ece-305: spring 2018 bjt and hbt nonidealities
TRANSCRIPT
ECE-305: Spring 2018
BJT and HBT Nonidealities
Professor Peter BermelElectrical and Computer Engineering
Purdue University, West Lafayette, IN [email protected]
Pierret, Semiconductor Device Fundamentals (SDF)Chapters 10 and 11 (pp. 371-385, 389-403)
4/23/2018 Bermel ECE 305 S18
MOSFETs vs. BJTs
IC
IBVCE
IE
VBEsiliconS
D
G
VBE
VDS
IG 0
ID
MOSFET characteristics:• simple to make• no gate current• moderate gm• low capacitance
BJT characteristics:• more complex to make• base current• large gm• high capacitance
C
E
B
Bermel ECE 305 S184/23/2018 2
active region example 1 (NPN)
Bermel ECE 305 S18
AE 10 m ´ m
IC 50 A
NDE 1.5 ´1019 cm3
NAB 1.0 ´1017 cm3
NDC 2.0 ´1016 cm3
WB 0.25 m
WB 0.50 m
WC 1.50 m
t pE 0.1 ns
t nB 75 ns
t pC 150 ns
4/23/2018 8
Ebers-Moll parameters summary
IC VBE ,VBC aF IF0 eqVBE kBT 1 IR0 eqVBC kBT 1
IE VBE ,VBC IF0 eqVBE kBT 1 a RIR0 eqVBC kBT 1
a R aFIF0IR0
aF 0.9987
IR0 1.91´1016A
IF0 1.33´1016A
a R 0.70 bR a R1a R
2.3
Bermel ECE 305 S184/23/2018 9
What is IC?
VCE 0
IE
IC
IB
IB 50 nA
IC aF IF0 eqVBE kBT 1 IR0 eqVBC kBT 1
IE IF0 eqVBE kBT 1 aRIR0 eqVBC kBT 1
VCE VBE VBC 0
VBC
VBE
VBC VBE
IC aF IF0 IR0 eqVBE kBT 1
IE IF0 aRIR0 eqVBE kBT 1 4/23/2018 Bermel ECE 305 S18 10
What is IC?
VCE 0
IE
IC
IB
IB 50 nA
IB IE IC
IB 1aF IF0 1aF IR0éë ùû eqVBE kBT 1
VBE
VBE
IC aF IF0 IR0 eqVBE kBT 1
IE IF0 aRIR0 eqVBE kBT 1
eqVBE kBT 1 IB1aF IF0 1aF IR0éë ùû
Bermel ECE 305 S184/23/2018 11
What is IC?
IC aF IF0 IR0
1aF IF0 1aF IR0
ìíîï
üýþïIB
IB 50 nA
aF 0.9987
IR0 1.91´1016A
IF0 1.33´1016A
a R 0.70
IC 1.01´ IB
IB 50 nA
Bermel ECE 305 S184/23/2018 12
result
Bermel ECE 305 S18
VCE
E: emitter
C
B: base
IC
NPN BJT
IC
IE
IB 50 nA bF 768
IB 50 nA
IC 0.38 mA
IC IB
IC 0 atVCE VOS
Why?4/23/2018 13
Gummel Plot and Output Characteristics
2 2, , //( 1) ( 1)BCBEi B i B q kTq kTn n
B B B B
C VVn nqD qDe e
A W N W N
I ;
2, /( 1)BEp i E V kT
E E
B qqD n
eA W N
I
CDC
BI
Ib
DCbCommon emitter Current Gain
VBE
4/23/2018
=
Bermel ECE 305 S18 14
Bermel ECE 305 S18
NE
NB
NC
Emitter doping: As high as possible withoutband gap narrowing
Base doping: As low as possible, withoutcurrent crowding, Early effect
Collector doping: Lower than base dopingwithout Kirk Effect
Base Width: As thin as possible withoutpunch through (~1 mm in ‘50s, 200 Å now)
4/23/2018
2,
2,
i Bn E EDC
B p i E B
nD W N
W D n Nb
How to make a Good Silicon Transistor
~1, same material
15
doping for maximum gain
n+emitter
pbase
ncollector
n+
FB RB
IEn
IEp
Bermel ECE 305 S18
log10 ND NA { }
NDE NAB
NDC
ND
4/23/2018 16
doping for maximum gain
n-collector
n+
n+
IBIB
high base doping lowers the base resistance
high sub-collector doping lowers collector series resistance
light collector doping increases the breakdown voltage
4/23/2018 Bermel ECE 305 S18 17
emitter-base doping
n+emitter
pbase
ncollector
n+
FB RB
IEn
IEp
IEn µni2
NAB
IEp µni2
NDE
High emitter doping increases the emitter injection efficiency.
g F 1
1DpE
DnB
WB
WE
NABNDE
NDE >> NAB
Bermel ECE 305 S184/23/2018 18
High emitter doping benefits
n+emitter
pbase
ncollector
n+
FB RB
IEn
IEp
IEn µni2
NAB
IEp µni2
NDE
High emitter doping increases the emitter injection efficiency.
g F 1
1DpE
DnB
WB
WE
NABNDE
NDE >> NAB
Bermel ECE 305 S184/23/2018 19
high emitter doping bandgap narrowing
Bermel ECE 305 S18
2,
2,
i Bn E
B p i E B
EnD W
W D n
N
Nb
Band-gap narrowing reduces gain significantly …
(Esaki-like) tunneling cause loss of base control …
,
,
/
/
B
g E
gE kT
n C VE
B p BC V
EE
kT
D N NW e
W D N e NN
N
/gE kT E
B
Ne
N
D
4/23/2018 20
Low base doping emitter current crowding
n-collector
n+
n+
IBIB
Bermel ECE 305 S184/23/2018
AE
AC
AC >> AE
21
Low Base Width punch-through
Bermel ECE 305 S18
NE
NB
NC
0
,
2 s Ep BE bi BE
B E B
k Nx V V
q N N N
0,
2 s Cp BC bi BC
B C B
k Nx V V
q N N N
NN+
4/23/2018 22
“inverted” base doping
Wide gapemitter
pbase
ncollector
n+
FB RB
IEn
IEp
Bermel ECE 305 S18
log10 ND NA { }NDE
NAB
NDC
ND
4/23/2018 23
High collector doping Kirk effect
n+emitter
pbase
ncoll
n+
FB RB
IEn
IEp
r qND
ICn qAEDnWB
Dn 0
IC qAWnusat n ND
“base push out”/ Kirk effect impliescannot reduce base doping arbitrarily
Bermel ECE 305 S184/23/2018 24
Low base doping early effect
Bermel ECE 305 S18
2,
2. . ,
i Bn E EDC
p B p C i EB p B
D W n N
W x x D n Nb
2 2, ,( / ) ( / )
, ( 1) 1' '
( )BE BCi B i BqV kT qV kTn nn
B
C
B B B
qD n qD nI e e
NW W N
C C C
BC BC A A
dI I I
dV V V V
VBC
VA
IC
Ideally
In practice
Ref. Fig 11.5 c,d B B
A
CB
qN WV
C
4/23/2018
Gain depends on collector voltage (bad) …
25
how to make better transistors
Bermel ECE 305 S18
2,
2,
i Bn E E
B p i E B
nD W N
W D n Nb
Classical Shockley Transistor
Heterojunction Bipolar Transistor
Graded Base transport
Polysilicon Emitter
4/23/2018 26
Bermel ECE 305 S18
heterojunction bipolar transistors (HBTs)
A heterojunction bipolar transistor
Shockley realized that HBT is possible, but Kroemer really provided the foundation of the field and worked out the details.
Kroemer
4/23/201827
SiGe HBTs
Si SiGe Si n+
FB RB
Bermel ECE 305 S18
EG Si 1.12 eV
EG Ge 0.66 eV
EG Si 1.12 eV
4/23/2018 28
Bermel ECE 305 S18
SiGe HBT applications
1) optical fiber communications-40Gb/s…….160Gb/s
2) Wideband, high-resolution DA/AD convertersand digital frequency synthesizers
-military radar and communications
3) Monolithic, millimeter-wave IC’s (MMIC’s)
-front ends for receivers and transmitters
future need for transistors with 1 THz power-gain cutoff freq.
4/23/2018 29
SiGe HBTs
Bermel ECE 305 S184/23/2018
https://www.macrumors.com/2017/11/02/fight-for-space-iphone-x/
30
HBT (wide gap emitter)
n+emitter
pbase
ncollector
n+
FB RB
EGE EGB EGC
IEn
IEp
IEn µniB2
NAB
IEp µniE2
NDE
g F 1
1DpE
DnB
WB
WE
NABNDE
niE2
niB2
ni2 µ eEG kBT
Bermel ECE 305 S184/23/2018
Choose an emitter with a higher bandgap than the base
IEn µniB2
NAB
31
SiGe HBTs
Si SiGe Si n+
FB RB
Bermel ECE 305 S18
EG Si 1.12 eV
EG Ge 0.66 eV
EG Si 1.12 eV
4/23/2018
g F 1
1DpE
DnB
WB
WE
NABNDE
niE2
niB2
ni2 µ eEG kBT
emitter collectorbase
Choose an emitter with a higher bandgap than the base
32
poly-silicon HBT emitter
Bermel ECE 305 S18
N+
P N
SiGe intrinsic base Dielectric trench
N+P+
N
P-
N-N+
CollectorEmitterBase
N+
Poly-silicon
emitter
4/23/2018 33
Bermel ECE 305 S18
mesa HBTs
nemitter
ncollector n+
EG1>EG2
p+base
EG2 EG3>EG2
p+ basen-collector
n+
semi-insulating substrate
nmesa HBT
4/23/2018 34
Bermel ECE 305 S18
collector breakdown
VBE
Common Emitter(IE variable, IB fixed)
Common emitter breakdown voltage is smaller than common base breakdown voltage.
Common Base(IE fixed, IB variable)
4/23/201835
Bermel ECE 305 S18
conclusions
• There are many non-ideal effects of bipolar
junction transistors that give rise to design
constraints
• Classical (MSM or Shockley) BJTs are
almost nonexistent today.
• Modern BJTs use poly-Silicon emitter, graded
based, hetero E-B junction, etc., to enhance
performance.
4/23/2018 36
Summary of ECE 305
Semiconductor Devices Fundamentals
Drift-diffusion model for semiconductors
D AD q p n N N
1
JP P P
pr g
t q
P P Pqp qD pJ E
1N N N
nr g
t q
J
N N Nqn qD nJ E
Band-diagram
Diffusion approximation,Minority carrier transport,
Ambipolar transport
4/23/2018 Bermel ECE 305 S18 38
Short-cut to Band-diagram
DN AN
… is equivalent to solving the Poisson equation
Vacuum level
EC
EV
EF
c2c1
Neutral Neutral
Space Charge
4/23/2018 Bermel ECE 305 S18
39
Minority Carrier Diffusion
( )
( )
( 0 )
( 0 )
b
b
n i
p i
F Ei
F E
i
n x n e
p x n e
( )2 2b b pn AqV
i i
F Fn e n enp
Fp
Fn
q(Vbi-VA)
-VA
2
(0 ) b Aqi
A
Vn eN
n
(0 ) Ap N
AN
0
2
(0 ) (0 ) (0 )
1b
D
G G
A
V V
qVi
A
n n n
ne
N
4/23/2018 Bermel ECE 305 S1840
Small Signal AC Response
RS
G
CD
CJ
x
Δρ
x
Δρ
VA<0
VA>0
np
x4/23/2018 Bermel ECE 305 S18
41
Concepts for Device Analysis
Equilibrium DC Small signal
Large Signal
Diode Band-diagram diffusion dn/dt~jwn Charge-control
Schottky Band-diagram TE Junction
capacitance
Majority transport
BJT/HBT Band-diagram diffusion/TE dn/dt~jwn Charge-control
MOSCAP
MOSFET
2D band-diagram
Drift/TE MOS capacitance
Charge-control
4/23/2018 Bermel ECE 305 S1842
Major Questions for Semiconductor Researchers
4/23/2018 Bermel ECE 305 S18
• Can we achieve computing power equivalent
to or greater than that of the human brain?
• Can we create and deploy sustainable energy
technologies to transition away from fossil
fuel usage?
• Can we integrate electronics with biology to
detect and fix debilitating diseases?
43
Grand Challenges in Electronics
1906-1950s 1947-1980s 1980-until now
Vacuum Tubes
Bipolar MOSFET
Spintronics
Bio Sensors
Displays ….
Now ??
?
1900 1920 1940 1960 1980 2000 2020
Te
mp
Tubes Bipolar MOS
4/23/2018 Bermel ECE 305 S1844
Emergence of Macroelectronics
Areasmall medium large
Perf
orm
ance
low
m
ed
ium
h
igh
mesoporous
NanoNetPoly-Si
Polymers
Flexible ElectronicsEnergy Biosensors
45
Thin Film Organic Transistors
www.faculty.iu-bremen.de/dknipp/group/research.htmM.G. Kanatzidis, Nature, 428, 2004.
pentacene
Samsung flexible phone
Can you draw the band-diagram?What type of transport theory would you use? Would you be able to use numerical simulators
from nanohub.org to explore the TFT?
Bermel ECE 305 S18 46
PV Simulations on nanoHUB.orga major resource for computational nanotechnology
enabled by the HUBzero platform for simulation, learning, and collaboration
• 25 Hubs operating or under construction• open source platform• 6,144 dedicated cores with over 17,000 on standby
4/23/2018 Bermel ECE 305 S18 47
Multijunction Photovoltaics
“ The NREL researchers improved the cell’s efficiency by enhancing the photon recycling in the lower, gallium-arsenide junction by using a gold back contact to reflect photons back into the cell, and by allowing a significant fraction of the luminescence from the upper, GaInP junction to couple into the GaAs junction. Both the open-circuit voltage and the short-circuit current were increased.”
—NREL News Release, June 24, 2013
• The new record obtain in June 2013, at 31.1%, is a significant jump from 29.5% prior record• Enhanced photon recycling is believed to cause this improvement.
4/23/2018 Bermel ECE 305 S18 48
Technology: Sequencing by Synthesis
Sanger method (1990s) Babbage computer (~1830s)
Intel Chip TodayIon-torrent system (2011)
4/23/2018 Bermel ECE 305 S18 49
Technology: Sequencing by Synthesis
Bermel ECE 305 S184/23/18 50
Superhydrophobic coatings
4/23/2018 Bermel ECE 305 S18
http://wyss.harvard.edu/viewpage/316
SLIPS Coatings ‘Waterproof’ circuit boards
http://www.cytonix.com/conformal-coating-s/1872.htm
51
Superhydrophobic coatings
4/23/2018 Bermel ECE 305 S18
https://www.youtube.com/watch?v=16RcYoXOvRM
Anti-soiling coatings for photovoltaic modules
52
Artificial Skin
4/23/2018
M.L. Hammock et al., "25th Anniversary Article: The Evolution of Electronic Skin (E‐Skin): A Brief History, Design Considerations, and Recent Progress.“ Advanced Materials 25, 5997-6038 (2013).
Bermel ECE 305 S18 53
Artificial Skin
4/23/2018 Bermel ECE 305 S18
Multiplex, flexible strain-gauge sensor based on the reversible interlocking of Pt-coated polymer nanofibres. Image: Nature Materials (2012) http://dx.doi.org/10.1038/nmat3380
54
Developing New Ideas
“New ideas pass through three periods: 1) It can't be done.2) It probably can be done, but it's not worth
doing.3) I knew it was a good idea all along!”
“I don't pretend we have all the answers. But the questions are certainly worth thinking about.”
-- Sir Arthur C. Clarke
4/23/2018 Bermel ECE 305 S1855