inp-based hbts: devices and ghz mixed-signal ics...rodwell, short course, 2002 ieee/osa conference...

Rodwell, short course, 2002 IEEE/OSA Conference on Indium Phosphide and Related Materials, May, Stockholm

InP-based HBTs: Devices and GHz mixed-signal ICs

Mark RodwellUniversity of California, Santa Barbara

[email protected] 805-893-3244, 805-893-3262 fax


Applications:


Applications: optical fiber transceivers at 40 Gb/s and higher

Key advantages for:TIA, LIA, Modulator driver

Closer competition with SiGe:MUX/CMU, DMUX/CDRlower powerproblems with integration scale

"40 Gb" is often 44, 48, or 52…increases InP leverage over SiGe

80 & 160 Gb may come in timeworld may not need capacity for some timeWDM might be better use of fiber bandwidth

clockPLL

AD DMUX

O/E, E/O interfaces

MUX


Applications: military mixed-signal ICs

Radar/Comms transmitter electronicsdirect digital frequency synthesisaccumulator, sine ROM, DAC

Radar/Comms receiver electronicshigh resolution ADC

Technology requirements3,000 to 30,000 transistorsFew GHz IF (operating ) bandwidths~160 dB/Hz dynamic range

high resolution drives technologyspeed far beyond signal bandwidth

50-100 GHz clock rate digital technologiessought


Applications: wireless / RF

Present Wireless/RF ICsGaAs HBTs at lower frequenciesInGaAs PHEMTs in higher bands

Opportunities for InP33 GHz LMDS and 60 GHz metropolitan area networks (IEEE 802.16)cheap GaAs HBT processes → cheap InP HBT processes

200 GHz ft and fmax , 8 V BVCEOquick migration to 6" wafers enabled by metamorphic growth on GaAs

Longer-term opportunities for InPwider range of RF/wireless applications IF SiGe-like integration scales can be reached.


0.0 0.2 0.4 0.6 0.8 1.01E-6

1E-5

1E-4

1E-3

0.01

0.1

11 kmSea level

Log

Tran

smis

sion

Frequency, THz

mmWave Transmission UCSB

Atmospheric attenuation is LOW(~4 dB/km) at bands of interest

60-80 GHz, 120-160 GHz, 220-300 GHz

(Weather permitting)

Geometric path losses are LOWdue to short wavelengths.

55 mW transmitter power sufficient for 10 Gb/stransmission over 500 meters range.

Bit rate 1.00E+10 1/s eccarrier frequency 1.50E+11 HzF 10 dB receiver nois e figureDis tance 5.00E+02 m trans mis s ion rangeatmos pheric los s 4.00E-03 dB/m dB los s per unit dis tanceDant, trans 0.1 m trans mit antenna diameterDant, rcvr 0.1 m receive antenna diameterbits /s ymbol 1kT -173.83 dBm (1Hz)P rec -48.27 dBm received power at 10 -9 B.E.R∆f 1.00E+10 Hz RF channel bandwidth requiredtrans mis s ion -63.68 geometric path los s , dBatmos pheric los s 2 dB total a tmos pheric los s , dB

P transmitte r 55.1 mW required trans mitter power


TransistorFigures of Merit


Short-circuit current gain cutoff frequency

short-circuit current gain:drive input, short output, measure H21=Iout/Iin

Vgen

Rgen

IinIout

fF9.6=cbxCfF2.3=cbiCΩ= 48bbR

Ω= 7.4exR

ebmVg 'ebV '

E

B C

=diffCfF172

=jeCfF34

=πRΩ433

Ω= 7000cbr

Ω=

500cer

)exp( cmom jgg ωτ−=

fmodiff gC τ=

=πR mgβ /

( ) ( )τβ fjffH

//11)(21 +

≈

0

10

20

30

40

50

1 10 102

Gai

ns

(dB)

Frequency (GHz)

h21

U

VCE

= 1 V, JC = 1.5 mA/um2

fMAX

= 295 GHz

f τ = 295 GHz


Current-gain cutoff frequency in HBTs

+++++= collex

Ebc

Ejecollectorbase RR

qIkTC

qIkTC

fττ

π τ21

nbbase DT 22≈τ satccollector vT 2≈τ

RC terms are quite important for > 200 GHz fτ devicesfτ is a questionable metric for high speed digital logic

where capacitance charging has proportionally larger role


Stabilize transistor and simultaneously match input and output of device

Approximate value for hybrid-πmodel

To first order MSG does not depend on fτ or Rbb

Simultaneously match input and output of device

K = Rollet stability factor

Measurement of power gains and fmax

( )1KKSS 2

12

21 −−=MAG

+

≈==

cexcb

12

21

12

21

qIkTRωC

1YY

SS

MSG

For Hybrid- ππππmodel, MSG rolls off at 10 dB/decade, MAG has no fixed slope.So, NEITHER can be used to accurately extrapolate fmax

Maximum Available Gain

Transistor must be unconditionally stable or MAG does not exist

Maximum Stable Gain

ge ne ra tor

los s le s sm a tchin gne twork

R ge n

Vge n

los s le s sma tch ingne twork

R L

loa d

ge ne ra tor

los s le s sm a tchin gne twork

R ge n

Vge n

los s le s sma tch ingne twork

R L

loa d

re s is tivelos s

(s ta b iliz -a tio n)

Miguel Urteaga

MSG/MAG is however of direct relevance in tuned RF amplifier design


Use lossless reactive feedback to cancel device feedback and stabilize the device, then match input/output.

( )12212211

21221

GGGG4YY−

−=U

Unilateral Power GainMason’s Unilateral Power Gain

0

5

10

15

20

25

30

35

40

1 10 100

Gai

ns, d

B

Frequency, GHz

MAG/MSGcommon base

U: all 3

MAG/MSGcommon collector

MAG/MSGcommon emitter

For Hybrid- ππππmodel, U rolls off at 20 dB/decade

ALL Power Gains must be unity at fmax

U is not changed by pad reactances

g e n e ra to r

lo s s le s sm a tc h in gn e tw o rk

R g e n

Vg e n

lo s s le s sm a tc h in gn e tw o rk

R L

lo a d

s e rie sfe e d b a c k

s h u n tfe e d b a c k

Miguel Urteaga

Monolithic amplifiers not easily made unilateral, so U of only historical relevance to IC design. U is usually valuable for fmax extrapolation


Excess Collector Capacitance, Fmax, and Device Utility

0

5

10

15

20

25

30

35

40

1 10 100

Gai

ns, d

BFrequency, GHz

MAG/MSGcommon base

U: all 3

MAG/MSGcommon collector

MAG/MSGcommon emitter

+

≈==

cex qI

kTRcb12

21

12

21

ωC

1YY

SS

CEMSG

Cbe

Ccbi

Ccbx

gmVbeRbe

RbbB C

E

Rex

Rc

cbibbCRff

πτ

8max ≅

speed. logic digitalupon impact large a has circuits. wave-mmin gain usableon impact large has hence

MSG,emitter -commonupon impact large a has or U. upon effect no has

later. discussed be will and between ngpartitioni The

max

cbx

cbx

cbx

cbxcbi

C

CfC

CC

high fmax does not mean low Ccb or fast logic


Tuned ICs (MIMICs, RF):fmax sets gain,

& max frequency, not ft.low ft/fmax ratio makes tuning design hard (high Q)high Ccbx reduces MSG

What do we need: fτ τ τ τ , fmax , or ?

Lumped analog circuitsneed high & comparable ft and fmax.

Ccb/Ic has major impactupon bandwidth

Distributed Amplifiersin principle, fmax-limited, ft not relevant.(low ft makes design hard)

digital ICs will be discussed in detail later


transistorlayer structures


SHBT layer structure

very low breakdown: scaling beyond ~75 GHz digital clock rate very difficult

high collector-base leakageparticularly at elevated temperatures. Serious difficulties in real applications

very high thermal resistanceInGaAs collector and subcollector

Layer Material Doping Thickness (Å)

Emitter cap In0.53Ga0.47As 2 × 1019 cm-3: Si 300

N+ emitter InP 2 × 1019 cm-3: Si 700

N- emitter InP 8 × 1017 cm-3: Si 500

Emitter-base grade

In0.53Ga0.26Al0.21As to In0.455Ga0.545As

P: 4 × 1017 cm-3: Si N: 8 × 1017 cm-3: C

233 47

Base In0.53Ga0.47As N: 4 × 1019 cm-3: C 400

Collector In0.53Ga0.47As N: 2× 1016 cm-3: Si 2000

Subcollector InP N: 1× 1019 cm-3: Si ~1000 Å


DHBT Layer structure

B-C grade design is critical

InGaAs or GaAsSb bases GaAsSb more easily passivatedotherwise comparable

high breakdownimportant for microwave powerimportant for logic

low thermal resistanceessential for high power density important for microwave powerimportant for logic

Performanceft and fmax good or better than SHBTs

Layer Material Doping Thickness (Å)

Emitter cap In0.53Ga0.47As 2 × 1019 cm-3: Si 300

N+ emitter InP 2 × 1019 cm-3: Si 700

N- emitter InP 8 × 1017 cm-3: Si 500

Emitter-base grade

In0.53Ga0.26Al0.21As to In0.455Ga0.545As

P: 4 × 1017 cm-3: Si N: 8 × 1017 cm-3: C

233 47

Base In0.53Ga0.47As N: 4 × 1019 cm-3: C 400 Base-

collector grade

In0.53Ga0.47As to In0.53Ga0.26Al0.21As N: 2 × 1016 cm-3: Si 240

Pulse doping InP 5.6 × 1018 cm-3: Si 30

Collector InP N: 2× 1016 cm-3: Si 1,630

Subcollector InP N: 1× 1019 cm-3: Si ~1000 Å

emitter

emittercap

gradedbase

collectorsubcollector

PK Sundararajan


Alternative InP DHBT base-collector junction designs

Several layer alternatives exist for DHBTs with:high fthigh current densitynegligible current blockinglow base sheet and contact resistivity

InP collector

InAlAs/InGaAssuperlattice

InGaAssetbackInGaAs

base

UCSB: InGaAs base, MBE

IEDM 2001

Mattias Dahlstrom


transistorprocess flow(research-lab-like)


scale: 1x1 um

1) emitter metal liftoff


scale: 1x1 um

2) base recess etch


scale: 1x1 um

3) base metal deposition


scale: 1x1 um

4) collector mesa etch


scale: 1x1 um

5) etch through base to subcollector, collector Ohmics


scale: 1x1 um

6) liftoff base-collector vias


scale: 1x1 um

7) planarize: spin on BCB or polyimide & etch back


scale: 1x1 um

8) deposit interconnect metal


Problems with mesa process flowLarge Parasitic Collector Junction, Large Excess Ccbresembles Si bipolar processes of 1960's !parasitic collector junction lies under base contactsbase contacts must be nonzero size: nonzero resistivitybase contacts must be nonzero size: lithographic impact on yield

Self-aligned emitter-base process flowbase-emitter short-circuitsproblems with wet-etch undercut controlproblems with dry-etch reproducibility

Nonplanar processloss of yield in back-end process


scale:1x1 um

emitter-base junction is 3 um^2collector-base junction is 12 um^2

collector/emitter area ratioeven worsein non-self-aligned processes...

Problems with mesa process flow

While research-lab processes have moderate Ccb, processes aimed at high yield at >3000 HBTs have very large collector junctions

Ccb then the dominant circuit parasitic, regardless of impact on fττττ & fmax.


transistorkey parasiticsand model


Emitter Resistance

evidence of edge depletion or damage

( )( )( )WWL

R

WWLR

EcEex

EE

cex

∆−=

∆−=

ρ

ρ

1

?

( )( ) drop voltage excessive high but

low high :devices speedhigh speedfor scalingin factor limiting one :resistanceEmitter

exE

ccb

RIJICJ

→→

Dino Mensa


Current Gain: surface leakage

evidence of surface conduction

Be: InGaAs4E19/cm3 doping

Dino Mensa

Surface Conduction:InGaAs has low surface recombination velocity.InGaAs has surface pinning near conduction band.→ weak surface inversion layer on base, surface conduction to base contactProblem aggravated by InP emitter, as this also pins near conduction band

emitter

base

bulkbnpoE

poE

c

bulk

c

surface

c

b

WDqnAqnkP

II

II

II

β

β1

)/()(

1

1 +=

+==


Current Gain: Auger recombination

ime transit tlow y,resistivitcontact base low basethin doping basehigh But,

doping basehigh through reduction constrains This

1/ )(1

.. 1 Since

1Augerby dominatedion recombinatBulk

feasible cm/10 above :doping baseCarbon

22

2base

2Auger

220

⇒+

∝∝

∝

∝

sheet

sheetBA

B

A

TNT

N

ρ

ρβτ

τ


Base Transit Time

( )( )

( )gbg

g

LWsatgngngbb

EkTWLL

evLDLDLW gb

∆=

−−−= −

/:length grading theis where

1/// /2τ

fs 35/ifcorrect modeldiffusion -Drift

* ≈≈>> kTmDτ nmb τ

cm/s 103

secV/cm 40:Assumes

7

2

⋅=

−=

exit

N

vD

Dino Mensa


Base Bandgap vs. Doping Grading

.resistancesheet degrading without fieldin -built introduceCan

)( maximum sets collapse induced-Auger doping,heavy With

field big induces change doping small :statistics doping Degeneratebase degenerate grading, doping base :3 Case

base of middlein somewhere land contacts :increased resistanceContact resistancesheet base increasedgreatly

1:12.01:by base of sidecollector at doping Reduceyreliabilitgrowth / by dconstraine sideemitter near doping Base

base degenerate-non grading, doping base :2 Case

(strained) AsGaIn AsGaIn :ratio Ga:InVary grading. bandgap base :1 Case

base. across drop potential meV 52 a introduce :Objective

0

2

0.470.530.5450.455

∫

⇒⇒

=

↔

−

bTdxxp

e

β


V=0

TcTc/2

--

+ +

V=0

Tc/4

+ +

--

+ +

----

+

+

Collector Transit TimeT. Ishibashi

base.near velocity tosensitive more is

2)()/1(

sketch) (refer to ticselectrosta elementary From

c

0c

τ

τ ∫ ≡−=CT

eff

cc

vTdx

xvTx

.scattering L- todue decreasesthen high, is velocity initial as ,Fortuitous

Γ


Collector Transit Time

layers A 2000~for cm/s105-3 also collectors InPin Velocities

layers A 2000~for cm/s105-3 collectors InGaAsin Velocities

atad RF fit tobest ...from

7

7

&

&

⋅

⋅


Current-induced Collector Velocity Overshoot (?)

0

0.5

1

1.5

2

2.5

0 0.5 1 1.5 2 2.5 3 3.5

tau ec

, ps

inverse current density, 1/J, µm2/mA

J=0

J= 8 mA/um2

bias. with variesalso )./ asnot vary doesoften (1/

current biash factor witty idealiemitter in modulation todue be

alsomay variation :BUT trend.predicted show *does* data

Ishibashiby predictedEffect

ec

ec

je

Eexm

CqIkTRg +

ττ

Mattias Dahlstrom

300 Å InGaAs base2000 Å InP collector280 GHz peak fτ


Base-Collector Distributed Model: exact

This "mesh model" can be entered into a microwave circuit simulator (e.g. Agilent ADS)to predict fmax , etc.


Components of Rbb and Ccb

ceeecb TWLC /, ε=

!contcontverthorizx RRRRR =+=

Ebcshoriz LWR 2/ρ=Pulfrey / Vaidyanathan

Miguel Urteaga


Components of base spreading resistance

. negligibleeven with emitters,

narrow with obtained is Low

.length emitter increasedin results dth emitter wi decreased

:fixed is areaemitter Given that

. and by dominated is doping) base 1E20~(or with

emitterssubmicron With

spread

bb

E

E

EEE

gapcontactbb

RR

LW

WLA

RRR

⇒

=


Typical base parameters

(Dvorak)s/cm 20 fs, 200150 ,m-Ohm 20 , eOhms/squar 0001

sA thicknes 250 grading, meV ?? base, GaAsSb doped-C /cm108

s/cm 40 fs, 100 ,m-Ohm 10 , eOhms/squar 700sA thicknes 300 grading, (doping) meV 52 base, InGaAs doped-C /cm107

s/cm 40 fs, 170 ,m-Ohm 100 , eOhms/squar 750sA thicknes 400 grading, meV 52 base, InGaAs doped-Be /cm104

22cs

319

22cs

319

22cs

319

≈−≈≈=

⋅

≈≈<=⋅

≈≈==⋅

nb

nb

nb

D

D

D

τµρρ

τµρρ

τµρρ


Pulfrey / Vaidyanathan fmax model

( )

speed analog and digitaluponimpact bigbut

,upon effect moderate has

small relatively is timecharging associated the...

1

. than less ,resistance low relatively a through charged is

ecapacitanc external that theNote

max,

,,

,

fC

T

RCRRCR

C

extcb

contactc

vertextcbvertcontextcb

vert

extcb

ρε=

<

Pulfrey / Vaidyanathan


Ccb Cancellation by Collector Space-Charge

collector space-charge layer

cbVδ

∂

∂−ε=∂

∂

sat

cc

cbccb

base

vTI

VTA

VQ

2

cb

cc

ccb V

ITAC

∂τ∂−ε=⇒

1.8

2

2.2

2.4

2.6

2.8

0 1 2 3 4 5 6 7

measured0.64 fF decrease

Ccb

tota

l

Ic

Collector space charge screens field, Increasing voltage decreases velocity, → modulates collector space-charge→ offsets modulation of base charge→ Ccb is reduced

Moll & Camnitz, Betser and Ritter

Even if you don't care about fmax,the effect can confuse HBT model extraction


equivalentcircuitmodel


Transistor Hybrid-Pi equivalent circuit model

Cbe

Ccbi

Ccbx

gmVbeRbe

RbbB C

E

Rex

Rc

kTqIg Em /0 =

)( cbmjebe gCC ττ ++=

)(0

cbjmm egg τγτω +−=


Comments regarding the Hybrid-Pi model

networkcollector -base ddistribute theto fitsrepresent and

design) ICfast in (importantdelay )(0.2 ~ associatedan requires also ssneverthelegenerator The

impedanceinput on

)( ofeffect themodels ecapacitanc The

.in order first toT thefit to a from results model pi-hybrid The

transportdependent -frequency modelsdirectly model (T) base-common The

cb

cb,

RCCCR

g

C

cbxcbibb

m

diffbe

ττ

ττ

ω

+⋅

+


Collector field-screening(Kirk Effect)


Kirk effect in DHBTs: not base pushout, but current-blocking

cm/s 104 ,kA/cm 1000 V, 2.1 7,

2 ⋅=== effsatece vJV

cm/s 104 ,kA/cm 0 V, 7.0 7,


cm/s 104 ,kA/cm 1000 V, 7.0 7,


τβ

εερφ

f

VJ

vJqNdxd

ce

d

and in decrease

blockingcurrent in results lowand high under gBandbendin

/

2

2

⇒

−==

Mattias Dahlstrom


Kirk effect in DHBTs

0

2

4

6

8

10

12

14

0

1 105

2 105

3 105

4 105

0 1 2 3 4 5 6

Je (A/cm

2 )I C (m

A)

VCE

(V)

0

50

100

150

200

250

0 1 105 2 105 3 105 4 105 5 105 6 105

Cut

off F

requ

enci

es, G

Hz f

τ at V

ce=1.5 V

fmax

at Vce

=1.5 V

Current Density, A/cm2

fmax

at Vce

=0.7 V

fτ at V

ce=0.7 V

ceV

Jff

increasedwith increases esholdeffect thr-Kirk

lower at and in Decrease maxτ

( )CEEeffective

effectivesat

c

c

ce

satce

TWLA

AvTR

dIdV

JV

2is areaflux current

collector effective thewhere2

increased with in Increase2

chargespace

,

+≈

== − ε

2min,

2min,max

/)(2

/)2(2

ccecesat

ccbcbsat

TVVv

TVVvJ

+≅

++=

ε

φε

Young-Min Kim


Kirk effect in SHBTs: base pushout, increased Ccb

increases. electrons compensate Holes

out. pushes Base

cbC

cm/s 103 ,kA/cm 500 V, 7.0 7,


Base pushes out

Ccb increases Ccb increases


Kirk effect with Nonuniform Collector Electron Velocity

base.near velocity tosensitive more are and

2)()/1(

analysis, ime transit tFrom

c

0c

eff

T

eff

cc

vvTdx

xvTxC

τ

τ ∫ ≡−=

2min,

2min,max

/)(2

/)2(2:velocitycollector uniformh effect witKirk

ccecesat

ccbcbsat

TVVv

TVVvJ

+≅

++=

ε

φε

2min,

2min,max

/)(2

/)2(2:velocitycollector NONuniformh effect witKirk

cceceeff

ccbcbeff

TVVv

TVVvJ

+≅

++=

ε

φε

Nonuniform collector electron velocity doesn't profoundly change Kirk effect


transistorscaling theory


HBT scaling: layer thicknessesWE

WBC

WEB

∆x

L E

base

emitter

base

collector

WCreduce Tb by √2:1→ → → → τb improved 2:1

reduce Tc by 2:1→ → → → τc improved 2:1

note that Ccb has been doubled..we had wanted it 2:1 smaller

nbb DT 2/2≅τ

satcb vT 2/≅τ

2:1 improved device speed: keep G's, R's, I's, V's constant, reduce 2:1 all C's, τ 's

EC WW ~ Assume

Rodwell


HBT scaling: lithographic dimensions

WE

WBC

WEB

∆x

L E

base

emitter

base

collector

WC

Ccb/Area has been doubled..we had wanted it 2:1 smaller…must make area=LeWe 4:1 smaller→ → → → must make We & Wc 4:1 smaller

Everticalcsheet

contact

contactspreadgapbb

L

RRRRR

2,ρρ=

≅

++=Base Resistance Rbb must remain constant→→→→ Le must remain ~ constant

reduce collector width 4:1reduce emitter width 4:1keep emitter length constant


EC WW ~ Assume

Rodwell


HBT scaling: emitter resistivity, current density

WE

WBC

WEB

∆x

L E

base

emitter

base

collector

WC

Emitter Resistance Rex must remain constantbut emitter area=LeWe is 4:1 smallerresistance per unit area must be 4:1 smaller

increase current density 4:1reduce emitter resistivity 4:1


Collector current must remain constantbut emitter area=LeWe is 4:1 smallerand collector area=LcWc is 4:1 smallercurrent density must be 4:1 larger

EC WW ~ AssumeRodwell


Scaling Laws, Collector Current Density, Ccb charging time

baseemitter

collector

subcollector

baseemitter

collector

subcollector

Collector Field Collapse (Kirk Effect)

Collector Depletion Layer Collapse

)2/)(/( 2 εφ cdsatcb TqNvJV −+>+

)2/)(( 2min, εφ cTqNV dcb +>+

2min,max /)2(2 ccbcbsat TVVvJ φε ++=⇒

Collector capacitance charging time is reduced by thinning the collector while increasing current

( )( ) ( )

+

∆=∆=∆sat

C

CECE

LOGICCLOGICcCLOGICcb v

TAA

VVVIVTAIVC

2/

emitter

collector

min,collectorε

cecbbe VVV ≅+≅ )( hence , that Note φφ

Rodwell


for x 2 improvement of all parasitics: ft, fmax, logic speed…base √2: 1 thinnercollector 2:1 thinneremitter, collector junctions 4:1 narrowercurrent density 4:1 higheremitter Ohmic 4:1 less resistive

Scaling Laws for fast HBTs

Challenges with Scaling:Collector mesa HBT: collector under base Ohmics. Base Ohmics must be one transfer lengthsets minimum size for collector Emitter Ohmic: hard to improvehow ?Current Density: dissipation, reliabilityLoss of breakdownavalanche Vbr never less than collector Egap (1.12 V for Si, 1.4 V for InP) .sufficient for logic, insufficient for power


digital circuitspeed


10.0

100.0

Dec-80 Dec-85 Dec-90 Dec-95 Dec-00 Dec-05

Divi

der F

requ

ency

(GHz

)

III-V HBTSi BJTSiGe HBTFET/HEMTCMOS

Benchmark: master-slave flip-flop configured as 2:1 static frequency dividerSource: M Sokolich, HRL, Rodwell, UCSB

Logic Speed: III-V vs. Silicon

0.1 um

0.15 um


75 GHz HBT master-slave latchconnected as Static frequency divider

technology:400 Å base, 2000 Å collector HBT0.7 um mask (0.6 um junction) x 12 um emitters1.5 um mask (1.4 um junction) x 14 um collectors

transistor performance:1.8×105 A/cm2 operation, 180 GHz ft, 260 GHz fmaxcollector/ emitter junction area ratio: 2.7:1 (low)Ccb/Ic: 0.9 ps/VRex*I=54 mV

simulations: 95 GHz clock rate in SPICE

UCSBThomas Mathew

Michelle LeeHwe-Jong Kim

-0.14

-0.12

-0.1

-0.08

-0.06

-0.04

-0.02

0

0 50 100 150 200

fin=75GHz, f

out=37.5GHz

Vout

(Vol

ts)

Time(ps)

-0.11

-0.1

-0.09

-0.08

-0.07

-0.06

-0.05

-0.04

-0.03

0 50 100 150 200

fin=69GHz, f

out=34.5GHz

Vout

(Vol

ts)

Time(PS)

modulation is synthesizer 6 GHz subharmonic

3.92 V, 224 mA, 0.88 W

~3.5 dBm input power


Neither fτ nor fmax predicts digital speed

Ccb∆Vlogic/Ic is very important→ collector capacitance reduction is critical→ increased III-V current density is critical

Rex must be very low for low ∆Vlogic at high Jc

InP: Rbb , (τb+τc) , are already low, must remain so

What do we need for fast logic ?

clock clock clock clock

inin

out

outECL M-S latch

( )

( )

( )

+>∆

∆

+

+

+

+

∆

cexLOGIC

LOGIC

Ccb

becb

becbC

LOGIC

IRq

kTV

VIR

CCR

CCI

V

6

leastat bemust swing logic The

resistance base the throughcharge stored

collector base Supplying

resistance base the throughcharging ecapacitancDepletion

swing logic the throughcharging ecapacitancDepletion

:by DeterminedDelay Gate

bb

depletion,bb

depletion,

ττ

Yoram Betser, Raja Pullela


What HBT parameters determine logic speed ?

( )exCicex

bbcbc

cbdiffcbje

RIqkTVRRIV

CCC

+•>∆

+∆

+

/6 as effect,indirect strong very has

from 17% ,)( from 12% ,/ from 68% :imes transit tand sresistanceby Delays Sorting

) (e.g. charging 18%only , charging 38% , charging 44% :escapacitancby Delays Sorting

log

logic ττ

ττ

Caveats: assumes a specific UCSB InP HBT (0.7 um emitter, 1.2 um collector 2kÅ thick, 400 Å base, 1.5E5 A/cm^2)

ignores interconnect capacitance and delay, which is very significant

Cje Ccbx Ccbi (τb+τc) ( I/∆V) total∆V/ I 33.5% 6.7% 27.8% 68.4%∆V/ I 12.3% 12.3%(kT/q) I 1.4% 0.1% 0.4% 0.5% 2.5%Rex -1.3% 0.1% 0.3% 0.9% 0.1%Rbb 10.2% 2.8% 3.7% 16.7%total 43.8% 6.8% 31.3% 17.5% 100.0%

38%

Yoram Betser, Raja Pullela


Logic Speed

Caveat: ignores interconnect capacitance and delay, which is very significant

( )exLOGIC

clock

jiijclockgate

ij

JqkTVf

crafTa

ρ+>∆

Σ==

/6 isswing voltagelogic minimum The frequency.clock maximum theis

where,2/1 form theof isdelay Gate analysis. handby

found flop,-flip slave-master ECLan for tscoefficiendelay eApproximat

max,

Yoram BetserRaja Pullela


Logic Speed: definition of terms

) area"-per" (e.g. areaemitter timesresistanceemitter :) area"-per" (e.g. areaemitter timesresistance base :

swing voltagelogic :areaemitter unit per current emitter :

mestransit ticollector and base of sum :areaemitter unit per ecapacitanc basecollector extrinsic :

areaemitter unit per ecapacitanc basecollector intrinsic :

areaemitter unit per ecapacitancdepletion baseemitter :

exex

bbbb

LOGIC

f

cbx

cbi

je

RRr

VJ

ccc

ρ

τ

∆


Why isn't base+collector transit time so important ?

reductionsuch no see escapacitancDepletion

1:10~ is which ,/

of ratioby reduced ecapacitancdiffusion signal-Large

)()(Q

:Operation Signal-LargeUnder /

)()()(Q

:Operation Signal-SmallUnder

base

base

∆

∆∆

+=+=∆

+=+=+=

qkTV

VV

II

VqkT

IVdVdII

LOGIC

LOGICLOGIC

dccbCcb

beCcb

bebe

CcbCcb

ττττ

δττδττδττδ


roadmap


Technology Roadmaps for 40 / 80 / 160 Gb/s

Rodwell


Technology Roadmaps for 40 / 80 / 160 Gb/s

80 Gb/s technology node:Change from 40 Gb/s does not fully follow scaling laws. Why ?Lithographic scaling eased by carbon base doping.Current density scaling eased by reduced excess collector area.

160 Gb/s technology node:Direct application of scaling laws.Aggressive current density and lithographic scaling required.If further improved base contact resistance → relax lithographic scalingFurther reduce Acollector/Aemitter ratio → relax current density scalingnote that Acollector/Aemitter <2.5 looks hard at deep submicron.


device structures


InP mesa HBT

BCB

collector


0.5 µm emitter, 0.25 µm basecontacts

Narrow-Mesa HBTs: high fτ & fmax if high base doping

Mattias Dahlstrom


Low Ccb HBT structures

emitterbase contact

collectorcontact

SI substrate

InGaAs subcollector

InP collector

InGaAscollector

InP subcollector

InGaAs baseundercut

collector junction

Narrow-mesa with ~1E20 carbon-doped base

undercut-collector

transferred-substrate Extremely high demonstrated fmax75 GHz (record) static frequency dividers

Too low yield for manufacturing (?)

Pursued by several research groups

Also has uncertain yield at submicron geometries

The conservative device structure

Yet, I assert that even this device is notviable of mass manufacturing if > 3000 transistors per IC are sought


yield and fabrication


InP HBT limits to yield: non-planar processEmitter contact

Etch to base

Liftoff base metal

Failure modes

Yield degrades as emitters arescaled to submicron dimensions

base contact

emittercontact

base contact

S.I. substrate

base

sub collector

S.I. substrate

base

sub collector

S.I. substrate

base

sub collector

emitter

S.I. substrate

base

sub collector

Emitter planarization, interconnects

base contact

liftoff failure:emitter-baseshort-circuit

S.I. substrate

base

sub collector

base contact

excessiveemitter undercut

S.I. substrate

base

sub collector

S.I. substrate

base

sub collector

planarization failure: interconnect breaks


InP

Front and side views

InAlAs

Front and side views

0.7 um

0.5 um metal~0.4 um junction

0.6 um metal~0.4 um junction

Smaller emitters → lower yield. Need better fabrication process


InP vs. SiGe Digital InP has slightly higher speed, much less powerInP can't meet integration scales of many complex fast ICs

Analog:Combined InP speed and breakdown are key advantages

mm-wave wireless / RF (60 GHz, etc)No significant market yetInP HBT could be strong contender (fast and cheap)

Fast, high-yield InP HBT IC processes are critically needed


InP vs. SiGe III-V literature, III-V research community:large inherent advantages in transport parameters over Siresearch focused on transport physics, poorly tied to circuit design→ → → → devices not well-tuned for circuits, poor parasitic reduction → → → → university-like fabrication, low yield, low scales of integration

Silicon research communityfocused on SCALING, closely tied to circuit design, focus on YEILDstrong extrinsic parasitic reductionresult: very good SiGe HBT digital circuit speed, large fast ICs

InP HBT has fundamental advantages which will allow it to scale beyond SiGe HBT scaling limits, but must address:

yield: Silicon-like planar implanted / regrowth processesspeed: device scaling informed by understanding of circuit design


InP vs. Si/SiGe HBTs: materials vs. scaling advantagesGood:Narrow emitter: 0.18 um High current density: 10 mA/um2

Large emitter contact: low resistancePolysilicon base contact: low resistance SiO2 trenches: small collector capacitancePlanar device : high yieldBad:High base sheet resistance, Low electron velocity, low breakdown limits scaling.Equal speed at 5x smaller scaling.Loss of breakdown may soon slow scaling

Good:20x lower base sheet resistance, 5 x higher electron velocity, 4x higher breakdown-at same ft.

Bad:Presently only scaled to ~ 1 umArchaic mesa fabrication process:

large emitters, poor emitter contact:low current density: 2 mA/um2

high collector capacitance nonplanar device : low yield


InAlAs/InGaAs/InP DHBT with polycrystalline extrinsic emitter regrowth.

emitterbase

collector

poly-InAsextrinsic emitter

subcollector

UCSBDennis Scott

ONR


thermal resistance andthermal runaway


Thermal resistance and effect of subcollector

collector

BInPB

B

BInPBBInGaAs

InGaAs

EEInP

cInPJA LKW

LLKLWK

TLWK

Tππ

θ 1ln12/, +

++≈

LB

WBApproximation:InGaAs dominates thermal resistance → heat flows through InGaAs in areaequal base mesa (excluding pad)

m- W/k68 m- W/k5V 1.2 ,mmA/ 4

m, 25.3 m, 7.0 m, 3 m, 5.02

====

====

InPInGaAs

CEE

EBEE

KKVJ

LWLWµ

µµµµ

7 127 16 117 13 16 11

∆T, 2000 Å InGaAs

∆T, 200 Å InGaAs

InGaAs subcollector

W. Liu, H-F Chau, E. Beam III, "Thermal properties and thermal instabilities of InP-based heterojunction bipolar transistors", IEEE Transactions on Electron Devices, vol.43, (no.3), IEEE, March 1996. p.388-95.


8 finger common emitter DHBTEmitter size: 16 um x 1 um Ballast resistor (design):9 Ohm/finger

0

20

40

60

80

100

120

0 1 2 3 4 5

I c, mA

Vce

, Volts

Ibstep = 380 µA

0

5

10

15

20

25

0 1 2 3 4 5 6

I c, mA

Vce

, Volts

Ibstep = 300 µA

0

5

10

15

20

25

1010 1011

Gai

ns, d

B

Frequency, Hz

H21

U

fmax

=120 GHz

fτ=91 GHz

Jc=5e4 A/cm2

Vce=1.5 V

Poor performance observed in multi-finger DHBT

current hogging observedfmax also low due to high

base feed resistance

UCSBYun Wei

ARO MURI



Excess Rbb, hence reduced fmax (big HBT has big Ccb, small Rbb, hence even small excess Rbb reduces fmax)

base

emitter

base feed-in contact

Restrictions on DHBT sizing: distributed base feed resistance

base contact

Self-aligned base contact thickness=0.08 µmLeads to feed sheet resistance:ρ = 0.3 Ω/restricts emitter length to ~15 µm

UCSBYun Wei

ARO MURI


DHBT thermal stability: multiple emitter fingers UCSBYun Wei

ARO MURI

1/

unless Unstable

/

1

constant at mV/K 1.1

fingers 2between difference re temperatuinitial Assume

stability thermal <++

=

=⇒=⇒++

=⇒=⇒

−=

Eballastex

JACEbe

JACCE

beEballastex

Cbe

be

cbe

qIkTRRV

dTdVK

PTIVP

VqIkTRR

ITdT

dVVT

IdT

dVT

θ

δθδδδ

δδδδδ

δ



With long emitter finger, current-crowding can occur within finger Long finger: temperature can vary along length of emitter finger

loss of strong thermal couplingTemperature gradients along finger results in nonuniform current distribution

center of stripe gets hotter → carries more current → gets hotter → Premature Kirk-effect-induced collapse in ft.

Thermal runaway within a finger UCSBYun Wei

ARO MURI

emitter


0

0.01

0.02

0.03

0.04

0.05

0 0.5 1 1.5 2

I_DC

SRC1

I_Probe

I_Probe1

I_Probe

I_Probe2

V_DC

SRC2

Current hogging observation: multi-finger DHBT UCSBYun Wei

ARO MURI



Measuring DHBT thermal resistance

∆Vbe

Ic

UCSBYun Wei

ARO MURI

( )

( )mV/K 1.11

mV/K 1.1

fixed

fixed

−×=⇒

⋅−==

CICE

beJA

CECJACECE

beIbe

IdVdV

VIVdVdP

dPdT

dTdVV

c

c

θ

δθδδ



Large current high breakdown voltage broadband InP DHBT

Objectives: fmax>300 GHz, BVCEO>6 V, Jmax~1x105 A/cm2

Approach: transferred-substrate multi-finger InP DHBTs, HBT thermal analysis

Simulations: large signal HBT spice model

Accomplishments:fmax>330 GHz, Bvce>7 V, Jmax>1x105 A/cm2

UCSBYun Wei

128 µµµµm2 common base DHBT

0

20

40

60

80

100

120

140

0 2 4 6 8 10

Ic, m

A

Vcb, V

AE=128 um2

0

2

4

6

8

10

12

14

-1 0 1 2 3 4 5 6 7

Ic, m

A

Vcb, V0

5

10

15

20

25

30

100 101 102 103

U, M

SG/M

AG, d

B

Frequency, GHz

MSG/MAGU

fmax=330 GHz

AE=128um2

IC=100mAVcb=2.9V

ARO MURI


On-wafercharacterizationof HBTs

accurate andotherwise


Ultra-high fmax Submicron HBTs

Electron beam lithography used to define submicron emitters and collectors

Minimum feature sizes⇒ 0.2 µm emitter stripe widths ⇒ 0.3 µm collector stripe widths

Improved collector-to-emitter alignment using local alignment marks

Aggressive scaling of transistor dimensions predicts progressive improvement of fmax

As we scale HBT to <0.4 um,fmax keeps increasing,measurements become very difficult

0.3 µµµµm Emitter before polyimide planarization

Submicron Collector Stripes(typical: 0.7 um collector)

Miguel Urteaga


HP8510C VNA, Oleson Microwave Lab mm-wave Extenders

GGB Industries coplanar wafer probes

connection via short length of WR-5 waveguide

Internal bias Tees in probes for biasing active devices

75-110 GHz set-up is similar

140-220 GHz On-Wafer Network Analysis

UCSB 140-220 GHz VNA Measurement Set-up


0

5

10

15

20

25

30

35

1 10 100Frequency, GHz

MSG

h21

Mason'sGain, U

Submicron HBTs have very low Ccb (< 3 fF) Characterization requires accurate measure of very small S12 Standard 12-term VNA calibrations do not correct S12 background error due to probe-to-probe coupling

SolutionEmbed transistors in sufficient length of transmission line to reduce coupling

Place calibration reference planes at transistor terminals

Line-Reflect-Line CalibrationStandards easily realized on-wafer

Does not require accurate characterization of reflect standards

Characteristics of Line Standards are well controlled in transferred-substrate microstrip wiring environment

Accurate Transistor Measurements Are Not Easy

Transistor in Embedded in LRL Test Structure

230 µm 230 µm

Corrupted 75-110 GHz measurements due toexcessive probe-to-probe coupling

Miguel Urteaga


140 150 160 170 180 190 200 210 220

freq, GHz

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

0.30

freq (75.00GHz to 110.0GHz)

Can we trust the calibration ?

freq (140.0GHz to 220.0GHz)

S11 of throughAbout 40 dB

140-220 GHz calibration looks OK75-110 GHz calibration looks Great

S11 of openAbout 0.1 dB / 3o error

dBS21 of through line is off by less than 0.05 dB

S11 of openS11 of short S11 of through

75 80 85 90 95 100 105 110

freq, GHz

-70

-65

-60

-55

-50

-45

-40

Probe-Probe couplingis better than 45 dB

Miguel Urteaga


140-220 GHz Calibration Verification:Measurement of Thru Line after Calibration

S11, S22 (dB)

Magnitude S21 (dB) Phase S21 (degrees)

Miguel Urteaga


transistorresults


Ultra-high fmax Transferred-Substrate HBTs Substrate transfer provides access to both sides of device epitaxy

Permits simultaneous scaling of emitter and collector widths

Maximum frequency of oscillation

⇒

Sub-micron scaling of emitter and collector widths has resulted in record values of extrapolated fmax

Extrapolation begins where measurements end

New 140-220 GHz Vector Network Analyzer (VNA) extends device measurement range 0

5

10

15

20

25

30

10 100 1000

Gai

ns, d

B

Frequency, GHz

fmax= 1.1 THz ??fτ

= 204 GHz

Mason's gain, U

H21

MSG

Emitter, 0.4 x 6 µm2

Collector, 0.7 x 6 µm2

Ic

= 6 mA, Vce

= 1.2 V

3000 Å collector400 Å base with 52 meV gradingAlInAs / GaInAs / GaInAs HBT

cbbbCRff πτ 8/max ≅

Michelle Lee


Submicron InAlAs/InGaAs HBTs: Unbounded Unilateral power gain 45-170 GHz

UCSBONR

emitter

collector

Miguel Urteaga

0

10

20

30

40

10 100 1000

Tran

sist

or G

ains

, dB

Frequency, GHz

U

UMSG/MAG

H21

unbounded U

Collector: 3000 A thickness , 1016/cm3 doping

Collector pulse doping: 50 A thickness 1017/cm3 doping, 250 A from base

Vce

= 1.1 V, Ic=5 mA

0.3 µm x 18 µm emitter, 0.7 µm x 18.6 µm collector,


Negative Unilateral Power Gain ???

YES, if denominator is negative

This may occur for device with a negative output conductance (G22) or some positive feedback (G12)

( )12212211

21221

GGGG4YY−

−=U

( )[ ]1221L2211

21221

GGGGG4YY

−+−

=U

2-portNetwork GL

Select GL such that denominator is zero:

Can U be Negative?

What Does Negative U Mean?Device with negative U will have infinite Unilateral Power Gain with the addition of a proper source or load impedance

AFTER Unilateralization Network would have negative output resistance Can support one-port oscillation Can provide infinite two-port power gain

∞=USimple Hybrid- ππππ HBT model will NOT show negative U

Miguel Urteaga


DC-200 GHz parameters of 0.3 µm Emitter / 0.7 µm Collector HBTs:

0 100

2 10-15

4 10-15

6 10-15

8 10-15

1 10-14

0 50 100 150 200

Ic = 1 mAIc = 2 mAIc = 3 mAIc = 4 mAIc = 5 mA

Cc

(Far

ads)

Frequency (GHz)

Vce

=1.2 V

-1 10-3

0 100

1 10-3

2 10-3

3 10-3

4 10-3

5 10-3

0 50 100 150 200

Ic = 1 mAIc = 2 mAIc = 3 mAIc = 4 mAIc = 5 mA

Gc

(S)

Frequency (GHz)

Vce

=1.2 V

-1

-0.5

0

0.5

1

0 50 100 150 200

magnitudereal partimaginary part

Frequency (GHz)

Com

mon

-bas

e cu

rrent

gai

n, a

lpha

Vce

=1.2 V

Ic=5 mA

0

2

4

6

8

10

0

2

4

6

8

10

0 0.5 1 1.5 2 2.5

1/2πft, 0.3 µm x 18 µm emitter

1/2πft, 0.4 µm x 6 µm emitter

Ccb

, 0.3 µm x 18 µm emitter

Ccb

, 0.4 µm x 6 µm emitter

Collector-base capacitance C

cb , fF

Inverse of collector current, Ic (1/mA)

Ic= 5 mA

Ic= 6 mA

No evidence whatsoever of the postulated base pushout phenomenon of Jäckel et al(this theory also uses an erroneous hole mobility, error due to calculus derivatives chain rule error)

Miguel Urteaga


0.0

1.0

2.0

3.0

4.0

5.0

6.0

0 0.5 1 1.5 2 2.5 3

Vce(V)

Ic(m

A)

transferred-substrate DHBTs UCSBSangmin Lee

0.0

1.0

2.0

3.0

0 1 2 3 4 5 6 7 8 9

Vce(V)

Ic(m

A)

BVCEO = 8 V at JE =0.4 mA/µm2

0

5

10

15

20

25

30

35

1.E+09 1.E+10 1.E+11 1.E+12

Frequency (Hz)

h21,

U (d

B)

fmax = 425 GHz, ft = 139 GHz

0.4 µm x 8 µm emitter 1.2 µm x 8.75 µm collectorIc=4.5 mA, Vce=1.9 V

Vce(sat) ~1 V at 1.8 mA/µm2

much wider bandwidth devices coming soon (we hope…)


Wideband Mesa InP/InGaAs/InP DHBTsWalsin

ONRUCSB IQEMattias Dahlstrom / Amy Liu

We have obtained high ft and very high fmaxin mesa DHBTs with C-doped InGaAs bases

Devices have very narrow base mesas and extremely low base contact resistivity

Unlike transferred-substrate HBTs, whichhave very low Ccbx , these devices have significant extrinsic collector-base junction areas.

→ further effort needed in excess Ccb reduction for >100 GHz digital ICs

Results to be presented soon


Comparing High-fmax HBTsMiguel Urteaga

S11

S22

S22

S111-45, 75-110, 140-220 GHz

Transferred-Substrate HBT:

0.3 um x 18 um emitter,0.7 um x 19 um collector. 130 GHz ft, fmax very high

Transferred-Substrate HBT:

0.4 um x 6 um emitter,0.7 um x 6.4 um collector. 130 GHz ft, fmax very high

Good (not record) mesa HBT

0.5 um x 7 um emitter,2.7 um x 12 um collector. 200 GHz ft, 200 GHz fmax

Miguel Urteaga

Fast C-doped-base mesa HBT

0.5 um x 7 um emitter,~1.6 um x 12 um collector. >250 GHz ft, high fmax

Verylow Ccb

Low Rbb

High Ccb

Mattias DahlstromPK Sundararajan

Low Rbb

Low Ccb

High Rbb

High Ccb

1-45 GHz, 140-220 GHz


InP/InGaAs/InP Metamorphic DHBTon GaAs substrate UCSB

triple-mesa device(not transferred-substrate)

Growth:400 Å base, 2000 Å collectorGaAs substrateInP metamorphic buffer layer

(high thermal conductivity)Processing

conventional mesa HBTnarrow 2 um base mesa, 0.4 um emitter

Results207 GHz ft, 140 GHz fmax, >6 Volt BVCEO, β=76

Young-Min Kim

0

10

20

30

40

0.1 1 10 100 1000frequency (GHz)

h21

U

ft = 207 GHz

fmax

= 140 GHz

0

2

4

6

8

10

12

14

0

1 105

2 105

3 105

4 105

0 1 2 3 4 5 6

J (A/cm2 )I C

(mA)

VCE

(V)


IC results


75 GHz HBT master-slave latchconnected as Static frequency divider

technology:400 Å base, 2000 Å collector HBT0.7 um mask (0.6 um junction) x 12 um emitters1.5 um mask (1.4 um junction) x 14 um collectors1.8×105 A/cm2 operation, 180 GHz ft, 260 GHz fmax

simulations: 95 GHz clock rate in SPICE

test data to date:tested, works over full 26-40 and 50-75 GHz bands

UCSBThomas Mathew

Michelle LeeHwe-Jong Kim

-0.14

-0.12

-0.1

-0.08

-0.06

-0.04

-0.02

0

0 50 100 150 200

fin=75GHz, f

out=37.5GHz

Vout

(Vol

ts)

Time(ps)

-0.11

-0.1

-0.09

-0.08

-0.07

-0.06

-0.05

-0.04

-0.03

0 50 100 150 200

fin=69GHz, f

out=34.5GHz

Vout

(Vol

ts)

Time(PS)

modulation is synthesizer 6 GHz subharmonic

3.92 V, 224 mA, 0.88 W

~3.5 dBm input power


18 GHz Σ−∆ ADC

Designcomparator is 75 GHz flip flopDC bias provided through 1 KΩ resistors Integration obtained with 3 pF capacitors RTZ gated DAC

Integrated Circuit150 HBTs, 1.2 x 1.5 mm, 1.5 W

gm2

Bias

Bias

C

C

Rz

Rz

Integrator-1 Integrator-2Bias

Bias

C

C

Master-Slave

Flip-flop

Clock

Out

Idac

gm1

DelayedClock

Vin

CurrentSumming

node

RL

RLRL

RL

UCSBS Jaganathan


High Speed Amplifiers

18 dB, DC--50+ GHz

UCSBDino Mensa

PK Sundararajan

8.2 dB, DC-80 GHz

-15

-10

-5

0

5

10

0 10 20 30 40 50 60 70 80

Gai

ns, d

B

Frequency, GHz

S21

S11

S22

-20

-15

-10

-5

0

5

10

15

20

0 10 20 30 40 50

>397 GHzgain x bandwidth from 2 HBTs

S22

S11

S21


HBT distributed amplifier UCSBPK Sundararajan11 dB, DC-87 GHz

AFOSR

-20

-15

-10

-5

0

5

10

15

0 20 40 60 80

Gai

ns, d

B

Frequency, GHz

S21

S22

S11

TWA with internal ft-doubler cells


175 GHz Single-Stage AmplifierSubmicron HBT Program

UCSBMiguel Urteaga

-20

-15

-10

-5

0

5

10

140 150 160 170 180 190 200 210 220

S21S11S22

dB

Freq. (GHz)

0.2pF

50Ω 30Ω,1.2ps

50Ω

30Ω,0.2ps

80Ω,1.2ps

30Ω,0.6ps

80Ω,1.2ps

50Ω

IN

OUT

6.3 dB gain at 175 GHz


40 mW, W-band InP DHBT power amplifiers

Objectives: W band, P1dB>9 dBm, Psat>12 dBm

Approach: transferred-substrate InP DHBTs, microwave amplifier design

Simulations: S-parameter and harmonic simulation in ADS

Accomplishments:f0=85 GHz, BW3dB=28 GHz,

GT=8.5 dB, P1dB=14.5 dBm, Psat=16dBm

UCSBYun Wei

common base PA

-5

0

5

10

15

20

0

2

4

6

8

10

-15 -10 -5 0 5 10 15

Pout

, dBm G

T , dB

Pin, dBm

GT Pout

-30

-25

-20

-15

-10

-5

0

5

10

80 90 100 110

S11

, S21

, S22

frequency, GHz

S21

S22

S11

0.5mm x 0.4 mm, AE=128 µµµµm2

ARO MURI

inp-based hbts: devices and ghz mixed-signal ics...rodwell, short course, 2002 ieee/osa conference...

Documents