Multiple Gate CMOS and Beyond

Similar documents
N ano scale l S il ii lco i n B ased N o nvo lat l i atl ie l M em ory r Chungwoo Kim, Ph.D.

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

Recent Development of FinFET Technology for CMOS Logic and Memory

Enhanced Mobility CMOS

Chapter 3 Basics Semiconductor Devices and Processing

FLCC Seminar. Spacer Lithography for Reduced Variability in MOSFET Performance

Scaling Issues in Planar FET: Dual Gate FET and FinFETs

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

Gold Nanoparticles Floating Gate MISFET for Non-Volatile Memory Applications

A Multi-Gate CMOS Compact Model BSIMMG

EE410 vs. Advanced CMOS Structures

Nanometer Transistors and Their Models. Jan M. Rabaey

How a single defect can affect silicon nano-devices. Ted Thorbeck

Faculty Presentation: Novel Technologies

Components Research, TMG Intel Corporation *QinetiQ. Contact:

Imaginary Band Structure and Its Role in Calculating Transmission Probability in Semiconductors

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

Suppression of Gate-Induced Drain Leakage by Optimization of Junction Profiles in 22 nm and 32 nm SOI nfets

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

Moores Law for DRAM. 2x increase in capacity every 18 months 2006: 4GB

MOSFET: Introduction

L ECE 4211 UConn F. Jain Scaling Laws for NanoFETs Chapter 10 Logic Gate Scaling

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

EE115C Winter 2017 Digital Electronic Circuits. Lecture 3: MOS RC Model, CMOS Manufacturing

CHAPTER I. Introduction. 1.1 State of the art for non-volatile memory

MSE 310/ECE 340: Electrical Properties of Materials Fall 2014 Department of Materials Science and Engineering Boise State University

Ultimately Scaled CMOS: DG FinFETs?

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

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

GMU, ECE 680 Physical VLSI Design 1

Investigation of the Dimension Effects of Sub-30nm Multiple-Gate SOI MOSFETs by TCAD Simulation

There s Plenty of Room at the Bottom and at the Top

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

Extending the Era of Moore s Law

The Future of CMOS. David Pulfrey. CHRONOLOGY of the FET. Lecture Lilienfeld s patent (BG FET) 1965 Commercialization (Fairchild)

CMPEN 411 VLSI Digital Circuits. Lecture 03: MOS Transistor

Random Access Memory. DRAM & SRAM Design DRAM SRAM MS635. Dynamic Random Access Memory. Static Random Access Memory. Cell Structure. 6 Tr.

ESE 570: Digital Integrated Circuits and VLSI Fundamentals

Fig The electron mobility for a-si and poly-si TFT.

EECS130 Integrated Circuit Devices

ECE-305: Fall 2017 MOS Capacitors and Transistors

Magnetic core memory (1951) cm 2 ( bit)

High Dielectric Constant (k) Materials

Comparative studies of Ge and Si p-channel metal oxide semiconductor field-effect-transistors with HfSiON dielectric and TaN metal gate

ESE 570: Digital Integrated Circuits and VLSI Fundamentals

Lecture 8. Detectors for Ionizing Particles

MOSFET SCALING ECE 663

SOI/SOTB Compact Models

The PSP compact MOSFET model An update

Future trends in radiation hard electronics

Gate Carrier Injection and NC-Non- Volatile Memories

Ion Implantation. alternative to diffusion for the introduction of dopants essentially a physical process, rather than chemical advantages:

Semiconductor Memories

Thin Film Transistors (TFT)

The Devices. Devices

Si Nanowire FET Modeling and Technology

MOS Transistor Properties Review

Introduction and Background

A 20 nm gate-length ultra-thin body p-mosfet with silicide source/drain

Chapter 2. Design and Fabrication of VLSI Devices

ELEN0037 Microelectronic IC Design. Prof. Dr. Michael Kraft

Simulation-based Study of Super-steep Retrograde Doped Bulk FinFET Technology and 6T-SRAM Yield

The Devices: MOS Transistors

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

SEMICONDUCTOR MEMORIES

Characteristics of MOSFET with Non-overlapped Source-Drain to Gate

! CMOS Process Enhancements. ! Semiconductor Physics. " Band gaps. " Field Effects. ! MOS Physics. " Cut-off. " Depletion.

Electric-Field Induced F - Migration in Self-Aligned InGaAs MOSFETs and Mitigation

The Devices. Digital Integrated Circuits A Design Perspective. Jan M. Rabaey Anantha Chandrakasan Borivoje Nikolic. July 30, 2002

Supporting Information for: Sustained sub-60 mv/decade switching via the negative capacitance effect in MoS 2 transistors

Lecture #27. The Short Channel Effect (SCE)

Lecture 6 PN Junction and MOS Electrostatics(III) Metal-Oxide-Semiconductor Structure

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

Section 12: Intro to Devices

Part 5: Quantum Effects in MOS Devices

Digital Integrated Circuits A Design Perspective

Feature-level Compensation & Control. Process Integration September 15, A UC Discovery Project

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

PROGRESS AND ISSUES IN DIELECTRIC MATERIALS FOR SUB-100NM DRAM TECHNOLOGY ABSTRACT

Circuits. L5: Fabrication and Layout -2 ( ) B. Mazhari Dept. of EE, IIT Kanpur. B. Mazhari, IITK. G-Number

Homework 2 due on Wednesday Quiz #2 on Wednesday Midterm project report due next Week (4 pages)

! CMOS Process Enhancements. ! Semiconductor Physics. " Band gaps. " Field Effects. ! MOS Physics. " Cut-off. " Depletion.

Split-gate charge trap memories: impact of scaling on performances and consumption for low-power embedded applications

RADIATION EFFECTS IN SEMICONDUCTOR MATERIALS AND DEVICES FOR SPACE APPLICATIONS. Cor Claeys and Eddy Simoen

JUNCTION LEAKAGE OF A SiC-BASED NON-VOLATILE RANDOM ACCESS MEMORY (NVRAM) K. Y. Cheong ABSTRACT INTRODUCTION

Single Event Effects: SRAM

Lecture 25. Semiconductor Memories. Issues in Memory

Solid-State Electronics

Lecture 5: CMOS Transistor Theory

Digital Integrated Circuits A Design Perspective. Semiconductor. Memories. Memories

+ V gate M O. Trend: As k, E g. Part 6: High Dielectric Constant (k), Gate Electrode, & Channel Materials. Bandgap versus Dielectric Constant (k) k k

Lecture 9. Strained-Si Technology I: Device Physics

Advanced and Emerging Devices: SEMATECH s Perspective

Effective Capacitance Enhancement Methods for 90-nm DRAM Capacitors

Optimization of the Dielectric Constant of a Blocking Dielectric in the Nonvolatile Memory Based on Silicon Nitride

Lecture 12: MOS Capacitors, transistors. Context

P. R. Nelson 1 ECE418 - VLSI. Midterm Exam. Solutions

Electrical and Reliability Characteristics of RRAM for Cross-point Memory Applications. Hyunsang Hwang

Characteristics Optimization of Sub-10 nm Double Gate Transistors

ANALYTICAL SOI MOSFET MODEL VALID FOR GRADED-CHANNEL DEVICES

Transcription:

Multiple CMOS and Beyond Dept. of EECS, KAIST Yang-Kyu Choi

Outline 1. Ultimate Scaling of MOSFETs - 3nm Nanowire FET - 8nm Non-Volatile Memory Device 2. Multiple Functions of MOSFETs 3. Summary 2

CMOS Evolution Scenario S G D S G D Fin Fin Fin Buried Oxide Bulk-single gate SOI single gate Double gate Ω-gate All-around No report for Sub-50nm L G =4nm NEC IEDM 2003 L G =8nm IBM IEDM 2003 L G =10nm AMD/Berkeley IEDM 2003 L G =10nm LETI VLSI 2005 This Work 3

ITRS Roadmap L G (nm) EOT Bulk SG (nm) DG I sd,leak (µa/µm) I d,sat (ma/mm) Bulk SG DG Bulk SG DG 2006 2008 2011 28 22 16 1.1 0.9 0.5 0.8 0.15 0.2 0.32 0.1 1130 1570 2490 1899 Solution exist Solution being pursued No known solution 4 2013 13 0.6 0.11 2220 2020 5 0.5 0.11 2981 Source: ITRS 2005 roadmap

All-Around (AAG) FinFET Source AAG-FinFET L G =3nm W Fin =3nm EOT(HfO 2 )=1.2nm Drain To fabricate sub-5nm silicon transistor, All-Around (AAG) FinFET was proposed. 5

Process Flow of AAG-FinFET (100) SOI wafer Silicon body thinning Fin patterning (dual-resist) Sacrificial oxidation dielectric (HfO 2 ) Poly-silicon deposition patterning (dual-resist) Spacer formation Source/Drain implantation Spike annealing (1000 o C) Forming gas annealing (450 o C) 6

3nm AAG FinFET: Silicon-Fin Silicon fin 20nm HfO 2 Source S Poly-si 3nm Drain Buried oxide 5nm 3nm 5nm 7 Y.-K. Choi et al., VLSI, 2006

3nm AAG FinFET: 5nm Poly-si Source 5nm Poly-si HfO 2 IFO 5nm Si Poly-si 3nm Drain HfO 2 IFO Silicon fin 5nm Si 8

I-V of 3nm AAG FinFET Drain Current [A/µm] 10-3 Simulated Measured 10-4 10-5 10-6 10-7 V D =1.0V V D =0.2V 160 140 120 100 80 DIBL=230mV/V SS=208 mv/dec 60 HfO 2 =1.4nm 40 L G =3nm 20 W Fin =3nm 0-1.0-0.6-0.2 0.2 0.6 1.0 Voltage [V] Transconductance [µs/µm] Drain Current [µa/µm] 350 300 250 200 150 100 50 HfO 2 =1.4nm L G =3nm W Fin =3nm 0.6V 0.4V V G =0.2V 0 0.0 0.2 0.4 0.6 0.8 1.0 Drain Voltage [V] Large DIBL and SS due to thick EOT ITRS requirement: 0.5nm EOT for DG 9

Fundamental Limit of Scaling Device scaling limit (Operating at 300K) Heisenberg s uncertainty principle Shannon - von Neumann - Landauer (SNL) expression SCALING LIMIT x min h = = p = 1.5nm h 2m E c bit = ( T = 300K) 2m k c h B T ln 2 1.5 Fabricated 3nm all-around gate FinFET is approaching to this fundamental limit. 10

Nanowire Structure Silicon nanowire non-volatile memory structure for ultimate scaling Poly-silicon Blocking oxide Nitride trap layer Tunneling oxide Si Buried Oxide 8nm Non-Volatile Memory Omega-shape gate ONO-structure 11

8nm Si Nanowire NVM Poly-silicon N O O S e - S Silicon Nanowire (7/10nm) trap sites BOX Tunneling oxide Nitride trap layer Blocking oxide 10nm Poly-silicon 12 Y.-K. Choi et al., VLSI, 2007

8nm Si Nanowire NVM 7/10nm 3.8nm 6.4nm 5.1nm Oxide Nitride Silicon T ONO =15nm > L G =8nm 8nm 5.1nm 6.4nm 3.8nm 13

Drain Current [A/µm] 8nm Silicon Nanowire NVM with ONO 1x10-4 1x10-5 10-6 10-7 10-8 10-9 10-10 ONO=3.8/6.4/5.1nm t Program =80µsec Initial 8V 10V V PG =12V L G =8nm W NW =7nm H NW =10.5nm -4-3 -2-1 0 1 2 3 Voltage [V] Drain Current [A/µm] 1x10-4 1x10-5 10-6 10-7 10-8 10-9 10-10 V ER =-12V 8nm L G with 7nm W NW using ONO-structure ONO=3.8/6.4/5.1nm t Erase =80µsec - Acceptable electrical performance by omega-gate -10-8V Initial L G =8nm W NW =7nm H NW =10.5nm -3-2 -1 0 1 2 3 Voltage [V] - Wide hysteresis shows the probability of multi-level NVM 14

8nm Si Nanowire Memory V T [V] 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ONO=3.8/6.4/5.1nm L G =8nm W NW =7nm V PG =12V V PG =10V V PG =8V 0.0 10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 10 0 10 1 Program Time [sec] V T [V] 3.0 2.5 2.0 1.5 1.0 L G =8nm W NW =7nm H NW =10nm -10V V ER =-8V 0.5 V 0.0 ER =-12V 10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 10 0 10 1 Erase Time [sec] 8nm NVM: V T window = 2V Program @ +12V/1msec, Erase @-12V/1msec 15

Scaling Limit in NVM Ultimately scaled NVM Nanowire + SONOS T O/N/O =15nm > L g =8nm Y.-K. Choi et. al., VLSI 2007 Scaling Limit! Close to end-point of semiconductor memory 16

Where to Go? Multi- Functioning This Work! Down Scaling Multi-Bit We are here 17

Example of Fusion Memory (System in Package) DRAM NVM SRAM. Increased volume Complexity in assembly Chip-to-chip interference 18 Wafer level fusion!!

Principal of SONOS Flash Memory S control oxide nitride tunnel oxide D S Substrate Substrate < program > < erase > D Drain current 1 0 S D Substrate Substrate < read 0 > < read 1 > S D voltage Features of flash High density Non-volatile Speed 19

Principal of Capacitorless DRAM S Buried oxide Substrate < program > D S D Buried oxide Substrate < erase > S Buried oxide Substrate < read 1 > Drain current D S 1 0 Time Buried oxide Substrate < read 0 > D Source : IEEE spectrum Dec. 2008 Features of 1T-DRAM High speed 2x denser than 1T/1C DRAM 5x denser than SRAM Volatile 20

Basis of URAM Operation (1) voltage capacitorless 1T-DRAM erase flash program capacitorless 1T-DRAM program Drain voltage flash erase Flash Memory Mode - Program : FN tunneling/hei - Erase : FN tunneling/hhi (High gate & drain voltage) Capacitorless 1T-DRAM Mode - Program : Impact ionization - Erase : Forward junction current (Low gate & drain voltage) Disturbance Issue - Periodical refresh 21

Multi-Function (URAM) Operation Flash Memory electron DRAM hole Oxide Nitride Oxide Source Drain 드레인 Quantum Barrier 22

Family of URAM S. Okonin et.al., SOI conf., 2001 R. Ranica et.al., VLSI symp., 2004 23

Process Flow (1) SOI Substrate Si Buried oxide Si substrate Initial SOI substrate SON Substrate SOC Substrate Si Si:C Si substrate Initial bulk substrate Si:C epitaxial growth Si epitaxial growth SOG Substrate Si n-well Si substrate Initial bulk substrate Buried n-well implantation p-body implantation 24 Si Si:Ge Si substrate Initial bulk substrate Si:Ge epitaxial growth Si epitaxial growth

Process Flow (2) fin Hole barrier layer Si substrate Puntch stop implantation Active fin lithography Photoresist trimming Fin patterning fin SiO 2 SiO 2 Hole barrier layer Si substrate HDP oxide deposition Oxide CMP Oxide recess formation 25

Process Flow (3) fin SiO 2 SiO 2 Hole barrier layer Si substrate fin SiO 2 SiO 2 Hole barrier layer Si substrate O/N/O formation Poly-Si deposition patterning S/D implantation Activation Forming gas annealing 26

URAM on SOI 27

Cross sectional view of URAMs URAM on SOI URAM on SiC URAM on Buried N-well URAM On SiGe 28

I D -V G Characteristics 10-5 V D = 1V Drain current, V D (A) I D 10-6 10-7 10-8 10-9 10-10 SOI SOC SOG SON 10-11 -1.0-0.5 0.0 0.5 1.0 1.5 2.0 voltage, V G (V) Superior device performances are result of the 3D nature. 29

I D -V D Characteristics (Kink) SOI SOC SON SOG 30

P/E in NVM Program Erase P/E sensing windows are acceptable. 31

Reliability in NVM Retention Endurance Data retention and endurance are acceptable. 32

Capacitorless 1T-DRAM in SOI FD URAM PD URAM 80 0.8V 0.7V 0.6V Source current, I S (µa) 33 70 60 50 40 30 I s =20µΑ I s =7µΑ SOI URAM 20 80msec 10 Conventional FinFET SONOS 0 0 50 100 150 200 Time, t (msec) < Potential Profile > < Measurement Results > Proposed SOI URAM exhibits wider sensing current window. Y.-K. Choi et. al., IEDM 2007

Capacitorless 1T-DRAM in SON bulk MOSFET Y.-K. Choi et. al., VLSI 2008 SON URAM Sensing current, I S (µa) 32 28 24 V SUB =0.5V 20 16 12 V SUB =0.3V I S =7µA I 8 S =4µA V SUB =0V 4 0 1 2 3 4 5 Time, t (msec) < Excess Hole Concentration > < Measurement Results > Capacitorless 1T-DRAM works in SON floating substrate. 34

Capacitorless 1T-DRAM at SOC SON URAM SOC URAM 28 Y.-K. Choi et. al., VLSI 2008 Source current, I S (µa) 24 20 16 12 8 4 V SUB =0.3V V SUB =0V I S =7µA I S =11µA 0 0 1 2 3 4 5 6 Time, t (msec) < Excess Hole Concentration > < Measurement Results > Capacitorless 1T-DRAM works in SOC floating substrate. 35

Summary IT only NT only NT IT 3nm MOSFET and 8nm non-volatile memory device were demonstrated. Close to scaling limit. Multi-function URAM was demonstrated. Continuously increased density 36

Built-in Potential for Buried n-well SIMS profile of the buried n-well Simulation Result of Energy Band 0.9 ev Built-in potential confines excess holes. 37

Band-Offset for Si:C & Si:Ge (2) 1.0 C-V for Band Engineered Substrate Energy Band Diagram Normalized capacitance 0.8 0.6 0.4 0.2 Si/Si 0.94 C 0.06 /Si E V =-180 mev Si/Si 0.7 Ge 0.3 /Si E V = 200 mev 0.0-3.0-2.5-2.0-1.5-1.0-0.5 0.0 0.5 voltage, V G (V) Double slopes appear in weak accumulation region. 38

Scaling Issue (Why capacitorless DRAM?) µ-processor Cache memory SRAM (32F 2 ) Cache portion 50% in 2006 Capacitorless DRAM is a promising candidate for embedded memory. 39 83% in 2008 90% in 2011 Is this memory or µ-processor? Cell capacitance in DRAM > 25fF (non-scalable) Cell area (~8F 2 ) is continuously reduced.

Soft-Program Issue voltage capacitorless 1T-DRAM erase flash program capacitorless 1T-DRAM program Drain voltage Disturbance Issue - Periodical refresh 1T-DRAM mode Initialization : Set V T of all cells to 0.2V by adjustment of trapped charges Capacitorless 1T-DRAM operation refresh yes Verification V T =0.2V? no 40

GIDL Program Method Bias (V) Source current, I S (µa) 2 1 0-1 -2 40 20 0 voltage Drain voltage Impact ionization 100 nsec 100 nsec 12µA 0 200 400 600 800 1000 Time, t (µsec) < 1T-DRAM Measurement > Threshold voltage, V T (V) 0.6 0.5 0.4 0.3 41 GIDL Impact ionization program V G =1, V D =2V GIDL program V G =-2V, V D =2V 10 0 10 1 10 2 10 3 10 4 10 5 Stress time, t (sec) < Soft-program test > GIDL program method mitigates soft-programming. [J.-W. Han et. al., EDL, Apr. 2009]

G-to-S/D Underlap Structure S Buried oxide Substrate D 50nm G-to-S/D Overlap junction S Buried oxide Substrate D S Buried oxide D G-to-S/D Underlap [J.-W. Han et. al., EDL, May 2009] 42

G-to-S/D Underlap Structure Bias (V) Source current, I S (µa/µm) 3 2-1 01-2 50 40 30 20 10 V G V D 50 nsec G-to-S/D underlap 8 µa G-to-S/D overlap 50 nsec 11 µa 0 0.0 0.5 1.0 Time, t (msec) < 1T-DRAM Measurement > Threshold voltage, V T (V) 0.45 0.40 0.35 0.30 0.25 0.20 0.15 Stress bais : V G =1V, V D =2.5V G-to-S/D overlap G-to-S/D underlap 10 0 10 1 10 2 10 3 10 4 10 5 Stress time, t (sec) < Soft-program test > G-to-S/D underlap structure mitigates soft-programming. [J.-W. Han et. al., EDL, May 2009] 43

S S Issue and Solution of G-to-S/D Underlap Device Structure Buried oxide Substrate SiO 2 Spacer Buried oxide Substrate SiO 2 D High-k D Source current (µa/µm) 120 100 80 60 40 underlap with HfO 2 spacer I S,HfO2 =55µA 20 0 underlap with I S,SiO2 =20µA -20 SiO 2 spacer read hold erase hold read -40 10 20 30 40 50 60 Time (nsec) 44 overlap with SiO 2 spacer I S =11µA High-k Spacer < Simulation Results > High-k spacer countervails the parasitic series resistance of the G-to-S/D underlap. [J.-W. Han et. al., EDL, May 2009]