58
Vijay K. Arora Vijay K. Arora Wilkes University Wilkes University E-mail: [email protected] E-mail: [email protected]

Vijay K. Arora Wilkes University E-mail: varora@wilkes

  • Upload
    dayton

  • View
    37

  • Download
    0

Embed Size (px)

DESCRIPTION

Vijay K. Arora Wilkes University E-mail: [email protected]. Emerging Technologies. Our Motivation and Economics. Adam Smith, “An Enquiry into Nature and Causes of the Wealth of Nations” (1776) The wealth is created by laisse-faire economy and free trade - PowerPoint PPT Presentation

Citation preview

Page 1: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Vijay K. AroraVijay K. AroraWilkes UniversityWilkes University

E-mail: [email protected]: [email protected]

Page 2: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Emerging TechnologiesEmerging Technologies

Page 3: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Our Motivation and Our Motivation and EconomicsEconomics

Adam Smith, “An Enquiry into Nature and Causes of the Wealth of Nations” (1776)The wealth is created by laisse-faire economy and free trade

John Maynard Keynes, “The General Theory of Employment, Interest, and Money” (1936)The wealth is created by careful government planning and government stimulation of economy

1990’s and BeyondThe wealth is created by innovations and inventions

Page 4: Vijay K. Arora Wilkes University E-mail: varora@wilkes

2020thth Century Paradigm Century Paradigm Formulate a hypothesis or theoryFormulate a hypothesis or theory

Accumulate dataAccumulate data

Do extensive experimentation and CheckDo extensive experimentation and Check

Publish if newsworthyPublish if newsworthy

Respect others’ work helping them to grow in the Respect others’ work helping them to grow in the professionprofession

Demonstrate character ethics that puts community Demonstrate character ethics that puts community interests above personal aggrandizementinterests above personal aggrandizement

Page 5: Vijay K. Arora Wilkes University E-mail: varora@wilkes

2121stst Century Paradigm Century Paradigm Formulate a hypothesis or theory or designFormulate a hypothesis or theory or design

Make a prototype structureMake a prototype structure

Patent itPatent it

Raise 17 million dollars and start an IPORaise 17 million dollars and start an IPO

Sue your competitor for stealing your ideaSue your competitor for stealing your idea

Demonstrate personality ethics that lubricates the Demonstrate personality ethics that lubricates the process of human interaction for personal process of human interaction for personal aggrandizementaggrandizement

Page 6: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Gross world product and Gross world product and sales volumessales volumes

Page 7: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Exponential GrowthExponential GrowthSIA roadmapSIA roadmap

Page 8: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Historical TrendsHistorical Trends

New Technology generation every three years

For each generation, memory density increase by 4 times and logic density increases by 2.5 times

Rule of Two: In every two generations (6 years), the feature size decreased by 2, transistor current density, circuit speed, chip area, chip current and maximum I/O pins increased by 2

Page 9: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Research ScenarioResearch Scenario A comprehensive transport theory for A comprehensive transport theory for

quantum processes at nanosaclequantum processes at nanosacle High-field distribution in quantum High-field distribution in quantum

wellswells Optimization of the shape and size of Optimization of the shape and size of

quantum wells for high frequenciesquantum wells for high frequencies Quantum Computing: Multi-state logic Quantum Computing: Multi-state logic

by using quantum statesby using quantum states Failure of Ohm’s Law: Re-assessment Failure of Ohm’s Law: Re-assessment

of the circuit theory principlesof the circuit theory principles

Page 10: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Goals for High Speed Goals for High Speed PerformancePerformance

Large transistor currentLarge transistor current• Time constantsTime constants• InterconnectsInterconnects• Cross talkCross talk

Reduced transit timeReduced transit time• Increased MobilityIncreased Mobility• High Saturation VelocityHigh Saturation Velocity• Reduced SizeReduced Size

Page 11: Vijay K. Arora Wilkes University E-mail: varora@wilkes

RC and Transit Time RC and Transit Time DelaysDelays

Source: CadenceSource: Cadence

Page 12: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Interconnect ProblemsInterconnect ProblemsRC Time DelaysRC Time Delays

RC time delay is increasing rapidlyRC time delay is increasing rapidly Wire resistance is rising Wires have larger cross-section …

introduce coupling Electromigration imposes current limits System performance, area and reliability

are determined by interconnect quality, not devices!!!

Page 13: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Increased cross-section improves performance but also increases noise and capacitive and inductive coupling

1

0.5

0.25

Increasin

g P

erform

ance

Decreasin

g C

ou

plin

g E

ffect

Interconnect PerformanceInterconnect Performance

Page 14: Vijay K. Arora Wilkes University E-mail: varora@wilkes

substrate

layer m

Cs CsCf CfCfCf

CcR1 R2

Cf

layer m

CoCf

CfCf

R3

layer n R4

Cint = Cf + Cs + Co + Cload

= Rint * ( Cint + Cc/(Cint+ Cc) )

= Rint * (Cint2 + Cint.Cc +Cc)/(Cint + Cc)

• Cc depends on dimensional shrink due to increased in cross-section• In VLSI, make Cc becomes insignificant as possible, then = Rint * Cint

RC Delay ConsiderationsRC Delay Considerations

Page 15: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Physical EffectsPhysical Effects

Quantum EffectsQuantum Effects nmfewaL D ,

TkqEorqE BD

cm

kV

m

V

L

VE 50

1

5

High-Field EffectsHigh-Field Effects

Field Broadening Field Broadening

Page 16: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Nano-Scale Nano-Scale Quantum EngineeringQuantum Engineering

Tkm

h

p

h

B

D

*3

Page 17: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Bulk SemiconductorsBulk Semiconductors

All 3 cartesian directions analog-type

DzyxL ,,

Density of States:

212

3

2

*24

1)( co

ec EE

h

m

dE

dN

VEg

Page 18: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quasi-Two-Dimensional Quasi-Two-Dimensional QWQW

z-direction digital-typex,y-directions analog-type

,......3,2,1

*2

)( 2222

n

nm

kkEE oz

e

yxconk

oz

2 2

2 me* Lz

2

DyxDz LL ,

oz

coec

EEInt

m

dE

dN

AEg

2

*

2

1)(

Density of States:

Page 19: Vijay K. Arora Wilkes University E-mail: varora@wilkes

AlGaAs/GaAs/AlGaAs AlGaAs/GaAs/AlGaAs Prototype Quantum WellPrototype Quantum Well

.

Pote

ntia

l

GaAs

AlGaAs

Ground State

x

y

x

y

z

Page 20: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quasi-One-Dimensional Quasi-One-Dimensional QWQW

y, z-direction digital-typex-directions analog-type (QWW)

,......3,2,1,

222

*

22

nm

nmm

kEE ozoy

xconk

e

2,

*

22

),( 2 zyezyo Lm

DxDzy LL ,

Density of States:

2

122

2/1*

1 )(21

)(

ozoycoe

xc nmEE

m

dE

dN

LEg

Page 21: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quasi-Zero-DimensionalQuasi-Zero-DimensionalQuantum WellQuantum Well

,......3,2,1,,

222

nm

nmEE ozoyoxcnk

2,,

*

22

),,( 2 zyxezyxo Lm

DzyxL ,,

All 3 cartesian directions digital-typeQuantum box (dot)

Page 22: Vijay K. Arora Wilkes University E-mail: varora@wilkes

AlGaAs

GaAs inside

Quantumwire

Quantum box

Quantum Well WireQuantum Well WireQuantum Box (Dot)Quantum Box (Dot)

Page 23: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quantum Well ArraysQuantum Well Arrays

Page 24: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Density of StatesDensity of States

N ( E) 1

Lx Ly Lz

E E s

0.0

0.2

0.4

0.6

0.8

1.0

1.2

DE

NS

ITY

OF

ST

AT

ES

( 1

026 e

V-1

m-3

)

0.0 0.2 0.4 0.6 0.8 1.0E - E c (eV)

3D2D1D

Page 25: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quantum Well with Finite Quantum Well with Finite BoundariesBoundaries

Lz 11

P

a

P

2m * E

2

1

2 a

2

Z n z 2

Lz

sinnz

Lz

Page 26: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Triangular Quantum WellTriangular Quantum Well

n

oonn z

zAi

zAizZ

2/1'

1)(

Ln 2

an2

zo

an 0.53556

Ai' n

Zn z 2

Ln

sinnz

Ln

Approximate:

Exact:

Page 27: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quantum-Confined Mobility Degradation

Changes in the Density of StatesChanges in the Density of States

DzD

z

isotropicbulk

QW LL

Changes in the relative strength Changes in the relative strength of each scattering interactionof each scattering interaction

Page 28: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Mobility Degradation Mobility Degradation Versus Quantum Versus Quantum

ConfinementConfinement

Page 29: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Gate-Field ConfinementGate-Field ConfinementMobility Degradation in a Mobility Degradation in a

TQWTQW

0.045

0.05

0.055

0.06

0.065

0.07

0.075

0.08

10 15 20 25 30 35 40 45 50

TheoryExperiment

MO

BIL

ITY

(m

2 /

V.s

)

ELECTRIC FIELD (V / m)

Page 30: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Electron and Hole Mobility in Electron and Hole Mobility in Submicron CMOSSubmicron CMOS

Courtesy: Y. Taur and E. Novak, IBM Microelectronics, IEDM97 Invited Talk.

Page 31: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Random Thermal MotionRandom Thermal Motion

0thv

smm

Tkv B

th /10*

3 5

Page 32: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quantum EmissionQuantum Emission

oQq EEq

oQ

EF

Page 33: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Randomness to Randomness to StreamliningStreamlining

Velocity Vectors in Equilibrium Randomness:

Velocity Vectors in a Very High Field Streamlined:

0 thd vv

2 ththd vvv

Page 34: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Saturation Velocity-BulkSaturation Velocity-Bulk

0 1)1(

1

x

j

j e

x

jF

2/13

2

FF1

thDvv

Tk

E

B

c

Fermi Integral

Normalized Fermi Energy

*

2

m

Tkv B

th

Page 35: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Saturation Velocity LimitsSaturation Velocity Limits

Bsat th *

8k T2v v

m

1

3

sat *

3 h 3nv

4 m 8

Non-degeneratelimit

Degeneratelimit

Page 36: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Saturation Velocity-Q2DSaturation Velocity-Q2D

0

2/12 2 F

FthD

vv

Tk

E

B

c

e 1ln0F

ozcoc EE

Page 37: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Saturation Velocity-Q1DSaturation Velocity-Q1D

2/1

01

1

FF

thDvv

Tk

E

B

c ozoycoc EE

Page 38: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Modeling TransportModeling Transport

c

thvv-

m

q

dt

dv

*

E

Transient Response:

c

t

c em

qv

1*

E

=0

EE oc

d m

qv

*

:ctStateSteady

Page 39: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Quantum EmissionQuantum Emission

Effective Collision time:

c

Q

eceff

1

Effective collision length:

o

o

Q

e

1

th

oQ vqE

oQq E

Eqo

Q

Page 40: Vijay K. Arora Wilkes University E-mail: varora@wilkes

1-D Random Walk in a 1-D Random Walk in a Bandgap semiconductorBandgap semiconductor

Page 41: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Modeling the DistributionModeling the Distribution

1

1

1

1 = ),(

x

Tk

q ee

f

B

E

E

o 1 eQ

o

cocoB

oo V

V

Tk

q

EEE

TkB

oQ

Tkx

B

Page 42: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Left-Right AsymmetryLeft-Right AsymmetryItinerant Electron Itinerant Electron

PopulationPopulation

)(cosh 2)(

e

ee

e

xn

xn

Page 43: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Streamlining the Streamlining the RandomnessRandomness

0

0.2

0.4

0.6

0.8

1

0 0.5 1 1.5 2 2.5

n+/n

n- /n

Page 44: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Drift-DiffusionDrift-Diffusion

dx

dnvq

vqxnxJ

th

th

tanh )( )(

tanhthd vv

otnothn VvD

**

n

cn

thn

ono m

q

m

q V

Drift Velocity

Diffusion

Drift

Diffusion Coefficient

q

TkV B

t

Page 45: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Single-Valley Single-Valley v-Ev-E CharacteristicsCharacteristics

Page 46: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Velocity-Field Velocity-Field CharacterisitcsCharacterisitcs

Page 47: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Effect of Degeneracy (2-D) Effect of Degeneracy (2-D)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 4 8 12 16 20

N=.01

N=.1

Nor

mal

ized

Dri

ft V

eloc

ity (

v d /(1/

2 vth

/2 )

)

N=1

Non-Degen

Tkm

h

B

D *2

2

DsnN

Page 48: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Mobility DegradationMobility Degradation

Page 49: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Diffusion Coefficient Diffusion Coefficient DegradationDegradation

Page 50: Vijay K. Arora Wilkes University E-mail: varora@wilkes

I-V Characteristics Microresistors

0.00

0.25

0.50

0.75

1.00

0.0

0

2.5

0

5.0

0

7.5

0

10

.00

V/Vc

Normalized I-V Characteristics

L=100 µm

L=10 µm

L=1 µm

I/Isat

Page 51: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Resistance Blow-UpResistance Blow-Up

0

2

4

6

8

10

0 0.2 0.4 0.6 0.8 1

R/Ro (Experiment)R/Ro(Theory)r/Ro(Experiment)r/Ro(Theory)

R/R

o

I/Isat

Page 52: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Multi-Valley Transport in Multi-Valley Transport in GaAsGaAsIntervalley Electron Intervalley Electron TransferTransfer

Page 53: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Multi-Valley Transport in Multi-Valley Transport in GaAsGaAsVelocity-Field Velocity-Field CharacteristicsCharacteristics

Page 54: Vijay K. Arora Wilkes University E-mail: varora@wilkes

High-Frequency TransportHigh-Frequency Transport

j tdc oE E e

E dc

o o

E

dc Conductivity Degradation

Ehf 2 2

eff

( E, )1

ac Conductivity Degradation

Page 55: Vijay K. Arora Wilkes University E-mail: varora@wilkes

ConclusionsConclusionsQuantum ConfinementQuantum Confinement

Transport properties function of Transport properties function of confinement length in QW’s because confinement length in QW’s because of the change in the Density of Statesof the change in the Density of States

Relative strength of each scattering Relative strength of each scattering changes.changes.

Electrons tend to stay away from the Electrons tend to stay away from the interface as wave function vanishes interface as wave function vanishes near the interfacenear the interface

Page 56: Vijay K. Arora Wilkes University E-mail: varora@wilkes

ConclusionsConclusionsHigh-Field Driven High-Field Driven

TransportTransport Electric field puts an order into otherwise Electric field puts an order into otherwise

completely random motioncompletely random motion

Higher mobility may not necessary lead to Higher mobility may not necessary lead to higher saturation velocity higher saturation velocity

Saturation velocity is limited by Fermi Saturation velocity is limited by Fermi /thermal velocity depending on degeneracy/thermal velocity depending on degeneracy

Saturation velocity is lowered by the Saturation velocity is lowered by the quantum remission processquantum remission process

RC time constants will dominate over RC time constants will dominate over transit time delay because of enhanced transit time delay because of enhanced resistanceresistance

Page 57: Vijay K. Arora Wilkes University E-mail: varora@wilkes

ConclusionsConclusionsFailure of Ohm’s LawFailure of Ohm’s Law

Effective resistance may rise Effective resistance may rise dramatically as current dramatically as current approaches saturation levelapproaches saturation level

Familiar voltage divider and Familiar voltage divider and current divider rule may not be current divider rule may not be valid on submicron scalesvalid on submicron scales

Page 58: Vijay K. Arora Wilkes University E-mail: varora@wilkes

Golden RuleGolden Rule

No matter what the size, make it smallerNo matter what the size, make it smaller

No matter what the speed, make it fasterNo matter what the speed, make it faster

No matter what the function, make it largerNo matter what the function, make it larger

No matter what the cost, make it cheaperNo matter what the cost, make it cheaper

No matter how little it heats up, make it No matter how little it heats up, make it coolercooler