Plenary, 2012 European Microwave Integrated Circuits Conference, October 29th, Amsterdam
Transistor & IC design
for Sub-mm-Wave & THz ICs
Mark Rodwell
University of California, Santa Barbara
Coauthors:
J. Rode, H.W. Chiang, T. Reed, S. Daneshgar, V. Jain, E. Lobisser,
A. Baraskar, J. Law, A. Carter, S. Lee, D. Elias, B. J. Thibeault,
B. Mitchell, S. Stemmer, A. C. Gossard, UCSB
Munkyo Seo, Jonathan Hacker, Adam Young, Zach Griffith, Richard
Pierson, Miguel Urteaga, Teledyne Scientific Company
rodwell@ece.ucsb.edu 805-893-3244
DC to Daylight. Far-Infrared Electronics
How high in frequency can we push electronics ?
1982: ~13 GHz
10
9
10
mm-wave
30-300 GHz
10
10
11
~2030: 3THz
far-IR
(sub-mm)
0.3-3THz
12
10
Frequency (Hz)
mid-IR
3-30 THz
10
13
near-IR
30-450 THz
10
14
optical
450-900 THz
microwave
3-30 GHz
2012: 820 GHz
10
15
...and what we would be do with it ?
100+ Gb/s wireless networks
Video-resolution radar
→ fly & drive through fog & rain
near-Terabit
optical fiber links
100-1000 GHz
Systems
100-1000 GHz Wireless Has High Capacity
very large bandwidths available
short wavelengths→ many parallel channels
N  B 2 / R  1
B  ND
angular resolution 
wavelength
array widt h
R
# channels  (aperture area) 2 /(waveleng th  distance) 2
100-1000 GHz Wireless Needs Phased Arrays
isotropic antenna → weak signal →short range
 Preceived

 Ptransmitted
  2  R
   2  e
 R 
highly directional antenna → strong signal, but must be aimed
 Preceived

 Ptransmitted

 2  R
  Dt Dr  2  e
R 

no good for mobile
must be precisely aimed →too expensive for telecom operators
beam steering arrays → strong signal, steerable
Preceived
2 R
 N receive N transmit 2 e
Ptransmit
R
32-element array → 30 (45?) dB increased SNR
100-1000 GHz Wireless Needs Mesh Networks
Object having area
~R
will block beam.
...high-frequency signals
are easily blocked.
Blockage is avoided
using beamsteering
and mesh networks.
...this is easier at high frequencies.
100-1000 GHz Wireless Has Low Attenuation ?
Wiltse, 1997
IEEE Int. APS
Symposium,
July
2-5 dB/km
200-300 GHz
125-165 GHz
75-110 GHz
Low attenuation on a sunny day
100-1000 GHz Wireless Has High Attenuation
High Rain Attenuation
High Fog Attenuation
Rain Attenuation, dB/km
100
100 mm/Hr
30 dB/km
50 mm/Hr
10
very heavy fog
4
0.1
9
10
10
~(25 dB/km)x(frequency/500 GHz)
50 GHz
five-9's rain @ 50-1000 GHz:
1 → 30 dB/km
11
10
10
Frequency, Hz
10
12
50-500 GHz links must tolerate ~30 dB/km attenuation
Olsen, Rogers, Hodge, IEEE Trans Antennas & Propagation Mar 1978
Liebe, Manabe, Hufford, IEEE Trans Antennas and Propagation, Dec. 1989
140 GHz, 10 Gb/s Adaptive Picocell Backhaul
140 GHz, 10 Gb/s Adaptive Picocell Backhaul
600 meters range in five-9's rain
Realistic packaging loss, operating & design margins
PAs: 30 dBm Psat (per element)→ GaN or InP
LNAs: 4 dB noise figure → InP HEMT
340 GHz, 160 Gb/s MIMO Backhaul Link
→ 1o beamwidth; 8o beamsteering
340 GHz, 160 Gb/s MIMO Backhaul Link
1o beamwidth; 8o beamsteering
600 meters range in five-9's rain
Realistic packaging loss, operating & design margins
PAs: 21 dBm Psat (per element)→ InP
LNAs: 7 dB noise figure → InP HEMT
100-1000 GHz Wireless Transceiver Architecture
III-V LNAs, III-V PAs → power, efficiency, noise
Si CMOS beamformer→ integration scale
...similar to today's cell phones.
Why
THz Transistors ?
THz Transistors: Not Just For THz Circuits
500 GHz digital logic
→ fiber optics
precision analog design
Higher-Resolution
at microwave frequencies
Microwave ADCs,
→ high-performance receivers DACs, DDSs
THz amplifiers→ THz radios
→ imaging, communications
THz InP HBTs
THz & nm Transistors: what it's all about
Metal-semiconductor interfaces (Ohmic contacts): very low resistivity
Dielectric-semiconductor interfaces (Gate dielectrics---FETs only): thin !
Ultra-low-resistivity (~0.25 W-mm2 ), ultra shallow (1 nm), ultra-robust (0.2 A/mm2 ) contacts
Mo
Ru
InGaAs
InGaAs
Heat
L
TIC 
Available quantum states to carry current
PIC
K th L
P
L
Ttransistor ~
ln  
K th L  W 
→ capacitance,
transconductance
contact resistance
Bipolar Transistor Design
We
Tb
 b  T 2 Dn
2
b
Wbc
Tc
 c  Tc 2v sat
Ccb  Ac /Tc
I c ,max  vsat Ae (Vce,operating  Vce,punch-through) / T
2
c
P
T 
LE

 Le 
1  ln  
 We 

Rex  contact /Ae
 We Wbc  contact
 
Rbb  sheet 

 12 Le 6 Le  Acontacts
emitter
length LE 
Bipolar Transistor Design: Scaling
We
Tb
 b  T 2 Dn
2
b
Wbc
Tc
 c  Tc 2v sat
Ccb  Ac /Tc
I c ,max  vsat Ae (Vce,operating  Vce,punch-through) / T
2
c
P
T 
LE

 Le 
1  ln  
 We 

Rex  contact /Ae
 We Wbc  contact
 
Rbb  sheet 

 12 Le 6 Le  Acontacts
emitter
length LE 
Scaling Laws, Scaling Roadmap
Tb
scaling laws: to double bandwidth
HBT parameter
emitter & collector junction widths
current density (mA/mm2)
current density (mA/mm)
collector depletion thickness
base thickness
emitter & base contact resistivities
We
change
decrease 4:1
increase 4:1
constant
decrease 2:1
decrease 1.4:1
decrease 4:1
emitter
Wbc
Tc
length LE 
150 nm device
HBT Fabrication Process Must Change... Greatly
32 nm width base & emitter contacts...self-aligned
32 nm width emitter semiconductor junctions
Contacts:
1 W-mm2 resistivities
70 mA/mm2 current density
~1 nm penetration depths
→ refractory contacts
nm III-V FET, Si FET processes have similar requirements
Needed: Greatly Improved Ohmic Contacts
textbook
with surface oxide with metal penetration
Interface barrier → resistance
Further intermixing during high-current operation → degradation
Needed: Greatly Improved Ohmic Contacts
Pt/Ti/Pd/Au
~5 nm
Pt contact
penetration
(into 25 nm base)
Ultra Low-Resistivity Refractory In-Situ Contacts
-6
Barasakar et al
IEEE IPRM 2012
P-InGaAs
2
Contact Resistivity (Wcm )
10
N-InGaAs
N-InAs
Mo
Mo
-7
10
Ir
W
Mo
-8
32 nm node
requirements
10
-9
10
18
19
20
21
10
10
10
10
-3
concentration (cm )
17
18
19
20
10
10
10
10
-3
concentration (cm )
18
19
20
21
10
10
10
10
-3
concentration (cm )
In-situ: avoids surface contaminants
Refractory: robust under high-current operation
Low penetration depth, ~ 1 nm
Contact performance sufficient for 32 nm /2.8 THz node.
Refractory
EmitterImproved
ContactsOhmic Contacts
Needed: Greatly
negligible
penetration
HBT Fabrication Process Must Change... Greatly
tall, narrow contacts: liftoff fails !
control undercut
→ thinner emitter
thinner emitter
→ thinner base metal
Undercutting of emitter ends
{101}A planes: fast
{111}A planes: slow
thinner base metal
→ excess base metal resistance
Sub-200-nm Emitter Anatomy
Refractory contact: high-J operation
Liftoff Sputter+dry etch→ sub-200nm contacts
High-stress emitters fall off
during subsequent lift-offs
TiW
W
TiW
SiNx
100 nm
W
Single sputtered metal has
non-vertical etch profile
Mo
slide: E. Lobisser. HBT: V. Jain. Process: Jain & Lobisser
Sub-200-nm Emitter Anatomy
Hybrid sputtered metal stack for
low-stress, vertical profile
W/TiW interfacial discontinuity
enables base contact lift-off
Very thin emitter epitaxial layer
for minimal undercut
TiW
Semiconductor wet etch
undercuts emitter contact
SiNx
100 nm
W
Interfacial Mo blanket-evaporated for low ρc
SiNx sidewalls protect emitter contact,
prevent emitter-base shorts during liftoff
Mo
slide: E. Lobisser. HBT: V. Jain. Process: Jain & Lobisser
Sub-200-nm Emitter Anatomy
emitter-base gap: only ~10 nm
→ greatly reduces link component of Rbb
bb.
slide: E. Lobisser. HBT: V. Jain. Process: Jain & Lobisser
TiW
SiNx sidewall
W
Wbc = 150 nm
Web = 155 nm
Pt/Ti/Pd/Au
RF Data: 25 nm thick base, 75 nm Thick Collector
25
U
Gains (dB)
20
H
15
10
5
21
f = 530 GHz

f
max
0
10
10
Required dimensions obtained
but poor base contacts on this run
= 750 GHz
11
10
Frequency (Hz)
12
10
E. Lobisser, ISCS 2012, August, Santa Barbara
DC, RF Data: 100 nm Thick Collector
f
32
f = 480 GHz
U
H
P = 20 mW/mm
2
P = 30 mW/mm
30
2
A = 0.22 x 2.7 mm
2

25
21
je
20
15
I
b,step
= 200 mA
10
20
5
0
16
A = 0.22 x 2.7 mm
2
BV
0
1
2
V
je
12
0 9
10
e
10
10
-3
10
-5
b
P = 33.5 mW/mm2
Vcb = 0.7 V
-1
11
10
10
Frequency (Hz)
ce
Solid line: V
Dashed: V
cb
cb
4
5
= 0.7V
20
= 0V
15
n = 1.19
c
10
n = 1.87
10
-7
10
-9
12
10
3
(V)
b
I
b

5
I
c
0
0.2 0.4 0.6 0.8
Vbe (V)
0
1

4
J = 20.4 mA/mm2
10
c
8
Ic = 12.1 mA
I , I (A)
Gain (dB)
e
24
= 1.0 THz
J (mA/mm )
28
max
Jain et al
IEEE DRC 2011
2
THz InP HBTs From Teledyne
Chart 33
Urteaga et al, DRC 2011, June
Towards & Beyond the 32 nm /2.8 THz Node
Base contact process:
Present contacts too resistive (4Wmm2)
Present contacts sink too deep (5 nm) for target 15 nm base
→ refractory base contacts
Emitter Degeneracy:
Target current density is almost 0.1 Amp/mm2 (!)
Injected electron density becomes degenerate.
transconductance is reduced.
→ Increased electron mass in emitter
Base Ohmic Contact Penetration
~5 nm
Pt contact
penetration
(into 25 nm base)
Refractory Base Process (1)
Blanket liftoff; refractory base metal
low contact resistivity
low penetration depth
Patterned liftoff; Thick Ti/Au
low bulk access resistivity
base surface not exposed to photoresist chemistry: no contamination
low contact resistivity, shallow contacts
low penetration depth allows thin base, pulsed-doped base contacts
Refractory Base Process (2)
-5
10
P-InGaAs
Contact Resistivity, Wcm
2
-6
10
-7
10
32 nm node
requirement
-8
10
B=0.8 eV
-9
10
0.6 eV
0.4 eV
0.2 eV
step-barrier
Landauer
-10
10
18
2.5 1020
doping, 1/cm3
19
20
21
10
10
10
10
-3
Hole Concentration, cm
Increased surface doping:
reduced contact resistivity,
but increased Auger recombination.
20
2 10
2 nm doping pulse
1.5 1020
1 1020
→ Surface doping spike at most 2-5 thick.
19
5 10
0 100
0
5
10
15
depth, nm
20
25
Refractory contacts do not penetrate;
compatible with pulse doping.
Refractory Base Ohmic Contacts
Ru / Ti / Au
<2 nm
Ru contact
penetration
(surface removal
during cleaning)
Degenerate Injection→ Reduced Transconductance
10
2
10
1
Boltzmann
(-V )>>kT/q
2
J(mA/mm )
be
10
10
-1
10
-2
10
-3
Current varies exponentially with Vbe
J  J s exp( qVbe / kT ).
Transconductance is high
g m / AE  J
0
-0.3
J  J s exp( qVbe / kT ).
-0.2
-0.1
0
V -
be
0.1
0.2
Degenerate Injection→ Reduced Transconductance
10
2
10
1
Fermi-Dirac
Boltzmann
(-V )>>kT/q
2
J(mA/mm )
be
10
10
-1
10
-2
10
-3
Current varies exponentially with Vbe
J  J s exp( qVbe / kT ).
Transconductance is reduced
0
-0.3
-0.2
-0.1
0
V -
be
0.1
0.2
Degenerate Injection→ Reduced Transconductance
10
2
10
1
Fermi-Dirac
Boltzmann
(-V )>>kT/q
2
J(mA/mm )
be
Highly degenerate limit:
10
0
Highly degenerate
(V ->>kT/q
be
10
-1
10
-2
10
-3
current varies as the square of bias
J  m (Vbe   )
*
E
2
-0.3
q 3m *
J  2 3 (Vbe   ) 2
8 
-0.2
-0.1
0
V -
be
0.1
0.2
Degenerate Injection→ Reduced Transconductance
10
2
10
1
Fermi-Dirac
Boltzmann
(-V )<<kT/q
2
J(mA/mm )
be
Highly degenerate limit:
10
0
Highly degenerate
(V ->>kT/q
be
10
-1
10
-2
10
-3
current varies as the square of bias
J  m (Vbe   )
*
E
2
Transconductance varies as J1/2
g m / AE  mE* J
...and as (m*)1/2
-0.3
q 3m *
J  2 3 (Vbe   ) 2
8 
-0.2
-0.1
0
V -
0.1
be
At & beyond 32 nm, we must increase the emitter effective mass.
0.2
Degenerate Injection→Solutions
At & beyond 32 nm, we must increase the emitter (transverse) effective mass.
Other emitter semiconductors:
no obvious good choices (band offsets, etc.).
Emitter-base superlattice:
increases transverse mass in junction
evidence that InAlAs/InGaAs grades are beneficial
Extreme solution (10 years from now):
partition the emitter into small sub-junctions, ~ 5 nm x 5 nm.
parasitic resistivity is reduced progressively as sub-junction areas are reduced.
IC Results
InP HBT Integrated Circuits: 600 GHz & Beyond
614 GHz
fundamental
VCO
M. Seo,
Vtune
VBB
VEE
Vout
340 GHz
dynamic
frequency
divider
M. Seo, UCSB/TSC
IMS 2010
565 GHz, 34 dB, 0.4 mW output power
amplifier
300 GHz
fundamental
PLL
J. Hacker, TSC
M. Seo, TSC
IMS 2011
204 GHz static
frequency divider
(ECL master-slave
latch)
220 GHz
90 mW
power
amplifier
Z. Griffith, TSC
CSIC 2010
T. Reed, UCSB
Integrated
300/350GHz
Receivers:
LNA/Mixer/VCO
M. Seo
600 GHz
Integrated
Transmitter
PLL + Mixer
M. Seo
Teledyne: 560 GHz Common-Base Amplifier IC
Chart 46
S-parameters
Output Power
• 10-Stage Common-base using
inverted CPW-G architecture
• 34 dB at 565 GHz
• Psat -3.9 dBm at 560 GHz
1200x230 mm2
J Hacker et al, Teledyne Scientific
90 mW, 220 GHz Power Amplifier
Reed (UCSB) and Griffith (Teledyne): CSIC 2012
Teledyne 250 nm InP HBT
20
3dB bandwidth = 240GHz
Amplifier gains (dB)
active area, 1.02 x 0.85 mm
die: 2.42 x 1.22 mm
S
10
21,mid-band
= 15.4dB
0
-10
S
22
S
-20
11
P
DC
210
220
= 4.46W
230
240
250
frequency (GHz)
8-cell, 2-stage PA
260
100
, mW
80
90mW
60
84mW
40
→ InP HBT is a competitive
mm-wave / sub-mm-wave
power technology.
88mW P out
80mW
72mW
P
out
RF output power densities
up to 0.5 W/mm @ 220 GHz.
20
62mW
42mW
220GHz operation
0
0
2
8-cell, 2-stage PA
4
6
8 10
P , mW
in
12
P
DC
14
= 4.46W
40
Operating Frequency = 220 GHz
35 Pdc = 12 W
30
25
20
15
10
5
0
-30 -25 -20 -15 -10 -5
400
350
300
250
200
150
100
50
0
0
5
Pout (mW)
Gain (dB), Pout (dBm)
220 GHz 330mW Power Amplifier Design
10
2.3 mm x 2.5 mm
T. Reed, UCSB
Z. Griffith, Teledyne
Teledyne 250 nm InP HBT
S-parameters (dB)
Pin (dBm)
30
30
20
20
10
10
0
0
-10
-10
-20
-20
50 100 150 200 250 300 350 400
Frequency (GHz)
0
50-G/s Track/Hold Amplifier; 250 nm InP HBT
S. Daneshgar, this conference
20.5 GHz input, 10 GHz clock
Linearity test
(third-order intercept)
Where Next ? → 2 THz Transistors, 1 THz Radios.
transmitter
interconnects
receiver
circuits
R1
Lout2
VOUT
L1
R3
Lout1
Q5
Q3
Q4
LC3
Q6
LB
Q1
LE1
CVar
Q8
L3
L4
Q10
R4
LC1
Q7
DIVOUT
R2
LC2
Q9
L2
Q2
Q3,4
Q5,6
R6
R7
Q11
CVar
L5
RFIN
Q1
R3
LE2
C1
C2
R4
R2
C1
R1
C3
C2
VEE VTUNE
VBB
C4
Q2
RE
R5
VEF
L6
VEE
C3
THz and Far-Infrared Electronics
IR today→ lasers & bolometers → generate & detect
Far-infrared ICs: classic device physics, classic circuit design
It's all about classic scaling: ...wire resistance,...
contact and gate dielectrics...
...heat,...
...& charge density.
band structure and
density of quantum states
(new!).
Even 1-3 THz ICs
will be feasible
(backup slides follow)
Weakly Degenerate→ Effective Added Resistance
10
2
10
1
Fermi-Dirac
Boltzmann
(-V )<<kT/q
2
J(mA/mm )
be
Vbe  (kT / q ) ln( I / I s )  I  Req
Req  1 / m
*
E
10
0
10
-1
10
-2
10
-3
Fit:
equivalent
series
resistance
-0.3
-0.2
-0.1
0
V -
0.1
be
At & beyond 32 nm, we must increase the emitter effective mass.
0.2
HBT Scaling Roadmap
emitter 128
4
64
2
32 nm width
1 Wmm2 access 
base
60
2.5
30 nm contact width,
1.25 Wmm2 contact 
53
36
2.75
37.5 nm thick,
72 mA/mm2 current density
2-2.5 V, breakdown
1000
2000
1000
480
1400 GHz
2800 GHz
1400 GHz
660 GHz
120
5
collector 75
18
3.3
f
fmax
RF-ICs
digital divider
730
1300
660
330
140 nm Device: RF Results
140 nm emitter junction
120 nm wide base contacts
75 nm thick collector
25 nm thick base
fmax impaired (780 GHz) :
excessive contact
penetration into base
433 GHz f
780 GHz fmax
Chiang & Rode
unpublished
DC to Daylight. Far-Infrared Electronics
How high in frequency can we push electronics ?
10
9
10
mm-wave
30-300 GHz
10
10
11
far-IR
(sub-mm)
0.3-3THz
mid-IR
3-30 THz
12
10
Frequency (Hz)
10
13
near-IR
30-450 THz
10
optical
450-900 THz
microwave
3-30 GHz
14
...and what would be do with it ?
0.3-3 THz radio: vast capacity
bandwidth, # channels
0.1-0.4 THz imaging systems
0.1-1 Tb/s
optical fiber links
10
15