)

Non-equilibrium Green s functions: Rough interfaces in THz quantum cascade lasers

GaN based transistors

Electrostatics of Nanowire Transistors

Analysis of Band-to-band. Tunneling Structures. Title of Talk. Dimitri Antoniadis and Judy Hoyt (PIs) Jamie Teherani and Tao Yu (Students) 8/21/2012

This is the author s final accepted version.

Electro-Thermal Transport in Silicon and Carbon Nanotube Devices E. Pop, D. Mann, J. Rowlette, K. Goodson and H. Dai

Analysis of InAs Vertical and Lateral Band-to-Band Tunneling. Transistors: Leveraging Vertical Tunneling for Improved Performance

Erik Lind

Energy dispersion relations for holes inn silicon quantum wells and quantum wires

NCN Student Research Symposium (NSRS)

Bandstructure Effects in Silicon Nanowire Electron Transport

Spin Lifetime Enhancement by Shear Strain in Thin Silicon-on-Insulator Films. Dmitry Osintsev, Viktor Sverdlov, and Siegfried Selberherr

ECE-305: Spring 2016 MOSFET IV

FLCC Seminar. Spacer Lithography for Reduced Variability in MOSFET Performance

Performance Enhancement of P-channel InGaAs Quantum-well FETs by Superposition of Process-induced Uniaxial Strain and Epitaxially-grown Biaxial Strain

Modeling and Computation of Gate Tunneling Current through Ultra Thin Gate Oxides in Double Gate MOSFETs with Ultra Thin Body Silicon Channel

Chapter 5 MOSFET Theory for Submicron Technology

1464 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 63, NO. 4, APRIL 2016

New Tools for the Direct Characterisation of FinFETS

arxiv: v2 [cond-mat.mtrl-sci] 11 Sep 2016

University of California Postprints

Ballistic Graphene Nanoribbon MOSFETs: a full quantum real-space simualtion study

Characteristics Optimization of Sub-10 nm Double Gate Transistors

IBM Research Report. Quantum-Based Simulation Analysis of Scaling in Ultra-Thin Body Device Structures

Quantification of Trap State Densities at High-k/III-V Interfaces

M R S Internet Journal of Nitride Semiconductor Research

Lecture 11: MOSFET Modeling

Extending the Era of Moore s Law

Lecture 9. Strained-Si Technology I: Device Physics

Part 5: Quantum Effects in MOS Devices

Electronics with 2D Crystals: Scaling extender, or harbinger of new functions?

IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 58, NO. 7, JULY

Normally-Off GaN Field Effect Power Transistors: Device Design and Process Technology Development

Directions for simulation of beyond-cmos devices. Dmitri Nikonov, George Bourianoff, Mark Stettler

Keywords MOSFET, FinFET, Silicon, Germanium, InGaAs, Monte Carlo, Drift Diffusion.

Electrical measurements of voltage stressed Al 2 O 3 /GaAs MOSFET

Non-equilibrium Green's function (NEGF) simulation of metallic carbon nanotubes including vacancy defects

A Verilog-A Compact Model for Negative Capacitance FET

CHAPTER 3. EFFECT OF STRUCTURAL AND DOPING PARAMETER VARIATIONS ON f t, NQS DELAY, INTRINSIC GAIN AND NF IN N-TYPE FINFETS

III-V field-effect transistors for low power digital logic applications

Fundamentals of Nanoelectronics: Basic Concepts

Transcription:

Comparison of Ultra-Thin InAs and InGaAs Quantum Wells and Ultra-Thin-Body Surface-Channel MOSFETs Cheng-Ying Huang 1, Sanghoon Lee 1, Evan Wilson 3, Pengyu Long 3, Michael Povolotskyi 3, Varistha Chobpattana 2, Susanne Stemmer 2, Arthur Gossard 1,2, Gerhard Klimeck 3, and Mark Rodwell 1 1 ECE, University of California, Santa Barbara 2 Materials Department, University of California, Santa Barbara 3 Network for Computational Nanotechnology, Purdue University, West Lafayette, IN CSW/IPRM 2015 Santa Barbara, CA

Why III-V FETs? Why Ultra-thin channel? III-V channel: low electron effective mass high velocity, high mobility higher current at lower V DD reducing switching power Channel thickness (T ch ) must be scaled in proportional to gate length (L g ) to maintain electrostatic integrity. For ultra-thin body (UTB) MOSFETs T ch ~1/4 L g, and for FinFETs T ch ~1/2 L g. At 7 or 5 nm nodes, channel thickness should be around 2-4 nm. Goal: Carefully examine InGaAs and InAs channels. The best design?? 300K Si InAs InGaAs m e * 0.19 0.023 0.041 μ e (cm 2 /V s) 1450 33000 12000 μ h (cm 2 /V s) 370 450 <300 t Si ~5nm W FIN ~8nm Eg(eV) 1.12 0.354 0.75 ε r 11.7 15.2 13.9 a(å) 5.43 6.0583 (InP) Quantum confinement effects!! STM-Leti-IBM Intel 14nm FinFET 14nm UTBSOI IEDM2014 Makr Bohr, IDF2014 2

Ultra-thin channel 2DEG: Hall results Quantum well (QW) 2DEGs were grown by solid source MBE. For wide wells: μ InAs > μ InGaAs For narrow wells (~2 nm): μ InAs μ InGaAs Carrier concentration decreases due to increased E 0. Mobility (cm 2 /Vs) 10000 8000 6000 4000 2000 InAs channel InGaAs channel 0 0 1 2 3 4 5 6 7 8 Well Thickness (nm) Carrier concentration(10 12 cm -2 ) 2.2 2.0 1.8 1.6 1.4 InAs channel InGaAs channel 1.2 0 1 2 3 4 5 6 7 8 Well Thickness (nm) 3

What happens in thin wells? Mobility is limited by interface roughness scattering. Strained InAs growth (S-K mode) might induce higher interface roughness. Electron effective mass are similar for ~2-3 nm InAs and InGaAs wells because of non-parabolic band effects. Mobility (cm 2 V -1 s -1 ) 10 7 10 6 10 5 10 4 10 3 10 2 InGaAs well alloy µ ~T ch -6 interface roughness remote impurity acoustic phonon polar optical phonon total 300K 5 10 15 20 Well thickness(nm) C. Y. Huang et al., J. Appl. Phys. 115, 123711 (2013) In-plane effective mass, m // (m 0 ) 0.10 0.09 0.08 0.07 0.06 0.05 InGaAs/InP, Nag1993 InAs/InP, Nag1993 InGaAs/InP, Wetzel1992 InGaAs/InP, Hrivnak1992 InAs, Mugny2015 Schneider1995 Wiesner1994 Wetzel1996 0.04 InGaAs 0.03 InAs 0.02 0 5 10 15 20 Well Thickness (nm) J. Appl. Phys. 77, 2828 (1995), Phys. Rev. B, 52, 1038 (1996), Appl. Phys. Lett 64, 2520 (1994), Appl. Phys. Lett 62, 2416 (1993), G. Mugny et al., EUROSOI-ULSI conference 2015. 4

Ultra-thin body III-V FETs: L g ~40 nm I D (ma/ m) I D (ma/ m) UTB FETs with 3 nm channels were fabricated to compare InAs and InGaAs channels. 1.6:1 I on and transcoaductance for InAs channels. 10:1 lower I off for InGaAs channels. 10 1 10 0 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-8 10-9 1.5 1.0 0.5 = 0.1 to 0.7 V 0.2 V increment InAs -0.2 0.0 0.2 0.4 0.6 V GS (V) V GS = -0.2 V to 1.0 V 0.2 V increment R on = 292 Ohm- m at V GS = 1.0 V InAs 0.0 0.0 0.2 0.4 0.6 V (V) = 0.1 to 0.7 V 0.2 V increment InGaAs -0.2 0.0 0.2 0.4 0.6 V GS (V) V GS = -0.2 V to 1.0 V 0.2 V increment R on = 400 Ohm- m at V GS = 1.0 V InGaAs 0.2 0.4 0.6 V (V) 2.4 2.0 1.6 1.2 0.8 0.4 0.0 gm (ms/ m) 5

I on ( A/ m) On-state performance Peak g m (ms/ m) R on (Ohm m) 500 400 300 200 100 0 3 nm InAs 3 nm InGaAs = 0.5 V I off = 100 na/ m 100 1000 Gate length (nm) 2.4 2.0 1.6 1.2 0.8 0.4 0.0 = 0.5 V 3 nm InAs 3 nm InGaAs 0.1 1 Gate length ( m) 1000 750 500 250 V GS = 1 V 3 nm InAs 3 nm InGaAs R S/D ~ 190 20 Ohm m 0 0.0 0.1 0.2 Gate length ( m) Higher I on and higher g m for UTB InAs FETs than InGaAs UTB FETs. InAs FETs achieve g m =2 ms/μm, and I on =400 μa/μm at =0.5V and I off =100 na/μm. Similar source/drain resistance (R S/D ) ensures that the performance degradation of InGaAs is not from source/drain, but from channel itself (slope). 6

Subthreshold Swing (mv/dec) Subthreshold swing and off-state current DIBL (mv/v) I OFF, MIN (na/ m) 120 100 =0.1 V,3 nm InAs =0.5 V,3 nm InAs =0.1 V,3 nm InGaAs =0.5 V,3 nm InGaAs 150 100 3 nm InAs 3 nm InGaAs V th at 1 A/ m 10 2 3 nm InAs 10 1 3 nm InGaAs 10 0 80 50 10-1 10-2 = 0.5 V 60 0.1 1 Gate Length ( m) 0 0.1 1 Gate length ( m) 10-3 100 1000 Gate length (nm) Superior SS~83 mv/dec. and DIBL~110 mv/v because of ultra-thin channels and improved electrostatics. Minimum I off is 10:1 lower for InGaAs channel at short L g, where leakage current limited by band-to-band tunneling. InGaAs FETs are limited by gate leakage at long L g. 7

Why QW-2DEGs and UTB-FETs show different results? 1 st possible cause: Electron population in L valley due to strong quantum confinement Unlikely. 2nm InAlAs barrier, 3nm InGaAs channel with H passivation on top 2nm InAlAs barrier, 3nm InAs channel with H passivation on top Courtesy of Evan Wilson, Pengyu Long, Michael Povolotskyi, and Gerhard Klimeck. In 0.53 Ga 0.47 As InAs m e * at Γ [m 0 ] 0.080 0.063 Γ L separation [ev] 0.596 0.905 E g at Γ [ev] 1.06 0.639 8

Why QW-2DEGs and UTB-FETs show different results? 2 nd possible cause: Electron interaction with oxide traps inside conduction band Likely. Electrons in high In% content channels have less scattering and less electron capture by the oxide traps. J. Robertson et al., J. Appl. Phys. 117, 112806 (2015) J. Robertson, Appl. Phys. Lett. 94, 152104 (2009) N. Taoka et al., Trans. Electron Devices. 13, 456 (2011) N. Taoka et al., IEEE IEDM 2011, 610. 9

UCSB L g ~12 nm III-V MOSFETs (DRC 2015) I D (ma/ m) I D (ma/ m) N+InGaAs N+InP InP spacer L g ~12nm ~ 8nm Ni 10 1 10 0 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-8 -0.2 0.0 0.2 0.4 0.6 0.8 V GS (V) 1.5 V GS = -0.2 V to 1.2 V 1.0 = 0.1 to 0.7 V, 0.2 V increment SS~107 mv/dec. SS~98 mv/dec. 0.2 V increment R on = 302 Ohm- m at V GS = 1.0 V 2.4 2.0 1.6 1.2 0.8 0.4 0.0 gm (ms/ m) 0.5 InAlAs Barrier t ch ~ 2.5 nm (1.5/1 nm InGaAs/InAs) 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 (V) I on /I off >8.3 10 5 10

Summary Below 10 nm logic nodes, ultrathin channels are required. In QW 2DEGs, the electron Hall mobility are similar for InGaAs and InAs wells as the wells thinned to 2~3nm. In UTB MOSFETs, 3 nm InAs channels significantly improve on-state current and transconductance (~1.6:1), and reduce channel resistance as compared to 3 nm InGaAs channel. Purdue s tight-binding calculations show large ~0.6 ev Γ L splitting in 3 nm InGaAs channels, ruling out the possibility of electron population in L-valley. UCSB C-V measurements show large dispersion in 3 nm InGaAs channels, possibly indicating the significant electron interactions with oxide traps. (As-As anti-bonding may be the culprit) 11

Acknowledgment Thanks for your attention! Questions? This research was supported by the SRC Non-classical CMOS Research Center (Task 1437.009) and GLOBALFOUNDRIES(Task 2540.001). A portion of this work was done in the UCSB nanofabrication facility, part of NSF funded NNIN network. This work was partially supported by the MRSEC Program of the National Science Foundation under Award No. DMR 1121053. cyhuang@ece.ucsb.edu

(backup slides follow)

Effective C G ( F/cm 2 ) Mobility (cm 2 /V s) Mobility in different channel design: 25 µm-l g 3.0 2.5 2.0 1.5 1.0 0.5 4.5 nm InGaAs 2.5 nm InAs 5.0 nm InAs Freq.: 200 khz W/L=25 m/21 m 0.0 0-0.2 0.0 0.2 0.4 0.6 0.8 V GS (V) 8 6 4 2 Carrier density ( 10 12 cm -2 ) 1200 1000 800 600 400 200 4.5 nm InGaAs 2.5 nm InAs 5.0 nm InAs 0 0 1 2 3 4 5 6 7 Carrier density (cm -2 ) m*, C g-ch, R S/D more important for ballistic FETs