Bipolar Junction Transistors (BJT)
A Bipolar Transistor essentially consists of a pair of PN Junction
Diodes that are joined back-to-back.
1
BJT operation
n
p
Built-in
voltage
Forward polarity is
applied to the left diode.
Positive battery polarity
attracts electrons.
If the battery voltage
exceeds the built-in
voltage, there is a high
current flowing through.
p
n
Built-in
voltage
Reverse polarity is applied
to the right diode
Positive battery polarity
attracts electrons.
There are no electrons on
the p-side of the diode.
The diode current is nearly
zero.
2
BJT operation
n
p
Battery 1
Forward polarity is
applied to the left diode.
There is a high current
flowing through it.
The electrons entering
p-region recombine with
holes to form a current
loop.
p
n
Battery 2
Reverse polarity is applied
to the right diode
There are still no electrons
on the p-side of the diode
(they have all recombined
at the left junction).
The current is nearly zero.
3
BJT operation
p
n
Battery 1
n
Battery 2
The P-layer is made VERY THIN
• Forward polarity is applied to the left diode. Electrons from
the left n-region get injected into the p-layer driven by a
positive terminal of the battery 1.
• The p-layer being very thin, most of the injected electrons
have low chances to recombine; they reach the 2nd junction
and drift toward positive terminal of the battery 2.
• Therefore high current flows through 2nd p-n junction.
4
n-p-n bipolar transistor
Device structure
Emitter
n
Circuit symbol
C
Base Collector
p
n
B
E
p-n-p bipolar transistor
Device structure
Emitter
p
Circuit symbol
C
Base Collector
n
p
B
E
5
Electron and hole fluxes in n-p-n transistor:
e
n
p
n
c
1
3
4
2
b
(1) electrons injected from emitter into the base and reaching the collector junction (as large as
possible);
(2) holes supplied by the base contact to initiate electron injection (as small as possible);
(3) injected electrons lost to recombination in the base (as small as possible);
(4) holes injected from the base into emitter recombining with electrons (as small as possible).
6
Electron and hole fluxes in p-n-p transistor:
e
p
n
p
c
1
3
4
2
b
(1) holes injected from emitter into the base and reaching the collector junction;
(2) electrons supplied by the base contact to initiate hole injection;
(3) injected holes lost to recombination in the base;
(4) electrons injected from the base into emitter recombining with holes.
7
Carrier concentrations and Terminal Currents
(p-n-p transistor case):
p(x)
e
p
Δpe
p
n
Δpc
0
c
x
b W
Hole concentration at the e-b junction:
Hole concentration at the b-c junction (Vcb <0):
Δpe = pb0 (eqVbe / kT − 1) >> pb0
Δpc = pb0 (eqVcb / kT − 1) ≈ − pb0
Within the base, the injected hole concentration decreases linearly with x
Why linearly?
8
p(x)
e
p
n
p
c
1
3
4
2
x
b
ƒ The base width W is larger than that of the e-b and c-b depletion regions.
Therefore, the electric field in most of the base, E ≈ 0.
ƒ Therefore, the hole current (1) through the base is mainly the diffusion current:
I p = Aq D p
∂p
∂x
ƒThe current (1) is nearly constant throughout the base because the
recombination current (3) is very small.
ƒTherefore: dp/dx =const;
ƒThe line that has a constant slope is a straight line.
9
Terminal Currents : Collector and Emitter currents
(p-n-p transistor case):
p(x)
e
p
Δpe
p
n
Δpc
0
c
x
b W
Assuming small recombination loss in the base,
Δpe
∂p
I c ≈ I e ≈ Aq D ≈ Aq D p
∂x
W
W is the BJT base thickness
10
Terminal Currents : Base recombination current
(p-n-p transistor case):
p(x)
e
p
Δpe
p
n
Δpc
0
c
x
b W
Total charge of the holes injected into the base (the area under the Δp(x) line:
Δpe W
Q p = Aq
2
This charge reproduces due to recombination; the required time is the hole lifetime τp
Therefore, the base recombination current, Ibrec = Q/t:
Δpe W
Ibrec = Aq
2τ p
11
Current gain (recombination limited)
Collector current:
Δpe
I c ≈ Aq D p
W
Base recombination current:
Δpe W
Ibrec = Aq
2τ p
Collector-base current gain (limited by recombination only) βrec = Ic/Ibrec:
β rec =
The diffusion length:
Ic
Ibrec
≈
Lp = D p τ p
Δpe
2τ p 2 D τ
p p
W
=
Aq Δpe W
W2
Aq D p
Therefore:
β rec =
2 L2p
W2
12
Emitter injection efficiency
e
p
n
p
c
1
3
4
2
b
Electrons injected from the base into emitter – the current component (4), increase total base
and emitter currents but DO NOT contribute into the collector current as they do not
increase the injection.
The ratio of the “useful” emitter current (the hole current in p-n-p device) over the total
emitter current is called emitter injection efficiency”; for a p-n-p transistor:
γ=
I ep
I ep + I en
13
Emitter injection efficiency (cont.)
For a p-n-p transistor:
p
n
p
e
c
1
I ep
1
γ=
=
I en
I ep + I en
1+
I ep
3
4
2
b
The electron and hole currents are mainly the diffusion currents:
For the hole current in the base: Iep ≈ qA Δpeb Dp/W;
The electron current from the base into the emitter: Ien ≈ qA Δnbe Dn/Ln;
The current ratio:
I en Δnbe Dn W
=
I ep Δpeb D p Ln
2
n
Δnbe = n p 0 eqV / kT = i eqV / kT
N Ae
Δpeb = pn0 e
qV / kT
ni2 qV / kT
e
=
N Db
must be as small as possible for the
highest injection efficiency
Δnbe N Db
=
<<1;
Δpeb N Ae
14
Emitter injection efficiency (cont.)
For any BJT type (p-n-p or n-p-n) high emitter injection efficiency requires:
Nb
<<1;
Ne
Concentration (cm -3 )
where Nb and Ne are the base and the emitter doping levels
10
21
10 19
10
..
17
15
n+
10
0
p
0.8
n
1.6
2.4
3.2
Distance (µm)
Typical doping profile in n-p-n BJT
15
BJT current gain
Current gain limited by recombination in the base:
p-n-p BJT:
β rec =
2 L2p
W
2
β rec =
n-p-n BJT:
2 L2n
W2
Current gain limited by injection efficiency:
N Ae D p Ln
p-n-p BJT: β0 =
N Db Dn W
n-p-n BJT:
β0 is related to emitter injection efficiency γ as
β=
The overall current gain:
β0 =
N De Dn L p
β0 =
N Ab D p W
γ
1− γ
1
⎛ 1
1 ⎞
+
⎜
⎟
β
β
0⎠
⎝ rec
16