sd_bjt_1

Semiconductor Devices
Bipolar Junction Transistors: Part 1
M. B. Patil
mbpatil@ee.iitb.ac.in
www.ee.iitb.ac.in/~sequel
Department of Electrical Engineering

Indian Institute of Technology Bombay
M. B. Patil, IIT Bombay

Bipolar junction transistors
E p n p C E n p n C
(Emitter) (Collector) (Emitter) (Collector)
B (Base) B (Base)

E p n p C E n p n C
B (Base) B (Base)
* A bipolar junction transistor (BJT) consists of two pn junction diodes connected back to back.

E p n p C E n p n C
B (Base) B (Base)
* A BJT is a three-terminal device; the terminals are called emitter, base, and collector.

E p n p C E n p n C
B (Base) B (Base)
* There are two types of BJTs: pnp and npn.

E p n p C E n p n C
B (Base) B (Base)
* The actual device construction is different than the above schematic diagram (to be discussed).

E p n p C E n p n C
B (Base) B (Base)
* The actual device construction is different than the above schematic diagram (to be discussed).
* For the device to work as a transistor (rather than two independent diodes), the two junctions must be
“close.”

p(x)
Basic operation
V1 + ∆V1
V1
pn0
0
W
E C
p n p
B
x=0 x=W
V1 = 0.6 V V2 = 5 V
D1
I2 = αI1
I1
V1 V2
p(x)
Basic operation
V1 + ∆V1
V1
pn0
0
W
E C
p n p
B
x=0 x=W
V1 = 0.6 V V2 = 5 V
D1
I2 = αI1
I1
V1 V2
dp
* If V1 is varied, p(x) varies → Ip (W ) ∝ (W ) varies, i.e., by changing V1 , I2 can be controlled. This is the basic transistor action.
dx
p(x) p(x)
Basic operation
V1 + ∆V1 V1 + ∆V1
V1 V1
pn0 pn0
0 0
W W
E C E C
p n p p n p
B B
x=0 x=W x=0 x=W
V1 = 0.6 V V2 = 5 V V1 = 0.6 V V2 = 5 V
D1
I2 = αI1 D1 D2
I1
V1 V2 V1 V2
dp
dx

p(x) p(x)
Basic operation
V1 + ∆V1 V1 + ∆V1
V1 V1
pn0 pn0
0 0
W W
E C E C
p n p p n p
B B
x=0 x=W x=0 x=W
V1 = 0.6 V V2 = 5 V V1 = 0.6 V V2 = 5 V
D1
I2 = αI1 D1 D2
I1
V1 V2 V1 V2
dp
dx
* If the two junctions are not sufficiently close, the device behaves like two independent diodes connected back-to-back, and there is no
transistor action.
W W
pnp and npn transistors
E p n p C E n p n C
B B
VEB VBC VBE VCB
Ec
EFn
EFp qVBE EFp
Ec qVBC
Ev qVCB
qVEB EFn
EFn
EFp
Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
IE IC = αIE IE IC = αIE
E C E C
(emitter) (collector) (emitter) (collector)
IB = (1 − α) IE
B (base) IB = (1 − α) IE B (base)

W W
E p n p C E n p n C
* In the “active” or “linear” mode, the B B

B-E junction is under forward bias, the VEB VBC VBE VCB
B-C junction is under reverse bias.
Ec
EFn
EFp qVBE EFp
Ec qVBC
Ev qVCB
qVEB EFn
EFn
EFp
Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

* The B-E voltage (magnitude) is Ec
EFn
restricted to about 0.8 V in a EFp qVBE EFp
low-power silicon BJT, as in a
Ec qVBC
forward-biased diode. Ev qVCB
qVEB EFn
EFn
EFp
Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

EFn
Ec qVBC
* The B-C voltage (magnitude) can be qVEB EFn
much larger than the B-E voltage and EFn
EFp
is limited by breakdown phenomena. Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

EFn
Ec qVBC
EFp
IE IC IE IC
* pnp transistor: Holes are injected
from the emitter. Most of them reach E C E C
the B-C depletion layer, get swept VEB VBC VBE VCB
away by the field there, and get
B B
collected by the collector. IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

EFn
Ec qVBC
EFp
IE IC IE IC
* pnp transistor: Holes are injected
the B-C depletion layer, get swept VEB VBC VBE VCB
away by the field there, and get
B B
collected by the collector. IB IB
* npn transistor: Electrons are injected
the B-C depletion layer, get swept
IB = (1 − α) IE
away by the field there, and get B (base) IB = (1 − α) IE B (base)
collected by the collector.

W W
E p n p C E n p n C
B B
VEB VBC VBE VCB
Ec
EFn
EFp qVBE EFp
Ec qVBC
Ev qVCB
qVEB EFn
EFn
EFp
Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C
* The emitter arrow in the BJT symbol B B

indicates the direction of the emitter VEB VBC VBE VCB
current when the BJT is operating in
the active mode (for both pnp and Ec
npn BJTs). EFn
EFp qVBE EFp
Ec qVBC
Ev qVCB
qVEB EFn
EFn
EFp
Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

npn BJTs). EFn
EFp qVBE EFp
* The collector current IC is a fraction
Ec qVBC
of the emitter current: IC = αIE . Ev qVCB
For a good transistor, α ≈ 1. EFn
qVEB
EFn
EFp
Ev
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

npn BJTs). EFn
EFp qVBE EFp
Ec qVBC
qVEB
* The three currents satisfy KCL, i.e., EFn
EFp
IE = IC + IB . Substituting for IC , we Ev
get IB = (1 − α) IE . IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE

W W
E p n p C E n p n C

npn BJTs). EFn
EFp qVBE EFp
Ec qVBC
qVEB
* The three currents satisfy KCL, i.e., EFn
EFp
IE = IC + IB . Substituting for IC , we Ev
get IB = (1 − α) IE . IE IC IE IC
* Note that the diagrams are not drawn E C E C
to scale. In reality, for a silicon BJT, VEB VBC VBE VCB
the depletion regions would be
B B
narrower. IB IB
E C E C
IB = (1 − α) IE

BJT in active mode
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE IB = (1 − α) IE
B (base) B (base)

BJT in active mode
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE IB = (1 − α) IE
B (base) B (base)
* The “common-emitter current gain” β – a figure of merit of a BJT – is defined as β = IC /IB .

BJT in active mode
IE IC IE IC
E C E C
VEB VBC VBE VCB
B B
IB IB
E C E C
IB = (1 − α) IE IB = (1 − α) IE
B (base) B (base)

IC αIE α
* β= = = .
IB (1 − α)IE 1−α

BJT in active mode
IE IC IE IC
E C E C
VEB VBC VBE VCB
α 1−α β = α/(1 − α)
B B
IB IB
0.9 0.1 9
IE IC = αIE IE IC = αIE 0.95 0.05 19

E C E C 0.99 0.01 99
0.995 0.005 199
IB = (1 − α) IE IB = (1 − α) IE
B (base) B (base)

IC αIE α
* β= = = .
IB (1 − α)IE 1−α

BJT in active mode
IE IC IE IC
E C E C
VEB VBC VBE VCB
α 1−α β = α/(1 − α)
B B
IB IB
0.9 0.1 9
IE IC = αIE IE IC = αIE 0.95 0.05 19

E C E C 0.99 0.01 99
0.995 0.005 199
IB = (1 − α) IE IB = (1 − α) IE
B (base) B (base)

IC αIE α
* β= = = .
IB (1 − α)IE 1−α
* For a typical discrete low-power transistor such as BC107A, β is in the range of 100 to 200.

BJT modes of operation
E p n p C E n p n C
B (Base) B (Base)
Mode B-E junction B-C junction
Active (linear) forward reverse

Cutoff reverse reverse
Saturation forward forward
Inverse active reverse forward

E p n p C E n p n C
B (Base) B (Base)

* In analog circuits, BJTs are generally biased to operate in the active mode.

E p n p C E n p n C
B (Base) B (Base)

* In analog circuits, BJTs are generally biased to operate in the active mode.
* BJT as a switch:
- Closed: saturation mode
- Open: cutoff mode

Device structure (pnp transistor)
Emitter
Base
insulator
z
p epi-layer
y
x n p+
rb
p n+
Rsub
p+
(substrate)
Collector

* The substrate thickness is hundreds of microns whereas

the p epi-layer and the rest of the device structure is
confined to a few microns. Emitter
Base
insulator
z
p epi-layer
y
x n p+
rb
p n+
Rsub
p+
(substrate)
Collector


Base
* The p + substrate can be treated as a small resistance
insulator
Rsub .
z
p epi-layer
y
x n p+
rb
p n+
Rsub
p+
(substrate)
Collector


Base
insulator
Rsub .
* The “active” region of the device is only the top few z
microns (the p epi-layer). p epi-layer
y
x n p+
rb
p n+
Rsub
p+
(substrate)
Collector


Base
insulator
Rsub .
y
* The figure is not drawn to scale. Typically, the y and z x n p+
dimensions are much larger than the p epi thickness. rb
p n+
Rsub
p+
(substrate)
Collector


Base
insulator
Rsub .
y
* Because the contact region dimensions are much larger p n+
than the base width (∼ the width of the n base region),
the device can be considered to be one-dimensional in
nature. Rsub
p+
(substrate)
Collector


Base
insulator
Rsub .
y
* Because the contact region dimensions are much larger p n+
than the base width (∼ the width of the n base region),
the device can be considered to be one-dimensional in
nature. Rsub
p+
* A “base resistance” rb exists between the base region and (substrate)
the base contact. To keep rb small, the base contact is
made close to the emitter.
Collector

Emitter
Base
insulator
y
x n p+
rb
p n+
Rsub
p+
(substrate)
Collector
Emitter
Base 1020
Doping density (cm−3 )

insulator Nd (B)
1018
z Na (E)
Na (C)
y 1016
x n p+
rb 1014
p n+
Rsub
p+
(substrate)
Collector
Emitter
Base 1020

insulator Nd (B)
1018
z Na (E)
Na (C)
y 1016
x n p+
rb 1014
p n+
Rsub
p+
(substrate)
Collector
Emitter
Base 1020

insulator Nd (B)
1018
z Na (E)
Na (C)
y 1016
x n p+
rb 1014
p n+
Rsub
p+
(substrate)
Collector
* For simplicity, we will assume the doping densities to be constant in the emitter, base, and collector regions.
Emitter
Base 1020

insulator Nd (B)
1018
z Na (E)
Na (C)
y 1016
x n p+
rb 1014
p n+ NEa
1020

Rsub 1018
p+ NBd
(substrate)
1016 NCa
Collector 1014
0 1.5 3
x (µm)

Emitter
Base 1020

insulator Nd (B)
1018
z Na (E)
Na (C)
y 1016
x n p+
rb 1014
p n+ NEa
1020

Rsub 1018
p+ NBd
(substrate)
1016 NCa
Collector 1014
0 1.5 3
x (µm)
* The relationship NaE > NdB > NaC , which is a consequence of the fabrication process, is also desirable from the
device performance angle.
BJT structure in integrated circuits (npn transistor)
insulator base emitter collector substrate
contact contact contact contact
p+ n+ n+
p+
p
epi-layer ∼ 5 µm
n
n+ buried layer
p+ substrate > 250 µm

(Not drawn to scale. Typically, the
horizontal dimensions are much larger
than the junction depths.)

p+ n+ n+
p+
p
epi-layer ∼ 5 µm
n
n+ buried layer

* To make an integrated circuit (IC), a large number of transistors are fabricated on a single silicon piece
and interconnected as required.

p+ n+ n+
p+
p
epi-layer ∼ 5 µm
n
n+ buried layer

* The BJTs are isolated from each other using reverse-biased pn junctions.

p+ n+ n+
p+
p
epi-layer ∼ 5 µm
n
n+ buried layer

* Contacts (E, B, C) are made on the top surface for connecting to other transistors.

p+ n+ n+
p+
p
epi-layer ∼ 5 µm
n
n+ buried layer

* Contacts (E, B, C) are made on the top surface for connecting to other transistors.
* An n+ buried layer is used to provide a low-resistance path for the electron current.

Dependence of α on device parameters
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
emitter collector
current current
recombination
base
x=0 x=W
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
emitter collector
current current
recombination
base
x=0 x=W
Consider a pnp transistor.

* IE has a hole component and an electron component. Of these, only the hole component contributes to IC .
IpE IpE
We define “emitter injection efficiency” (or simply “injection efficiency”) as γ = = .
IE IpE + InE
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
emitter collector
current current
recombination
base
x=0 x=W

IpE IpE
IE IpE + InE
* Some of the injected holes are lost due to recombination in the neutral base region (we will ignore G-R in the
B-C depletion regions).
IpC
We define “base transport factor” as αT = .
IpE
E-B C-B E-B C-B
depletion depletion depletion depletion
region region region region
E (p) B (n) C (p) E (n) B (p) C (n)
emitter collector emitter collector

current current current current
recombination recombination
base base
x=0 x=W x=0 x=W

IpE IpE
IE IpE + InE
* Some of the injected holes are lost due to recombination in the neutral base region (we will ignore G-R in the
B-C depletion regions).
IpC
We define “base transport factor” as αT = .
IpE
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
emitter collector
current current
recombination
base
x=0 x=W
Since the C-B junction is reverse biased, the pn junction current arising because of VCB is negligibly small.

E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
emitter collector
current current
recombination
base
x=0 x=W
→ IC is entirely due to the holes injected by the emitter which make it to the C-B depletion boundary (x = W ), i.e.,
IC
IC ≈ IpC = αT IpE = αT (γ IE ) → α = = γ αT .
IE

E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
emitter collector
current current
recombination
base
x=0 x=W
→ IC is entirely due to the holes injected by the emitter which make it to the C-B depletion boundary (x = W ), i.e.,
IC
IC ≈ IpC = αT IpE = αT (γ IE ) → α = = γ αT .
IE
→ For α ≈ 1, both γ and αT must be close to 1.

E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
p(x)
n(x)
n0E
xE 0 W
We assume that the emitter width is greater than 5 Ln .

E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
p(x)
n(x)
n0E
xE 0 W

Neglecting the drift components for minority carriers in the emitter and base neutral regions, we get
d 2 ∆n ∆n
DnE − = 0, x < xE , with
dx 2 τnE

VEB
∆n(xE ) = n0E exp −1 ,
VT
∆n(−∞) = 0.
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
p(x)
n(x)
n0E
xE 0 W

Neglecting the drift components for minority carriers in the emitter and base neutral regions, we get
d 2 ∆p ∆p
d 2 ∆n ∆n DpB − = 0, 0 < x < W , with
DnE − = 0, x < xE , with dx 2 τpB
dx 2 τnE
VEB
VEB ∆p(0) = p0B exp −1 ,
∆n(xE ) = n0E exp −1 , VT
VT
VCB
∆n(−∞) = 0. ∆p(W ) = p0B exp −1 .
VT
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
p(x)
n(x)
n0E
xE 0 W
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
p(x)
n(x)
n0E
xE 0 W
Solution: ∆n(x) = ∆n(xE ) e −(xE −x)/LnE , x < xE ,

∆p(x) = A e −x/LpB + B e +x/LpB , 0 < x < W .
E-B C-B
depletion depletion
region region
E (p) B (n) C (p)
p(x)
n(x)
n0E
xE 0 W

Using the boundary conditions (last slide), we get
W −x

x
sinh sinh
LpB LpB
∆p(x) = ∆p(0) + ∆p(W ) .
W W
sinh sinh
LpB LpB
E-B C-B
depletion depletion
region region
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
p(x) 1
n(x)
n0E 0
0 0.2 0.4 0.6 0.8 1
xE x/W
0 W

Using the boundary conditions (last slide), we get
W −x

x
sinh sinh
LpB LpB
∆p(x) = ∆p(0) + ∆p(W ) .
W W
sinh sinh
LpB LpB
E-B C-B
depletion depletion
region region
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
p(x) 1
n(x)
n0E 0
0 0.2 0.4 0.6 0.8 1
xE x/W
0 W
dn d∆n
InE = qADnE (xE ) = qADnE (xE )
dx dx
qADnE qADnE
= ∆n(xE ) = n0E e VEB /VT − 1 ,
LnE LnE

E-B C-B
depletion depletion
region region
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
p(x) 1
n(x)
n0E 0
0 0.2 0.4 0.6 0.8 1
xE x/W
0 W
dn d∆n
InE = qADnE (xE ) = qADnE (xE )
dx dx
qADnE qADnE
= ∆n(xE ) = n0E e VEB /VT − 1 ,
LnE LnE
dp d∆p
IpE = −qADpB (0) = −qADpB (0)
dx dx
qADpB cosh(W /L )
pB
= p0B e VEB /VT − 1 .
LpB sinh(W /LpB )

E-B C-B
depletion depletion
region region
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
p(x) 1
n(x)
n0E 0
0 0.2 0.4 0.6 0.8 1
xE x/W
0 W
dn d∆n
InE = qADnE (xE ) = qADnE (xE ) IpE 1
dx dx γ = =
IpE + InE 1 + (InE /IpE )
qADnE qADnE
= ∆n(xE ) = n0E e VEB /VT − 1 , 1
LnE LnE = ,
DnE LpB NdB sinh(W /LB )

dp d∆p 1+
IpE = −qADpB (0) = −qADpB (0) DpB LnE NaE cosh(W /LB )
dx dx
cosh(W /L ) n0E n2 NdB NdB
qADpB since = i × 2 = .

pB
= p0B e VEB /VT − 1 . p0B NaE ni NaE
LpB sinh(W /LpB )

Base transport factor
E (p) B (n) C (p) E (p) B (n) C (p)
p(x) emitter collector

current current
recombination
n(x)
base
W −x x=0 x=W

x
sinh sinh 1
LpB LpB
∆p(x) = ∆p(0) + ∆p(W ) .
W W W/LpB = 0.5
∆p(x)/∆p(0)
sinh sinh
LpB LpB 1
0
0 0.2 0.4 0.6 0.8 1
x/W

E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
W −x x=0 x=W

x
sinh sinh 1
LpB LpB
∆p(x) = ∆p(0) + ∆p(W ) .
W W W/LpB = 0.5
∆p(x)/∆p(0)
sinh sinh
LpB LpB 1
d∆p 2
IC ≈ IpC = −qADpB (W )
dx
0
0 0.2 0.4 0.6 0.8 1
x/W

E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
W −x x=0 x=W

x
sinh sinh 1
LpB LpB
∆p(x) = ∆p(0) + ∆p(W ) .
W W W/LpB = 0.5
∆p(x)/∆p(0)
sinh sinh
LpB LpB 1
d∆p 2
dx
qADpB 1
= p0B e VEB /VT − 1 .
LpB sinh(W /LB ) 0
0 0.2 0.4 0.6 0.8 1
x/W

E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
W −x x=0 x=W

x
sinh sinh 1
LpB LpB
∆p(x) = ∆p(0) + ∆p(W ) .
W W W/LpB = 0.5
∆p(x)/∆p(0)
sinh sinh
LpB LpB 1
d∆p 2
dx
qADpB 1
= p0B e VEB /VT − 1 .
LpB sinh(W /LB ) 0
0 0.2 0.4 0.6 0.8 1
The base transport factor is (using IpE from the last slide), x/W
IpC 1
αT = = .
IpE cosh(W /LpB )
αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
W −x

x cosh (x)
sinh sinh x2
LpB LpB 2 1+
∆p(x) = ∆p(0) 2
+ ∆p(W ) .
W W 1.5
sinh sinh
LpB LpB
1 sinh (x)
0.5 x
0
0 0.5 1 1.5
x

αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
W −x

x cosh (x)
sinh sinh x2
LpB LpB 2 1+
∆p(x) = ∆p(0) 2
+ ∆p(W ) .
W W 1.5
sinh sinh
LpB LpB
1 sinh (x)
IpC 1 1
αT = = ≈ 2 . 0.5 x
IpE cosh(W /LpB ) 1

W
1+
2 LpB 0
0 0.5 1 1.5
x

αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
W −x

x cosh (x)
sinh sinh x2
LpB LpB 2 1+
∆p(x) = ∆p(0) 2
+ ∆p(W ) .
W W 1.5
sinh sinh
LpB LpB
1 sinh (x)
IpC 1 1
αT = = ≈ 2 . 0.5 x
IpE cosh(W /LpB ) 1

W
1+
2 LpB 0
0 0.5 1 1.5
Remark: αT → 1 if the base width W is made small compared to LpB . x

γ with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
IpE 1 1 cosh (x)
γ= = = x2
NdB sinh(W /LB )

IpE + InE 1 + (InE /IpE ) DnE LpB 2 1+
1+ 2
DpB LnE NaE cosh(W /LB )
1.5
1 sinh (x)
0.5 x
0
0 0.5 1 1.5
x

γ with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
IpE 1 1 cosh (x)
γ= = = x2
NdB sinh(W /LB )

1+ 2
1.5
1
≈ 1
DnE LpB NdB W /LpB sinh (x)
1+
DpB LnE NaE 1 + 12 (W /LpB )2 0.5 x
0
0 0.5 1 1.5
x

γ with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
IpE 1 1 cosh (x)
γ= = = x2
NdB sinh(W /LB )

1+ 2
1.5
1
≈ 1
DnE LpB NdB W /LpB sinh (x)
1+
DpB LnE NaE 1 + 12 (W /LpB )2 0.5 x
1
≈ 0
DnE W NdB 0 0.5 1 1.5
1+ x
DpB LnE NaE
γ with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
1 cosh (x)
γ≈ . 2 x2
1+

DnE W NdB 2
1+
DpB LnE NaE 1.5
1 sinh (x)
0.5 x
0
0 0.5 1 1.5
x

γ with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
1 cosh (x)
γ≈ . 2 x2
1+

DnE W NdB 2
1+
DpB LnE NaE 1.5
* γ → 1 if NaE NdB . 1 sinh (x)
0.5 x
0
0 0.5 1 1.5
x

γ with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
2.5
1 cosh (x)
γ≈ . 2 x2
1+

DnE W NdB 2
1+
DpB LnE NaE 1.5
* γ → 1 if NaE NdB . 1 sinh (x)

* It is now clear why a higher doping density in the emitter region 0.5 x
(compared to the base doping density) is desirable.
0
0 0.5 1 1.5
x

Example E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
x=0 x=W
Consider a pnp BJT with NaE = 1018 cm−3 , NdB = 5 × 1016 cm−3 , NaC = 1015 cm−3 , and with
a base width W = 2 µm (T = 300 K).
(a) Calculate αT , γ, α, and β, using the following parameters.
µnE = 250 cm2 /V-s, µpB = 500 cm2 /V-s, µnC = 1500 cm2 /V-s,
τnE = 0.2 µs, τpB = 1 µs, τnC = 1 µs.


current current
recombination
n(x)
base
x=0 x=W
Consider a pnp BJT with NaE = 1018 cm−3 , NdB = 5 × 1016 cm−3 , NaC = 1015 cm−3 , and with
a base width W = 2 µm (T = 300 K).
(a) Calculate αT , γ, α, and β, using the following parameters.
µnE = 250 cm2 /V-s, µpB = 500 cm2 /V-s, µnC = 1500 cm2 /V-s,
τnE = 0.2 µs, τpB = 1 µs, τnC = 1 µs.
(b) Repeat (a) for the BJT operating in the reverse active mode.


current current
recombination
n(x)
base
x=0 x=W
Solution:
The minority carrier diffusion lengths are
√ √ √
LnE = DnE τnE = VT µnE τnE = 0.0258 × 250 × 0.2 × 10−6 = 1.14 × 10−3 cm = 11.4 µm.
√
LpB = DpB τpB = VT µpB τpB = 0.0258 × 500 × 1 × 10−6 = 3.59 × 10−3 cm = 35.9 µm.
p p
√ √ √
LnC = DnC τnC = VT µnC τnC = 0.0258 × 1500 × 1 × 10−6 = 6.22 × 10−3 cm = 62.2 µm.
Note that LpB W (2 µm).


current current
recombination
n(x)
base
x=0 x=W
DnE W NdB µnE W NdB 250 2 × 10−4 5 × 1016

(a) = = = 4.386 × 10−3 .
DpB LnE NaE µpB LnE NaE 500 1.14 × 10−3 1018


current current
recombination
n(x)
base
x=0 x=W
DnE W NdB µnE W NdB 250 2 × 10−4 5 × 1016

(a) = = = 4.386 × 10−3 .
1
γ= = 0.9956.
1 + 4.386 × 10−3


current current
recombination
n(x)
base
x=0 x=W
DnE W NdB µnE W NdB 250 2 × 10−4 5 × 1016

(a) = = = 4.386 × 10−3 .
1
γ= = 0.9956.
1 + 4.386 × 10−3
1 1
αT = = = 0.9985.
1 + 12 (W /LpB )2 1 + 12 (2.0/35.9)2


current current
recombination
n(x)
base
x=0 x=W
DnE W NdB µnE W NdB 250 2 × 10−4 5 × 1016

(a) = = = 4.386 × 10−3 .
1
γ= = 0.9956.
1 + 4.386 × 10−3
1 1
αT = = = 0.9985.
1 + 12 (W /LpB )21 + 12 (2.0/35.9)2
α
α = γαT = 0.9940 → β = = 166.
1−α


current current
recombination
n(x)
base
x=0 x=W
(b) With E ↔ C ,
DnE W NdB DnC W NdB µnC W NdB 1500 2 × 10−4 5 × 1016
→ = = = 4.823.
DpB LnE NaE DpB LnC NaC µpB LnC NaC 500 6.22 × 10−3 1015
1 1
γ= = 0.1717, αT = = 0.9985.
1 + 4.823 1 + 12 (2/35.9)2
→ α = γ αT = 0.1714 → β = 0.2, a disaster.
Conclusion: NaE NdB is crucial.
(Note that we have treated W as a constant, but it would vary with bias conditions.)

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
xE − x

d VEB
InE = −q A DnE n0E exp exp − n at x = xnE
dx VT LnE

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
xE − x

d VEB
dx VT LnE
1

VEB
= q A DnE n0E exp ×
VT LnE

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
xE − x

d VEB
dx VT LnE
1

VEB
VT LnE
1

VEB
IpE ≈ q A DpB p0B exp ×
VT W

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
xE − x

d VEB
dx VT LnE
1

VEB
VT LnE
1

VEB
VT W
InE DnE n0E W
≈
IpE DpB p0B LnE

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
xE − x

d VEB
dx VT LnE
1

VEB
VT LnE
1

VEB
VT W
InE DnE n0E W DnE ni2 NdB W
≈ =
IpE DpB p0B LnE DpB NaE ni2 LnE

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
xE − x

d VEB
dx VT LnE
1

VEB
VT LnE
1

VEB
VT W
InE DnE n0E W DnE ni2 NdB W DnE NdB W
≈ = = .
IpE DpB p0B LnE DpB NaE ni2 LnE DpB NaE LnE

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
InE DnE NdB W

≈
IpE DpB NaE LnE

γ with W << LpB
E (p) B (n) C (p) E (p) B (n) C (p)

current current
recombination
n(x)
base
xEn x=0 x=W
InE DnE NdB W

≈
IpE DpB NaE LnE
IpE 1 1
→γ= = ≈ .
InE

IpE + InE DnE W NdB
1+ 1+
IpE DpB LnE NaE

αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W
When W LpB , ∆p(x) is linear.

IpC IpC 1
αT = = = .
IpE IpC + IpB IpB
1+
IpC

αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W

IpC IpC 1
αT = = = .
IpE IpC + IpB IpB
1+
IpC
dp ∆p(0)
IpC = −q A DpB (W ) ≈ q A DpB .
dx W

αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W

IpC IpC 1
αT = = = .
IpE IpC + IpB IpB
1+
IpC
dp ∆p(0)
dx W
1
Qp qA 2
∆p(0) W
IpB = =
τpB τpB

αT with W LpB
1
E (p) B (n) C (p)
W/LpB = 0.5
∆p(x)/∆p(0)
emitter collector
1
current current
2
recombination
0
base 0 0.2 0.4 0.6 0.8 1
x=0 x=W x/W

IpC IpC 1
αT = = = .
IpE IpC + IpB IpB
1+
IpC
dp ∆p(0)
dx W
1
Qp qA 2
∆p(0) W 1
IpB = = → αT ≈ 2 .
τpB τpB 1

W
1+
2 LpB

sd_bjt_1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

sd_bjt_1

Uploaded by

Copyright:

Available Formats

Semiconductor Devices

Bipolar Junction Transistors: Part 1

Department of Electrical Engineering

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

* In the “active” or “linear” mode, the B B

M. B. Patil, IIT Bombay

* In the “active” or “linear” mode, the B B

M. B. Patil, IIT Bombay

* In the “active” or “linear” mode, the B B

M. B. Patil, IIT Bombay

* In the “active” or “linear” mode, the B B

M. B. Patil, IIT Bombay

* In the “active” or “linear” mode, the B B

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

* The emitter arrow in the BJT symbol B B

M. B. Patil, IIT Bombay

* The emitter arrow in the BJT symbol B B

M. B. Patil, IIT Bombay

* The emitter arrow in the BJT symbol B B

M. B. Patil, IIT Bombay

* The emitter arrow in the BJT symbol B B

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

* The “common-emitter current gain” β – a figure of merit of a BJT – is defined as β = IC /IB .

M. B. Patil, IIT Bombay

* The “common-emitter current gain” β – a figure of merit of a BJT – is defined as β = IC /IB .

M. B. Patil, IIT Bombay

IE IC = αIE IE IC = αIE 0.95 0.05 19

* The “common-emitter current gain” β – a figure of merit of a BJT – is defined as β = IC /IB .

M. B. Patil, IIT Bombay

IE IC = αIE IE IC = αIE 0.95 0.05 19

* The “common-emitter current gain” β – a figure of merit of a BJT – is defined as β = IC /IB .

M. B. Patil, IIT Bombay

Mode B-E junction B-C junction

Active (linear) forward reverse

M. B. Patil, IIT Bombay

Mode B-E junction B-C junction

Active (linear) forward reverse

M. B. Patil, IIT Bombay

Mode B-E junction B-C junction

Active (linear) forward reverse

M. B. Patil, IIT Bombay

M. B. Patil, IIT Bombay

* The substrate thickness is hundreds of microns whereas

M. B. Patil, IIT Bombay

* The substrate thickness is hundreds of microns whereas

M. B. Patil, IIT Bombay

* The substrate thickness is hundreds of microns whereas

M. B. Patil, IIT Bombay

* The substrate thickness is hundreds of microns whereas

M. B. Patil, IIT Bombay

* The substrate thickness is hundreds of microns whereas

M. B. Patil, IIT Bombay

* The substrate thickness is hundreds of microns whereas

M. B. Patil, IIT Bombay

* γ → 1 if NaE NdB . 1 sinh (x)

* γ → 1 if NaE NdB . 1 sinh (x)