14

C H A P T E R

Optical Devices


n previous chapters, we have considered the basic physics of transistors that are
used to amplify or switch electrical signals. Semiconductor devices can be de-
signed to convert optical energy into electrical energy, and to convert electrical
signals into optical signals. These devices are used in broadband communications
and data transmission over optical fibers. The general classification of these devices
is called optoelectronics.
In this chapter, we discuss the basic principles of solar cells, several photodetec-
tors, light emitting diodes, and laser diodes. Solar cells and photodetectors convert
optical energy into electrical energy; light emitting diodes and laser diodes convert
electrical signals into optical signals. ■

14.0 | PREVIEW
In this chapter, we will:
■ Discuss and analyze photon absorption in a semiconductor and present
absorption coefficient data for several semiconductor materials.
■ Consider the basic principles of solar cells, analyze their I–V characteristics,
and discuss the conversion efficiency.
■ Present various types of solar cells, including homojunction, heterojunction,
and amorphous silicon solar cells.
■ Discuss the basic principles of photodetectors, including photoconductors,
photodiodes, and phototransistors.
■ Derive the output current characteristics of the various photodectors.
■ Present and analyze the basic operation of the Light Emitting Diode (LED).
■ Discuss the basic principles and operation of the laser diode.

618

nea29583_ch14_618-669.indd 618 12/11/10 12:51 PM

 14.1 Optical Absorption 619

14.1 | OPTICAL ABSORPTION

In Chapter 2, we discussed the wave–particle duality principle and indicated that light
waves could be treated as particles, which are referred to as photons. The energy of
a photon is E  h where h is Plank’s constant and  is the frequency. We can also
relate the wavelength and energy by
c _
_ hc _
1.24
  E  E m (14.1)

where E is the photon energy in eV and c is the speed of light.

There are several possible photon–semiconductor interaction mechanisms. For
example, photons can interact with the semiconductor lattice whereby the photon
energy is converted into heat. Photons can also interact with impurity atoms, either
donors or acceptors, or they can interact with defects within the semiconductor. How-
ever, the basic photon interaction process of greatest interest is the interaction with
valence electrons. When a photon collides with a valence electron, enough energy
may be imparted to elevate the electron into the conduction band. Such a process
generates electron–hole pairs and creates excess carrier concentrations. The behavior
of excess carriers in a semiconductor was considered in Chapter 6.

14.1.1 Photon Absorption Coefficient

When a semiconductor is illuminated with light, the photons may be absorbed or
they may propagate through the semiconductor, depending on the photon energy and
on the bandgap energy Eg. If the photon energy is less than Eg, the photons are not
readily absorbed. In this case, the light is transmitted through the material and the
semiconductor appears to be transparent.
If E  h  Eg, the photon can interact with a valence electron and elevate the
electron into the conduction band. The valence band contains many electrons and the
conduction band contains many empty states, so the probability of this interaction is
high when h  Eg. This interaction creates an electron in the conduction band and
a hole in the valence band—an electron–hole pair. The basic absorption processes
for different values of h are shown in Figure 14.1. When h  Eg, an electron–hole


 ()
Ec

h

Ev
()
h  Eg h  Eg
h  Eg

Figure 14.1 | Optically generated

electron–hole pair formation in a
semiconductor.

nea29583_ch14_618-669.indd 619 12/11/10 12:51 PM

 620 CHAPTER 14 Optical Devices

pair is created and the excess energy may give the electron or hole additional kinetic
energy, which will be dissipated as heat in the semiconductor.
The intensity of the photon flux is denoted by I(x) and is expressed in terms of
energy/cm2-s. Figure 14.2 shows an incident photon intensity at a position x and the
photon flux emerging at a distance x  dx. The energy absorbed per unit time in the
distance dx is given by
I (x) dx (14.2)
where  is the absorption coefficient. The absorption coefficient is the relative num-
ber of photons absorbed per unit distance, given in units of cm1.
From Figure 14.2, we can write
dI (x)
I (x  dx)I (x)  _ ⴢ dx  I (x) dx (14.3)
dx
or
dI (x)
_  I (x) (14.4)
dx
If the initial condition is given as I (0)  I0, then the solution to the differential
equation, Equation (14.4), is
I (x)  I0ex (14.5)
The intensity of the photon flux decreases exponentially with distance through the
semiconductor material. The photon intensity as a function of x for two general val-
ues of absorption coefficient is shown in Figure 14.3. If the absorption coefficient is
large, the photons are absorbed over a relatively short distance.
The absorption coefficient in the semiconductor is a very strong function of pho-
ton energy and bandgap energy. Figure 14.4 shows the absorption coefficient  plot-
ted as a function of wavelength for several semiconductor materials. The absorption
coefficient increases very rapidly for h  Eg, or for   1.24Eg. The absorption

I0

I(x) I(x  dx)

Large 
Small 
I

dx

x
Figure 14.2 | Optical
absorption in a differential Figure 14.3 | Photon intensity versus
length. distance for two absorption coefficients.

nea29583_ch14_618-669.indd 620 12/11/10 12:51 PM

 14.1 Optical Absorption 621

Photon energy (eV)

3 2 1.5 1 0.7
106
Ge

105
Ga 0.3In0.7As0.64P0.36

InGaAs
Absorption coefficient (cm1)

GaAs
104 Si

InP

103

GaP

102
Amorphous
Si

10
0.2 0.6 1 1.4 1.8
Wavelength (m)

Figure 14.4 | Absorption coefficient as a function of

wavelength for several semiconductors.
(From Shur [13].)

coefficients are very small for h  Eg, so the semiconductor appears transparent to
photons in this energy range.

Objective: Calculate the thickness of a semiconductor that will absorb 90 percent of the EXAMPLE 14.1
incident photon energy.
Consider silicon and assume that in the first case the incident wavelength is   1.0 m
and in the second case, the incident wavelength is   0.5 m.

■ Solution
From Figure 14.4, the absorption coefficient is   102 cm1 for   1.0 m. If 90 percent
of the incident flux is to be absorbed in a distance d, then the flux emerging at x  d will be
10 percent of the incident flux. We can write
I (d )
_  0.1  ed
I0
Solving for the distance d, we have
1
d_  
1 1
 ln 0.1  102 ln (10)  0.0230 cm
_ _

nea29583_ch14_618-669.indd 621 12/11/10 12:51 PM

 622 CHAPTER 14 Optical Devices

In the second case, the absorption coefficient is   104 cm1 for   0.5 m. The distance d,
then, in which 90 percent of the incident flux is absorbed, is

d_1 ln _
104 0.1  
1  2.30 104 cm  2.30 m

■ Comment
As the incident photon energy increases, the absorption coefficient increases rapidly, so
that the photon energy can be totally absorbed in a very narrow region at the surface of the
semiconductor.

■ EXERCISE PROBLEM
Ex 14.1 Consider a slab of silicon 5 m thick. Determine the percentage of photon energy
that will pass through the slab if the photon wavelength is (a)   0.8 m and
(b)   0.6 m.
[Ans. (a) 60.7%; (b) 10.5%]

The relation between the bandgap energies of some of the common semicon-
ductor materials and the light spectrum is shown in Figure 14.5. We may note that
silicon and gallium arsenide will absorb all of the visible spectrum, whereas gallium
phosphide, for example, will be transparent to the red spectrum.

14.1.2 Electron–Hole Pair Generation Rate

We have shown that photons with energy greater than Eg can be absorbed in a semi-
conductor, thereby creating electron–hole pairs. The intensity I(x) is in units of

Relative
eye response

Full width, half maximum

m  0.555 m
Infrared Red Green Violet Ultraviolet
Orange Yellow Blue

Si GaAs CdSe GaP CdS SiC GaN ZnS

GaAs1y Py

 (m)
1.0 0.9 0.8 0.7 0.6 0.5 0.45 0.4 0.35

1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6
Eg (eV)

Figure 14.5 | Light spectrum versus wavelength and

energy. Figure includes relative response of the human eye.
(From Sze [18].)

nea29583_ch14_618-669.indd 622 12/11/10 12:51 PM

 14.1 Optical Absorption 623

energy/cm2-s and I (x) is the rate at which energy is absorbed per unit volume. If
we assume that one absorbed photon at an energy h creates one electron–hole pair,
then the generation rate of electron–hole pairs is
I(x)
g
 _ (14.6)
h
which is in units of #/cm3-s. We may note that the ratio I (x)h is the photon flux. If,
on the average, one absorbed photon produces less than one electron–hole pair, then
Equation (14.6) must be multiplied by an efficiency factor.

Objective: Calculate the generation rate of electron–hole pairs given an incident intensity EXAMPLE 14.2
of photons.
Consider gallium arsenide at T  300 K. Assume the photon intensity at a particular point
is I (x)  0.05 W/cm2 at a wavelength of   0.75 m. This intensity is typical of sunlight,
for example.

■ Solution
The absorption coefficient for gallium arsenide at this wavelength is   0.9 104 cm1. The
photon energy, using Equation (14.1), is
1.24  1.65 eV
E  h  _
0.75
Then, from Equation (14.6) and including the conversion factor between joules and eV, we
have, for a unity efficiency factor,
I (x) ____
(0.9 104)(0.05)
g
 __   1.70 1021 cm3-s1
h (1.6 1019)(1.65)
If the incident photon intensity is a steady-state intensity, then, from Chapter 6, the steady-
state excess carrier concentration is n  g
, where  is the excess minority carrier lifetime.
If   107 s, for example, then

n  (1.70 1021)(107)  1.70 1014 cm3

■ Comment
This example gives an indication of the magnitude of the electron–hole generation rate and the
magnitude of the excess carrier concentration. Obviously, as the photon intensity decreases
with distance in the semiconductor, the generation rate also decreases.

■ EXERCISE PROBLEM
Ex 14.2 A photon flux with an intensity of I0  0.10W/cm2 and at a wavelength of
  1 m is incident on the surface of silicon. Neglecting any reflection from
the surface, determine the generation rate of electron–hole pairs at a depth of
(a) x  5 m and (b) x  20 m from the surface.
[Ans. (a) 4.79 1019 cm3 s1; (b) 4.13 1019 cm3 s1]

nea29583_ch14_618-669.indd 623 12/11/10 12:51 PM

 624 CHAPTER 14 Optical Devices

TEST YOUR UNDERSTANDING

TYU 14.1 (a) A photon flux with an intensity of I0  0.10 W/cm2 is incident on the sur-
face of silicon. The wavelength of the incident photon signal is   1 m. Ne-
glecting any reflection from the surface, determine the photon flux intensity at a
depth of (i) x  5 m and (ii) x  20 m from the surface. (b) Repeat part (a)
for a wavelength of   0.60 m.
(ii) 3.35 105 W/cm2]
[Ans. (a) (i) 0.0951 W/cm2, (ii) 0.0819 W/cm2; (b) (i) 0.0135 W/cm2,

14.2 | SOLAR CELLS

A solar cell is a pn junction device with no voltage directly applied across the junc-
tion. The solar cell converts photon power into electrical power and delivers this
power to a load. These devices have long been used for the power supply of satellites
and space vehicles, and also as the power supply to some calculators. We will first
consider the simple pn junction solar cell with uniform generation of excess carriers.
We will also discuss briefly the heterojunction and amorphous silicon solar cells.

14.2.1 The pn Junction Solar Cell

Consider the pn junction shown in Figure 14.6 with a resistive load. Even with zero
bias applied to the junction, an electric field exists in the space charge region as
shown in the figure. Incident photon illumination can create electron–hole pairs in
the space charge region that will be swept out producing the photocurrent IL in the
reverse-biased direction as shown.
The photocurrent IL produces a voltage drop across the resistive load which forward
biases the pn junction. The forward-bias voltage produces a forward-bias current IF as
indicated in the figure. The net pn junction current, in the reverse-biased direction, is

I  IL  IF  IL  IS exp _
kT   
eV  1 (14.7)

h

E-field
p n
IL

IF

I
 V

Figure 14.6 | A pn junction solar cell with resistive load.

nea29583_ch14_618-669.indd 624 12/11/10 12:51 PM

 14.2 Solar Cells 625

Isc

0 Voc
V

Figure 14.7 | I–V characteristics of a

pn junction solar cell.

where the ideal diode equation has been used. As the diode becomes forward biased,
the magnitude of the electric field in the space charge region decreases, but does
not go to zero or change direction. The photocurrent is always in the reverse-biased
direction and the net solar cell current is also always in the reverse-biased direction.
There are two limiting cases of interest. The short-circuit condition occurs when
R  0 so that V  0. The current in this case is referred to as the short-circuit cur-
rent, or
I  Isc  IL (14.8)
The second limiting case is the open-circuit condition and occurs when R → . The
net current is zero and the voltage produced is the open-circuit voltage. The photo-
current is just balanced by the forward-biased junction current, so we have

   
eVoc
I  0  ILIS exp _
kT
1 (14.9)

We can find the open circuit voltage Voc as

 I
Voc  Vt ln 1  _L
IS  (14.10)
A plot of the diode current I as a function of the diode voltage V from Equa-
tion (14.7) is shown in Figure 14.7. We may note the short-circuit current and open-
circuit voltage points on the figure.

Objective: Calculate the open-circuit voltage of a silicon pn junction solar cell. EXAMPLE 14.3
Consider a silicon pn junction at T  300 K with the following parameters:
Na  5 1018 cm3 Nd  1016 cm3
Dn  25 cm2/s Dp  10 cm2/s
n0  5 107 s p0  107 s
Let the photocurrent density be JL  ILA  15 mA/cm2.

nea29583_ch14_618-669.indd 625 12/11/10 12:51 PM

 626 CHAPTER 14 Optical Devices

■ Solution
We have that
I

eDnnp0 eDppn0
JS  _S  __  __  en2i _
A Ln Lp
Dn
_
Dp

LnNa LpNd  
We may calculate
_____ _____________
Ln   Dnn0  (25)(5 107)  35.4 m
and
_____ _________
Lp  Dpp0  (10)(107)  10.0 m

Then

JS  (1.6 1019)(1.5 1010)2 ____ 25  ___ 10

(35.4 104)(5 1018) (10 104)(1016) 
 3.6 1011 A/cm2
Then from Equation (14.10), we can find

 I
 
J
 
15 103  0.514 V
Voc  Vt ln 1  _L  Vt ln 1  _L  (0.0259) ln 1  __
IS JS 3.6 1011 
■ Comment
We may determine the built-in potential barrier of this junction to be Vbi  0.8556 V. Taking
the ratio of the open-circuit voltage to the built-in potential barrier, we find that VocVbi  0.60.
The open-circuit voltage will always be less than the built-in potential barrier.

■ EXERCISE PROBLEM
Ex 14.3 Consider a GaAs pn junction solar cell with the following parameters:
Na  1017 cm3, Nd  2 1016 cm3, Dn  190 cm2 /s, Dp  10 cm2 /s, n0  107 s,
and p0  108 s. Assume a photocurrent density of JL  20 mA /cm2 is generated
in the solar cell. (a) Calculate the open-circuit voltage and (b) determine the ratio
of open-circuit voltage to built-in potential barrier.
[Ans. (a) Voc  0.971 V; (b) Voc Vbi  0.783]

The power delivered to the load is

P  I ⴢ V  IL ⴢ V  IS exp _
kT    
eV 1 ⴢ V (14.11)

We may find the current and voltage which will deliver the maximum power to the load
by setting the derivative equal to zero, or dPdV  0. Using Equation (14.11), we find
_
dV L S
eVm
dP  0  I I exp _
  kT   eVm
e exp _
1  ISVm _  kT   kT  (14.12)

where Vm is the voltage that produces the maximum power. We may rewrite Equa-
tion (14.12) in the form
1_ 
Vm
Vt
exp _
eVm
kT   I
 1  _L
IS
(14.13)

The value of Vm may be determined by trial and error. Figure 14.8 shows the maxi-
mum power rectangle where Im is the current when V  Vm.

nea29583_ch14_618-669.indd 626 12/11/10 12:51 PM

 14.2 Solar Cells 627

Isc

Im

0 Vm Voc
V

Figure 14.8 | Maximum power rectangle

of the solar cell I–V characteristics.

14.2.2 Conversion Efficiency and Solar Concentration

The conversion efficiency of a solar cell is defined as the ratio of output electrical
power to incident optical power. For the maximum power output, we can write
Pm ImVm
_ 100%  _ 100% (14.14)
Pin Pin
The maximum possible current and the maximum possible voltage in the solar cell
are Isc and Voc, respectively. The ratio ImVmIscVoc is called the fill factor and is a mea-
sure of the realizable power from a solar cell. Typically, the fill factor is between 0.7
and 0.8.
The conventional pn junction solar cell has a single semiconductor bandgap
energy. When the cell is exposed to the solar spectrum, a photon with energy less
than Eg will have no effect on the electrical output power of the solar cell. A photon
with energy greater than Eg will contribute to the solar cell output power, but the
fraction of photon energy that is greater than Eg will eventually only be dissipated
as heat. Figure 14.9 shows the solar spectral irradiance (power per unit area per unit
wavelength) where air mass zero represents the solar spectrum outside the earth’s
atmosphere and air mass one is the solar spectrum at the earth’s surface at noon.
The maximum efficiency of a silicon pn junction solar cell is approximately 28 per-
cent. Nonideal factors, such as series resistance and reflection from the semicon-
ductor surface, will lower the conversion efficiency typically to the range of 10 to
15 percent.
A large optical lens can be used to concentrate sunlight onto a solar cell so that
the light intensity can be increased up to several hundred times. The short-circuit
current increases linearly with light concentration while the open-circuit voltage in-
creases only slightly with concentration. Figure 14.10 shows the ideal solar cell effi-
ciency at 300 K for two values of solar concentration. We can see that the conversion
efficiency increases only slightly with optical concentration. The primary advantage
of using concentration techniques is to reduce the overall system cost since an optical
lens is less expensive than an equivalent area of solar cells.

nea29583_ch14_618-669.indd 627 12/11/10 12:51 PM

 628 CHAPTER 14 Optical Devices

50
AM 1
300 K
Si
40 GaAs
Ge

Efficiency (%)
30
2400 C  1000
Air mass zero, 1353 W/m2
Spectral irradiance (W/m2-m)

20
Air mass one, 925 W/m2 CdS
1600 C1
GaAs (c  0.87 m)
10

Si (c  1.1 m)

800
0
0 1 2 3
Visible Eg (eV )
0 Figure 14.10 | Ideal solar cell efficiency
0.2 0.8 1.4 2.0 2.6
Wavelength (m) at T  300 K for C  1 sun and for a
C  1000 sun concentrations as a
Figure 14.9 | Solar spectral irradiance. function of bandgap energy.
(From Sze [18].) (From Sze [18].)

14.2.3 Nonuniform Absorption Effects

We have seen from the previous section that the photon absorption coefficient in a
semiconductor is a very strong function of the incident photon energy or wavelength.
Figure 14.4 shows the absorption coefficient as a function of wavelength for several
semiconductor materials. As the absorption coefficient increases, more photon en-
ergy will be absorbed near the surface than deeper into the semiconductor. In this
case, then, we will not have uniform excess carrier generation in a solar cell.
The number of photons absorbed per cm3 per second as a function of distance x
from the surface can be written as
 0ex (14.15)
where 0 is the incident photon flux (cm2 s1) on the surface of the semiconductor.
We can also take into account the reflection of photons from the surface. Let R()
be the fraction of photons that are reflected. (For bare silicon, R  35 percent.) If we
assume that each photon absorbed creates one electron–hole pair, then the generation
rate of electron–hole pairs as a function of distance x from the surface is
GL  () 0()[1R()]e()x (14.16)
where each parameter may be a function of the incident wavelength. Figure 14.11
shows the excess minority carrier concentrations in this pn solar cell for two values
of wavelength and for the case when s  0 at the surface.

nea29583_ch14_618-669.indd 628 12/11/10 12:51 PM

 14.2 Solar Cells 629

45
n p
40
35

np1 or pn1 (relative scale)

30
25
  3 102 (  900 nm)
20   3 103 (  650 nm)
15
10 np1
5
pn1

x  0 xj  2 x
 0 16 32 48 64 80
x (m) x
(m)

Figure 14.11 | Steady-state, photon-induced normalized

minority carrier concentration in the pn junction solar cell
for two values of incident photon wavelength (xj  2 m,
W  1 m, Lp  Ln  40 m).

Ec
Ec
eVbip
eVbiN
Egp

EF
h Ev
EgN
Ev xp

xN

Figure 14.12 | The energy-band diagram of a pN

heterojunction in thermal equilibrium.

14.2.4 The Heterojunction Solar Cell

As we have mentioned in previous chapters, a heterojunction is formed between
two semiconductors with different bandgap energies. A typical pN heterojunction
energy-band diagram in thermal equilibrium is shown in Figure 14.12. Assume that
photons are incident on the wide-bandgap material. Photons with energy less than
EgN will pass through the wide-bandgap material, which acts as an optical window,
and photons with energies greater than Egp will be absorbed in the narrow bandgap
material. On the average, excess carriers created in the depletion region and within a
diffusion length of the junction will be collected and will contribute to the photocur-
rent. Photons with an energy greater than EgN will be absorbed in the wide-bandgap

nea29583_ch14_618-669.indd 629 12/11/10 12:51 PM

 630 CHAPTER 14 Optical Devices

Ec
h Eg1
1.4 p AlxGa1x As Eg2
p GaAs Voc
Ev
1.2
n GaAs

Normalized spectral response

1.0

0.8

0.6

x  0.23 x  0.45 x  0.86

0.4
x  0.73

0.2

0
1.4 1.6 1.8 2.0 2.2 2.4 2.6
h (eV)

Figure 14.13 | The normalized spectral response of several

AlGaAs/GaAs solar cells with different compositions.
(From Sze [17].)

material, and excess carriers generated within one diffusion length of the junction
will be collected. If EgN is large enough, then the high-energy photons will be ab-
sorbed in the space charge region of the narrow-bandgap material. This heterojunc-
tion solar cell should have better characteristics than a homojunction cell, especially
at the shorter wavelengths.
A variation of the heterojunction is shown in Figure 14.13. A pn homojunc-
tion is formed and then a wide-bandgap material is grown on top. Again, the wide-
bandgap material acts as an optical window for photon energies h  Eg1. Photons
with energies Eg2  h  Eg1 will create excess carriers in the homojunction and pho-
tons with energies h  Eg1 will create excess carriers in the window type material.
If the absorption coefficient in the narrow bandgap material is high, then essentially
all of the excess carriers will be generated within a diffusion length of the junction,
so the collection efficiency will be very high. Figure 14.13 also shows the normalized
spectral response for various mole fractions x in the AlxGa1x As.

14.2.5 Amorphous Silicon Solar Cells

Single-crystal silicon solar cells tend to be expensive and are limited to approxi-
mately 6 inches in diameter. A system powered by solar cells requires, in general,

nea29583_ch14_618-669.indd 630 12/11/10 12:51 PM

 14.2 Solar Cells 631

a very large area solar cell array to generate the required power. Amorphous silicon
solar cells provide the possibility of fabricating large area and relatively inexpensive
solar cell systems.
When silicon is deposited by CVD techniques at temperatures below 600 C, an
amorphous film is formed regardless of the type of substrate. In amorphous silicon,
there is only very short range order, and no crystalline regions are observed. Hy-
drogen may be incorporated in the silicon to reduce the number of dangling bonds,
creating a material called hydrogenated amorphous silicon.
The density of states versus energy for amorphous silicon is shown in Fig-
ure 14.14. Amorphous silicon contains large numbers of electronic energy states
within the normal bandgap of single-crystal silicon. However, because of the short-
range order, the effective mobility is quite small, typically in the range between 106
and 103 cm2/V-s. The mobilities in the states above Ec and below E are between
1 and 10 cm2/V-s. Consequently, conduction through the energy states between
Ec and E is negligible because of the low mobility. Because of the difference in
mobility values, Ec and E are referred to as the mobility edges and the energy be-
tween Ec and E is referred to as the mobility gap. The mobility gap can be modified
by adding specific types of impurities. Typically, the mobility gap is on the order of
1.7 eV.
Amorphous silicon has a very high optical absorption coefficient, so most sun-
light is absorbed within approximately 1 m of the surface. Consequently, only a

E
Conduction
band

Ec

Band-gap
states Mobility
edges

Ev

Valence
band

Density of states N(E )

Figure 14.14 | Density of states versus

energy of amorphous silicon.
(From Yang [22].)

nea29583_ch14_618-669.indd 631 12/11/10 12:51 PM

 632 CHAPTER 14 Optical Devices

Ec
Indium tin oxide
EF Ev

Glass
h n i p

(a) (b)

Ec

EFn
Voc
Ev
h
EFp

(c)

Figure 14.15 | The (a) cross section, (b) energy-band diagram at thermal equilibrium, and
(c) energy-band diagram under photon illumination of an amorphous silicon PIN solar cell.
(From Yang [22].)

very thin layer of amorphous silicon is required for a solar cell. A typical amorphous
silicon solar cell is a PIN device shown in Figure 14.15. The amorphous silicon
is deposited on an optically transparent indium tin oxide–coated glass substrate. If
aluminum is used as the back contact, it will reflect any transmitted photons back
through the PIN device. The n and p regions can be quite thin while the intrinsic
region may be in the range of 0.5 to 1.0 m thick. The energy-band diagram for the
thermal equilibrium case is shown in the figure. Excess carriers generated in the
intrinsic region are separated by the electric field and produce the photocurrent, as
we have discussed. Conversion efficiencies are smaller than in single-crystal silicon,
but the reduced cost makes this technology attractive. Amorphous silicon solar cells
approximately 40 cm wide and many meters long have been fabricated.

TEST YOUR UNDERSTANDING

TYU 14.2 Consider a silicon pn junction solar cell with the parameters given in Example 14.3.
Determine the required photocurrent density to produce an open-circuit voltage of
Voc  0.60 V.
(Ans. JL  0.414 A/cm2)
TYU 14.3 Consider the silicon pn junction solar cell described in Example 14.3. Let the
solar intensity increase by a factor of 10. Calculate the open-circuit voltage.
(Ans. Voc  0.574 V)
TYU 14.4 The silicon pn junction solar cell described in TYU 14.2 has a cross-sectional
area of 1 cm2. Determine the maximum power that can be delivered to a load.
(Ans. 0.205 W)

nea29583_ch14_618-669.indd 632 12/11/10 12:51 PM

 14.3 Photodetectors 633

14.3 | PHOTODETECTORS
There are several semiconductor devices that can be used to detect the presence of
photons. These devices are known as photodetectors; they convert optical signals
into electrical signals. When excess electrons and holes are generated in a semi-
conductor, there is an increase in the conductivity of the material. This change
in conductivity is the basis of the photoconductor, perhaps the simplest type of
photodetector. If electrons and holes are generated within the space charge region
of a pn junction, then they will be separated by the electric field and a current will
be produced. The pn junction is the basis of several photodetector devices includ-
ing the photodiode and the phototransistor.

14.3.1 Photoconductor
Figure 14.16 shows a bar of semiconductor material with ohmic contacts at each
end and a voltage applied between the terminals. The initial thermal-equilibrium
conductivity is

␴0 ⫽ e(␮n n0 ⫹ ␮p p0) (14.17)

If excess carriers are generated in the semiconductor, the conductivity becomes

␴ ⫽ e[␮n(n0 ⫹ ␦n) ⫹ ␮p(p0 ⫹ ␦p)] (14.18)

where ␦n and ␦p are the excess electron and hole concentrations, respectively. If we
consider an n-type semiconductor, then, from charge neutrality, we can assume that

I
⫹V

h␯

Area ⫽ A

Figure 14.16 | A photoconductor.

nea29583_ch14_618-669.indd 633 12/13/10 1:51 PM

 634 CHAPTER 14 Optical Devices

n  p
p. We will use p as the concentration of excess carriers. In steady state,
the excess carrier concentration is given by p  GLp, where GL is the generation
rate of excess carriers (cm3-s1) and p is the excess minority carrier lifetime.
The conductivity from Equation (14.18) can be rewritten as

 e(n n0  p p0)  e(p)(n  p) (14.19)

The change in conductivity due to the optical excitation, known as the photoconduc-
tivity, is then


 e(p)(n  p) (14.20)

An electric field is induced in the semiconductor by the applied voltage, which

produces a current. The current density can be written as
J  (J0  JL)  (
0  
)E (14.21)
where J0 is the current density in the semiconductor prior to optical excitation and JL
is the photocurrent density. The photocurrent density is JL  
ⴢ E. If the excess
electrons and holes are generated uniformly throughout the semiconductor, then the
photocurrent is given by
IL  JL ⴢ A  
ⴢ AE  eGLp(n  p)AE (14.22)
where A is the cross-sectional area of the device. The photocurrent is directly pro-
portional to the excess carrier generation rate, which in turn is proportional to the
incident photon flux.
If excess electrons and holes are not generated uniformly throughout the semi-
conductor material, then the total photocurrent is found by integrating the photocon-
ductivity over the cross-sectional area.
Since n E is the electron drift velocity, the electron transit time, that is, the time
required for an electron to flow through the photoconductor, is

tn  _L (14.23)
n E
The photocurrent, from Equation (14.22), can be rewritten as
p 
IL  eGL _  1  _
tn   AL
p
n
(14.24)

We may define a photoconductor gain, ph, as the ratio of the rate at which
charge is collected by the contacts to the rate at which charge is generated within the
photoconductor. We can write the gain as
IL
ph  __ (14.25)
eGL AL
which, using Equation (14.24), can be written
p p
ph  _ 
tn 1  n
_
 (14.26)

nea29583_ch14_618-669.indd 634 12/11/10 12:51 PM

 14.3 Photodetectors 635

Objective: Calculate the photoconductor gain of a silicon photoconductor. EXAMPLE 14.4

Consider an n-type silicon photoconductor with a length L  100 m, cross-sectional area
A  107 cm2, and minority carrier lifetime p  106 s. Let the applied voltage be V  10 volts.

■ Solution
The electron transit time is determined as
L _ (100 104)2
L2  ___
tn  _  7.41 109 s
n E nV (1350)(10)
The photoconductor gain is then
p p
ph  _   106
 480

tn 1  n  7.41 109 1  1350  1.83 10
_ __ _ 2

■ Comment
The fact that a photoconductor—a bar of semiconductor material—has a gain may be
surprising.

■ EXERCISE PROBLEM
Ex 14.4 Consider the photoconductor described in Example 14.4. Determine the photocur-
rent if GL  1021 cm3 s1 and E  10 V /cm. Also assume that n  1000 cm2 /V-s
and p  400 cm2 /V-s.
(Ans. IL  0.224 A)

Let’s consider physically what happens to a photon-generated electron, for ex-

ample. After the excess electron is generated, it drifts very quickly out of the photo-
conductor at the anode terminal. In order to maintain charge neutrality throughout
the photoconductor, another electron immediately enters the photoconductor at the
cathode and drifts toward the anode. This process will continue during a time period
equal to the mean carrier lifetime. At the end of this period, on the average, the
photoelectron will recombine with a hole.
The electron transit time, using the parameters from Example 14.4, is tn 
7.41 109 s. In a simplistic sense, the photoelectron will circulate around the photo-
conductor circuit 135 times during the 106 s time duration, which is the mean car-
rier lifetime. If we take into account the photon-generated hole, the total number of
charges collected at the photoconductor contacts for every electron generated is 183.
When the optical signal ends, the photocurrent will decay exponentially with a
time constant equal to the minority carrier lifetime. From the photoconductor gain
expression, we would like a large minority carrier lifetime, but the switching speed is
enhanced by a small minority carrier lifetime. There is obviously a trade-off between
gain and speed. In general, the performance of a photodiode, which we will discuss
next, is superior to that of a photoconductor.

14.3.2 Photodiode
A photodiode is a pn junction diode operated with an applied reverse-biased volt-
age. We will initially consider a long diode in which excess carriers are generated

nea29583_ch14_618-669.indd 635 12/11/10 12:51 PM

 636 CHAPTER 14 Optical Devices

VR
 

n p

(a)

W
n p
E-field

pn0 np0

x
 0 x0
x
x
(b)

Figure 14.17 | (a) A reverse-biased pn junction. (b) Minority

carrier concentration in the reverse-biased pn junction.

uniformly throughout the semiconductor device. Figure 14.17a shows the reverse-
biased diode and Figure 14.17b shows the minority carrier distribution in the reverse-
biased junction prior to photon illumination.
Let G L be the generation rate of excess carriers. The excess carriers generated
within the space charge region are swept out of the depletion region very quickly
by the electric field; the electrons are swept into the n region and the holes into
the p region. The photon-generated current density from the space charge region is
given by

JL1  e GL dx (14.27)

where the integral is over the space charge region width. If GL is constant throughout
the space charge volume, then
JL1  eGLW (14.28)
where W is the space charge width. We may note that JL1 is in the reverse-biased
direction through the pn junction. This component of photocurrent responds very
quickly to the photon illumination and is known as the prompt photocurrent.
We may note, by comparing Equations (14.28) and (14.25), that the photodiode
gain is unity. The speed of the photodiode is limited by the carrier transport through

nea29583_ch14_618-669.indd 636 12/11/10 12:51 PM

 14.3 Photodetectors 637

the space charge region. If we assume that the saturation drift velocity is 107 cm/s
and the depletion width is 2 m, the transit time is t  20 ps. The ideal modulat-
ing frequency has a period of 2t, so the frequency is f  25 GHz. This frequency
response is substantially higher than that of photoconductors.
Excess carriers are also generated within the neutral n and p regions of the diode.
The excess minority carrier electron distribution in the p region is found from the
ambipolar transport equation, which is
2(np) np __ (np)
Dn __  GL  _ n0  t (14.29)
x2
We will assume that the E-field is zero in the neutral regions. In steady state,
(np)t  0, so that Equation (14.29) can be written as
2
d (np) _np G
__    _L (14.30)
dx2 L2n Dn
where L2n  Dnn0.
The solution to Equation (14.30) can be found as the sum of the homogeneous
and particular solutions. The homogeneous solution is found from the equation
2
d (nph) _nph
__  0 (14.31)
dx2 L2n
where nph is the homogeneous solution and is given by
nph  AexLn  BexLn (x  0) (14.32)
One boundary condition is that nph must remain finite, which implies that B
0 for
the “long” diode.
The particular solution is found from
npp G
_   _L (14.33)
L2n Dn
which yields
GLL2 GL(Dnn0)
npp  _n  __  GLn0 (14.34)
Dn Dn
The total steady-state solution for the excess minority carrier electron concentra-
tion in the p region is then
np  AexLn  GLn0 (14.35)
The total electron concentration is zero at x  0 for the reverse-biased junction. The
excess electron concentration x  0 is then
np(x  0)  np0 (14.36)
Using the boundary condition from Equation (14.36), the electron concentration
given by Equation (14.35) becomes
np  GLn0  (GLn0  np0)exLn (14.37)

nea29583_ch14_618-669.indd 637 12/11/10 12:51 PM

 638 CHAPTER 14 Optical Devices

h

E-field
n p


Jp1 JL1
Jn1
np  GLp0 np  GLn0

np(x
) np(x)

x
 0 x0
x
x
JL

Figure 14.18 | Steady-state, photoinduced minority carrier

concentrations and photocurrents in a “long” reverse-biased
pn junction.

We can find the excess minority carrier hole concentration in the n region using the
same type of analysis. Using the x
notation shown in Figure 14.17, we can write
pn  GLp0  (GLp0  pn0)ex
L p
(14.38)
Equations (14.37) and (14.38) are plotted in Figure 14.18. We may note that the steady-
state values far from the space charge region are the same as were given previously.
The gradient in the minority carrier concentrations will produce diffusion cur-
rents in the pn junction. The diffusion current density at x  0 due to minority carrier
electrons is
Jn1  eDn __
d(np)

dx x  0
d [G   (G   n )exLn]
 eDn _
dx L n0 L n0 p0
x0 
eD (14.39)
 _ n
(GLn0  np0)
Ln
Equation (14.39) can be written as
eDn np0
Jn1  eGL Ln  __ (14.40)
Ln
The first term in Equation (14.40) is the steady-state photocurrent density while the
second term is the ideal reverse saturation current density due to the minority carrier
electrons.
The diffusion current density (in the x direction) at x
 0 due to the minority
carrier holes is
eDp pn0
Jp1  eGL Lp  __ (14.41)
Lp

nea29583_ch14_618-669.indd 638 12/11/10 12:51 PM

 14.3 Photodetectors 639

Similarly, the first term is the steady-state photocurrent density and the second term
is the ideal reverse saturation current density.
The total steady-state diode photocurrent density for the long diode is now

JL  eGLW  eGL Ln  eGL Lp  e(W  Ln  Lp)GL (14.42)

Again note that the photocurrent is in the reverse-biased direction through the diode.
The photocurrent given by Equation (14.42) is the result of assuming uniform gen-
eration of excess carriers throughout the structure, a long diode, and steady state.
The time response of the diffusion components of the photocurrent is relatively
slow, since these currents are the results of the diffusion of minority carriers toward
the depletion region. The diffusion components of photocurrent are referred to as the
delayed photocurrent.

Objective: Calculate the steady-state photocurrent density in a reverse-biased, long pn diode. EXAMPLE 14.5
Consider a silicon pn diode at T  300 K with the following parameters:

Na  1016 cm3 Nd  1016 cm3

Dn  25 cm2/s Dp  10 cm2/s
n0  5 107 s p0  107 s

Assume that a reverse-biased voltage of VR  5 volts is applied and let GL  1021 cm3-s1.

■ Solution
We may calculate various parameters as follows:
_____ _____________
Ln   Dnn0  (25)(5 107)  35.4 m
_____ _________
Lp   Dpp0  (10)(107)  10.0 m

Vbi  Vt ln _
ni2 
Na Nd
 (0.0259) ln ___ 
(1016)(1016)
(1.5 1010)2 
 0.695 V

 
Na  Nd

2s __ 12
W _e (Vbi  VR)
NaNd


2(11.7)(8.85 1014) __(2 1016)

12
 ____ ⴢ ⴢ (0.695  5)  1.21 m
1.6 1019 (1016)(1016)

Finally, the steady-state photocurrent density is

JL  e(W  Ln  Lp)GL
 (1.6 1019)(1.21  35.4  10.0) 104(1021)  0.75 A/cm2

■ Comment
Again, keep in mind that this photocurrent is in the reverse-biased direction through the diode
and is many orders of magnitude larger than the reverse-biased saturation current density in
the pn junction diode.

nea29583_ch14_618-669.indd 639 12/11/10 12:51 PM

 640 CHAPTER 14 Optical Devices

■ EXERCISE PROBLEM
Ex 14.5 The doping concentrations of the photodiode described in Example 14.5 are
changed to Na  Nd  1015 cm3. (a) Determine the steady-state photocurrent den-
sity. (b) Calculate the ratio of prompt photocurrent to steady-state photocurrent.
[Ans. (a) JL  0.787 A /cm2; (b) JL1JL  0.0773]

In this example calculation, Ln  W and Lp  W. In many pn junction structures,

the assumption of a long diode will not be valid, so the photocurrent expression will
have to be modified. In addition, the photon energy absorption may not be uniform
throughout the pn structure. The effect of nonuniform absorption will be considered
in the next section.

14.3.3 PIN Photodiode

In many photodetector applications, the speed of response is important; therefore, the
prompt photocurrent generated in the space charge region is the only photocurrent of
interest. To increase the photodetector sensitivity, the depletion region width should
be made as large as possible. This can be achieved in a PIN photodiode.
The PIN diode consists of a p region and an n region separated by an intrinsic
region. A sketch of a PIN diode is shown in Figure 14.19a. The intrinsic region width
W is much larger than the space charge width of a normal pn junction. If a reverse
bias is applied to the PIN diode, the space charge region extends completely through
the intrinsic region.

VR
 

p i n

Wp W Wn

(a)

p i n
0 (x)

x0 xW
(b)

Figure 14.19 | (a) A reverse-biased PIN

photodiode. (b) Geometry showing
nonuniform photon absorption.

nea29583_ch14_618-669.indd 640 12/11/10 12:51 PM

 14.3 Photodetectors 641

Assume that a photon flux 0 is incident on the p region. If we assume that the

p region width Wp is very thin, then the photon flux, as a function of distance, in the
intrinsic region is (x)  0ex, where  is the photon absorption coefficient. This
nonlinear photon absorption is shown in Figure 14.19b. The photocurrent density
generated in the intrinsic region can be found as
 W  W
JL  e GL dx  e 0ex dx  e 0(1eW ) (14.43)
0 0

This equation assumes that there is no electron–hole recombination within the space
charge region and also that each photon absorbed creates one electron–hole pair.

Objective: Calculate the photocurrent density in a PIN photodiode. EXAMPLE 14.6

Consider a silicon PIN diode with an intrinsic region width of W  20 m. Assume that
the photon flux is 1017 cm2-s1 and the absorption coefficient is   103 cm1.

■ Solution
The generation rate of electron–hole pairs at the front edge of the intrinsic region is

GL1   0  (103)(1017)  1020 cm3-s1

and the generation rate at the back edge of the intrinsic region is

GL2   0eW  (103)(1017)exp [(103)(20 104)]

 0.135 1020 cm3-s1

The generation rate is obviously not uniform throughout the intrinsic region. The photocurrent
density is then

JL  e 0(1eW )
 (1.6 1019)(1017){1exp [(103)(20 104)]}
 13.8 mA/cm2
■ Comment
The prompt photocurrent density of a PIN photodiode will be larger than that of a regular
photodiode since the space charge region is larger in a PIN photodiode.

■ EXERCISE PROBLEM
Ex 14.6 Repeat Example 14.6 for photon absorption coefficients of (a)   102 cm1 and
(b)   104 cm1.
[Ans. (a) JL  2.90 mA /cm2; (b) JL  16.0 mA /cm2]

In most situations, we will not have a long diode; thus, the steady-state photo-
current described by Equation (14.42) will not apply for most photodiodes.

14.3.4 Avalanche Photodiode

The avalanche photodiode is similar to the pn or PIN photodiode except that the bias
applied to the avalanche photodiode is sufficiently large to cause impact ionization.

nea29583_ch14_618-669.indd 641 12/11/10 12:51 PM

 642 CHAPTER 14 Optical Devices

Electron–hole pairs are generated in the space charge region by photon absorption,
as we have discussed previously. The photon-generated electrons and holes now
generate additional electron–hole pairs through impact ionization. The avalanche
photodiode now has a current gain introduced by the avalanche multiplication factor.
The electron–hole pairs generated by photon absorption and by impact ioniza-
tion are swept out of the space charge region very quickly. If the saturation velocity
is 107 cm/s in a depletion region that is 10 m wide, then the transit time is
7
t  __ 10  100 ps
10 104
The period of a modulation signal would be 2t, so that the frequency would be

f_ 1  __ 1  5 GHz
2t 200 1012
If the avalanche photodiode current gain is 20, then the gain-bandwidth product is
100 GHz. The avalanche photodiode could respond to light waves modulated at
microwave frequencies.

14.3.5 Phototransistor
A bipolar transistor can also be used as a photodetector. The phototransistor can have
high gain through the transistor action. An npn bipolar phototransistor is shown in
Figure 14.20a. This device has a large base–collector junction area and is usually
operated with the base open circuited. Figure 14.20b shows the block diagram of the
phototransistor. Electrons and holes generated in the reverse-biased B–C junction are
swept out of the space charge region, producing a photocurrent IL. Holes are swept

h␯

Base Emitter

p n

Collector
(a)

n p n
␣IE C
E
IE VCE
IL

B
(b)

Figure 14.20 | (a) A bipolar phototransistor. (b) Block

diagram of the open-base phototransistor.

nea29583_ch14_618-669.indd 642 12/11/10 12:51 PM

 14.4 Photoluminescence and Electroluminescence 643

into the p-type base, making the base positive with respect to the emitter. Since the
B–E becomes forward-biased, electrons will be injected from the emitter back into
the base, leading to the normal transistor action.
From Figure 14.20b, we see that
IE  IE  IL (14.44)
where IL is the photon-generated current and  is the common base current gain. Since
the base is an open circuit, we have IC  IE, so Equation (14.44) can be written as
IC  IC  IL (14.45)
Solving for IC, we find
IL
IC  __ (14.46)
1
Relating  to , the dc common emitter current gain, Equation (14.46) becomes
IC  (1  )IL (14.47)
Equation (14.47) shows that the basic B–C photocurrent is multiplied by the factor
(1  ). The phototransistor, then, amplifies the basic photocurrent.
With the relatively large B–C junction area, the frequency response of the photo-
transistor is limited by the B–C junction capacitance. Since the base is essentially the
input to the device, the large B–C capacitance is multiplied by the Miller effect, so
the frequency response of the phototransistor is further reduced. The phototransistor,
however, is a lower-noise device than the avalanche photodiode.
Phototransistors can also be fabricated in heterostructures. The injection
efficiency is increased as a result of the bandgap differences, as we discussed in
Chapter 12. With the bandgap difference, the lightly doped base restriction no longer
applies. A fairly heavily doped, narrow-base device can be fabricated with a high
blocking voltage and a high gain.

TEST YOUR UNDERSTANDING

TYU 14.5 Consider a long silicon pn junction photodiode with the parameters given in
Example 14.5. The cross-sectional area is A  103 cm2. Assume the photodiode
is reverse biased by a 5-volt battery in series with a 5 k load resistor. An opti-
cal signal at a wavelength of   1 m is incident on the photodiode producing
a uniform generation rate of excess carriers throughout the entire device. Deter-
mine the incident intensity such that the voltage across the load resistor is 0.5 V.
(Ans. Iv  0.266 W/cm2)

14.4 | PHOTOLUMINESCENCE AND

ELECTROLUMINESCENCE
In the first section of this chapter, we have discussed the creation of excess electron–
hole pairs by photon absorption. Eventually, excess electrons and holes recombine,
and in direct bandgap materials the recombination process may result in the emission
of a photon. The general property of light emission is referred to as luminescence.

nea29583_ch14_618-669.indd 643 12/11/10 12:51 PM

 644 CHAPTER 14 Optical Devices

When excess electrons and holes are created by photon absorption, photon emission
from the recombination process is called photoluminescence.
Electroluminescence is the process of generating photon emission when the
excitation of excess carriers is a result of an electric current caused by an applied
electric field. We are mainly concerned here with injection electroluminescence, the
result of injecting carriers across a pn junction. The light emitting diode and the pn
junction laser diode are examples of this phenomenon. In these devices, electric en-
ergy, in the form of a current, is converted directly into photon energy.

14.4.1 Basic Transitions

Once electron–hole pairs are formed, there are several possible processes by which
the electrons and holes can recombine. Some recombination processes may result in
photon emission from direct bandgap materials, whereas other recombination pro-
cesses in the same material may not.
Figure 14.21a shows the basic interband transitions. Curve (i) corresponds
to an intrinsic emission very close to the bandgap energy of the material. Curves
(ii) and (iii) correspond to energetic electrons or holes. If either of these recombina-
tions result in the emission of a photon, the energy of the emitted photon will be
slightly larger than the bandgap energy. There will then be an emission spectrum and
a bandwidth associated with the emission.
The possible recombination processes involving impurity or defect states are
shown in Figure 14.21b. Curve (i) is the conduction band to acceptor transition,
curve (ii) is the donor to valence-band transition, curve (iii) is the donor to accep-
tor transition, and curve (iv) is the recombination due to a deep trap. Curve (iv)
is a nonradiative process corresponding to the Shockley–Read–Hall recombination
process discussed in Chapter 6. The other recombination processes may or may not
result in the emission of a photon.
Figure 14.21c shows the Auger recombination process, which can become im-
portant in direct bandgap materials with high doping concentrations. The Auger
recombination process is a nonradiative process. The Auger recombination, in one case,
shown in curve (i), is a recombination between an electron and hole, accompanied by
the transfer of energy to another free hole. Similarly, in the second case, the recombina-
tion between an electron and hole can result in the transfer of energy to a free electron
as shown in curve (ii). The third particle involved in this process will eventually lose its
energy to the lattice in the form of heat. The process involving two holes and an electron
would occur predominantly in heavily doped p-type materials, and the process involving
two electrons and a hole would occur primarily in a heavily doped n-type material.
The recombination processes shown in Figure 14.21a indicate that the emission
of a photon is not necessarily at a single, discrete energy, but can occur over a range
of energies. The spontaneous emission rate generally has the form


(h  Eg)
I()  2 (h  Eg)12 exp __
kT  (14.48)

where Eg is the bandgap energy. Figure 14.22 shows the emission spectra from gal-
lium arsenide. The peak photon energy decreases with temperature because the

nea29583_ch14_618-669.indd 644 12/11/10 12:51 PM

 14.4 Photoluminescence and Electroluminescence 645


 
Ec

Ev
 


(i) (ii) (iii)
(a)

  Ec Wavelength (m)
  E 1.0 0.95 0.9 0.85
d
100
Et 0.018 eV
0.036 eV

  Ea
Ev
 

Photon flux (relative)

(i) (ii) (iii) (iv) 101
(b)

295 K 77 K
  E
c
102

Resolution

  
Ev 103
1.24 1.28 1.32 1.36 1.40 1.44 1.48 1.52
Photon energy (eV)
(i) (ii)
(c) Figure 14.22 | GaAs diode emission
spectra at T  300 K and T  77 K.
Figure 14.21 | Basic transitions in a semiconductor. (From Sze and Ng [17].)

bandgap energy decreases with temperature. We will show that the bandwidth of the
emission spectra can be greatly reduced in a laser diode by using an optical resonator.

14.4.2 Luminescent Efficiency

We have shown that not all recombination processes are radiative. An efficient lu-
minescent material is one in which radiative transitions predominate. The quantum
efficiency is defined as the ratio of the radiative recombination rate to the total re-
combination rate for all processes. We can write
R
q  _r (14.49)
R
where q is the quantum efficiency, Rr is the radiative recombination rate, and R is
the total recombination rate of the excess carriers. Since the recombination rate is

nea29583_ch14_618-669.indd 645 12/11/10 12:51 PM

 646 CHAPTER 14 Optical Devices

inversely proportional to lifetime, we can write the quantum efficiency in terms of

lifetimes as
nr
q  __ (14.50)
nr  r
where nr is the nonradiative lifetime and r is the radiative lifetime. For a
high luminescent efficiency, the nonradiative lifetimes must be large; thus, the
probability of a nonradiative recombination is small compared to the radiative
recombination.
The interband recombination rate of electrons and holes will be directly propor-
tional to the number of electrons available and directly proportional to the number of
available empty states (holes). We can write
Rr  Bnp (14.51)
where Rr is the band-to-band radiative recombination rate and B is the constant of
proportionality. The values of B for direct-bandgap materials are on the order of
106 larger than for indirect bandgap materials. The probability of a direct band-
to-band radiative recombination transition in an indirect bandgap material is very
unlikely.
One problem encountered with the emission of photons from a direct bandgap
material is the reabsorption of the emitted photons. In general, the emitted photons
will have energies h  Eg, which means that the absorption coefficient is not zero for
this energy. In order to generate a light output from a light emitting device, the process
must take place near the surface. One possible solution to the reabsorption problem is
to use heterojunction devices. These are discussed in later sections.

14.4.3 Materials
An important direct bandgap semiconductor material for optical devices is gallium
arsenide. Another compound material that is of great interest is AlxGa1x As. This
material is a compound semiconductor in which the ratio of aluminum atoms to
gallium atoms can be varied to achieve specific characteristics. Figure 14.23 shows
the bandgap energy as a function of the mole fraction between aluminum and gal-
lium. We can note from the figure that for 0  x  0.45, the alloy material is a
direct bandgap material. For x  0.45, the material becomes an indirect bandgap
material, not suitable for optical devices. For 0  x  0.35, the bandgap energy can
be expressed as
Eg  1.424  1.247x eV (14.52)

Another compound semiconductor used for optical devices is the GaAs1xPx

system. Figure 14.24a shows the bandgap energy as a function of the mole frac-
tion x. For 0  x  0.45, this material is also a direct bandgap material, and for
x  0.45, the bandgap becomes indirect. Figure 14.24b is the E versus k diagram,
showing how the bandgap changes from direct to indirect as the mole fraction
changes.

nea29583_ch14_618-669.indd 646 12/11/10 12:51 PM

 14.4 Photoluminescence and Electroluminescence 647

3.0 Eg  3.018

2.5
2.168

Bandgap Eg (eV)
Eg  1.9
2.0
Indirect
band
Direct
1.5 band
Eg  1.424
1.0
AlxGa1x As
0.5 T  297 K

0 0.5 1.0
GaAs AlAs
Mole fraction AlAs, x

Figure 14.23 | Bandgap energy of

AlxGa1x As as a function of the mole
fraction x.
(From Sze [18].)

3.0
x  1.0
T  300 K 3
GaAs1x Px
0.85
0.65
2.5 0.4
2
Energy gap Eg (eV)

Eg 
2.261 0
Energy (eV)

Conduction
Indirect band
2.0 band 1
x  0.45
Eg  1.977 h␯

Direct
1.5 band 0
Valence
Eg  1.424 band

0 –p
max
0 0.2 0.4 0.6 0.8 1.0 Crystal momentum –p
GaAs GaP
Mole fraction GaP, x
(a) (b)

Figure 14.24 | (a) Bandgap energy of GaAs1x Px as a function of mole fraction x.

(b) E versus k diagram of GaAs1x Px for various values of x.
(From Sze [18].)

nea29583_ch14_618-669.indd 647 12/11/10 12:51 PM

 648 CHAPTER 14 Optical Devices

EXAMPLE 14.7 Objective: Determine the output wavelength of a GaAs1xPx material for two different mole
fractions.
Consider first GaAs and then GaAs1x Px.

■ Solution
GaAs has a bandgap energy of Eg  1.42 eV. This material would produce a photon output at
a wavelength of

1.24  _
_ 1.24  0.873 m
E 1.42
This wavelength is in the infrared range and not in the visible range. If we desire a visible
output with a wavelength of   0.653 m, for example, the bandgap energy would have to be

1.24  _
E_ 1.24  1.90 eV
 0.653
This bandgap energy would correspond to a mole fraction of approximately x  0.4.

■ Comment
By changing the mole fraction in the GaAs1xPx system, the output can change from the infra-
red to the red spectrum.

■ EXERCISE PROBLEM
Ex 14.7 Determine the output wavelength of a GaAs1xPx material for mole fractions of
(a) x  0.15 and (b) x  0.30.
[Ans. (a)   0.775 m; (b)   0.705 m]

14.5 | LIGHT EMITTING DIODES

Photodetectors and solar cells convert optical energy into electrical energy—the
photons generate excess electrons and holes, which produce an electric current. We
might also apply a voltage across a pn junction resulting in a diode current, which in
turn can produce photons and a light output. This inverse mechanism is called injec-
tion electroluminescence. This device is known as a Light Emitting Diode (LED).
The spectral output of an LED may have a relatively wide wavelength bandwidth of
between 30 and 40 nm. However, this emission spectrum is narrow enough so that a
particular color is observed, provided the output is in the visible range.

14.5.1 Generation of Light

As we have discussed previously, photons may be emitted if an electron and hole
recombine by a direct band-to-band recombination process in a direct bandgap mate-
rial. The emission wavelength, from Equation (14.1), is
hc  _
_ 1.24 m (14.53)
Eg Eg
where Eg is the bandgap energy measured in electron-volts.

nea29583_ch14_618-669.indd 648 12/11/10 12:51 PM

 14.5 Light Emitting Diodes 649

When a voltage is applied across a pn junction, electrons and holes are injected
across the space charge region where they become excess minority carriers. These
excess minority carriers diffuse into the neutral semiconductor regions where they
recombine with majority carriers. If this recombination process is a direct band-to-
band process, photons are emitted. The diode diffusion current is directly proportional
to the recombination rate, so the output photon intensity will also be proportional to
the ideal diode diffusion current. In gallium arsenide, electroluminescence originates
primarily on the p side of the junction because the efficiency for electron injection is
higher than that for hole injection.

14.5.2 Internal Quantum Efficiency

The internal quantum efficiency of an LED is the fraction of diode current that
produces luminescence. The internal quantum efficiency is a function of the injec-
tion efficiency and a function of the percentage of radiative recombination events
compared with the total number of recombination events.
The three current components in a forward-biased diode are the minority car-
rier electron diffusion current, the minority carrier hole diffusion current, and the
space charge recombination current. These current densities can be written, respec-
tively, as

   
eDnnp0 eV  1
Jn  __ exp _ (14.54a)
Ln kT

   
eDppn0 eV  1
Jp  __ exp _ (14.54b)
Lp kT
and
eniW
JR  _
20    
eV  1
exp _
2kT
(14.54c)

The recombination of electrons and holes within the space charge region is, in
general, through traps near midgap and is a nonradiative process. Since lumines-
cence is due primarily to the recombination of minority carrier electrons in GaAs, we
can define an injection efficiency as the fraction of electron current to total current.
Then
Jn
 ___ (14.55)
Jn  Jp  JR
where is the injection efficiency. We can make approach unity by using an np
diode so that Jp is a small fraction of the diode current and by forward biasing the
diode sufficiently so that JR is a small fraction of the total diode current.
Once the electrons are injected into the p region, not all electrons will recombine
radiatively. We can define the radiative and nonradiative recombination rates as
n
Rr  _ (14.56a)
r

nea29583_ch14_618-669.indd 649 12/11/10 12:51 PM

 650 CHAPTER 14 Optical Devices

and
n
Rnr  _ (14.56b)
nr
where r and nr are the radiative and nonradiative recombination lifetimes, respec-
tively, and n is the excess carrier concentration. The total recombination rate is
n  _
R  Rr  Rnr  _ n  _
n (14.57)
 r nr
where  is the net excess carrier lifetime.
The radiative efficiency is defined as the fraction of recombinations that are
radiative. We can write
_1
Rr r 
 __  __ _r (14.58)
Rr  Rnr _1 _ 1
r nr
where  is the radiative efficiency. The nonradiative recombination rate is propor-
tional to Nt , which is the density of nonradiative trapping sites within the forbidden
bandgap. Obviously, the radiative efficiency increases as Nt is reduced.
The internal quantum efficiency is now written as
i   (14.59)
The radiative recombination rate is proportional to the p-type doping. As the p-type
doping increases, the radiative recombination rate increases. However, the injection
efficiency decreases as the p-type doping increases; therefore, there is an optimum
doping that maximizes the internal quantum efficiency.

14.5.3 External Quantum Efficiency

One very important parameter of the LED is the external quantum efficiency: the frac-
tion of generated photons that are actually emitted from the semiconductor. The exter-
nal quantum efficiency is normally a much smaller number than the internal quantum
efficiency. Once a photon has been produced in the semiconductor, there are three loss
mechanisms the photon may encounter: photon absorption within the semiconductor,
Fresnel loss, and critical angle loss.
Figure 14.25 shows a pn junction LED. Photons can be emitted in any direc-
tion. Since the emitted photon energy must be h  Eg, these emitted photons can be
reabsorbed within the semiconductor material. The majority of photons will actually
be emitted away from the surface and reabsorbed in the semiconductor.
Photons must be emitted from the semiconductor into air; thus, the photons
must be transmitted across a dielectric interface. Figure 14.26 shows the incident,
_
reflected, and transmitted waves. The parameter n2 is the index of refraction for the
_
semiconductor and n1 is the index of refraction for air. The reflection coefficient is
_ _

n_2  n_1

2
  __ (14.60)
n2  n1
This effect is called Fresnel loss. The reflection coefficient  is the fraction of inci-
dent photons that are reflected back into the semiconductor.

nea29583_ch14_618-669.indd 650 12/11/10 12:51 PM

 14.5 Light Emitting Diodes 651

n2 n1
Emitted photon
Incident wave

p Transmitted wave
Reflected wave

Reflective coating
Figure 14.26 | Schematic of
Figure 14.25 | Schematic of photon incident, reflected, and transmitted
emission at the pn junction of an LED. photons at a dielectric interface.

Objective: Calculate the reflection coefficient at a semiconductor–air interface. EXAMPLE 14.8

Consider the interface between a GaAs semiconductor and air.

■ Solution
_
The index of refraction for GaAs is n2  3.8 at a wavelength of   0.70 m and the index of
_
refraction for air is n1  1.0. The reflection coefficient is
_ _
  __
_

n2  _n1 2
n2  n1
 __  
3.8  1.0 2  0.34
3.8  1.0 
■ Comment
A reflection coefficient of   0.34 means that 34 percent of the photons incident from
the gallium arsenide onto the semiconductor–air interface are reflected back into the
semiconductor.

■ EXERCISE PROBLEM
_
Ex 14.8 At a wavelength of   0.70 m, the index of refraction for GaAs is n2  3.8
_
and that for GaP is n2  3.2. Consider a GaAs1xPx material with a mole fraction
x  0.40. Assuming the index of refraction is a linear function of the mole frac-
tion, determine the reflection coefficient, , at the GaAs0.6P0.4–air interface.
(Ans.   0.315)

Photons incident on the semiconductor–air interface at an angle are refracted as

shown in Figure 14.27. If the photons are incident on the interface at an angle greater
than the critical angle c, the photons experience total internal reflection. The critical
angle is determined from Snell’s law and is given by
_
 
n_1
c  sin1 _
n2
(14.61)

nea29583_ch14_618-669.indd 651 12/11/10 12:51 PM

 652 CHAPTER 14 Optical Devices

n1

n2  n1

␪c

Figure 14.27 | Schematic showing

refraction and total internal reflection
at the critical angle at a dielectric
interface.

EXAMPLE 14.9 Objective: Calculate the critical angle at a semiconductor–air interface.

Consider the interface between GaAs and air.

■ Solution
_ _
For GaAs, n2  3.8 at a wavelength of   0.70 m and for air, n1  1.0. The critical angle is
_
c  sin1  n   sin  _
n1
_
_
2
1
3.8 
1.0  15.3°

■ Comment
Any photon that is incident at an angle greater than 15.3° will be reflected back into the
semiconductor.

■ EXERCISE PROBLEM
Ex 14.9 Repeat Example 14.9 for GaAs 0.6 P0.4. See Exercise Problem Ex 14.8 for a discus-
sion of the dielectric constant.
(Ans. c  16.3°)

Figure 14.28a shows the external quantum efficiency plotted as a function of the
p-type doping concentration and Figure 14.28b is a plot of the external efficiency as
a function of junction depth below the surface. Both figures show that the external
quantum efficiency is in the range of 1 to 3 percent.

14.5.4 LED Devices

The wavelength of the output signal of an LED is determined by the bandgap energy
of the semiconductor. Gallium arsenide, a direct bandgap material, has a bandgap
energy of Eg  1.42 eV, which yields a wavelength of   0.873 m. Comparing
this wavelength to the visible spectrum, which is shown in Figure 14.5, the output

nea29583_ch14_618-669.indd 652 12/11/10 12:51 PM

 14.5 Light Emitting Diodes 653

10 (Zn.O) pairs
unencapsulated
␩ext (%)

0.1
1016 1017 1018 1019
Photon energy (eV)
Na (cm3)
1.8 1.9 2.0 2.1 2.2
(a)

1.4 800 T  300 K

Brightness (fL) at 10 A/cm2

GaAs1x Px
Without nitrogen
1.2
600
␩ext (%)

1.0

400
0.8
Substrate with Nd  1018 cm3
T = 300 K Yellow
200 Green
Orange
10 15 20 25 30
x ( ␮m) Red

(b) 690 650 610 570

Wavelength  (nm)
Figure 14.28 | (a) External quantum
efficiency of a GaP LED versus acceptor Figure 14.29 | Brightness of GaAsP diodes
doping. (b) External quantum efficiency of versus wavelength (or versus bandgap
a GaAs LED versus junction depth. energy).
(From Yang [22].) (From Yang [22].)

of a GaAs LED is not in the visible range. For a visible output, the wavelength of
the signal should be in the range of 0.4 to 0.72 m. This range of wavelengths cor-
responds to bandgap energies between approximately 1.7 and 3.1 eV.
GaAs1xPx is a direct bandgap material for 0  x  0.45, as shown in Fig-
ure 14.24. At x  0.40, the bandgap energy is approximately Eg  1.9 eV, which
would produce an optical output in the red range. Figure 14.29 shows the bright-
ness of GaAs1xPx diodes for different values of x. The peak also occurs in the red
range. By using planar technology, GaAs0.6P0.4 monolithic arrays have been fabri-
cated for numeric and alphanumeric displays. When the mole fraction x is greater
than 0.45, the material changes to an indirect bandgap semiconductor so that the
quantum efficiency is greatly reduced.
GaAlxAs1x can be used in a heterojunction structure to form an LED. A device
structure is shown in Figure 14.30. Electrons are injected from the wide-bandgap
N-GaAl0.7As0.3 into the narrow-bandgap p-GaAl0.6As0.4. The minority carrier electrons

nea29583_ch14_618-669.indd 653 12/11/10 12:51 PM

 654 CHAPTER 14 Optical Devices

Contact

N GaAl0.7As0.3
P GaAl0.6As0.4

p GaAs

Contact
(a)

2.0 eV 1.42 eV
EF

2.1 eV

N-GaAl0.7As0.3 P-GaAl0.6As0.4 p-GaAs

(b)

Figure 14.30 | The (a) cross section and

(b) thermal equilibrium energy-band
diagram of a GaAlAs heterojunction LED.
(From Yang [22].)

in the p material can recombine radiatively. Since Egp  EgN, the photons are emit-
ted through the wide-bandgap N material with essentially no absorption. The wide
bandgap N material acts as an optical window and the external quantum efficiency
increases.

14.6 | LASER DIODES

The photon output of the LED is due to an electron giving up energy as it makes
a transition from the conduction band to the valence band. The LED photon emis-
sion is spontaneous in that each band-to-band transition is an independent event.
The spontaneous emission process yields a spectral output of the LED with a fairly
wide bandwidth. If the structure and operating condition of the LED are modified,
the device can operate in a new mode, producing a coherent spectral output with
a bandwidth of wavelengths less than 0.1 nm. This new device is a laser diode,
where laser stands for Light Amplification by Stimulated Emission of Radiation.
Although there are many different types of lasers, we are here concerned only with
the pn junction laser diode.

nea29583_ch14_618-669.indd 654 12/11/10 12:51 PM

 14.6 Laser Diodes 655

14.6.1 Stimulated Emission and Population Inversion

Figure 14.31a shows the case when an incident photon is absorbed and an electron
is elevated from an energy state E1 to an energy state E2. This process is known as
induced absorption. If the electron spontaneously makes the transition back to the
lower energy level with a photon being emitted, we have a spontaneous emission
process as indicated in Figure 14.31b. On the other hand, if there is an incident pho-
ton at a time when an electron is in the higher energy state as shown in Figure 14.31c,
the incident photon can interact with the electron, causing the electron to make a
transition downward. The downward transition produces a photon. Since this process
was initiated by the incident photon, the process is called stimulated or induced emis-
sion. Note that this stimulated emission process has produced two photons; thus, we
can have optical gain or amplification. The two emitted photons are in phase so that
the spectral output will be coherent.
In thermal equilibrium, the electron distribution in a semiconductor is deter-
mined by the Fermi–Dirac statistics. If the Boltzmann approximation applies, then
we can write
_N2
N1
 exp __ 
(E2  E1)
kT  (14.62)

where N1 and N2 are the electron concentrations in the energy levels E1 and E2,
respectively, and where E2  E1. In thermal equilibrium, N2  N1. The probability of
an induced absorption event is exactly the same as that of an induced emission event.
The number of photons absorbed is proportional to N1 and the number of additional
photons emitted is proportional to N2. In order to achieve optical amplification or for
lasing action to occur, we must have N2  N1; this is called population inversion. We
cannot achieve lasing action at thermal equilibrium.

E2
Induced
h
absorption
E1
(a)

E2
Spontaneous h
emission
E1
(b)

E2 Stimulated
h h
or induced
h
E emission
1

(c)

Figure 14.31 | Schematic diagram showing (a) induced

absorption, (b) spontaneous emission, and (c) stimulated
emission processes.

nea29583_ch14_618-669.indd 655 12/11/10 12:51 PM

 656 CHAPTER 14 Optical Devices

N2
E2
I
E1
N1
z

Figure 14.32 | Light propagating in

z direction through a material with
two energy levels.

Figure 14.32 shows the two energy levels with a light wave at an intensity I
propagating in the z direction. The change in intensity as a function of z can be
written as
# photons emitted ____
dI ___ # photons absorbed
_  
dz cm3 cm3
or
dI
_  N2Wi ⴢ h  N1Wi ⴢ h (14.63)
dz
where Wi is the induced transition probability. Equation (14.63) assumes no loss
mechanisms and neglects the spontaneous transitions.
Equation (14.63) can be written as
dI
_  ()I  (14.64)
dz
where ()  (N2  N1) and is the amplification factor. From Equation (14.64), the
intensity is
I  I(0)e ()z (14.65)
Amplification occurs when ()  0 and absorption occurs when ()  0.
We can achieve population inversion and lasing in a forward-biased pn homo-
junction diode, if both sides of the junction are degenerately doped. Figure 14.33a
shows the energy-band diagram of a degenerately doped pn junction in thermal equi-
librium. The Fermi level is in the conduction band in the n-region and the Fermi level
is in the valence band in the p region. Figure 14.33b shows the energy bands of the pn
junction when a forward bias is applied. The gain factor in a pn homojunction diode
is given by

 
h  (EFn  EFp)
()  1  exp ___
kT  (14.66)

In order for ()  1, we must have h  (EFn  EFp), which implies that the junction
must be degenerately doped since we also have the requirement that h  Eg. In the
vicinity of the junction, there is a region in which population inversion occurs. There
are large numbers of electrons in the conduction band directly above a large number
of empty states. If band-to-band recombination occurs, photons will be emitted with
energies in the range Eg  h  (EFn  EFp).

nea29583_ch14_618-669.indd 656 12/11/10 12:51 PM

 14.6 Laser Diodes 657

Laser
output
Active
region

p n
Ohmic
contact

L
EF
Ec
I
z
Ev
(a)

t
p n
EFn
Ec
h
EFp Ev
Optically flat
(b) Laser cleaved (110)
output surfaces
Figure 14.33 | (a) Degenerately doped
pn junction at zero bias. (b) Degenerately Figure 14.34 | A pn junction laser diode with cleaved (110)
doped pn junction under forward bias planes forming the Fabry-Perot cavity.
with photon emission. (After Yang [22].)

14.6.2 Optical Cavity

Population inversion is one requirement for lasing action to occur. Coherent emission
output is achieved by using an optical cavity. The cavity will cause a buildup of the
optical intensity from positive feedback. A resonant cavity consisting of two parallel
mirrors is known as a Fabry–Perot resonator. The resonant cavity can be fabricated, for
example, by cleaving a gallium arsenide crystal along the (110) planes as shown in Fig-
ure 14.34. The optical wave propagates through the junction in the z direction, bounc-
ing back and forth between the end mirrors. The mirrors are actually only partially
reflecting so that a portion of the optical wave will be transmitted out of the junction.
For resonance, the length of the cavity L must be an integral number of half
wavelengths, or

N _  L
2   (14.67)

where N is an integer. Since  is small and L is relatively large, there can be many
resonant modes in the cavity. Figure 14.35a shows the resonant modes as a function
of wavelength.

nea29583_ch14_618-669.indd 657 12/11/10 12:51 PM

 658 CHAPTER 14 Optical Devices

Longitudinal modes

2 Wavelength
2L
(a)

Spontaneous
emission or
Optical
optical gain
loss
spectrum


(b)

Lasing modes


(c)

Figure 14.35 | Schematic diagram

showing (a) resonant modes of a cavity
with length L, (b) spontaneous emission
curve, and (c) actual emission modes of
a laser diode.
(After Yang [22].)

When a forward-bias current is applied to the pn junction, spontaneous emission

will initially occur. The spontaneous emission spectrum is relatively broadband and
is superimposed on the possible lasing modes as shown in Figure 14.35b. In order for
lasing to be initiated, the spontaneous emission gain must be larger than the optical
losses. By positive feedback in the cavity, lasing can occur at several specific wave-
lengths as indicated in Figure 14.35c.

14.6.3 Threshold Current

The optical intensity in the device can be written from Equation (14.65) as
I  e ()z, where () is the amplification factor. We have two basic loss mecha-
nisms. The first is the photon absorption in the semiconductor material. We can
write
I  e()z (14.68)
where () is the absorption coefficient. The second loss mechanism is due to the
partial transmission of the optical signal through the ends, or through the partially
reflecting mirrors.

nea29583_ch14_618-669.indd 658 12/11/10 12:51 PM

 14.6 Laser Diodes 659

At the onset of lasing, which is known as threshold, the optical loss of one round
trip through the cavity is just offset by the optical gain. The threshold condition is
then expressed as
12 exp [(2 t ()  2())L]  1 (14.69)
where 1 and 2 are the reflectivity coefficients of the two end mirrors. For the case
when the optical mirrors are cleaved (110) surfaces of gallium arsenide, the reflectiv-
ity coefficients are given approximately by
_ _

n  n_1

2
1  2  __
_2 (14.70)
n2  n1
_ _
where n2 and n1 are the index of refraction parameters for the semiconductor and air,
respectively. The parameter t () is the optical gain at threshold.
The optical gain at threshold, t (), may be determined from Equation (14.69) as

t()    _1 ln _
2L
1
12   (14.71)

Since the optical gain is a function of the pn junction current, we can define a thresh-
old current density as
1 _
Jth  _
 
1 ln _
2L
1
12   (14.72)

where  can be determined theoretically or experimentally. Figure 14.36 shows the

threshold current density as a function of the mirror losses. We may note the rela-
tively high threshold current density for a pn junction laser diode.

3500

3000
300 K

2500
1 –
Jth  –  ( 1 1
)
Jth (A/cm2)

2000 ln
 2L 12
–  15 cm1

–
1500   2.1 102 cm /A

1000

500

0
0 10 20 30 40 50 60 70 80
1 1
ln (cm1)
2L 12

Figure 14.36 | Threshold current density of a laser diode as a

function of Fabry-Perot cavity end losses.
(After Yang [22].)

nea29583_ch14_618-669.indd 659 12/11/10 12:51 PM

 660 CHAPTER 14 Optical Devices

n p p
AlxGa1x As GaAs AlxGa1x As (a)

Jn
3.6

h  1.38 eV p
3.5 h
T  297 K

Energy
(b)
n
3.4 
Refractive index –n

d
3.3 Jp

Light Refractive index

(1 m) 5%
3.2 (c)

3.1

3.0 (d)

2.9  0.1 m
0 0.5 1.0
GaAs AlAs
Mole fraction AlAs, x
Figure 14.38 | (a) Basic double hetero-
junction structure. (b) Energy-band diagram
Figure 14.37 | Index of refraction under forward bias. (c) Refractive index
of AlxGa1xAs as a function of mole change through the structure. (d) Confine-
fraction x. ment of light in the dielectric waveguide.
(From Sze [18].) (From Yang [22].)

14.6.4 Device Structures and Characteristics

We have seen that in a homojunction LED, the photons may be emitted in any direc-
tion, which lowers the external quantum efficiency. Significant improvement in de-
vice characteristics can be made if the emitted photons are confined to a region near
the junction. This confinement can be achieved by using an optical dielectric wave-
guide. The basic device is a three-layered, double heterojunction structure known as
a double heterojunction laser. A requirement for a dielectric waveguide is that the
index of refraction of the center material be larger than that of the other two dielec-
trics. Figure 14.37 shows the index of refraction for the AlGaAs system. We may
note that GaAs has the highest index of refraction.
An example of a double heterojunction laser is shown in Figure 14.38a. A thin
p-GaAs layer is between P-AlGaAs and N-AlGaAs layers. A simplified energy-band
diagram is shown in Figure 14.38b for the forward-biased diode. Electrons are in-
jected from the N-AlGaAs into the p-GaAs. Population inversion is easily obtained
since the conduction band potential barrier prevents the electrons from diffusing
into the P-AlGaAs region. Radiative recombination is then confined to the p-GaAs
region. Since the index of refraction of GaAs is larger than that of AlGaAs, the light
wave is also confined to the GaAs region. An optical cavity can be formed by cleav-
ing the semiconductor perpendicular to the N-AlGaAs–p-GaAs junction.

nea29583_ch14_618-669.indd 660 12/11/10 12:51 PM

 14.7 Summary 661

14 AlGaAs DH laser
 (20 C)  827 nm
W  12 m
L  130 m
12

0
CW power emission (one facet) (mW)

10
10
20
30
40
8 50
60
70
6

Threshold
current
2 at 70 C

0
0 20 40 60 80 100
Diode current (mA)

Figure 14.39 | Typical output power versus laser diode

current at various temperatures.
(From Yang [22].)

Typical optical output versus diode current characteristics are shown in Fig-
ure 14.39. The threshold current is defined to be the current at the breakpoint. At
low currents, the output spectrum is very wide and is the result of the spontaneous
transitions. When the diode current is slightly above the threshold value, the various
resonant frequencies are observed. When the diode current becomes large, a single
dominant mode with a narrow bandwidth is produced.
The performance of the laser diode can be further improved if a very narrow
recombination region is used with a somewhat wider optical waveguide. Very com-
plex structures using multilayers of compound semiconductor materials have been
fabricated in a continuing effort to improve semiconductor laser performance.

14.7 | SUMMARY
■ The absorption or emission of light (photons) in semiconductors leads to the study of a
general class of devices called optoelectronics. A few of these devices have been dis-
cussed and analyzed in this chapter.

nea29583_ch14_618-669.indd 661 12/11/10 12:51 PM

 662 CHAPTER 14 Optical Devices

■ The photon absorption process has been discussed and the absorption coefficient data
for semiconductors has been presented.
■ Solar cells convert optical power into electrical power. The simple pn junction solar cell
was initially considered. The short-circuit current, open-circuit voltage, and maximum
power were considered.
■ Heterojunction and amorphous silicon solar cells were also considered. Heterojunction
cells can be fabricated that tend to increase the conversion efficiency and produce rela-
tively large open-circuit voltages. Amorphous silicon offers the possibility of low-cost,
large-area solar cell arrays.
■ Photodetectors are semiconductor devices that convert optical signals into electrical sig-
nals. The photoconductor is perhaps the simplest type of photodetector. The change in
conductivity of the semiconductor due to the creation of excess electrons and holes by
the incident photons is the basis of this device.
■ Photodiodes are diodes that have reverse-biased voltages applied. Excess carriers
that are created by incident photons in the space-charge region are swept out by the
electric field creating a photocurrent. The photocurrent is directly proportional to the
incident photon intensity. PIN and avalanche photodiodes are variations of the basic
photodiode.
■ The photocurrent generated in a phototransistor is multiplied by the transistor gain.
However, the time response of the phototransistor may be slower than that of a photodi-
ode because of the Miller effect and Miller capacitance.
■ The inverse mechanism of photon absorption in a pn junction is injection electro-
luminescence. The recombination of excess electrons and holes in a direct bandgap
semiconductor can result in the emission of photons.
■ The light emitting diodes (LEDs) are the class of pn junction diodes whose photon
output is a result of spontaneous recombinations of excess electrons and holes. A fairly
wide bandwidth in the output signal, on the order of 30 nm, is a result of the sponta-
neous process.
■ The output of a laser diode is the result of stimulated emission. An optical cavity, or
Fabry–Perot resonator, is used in conjunction with a diode so that the photon output
is in phase, or coherent. Multilayered heterojunction structures can be fabricated to
improve the laser diode characteristics.

GLOSSARY OF IMPORTANT TERMS

absorption coefficient The relative number of photons absorbed per unit distance in a semi-
conductor and denoted by the parameter .
conversion efficiency The ratio of output electrical power to incident optical power in a
solar cell.
delayed photocurrent The component of photocurrent in a semiconductor device due to
diffusion currents.
external quantum efficiency The ratio of emitted photons to generated photons in a semi-
conductor device.
fill factor The ratio ImVm to IscVoc, which is a measure of the realizable power from a solar
cell. The parameters Im and Vm are the current and voltage at the maximum power point,
respectively, and Isc and Voc are the short-circuit current and open-circuit voltage.
fresnel loss The ratio of reflected to incident photons at an interface due to a change in the
index of refraction.

nea29583_ch14_618-669.indd 662 12/11/10 12:51 PM

 Review Questions 663

internal quantum efficiency The fraction of diode current that produces luminescence.
LASER diode An acronym for Light Amplification by Stimulated Emission of Radiation;
the stimulated emission of photons produced in a forward-biased pn junction in conjunc-
tion with an optical cavity.
LED An acronym for Light Emitting Diode; the spontaneous photon emission due to
electron–hole recombination in a forward-biased pn junction.
luminescence The general property of light emission.
open-circuit voltage The voltage generated across the open-circuited terminals of a solar
cell.
photocurrent The current generated in a semiconductor device due to the flow of excess
carriers generated by the absorption of photons.
population inversion The condition whereby the concentration of electrons in one energy
state is greater than that in a lower energy state; a nonequilibrium condition.
prompt photocurrent The component of photocurrent generated within the space charge
region of a semiconductor device.
radiative recombination The recombination process of electrons and holes that produces a
photon, such as the direct band-to-band transition in gallium arsenide.
short-circuit current The current produced in a solar cell when the two terminals are
shorted together.
stimulated emission The process whereby an electron is induced by an incident photon to
make a transition to a lower energy state, emitting a second photon.

CHECKPOINT
After studying this chapter, the reader should have the ability to:
■ Describe the optical absorption process in semiconductors. When is optical absorption
essentially zero?
■ Describe the basic operation and characteristics of a solar cell, including the short-
circuit current and open-circuit voltage.
■ Discuss the factors that contribute to the solar cell conversion efficiency.
■ Describe the advantages and disadvantages of an amorphous silicon solar cell.
■ Describe the characteristics of a photoconductor, including the concept of the photocon-
ductor gain.
■ Discuss the operation and characteristics of a simple pn junction photodiode.
■ Discuss the advantages of PIN and avalanche photodiodes compared to the simple
pn junction photodiode.
■ Discuss the operation and characteristics of a phototransistor.
■ Describe the operation of an LED.
■ Describe the operation of a laser diode.

REVIEW QUESTIONS
1. Sketch the general shape of the optical absorption coefficient in a semiconductor as a
function of wavelength. When does the absorption coefficient become zero?
2. Sketch the I–V characteristic of a pn junction solar cell. Define short-circuit current and
open-circuit voltage.

nea29583_ch14_618-669.indd 663 12/11/10 12:51 PM

 664 CHAPTER 14 Optical Devices

3. Discuss how a pn junction solar cell becomes forward biased.

4. Write an expression for the steady-state photocurrent in a simple photoconductor.
5. What is the source of prompt photocurrent in a photodiode? Does the prompt photocur-
rent depend on the reverse-biased voltage? Why or why not.
6. Sketch the cross section of a phototransistor and show the currents that are created by
incident photons. Explain how current gain is achieved.
7. Explain the basic operation of an LED. State two factors that affect the efficiency of the
device.
8. How can different colors be obtained in an LED?
9. Discuss the difference between an LED and a laser diode.
10. Discuss the concept of population inversion in a laser diode.

PROBLEMS
Section 14.1 Optical Absorption
14.1 Determine the maximum wavelength  of a light source that can generate electron–
hole pairs in (a) Si, (b) Ge, (c) GaAs, and (d ) InP.
14.2 (a) Two sources generate light at wavelengths of   480 nm and   725 nm,
respectively. What are the corresponding photon energies? (b) Three sources gen-
erate light with photon energies of E  0.87 eV, E  1.32 eV, and E  1.90 eV,
respectively. What are the corresponding wavelengths?
14.3 (a) A sample of GaAs is 1.2 m thick. The sample is illuminated with a light source
that generates photons with energies of h  1.65 eV. Determine the (i) absorption
coefficient and (ii) fraction of energy that is absorbed in the material. (b) Repeat
part (a) for a sample of GaAs that is 0.80 m thick and is illuminated with photons
with energies of h  1.90 eV.
14.4 A light source with h  1.3 eV and at a power density of 102 W/cm2 is incident
on a thin slab of silicon. The excess minority carrier lifetime is 106 s. Determine
the electron–hole generation rate and the steady-state excess carrier concentration.
Neglect surface effects.
14.5 An n-type GaAs sample has a minority carrier lifetime of p  2 107 s. Incident
photons with energies h  1.65 eV generate an excess carrier concentration of
p  5 1015 cm3 at the surface of the semiconductor. (a) Determine the incident
power required. (b) At what distance in the semiconductor does the generation rate
drop to 10 percent of that at the surface?
14.6 Consider a silicon semiconductor that is illuminated with photons with energies
h  1.40 eV. (a) Determine the thickness of the material such that 90 percent of
the energy is absorbed. (b) Determine the thickness of the material such that 30 per-
cent of the energy is transmitted through the material.
14.7 If the thickness of a GaAs semiconductor is 1 m and 50 percent of the incident
monochromic photon energy is absorbed, determine the incident photon energy and
wavelength.
*14.8 Consider monochromatic light at an intensity I 0 incident on the surface at x  0
of an n-type semiconductor that extends to x  . Assume the electric field is zero
in the semiconductor and assume a surface recombination velocity, s. Taking into

*Asterisks next to problems indicate problems that are more difficult.

nea29583_ch14_618-669.indd 664 12/11/10 12:51 PM

 Problems 665

I0

s⬁ s  s0

x0 xW

Figure P14.9 | Figure for

Problem 14.9.

account the absorption coefficient, determine the steady-state excess hole concen-
tration as a function of x.
*14.9 Monochromatic light with intensity I 0 is incident on a p-type semiconductor as
shown in Figure P14.9. Assume the surface recombination velocity at x  0 is
s   and assume the surface recombination velocity at x  W is s  s0. Derive the
expression for the steady-state excess electron concentration as a function of x.

Section 14.2 Solar Cells

14.10 A long silicon pn junction solar cell at T  300 K has the following parameters:
Na  1016 cm3, Nd  1015 cm3, Dn  25 cm2/s, Dp  10 cm2/s, n0  106 s, and
p0  5 107 s. The cross-sectional area of the solar cell is 5 cm2. The entire junc-
tion is uniformly illuminated such that the generation rate of electron–hole pairs is
GL  5 1021 cm3 s1. (a) Calculate the short circuit photocurrent generated in
the space charge region. (b) Using the results of part (a), calculate the open-circuit
voltage. (c) Determine the ratio of Voc to Vbi.
14.11 A long silicon pn junction solar cell has the same parameters as described in
Problem 14.10. The generated photocurrent in the cell is IL  120 mA. Determine
the (a) open-circuit voltage, (b) the voltage across the junction that will produce
a total solar cell current of I  100 mA, (c) the maximum power output of
the solar cell, and (d) the external load resistance that will produce the maximum
power.
14.12 Consider the solar cell described in Problem 14.10. (a) The generated photocurrent
is IL  10 mA. Determine (i) the open-circuit voltage and (ii) the maximum power
output. (b) The solar cell now uses a solar concentrator such that the photocurrent
increases by a factor of 10. Determine the new values of (i) open circuit voltage
and (ii) maximum power output. (c) Determine the ratio of maximum power from
part (b) to that from part (a).
14.13 Consider an ideal long np junction GaAs solar cell at T  300 K in which excess
carriers are uniformly generated. The parameters of the diode are as follows:
Nd  1019 cm3 Dn  225 cm2/s
n0  p0  5 108 s Dp  7 cm2/s.
The generated photocurrent density is JL  30 mA/cm2. Plot the open-circuit volt-
age as a function of the acceptor doping concentration for 1015  Na  1018 cm3.

nea29583_ch14_618-669.indd 665 12/11/10 12:51 PM

 666 CHAPTER 14 Optical Devices

14.14 A long silicon pn junction solar cell with an area of 2 cm2 has the following
parameters:
Nd  1019 cm3 Na  3 1016 cm3
Dp  6 cm /s
2
Dn  18 cm2/s
p0  5 10 s
7
n0  5 106 s
Assume that excess carriers are uniformly generated in the solar cell and that
JL  25 mA/cm2. Let T  300 K. (a) Plot the I–V characteristics of the diode,
(b) determine the maximum power output of the solar cell, and (c) calculate the
external load resistance that will produce the maximum power.
14.15 A silicon solar cell at T  300 K has a cross-sectional area of 6 cm2 and a reverse
saturation current of IS  2 109 A. The induced short-circuit photocurrent is
IL  180 mA. Determine the (a) open-circuit voltage, (b) maximum power output,
and (c) load resistance that will produce the maximum output power. (d ) If the load
resistance determined in part (c) is increased by 50 percent, what is the new value
of the maximum output power?
14.16 Consider a silicon solar cell at T  300 K with a reverse saturation current of
IS  1010 A and an induced short-circuit photocurrent of IL  100 mA. (a) Deter-
mine Voc. (b) Find Vm, Im, and Pm. (c) How many cells, operating at the maximum
output power, must be connected in series to produce an output voltage of at least
10 V? (d) How many of the 10 V cells in part (c) must be connected in parallel to
produce an output power of at least 5.2 W? (e) Considering the results of part (d ),
what must be the load resistance connected across the solar cell system to produce
the maximum output power?
*14.17 Consider the pn junction solar cell with nonuniform absorption. Derive the expres-
sion for the excess minority carrier electron concentration for the short-circuit con-
dition and for the case when the p region is very long and the n region is short.
14.18 The absorption coefficient in amorphous silicon is approximately 104 cm1 at
h  1.7 eV and 105 cm1 at h  2.0 eV. Determine the amorphous silicon
thickness for each case so that 90 percent of the photons are absorbed.

Section 14.3 Photodetectors

14.19 Consider an n-type silicon photoconductor at T  300 K doped at
Nd  5 1015 cm3. The cross-sectional area is A  5 104 cm2 and the length
is L  120 m. The carrier parameters are n  1200 cm2/V-s, p  400 cm2/V-s,
n0  5 107 s, and p0  107 s. The photoconductor is uniformly illuminated
such that the generation rate of electron–hole pairs is GL  1021 cm3 s1. For
3 volts applied to the photoconductor, determine (a) the thermal equilibrium cur-
rent, (b) the steady-state excess carrier concentration, (c) the photoconductivity,
(d) the steady-state photocurrent, and (e) the photocurrent gain.
14.20 Excess carriers are uniformly generated in a GaAs photoconductor at a rate of
GL  1021 cm3-s1. The area is A  104 cm2 and the length is L  100 m. The
other parameters are:
Nd  5 1016 cm3 Na  0
n  8000 cm2/V-s p  250 cm2/V-s
n0  107 s p0  108 s.

nea29583_ch14_618-669.indd 666 12/11/10 12:51 PM

 Problems 667

If a voltage of 5 volts is applied, calculate (a) the steady-state excess carrier con-
centration, (b) the photoconductivity, (c) the steady-state photocurrent, and (d ) the
photoconductor gain.
*14.21 Consider an n-type silicon photoconductor that is 1 m thick, 50 m wide,
and has an applied electric field in the longitudinal dimension of 50 V/cm. If
the incident photon flux is 0  1016 cm2-s1 and the absorption coefficient is
  5 104 cm1, calculate the steady-state photocurrent if n  1200 cm2/V-s,
p  450 cm2/V-s, and p0  2 107 s.
14.22 A long silicon pn junction photodiode has the following parameters at T  300 K:
Na  1016 cm3, Nd  2 1015 cm3, Dp  10 cm2/s, Dn  25 cm2/s, p0  107 s,
and n0  5 107 s. The cross-sectional area of the diode is A  103 cm2.
Assume that a reverse-biased voltage of 5 volts is applied and that a uniform
generation rate for electron–hole pairs of GL  1021 cm3 s1 exists throughout the
entire photodiode. (a) Determine the prompt component of photocurrent. (b) Find
the steady-state excess carrier concentrations in the p and n regions far from the
junction. (c) Determine the total steady-state photocurrent.
*14.23 Starting with the ambipolar transport equation for minority carrier holes, derive
Equation (14.41) using the geometry shown in Figure 14.17.
14.24 Three silicon PIN photodiodes A, B, and C, at T  300 K have intrinsic region
widths of 2, 10, and 80 m, respectively. A photon flux of 0  5 1017 cm2 s1
is incident on the surface of each diode as shown in Figure 14.19. (a) For an ab-
sorption coefficient of   104 cm1, calculate the prompt photocurrent density in
each diode. (b) Repeat part (a) for an absorption coefficient of   5 102 cm1.
14.25 Consider a silicon PIN photodiode at T  300 K with the geometry shown in Figure
14.19. The intrinsic region width is 100 m. Assume that a reverse-biased voltage
is applied such that the intrinsic region is completely depleted. The incident photon
power is I 0  0.080 W/cm2, the absorption coefficient is   103 cm1, and the
photon energy is 1.5 eV. Neglect any absorption in the p top layer of the photodi-
ode. (a) Determine the steady-state electron–hole generation rate, GL, versus dis-
tance in the intrinsic region. (b) Determine the steady-state photocurrent density.
14.26 A silicon PIN photodiode at T  300 K has the geometry shown in Figure 14.19.
The intrinsic region width is 20 m and is fully depleted. (a) The electron–hole
pair generation rate in the intrinsic region is GL  1021 cm3 s1 and is uniform
throughout the intrinsic region. Calculate the steady-state photocurrent density for
this condition. (b) The generation rate of electron–hole pairs is GL  1021 cm3 s1
at x  0 and the absorption coefficient is   103 cm1. Determine the steady-state
photocurrent density for this situation.
14.27 Consider a silicon PIN photodiode exposed to sunlight. Calculate the intrinsic
region width so that at least 90 percent of all photons with wavelengths   1 m
are absorbed in the intrinsic region. Neglect any absorption in the p or n regions.

Section 14.4 Photoluminescence and Electroluminescence

14.28 Consider the AlxGa1xAs system. Determine the range of the direct bandgap ener-
gies possible and the corresponding range of wavelengths.
14.29 Consider the GaAs1xPx system. (a) For a mole fraction x  0.2, determine the
(i) bandgap energy and (ii) corresponding photon wavelength. (b) Repeat part (a)
for a mole fraction x  0.32.

nea29583_ch14_618-669.indd 667 12/11/10 12:51 PM

 668 CHAPTER 14 Optical Devices

14.30 Using Figure 14.23, determine the mole fraction x in AlxGa1x As such that the
material would emit light at a wavelength of   0.670 m. What is the corre-
sponding bandgap energy?
14.31 Repeat Problem 14.30 for the GaAs1xPx system.

Section 14.5 Light Emitting Diodes

14.32 Consider a pn junction GaAs LED. Assume that photons are generated uniformly in
all directions in a plane perpendicular to the junction at a distance of 0.50 m from
the surface. (a) Taking into account total internal reflection, calculate the fraction of
photons that have the potential of being emitted from the semiconductor. (b) Using
the results of part (a) and including Fresnel loss, determine the fraction of gener-
ated photons that will be emitted from the semiconductor into air (neglect absorp-
tion losses).
*14.33 In a pn junction LED, consider a point source in the semiconductor at the junction
and assume that photons are emitted uniformly in all directions. Show that (neglect-
ing photon absorption) the external quantum efficiency of the LED is given by
_ _
2n n
ext  __
_ 1 _2 (1  cos c)
(n1  n2)2
_ _
where n1 and n2 are the index of refraction parameters for the air and semiconductor,
respectively, and c is the critical angle.

Section 14.6 Laser Diodes

14.34 Consider an optical cavity. If N  1, show that the wavelength separation between
two adjacent resonant modes is   22L.
14.35 If the photon output of a laser diode is equal to the bandgap energy, find the
wavelength separation between adjacent resonant modes in a GaAs laser with
L  75 m.

READING LIST
1. Bhattacharya, P. Semiconductor Optoelectronic Devices, 2nd ed. Upper Saddle River,
NJ: Prentice Hall, 1997.
2. Carlson, D. E. “Amorphous Silicon Solar Cells.” IEEE Transactions on Electron
Devices ED-24 (April 1977), pp. 449–53.
3. Fonash, S. J. Solar Cell Device Physics. New York: Academic Press, 1981.
4. Kano, K. Semiconductor Devices. Upper Saddle River, NJ: Prentice Hall, 1998.
5. Kressel, H. Semiconductor Devices for Optical Communications: Topics in Applied
Physics. Vol. 39. New York: Springer-Verlag, 1987.
6. MacMillan, H. F., H. C. Hamaker, G. F. Virshup, and J. G. Werthen. “Multijunction
III-V Solar Cells: Recent and Projected Results.” Twentieth IEEE Photovoltaic
Specialists Conference (1988), pp. 48–54.
7. Madan, A. “Amorphous Silicon: From Promise to Practice.” IEEE Spectrum 23
(September 1986), pp. 38–43.
8. Pankove, J. I. Optical Processes in Semiconductors. New York: Dover Publications, 1971.

nea29583_ch14_618-669.indd 668 12/11/10 12:51 PM

 Reading List 669

9. Pierret, R. F. Semiconductor Device Fundamentals. Reading, MA: Addison-Wesley,

1996.
10. Roulston, D. J. An Introduction to the Physics of Semiconductor Devices. New York:
Oxford University Press, 1999.
11. Schroder, D. K. Semiconductor Material and Devices Characterization, 3rd ed.
Hoboken, NJ: John Wiley and Sons, 2006.
12. Shur, M. Introduction to Electronic Devices. New York: John Wiley and Sons, 1996.
*13. . Physics of Semiconductor Devices. Englewood Cliffs, NJ: Prentice Hall,
1990.
14. Singh, J. Optoelectronics: An Introduction to Materials and Devices. New York:
McGraw-Hill, 1996.
15. . Semiconductor Devices: Basic Principles. New York: John Wiley and Sons,
2001.
16. Streetman, B. G., and S. K. Banerjee. Solid State Electronic Devices, 6th ed. Upper
Saddle River, NJ: Pearson Prentice-Hall, 2006.
17. Sze, S. M. Semiconductor Devices: Physics and Technology. New York: John Wiley
and Sons, 1985.
18. Sze, S. M. and K. K. Ng. Physics of Semiconductor Devices, 3rd ed. Hoboken,
NJ: John Wiley and Sons, 2007.
*19. Wang, S. Fundamentals of Semiconductor Theory and Device Physics. Englewood
Cliffs, NJ: Prentice Hall, 1989.
20. Wilson, J., and J. F. B. Hawkes. Optoelectronics: An Introduction. Englewood Cliffs,
NJ: Prentice Hall, 1983.
*21. Wolfe, C. M, N. Holonyak, Jr., and G. E. Stillman. Physical Properties of Semicon-
ductors. Englewood Cliffs, NJ: Prentice Hall, 1989.
22. Yang, E. S. Microelectronic Devices. New York: McGraw-Hill, 1988.

*Indicates reference that is at an advanced level compared to this text.

nea29583_ch14_618-669.indd 669 12/11/10 12:51 PM

 15
C H A P T E R

Semiconductor Microwave and

Power Devices


n previous chapters, we have discussed the basic physics, operation, and char-
acteristics of diodes and transistors. We have analyzed the frequency response
as well as the current–voltage characteristics of these semiconductor devices.
However, we have not specifically considered the generation of microwave signals
using semiconductor devices or the power capabilities of semiconductor transistors.
In this chapter, we first consider three semiconductor devices that are used to gen-
erate microwave signals. These devices include the tunnel diode, GUNN diode, and
IMPATT diode. A basic principle of oscillators is that a region of negative differential
resistance must exist. We consider the process by which a region of negative differen-
tial resistance is created in each device and discuss the basic operation of these devices.
Second, we discuss three specialized semiconductor power devices, including
power bipolar transistors and power MOSFETs. We have considered the basic phys-
ics of these devices in previous chapters, and analyzed the current–voltage character-
istics without specifically considering the current or voltage limitations or the power
dissipation within the devices. In this chapter, we discuss the limitations in current
and voltage, and the power capabilities of the devices. Finally, we discuss the opera-
tion and characteristics of a four-layered structure called a thyristor. ■

15.0 | PREVIEW
In this chapter, we will:
■ Discuss the concept of negative differential resistance in a tunnel diode and
derive an expression for the maximum resistance cutoff frequency.
■ Discuss the concept of negative differential mobility in GaAs and discuss the
process by which this characteristic can lead to microwave oscillations in a
GUNN diode.
■ Discuss the operation of an IMPATT diode oscillator and determine the pro-
cess by which a dynamic negative resistance is created.

670

nea29583_ch15_670-706.indd 670 12/11/10 12:57 PM

 15.1 Tunnel Diode 671

■ Present the basic geometry and electrical characteristics of a power bipolar

transistor. The limiting current and voltage factors will be analyzed, and the
safe operating area of the BJT will be considered.
■ Present the basic geometry and electrical characteristics of a power MOSFET.
The limiting current and voltage factors will be analyzed, and the safe operat-
ing area of the MOSFET will be considered.
■ Discuss the operation of a four-layer switching device that is generally referred
to as a Thyristor. The operation of several structures will be analyzed.

15.1 | TUNNEL DIODE

The tunnel diode, also known as the Esaki diode, has been briefly discussed in Sec-
tion 8.5 of the book. Recall that the device is a pn junction in which both the n and p
regions are degenerately doped. With the very high doping concentrations, the space
charge region width is very narrow (W  0.5  106 cm  50 Å).
The forward-bias current–voltage characteristics are again shown in Fig-
ure 15.1a. For small forward-bias voltages (V  Vp), electrons in the conduction
band on the n side are directly opposite empty states in the valence band of the p
region (see Figure 8.29). Electrons tunnel through the potential barrier into the empty
states producing a tunneling current. For forward-bias voltages in the range Vp  V
 Vv, the number of electrons on the n side directly opposite empty states on the p
side decreases so that the tunneling current decreases. For V  Vv, the normal diode
diffusion currents dominate.
A decrease in current with an increase in voltage produces a region of negative
differential resistance in the range Vp  V  Vv. A negative differential resistance
phenomenon is necessary for oscillators.

I I
Peak
Ip
Tunneling
current
Ip
Rmin
Iv

Diffusion Iv
current
Valley

Vp Vv V Vp Vv V
(a) (b)

Figure 15.1 | (a) Forward-bias current–voltage characteristics of a tunnel diode.

(b) Expanded plot of I–V characteristics.

nea29583_ch15_670-706.indd 671 12/11/10 12:57 PM

 672 CHAPTER 15 Semiconductor Microwave and Power Devices

Lp Rp

z Cj Rmin

Figure 15.2 | Equivalent circuit of the tunnel

diode.

Figure 15.1b shows an expanded plot of the I–V characteristics in the tunnel-
ing range. A point is shown on the curve where the minimum value of negative
resistance occurs. (Note that Rmin is a positive quantity.) The equivalent circuit of
the tunnel diode for the case when the diode is biased at the  Rmin point is shown in
Figure 15.2. The parameter Cj is the junction capacitance, and the parameters Lp and
Rp are the parasitic or interconnect line inductance and resistance, respectively.
The small signal input impedance can be written as

   
Rmin R2minCj
Z  Rp  ___  j L p  ___ (15.1)
1  2 Rmin Cj
2 2
1  2 R2min C2j
The resistive part of the impedance goes to zero at a frequency of
________

R
1 Rmin
fr  __ _ 1 (15.2)
2Rmin Cj p

For frequencies f  fr, the resistive part of the impedance becomes positive so that
the diode loses its negative differential resistance characteristic. The operating fre-
quency must then occur at fo  fr. The frequency fr is referred to as the maximum
resistive cutoff frequency.
The tunneling process is a majority carrier effect so the diode does not exhibit
time delays due to minority carrier diffusion, which means that the diode is capable
of operating at microwave frequencies. However, due to the relatively small voltage
range in which the diode exhibits the negative resistance characteristic, the tunnel
diode is not used extensively.

15.2 | GUNN DIODE

Another negative differential resistance device is the GUNN diode, or Transferred-
Electron Device (TED). The transferred-electron phenomenon is demonstrated in a
few semiconductors in which conduction electrons in a high-mobility band are scat-
tered to a low-mobility band by a high electric field. In Chapter 5, we discussed the
drift velocity of electrons in GaAs versus electric field. Figure 15.3 again shows a
plot of this characteristic. InP also shows this same characteristic.
Figure 15.4 shows an expanded plot of the energy-band structure in GaAs that
is given in Figure 5.8. For small electric fields, essentially all of the electrons in the

nea29583_ch15_670-706.indd 672 12/11/10 12:57 PM

 15.2 GUNN Diode 673

Drift velocity (107cm-s)

2

0 Eth 5 10 15
Ebias E(kV/cm)

Figure 15.3 | Electron drift velocity versus electric

field for GaAs.

Upper valley
m*n
 0.55
m0
Lower valley
m*n Conduction
 0.067 band
0.3 eV m0

1.42 eV

k
Valence
band

Figure 15.4 | Energy-band structure of GaAs showing the

lower valley and upper valley in the conduction band.

conduction band exist in the lower valley of the E versus k diagram, where the den-
sity of states electron effective mass is small. A small effective mass leads to a large
mobility value.
As the electric field increases above a threshold or critical value, Eth, the electrons
gain more than the 0.3 eV energy separating the two valleys so that electrons can be
scattered into the upper valley, where the density of states electron effective mass is
much larger. The larger effective mass yields a smaller mobility. The intervalley trans-
fer mechanism with a change in mobility results in a decreasing average drift velocity
of electrons with electric field, or a negative differential electron mobility. The maxi-
mum negative differential electron mobility in GaAs is approximately 2400 cm2 /V-s.
Consider a two-terminal n-type GaAs device with ohmic contacts at the ends
that is biased in the negative mobility region (Ebias  Eth) as shown in Figure 15.5a.
A small space charge region may develop in the material near the cathode as shown
in Figure 15.5b. As a result, the electric field increases in this region as shown in
Figure 15.5c. (Special device structures can be fabricated to ensure that the space
charge fluctuations are generated near the cathode.)

nea29583_ch15_670-706.indd 673 12/11/10 12:57 PM

 674 CHAPTER 15 Semiconductor Microwave and Power Devices

Cathode Anode

Electron drift  Vbias

n  type

Ebias
(a)

n Gate
n0

(b)

E
Eth

(c)

Figure 15.5 | (a) A simplified two-terminal

GaAs device. (b) Electron concentration versus
distance showing a space charge formation.
(c) Electric field versus distance.

In discussing excess carrier behavior in Chapter 6, we found the time behavior

of a net charge density in a semiconductor to be given by
Q(t)  Q(0)et d
(15.3)
where d is the dielectric relaxation time constant and is on the order of a picosecond.
Normally, a small space charge region would be quickly neutralized. The dielectric
relaxation time constant is given by d  , where  is the semiconductor conduc-
tivity. If the GaAs is biased in the negative mobility region, then the conductivity is
negative and the exponent in Equation (15.3) becomes positive, so the space charge
region, now called a domain, can actually build up as it drifts toward the anode. As the
domain grows (Figure 15.6a), the electric field in this region increases which means
that the electric field in the remaining material decreases. The E field in the material
outside of the domain can drop below the critical value, as indicated in Figure 15.6b,
while the E field within the domain remains above the critical value. For this reason,
only one domain will normally be established in the material at any given time.
As the domain reaches the anode, a current pulse is induced in the external cir-
cuit. After the domain reaches the anode, another domain may form near the cathode
and the process repeats itself. Thus, a series of current pulses may be generated as
shown in Figure 15.7. The time between current pulses is the time for the domain to
drift through the device. The oscillation frequency is given by
f  1  vdL (15.4)
where vd is the average drift velocity and L is the length of the drift region.

nea29583_ch15_670-706.indd 674 12/11/10 12:57 PM

 15.3 IMPATT Diode 675

n
n0

(a)

E
Eth
J

(b)

Figure 15.6 | (a) Electron t

(t )
concentration versus distance
showing a domain. (b) Electric field Figure 15.7 | Current pulses versus
versus distance. time in the GaAs device.

The oscillation mechanism just described is called the transit-time mode. More
complex modes of operation are possible. Studies have shown that the efficiency of
the transit-time device is largest when the product n0 L is a few times 1012 cm2. For
this case, the domain fills about one-half of the drift region length and produces a
current output that is nearly sinusoidal. The maximum dc-to-rf conversion efficiency
is approximately 10 percent.
Oscillations in the frequency range of 1 to 100 GHz or higher can be obtained. If
the device is operated in a pulsed mode, a peak output power in the range of hundreds
of watts can be produced. Transferred-electron devices are now used as the micro-
wave source in many radar systems.

15.3 | IMPATT DIODE

The term IMPATT stands for IMPact ionization Avalanche Transit-Time. The
IMPATT diode consists of a high-field avalanche region and a drift region that pro-
duces a dynamic negative resistance at microwave frequencies. The negative resis-
tance characteristic produced in this device is a result of a time delay so that the ac
current and voltage components are out of phase, and is a different phenomenon
compared to the tunnel diode, for example. The tunnel diode has a negative dIdV
region in the I–V characteristic.
One example of an IMPATT diode is a p-n-i-n structure as shown in
Figure 15.8a. Typical doping concentrations (magnitudes) are shown in Figure 15.8b.
The device is reverse biased so that the n and intrinsic regions are completely de-
pleted. The electric field in the device is shown in Figure 15.8c. We may note that

Edx  VB where VB is the applied reverse-biased voltage. The value of VB is very
close to the breakdown voltage. The avalanche region is localized near the pn junc-
tion. The electric field in the intrinsic region is nearly constant and the intrinsic layer
provides the drift region.

nea29583_ch15_670-706.indd 675 12/11/10 12:57 PM

 676 CHAPTER 15 Semiconductor Microwave and Power Devices

P n i n

(a)
N (cm3)
1020

1017

1013
 v 
(b)
E
 IMPATT

VB


(c)
RL
Figure 15.8 | (a) An IMPATT
diode structure. (b) Typical doping
concentrations in the IMPATT diode.
(c) Electric field versus distance
through the IMPATT diode. Figure 15.9 | Circuit for an IMPATT diode oscillator.

Figure 15.9 shows the circuit for an IMPATT diode oscillator. An LC resonant
circuit is required for the oscillator operation. During the positive ac voltage across
the LC circuit as shown in the figure, the diode goes into breakdown and electron–
hole pairs are generated at the pn junction. The generated electrons flow back into
the p region, while the holes start drifting through the depleted intrinsic region. In
general, the holes will travel at their saturation velocity. During the negative ac volt-
age, the device operates below the breakdown voltage so electron–hole pairs are no
longer produced.
There is an inherent 2 phase shift between the peak value of the avalanche
voltage at the pn junction and the injection of the holes into the intrinsic drift re-
gion due to the finite buildup time of the avalanche generated electron–hole pairs.
A further delay of 2 is then required during the drift process to provide the total
180 degrees of phase shift between the current and voltage at the output terminal.
The transit time of the holes is   Lvs, where L is the length of the drift region and
vs is the saturation velocity of the holes. The LC circuit resonant frequency must be
designed to be equal to the device resonant frequency, which is given by
1 _ vs
f_ (15.5)
2 2L

When the holes reach the n cathode, the current is at a maximum value and the
voltage is at its minimum value. The ac current and ac voltage are 180 degrees out of
phase with respect to each other producing the dynamic negative resistance.

nea29583_ch15_670-706.indd 676 12/11/10 12:57 PM

 15.4 Power Bipolar Transistors 677

Devices can be designed to operate in the 100 GHz or higher frequency range
and produce power outputs of a few watts. The efficiency of these devices is in the
range of 10 to 15 percent, and these devices provide the highest continuous out-
put power of all the semiconductor microwave devices. As with most semiconduc-
tor device designs, other structures can be fabricated to provide specialized output
characteristics.

15.4 | POWER BIPOLAR TRANSISTORS

In our previous discussions, we have ignored any physical transistor limitations in
terms of maximum current, voltage, and power. We implicitly assumed that the tran-
sistors are capable of handling the current and voltage, and could handle the power
dissipated within the device without suffering any damage.
However, with power transistors, we must be concerned with various transistor
limitations. The limitations involve maximum rated current (on the order of am-
peres), maximum rated voltage (on the order of 100 V), and maximum rated power
(on the order of watts or tens of watts).1

15.4.1 Vertical Power Transistor Structure

Figure 15.10 shows the structure of a vertical npn power transistor. We have con-
sidered vertical npn bipolar transistors previously. However, with small switching
devices, the collector terminal is still formed at the surface. In the vertical configura-
tion for the power bipolar transistor, the collector terminal is at the “bottom” of the
device. This configuration is preferred since it maximizes the cross-sectional area
through which current is flowing in the device. In addition, the doping concentrations

Base Emitter

n  1019 cm3

Base 5 –20 ␮m p  1016 cm3

Collector
drift 50 –200 ␮m n  1014 cm3
region

200 ␮m n  1019 cm3

Collector

Figure 15.10 | Cross section of typical vertical npn power BJT.

1
We must note that, in general, the maximum rated current and maximum rated voltage cannot occur at
the same time.

nea29583_ch15_670-706.indd 677 12/11/10 12:57 PM

 678 CHAPTER 15 Semiconductor Microwave and Power Devices

Base

Emitter

(a)

Base Emitter
terminal terminal

n n n n n
p base

n collector

(b)

Figure 15.11 | An interdigitated bipolar transistor structure

showing the top view and cross-sectional view.

and dimensions are not the same as we have encountered in small switching transis-
tors. The primary collector region has a low-doped impurity concentration so that
a large base–collector voltage can be applied without initiating breakdown. An-
other n region, with a higher doping concentration, reduces collector resistance and
makes contact with the external collector terminal. The base region is also much
wider than normally encountered in small devices. A large base–collector voltage
implies a relatively large space charge width being induced in both the collector
and base regions. A relatively large base width is required to prevent punch-through
breakdown.
Power transistors must also be large-area devices in order to handle large cur-
rents. We have previously considered the interdigitated structure that is repeated in
Figure 15.11. Relatively small emitter widths are required to prevent the emitter cur-
rent crowding effects that were discussed in Section 12.4.4.

15.4.2 Power Transistor Characteristics

The relatively wide base width implies a much smaller current gain  for power
transistors compared to small switching transistors, and large area device implies a
larger junction capacitance and hence lower cutoff frequency for a power transistor
compared to a small switching transistor. Table 15.1 compares the parameters of a

nea29583_ch15_670-706.indd 678 12/11/10 12:57 PM

 15.4 Power Bipolar Transistors 679

Table 15.1 | Comparison of the characteristics and maximum ratings of small-signal

and power BJTs
Small-signal BJT Power BJT Power BJT
Parameter (2N2222A) (2N3055) (2N6078)
VCE (max) (V) 40 60 250
TC (max) (A) 0.8 15 7
PD (max) (W) 1.2 115 45
(at T  25 C)
 35–100 5–20 12–70
fT (MHz) 300 0.8 1

general-purpose small-signal BJT to those of two power BJTs. The current gain is
generally smaller in the power transistors, typically in the range of 20 to 100, and
may be a strong function of collector current and temperature. Figure 15.12 shows
typical current gain versus collector current characteristics for the 2N3055 power
BJT at various temperatures.
The maximum rated collector current IC,max may be related to the maximum cur-
rent that the wires connecting the semiconductor to the external terminals can handle,
the collector current at which the current gain falls below a minimum specified value,
or the current that leads to the maximum power dissipation when the transistor is
biased in saturation.
The maximum rated voltage in a BJT is generally associated with avalanche
breakdown in the reverse-biased base–collector junction. In the common-emitter
configuration, the breakdown voltage mechanism also involves the transistor
gain, as well as the breakdown phenomenon in the pn junction. This is discussed
in Section 12.4.6. Typical IC versus VCE characteristics are shown in Figure 15.13.

10
1000 9
Common emitter
Collector current IC (mA)

500 VCE  4 V 8
TC  100C IB  75
300 7
DC current gain hFE

25 6 60
100
5
45
50 55 4
30 3 30
2 15
10
1
Base current IB  0 ␮A
5
3 0 20 40 60 80 100 120 140
0.01 0.03 0.1 0.3 1 3 10 20 Collector-to-emitter voltage VCE (V)
Collector current IC (A)
Figure 15.13 | Typical collector current versus collector–
Figure 15.12 | Typical dc beta characteristics emitter voltage characteristics of a bipolar transistor,
(hFE versus IC) for 2N3055. showing breakdown effects.

nea29583_ch15_670-706.indd 679 12/11/10 12:57 PM

 680 CHAPTER 15 Semiconductor Microwave and Power Devices

IC (A)
IC (A)
1.0 IC, max Maximum
current limit
1.0
PT
PT VCE, sus
Second Second
0.5 breakdown 0.10 breakdown

VCE, sus
0.01
0 10 30 50 70 90 VCE (V) 1 10 100 VCE (V)
(a) (b)

Figure 15.14 | The safe operating area (SOA) of a bipolar transistor plotted on (a) linear scales and (b) logarithmic
scales.

When the transistor is biased in the forward-active mode, the collector current be-
gins to increase significantly before the actual breakdown voltage is reached. All
the curves tend to merge to the same collector–emitter voltage once breakdown has
occurred. This voltage, VCE, sus, is the minimum voltage necessary to sustain the tran-
sistor in breakdown.
Another breakdown effect is called second breakdown, which occurs in a BJT
operating at high voltage and high current. Slight nonuniformities in current den-
sity produce local regions of increased heating that increases the minority carrier
concentrations in the semiconductor material, which in turn increases the current
in these regions. This effect results in positive feedback, and the current continues
to increase, producing a further increase in temperature, until the semiconduc-
tor material may actually melt, creating a short circuit between the collector and
emitter.
The average power dissipated in a BJT must be kept below a specified maximum
value, to ensure that the temperature of the device remains below a maximum value.
If we assume the collector current and collector–emitter voltage are dc values, then
at the maximum rated power PT for the transistor, we can write
PT  VCE IC (15.6)
Equation (15.6) neglects the VBE IB component of power dissipation in the transistor.
The maximum current, voltage, and power limitations can be illustrated on the
IC versus VCE characteristics as shown in Figure 15.14. The average power limitation,
PT, is a hyperbola described by Equation (15.6). The region where the transistor can
be operated safely is known as the safe operating area (SOA) and is bounded by
IC,max, VCE, sus, PT, and the transistor’s second breakdown characteristic curve. Fig-
ure 15.14a shows the safe operating area using linear current and voltage scales.
Figure 15.14b shows the same characteristics using log scales.

nea29583_ch15_670-706.indd 680 12/11/10 12:57 PM

 15.4 Power Bipolar Transistors 681

Objective: Determine the required current, voltage, and power rating of a power BJT. EXAMPLE 15.1
Consider the common-emitter circuit in Figure 15.15. The parameters are RL  10
and
VCC  35 V. VCC

■ Solution
For VCE  0, the maximum collector current is RL
VCC _
IC (max)  _  35  3.5 A
RL 10 vO
For IC  0, the maximum collector–emitter voltage is vI
VCE (max)  VCC  35 V

The load line is given by

VCE  VCC  IC RL Figure 15.15 |
and must remain within the SOA, as shown in Figure 15.16. Bipolar common-
The transistor power dissipation is emitter circuit.

PT  VCE IC  (VCC  IC RL)IC  VCC IC  IC2 RL

The current at which the maximum power occurs is found by setting the derivative of this
equation equal to zero as follows:
dPT
_  0  VCC  2ICRL
dIC
which yields
VCC _
IC  _  35  1.75 A
2RL 2(10)

IC (A)
4

IC (max)  3.5 A
PT
3

2
1.75

1
Load
line

0 10 17.5 20 30 VCC 40
 35
VCE (V)

Figure 15.16 | Load line and maximum power curve for

Example 15.1.

nea29583_ch15_670-706.indd 681 12/11/10 12:57 PM

 682 CHAPTER 15 Semiconductor Microwave and Power Devices

The collector–emitter voltage at this maximum power point is

VCE  VCC  ICRL  35  (1.75)(10)  17.5 V

The maximum power dissipated in the transistor occurs at the center of the load line. The
maximum transistor power dissipation is therefore

PT  VCEIC  (17.5)(1.75)  30.6 W

■ Comment
To find a transistor for a given application, safety factors are normally used. For this example,
a transistor with a current rating greater than 3.5 A, a voltage rating greater than 35 V, and a
power rating greater than 30.6 W would be required for the application just described.

■ EXERCISE PROBLEM
Ex 15.1 Assume the BJT in the common-emitter circuit shown in Figure 15.15 has limit-
ing factors of IC,max  5A, VCE, sus  75 V, and PT  30 W. Neglecting second
breakdown effects, determine the minimum value of RL such that the Q-point
of the transistor always stays within the safe operating area for (a) VCC  60 V,
(b) VCC  40 V, and (c) VCC  20 V. In each case, determine the maximum collec-
tor current and maximum transistor power dissipation.
IC (max)  3 A, P(max)  30 W; (c) RL  4
, IC (max)  5 A, P(max)  25 W]
[Ans. (a) RL  30
, IC (max)  2 A, P(max)  30 W; (b) RL  13.3
,

15.4.3 Darlington Pair Configuration

As mentioned, the base width of a power BJT is relatively wide so that the current
gain is then relatively small. One method that is used to increase the effective cur-
rent gain is to use a Darlington pair such as shown in Figure 15.17. Considering the
currents, we see that
iC  iCA  iCB  AiB  BiEA  AiB  B (1  A) iB (15.7)

iC
iCA

iB iCB
QA
B iEA
QB

D1
iE

Figure 15.17 | An npn Darlington pair

configuration.

nea29583_ch15_670-706.indd 682 12/11/10 12:57 PM

 15.4 Power Bipolar Transistors 683

B E

n n
p SiO2 p

QA QB
n

n

Figure 15.18 | An integrated circuit implementation of the

npn Darlington pair configuration.

The overall common-emitter current gain is then

iC
_  AB  A  B (15.8)
iB
Thus, if the gain of each individual transistor is A  B  15, then the overall gain
of the Darlington pair is iC iB  255. This overall gain is then substantially larger
than that of the individual device. A diode may be incorporated as shown in Fig-
ure 15.17 to aid in turning off the transistor QB. A reverse current out of the base of
QB through the diode will pull charge out of the base of this transistor and turn the
device off faster than when no diode is used.
The Darlington pair shown in Figure 15.17 is typically used in the output stage
of a power amplifier when an npn bipolar transistor is required. A pnp Darlington
pair may also be used to increase the effective current gain of a power pnp device.
The integrated circuit configuration of the npn Darlington pair may be as shown
in Figure 15.18. The silicon dioxide that is shown completely penetrates through the
p-type base region so that the base regions of the two transistors are isolated.

TEST YOUR UNDERSTANDING

VCC
TYU 15.1 Consider the vertical power silicon BJT shown in Figure 15.10. Assume that
a reverse-biased voltage of 200 V is applied to the base–collector junction.
vI
Calculate the space charge width that extends into the (a) collector region and
(b) base region. vO
[Ans. (a) xn  50.6 m; (b) xp  0.506 m]
RE
TYU 15.2 For the emitter–follower circuit in Figure 15.19, the parameters are VCC  10 V
and RE  200
. The transistor current gain is   150, and the current and
voltage limitations are IC,max  200 mA and VCE,sus  50 V. Determine the mini- VCC
mum transistor power rating such that the transistor Q-point is always inside the
safe operating area. Figure 15.19 |
(Ans. Pmax  0.5 W) Figure for
Exercise TYU 15.2.

nea29583_ch15_670-706.indd 683 12/11/10 12:57 PM

 684 CHAPTER 15 Semiconductor Microwave and Power Devices

15.5 | POWER MOSFETs

The basic operation of the power MOSFET is the same as that of any MOSFET. How-
ever, the current handling capability of these devices is usually in the ampere range,
and the drain-to-source blocking voltage may be in the range of 50 to 100 volts or even
higher. One big advantage that a power MOSFET has over a bipolar power device is
that the control signal is applied to the gate whose input impedance is extremely large.
Even during switching between on and off states, the gate current is small, so that
relatively large currents can be switched with very small control currents.

15.5.1 Power Transistor Structures

Large currents can be obtained in a MOSFET with a very large channel width. To achieve
a large channel width device with good characteristics, power MOSFETs are fabricated
with a repetitive pattern of small cells operating in parallel. To achieve a large blocking
voltage, a vertical structure is used. There are two basic power MOSFET structures. The
first is called a DMOS device and is shown in Figure 15.20. The DMOS device uses a
double diffusion process: The p-base or the p-substrate region and the n source contact
are diffused through a common window defined by the edge of the gate. The p-base
region is diffused deeper than the n source, and the difference in the lateral diffusion
distance between the p-base and the n source defines the surface channel length.
Electrons enter the source terminal and flow laterally through the inversion layer
under the gate to the n-drift region. The electrons then flow vertically through the
n-drift region to the drain terminal. The conventional current direction is from the
drain to the source. The n-drift region must be moderately doped so that the drain
breakdown voltage is sufficiently large. However, the thickness of the n-drift region
should also be as thin as possible to minimize drain resistance.
The second power MOSFET structure, shown in Figure 15.21, is a VMOS struc-
ture. The vertical channel or VMOS power device is a nonplanar structure that requires
a different type of fabrication process. In this case, a p-base or p-“substrate” diffusion

Gate
Source Source
Gate
Source Source
n p base p n
Channel n p base p n

Channel
n-drift region
n-drift region

n
n
Drain
Drain
Figure 15.20 | Cross section of a
double-diffused MOS (DMOS) Figure 15.21 | Cross section of a vertical
transistor. channel MOS (VMOS) transistor.

nea29583_ch15_670-706.indd 684 12/11/10 12:57 PM

 15.5 Power MOSFETs 685

Multiple source cells Silicon gate

interconnected by Multiple
metallization source
cells

W n

nt
p u rre
ai nc
n Dr

Drain
metallization

Figure 15.22 | A HEXFET structure.

is performed over the entire surface followed by the n source diffusion. A V-shaped
groove is then formed, extending through the n-drift region. It has been found that
certain chemical solutions etch the (111) planes in silicon at a much slower rate than
the other planes. If (100) oriented silicon is etched through a window at the surface,
these chemical etches will create a V-shaped groove. A gate oxide is then grown in the
V-shaped groove and the metal gate material is deposited. An electron inversion layer
is formed in the base or substrate so that current is again essentially a vertical current
between the source and the drain. The relatively low-doped n-drift region supports the
drain voltage since the depletion region extends mainly into this low-doped region.
We mentioned that many individual MOSFET cells are connected in parallel
to fabricate a power MOSFET with the proper width-to-length ratio. Figure 15.22
shows a HEXFET structure. Each cell is a DMOS device with an n polysilicon gate.
The HEXFET has a very high packing density—it may be on the order of 105 cells
per cm2. In the VMOS structure, the anisotropic etching of the grooves must be along
the [110] direction on the (100) surface. This constraint limits the design options
available for this type of device.

15.5.2 Power MOSFET Characteristics

Table 15.2 lists the basic parameters of two n-channel power MOSFETs. The drain
currents are in the ampere range and the breakdown voltages are in the hundreds of
volts range.
An important parameter of a power MOSFET is the on resistance, which can be
written as
Ron  RS  RCH  RD (15.9)
where RS is the resistance associated with the source contact, RCH is the channel
resistance, and RD is the resistance associated with the drain contact. The RS and RD

nea29583_ch15_670-706.indd 685 12/11/10 12:57 PM

 686 CHAPTER 15 Semiconductor Microwave and Power Devices

Table 15.2 | Characteristics of two power MOSFETs

Parameter 2N6757 2N6792
VDS (max) (V) 150 400
ID (max) (at T  25°C) 8 2
PD (W) 75 20

resistance values are not necessarily negligible in power MOSFETs since small re-
sistances and high currents can produce considerable power dissipation.
In the linear region of operation, we may write the channel resistance as
RCH  ___ L (15.10)
W nCox (VGS  VT)
We have noted in previous chapters that mobility decreases with increasing tempera-
ture. The threshold voltage varies only slightly with temperature so that, as current in
a device increases and produces additional power dissipation, the temperature of the
device increases, the carrier mobility decreases, and RCH increases, which inherently
limits the channel current. The resistances RS and RD are proportional to semiconduc-
tor resistivity and so are also inversely proportional to mobility and have the same
temperature characteristics as RCH. Figure 15.23 shows a typical “on-resistance”
characteristic as a function of drain current.
The increase in resistance with temperature provides stability for the power
MOSFET. If the current in any particular cell begins to increase, the resulting tem-
perature rise will increase the resistance, thus limiting the current. With this particu-
lar characteristic, the total current in a power MOSFET tends to be evenly distributed
among the parallel cells, not concentrated in any single cell, a condition that can
cause burnout.

1.0
VGS  10 V
RDS(on), drain-to-source on resistance ()

0.9 VGS  20 V

0.8

0.7

0.6

0.5 RDS(on) measured with current

pulse of 2.0-␮s duration,
0.4 initial Tj  25C (heating
effect of 2.0-␮s pulse is minimal)
0.3
0 10 20 30 40 50 60 70
ID, drain current (A)

Figure 15.23 | Typical drain-to-source resistance versus

drain current characteristics of a MOSFET.

nea29583_ch15_670-706.indd 686 12/11/10 12:57 PM

 15.5 Power MOSFETs 687

10 10
80-␮s pulse test
VDS  15 V
Transconductance (siemens)
8 TJ  55C 8

Drain current ID (A)

TJ  25C
6 TJ  125C 6

TJ  125C
4 4
TJ  25C
TJ  55C
2 2
VDS  15 V
80-␮s pulse test
0 2 4 6 8 10 0 1 2 3 4 5 6 7
Drain current ID (A) Gate-to-source voltage (V)
(a) (b)

Figure 15.24 | Typical characteristics for high-power MOSFETs at various

temperatures: (a) transconductance versus drain current; (b) drain current versus
gate-to-source voltage.

Power MOSFETs differ from bipolar power transistors in both operating principles
and performance. The superior performance characteristics of power MOSFETs are
faster switching times, no second breakdown, and stable gain and response time over
a wide temperature range. Figure 15.24a shows the transconductance of the 2N6757
versus temperature. The variation with temperature of the MOSFET transconductance
is less than the variation in the BJT current gain that is shown in Figure 15.12.
Figure 15.24b is a plot of drain current versus gate-to-source voltage at three different
temperatures. We may note that at high current, the current decreases with temperature
at a constant gate-to-source voltage, providing the stability that has been discussed.
Power MOSFETs must operate in a SOA. As with power BJTs, the SOA is
defined by three factors: the maximum drain current, ID,max, rated breakdown voltage,
BVDSS, and the maximum power dissipation given by PT  VDSID. The SOA is shown
in Figure 15.25a in which the current and voltage are plotted on linear scales. The

ID (A) log ID
ID, max log ID, max

PT

PT

0 BVDSS 0 log BVDSS

VDS (V) log VDS
(a) (b)

Figure 15.25 | The safe operating area (SOA) of a MOSFET plotted on (a) linear scales and
(b) logarithmic scales.

nea29583_ch15_670-706.indd 687 12/11/10 12:57 PM

 688 CHAPTER 15 Semiconductor Microwave and Power Devices

same SOA curve is shown in Figure 15.25b in which the current and voltage are
plotted on log scales.

EXAMPLE 15.2 Objective: Find the optimum drain resistor in a MOSFET inverter circuit.
A MOSFET inverter circuit is shown in Figure 15.26. Two different MOSFETs are being
considered for use in the circuit. The parameters for devices A and B are given.
VDD
Device A Device B
ID RD BVDSS  35 V BVDSS  35 V
PT  30 W PT  30 W
vO ID,max  6 A ID,max  4 A

vI VDS ■ Solution

The SOA curves for the two devices are shown in Figure 15.27.
The load line for the inverter circuit using device A is shown as curve A. The load line
intersects the voltage axis at VDD  24 V. This curve is tangent to the maximum power curve
Figure 15.26 | A and intersects the current axis at ID  5 A. Note that, if we had wanted the load line to intersect
MOSFET inverter the maximum rated current of ID, max  6 A, the load line would have gone outside of the SOA.
circuit. For the load line A, the drain resistance is
VDD 24  4.8

RD  _ _
ID 5

6 ID, max  6 A

5
PT

4 ID, max  4 A
A
ID (A)

B
3
Maximum power
dissipated

VDD  24 V

0 5 10 15 20 25 30 35  BVDSS
VDS (V)

Figure 15.27 | Safe operating area and load lines for devices in Example 15.2.

nea29583_ch15_670-706.indd 688 12/11/10 12:57 PM

 15.5 Power MOSFETs 689

The current at the maximum power point (using the results from Example 15.1) is
VDD 24  2.5 A
ID  _  __
2RD 2(4.8)
and the corresponding drain-to-source voltage is

VDS  VDD  IDRD  24  (2.5)(4.8)  12 V

The maximum power that may be dissipated in the transistor is P  VDS ID  (12)(2.5) 
30 W  PT, which corresponds to the maximum rated power. This point is shown on the curve.
The load line for the inverter circuit using device B is shown as curve B. The load line
intersects the voltage axis at VDD  24 V as before. This curve can now intersect the current
axis at the maximum rated drain current of ID, max  4 A. We see that the load line falls within
the SOA of the transistor.
For load line B, the drain resistance is
VDD _
RD  _  24  6

ID 4
The current at the maximum power point is
VDD 24  2 A
ID  _ _
2RD 2(6)
and the corresponding drain-to-source voltage is

VDS  VDD  IDRD  24  (2)(6)  12 V

The maximum power that may be dissipated in the transistor is P  VDD ID  (12)(2)  24 W,
which is less than the maximum rated power. This point is also shown on the curve.

■ Conclusion
We see that if device A is used, the drain resistor is determined by the maximum power.
However, if device B is used, the drain resistor is determined by the maximum rated current
of the device.

■ EXERCISE PROBLEM
Ex 15.2 Consider the common-source circuit shown in Figure 15.26. Determine the
required current, voltage, and power ratings of the MOSFET for (a) RD  12
,
VDD  24 V and (b) RD  8
, VDD  40 V.
PT  50 W]
[Ans. (a) BVDSS  24 V, ID, max  2 A, PT  12 W; (b) BVDSS  40 V, ID, max  5 A,

15.5.3 Parasitic BJT

The MOSFET has a parasitic BJT as an inherent part of its structure. The parasitic
BJT may be seen in both the DMOS and VMOS structures shown in Figures 15.20
and 15.21. The source terminal corresponds to the n-type emitter, the p-type base or
substrate region corresponds to the p-type base, and the n-type drain corresponds to
the n-type collector. This is also shown schematically in Figure 15.28. The channel
length of the MOSFET corresponds to the base width of the parasitic BJT. Since this
length is normally quite small, the current gain  of the BJT can be larger than unity.

nea29583_ch15_670-706.indd 689 12/11/10 12:57 PM

 690 CHAPTER 15 Semiconductor Microwave and Power Devices

Gate
Source

n n
p Cgd p

n
n

Drain
(a)

Cgd
Parasitic
G BJT

S
(b)

Figure 15.28 | (a) Cross section of vertical MOSFET showing parasitic BJT and
distributed resistance; (b) equivalent circuit of MOSFET and parasitic BJT with
distributed parameters.

The BJT should be cutoff at all times, which means the source-to-body volt-
age (emitter-to-base voltage) should be as close to zero as possible. We see from
the geometries of Figures 15.20 and 15.21 that the source ohmic contact also goes
across the p-type body region so that this junction voltage is zero during steady-state
operation of the transistor. However, the BJT may be turned on during high-speed
switching of the MOSFET.
Figure 15.28b shows that the base and the collector of the parasitic BJT are
connected by the gate-to-drain capacitance. A parasitic or distributed resistance also
connects the base to the emitter of the BJT. When the MOSFET is being turned
off, the drain-to-source voltage increases and induces a current in the gate-to-drain
capacitance in the direction from the parasitic collector terminal to the parasitic
base terminal. This induced current may be large enough to induce a voltage in
the parasitic resistance that is sufficient to forward bias the base–emitter junction
and therefore turn the BJT on. The turned on BJT may then induce a large drain
current that can cause burnout of the MOSFET. This breakdown mechanism is
known as snapback breakdown and has been discussed briefly in Section 11.4.1.

nea29583_ch15_670-706.indd 690 12/11/10 12:57 PM

 15.6 The Thyristor 691

The current–voltage characteristics are shown in Figure 11.22. Devices can be de-
signed to minimize the parasitic or distributed base–emitter resistance to minimize
this problem.

15.6 | THE THYRISTOR

One of the important applications of electronic devices is in switching between an
off or blocking state to an on or low-impedance state. Thyristor is the name given
to a general class of semiconductor pnpn switching devices that exhibit bistable
regenerative switching characteristics. We have considered the transistor, which
may be switched on with the application of a base drive or a gate voltage. The
base drive or gate voltage must be applied as long as the transistor is to remain on.
There are a number of applications in which it is useful to have a device remain
in a blocking state until switched to the low-impedance state by a control signal,
which then does not necessarily have to remain on. These devices are efficient
in switching large currents at low frequencies, such as industrial control circuits
operating at 60 Hz.
A Semiconductor Controlled Rectifier (SCR) is the common name given to a
three-terminal thyristor. The SCR (sometimes referred to as a silicon controlled rec-
tifier) is a four-layer pnpn structure with a gate control terminal. As with most semi-
conductor devices, there are several variations of the device structure. We consider
the basic SCR operation and limitations, and then discuss some variations of the
basic four-layer device.

15.6.1 The Basic Characteristics

The four-layer pnpn structure is shown in Figure 15.29a. The upper p region is called
the anode and the lower n region is called the cathode. If a positive voltage is applied
to the anode, the device is said to be forward biased. However, the junction J2 is
reverse biased so that only a very small current exists. If a negative voltage is applied
to the anode, then junctions J1 and J3 are reverse biased—again only a very small
current will exist. Figure 15.29b shows the I–V characteristics for these conditions.
The voltage Vp is the breakdown voltage of the J2 junction. For properly designed
devices, the blocking voltage can be several thousand volts.
To consider the characteristics of the device as it goes into its conducting state,
we can model the structure as coupled npn and pnp bipolar transistors. Figure 15.30a
shows how we can split the four-layer structure and Figure 15.30b shows the two-
transistor equivalent circuit with the associated currents. Since the base of the pnp
device is the same as the collector of the npn transistor, the base current IB1 must in
fact be the same as the collector current IC2. Similarly, since the collector of the pnp
transistor is the same as the base of the npn device, the collector current IC1 must
be the same as the base current IB2. In this bias configuration, the B–C of the pnp
and the B–C of the npn devices are reverse biased, while the B–E junctions are both
forward biased. The parameters
1 and
2 are the common base current gains of the
pnp and npn transistors, respectively.

nea29583_ch15_670-706.indd 691 12/11/10 12:57 PM

 692 CHAPTER 15 Semiconductor Microwave and Power Devices

VA
IA
A
IA Anode

p
n
J1

n p
⫹
J2 VA
⫺ n
p
J3 IK
n

(a)

Cathode VA
K IA
(a)

p
IB1 ⫽ IC2
IA
n n

p p
IC1 ⫽ IB2
n
Vp

Reverse Forward VA IK
blocking blocking

(b) (b)

Figure 15.29 | (a) The basic four-layer Figure 15.30 | (a) The splitting of the
pnpn structure. (b) The initial current– basic pnpn structure. (b) Two two-
voltage characteristic of the pnpn transistor equivalent circuit of the
device. four-layer pnpn device.

We can write

IC1 ⫽ ␣1 IA ⫹ IC01 ⫽ IB2 (15.11a)

and

IC2 ⫽ ␣2 IK ⫹ IC02 ⫽ IB1 (15.11b)

nea29583_ch15_670-706.indd 692 12/13/10 1:53 PM

 15.6 The Thyristor 693

where IC01 and IC02 are the reverse B–C junction saturation currents in the two de-
vices. In this particular configuration, IA = IK and IC1 + IC2  IA. If we add Equa-
tions (15.11a) and (15.11b), we obtain
IC1  IC2  IA  (
1 
2) IA  IC01  IC02 (15.12)
The anode current IA, from Equation (15.12), can be found as
IC01  IC02
IA  ___ (15.13)
1  (
1 
2)
As long as (
1 
2) is much smaller than unity, the anode current is small, as we
have indicated in Figure 15.29b.
The common base current gains,
1 and
2, are very strong functions of collector
current as we discussed in Chapter 12. For small values of VA, the collector current in
each device is just the reverse saturation current, which is very small. The small col-
lector current implies that both
1 and
2 are much smaller than unity. The four-layer
structure maintains this blocking condition until the junction J2 starts into breakdown
or until a current is induced in the J2 junction by some external means.
Consider, initially, the condition when the applied anode voltage is sufficiently
large to cause the J2 junction to start into avalanche breakdown. This effect is shown
in Figure 15.31a. The electrons generated by impact ionization are swept into the
n1 region, making the n1 region more negative, and the holes generated by impact
ionization are swept into the p2 region, making the p2 region more positive. The more
negative voltage of the n1 region and the more positive voltage of the p2 region means
that the forward-bias junction voltages V1 and V3 both increase. The increase in the
respective B–E junction voltages causes an increase in current, which results in an
increase in the common-base current gains
1 and
2, causing a further increase in

 V1   V3 

e
VA p1 n1 p2 n2
IA h
J1 J2 J3
(a)

 V1   V2   V3 

VA p1 n1 p2 n2
IA
J1 J2 J3
(b)

Figure 15.31 | (a) The pnpn device when the J2 junction

starts into avalanche breakdown. (b) The junction voltages
in the pnpn structure when the device is in the high-current,
low-impedance state.

nea29583_ch15_670-706.indd 693 12/11/10 12:57 PM

 694 CHAPTER 15 Semiconductor Microwave and Power Devices

VA
IA VA
IA

p1
p1
IB1 ⫽ IC2
Forward n1
IA conducting n1 n1
IC1 IB2
G p2
p2 p2
Ig
n2 n2

Vp G
IK IK
Ig
Reverse Forward VA
blocking blocking
(a) (b)

Figure 15.32 | The current–voltage Figure 15.33 | (a) The three-terminal SCR. (b) The two-transistor
characteristics of the pnpn device. equivalent circuit of the three-terminal SCR.

IA as seen in Equation (15.13). We now have a regenerative positive feedback situa-

tion, so the current IA will increase very rapidly.
As the anode current IA increases and ␣1 ⫹ ␣2 increases, the two equivalent bipo-
lar transistors are driven into saturation and the junction J2 becomes forward biased.
The total voltage across the device decreases and is approximately equal to one diode
drop as shown in Figure 15.31b. The current in the device is limited by the external
circuit. If the current is allowed to increase, ohmic losses may become important
so that the voltage drop across the device may increase slightly with current. The
IA versus VA characteristic is shown in Figure 15.32.

15.6.2 Triggering the SCR

In the last section, we considered the case when the four-layer pnpn device is turned
on by the avalanche breakdown process in the center junction. The turn-on condition
can also be initiated by other means. Figure 15.33a shows three-terminal SCR in
which the third terminal is the gate control. We can determine the effect of the gate
current by reconsidering Equations (15.11a) and (15.11b).
Figure 15.33b again shows the two-transistor equivalent circuit including the
gate current. We can write
IC1 ⫽ ␣1IA ⫹ IC01 (15.14a)
and
IC2 ⫽ ␣2IK ⫹ IC02 (15.14b)
We now have IK ⫽ IA ⫹ Ig and we can still write IC1 ⫹ IC2 ⫽ IA. Adding Equa-
tions (15.14a) and (15.14b), we find that
IC1 ⫹ IC2 ⫽ IA ⫽ (␣1 ⫹ ␣2)IA ⫹ ␣2Ig ⫹ IC01 ⫹ IC02 (15.15)

nea29583_ch15_670-706.indd 694 12/13/10 1:53 PM

 15.6 The Thyristor 695

Solving for IA, we find

2Ig  (IC01  IC02)
IA  ____ (15.16)
1  (
1 
2)
We can think of the gate current as the flow of holes into the p2 region.
The additional holes increase the potential of this region, which increases the
forward-biased B–E voltage of the npn bipolar transistor, and the transistor ac-
tion. The transistor action of the npn increases the collector current IC2, which starts
the transistor action of the pnp bipolar transistor, and the entire pnpn device can
be turned on into its low-impedance state. The gate current required to switch the
SCR into its on condition is typically in the milliamp range. SCR can be turned on
with a small gate current, which can control hundreds of amperes of anode current.
The gate current can be turned off and the SCR will remain in its conducting state.
The gate loses control of the device once the SCR is triggered into its conducting
state. The current–voltage characteristics of the SCR as a function of gate current is
shown in Figure 15.34.
A simple application of an SCR in a half-wave control circuit is shown in Fig-
ure 15.35a. The input signal is an ac voltage and a trigger pulse will control the
turn-on of the SCR. We assume that the trigger pulse occurs at time t1 during the
ac voltage cycle. Prior to t1, the SCR is off so that the current in the load is zero; thus,
there is a zero output voltage. At t  t1, the SCR is triggered on and the input voltage
appears across the load (neglecting the voltage drop across the SCR). The SCR turns
off when the anode-to-cathode voltage becomes zero even though the trigger pulse
has been turned off prior to this time. The time at which the SCR is triggered dur-
ing the voltage cycle can be varied, changing the amount of power delivered to the
load. Full-wave control circuits can be designed to increase efficiency and degree of
control.
The gate allows control of the turn-on of the SCR. However, the four-layer pnpn
structure can also be triggered on by other means. In many integrated circuits, para-
sitic pnpn structures exist. One such example is the CMOS structure that we con-
sidered in Chapter 10. A transient ionizing radiation pulse can trigger the parasitic

IA

Ig3
Ig2
Ig1 Ig  0
Ig  0

0 VAK
Ig1
Ig3
Ig2

0  Ig1  Ig2  Ig3

Figure 15.34 | Current–voltage characteristics of an SCR.

nea29583_ch15_670-706.indd 695 12/11/10 12:57 PM

 696 CHAPTER 15 Semiconductor Microwave and Power Devices

SCR
A C
Gate

⫹ ⫹
Trigger RL Vout
Vac
pulse ⫺
⫺
AC voltage

(a)

Trigger pulse

Vac

t1 t1⬘

(b)

Vout

t1 t1⬘
Vac

(c)

Figure 15.35 | (a) Simple SCR circuit. (b) Input ac voltage

signal and trigger pulse. (c) Output voltage versus time.

four-layer device by generating electron–hole pairs, particularly in the J2 junction,

producing a photocurrent. The photocurrent is equivalent to a gate current in an SCR
so the parasitic device can be switched into its conducting state. Again, once the
device is switched on, it will remain in its conducting state even when the radiation
ceases. An optical signal can also trigger the device in the same manner by generat-
ing electron–hole pairs.
Another triggering mechanism in the pnpn device is dV 兾dt triggering. If the
forward-bias anode voltage is applied rapidly, the voltage across the J2 junction will
also change quickly. This changing reverse-biased J2 junction voltage means that the
space charge region width is increasing; thus, electrons are being removed from the
n1 side of the junction and holes are being removed from the p2 side of the junction.

nea29583_ch15_670-706.indd 696 12/13/10 1:53 PM

 15.6 The Thyristor 697

If dV dt is large, the rate of removal of these carriers is rapid, which leads to a large
transient current that is equivalent to a gate current and can trigger the device into a
low-impedance conducting state. In SCR devices, a dV dt rating is usually specified.
However, in parasitic pnpn structures, the dVdt triggering mechanism is a potential
problem.

15.6.3 SCR Turn-Off

Switching the four-layer pnpn structure from its conducting state to its blocking
state can be accomplished if the current IA is reduced below the value creating the

1 
2  1 condition. This critical IA current is called the holding current. If a para-
sitic four-layer structure is triggered into the conducting state, the effective anode cur-
rent in the device must be reduced below the corresponding holding current in order to
turn off the device. This requirement essentially implies that all power supplies must
be turned off in order to bring the parasitic device back into its blocking state.
The SCR can be triggered on by supplying holes to the p2 region of the device.
The SCR can perhaps be turned off by removing holes from this same region. If the
reverse gate current is large enough to bring the npn bipolar transistor out of satura-
tion, then the SCR can be switched from the conducting state into the blocking state.
However, the lateral dimensions of the device may be large enough so that nonuni-
form biasing in the J2 and J3 junctions occurs during a negative gate current and
the device will remain in the low-impedance conducting state. The four-layer pnpn
device must be specifically designed for a turn-off capability.

15.6.4 Device Structures

Many thyristor structures have been fabricated with specific characteristics for spe-
cific applications. We consider a few of these types of device to gain an appreciation
for the variety of structures.
Basic SCR There are many variations of diffusion, implantation, and epitaxial
growth that can be used in the fabrication of the SCR device. The basic structure is
shown in Figure 15.36. The p1 and p2 regions are diffused into a fairly high resistivity
n1 material. The n cathode is formed and the p gate contact is made. High thermal
conductivity materials can be used for the anode and cathode ohmic contacts to aid in
heat dissipation for high-power devices. The n1 region width may be on the order of
250 m in order to support very large reverse-biased voltages across the J2 junction.
The p1 and p2 regions may be on the order of 75 m wide, while the n and p regions
are normally quite thin.
Bilateral Thyristor Since thyristors are often used in ac power applications, it may
be useful to have a device that switches symmetrically in the positive and negative
cycles of the ac voltage. There are a number of such devices, but the basic concept
is to connect two conventional thyristors in antiparallel as shown in Figure 15.37a.
The integration of this concept into a single device is shown in Figure 15.37b. Sym-
metrical n regions can be diffused into a pnp structure. Figure 15.37c shows the
current–voltage characteristics in which the triggering into the conduction mode

nea29583_ch15_670-706.indd 697 12/11/10 12:57 PM

 698 CHAPTER 15 Semiconductor Microwave and Power Devices

n2 p1
p2
n2
p1
p2
n1
n1
n1 n1
p2
p1 n2
p2
p1
n2

(a) (b)

I
K G

n p
J3 p2 V
J2

n1

J1
p1
(c)

Figure 15.37 | (a) The antiparallel connection of two

A thyristors to form a bilateral device. (b) The bilateral
thyristor as an integrated device. (c) The current–
Figure 15.36 | The basic SCR device voltage characteristics of the bilateral thyristor.
structure. (From Ghandhi [7].)

would be due to breakdown triggering. The two terminals alternately share the role
of anode and cathode during successive half cycles of the ac voltage.
Triggering by a gate control is more complex for this device since a single gate
region must serve for both of the antiparallel thyristors. One such device is known as a
triac. Figure 15.38a shows the cross section of such a device. This device can be trig-
gered into conduction by gate signals of either polarity and with anode-to-cathode
voltages of either polarity.

nea29583_ch15_670-706.indd 698 12/11/10 12:57 PM

 15.6 The Thyristor 699

1 G
(IG)

n2 n3
J3 p2 J4
J2

n1

J1
p1 A B
n4 J5

I
2
(a)

1 G
 (IG)
I
n2 n3
J3 p2 J4 IG
J2

n1

J1 V12
p1 A B
n4 J5

 I IG
2
(b)

Figure 15.38 | (a) The triac device. (b) The

triac with a specific bias configuration. Figure 15.39 | The current–voltage
(From Ghandhi [7].) characteristics of the triac.

One particular gate control situation is shown in Figure 15.38b. Terminal 1 is

positive with respect to terminal 2, and a negative gate voltage is applied with respect
to terminal 1, so the gate current is negative. This polarity arrangement induces the
current I1 and the junction J4 becomes forward biased. Electrons are injected from
n3, diffuse across p2, and are collected in the n1 region. In this case n3 p2 n1 behaves
like a saturated transistor. The collected electrons in n1 lower the potential of n1 with
respect to p2. The current across the p2n1 junction increases, which can trigger the
p2 n1 p1 n4 thyristor into its conducting mode.
We can show that the other combinations of gate, anode, and cathode volt-
ages will also trigger the triac into conduction. Figure 15.39 shows the terminal
characteristics.

nea29583_ch15_670-706.indd 699 12/11/10 12:57 PM

 700 CHAPTER 15 Semiconductor Microwave and Power Devices

Cathode
Gate
n p base W␮ n p base

Depletion
layer edge n-drift region

p anode

Anode

Figure 15.40 | The V groove MOS gated thyristor.

(From Baliga [1].)

MOS Gated Thyristor The operation of a MOS gated thyristor is based upon
controlling the gain of the npn bipolar transistor. Figure 15.40 shows a V-groove
MOS gated thyristor. The MOS gate structure must extend into the n-drift region. If
the gate voltage is zero, the depletion edge in the p-base remains essentially flat and
parallel to the junction J2; the gain of the npn transistor is low. This effect is shown in
the figure by the dashed line. When a positive gate voltage is applied, the surface of
the p base becomes depleted—the depletion region in the p base adjacent to the gate
is shown by the dotted line. The undepleted base width W of the npn bipolar device
narrows and the gain of the device increases.
At a gate voltage approximately equal to the threshold voltage, electrons from
the n emitter are injected through the depletion region into the n-drift region. The
potential of the n-drift region is lowered, which further forward biases the p anode
to n-drift junction voltage, and the regenerative process is initiated. The gate volt-
age required to initiate turn-on is approximately the threshold voltage of the MOS
device. One advantage of this device is that the input impedance to the control ter-
minal is very high; relatively large currents can be switched with very small capacity
coupled gate currents.

MOS Turn-Off Thyristor The MOS turn-off thyristor can both turn on and turn
off the anode current by applying a signal to a MOS gate terminal. The basic device
structure is shown in Figure 15.41. By applying a positive gate voltage, the npn
bipolar transistor can be turned on as just discussed. Once the thyristor is turned on,
the device can be turned off by applying a negative gate voltage: the negative gate
voltage turns on the p-channel MOS transistor that effectively short circuits the B–E
junction of the npn bipolar transistor. Holes that now enter the p-base have an alter-
native path to the cathode. If the resistance of the p-channel MOSFET becomes low
enough, all current will be diverted away from the np emitter and the npn device
will effectively be turned off.

nea29583_ch15_670-706.indd 700 12/11/10 12:57 PM

 15.7 Summary 701

Gate Cathode Anode

p n n
n p pnp
p
Turn-on
Turn-off gate
gate

n
Main current
flow path npn

Gate
p

Anode Cathode
(a) (b)

Figure 15.41 | (a) The MOS turn-off thyristor. (b) Equivalent circuit for the MOS
turn-off thyristor.
(From Baliga [1].)

15.7 | SUMMARY
■ The concept of a negative differential resistance in the I–V characteristic of the tunnel
diode is used in the design of a microwave tunnel diode oscillator. The expression for
the maximum resistance cutoff frequency is derived.
■ The operation of a microwave GUNN diode oscillator is based on the concept of
negative differential mobility.
■ The IMPATT diode oscillator uses injection and drift time delays to create a region of
differential negative resistance.
■ The power BJT has a vertical configuration and an interdigitated base–emitter surface
structure. The collector drift region (doping and width) determines the rated blocking
voltage of the BJT, while the base width must be sufficiently wide to avoid punch-
through breakdown at the rated blocking voltage.
■ A power BJT is characterized by the maximum rated collector current, maximum rated
voltage, and maximum rated power dissipation. These three parameters define the SOA
of the transistor.
■ A power MOSFET has a vertical configuration and an interdigitated gate–source sur-
face structure. Two specific devices considered are the DMOS and VMOS structures.
The drain-drift region (doping and width) determines the rated blocking voltage of
the MOSFET, while the channel length of the base (body) must be sufficiently wide to
avoid punch-through breakdown at the rated blocking voltage.
■ A power MOSFET is characterized by the maximum rated drain current, maximum
rated voltage, and maximum rated power dissipation. These three parameters define the
SOA of the transistor.

nea29583_ch15_670-706.indd 701 12/11/10 12:57 PM

 702 CHAPTER 15 Semiconductor Microwave and Power Devices

■ The “on resistance” of a MOSFET has a positive temperature coefficient so that the
power MOSFET is more stable versus temperature than a power BJT. This charac-
teristic allows MOSFETs to be fabricated in parallel to increase the current capability
of the device.
■ The thyristor refers to a general class of pnpn switching devices that can be switched
between a high-impedance, low-current state and a low-impedance, high-current state.
These devices exhibit a bistable regenerative positive feedback switching characteristic.
■ The basic pnpn device can be modeled as coupled npn and pnp bipolar transistors. In
the “on” state, both bipolar transistors are driven into saturation, creating the high-
current, low-voltage condition. In the “off” or blocking state, large voltages can be
applied to the device and the current is essentially zero.
■ The turn-on characteristics of the thyristor can be controlled through a gate control
terminal. The three-terminal thyristors are referred to as semiconductor controlled recti-
fiers (SCRs).

GLOSSARY OF IMPORTANT TERMS

double-diffused MOSFET (DMOS) A power MOSFET in which the source and channel
regions are formed using a double diffusion process.
HEXFET The structure of a power MOSFET in which many individual MOSFETs are
placed in parallel in a hexagonal configuration.
maximum rated current The maximum allowed current in a power transistor such that
proper operation is maintained.
maximum rated power The maximum allowed power dissipation in a power transistor
such that no permanent damage is done to the transistor.
maximum rated voltage The maximum allowed applied voltage to a power transistor such
that breakdown is not initiated.
negative differential mobility A region in the drift velocity versus electric field character-
istic of a semiconductor material in which the drift velocity decreases with an increase in
the electric field.
negative differential resistance A region in the I–V characteristic of a device in which the
current decreases while the voltage increases.
on resistance The effective resistance between source and drain of a power MOSFET.
safe operating area The allowed current–voltage regions of operation for a power transistor
bounded by the maximum rated current, maximum rated voltage, and maximum power.
second breakdown A breakdown effect in a power BJT in which high temperature causes a
thermal runaway process.
SCR (semiconductor controlled rectifier) The common name given to a three-terminal
thyristor.
thyristor The name given to a general class of semiconductor pnpn switching devices ex-
hibiting bistable regenerative switching characteristics.
transferred-electron effect The phenomenon in which conduction electrons are scattered
from a lower energy, high-mobility band into a higher energy, low-mobility band.
triac The name of a bilateral three-terminal thyristor.
V-groove MOSFET (VMOS) A power MOSFET in which the channel region is formed
along a V-shaped groove formed in the surface of the semiconductor.

nea29583_ch15_670-706.indd 702 12/11/10 12:57 PM

 Problems 703

CHECKPOINT
After studying this chapter, the reader should have the ability to:
■ Explain how a region of negative differential resistance is developed in the I–V charac-
teristic of the tunnel diode.
■ Discuss the concept of negative differential mobility in GaAs and discuss how this phe-
nomenon leads to the generation of domains in a GUNN diode.
■ Discuss the operation of an IMPATT diode oscillator.
■ Sketch the cross section of a power BJT and discuss the voltage and current limitations
of the device.
■ Discuss the reason the current gain of a power BJT is generally smaller than that of a
small switching BJT.
■ Sketch the safe operating area of a power BJT.
■ Describe the reason for and the operation of a Darlington configuration.
■ Sketch the cross section of the DMOS and VMOS power MOSFET structures.
■ Sketch the safe operating area of a power MOSFET.
■ Describe why the “on resistance” of a power MOSFET has a positive temperature coefficient.
■ Describe the switching characteristics of a pnpn device.
■ Describe the switching characteristics of a semiconductor controlled rectifier.

REVIEW QUESTIONS
1. Describe how a negative differential resistance region in the I–V characteristic of the
tunnel diode is generated.
2. Describe how a negative differential mobility region in the drift velocity versus electric
field characteristic in GaAs is developed.
3. Describe how a negative differential resistance characteristic is produced in the
IMPATT diode.
4. Why is the doping concentration in the collector drift region low and why is the drift
region width large in a power BJT?
5. Why does a power BJT have an interdigitated base–emitter structure?
6. Sketch the safe operating area of a power BJT.
7. Discuss how a DMOS structure of a power MOSFET is formed.
8. Discuss the voltage limitation of a power MOSFET.
9. Define the “on resistance” of a power MOSFET and show that the on resistance has a
positive temperature coefficient.
10. Discuss how the gate terminal of a semiconductor controlled rectifier can control the
switching characteristics.

PROBLEMS
Section 15.1 Tunnel Diode
15.1 Sketch the energy band diagrams of a tunnel diode in which both the n and p
regions are degenerately doped for the case of (a) zero bias, (b) 0  V  Vp,
(c) Vp  V  Vv, and (d) V  Vv.

nea29583_ch15_670-706.indd 703 12/11/10 12:57 PM

 704 CHAPTER 15 Semiconductor Microwave and Power Devices

15.2 The parameters in Figure 15.1b are Ip  20 mA, Iv  2 mA, Vp  0.15 V, and
Vv  0.60 V. Assuming a straight-line approximation to the I–V characteristics
between these two points, calculate the value of differential negative resistance.
15.3 For values of Rmin  10
, Rp  1
, and Cj  2 nF, determine the maximum
resistance cutoff frequency of a tunnel diode.

Section 15.2 GUNN Diode

15.4 (a) A GaAs transferred-electron device has a doping concentration of
Nd  1015 cm3. Determine (i) the minimum device length, (ii) the time between
current pulses, and (iii) the oscillation frequency (assume vd  1.5  107 cm/s).
(b) Repeat part (a) for a doping concentration of Nd  1016 cm3.
15.5 The drift region length of a GUNN diode is L  15 m. The voltage across the
diode oscillates between 8 and 10 V. (a) Determine the average electric field in the
device. (b) Using Figure 15.3, find the average electron drift velocity. (c) Using the
results of part (b), find the frequency of oscillation.

Section 15.3 IMPATT Diode

15.6 Find the frequency of oscillation of a silicon IMPATT diode with a drift region
length of L  10 m.

Section 15.4 Power Bipolar Transistors

15.7 Consider the vertical npn power bipolar transistor shown in Figure 15.10.
The doping concentrations are NE  1018 cm3, NB  8  1015 cm3, and
NC  6  1014 cm3. The neutral base width is 2 m, the electron diffusion
coefficient in the base is DB  20 cm2/s, and the B–E cross-sectional area is
0.4 cm2. (a) The excess electron concentration in the base at the edge of the B–E
junction is np (0)  1014 cm3. Determine (i) the B–E voltage and (ii) the approxi-
mate collector current. (b) Determine the (i) B–E voltage at the edge of high injec-
tion and (ii) the approximate resulting collector current.
15.8 Consider the npn power bipolar transistor described in Problem 15.7. (a) Determine
the expected B–C avalanche breakdown voltage. (b) Find the punch-through volt-
age. (c) What is the expected B–E avalanche breakdown voltage?
15.9 A silicon pnp power BJT is to be designed. The base doping concentration is
NB  5  1015 cm3. The base–collector junction breakdown voltage is to be
BVCBO  1000 V. Determine the maximum collector doping concentration and the
minimum base and collector region widths.
15.10 (a) Assume that BVCBO  300 V for a power BJT. Determine BVCEO for (i)   10
and (ii)   50. Assume n  3 (see Equation (12.63)). (b) Repeat part (a) for
BVCBO  125 V.
15.11 The effective  of a Darlington pair is found to be eff  180. The driver BJT, QA,
has a current gain A  25. (a) What is  of the output transistor QB? (b) If the
rated collector current of QB is ICB, max  20 A, what must be the rated collector
current of QA?
15.12 The maximum current, voltage, and power rating of an npn power BJT are 2 A,
120 V, and 30 W, respectively. (a) Sketch and label the safe operating area for this
transistor using linear current and voltage scales. (b) Determine RL such that the

nea29583_ch15_670-706.indd 704 12/11/10 12:57 PM

 Problems 705

maximum power is delivered to the load if the quiescent collector–emitter voltage

is 60 V. For this value of RL, what is the maximum current and maximum voltage?
(c) Determine the value of RL such that the maximum current and maximum power
can be obtained. (d) Determine the value of RL such that the maximum voltage and
maximum power can be obtained.
15.13 The common-emitter circuit in Figure 15.15 is biased at VCC  12 V. The power rat-
ing of the transistor is PT  10 W. (a) Determine RL such that the maximum power
is delivered to the load. (b) What must be the current rating of the transistor, IC, max?
15.14 The transistor in the common-emitter circuit in Figure 15.15 has parameters
PT  2.5 W, VCE,sus  25 V, and IC, max  500 mA. Let RL  100
. What is the value
of VCC such that the maximum power is delivered to the load?

Section 15.5 Power MOSFETs

15.15 A power MOSFET is used in the inverter circuit shown in Figure 15.26 in which
VDD  200 V and RD  100
. The on resistance of the transistor is Ron  2
at
a junction temperature of 25 C. The on resistance increases linearly with tempera-
ture and is 3
at a junction temperature of 100 C. Plot the power dissipated in the
transistor as a function of junction temperature.
15.16 Three MOSFETs are to be used in parallel to sink 5 A of load current when they
are on. (a) The on resistances of the three devices are Ron1  1.8
, Ron2  2
,
and Ron3  2.2
. Calculate the current in each device and the power dissipated in
each device. (b) For some unknown reason, the on resistance of the second device
increases to Ron2  3.6
. Recalculate the current in each device and the power
dissipated in each device.
15.17 Consider a silicon DMOS power MOSFET shown in Figure 15.20. The source dop-
ing concentration is 5  1017 cm3 and the base doping concentration is 1015 cm3.
(a) Design the drain doping concentration, channel length, and drain drift region
width to support a blocking voltage of 200 V. (b) Repeat part (a) such that the
blocking voltage is 80 V.
15.18 A power MOSFET is connected in a common-source configuration as shown in
Figure 15.26. The transistor parameters are Kn  0.20 A/V2, VT  2 V, ID, max  8 A,
BVDSS  80 V, and PT  45 W. The circuit parameters are VDD  60 V and
RL  10
. (a) Sketch and label the safe operating area for the transistor using lin-
ear current and voltage scales. Sketch the load line on the same curve. (b) Calculate
the power dissipated in the transistor for VGS  4, 6, and 8 V. Is there a possibility
of damaging the transistor? Explain.
15.19 Consider the power MOSFET described in Problem 15.18. (a) For VDD  60 V,
determine the value of RL such that the maximum power is delivered to the load
and the transistor remains biased in the safe operating area. For this case, what is
the maximum allowed drain current? (b) For RL  10
, determine the maximum
value VDD such that the maximum power is delivered to the load and the transistor
remains biased in the safe operating area.

Section 15.6 The Thyristor

15.20 One condition for switching a thyristor is that
1 
2  1. Show that this condi-
tion corresponds to 1 2  1, where 1 and 2 are the common-emitter current
gains of the pnp and npn bipolar transistors in the equivalent circuit of the thyristor.

nea29583_ch15_670-706.indd 705 12/11/10 12:57 PM

 706 CHAPTER 15 Semiconductor Microwave and Power Devices

15.21 Explain how a pulse of ionizing radiation could trigger a basic CMOS structure
into a high-current, low-impedance state.
15.22 Show that the triac can be triggered into its ON state by gate signals of either polar-
ity and with anode-to-cathode voltages of either polarity. Consider each voltage
polarity combination.

READING LIST
1. Baliga, B. J. Modern Power Devices. New York: John Wiley and Sons, 1987.
2. . Power Semiconductor Devices. Boston: PWS Publishing, 1996.
3. Dimitrijev, S. Principles of Semiconductor Devices. New York: Oxford University
Press, 2006.
4. Esaki, L. “Discovery of the Tunnel Diode.” IEEE Trans. Elec. Dev., ED-23 (1976).
5. Fisher, M. J. Power Electronics. Boston: PWS-Kent Publishing, 1991.
6. Gentry, F. E., F. W. Gutzwiller, N. Holonyak, Jr., and E. E. Von Zastrow.
Semiconductor Controlled Rectifiers: Principles and Applications of pnpn Devices.
Englewood Cliffs, NJ: Prentice Hall, 1964.
7. Ghandhi, S. K. Semiconductor Power Devices: Physics of Operation and Fabrication
Technology. New York: John Wiley and Sons, 1977.
8. Gunn, J. B. “Microwave Oscillations of Current in III-V Semiconductors.” Solid State
Comm., 1 (1963).
9. Oxner, E. S. Power FETs and Their Applications. Englewood Cliffs, NJ: Prentice
Hall, 1982.
10. Read, W. T. “A Proposed High Frequency, Negative Resistance Diode.” Bell Syst.
Tech. J., 37 (1958).
11. Ridley, B. K., and T. B. Watkins. “The Possibility of Negative Resistance Effects in
Semiconductors.” Proc. Phys. Soc. Lond., 78 (1961).
12. Roulston, D. J. Bipolar Semiconductor Devices. New York: McGraw-Hill, 1990.
13. Schroder, D. K. Advanced MOS Devices: Modular Series on Solid State Devices.
Reading, MA: Addison-Wesley, 1987.
14. Shur, M. Introduction to Electronic Devices. New York: John Wiley and Sons, 1996.
15. Streetman, B. G., and S. K. Banerjee. Solid State Electronic Devices, 6th ed. Upper
Saddle River, NJ: Pearson Prentice Hall, 2006.
16. Sze, S. M. Semiconductor Devices: Physics and Technology. New York: John Wiley
and Sons, 1985.
17. Sze, S. M. and K. K. Ng. Physics of Semiconductor Devices, 3rd ed. Hoboken, NJ:
John Wiley and Sons, 2007.
*18. Wang, S. Fundamentals of Semiconductor Theory and Device Physics. Englewood
Cliffs, NJ: Prentice Hall, 1989.
19. Yang, E. S. Microelectronic Devices. New York: McGraw-Hill, 1988.

*Indicates reference that is at an advanced level compared to this text.

nea29583_ch15_670-706.indd 706 12/11/10 12:57 PM

 A

A
P P E N D I X

Selected List of Symbols

T his list does not include some symbols that are defined and used specifically
in only one section. Some symbols have more than one meaning; however, the
context in which the symbol is used should make the meaning unambiguous. The
usual unit associated with each symbol is given.
a Unit cell dimension (Å), potential well width, acceleration,
gradient of impurity concentration, channel thickness of a
one-sided JFET (cm)
a0 Bohr radius (Å)
c Speed of light (cm/s)
d Distance (cm)
e Electronic charge (magnitude) (C), Napierian base
f Frequency (Hz)
fF (E ) Fermi–Dirac probability function
fT Cutoff frequency (Hz)
g Generation rate (cm3 s1)
g Generation rate of excess carriers (cm3 s1)
g(E ) Density of states function (cm3 eV1)
gc, gv Density of states function in the conduction band and
valence band (cm3 eV1)
gd Channel conductance (S), small-signal diffusion conductance (S)
gm Transconductance (A/V)
gn, gp Generation rate for electrons and holes (cm3 s1)
h Planck’s constant (J-s), induced space charge width in a
JFET (cm)
 Modified Planck’s constant (h/2)
hf Small-signal common-emitter current gain

707

nea29583_appA-G_707-729.indd 707 12/11/10 12:54 PM

 708 A P P E N D I X A Selected List of Symbols

___
j Imaginary constant,  1
k Boltzmann’s constant (J/K), wavenumber (cm1)
kn Conduction parameter (A/V2)
m Mass (kg)
m0 Rest mass of the electron (kg)
m* Effective mass (kg)
m*cn, m*cp Conductivity effective mass of electron and hole (kg)
m*dn, m*dp Density of states effective mass of electron and hole (kg)
m*n , m*p Effective mass of electron and hole (kg)
n Integer
n, l, m, s Quantum numbers
n, p Electron and hole concentration (cm3)
_
n Index of refraction
n, p Constants related to the trap energy (cm3)
nB0, pE0, pC0 Thermal-equilibrium minority carrier electron
concentration in the base and minority carrier hole
concentration in the emitter and collector (cm3)
nd Density of electrons in the donor energy level (cm3)
ni Intrinsic concentration of electrons (cm3)
n0, p0 Thermal-equilibrium concentration of electrons and
holes (cm3)
np, pn Minority carrier electron and minority carrier hole
concentration (cm3)
np0, pn0 Thermal-equilibrium minority carrier electron and minority
carrier hole concentration (cm3)
ns Density of a two-dimensional electron gas (cm2)
p Momentum
pa Density of holes in the acceptor energy level (cm3)
pi Intrinsic hole concentration ( ni)(cm3)
q Charge (C)
r, ,  Spherical coordinates
rd, r Small-signal diffusion resistance ()
rds Small-signal drain-to-source resistance ()
ro Output resistance ()
s Surface recombination velocity (cm/s)
t Time (s)
td Delay time (s)
tox Gate oxide thickness (cm or Å)
ts Storage time (s)

nea29583_appA-G_707-729.indd 708 12/11/10 12:54 PM

 A P P E N D I X A Selected List of Symbols 709

u(x) Periodic wave function

v Velocity (cm/s)
vd Carrier drift velocity (cm/s)
vds, vs, vsat Carrier saturation drift velocity (cm/s)
x, y, z Cartesian coordinates
x Mole fraction in compound semiconductors
xB, xE, xC Neutral base, emitter, and collector region widths (cm)
xd Induced space charge width (cm)
xdB, xdC Space charge width in base and collector (cm)
xBO Metallurgical base width (cm)
xdT Maximum space charge width (cm)
xn, xp Depletion width from the metallurgical junction into n-type
and p-type semiconductor regions (cm)
A Area (cm2)
A* Effective Richardson constant (A/K2/cm2)
B Magnetic flux density (Wb/m2)
B, E, C Base, emitter, and collector
BVCBO Breakdown voltage of collector–base junction with emitter
open (V)
BVCEO Breakdown voltage of collector–emitter with base open (V)
C Capacitance (F)
C Capacitance per unit area (F/cm2)
Cd, C Diffusion capacitance (F)
CFB Flat-band capacitance (F)
Cgs, Cgd, Cds Gate-source, gate-drain, and drain-source capacitance (F)
Cj Junction capacitance per unit area (F/cm2)
CM Miller capacitance (F)
Cn, Cp Constants related to capture rate of electrons and holes
Cox Gate oxide capacitance per unit area (F/cm2)
C Reverse-biased B–C junction capacitance (F)
D, S, G Drain, source, and gate of an FET
D Ambipolar diffusion coefficient (cm2/s)
DB, DE, DC Base, emitter, and collector minority carrier diffusion
coefficients (cm2/s)
Dit Density of interface states (#/eV-cm3)
Dn, Dp Minority carrier electron and minority carrier hole
diffusion coefficient (cm2/s)
E Energy ( J or eV)
Ea Acceptor energy level (eV)

nea29583_appA-G_707-729.indd 709 12/11/10 12:54 PM

 710 A P P E N D I X A Selected List of Symbols

Ec, Ev Energy at the bottom edge of the conduction band and top
edge of the valence band (eV)
Ec, Ev Difference in conduction band energies and valence band
energies at a heterojunction (eV)
Ed Donor energy level (eV)
EF Fermi energy (eV)
EFi Intrinsic Fermi energy (eV)
EFn, EFp Quasi-Fermi energy levels for electrons and holes (eV)
Eg Bandgap energy (eV)
Eg Bandgap narrowing factor (eV), difference in bandgap
energies at a heterojunction (eV)
Et Trap energy level (eV)
F Force (N)
Fn , Fp Electron and hole particle flux (cm2 s1)
F1/2() Fermi–Dirac integral function
G Generation rate of electron–hole pairs (cm3 s1)
GL Excess carrier generation rate (cm3 s1)
Gn0, Gp0 Thermal-equilibrium generation rate for electrons and
holes (cm3 s1)
G01 Conductance (S)
I Current (A)
Ib, Ie, Ic Small-signal base, emitter, and collector currents (A)
IA Anode current (A)
IB, IE, IC Base, emitter, and collector current (A)
ICBO Reverse-biased collector–base junction current with
emitter open (A)
ICEO Reverse-biased collector–emitter current with base open (A)
ID Diode current (A), drain current (A)
ID (sat) Saturation drain current (A)
IL Photocurrent (A)
IP1 Pinchoff current (A)
IS Ideal reverse-biased saturation current (A)
ISC Short-circuit current (A)
Iv Photon intensity (energy/cm2/s)
J Electric current density (A/cm2)
Jgen Generation current density (A/cm2)
JL Photocurrent density (A/cm2)
Jn, Jp Electron and hole electric current density (A/cm2)
Jn , Jp Electron and hole particle current density (cm2 s1)

nea29583_appA-G_707-729.indd 710 12/11/10 12:54 PM

 A P P E N D I X A Selected List of Symbols 711

Jrec Recombination current density (A/cm2)

Jr0 Zero-bias recombination current density (A/cm2)
JR Reverse-biased current density (A/cm 2)
JS Ideal reverse-biased saturation current density (A/cm2)
JsT Ideal reverse-saturation current density in a Schottky
diode (A/cm2)
L Length (cm), inductance (H), channel length (cm)
L Channel length modulation factor (cm)
LB, LE, LC Minority carrier diffusion length in the base, emitter, and
collector (cm)
LD Debye length (cm)
Ln, Lp Minority carrier electron and hole diffusion length (cm)
M, Mn Multiplication constant
N Number density (cm3)
Na Density of acceptor impurity atoms (cm3)
NB, NE, NC Base, emitter, and collector doping concentrations (cm3)
Nc, Nv Effective density of states function in the conduction band
and valence band (cm3)
Nd Density of donor impurity atoms (cm3)
Nit Interface state density (cm2)
Nt Trap density (cm3)
P Power (W)
P(r) Probability density function
Q Charge (C)
Q Charge per unit area (C/cm2)
QB Gate-controlled bulk charge (C)
Qn Inversion channel charge density per unit area (C/cm2)
Qsig Signal charge density per unit area (C/cm2)
QSD (max) Maximum space charge density per unit area (C/cm2)
QSS Equivalent trapped oxide charge per unit area (C/cm2)
R Reflection coefficient, recombination rate (cm3 s1),
resistance ()
R(r) Radial wave function
Rc Specific contact resistance (-cm2)
Rcn, Rcp Capture rate for electrons and holes (cm3 s1)
Ren, Rep Emission rate for electrons and holes (cm3 s1)
Rn, Rp Recombination rate for electrons and holes (cm3 s1)
Rn0, Rp0 Thermal-equilibrium recombination rate of electrons and
holes (cm3 s1)

nea29583_appA-G_707-729.indd 711 12/11/10 12:54 PM

 712 A P P E N D I X A Selected List of Symbols

T Temperature (K), kinetic energy (J or eV), transmission

coefficient
V Potential (V), potential energy (J or eV)
Va Applied forward-bias voltage (V)
VA Early voltage (V), anode voltage (V)
Vbi Built-in potential barrier (V)
VB Breakdown voltage (V)
VBD Breakdown voltage at the drain (V)
VBE , VCB , VCE Base–emitter, collector–base, and collector–emitter
voltage (V)
VDS, VGS Drain-source and gate-source voltage (V)
VDS (sat) Drain-source saturation voltage (V)
VFB Flat-band voltage (V)
VG Gate voltage (V)
VH Hall voltage (V)
Voc Open-circuit voltage (V)
Vox Potential difference across an oxide (V)
Vp0 Pinchoff voltage (V)
Vpt Punch-through voltage (V)
VR Applied reverse-biased voltage (V)
VSB Source-body voltage (V)
Vt Thermal voltage (kT/e)
VT Threshold voltage (V)
VT Threshold voltage shift (V)
W Total space charge width (cm), channel width (cm)
WB Metallurgical base width (cm)
Y Admittance
 Photon absorption coefficient (cm1), ac common-base
current gain
n, p Electron and hole ionization rates (cm1)
0 dc common-base current gain
T Base transport factor
 Common-emitter current gain
Emitter injection efficiency factor

Recombination factor

n,
p Excess electron and hole concentration (cm3)

np,
pn Excess minority carrier electron and excess minority
carrier hole concentration (cm3)
 Permittivity (F/cm 2)

nea29583_appA-G_707-729.indd 712 12/11/10 12:54 PM

 A P P E N D I X A Selected List of Symbols 713

0 Permittivity of free space (F/cm 2)

ox Permittivity of an oxide (F/cm 2)
r Relative permittivity or dielectric constant
s Permittivity of a semiconductor (F/cm2)
 Wavelength (cm or m)
 Permeability (H/cm)
 Ambipolar mobility (cm 2/V-s)
n, p Electron and hole mobility (cm 2/V-s)
0 Permeability of free space (H/cm)
Frequency (Hz)
 Resistivity (-cm), volume charge density (C/cm3)
 Conductivity (1 cm 1)
 Photoconductivity (1 cm1)
i Intrinsic conductivity (1 cm1)
n, p Conductivity of n-type and p-type semiconductors
(1 cm1)
 Lifetime (s)
n, p Electron and hole lifetime (s)
n0, p0 Excess minority carrier electron and hole lifetime (s)
0 Lifetime in space charge region (s)
 Potential (V)
(t) Time-dependent wave function
 Schottky barrier lowering potential (V)
Bn Schottky barrier height (V)
B0 Ideal Schottky barrier height (V)
fn, fp Potential difference (magnitude) between EFi and EF
in n-type and p-type semiconductors (V)
Fn, Fp Potential difference (with sign) between EFi and EF
in n-type and p-type semiconductors (V)
m Metal work function (V)
m Modified metal work function (V)
ms Metal–semiconductor work function difference (V)
n, p Potential difference (magnitude) between Ec and EF
in n-type and between Ev and EF in p-type
semiconductor (V)
s Semiconductor work function (V), surface potential (V)
 Electron affinity (V)
 Modified electron affinity (V)
(x) Time-independent wave function

nea29583_appA-G_707-729.indd 713 12/11/10 12:54 PM

 714 A P P E N D I X A Selected List of Symbols

 Radian frequency (s1)

Reflection coefficient
E Electric field (V/cm)
EH Hall electric field (V/cm)
Ecrit Critical electric field at breakdown (V/cm)

() Angular wave function
 Photon flux (cm2 s1)
() Angular wave function
(x, t) Total wave function

nea29583_appA-G_707-729.indd 714 12/11/10 12:54 PM

 A

B
P P E N D I X

System of Units,
Conversion Factors,
and General Constants

Table B.1 | International system of units*

Quantity Unit Symbol Dimension
Length meter m
Mass kilogram kg
Time second s or sec
Temperature kelvin K
Current ampere A
Frequency hertz Hz 1/s
Force newton N kg-m/s2
Pressure pascal Pa N/m2
Energy joule J N-m
Power watt W J/s
Electric charge coulomb C A-s
Potential volt V J/C
Conductance siemens S A/V
Resistance ohm  V/A
Capacitance farad F C/V
Magnetic flux weber Wb V-s
Magnetic flux density tesla T Wb/m2
Inductance henry H Wb/A
*The centimeter is the common unit of length and the electron-volt is the common unit of
energy (see Appendix D) used in the study of semiconductors. However, the joule and in
some cases the meter should be used in most formulas.

715

nea29583_appA-G_707-729.indd 715 12/11/10 12:54 PM

 716 APPENDIX B System of Units, Conversion Factors, and General Constants

Table B.2 | Conversion factors

Prefixes
1 Å (angstrom)  10 cm  10
8 10
m 1015 femto- f
1 m (micrometer)  104 cm 1012 pico- p
1 mil  103 in.  25.4 m 109 nano- n
2.54 cm  1 in. 106 micro- 
1 eV  1.6 1019 J 103 milli- m
1 J  107 erg 103 kilo- k
106 mega- M
109 giga- G
1012 tera T

Table B.3 | Physical constants

Avogadro’s number NA  6.02 1023
atoms per gram
molecular weight
Boltzmann’s constant k  1.38 1023 J/K
 8.62 105 eV/K
Electronic charge e  1.60 1019 C
(magnitude)
Free electron rest mass m0  9.11 1031 kg
Permeability of free space 0  4 107 H/m
Permittivity of free space 0  8.85 1014 F/cm
 8.85 1012 F/m
Planck’s constant h  6.625 1034 J-s
 4.135 1015 eV-s
_h    1.054 1034 J-s
2
Proton rest mass M  1.67 1027 kg
Speed of light in vacuum c  2.998 1010 cm/s
Thermal voltage (T  300 K) kT  0.0259 V
Vt  _e
kT  0.0259 eV

nea29583_appA-G_707-729.indd 716 12/11/10 12:54 PM

 APPENDIX B System of Units, Conversion Factors, and General Constants 717

Table B.4 | Silicon, gallium arsenide, and germanium properties (T  300 K)

Property Si GaAs Ge
Atoms (cm ) 3
5.0 10 22
4.42 10 22
4.42 1022
Atomic weight 28.09 144.63 72.60
Crystal structure Diamond Zincblende Diamond
Density (g/cm3) 2.33 5.32 5.33
Lattice constant (Å) 5.43 5.65 5.65
Melting point (C) 1415 1238 937
Dielectric constant 11.7 13.1 16.0
Bandgap energy (eV) 1.12 1.42 0.66
Electron affinity,  (V) 4.01 4.07 4.13
Effective density of states in 2.8 1019 4.7 1017 1.04 1019
conduction band, Nc (cm3)
Effective density of states in 1.04 1019 7.0 1018 6.0 1018
valence band, Nv (cm3)
Intrinsic carrier concentration (cm3) 1.5 1010 1.8 106 2.4 1013
Mobility (cm2/V-s)
Electron, n 1350 8500 3900
Hole, p 480 400 1900
m*
Effective mass _
m  0

Electrons m*I  0.98 0.067 1.64

m*t  0.19 0.082
Holes m*lh  0.16 0.082 0.044
m*hh  0.49 0.45 0.28
Density of states effective mass
Electrons _  
m*dn
mo 1.08 0.067 0.55

Holes _mo 
m*dp
0.56 0.48 0.37
Conductivity effective mass
Electrons _ 
m*cn
mo 0.26 0.067 0.12

Holes _mo 
m*cp
0.37 0.34 0.21

Table B.5 | Other semiconductor parameters

__
Material Eg (eV) a (Å) r χ n
Aluminum arsenide 2.16 5.66 12.0 3.5 2.97
Gallium phosphide 2.26 5.45 10 4.3 3.37
Aluminum phosphide 2.43 5.46 9.8 3.0
Indium phosphide 1.35 5.87 12.1 4.35 3.37

nea29583_appA-G_707-729.indd 717 12/11/10 12:54 PM

 718 APPENDIX B System of Units, Conversion Factors, and General Constants

Table B.6 | Properties of SiO2 and Si3N4 (T  300 K)

Property SiO2 Si3N4
Crystal structure [Amorphous for most integrated
circuit applications]
Atomic or molecular 2.2 1022 1.48 1022
density (cm3)
Density (g/cm3) 2.2 3.4
Energy gap  9 eV 4.7 eV
Dielectric constant 3.9 7.5
Melting point (C) 1700 1900

nea29583_appA-G_707-729.indd 718 12/11/10 12:54 PM

 A

C
P P E N D I X

The Periodic Table

Group I Group II Group III Group IV Group V Group VI Group VII Group VIII

Period a b a b a b a b a b a b a b a b

I lH 2 He
1.0079 4.003
II 3 Li 4 Be 5B 6C 7N 8O 9F 10 Ne
6.94 9.02 10.82 12.01 14.01 16.00 19.00 20.18
III 11 Na 12 Mg 13 Al 14 Si 15 P 16 S 17 Cl 18 Ar
22.99 24.32 26.97 28.06 30.98 32.06 35.45 39.94
IV 19 K 20 Ca 21 Sc 22 Ti 23 V 24 Cr 25 Mn 26 Fe 27 Co 28 Ni
39.09 40.08 44.96 47.90 50.95 52.01 54.93 55.85 58.94 58.69
29 Cu 30 Zn 31 Ga 32 Ge 33 As 34 Se 35 Br 36 Kr
63.54 65.38 69.72 72.60 74.91 78.96 79.91 83.7
V 37 Rb 38 Sr 39 Y 40 Zr 41 Nb 42 Mo 43 Tc 44 Ru 45 Rh 46 Pd
85.48 87.63 88.92 91.22 92.91 95.95 99 101.7 102.91 106.4
47 Ag 48 Cd 49 In 50 Sn 51 Sb 52 Te 53 I 54 Xe
107.88 112.41 114.76 118.70 121.76 127.61 126.92 131.3
VI 55 Cs 56 Ba 57–71 72 Hf 73 Ta 74 W 75 Re 76 Os 77 Ir 28 Pt
132.91 137.36 Rare earths 178.6 180.88 183.92 186.31 190.2 193.1 195.2
79 Au 80 Hg 81 Tl 82 Pb 83 Bi 84 Po 85 At 86 Rn
197.2 200.61 204.39 207.21 209.00 210 211 222
VII 87 Fr 88 Ra 89 Ac 90 Th 91 Pa 92 U 93 Np 94 Pu 95 Am 96 Cm 97 Bk 98 Ct 99 Es 100 Fm 101 Md
223 226.05 227 232.12 231 238.07 237 239 241 242 246 249 254 256 256

Rare earths
VI 57 La 58 Ce 59 Pr 60 Nd 61 Pm 62 Sm 63 Eu 64 Gd 65 Tb 66 Dy 67 Ho 68 Er 69 Tm 70 Yb 71 Lu
57-71 138.92 140.13 140.92 144.27 147 150.43 152.0 156.9 159.2 162.46 164.90 167.2 169.4 173.04 174.99
The numbers in front of the symbols of the elements denote the atomic numbers; the numbers underneath are the atomic weights.

719

nea29583_appA-G_707-729.indd 719 12/11/10 12:54 PM

 A

D
P P E N D I X

Unit of Energy—The
Electron Volt

T he electron volt (eV) is a unit of energy that is used constantly in the study of
semiconductor physics and devices. This short discussion may help in “getting
a feel” for the electron-volt.
Consider a parallel-plate capacitor with an applied voltage as shown in
Figure D.1. Assume that an electron is released at x  0 at time t  0. We may write
2
F  m0a  m0 _ d x  eE (D.1)
dt2
where e is the magnitude of the electronic charge and E is the magnitude of the electric
field as shown. Upon integrating, the velocity and distance versus time are given by

eEt
v_ (D.2)
m0

⫺V⫹

E-field

x⫽0 x⫽d

Figure D.1 | Parallel-plate

capacitor.

720

nea29583_appA-G_707-729.indd 720 12/11/10 12:54 PM

 APPENDIX D Unit of Energy—The Electron Volt 721

and
2
x_ eEt (D.3)
2m0
where we have assumed that v  0 at t  0.
Assume that at t  t0 the electron reaches the positive plate of the capacitor so
that x  d. Then
eEt2
d  _0 (D.4a)
2m0
or _____

_
2m0 d
t0  (D.4b)
eE
The velocity of the electron when it reaches the positive
_____
plate of the capacitor is


eEt 2eE d
v(t0)  _m0 
0 _
m0 (D.5)
The kinetic energy of the electron at this time is
T_ 1 m v(t )2  _
2 0 0
1m _
2 0 m0 
2eE d  eEd
 (D.6)
The electric field is
E_ V (D.7)
d
so that the energy is
Te·V (D.8)
If an electron is accelerated through a potential of 1 V, then the energy is
T  e · V  (1.6 1019)(1)  1.6 1019 joule (J) (D.9)
The electron-volt (eV) unit of energy is defined as
joule
Electron-volt  _ e (D.10)

Then, the electron that is accelerated through a potential of 1 V will have an energy
of

1.6 1019 (eV)

T  1.6 1019 J  __ (D.11)
1.6 1019
or 1 eV.
We may note that the magnitude of the potential (1 V) and the magnitude of the
electron energy (1 eV) are the same. However, it is important to keep in mind that the
unit associated with each number is different.

nea29583_appA-G_707-729.indd 721 12/11/10 12:54 PM

 A P

E P E N D I X

“Derivation” of Schrodinger’s
Wave Equation

S chrodinger’s wave equation is stated in Equation (2.6). The time-independent

form of Schrodinger’s wave equation is then developed and given by Equation
(2.13). The time-independent Schrodinger’s wave equation can also be developed
from the classical wave equation. We may think of this development more in terms
of a justification of the Schrodinger’s time-independent wave equation rather than a
strict derivation.
The time-independent classical wave equation, in terms of voltage, is given as

v 
 V(x)
2
2 V(x)  0
__  _ (E.1)
x2 2
p

where  is the radian frequency and vp is the phase velocity.

If we make a change of variable and let (x)  V(x), then we have

v 
 (x)
2
2 (x)  0
__  _ (E.2)
x2 2
p

We can write that

2  _
 v   _ 
2 2
_ 2 2 (E.3)
v2p p

where v and  are the wave frequency and wavelength, respectively.

From the wave–particle duality principle, we can relate the wavelength and
momentum as
h
_ (E.4)
p
Then

_   2h · p 
2 2
2  _ (E.5)

722

nea29583_appA-G_707-729.indd 722 12/11/10 12:54 PM

 APPENDIX E “Derivation” of Schrodinger’s Wave Equation 723

and since   _h , we can write

2
_
    
p 2 _
2 2  _

p2
 2m2 _
 2m   (E.6)

Now
2
p
_ TEV (E.7)
2m
where T, E, and V are the kinetic energy, total energy, and potential energy terms,
respectively.
We can then write

  2m 
   _
2  _
2
2 2
2m _p 2m (E  V)
_ _ (E.8)
v2
p  2 2

Substituting Equation (E.8) into Equation (E.2), we have

 (x) _
2
__  2m (E  V )(x)  0 (E.9)
x2 2
which is the one-dimensional, time-independent Schrodinger’s wave equation.

nea29583_appA-G_707-729.indd 723 12/11/10 12:54 PM

 A P

F
P E N D I X

Effective Mass Concepts

I n Chapter 3, we have discussed the relationship between the effective masses of

electrons and holes and the E versus k diagrams. In that discussion, we have lim-
ited ourselves to a one-dimensional analysis in k space.

F.1 | ENERGY-BAND STRUCTURES

GaAs Energy Bands: The E versus k diagram for GaAs is given in Figure 3.25a.
The minimum conduction band energy and the maximum valence band energy occur
at k  0. In a three-dimensional kx  ky  kz coordinate system, the constant energy
surface near the minimum conduction band energy is essentially spherical as shown
in Figure F.1. The electron effective mass can be determined as previously discussed
and is found to be m*n  0.067mo, where mo is the rest mass of an electron.
Silicon Conduction Energy Band: The E versus k diagram for silicon is given in
Figure 3.25b. The minimum conduction band energy is in the [100] direction. The
constant energy surface near the minimum conduction band energy in the three-
dimensional kx  ky  kz coordinate system is approximately an ellipsoid. There
are actually six ellipsoid energy surfaces corresponding to the six equivalent [100]
directions in the crystal as shown in Figure F.2a. The effective mass in both the
kx and ky directions is called a transverse effective mass mt, and the effective mass in
the kz direction is called a longitudinal effective mass ml. These effective masses are
indicated in a single ellipsoid as shown in Figure F.2b. The values of these effective
masses in silicon are found to be mt  0.19mo and ml  0.98mo.
Electrons are continually undergoing random scattering effects (see Chapter 5)
so that, at any given time, one-third of the electrons are moving in the kx direction with
an effective mass mt , one-third of the electrons are moving in the ky direction with an
effective mass mt , and one-third of the electrons are moving in the kz direction with
an effective mass ml . The effective mass parameter in the density of states function
and the effective mass parameter in conductivity calculations must therefore involve
some type of averaging of the transverse and longitudinal effective masses.

724

nea29583_appA-G_707-729.indd 724 12/11/10 12:54 PM

 APPENDIX F Effective Mass Concepts 725

kz kz
(001) mt
kz

(010) (100)
ky
mt

kx (100) (010) ky mt
(001)
kx
ky
(a) (b)
kx
Figure F.2 | (a) Six equivalent ellipsoidal constant
Figure F.1 | Spherical energy surfaces in the conduction band of silicon.
constant energy surface in the (b) A single ellipsoidal energy surface showing the
conduction band of GaAs. effective masses.

Silicon Valence Energy Band: The maximum energy of the valence band in sili-
con occurs at k  0. The valence band actually has two branches (not shown in
Figure 3.25b) with approximately parabolic shapes. The sharper parabola (larger
d 2Edk2) corresponds to light holes and the wider parabola (smaller d 2Edk2) cor-
responds to heavy holes. The effective masses of the light and heavy holes in silicon
are mlh  0.16mo and mhh  0.49mo, respectively.

F.2 | DENSITY OF STATES EFFECTIVE MASS

Density of States Effective Mass—Electrons: The kinetic energy of an electron,
corresponding to a constant ellipsoidal energy surface in silicon (see Figure F.2b)
can be written as
p2x p2y p2z
E_ __
2mt 2mt 2ml
or
p2x p2y p2z
1_ __
2mt E 2mt E 2ml E
The general equation of an ellipsoid in momentum space can be written as
p2 p2y p2
1  _2x  _2  _2z
a b c
where a, b, and c are the axes of the ellipsoid. For the energy ellipsoid in Figure F.2b,
we can write
a2  2mt E, b2  2mt E, c2  2ml E
The volume of an ellipsoid is proportional to the product a ⴢ b ⴢ c, so we have
______
Volume   (mt)2 ml

nea29583_appA-G_707-729.indd 725 12/11/10 12:54 PM

 726 APPENDIX F Effective Mass Concepts

There are six energy ellipsoids, so the total volume is proportional to

______
Total volume  6 (mt)2 ml
In the derivation of the density of states function in the conduction band, the vol-
ume in k space (momentum space) is included. So, from Equation (3.72), the density
of states function is proportional to
_______
gc (E)  Volume  (mdn
* )32  6 (mt)2 ml
The density of states function electron effective mass can then be written as
m*dn  623 [(mt)2 ml]13
For silicon, we have mt  0.19mo and ml  0.98mo. Then, we find
m*dn  623 [(0.19mo)2 (0.98mo)]13  1.08mo
where m*dn is the density of states electron effective mass.
Density of States Effective Mass—Holes: In the three-dimensional kx  ky  kz
coordinate system, the constant energy is essentially spherical for both the heavy and
light holes. The volume of a sphere in momentum space is
Volume  p3
where, for the heavy and light holes, respectively, we have
p2hh  2mhh E and p2lh  2mlh E
The total volume is the sum of the two spherical volumes, so that
Total volume  (mhh)32  (mlh)32
In the derivation of the density of states function in the valence band, the volume
in k space (momentum space) is included. So, from Equation (3.75), the density of
states function for holes is proportional to
gv (E )  Volume  (m*dp)32  (mhh)32  (mlh)32
The density of states function effective hole mass is then
m*dp  [(mhh)32  (mlh)3.2]23
For silicon, we have mhh  0.49mo and mlh  0.16mo, so that
m*dp  [(0.49mo)3.2  (0.16mo)32]23  0.55mo
where mdp
* is the density of states hole effective mass.

F.3 | CONDUCTIVITY EFFECTIVE MASS

Conductivity Effective Mass—Electrons: From Chapter 5, the average drift
velocity of a carrier due to an applied electric field is given by

 
1 _ec
vd  _ ⴢE
2 m*c

nea29583_appA-G_707-729.indd 726 12/11/10 12:54 PM

 APPENDIX F Effective Mass Concepts 727

where c is the mean time between collisions, E is the electric field, and m*c is now
the conductivity effective mass.
For a simple electron gas, the electron kinetic energy can be written as
1 m* (v )2  _ p2x p2y p2z
E_ d  _ _
2 2mcn 2mcn 2m*cn
* *

For the case of silicon and the ellipsoid energy surface, we have
p2x p2y p2z
E_ __
2mt 2mt 2ml
The two expressions for kinetic energy are equal if

3 _  
1 _
2m*cn
2 _
2mt
1
2ml
or
3 _
_ 2 _
1
m*cn mt ml
Again, for electrons in silicon, we have mt  0.19mo and ml  0.98mo. Then
3  __
_ 2  __
1
m*cn 0.19mo 0.98mo
which gives m*cn  0.26mo, where m*cn is the conductivity effective mass for electrons.
Conductivity Effective Mass—Holes: From Chapter 5, the drift current density
due to holes is given by
e
J  eppE  e _c pE
m*  
Assuming the mean time between collisions is the same for heavy holes and light
holes, we can write
JTotal  Jhh  Jlh
which can be written as

 
ec
JTotal  e _ (m*dp)32
m*cp
where p is the total hole concentration and is proportional to (m*dp)32. The parameter
m*cp is the conductivity effective mass for holes and m*dp is the density of states effec-
tive mass for holes.
The individual currents for heavy holes and light holes are proportional to


Jhh  e _
ec

mhh (mhh)  e(ec)(mhh)
32 12

and

 
ec
Jlh  e _
mlh (mlh)  e(ec)(mlh)
32 12

We then have
32
(mdp)
*
__  (mhh)12  (mlh)12
mcp
*

nea29583_appA-G_707-729.indd 727 12/11/10 12:54 PM

 728 APPENDIX F Effective Mass Concepts

or
(m*dp)32 (mhh)32  (mlh)32
m*cp  ___  ___
(mhh)  (mlh)
12 12
(mhh)12  (mlh)12
For silicon, we again have mhh  0.49mo and mlh  0.16mo, so that
(0.49mo)32  (0.16mo)32
m*cp  _____  0.37mo
(0.49mo)12  (0.16mo)12
and m*cp is the conductivity effective mass for holes.

F.4 | SUMMARY
The energy-band structure of germanium is essentially the same as silicon with four
ellipsoidal energy surfaces in the conduction band and two spherical energy surfaces
in the valence band corresponding to heavy and light holes. The calculations for the
density of states effective masses and conductivity effective masses are then identi-
cal to those for silicon. Gallium arsenide also has two spherical energy surfaces in
the valence band corresponding to heavy and light holes. So the calculations for the
density of states effective mass for holes and conductivity effective mass for holes
are also identical to those for silicon.
The density of states effective masses for electrons and holes are denoted as
m*dn and m*dp, respectively. The conductivity effective masses for electrons and holes
are denoted as m*cn and m*cp, respectively. In analyses and calculations in the text, the
effective masses for electrons and holes are usually denoted simply as m*n and m*p ,
respectively. Whether the density of states effective mass or the conductivity effec-
tive mass is to be used should be clear from the context of the problem.

nea29583_appA-G_707-729.indd 728 12/11/10 12:54 PM

 A

GP P E N D I X

The Error Function

 z
2__ et dt
2

erf(z)  _
 0

erf(0)  0 erf()  1
erfc(z)  1  erf(z)

z erf(z) z erf(z)
0.00 0.00000 1.00 0.84270
0.05 0.05637 1.05 0.86244
0.10 0.11246 1.10 0.88021
0.15 0.16800 1.15 0.89612
0.20 0.22270 1.20 0.91031
0.25 0.27633 1.25 0.92290
0.30 0.32863 1.30 0.93401
0.35 0.37938 1.35 0.94376
0.40 0.42839 1.40 0.95229
0.45 0.47548 1.45 0.95970
0.50 0.52050 1.50 0.96611
0.55 0.56332 1.55 0.97162
0.60 0.60386 1.60 0.97635
0.65 0.64203 1.65 0.98038
0.70 0.67780 1.70 0.98379
0.75 0.71116 1.75 0.98667
0.80 0.74210 1.80 0.98909
0.85 0.77067 1.85 0.99111
0.90 0.79691 1.90 0.99279
0.95 0.82089 1.95 0.99418
1.00 0.84270 2.00 0.99532

729

nea29583_appA-G_707-729.indd 729 12/11/10 12:54 PM

 A

H
P P E N D I X

ANSWERS TO SELECTED PROBLEMS

Chapter 1 Chapter 2
1.1 (a) 4 atoms, (b) 2 atoms, (c) 8 atoms 2.5   0.254 m (gold),   0.654 m (cesium)
1.3 (a) 2.35 Å, (b) 5  1022 atoms/cm3 2.7 (a) (i) 11.2 Å, (ii) 3.54 Å, (iii) 1.12 Å; (b) 0.262 Å
(c) 2.33 gm/cm3 2.9 10.3 keV
1.5 (a) 2.447 Å, (b) 3.995 Å 2.11 (a) 12.4 kV, (b) 0.11 Å
1.7 (a) 3.9 Å, (b) 5.515 Å, 2.13 (a) (i) p  8.783  1026 kg-m/s,
(c) 4.503 Å, (d) 9.007 Å (ii) E  1.31 eV;
1.9 (a) 0.228 gm/cm3, (b) 0.296 gm/cm3 (b) (i) p  8.783  1026 kg-m/s,
1.11 (b) a  2.8 Å, (c) 2.28  1022 cm3 for both (ii) E  5.55  102 eV
Na and Cl, (d) 2.21 gm/cm3 2.15 (a) t  8.23  1016 s, (b) p  7.03  1025 kg-m/s
1.13 (a) For A and B atoms, 4.687  1014 cm2, 2.17 A  _ 1__
2
(b) For A and B atoms, 3.315  1014 cm2
2.19 (a) P  0.393, (b) P  0.239, (c) P  0.865
1.15 (a) (i) See Figure 1.10b,
2.21 (a) P  0.25, (b) P  0.25, (c) P  1
(ii) See Figure 1.10c,
2.23 (a) (x, t)  A exp [j(kx  t)],
(iii) Same as (110) plane,
(b) k  8.097  108 m1,   7.76  109 m,
(iv) Intercepts at p  2, q  3, s  6;   7.586  1013 rad/s
(b) Directions perpendicular to planes 2.25 E1  6.69  103 eV, E2  2.67  102 eV,
1.17 (634) plane E3  6.02  102 eV
1.19 (a) (i) 4.47  1014 cm2, (ii) 3.16  1014 cm2, 2.27 (a) n  7.688  1029, (b) En1  15 mJ, (c) No
(iii) 2.58  1014 cm2;
(b) (i) 4.47  1014 cm2, (ii) 6.32  1014 cm2,
2.29 1  A cos  _ a  2 
x ,   B sin _

2x ,
a
(iii) 2.58  1014 cm2; 3x

3  C cos a , 4  D sin _
_
 
4x
a
(c) (i) 8.94  1014 cm2, (ii) 6.32  1014 cm2,
  
2 2
h2 n 
2 2
n y
(iii) 1.03  1015 cm2 2.31 (a) Enxny  _ _ x
_
2m a2 b2 _____


1.21 (a) 1.328  1022 cm3, 2.33 (a) 1 (x)  B1 exp (jk1 x), k1  _ 2mE ;
h2
(b) 3.148  1014 cm2, 2 (x)  A2 exp ( jk2x)  B2 exp (jk2 x),
(c) 4.74 Å, (d) 5.14  1014 cm2, 3.87 Å ___________


k2  _ 2m (E  V )
1.23 1.77  1023 cm3 h2
O

1.25 (a) 1.542  107, (b) 2.208  105

 k2  k1

2
4k1 k2
(b) R  __ , T  __
1.27 dao  116 k2  k1 (k1  k2)2

730

nea29583_appH_730-737.indd 730 12/11/10 12:52 PM

 A P P E N D I X H Answers to Selected Problems 731

2.35 (a) T  0.0295, (b) T  1.24  105, (c) ni  1.38 cm3; 3.28  109 cm3;
(c) N  1.357  1010 cm3 5.72  1012 cm3
2.37 (a) T  5.875  107, (b) a  0.842  1014 m 4.3 (a) T  367.5 K, (b) T  417.5 K
4k1 k3 4.5 (a) 9.325  106, (b) 4.43  104, (c) 3.05  103
2.39 T  __
(k1  k3)2 4.7 0.0854
2.41 E1  13.58 eV, E2  3.395 eV, 4.11 For T  200 K, EFi  Emidgap  0.0086 eV;
E3  1.51 eV, E4  0.849 eV For T  400 K, EFi  Emidgap  0.0171 eV;
Chapter 3 For T  600 K, EFi  Emidgap  0.0257 eV
3.5 (b) (i) a  , a  1.729; 4.13 
no  K ⴢ kT exp __
kT
(Ec  EF)

(ii) a  2, a  2.617

4.15 r1  15.4 Å, E  0.029 eV
3.9 (a) E  0.559 eV, (b) E  2.15 eV
4.17 (a) 0.2148 eV, (b) 0.9052 eV, (c) 6.90  103 cm3,
3.11 (a) E  1.005 eV, (b) E  3.635 eV
(d) Holes, (e) 0.338 eV
3.13 m*(A)  m*(B)
4.19 (a) 0.2764 eV, (b) 2.414  1014 cm3, (c) p type
3.15 A,B: velocity  x direction;
4.21 (a) no  6.86  1015 cm3, po  7.84  107 cm3;
C,D: velocity  x direction;
(b) Ec  EF  0.2153 eV, po  7.04  103 cm3
B,C: positive mass; A,D: negative mass
4.23 (a) no  7.33  1013 cm3, po  3.07  106 cm3;
3.17 A: m*  0.976mo; B: m*  0.0813mo
(b) no  8.80  109 cm3, po  3.68  102 cm3
3.21 (a) m*dn  0.56m o, (b) m*cn  0.12mo
____ 4.25 (a) 0.2787 eV, (b) 0.8413 eV, (c) 1.134  109 cm3,
 
1 2mn* ___ 1.055 18
__ 10 m3 J1
3.25 g(E)  _ _  (d) Holes, (e) 0.2642 eV
 E E
4.27 (a) po  6.68  1014 cm3, no  7.23  104 cm3;
3.27 (a) (i) gv  4.12  1019 cm3,
(b) EF  Ev  0.3482 eV, no  8.49  109 cm3
(ii) gv  6.34  1019 cm3;
4.29 0.0777 eV
(b) (i) gv  3.27  1019 cm3, 1 kT, E  E  _ 1 kT
(ii) gv  5.03  1019 cm3 4.31 E  Ec  _ v
2 2
3.29 (a) 2.68, (b) 0.0521 4.35 (a) po  3  1015 cm3, no  1.08  103 cm3;
3.31 (a) 120; (b) (i) 66, (ii) 495 (b) no  3  1016 cm3, po  1.08  104 cm3;
3.33 (a) 0.269, (b) 6.69  103, (c) 4.54  105 (c) no  po  1.8  106 cm3;
Ec  Ev (d) po  4  1015 cm3, no  1.44  102 cm3;
3.35 EF  __  Emidgap
2 (e) no  1014 cm3, po  1.48  107 cm3
3.37 (a) EF  2.35 eV, (b) EF  5.746 eV n
4.37 (a) _d  8.85  104, (b) fF (E)  2.87  105
3.39 (a) E1  EF  4.6kT, (b) f (E1)  0.01 Nd
3.41 (a) 0.00304, (b) 0.1496, (c) 0.997, (d) 0.50 4.39 (a) n type; (b) no  8  1014 cm3,
3.43 (a) At E  E1, f (E)  9.3  106; po  2.81  105 cm3;
At E  E2, 1  f (E)  1.66  1019 (c) Na
 4.8  1015 cm3, no  5.625  104 cm3
(b) At E  E1, f (E)  7.88  1018; 4.41 no  6.88  1011 cm3, po  2.75  1012 cm3,
At E  E2, 1  f (E)  1.96  107 Na  2.064  1012 cm3
3.45 (a) Si: f (E)  4.07  1010; Ge: f (E)  2.93  106; 4.45 ni  5.74  1013 cm3, po  3  1013 cm3
GaAs: f (E)  1.24  1012; 4.47 (a) n type; (b) no  1.125  1016 cm3,
(b) Same values as part (a) po  2  104 cm3;
3.47 (a) E  0.1017 eV, (b) E  0.2034 eV (c) Nd  1.825  1016 cm3
4.49 For 1014 cm3, Ec  EF  0.3249 eV,
Chapter 4 EF  EFi  0.2280 eV;
4.1 (a) ni  7.68  104 cm3; 2.38  1012 cm3; 1015 cm3, Ec  EF  0.2652 eV,
9.74  1014 cm3, EF  EFi  0.2877 eV;
(b) ni  2.16  1010 cm3; 8.60  1014 cm3; 1016 cm3, Ec  EF  0.2056 eV,
3.82  1016 cm3, EF  EFi  0.3473 eV,

nea29583_appH_730-737.indd 731 12/11/10 12:52 PM

 732 APPENDIX H Answers to Selected Problems

1017 cm3, Ec  EF  0.1459 eV, (d) n type: E  13.6 V/cm;

EF  EFi  0.4070 eV p type: E  93.75 V/cm;
4.51 T  200 K, EFi  EF  0.4212 eV, compensated: E  25 V/cm
T  400 K, EFi  EF  0.2465 eV, 5.25 (a) 2388 cm2/V-s, (b) 844 cm2/V-s
T  600 K, EFi  EF  0.0630 eV 5.29 n(0)  0.25  1014 cm3
4.53 (a) EFi  Emidgap  0.0447 eV; 5.31 (a) n(x1)  1.67  1014 cm3,
(b) (i) Acceptors, (ii) Na  1.97  1013 cm3 (b) n(x1)  8.91  1014 cm3
4.55 (a) (i) Ec  EF  0.2188 eV, 5.33 JTotal  18 A/cm2
(ii) N
d  1.031  1016 cm3;
(b) (i) Ec  EF  0.1594 eV,
5.35 E  14.5  26 exp _  
x V/cm
18
x cm3;
(ii) Nd
 1.718  1015 cm3 5.37 (a) n(x)  6.51  1015  (3.255  1015) exp _
d  
4.57 Add acceptors, Na  4  1015 cm3 (b) n(0)  3.26  10 cm ,
15 3

4.59 (a) 0.2009 eV, (b) 1.360 eV, (c) 0.7508 eV, n(50)  6.19  1015 cm3;
(d) 0.2526 eV, (e) 1.068 eV (c) Jdrf  95.08 A/cm2, Jdiff  4.92 A/cm2
(a) E  __ 24.1 13.4
x  1 , (b) E  1  _
5.39 __
Chapter 5  _
L   x
L 
5.1 (a)   4.808 -cm, (b)  0.208(-cm) 1
5.41 V  2.73 mV
5.3 (a) Nd  6  1016 cm3, n  1050 cm2/V-s; 5.43 (a) Jdiff  (1.24  105) exp _  L 
x A/cm2,
(b) Na  1017 cm3, p  320 cm2/V-s (b) E  2.59  10 V/cm
3

5.5 n  1116 cm2/V-s 5.45 (a) (i) 29.8 cm2/s, (ii) 160.6 cm2/s;
5.7 (a) R  100 , (b)  0.01(-cm)1, (b) (i) 308.9 cm2/V-s, (ii) 1351 cm2/V-s
(c) Nd  4.63  1015 cm3, 5.47 (a) VH  0.3125 mV, (b) EH  1.56  102 V/cm,
(d) Na  1.13  1015 cm3 (c) n  3125 cm2/V-s
5.9 (a) L  0.0256 cm, (b) vd  1.56  106 cm/s, 5.49 (a) VH  0.825 mV, (b) n type,
(c) I  80 mA (c) n  4.92  1015 cm3, (d) n  1015 cm2/V-s
5.11 (a) Si: tt  8.33  1011 s, GaAs: tt  1.33  1011 s;
(b) Si: tt  1.05  1011 s, GaAs: tt  1.43  1011 s Chapter 6
5.13 (a) po  1.3  1017 cm3, no  2.49  105 cm3; 6.1 (a) no  5  1015 cm3, po  4.5  104 cm3;
(b) no  5.79  1014 cm3, po  3.89  105 cm3 (b) R
 5  1020 cm3 s1
5.15 (a) (i) 4.39  106 (-cm)1, 6.3 (a)
n0  8.89  106 s,
(ii) 2.23  102 (-cm)1, (b) G  1.125  109 cm3 s 1,
(iii) 2.56  109 (-cm)1; (c) G  R  1.125  109 cm3 s1
(b) (i) 5.36  109 , (ii) 1.06  106 , Fp
(iii) 9.19  1012  6.7 _  2  1019 cm3 s1
x
5.17 avg  3.97 (-cm)1 6.9 (a) 
 n  1300 cm2/V-s;
5.21 (a) J  1.60 A/cm2, (b) T  456 K (b) D
 Dn  33.67 cm2/s;
5.23 (a) n type: no  5  1016 cm3, po  4.5  103 cm3; (c)
nt 
nO  107 s,
pt  2.18  104 s
p type: po  2  1016 cm3, 6.13 (a) For 0 t 106 s:
no  1.125  104 cm3; compensated:
no  3  1016 cm3, po  7.5  103 cm3; 
n  p  (2  1014) 1  exp _ t
 

pO cm ,
3

For t  10 s:
6
(b) n type: n  1100 cm2/V-s;
p type: p  400 cm2/V-s;
n  p  (2  1014) exp __ 
(t  106)

pO cm3;
(b) For 0 t 10 s:
6
compensated: n  1000 cm2/V-s;
(c) n type:  8.8 (-cm)1; 
 6.0  0.250 1  exp _ t
 

pO (-cm) ,
1

For t  10 s:
6
p type:  1.28 (-cm)1;
compensated:  4.8 (-cm)1;
 6.0  0.250 exp __ 
(t  106)

pO
(-cm)1

nea29583_appH_730-737.indd 732 12/11/10 12:52 PM

 A P P E N D I X H Answers to Selected Problems 733

6.15 (a)
nO  2.5  107 s; Chapter 7
(b) n  p  (5  1014) 1  exp _ 
t 3

nO cm ,   7.1 (a) (i) 0.611 V, (ii) 0.671 V, (iii) 0.731 V;


R
 (2  1021) 1  exp _ t
 
3 1

nO cm s ;
(b) (i) 0.731 V, (ii) 0.790 V, (iii) 0.850 V
7.3 (a) For Na  Nd  1014 cm3, Vbi  0.4561 V
(c) (i) 7.19  10 s, (ii) 1.73  10 s,
8 7
1015 cm3, 0.5754 V
(iii) 3.47  107 s, (iv) 7.49  107 s
1016 cm3, 0.6946 V
6.17 (a) (i) For 0 t 5  107 s:
1017 cm3, 0.8139 V

p  (2.5  1014) 1  exp _ t

pO cm ,
3
  (b) For Na  Nd  1014 cm3, Vbi  0.9237 V
For t  5  10 s:7

p  (1.58  1014) exp ___

pO 
(t  5  107)
cm3; 1015 cm3,
1016 cm3,
1.043 V
1.162 V
(ii) At t  5  107 s: p  1.58  1014 cm3; 1017 cm3, 1.282 V
(b) (i) For 0 t 2  106 s: (c) Silicon:

p  (2.5  1014) 1  exp _  t cm3,

pO   For Na  Nd  1014 cm3, Vbi  0.2582 V
For t  2  106 s: 1015 cm3,

(t  2 _
106) 0.4172 V
p  (2.454  1014) exp __
pO cm3; 1016 cm3, 0.5762 V
(ii) At t  2  106 s: p  2.454  1014 cm3 1017 cm3, 0.7353 V
x cm3,
6.19 (a) n  p  (2  1014) exp _
Ln   GaAs:
Ln  5.575  10 cm; 3
For Na  Nd  1014 cm3, Vbi  0.7129 V
x A/cm2,
(b) Jn  0.1784 exp _   Ln
1015 cm3,
1016 cm3,
0.8719 V
1.031 V
x A/cm2
Jp  0.1784 exp _  Ln 1017 cm3, 1.190 V
x cm3, L  5  103 cm;
6.21 n(x)  (5  1014) exp _   Ln n 7.5 (a) n side: EF  EFi  0.3653 eV,
p side: EFi  EF  0.3653 eV;
x
Jn (x)  0.4 exp  _
Ln
A/cm , 2
(b) Vbi  0.7306 V;
x (c) Vbi  0.7305 V;
 
Jp (x)  0.4 exp _ A/cm2
Ln (d) xn  0.154 m, xp  0.154 m,
6.25 For 0 t T: n  G
ot, Emax  4.75  104 V/cm
For t  T: n  G
oT 7.7 For T  200 K, Vbi  1.257 V
6.27 p  390.6 cm2/V-s, Dp  10.42 cm2/s 300 K, 1.157 V
6.31 (a) EFi  EF  0.3294 eV; 400 K, 1.023 V
(b) EFn  EFi  0.2697 eV, EFi  EFp  0.3318 eV 7.9 (a) Vbi  0.635 V;
6.33 (a) n  p  5.05  1014 cm3; (b) xn  0.864 m,
(b) EFi  EFp  0.3362 eV; xp  0.0864 m;

(c) (i) EF  EFp  kT ln __ 

po  p
po ,  (d) Emax  1.34  104 V/cm
7.11 T  380 K
(ii) EF  EFp  2.093 meV 7.13 (a) Vbi  0.456 V,
ni
6.39 (a) R  __ (b) xn  2.43  107 cm,

pO 
nO
6.41 (a) (i) p  1014 cm3, (c) xp  2.43  103 cm,
x cm3,

(ii) p  1014 1  0.167 exp _
Lp   (d) Emax  3.75  102 V/cm
7.17 (a) Vbi  0.8081 V;
x
14

(iii) p  10 1  exp _
Lp  
cm3, Lp  103 cm; (b) xn  0.2987 m,
(b) (i) p(0)  1014 cm3, xp  0.0597 m, W  0.3584 m;
(ii) p(0)  0.833  1014 cm3, (c) Emax  1.85  105 V/cm;
(iii) p(0)  0 (d) C  5.78 pF
6.43 (a) p(x)  1018 (20  104  x) cm3, 7.19 (a) Vbi  0.02845 V, (b) 1.732
(b) p(x)  1018 (70  104  x) cm3 7.21 (a) 3.13, (b) 0.316, (c) 0.319

nea29583_appH_730-737.indd 733 12/11/10 12:52 PM

 734 APPENDIX H Answers to Selected Problems

7.23 VR2  2.58 V 8.23 T  502 K, reverse-biased current

7.25 (a) L  3.306 mH; 8.29 (a) T  567 K, IS  Igen  2.314 A;
(b) (i) f  0.794 MHz, (ii) f  1.069 MHz (b) Va  0.5366 V
7.27 (a) Na  6.016  1015 cm3, Nd  1.504  1015 cm3; 8.31 Va  0.4 V: Id  7.64  1016 A, Irec  1.35  1010 A;
(b) Na  1.19  1016 cm3, Nd  2.976  1015 cm3 0.6 V: 1.73  1012 A, 6.44  109 A;
7.29 (a) VR  193 V, (b) xn  0.5 m, 0.8 V: 3.90  10 A,9
3.06  107 A;
(c) Emax  7.65  104 V/cm 1.0 V: 8.80  10 A,6
1.45  105 A;
7.31 (a) N  5.36  1015 cm3, (b) A  7.56  105 cm2, 1.2 V: 1.99  10 A,2
6.90  104 A
(c) VR  2.96 V 8.35 Jgen  1.5  10 A/cm
3 2

7.33 (a) Vbi  Vt ln __


NaO NdO
n2i
,
8.37 (a) rd  21.6 , Cd  11.6 nF;
(b) rd  216 , Cd  1.16 nF
eNaO 8.39 For 10 kHz, Z  25.9  j 0.0814;
(c) p region: E  __  (x  xp);
eNdO x _eN For 100 kHz, Z  25.9  j 0.814;
n region: 0  x  xo, E  _  dO For 1 MHz, Z  23.6  j 7.41;
2

 xn 
xo
_
2
;  For 10 MHz, Z  2.38  j 7.49
eNdO 8.41
p0  1.3  107 s; Cd  2.5  109 F
xo  x  xn, E  __  (xn  x)
7.35 (a) Na  1.29  10 cm3,
16 8.43 (a) R  72.3 , I  1.38 mA
(b) Na  2.59  1016 cm3 8.45 (a) Va  0.4896 V, (b) Va  0.4733 V
ts ts
8.47 (a) _
pO  0.956, (b)
pO  0.228
_
7.37 (a) VB  75 V, (b) VB  450 V
7.39 xn (min)  5.09 m 8.49 2.21  107 s
7.41 (a) VR  4.35  103 V, (b) VR  1.74  104 V
Chapter 9
(Note that breakdown is reached first in each case.)
9.1 (c) n  0.206 V, B0  0.27 V,
Chapter 8 Vbi  0.064 V, Emax  1.41  104 V/cm,
8.1 (a) 60 mV, (b) 120 mV (d) Bn  0.55 V, Emax  3.26  104 V/cm
8.3 (a) pn (xn)  4.0  1011 cm3, 9.3 (a) BO 1.09 V;
np (xp)  1.0  1011 cm3; (b) Vbi  0.8844 V;
(b) pn (xn)  9.03  1014 cm3, (c) (i) xn  0.4939 m, Emax  7.63  104 V/cm;
np (xp)  2.26  1014 cm3; (ii) xn  0.8728 m, Emax  1.35  105 V/cm
(c) pn (xn)  0, np (xp)  0 9.5 (b) n  0.1177 V;
8.5 (a) In  1.85 mA, (b) Ip  4.52 mA, (c) I  6.37 mA (c) Vbi  0.7623 V;
8.7 (a) I  0.244 mA, (b) I  1.568  108 A (d) (i) xn  0.7147 m, Emax  4.93  104 V/cm,
8.9 V  59.6 mV (ii) xn  1.292 m, Emax  8.92  104 V/cm
N N
8.11 (a) _d  12.73, (b) _d  0.354 9.7 (a) Vbi  0.90 V, (b) Nd  1.05  1016 cm3,
Na Na (c) n  0.0985 V, (d) Bn  0.9985 V
8.15 (a) p side: EFi  EF  0.329 eV, n side:
9.13 D
it  4.97  1011 cm2 eV1
EF  EFi  0.407 eV;
9.15 (a) BO  0.63 V; (i) 0.151 V, (ii) 0.211 V,
(b) IS  4.426  1015 A, I  1.07 A; (iii) 0.270 V;
Ip
(c) _  0.0741 (b) (i) 0.0654 V, (ii) 0.1317 V, (iii) 0.201 V
I
x
8.17 (a) pn (x)  (3.81  1014) exp __ 
2.83  104
cm3, 9.21 pn junction: (a) 0.678 V, (b) 0.718 V, (c) 0.732 V;
Schottky junction: (a) 0.447 V, (b) 0.487 V,
(b) Jp  0.597 A/cm2, (c) Jn  1.39 A/cm2 (c) 0.501 V
8.19 (a) Np  1.51  104, Nn  2.41  103; 9.23 pn junction: (a) 0.691 V, (b) I  120 mA;
(b) Np  7.17  105, Nn  1.15  105; Schottky junction: (a) 0.445 V, (b) I  53.3 mA
(c) Np  3.40  107, Nn  5.45  106 9.25 (a) R  0.1 , (b) R  1 , (c) R  10 
IS (400) IS (400) 9.27 (a) Bn  0.258 V, (b) Bn  0.198 V
8.21 (b) (i) __  1383, (ii) __  1.17  105
IS (300) IS (300)

nea29583_appH_730-737.indd 734 12/11/10 12:52 PM

 A P P E N D I X H Answers to Selected Problems 735

9.29 Nd  3.5  1018 cm3 W  7.24;

10.47 (a) (i) kn
 86.29 A/V2, (ii) _
L
9.33 Ec  0.17 eV W  15.5
(b) (i) kp
 40.27 A/V2, (ii) _
L
Chapter 10 10.49 (a) gmL  0.192 mA/V, (b) gms  2.21 mA/V
10.1 (a) p type, inversion; 10.51 VTO  0.386 V
(b) p type, depletion; 10.53 (a) VT  0.357 V, (b) VSB  5.43 V
(c) p type, accumulation; 10.55 (a) rs  198 , (b) 12% reduction
(d) n type, inversion 10.57 (a) fT  3.18 GHz, (b) fT  0.83 GHz
10.3 (a) Nd  8.38  1014 cm3, (b) s  0.566 V
Chapter 11
10.5 ms  0.9932 V
10.7 (a) VFB  1.04 V, (b) VFB  1.012 V 11.1 ID  1015 exp __ VGS
, I  (106)ID,
(2.1)Vt T
10.9 Q
sse  1.2  1010 cm−2 P  IT ⴢ VDD; for VGS  0.5 V,
10.11 (a) VTP  1.20 V, (b) VTP  0.210 V, ID  9.83 pA, IT  9.83 A,
(c) VTP  1.08 V P  49.2 W; for VGS  0.7 V,
10.13 Na  4  1016 cm3 ID  0.388 nA, IT  0.388 mA,
10.15 Nd  5  1014 cm3 P  1.94 mW; for VGS  0.9 V,
10.17 (a) tC  0.863 m, (b) VT  1.07 V ID  15.4 nA, IT  15.4 mA, P  77 mW
10.23 (a) Cox  2.876  107 F/cm2, 11.3 (a) L  0.1413 m, (b) L  0.2816 m,

 1.346  107 F/cm2,
CFB (c) L  0.0346 m, (d) L  0.1749 m

 3.083  108 F/cm2,
Cmin 11.5 (a) (i) L  0.0735 m, (ii) L  0.1303 m,
C
(inv)  2.876  107 F/cm2;
(iii) L  0.2205 m;

, and Cmin
(b) Cox, CFB
unchanged from part (a),
(b) L  1.84 m
C
(inv)  3.083  108 F/cm2;
11.7 (a) (i) ID  75.94 A, (ii) ID
 78.22 A,
(c) VFB  1.10 V, VT  0.2385 V
(iii) ro  658 k;
10.29 (b) VFB  0.695 V;
(b) (i) ID  0.30375 mA, (ii) ID
 0.3129 mA,
(c) (i) For VG  3 V, Vox  0.359 V
(iii) ro  165 k
10.31 Point 1: Inversion, 2: Threshold, 3: Depletion,
11.9 (a) Assume VDS(sat)  1 V; then
4: Flat band, 5: Accumulation
L  3 m ⇒ Esat  3.33  103 V/cm
10.33 (a) 0.0864 mA, (b) 0.1152 mA, (c) 0.1152 mA,
L  1 m ⇒ Esat  104 V/cm
(d) 0.4608 mA
W  9.26, (b) I  3.06 mA, (c) I  0.271 mA L  0.5 m ⇒ Esat  2  104 V/cm
10.35 (a) _
L D D
(b) Assume n  500 cm2/V-s,  nEsat,
10.37 (a) VGS  0.6 V, ID (sat)  0.025 mA L  3 m ⇒  1.67  106 cm/s
1.2 V, 0.625 mA L  1 m ⇒  5  106 cm/s
1.8 V, 2.025 mA L 0.5 m ⇒ ≈ 107 cm/s
2.4 V, 4.225 mA 11.13 (a) (i) ID  0.7175 mA, (ii) ID  1.23 mA,
(c) VGS  0.6 V, ID  0.0222 mA (iii) ID  1.409 mA, (iv) ID  1.64 mA;
1.2 V, 0.156 mA (b) (i) ID  0.552 mA, (ii) ID  1.10 mA,
1.8 V, 0.289 mA (iii) ID  1.38 mA, (iv) ID  1.38 mA;
2.4 V, 0.422 mA (c) For (a), VDS (sat)  2 V; for (b), VDS (sat)  1.25 V
10.39 (a) VGS  0 V, ID (sat)  0.711 mA 11.15 (a) Both bias conditions, ID
kID,
0.8 V, 2.84 mA (b) P
k2P
1.6 V, 6.40 mA 11.17 (a) (i) ID (max)  2.438 mA, (ii) ID (max)  1.298 mA;
10.43 For VSG  0.35 V, gd  0; (b) (i) P(max)  7.314 mW, (ii) P(max)  2.531 mW
For VSG  0.35 V, gd  2(0.961)(VSG  0.35) 11.19 VTO  0.389 V
10.45 VT
0.2 V, n  342 cm2/V-s 11.25 VT → kVT

nea29583_appH_730-737.indd 735 12/11/10 12:52 PM

 736 APPENDIX H Answers to Selected Problems

11.27 W  1.11 m 12.27 (a) xBLB  0.01: T  0.99995,   19,999

11.29 VT → kVT 0.10: 0.995 199
11.31 (a) tox  400 Å, (b) tox  600 Å 1.0: 0.648 1.84
11.33 Near punch-through, Vpt  2.08 V; 10.0: 0 0
Ideal punch-through, Vpt  4.9 V (b) NBNE  0.01:   0.990,   99
11.35 L  1.08 m 0.10: 0.909 9.99
11.37 Donor ions, DI  7.19  1011 cm2 1.0: 0.50 1.0
11.39 (a) VTO  0.0969 V; 10.0: 0.0909 0.10
(b) Donor ions, DI  3.63  1011 cm2 12.29 (a) Let xB  0.80 m, then NE  4.61  1018 cm3;
11.41 For VSB  1 V: VT  0.0443 V (b) T  0.99930,   0.99656
3 V: 0.0987 V 12.35 (a) (i) ro  101.7 k, (ii) go  9.84  106 ()1,
5 V: 0.138 V (iii) IC  1.22 mA;
11.43 VT  2.09 V (b) (i) ro  648 k, (ii) go  1.54  106 ()1,
(iii) IC  0.253 mA
Chapter 12 12.37 (a) (i) JC  52.16 A/cm2, (ii) JC  57.18 A/cm2,
12.3 (a) IS  7.2  1015 A; (iii) JC  61.85 A/cm2
(b) (i) IC  38.27 A, (ii) IC  0.571 mA, (b) VA  38.4 V
(iii) IC  8.519 mA
12.39 (a) xdB  0.1188 m, (b) IC  0.519 mA,
12.5 (a)   65.7;
(c) VA  13.3 V, (d) ro  7.705 k
(b) (i) IB  0.5828 A, IE  38.85 A;
12.41 (a) NB  1.83  1015 cm3,
(ii) IB  8.695 A, IE  0.5797 mA;
(b) NB  4.02  1016 cm3
(iii) IB  0.1297 mA, IE  8.649 mA;
12.43 S  1.42 m
(c)   165.7;
12.45 (a) BVBCO  180 V, (b) BVECO  34.5 V,
(i) IB  0.2310 A, IE  38.50 A;
(c) BVEB  19 V
(ii) IB  3.446 A, IE  0.5744 mA;
12.47 (a) BVCBO  64 V, (b) Vpt  70.0 V
(iii) IB  51.41 A, IE  8.570 mA
12.49 xBO  0.1483 m
12.9 (a) pEO  2.8125  102 cm3,
12.55 (a) (i)
e  36.26 ps, (ii)
b  84.5 ps,
nBO  1.125  104 cm3,
(iii)
d  22 ps, (iv)
c  0.72 ps;
pCO  2.25  105 cm3;
(b)
ec  143.48 ps;
(b) nB (0)  6.064  1014 cm3,
(c) fT  1.109 GHz;
pE (0)  1.516  1013 cm3
(d) f  8.87 MHz
12.11 (a) VBE  0.6709 V, (b) pE (0)  5.0  1013 cm3
12.15 (a) 0.126%, (b) 11.32% Chapter 13
12.19 (b) nB (xB)  6.7  1012 cm3, 13.3 (a) (i) VpO  3.312 V, (ii) Vp  1.984 V;
pC (0)  9.56  1013 cm3; (b) (i) a  h  0.103 m, (ii) a  h  0.065 m,
(c) xB  0.994 m (iii) a  h  0;
12.21 (a) (i)   0.99305, (ii) T  0.990, (c) (i) VDS (sat)  1.984 V, (ii) VDS (sat)  0.984 V
(iii)  0.990167, (iv)   0.97345, 13.5 (a) Na  9.433  1015 cm3, (b) Vp  1.47 V,
(v)   36.7; (c) VGS  0.347 V, (d) VSD  1.47 V
(b) InC  0.4986 mA, IpE  1.38 A, IR  1.39 A 13.7 (a) a  0.50 m;
12.23 (a) JnE  1.779 A/cm2, JpE  0.0425 A/cm2, (b) VpO  3.86 V;
JnC  1.773 A/cm2, JR  3.22  103 A/cm2; (c) (i) VSD (sat)  3.0 V, (ii) VSD (sat)  1.5 V
(b)   0.9767, T  0.9966,   0.9982, 13.9 (a) a  0.436 m;
  0.9716,   34.2
T (B) T (C) (b) (i) VpO  5.886 V, (ii) Vp  5.0 V
12.25 (b) (i) _  1, (ii) _ 1
T (A) T (A)

nea29583_appH_730-737.indd 736 12/11/10 12:52 PM

 A P P E N D I X H Answers to Selected Problems 737

13.11 (a) Ip1  1.03 mA; 14.11 (a) Voc  0.4847 V, (b) V  0.4383 V,
(b) (i) VDS (sat)  1.056 V, (ii) VDS (sat)  0.792 V, (c) Pm  46.5 mW, (d) RL  3.65 
(ii) VDS (sat)  0.528 V, (iv) VDS (sat)  0.264 V; 14.15 (a) Voc  0.474 V, (b) Pm  67.9 mW,
(c) (i) ID1  0.258 mA, (ii) ID1  0.141 mA, (c) RL  2.379 , (d) P  55.2 mW
o
n x  exp (x)
(iii) ID1  0.061 mA, (iv) ID1  0.0148 mA
13.13 (a) GO1  2.69  103 S;
14.17 np  __   
2L2n  1
exp _
Ln
(b) (i) VDS (sat)  0.35 V, (ii) VDS (sat)  0.175 V; 14.19 (a) I  120 mA, (b) p  1014 cm3,
(c) (i) ID1 (sat)  50.6 A, (ii) ID1 (sat)  12.4 A (c)   2.56  102 (-cm)1,
13.15 (a) gms (max)  0.295 mS, (b) gms (max)  1.48 mS (d) IL  3.2 mA, (e) ph  3.33
13.17 (a) Nd  8.1  1015 cm3, (b) VT  0.051 V 14.21 IL  0.131 A
13.19 (a) VT  0.1103 V, (b) a  0.2095 m 14.25 (a) GL (x)  (3.33  1020) exp [(103)x] cm3 s1,
13.21 (a) a  0.26 m,VT  0.092 V; (b) JL  53.3 mA/cm2
(b) VDS (sat)  0.258 V 14.27 d  230 m
13.23 (a) kn  1.206 mA/V2; 14.29 (a) (i) Eg  1.64 eV, (ii)   0.756 m;
(b) (i) ID1 (sat)  12.06 A, (ii) ID1 (sat)  0.1085 mA; (b) (i) Eg  1.78 eV, (ii)   0.697 m
(c) (i) VDS (sat)  0.10 V, (ii) VDS (sat)  0.30 V 14.31 x  0.38, Eg  1.85 eV
13.27 (a) L  2.333 m, (b) L  2.946 m 14.35   5.08  103 m
13.29 (a) VDS  2 V, (b) hsat  0.306 m,
Chapter 15
(c) ID1 (sat)  3.72 mA, (d) ID1 (sat)  9.05 mA
13.31 (a) td  5 ps, (b) td  20 ps 15.1 See Figure 8.29
13.33 (a) IDG  0.39 pA, (b) IDG  0.42 pA, 15.3 fr  23.9 MHz
(c) IDG  0.50 pA 15.5 (a) E  6  103 V/cm, (b) vd  1.5  107 cm/s,
13.35 fT  9.76 GHz (c) f  10 GHz
13.37 (a) fT  8.74 GHz, (b) fT  35.0 GHz 15.7 (a) (i) VBE  0.5696 V, (ii) IC  0.640 A;
13.39 (a) Voff  2.07 V, (b) ns  3.25  1012 cm2 (b) (i) VBE  0.6234 V, (ii) IC  5.12 A
13.41 d  251 Å 15.9 NC  2  1014 cm3, base width  3.16 m,
collector width  78.9 m
15.11 (a) B  5.96, (b) ICA  3.23 A
Chapter 14
15.13 (a) RL  3.60 , (b) IC, max  3.33 A
14.1 (a) 1.11 m, (b) 1.88 m, (c) 0.873 m, 15.17 (a) Let Nd  1014 cm3, channel length  4.86 m,
(d) 0.919 m drift region  48.6 m;
14.3 (a) (i)   9  103 cm1, (ii) 0.66; (b) Let Nd  1014 cm3, channel length  3.08 m,
(b) (i)   2.6  104 cm1, (ii) 0.875 drift region  30.8 m
14.5 (a) IvO  0.733 W/cm2, (b) d  2.56 m 15.19 (a) RL  20 , ID, max  3 A;
14.7 E  1.65 eV,   0.75 m (b) VDD  42.4 V

nea29583_appH_730-737.indd 737 12/11/10 12:52 PM

 INDEX

A Aluminum phosphide (AlP), 2

Abrupt junction approximation, 268 Ambipolar diffusion, 201
Absorption coefficient, 658, 662 Ambipolar diffusion coefficient, 203, 231
Accelerating field, 531 Ambipolar mobility, 231
Acceptor atoms, 148 Ambipolar mobility coefficient, 203
Acceptor concentration, 244 Ambipolar phenomenon, 206
Acceptor impurity atom, 120 Ambipolar transport, 198, 201–219, 231, 503, 524
Acceptor states, 401 Ambipolar transport equation, 201–203, 206–214,
Acceptor-type trap, 222 232, 637
Accumulation layer, 373 Ambipolar transport equation simplifications, 206
Accumulation layer charge, 431 Amorphous silicon, 662
Accumulation layer of electrons, 375–376 Amorphous silicon solar cells, 631–632
Accumulation layer of holes, 394–395 Amorphous solids, 2–3
Accumulation mode, 395 Amphoteric impurities, 123
Accumulation of electrons, 377 Amplification, 500–501, 537
Active device, 491 Anisotype junction, 355, 364
AIAs (aluminum arsenide), 2, 333 Anode, 674, 691–692, 700–701
AlGaAs (aluminum gallium arsenide) Anode current, 693, 700
AlxGal-xAs, 2, 630, 646, 660 Areal hole trap densities, 476–477
grown on substrates, 20 Arsenic (As), 122
heteroepitaxy, 19 Atomic bonding, 12–14
heterojunction bipolar transistor (HBT), 556–557 Atomic thermal vibration, 14
heterojunction LED, 653–654 Auger recombination, 644
AlGaAs-GaAs (aluminum gallium arsenide-gallium Auger recombination process, 644
arsenide) Avalanche breakdown
HEMT, 604–605 bipolar transistor, 464–465, 467, 533–534
heterojunction, 556–557 defined, 258, 260
junctions, 354 pn junction, 258–262
Allowed energy bands, 60–61, 69–72, 76–77, 79–82, 98 thyristor, 693
Allowed energy state, 221 Avalanche breakdown condition, 260
Almost empty band, 81 Avalanche breakdown voltage, 260
Almost full band, 81 Avalanche effect, 258
AlP (aluminum phosphide), 2 Avalanche photodiode, 641–642
Alpha cutoff frequency, 547, 559 Average drift velocity, 157–158, 170, 183
Aluminum (Al), 122, 333 Azimuthal (angular) quantum number (l), 48
Aluminum arsenide (AIAs), 2, 333
Aluminum gallium arsenide (AlGaAs) B
AlxGal-xAs, 2, 630, 646, 660 Ballistic transport, 453–455
GaAs-AlGaAs HEMT, 604–605 Band splitting, 80, 82
GaAs-AlGaAs junctions, 354 Bandgap energy, 63, 81
grown on substrates, 20 Bandgap narrowing, 526–528, 556–557, 559
heteroepitaxy, 19 Bandgap narrowing factor, 527–528
heterojunction bipolar transistor (HBT), 556–557 Barrier height. See Schottky barrier height
heterojunction LED, 653–654 Base current, 497–498, 554

738

nea29583_index_738-762.indd 738 12/13/10 2:24 PM

 Index 739

Base Gummel number, 541 emitter bandgap narrowing, 526–528

Base region, 492, 503–506, 528–529, 554 emitter current, 496–497
Base transit time, 545–546, 556, 559 emitter injection efficiency factor, 511–513, 518, 526,
Base transport factor, 511, 513–514, 518, 559 556, 559
Base width modulation, 522–524, 541, 559 emitter region, 492, 506–507
Base-collector (B-C) pn junction, 493 emitter-base (EùB) charging time, 556
Base-collector (B-C) space charge region, 546 equivalent circuit models, 536–541
Base-emitter (B-E) pn junction, 493 forward-active mode, 494–495, 500, 502–510, 559
Base-emitter (B-E) voltage, 496–497, 526, 543, 558 frequency limitations, 544–549
Basic Ebers-Moll equivalent circuit, 538 glossary of terms, 559
Basic interband transitions, 644 Gummel-Pool model, 540–541
Basic MOS capacitor structure, 372 HBT (heterojunction bipolar transistor), 552, 556–559
Basic SCR device, 697–698 high injection, 524–526
Basic transistor action, 491, 496, 558 hybrid-pi model, 537, 541–544
B-C (base-collector) pn junction, 493 interaction pn junction, 495
B-C (base-collector) space charge region, 546 inverse-active mode, 499–500, 508, 559
Bcc (body-centered cubic structure), 4–5, 7 large-signal switching, 549–552
B-E (base-emitter) pn junction, 493 low-frequency common-base current gain, 509–521, 546
B-E (base-emitter) voltage, 496–497, 526, 543, 558 minority carrier distribution, 501–509
Beryllium (Be), 50, 122–123 modes of operation, 498–500, 508–509
Beta cutoff frequency, 547–548, 559 nonideal effects, 522–536
Bilateral thyristor, 697–698 nonuniform base doping, 530–531
Binary semiconductor, 2, 20 notation, 502
Bipolar junction transistor (BJT), 491 polysilicon emitter BJT (bipolar junction transistor),
Bipolar phototransistor, 642 552–554
Bipolar transistor, 491–570 punch-through, 531–532
amplification, 500–501, 537 reading list, 569–570
avalanche breakdown, 533–534 recombination factor, 511, 515–516, 518
bandgap narrowing, 526–528, 556–557, 559 review and problems, 559–569, 736
base current, 497–498, 554 saturation mode, 499
base region, 492, 503–506, 528–529, 554 Schottky-clamped transistor, 551–552
base transit time, 545–546, 556, 559 silicon-germanium (SiGe)-base transistor, 552,
base transport factor, 511, 513–514, 518, 559 554–556
base width modulation, 522–524, 541, 559 simplified transistor current relation, 495–498
base-collector (B-C) pn junction, 493 specialized structures, 552–559
base-emitter (B-E) pn junction, 493 summary, 558
base-emitter (B-E) voltage, 496–497, 526, 543, 558 time-delay factors, 544–546
basic principle of operation, 493–495 transistor currents, 509–521
breakdown voltage, 531–536 transistor cutoff frequency, 546–549
collector current, 495–499, 509, 522–526, 529, 537, Bipolar transistor action, 492–501
550, 555, 558 BJT (bipolar junction transistor), 491
collector region, 492, 507–508 Bloch theorem, 63
collector-emitter (C-E) voltage, 499, 523 Bode plot, 548
common-base current gain, 497, 509–521, 534–535, Body diagonal, 9
537–538, 546, 559 Body-centered cubic structure (bcc), 4–5, 7
current crowding, 528–530, 559 Body-effect coefficient, 420–421
current gain effects, 555 Bohr model of the atom, 120
cutoff frequency, 547–548, 558–559 Bohr radius, 120–121
cutoff mode, 500, 508, 559 Bohr theory, 46, 120
Early effect, 522–523, 559 Boltzmann approximation, 96, 98
Ebers-Moll model, 537–540, 551 Born, Max, 32–33, 51

nea29583_index_738-762.indd 739 12/13/10 2:24 PM

 740 Index

Boron (B), 17, 50, 119–120, 122, 554 glossary of terms, 183
Bose-Einstein function, 91 graded impurity distribution, 176–180, 183
Bound particle, 36 Hall effect, 180–182
Boundary conditions induced electric field, 176–178
minority carrier concentrations, 284 mobility effects, 159–164
pn junction diode, 279–283 reading list, 191
Schrodinger’s wave equation, 33–34 resistivity, 164–166, 183
short diode, 294 review and problems, 184–191, 732
Breakdown voltage, 258, 261–262, 464–468, 531–536 summary, 183
Broken gap, 354–355 total current density, 175–176
Built-in potential barrier, 243–246, 267–268 velocity saturation, 169–172
Bulk charge effect, 418 Carrier velocity, 452
Carrier velocity saturation, 167–169, 452–453
C Cathode, 674, 691–692, 700–701
Cgd, 423 C-E (collector-emitter) loop, 498–499
Cgdp, 423 C-E (collector-emitter) saturation voltage, 539
CgdT, 424 Channel conductance, 406, 431, 585, 609
Cgs, 423 Channel conductance modulation, 431, 609
Cgsp, 423 Channel conductivity, 410
CgsT, 424 Channel length modulation, 424, 446–450, 481,
C⬘FB, 397 594–596, 610
C⬘min, 396 Channel length modulation effect, 447–448
Cox, 397 Channel length modulation parameter, 449
CoxT, 398 Channel space charge region, 577
Cadmium (Cd), 122–123 Channel transit time
Capacitance, 394 JFET, 600–601
Capacitance charging time MOSFET, 425
JFET, 600, 609 Channel width, 411, 461–464
MOSFET, 425 Charge carriers, 107–118, 148
Capacitance-voltage characteristics (MOSFET), 394–403 Charge distribution (MOSFET), 387–388
Carbon (C), 50 Charge neutrality, 135–148, 411
Carrier density gradient, 183 Charge sharing, 459
Carrier diffusion, 172–176 Charge storage, 551
Carrier diffusion coefficient, 183 Charge storage and diode transients, 314–317
Carrier diffusion current density, 172–175 Charging time constant, 546, 556
Carrier drift, 157–172 Chemical bonds, 14
Carrier drift current density, 157–159 Chemical vapor-phase deposition (CVD), 19
Carrier drift velocity, 170, 454 Chromium (Cr), 333
Carrier generation, 232 Circuit layout techniques, 430
Carrier generation and recombination, 193–198 Classical mechanics, compared to quantum mechanics,
Carrier injection, 322 33, 38, 43, 45, 80
Carrier mobility, 159–164, 183, 450, 452, 478 CMOS (complementary MOS), 427–431
Carrier recombination, 232 CMOS circuit, 371
Carrier transport phenomenon, 156–191 CMOS inverter, 428–429
carrier diffusion, 172–176 Collector, 492, 495
carrier drift, 157–172 Collector capacitance charging time, 545–546, 559
carrier mobility, 159–164, 183 Collector current, 495–499, 509, 522–526, 529, 537,
conductivity, 164–169, 183 550, 555, 558
diffusion current density, 172–175, 183 Collector depletion region transit time, 545–546, 559
drift current density, 157–159, 183 Collector region, 492, 507–508
Einstein’s relation, 179, 183 Collector series resistance, 546

nea29583_index_738-762.indd 740 12/13/10 2:24 PM

 Index 741

Collector-emitter (C-E) loop, 498–499 diamond structure, 10–11

Collector-emitter (C-E) saturation voltage, 539 glossary of terms, 20–21
Collector-emitter (C-E) voltage, 499, 523 growth of semiconductor materials, 17–20
Collector-to-substrate capacitance, 546 imperfections and impurities, 14–17
Common-base current gain, 497, 509–521, 534–535, reading list, 24
537–538, 546, 559 review and problems, 21–24, 730
Common-emitter circuit configuration, 500 semiconductor materials, 1–2
Common-emitter current gain, 498–499, 517–518, 525, space lattices, 3–9
534–535, 547–548, 559 summary, 20
Compensated semiconductors, 135–136, 148 types of, 4–5
Complementary MOS (CMOS), 427–431 Current, 107
Complete hybrid-pi equivalent circuit, 543 Current capability/current handling capability, 415
Complete ionization, 133–134, 148 Current crowding, 528–530, 559
Complete small-signal equivalent circuit, 313 Current density, 510
Compound semiconductor, 2 Current gain effects, 555
Compton effect, 27–28 Current gain factors. See Common-base current gain;
Conduction bands, 74–75, 77, 81 Common-emitter current gain
Conduction parameter, 410, 418, 431, 591, 610 Current-voltage (C-V) relationship. See I-V relationship/
Conduction-band edge characteristics
graded heterojunction, 357 Curvature effect, 465
N-AlGaAs, n-GaAs heterojunction, 357 Cutoff, 498
N-AlGaAs-intrinsic GaAs abrupt heterojunction, 602 Cutoff frequency
N-AlGaAs-undoped AlGaAs-undoped GaAs bipolar transistor, 547–548, 558–559
heterojunction, 603 JFET, 600–602, 610
Conduction-band energy, 279 MOSFET, 426–427, 431, 453
Conductivity, 164–169, 183 Cutoff mode, 500, 508, 559
Conductivity effective mass, 157n1, 726–728 C-V (current-voltage) relationship. See I-V relationship/
Constant of motion (k), 65 characteristics
Constant-field device scaling, 456 CVD (chemical vapor-phase deposition), 19
Constant-field scaling, 455–456 Czochralski method, 17, 19
Constants
charging time, 546, 556 D
dielectric relaxation time, 214–216, 674 DEc, 359
physical, 716 DEv, 359
Planck’s, 26, 30, 619 Darlington pair configuration, 682–684
Richardson, 343–345, 364 Davisson-Germer experiment, 28
separation-of-variables, 47n7, 48 Dc bias current, 311
Continuity equations, 198–199, 214 Dc common-base current gain, 511
Conversion efficiency, 627–628 Dc emitter current, 545
Conversion efficiency of solar cell, 627–628, 662 Dc voltage sources, 500
Conversion factors, 623, 716 De Broglie, Louis, 28, 31
Coulomb attraction, 47 De Broglie wavelength, 28, 36, 51
Covalent bonding, 13, 20, 72–73 Decoupling, 430
Critical angle, 651–652 Deep traps, 430
Critical angle loss, 650 Definitions. See Glossary of terms
Critical electric field, 268 Degenerate n-type semiconductor, 130
Crystal momentum, 72 Degenerate p-type semiconductor, 130
Crystal planes, 6–7 Degenerate semiconductor, 130–131, 148
Crystal pullers, 18 Delay time, 545, 550
Crystal structure of solids, 1–24 Delayed photocurrent, 639, 662
atomic bonding, 12–14 Density gradient, 242–243

nea29583_index_738-762.indd 741 12/13/10 2:24 PM

 742 Index

Density of states effective mass, 725–726, 728 Drain-to-source voltage, 408, 410–419, 578, 582,
Density of states function, 85–90, 98 585, 608
Dependent current source, 543 Drain-to-substrate capacitance, 598
Depletion layer capacitance, 255, 268 Drain-to-substrate pn junction, 404
Depletion layer thickness, 376–379 Drain-to-substrate pn junction capacitance, 423
Depletion mode Drift, 157, 183
JFET, 578–580, 582–587, 610 Drift current, 74–75, 183
MOSFET, 394–395, 403–405, 408–409, 415, Drift current density, 157–159, 178
431, 477 Drift velocity, 158, 169–171, 183, 452–454
Depletion mode device, 404, 573 DV/dt triggering, 696–697
Depletion region, 243, 267–268, 461
Depletion width, 446 E
Diamond lattice, 10–11, 13, 20, 83 E, 619
DIBL (drain-induced barrier lowering), 468, 470, 481 Ef, 114
Dielectric relaxation time constant, 214–216, 674 EFi, 114
Differential voltage, 583 Eg, 619
Diffusion, 172, 183 EV (electron-volt), 648, 720–721
Diffusion capacitance, 306–307, 311–313, 322 E versus k diagram
Diffusion coefficient, 179, 183 asymmetric distribution of electrons, 75
Diffusion conductance, 311, 322 conduction/valence bands, 73–74
Diffusion current, 172, 175, 183, 495, 522 displacements of allowed energy bands, 71
Diffusion current density, 172–175, 183 electron in bottom of conduction band, 89
Diffusion force, 242 empty states, 78
Diffusion of impurities, 16 free electron, 76
Diffusion resistance, 305–306, 322 GaAs1-xPx, 647
Diode current-voltage (C-V) relationship, 344–345, 364 gallium arsenide (GaAs), 83–84
Direct bandgap semiconductor, 84 one-dimensional, 85
Direct band-to-band generation, 193 parabolic approximation, 88
Directions in crystals, 9 reduced-zone representation, 71
Distribution laws, 91 silicon (Si), 83–84
DMOS (double-diffused MOSFET), 684–685, 702 Early effect, 522–523, 559
Domain, 674 Early voltage, 522–523, 555, 559
Donor atoms, 148 E-B (emitter-base) charging time, 556
Donor concentration, 130 E-B (emitter-base) junction, 516
Donor electron, 118–120 E-B (emitter-base) junction capacitance charging time,
Donor impurity atom, 119 545, 559
Donor impurity concentration, 527 E-B (emitter-base) space charge region, 494
Donor states, 401 Ebers-Moll model, 537–540, 551
Dopant atoms, 20, 118, 130 Effective density of states, 130, 148, 725–726
Doping, 16–17, 19–20, 118, 141, 144, 472 Effective density of states function in the conduction
Doping concentration, 167 band (Nc), 110, 113
Double heterojunction laser, 660 Effective density of states function in the valence band
Double-diffused MOSFET (DMOS), 684–685, 702 (Nv), 112–113, 148
Double-diffused npn bipolar transistor, 531 Effective density of states functions, 109, 113
Draft, 59 Effective electric field, 452
Drain current, 424–425 Effective electron mobility, 451
Drain overlap capacitance, 423, 426 Effective inversion charge mobility, 451–452
Drain-induced barrier lowering (DIBL), 468, 470, 481 Effective mass, 75, 77, 114, 724–728
Drain-to-source parasitic capacitance, 598 Effective mass values, 113
Drain-to-source resistance, 686 Effective mobility, 451–452
Drain-to-source saturation voltage, 582 Effective Richardson constant, 343–345

nea29583_index_738-762.indd 742 12/13/10 2:24 PM

 Index 743

Effective transverse electric field, 451–452 Emitter injection efficiency factor, 511–513, 518, 526,
Effective trapped oxide charge, 401 556, 559
Efficient luminescent material, 645 Emitter region, 492, 506–507
Einstein, Albert, 26 Emitter-base (E-B) charging time, 556
Einstein’s relation, 179, 183, 540 Emitter-base (E-B) junction, 516
Electric field (pn junction), 246–254, 267 Emitter-base (E-B) junction capacitance charging time,
Electrical conduction in solids, 72–82 545, 559
Electroluminescence, 644 Emitter-base (E-B) space charge region, 494
Electromagnetic frequency spectrum, 29 Emitter-to-collector transit time, 548
Electromagnetic waves, 30 Empty band, 71
Electron, 30, 98, 107 Empty state, 73, 78, 99, 107
Electron affinity, 333 Energy band theory (single crystal), 61, 63, 72, 80
Electron affinity rule, 356, 364 Energy bands, 59–63. See also Allowed energy bands;
Electron and hole Forbidden energy bands
concentrations, 107, 113, 123–124, 135–141 Energy quanta, 26–27
mobilities, 162–163 Energy shells, 12–13, 49
Electron behavior. See Quantum mechanics Energy-band diagrams
Electron capture, 222 adding donors, 138
Electron conductivity effective mass, 726–727 amorphous silicon PIN solar cell, 632
Electron density of states effective mass, bandgap materials, 355
725–726, 728 channel length (accumulation/weak inversion/
Electron diffusion coefficient, 174, 176 inversion), 445
Electron diffusion current, 173–174 degenerate semiconductors, 131
Electron diffusion current density, 176, 287 discrete acceptor energy state, 120
Electron drift, 175 discrete donor energy state, 119
Electron effective mass, 75–77, 80, 85, 98–99 double heterojunction laser, 660
Electron emission, 222 forward bias, 223, 277, 280, 299, 308, 342
Electron hole generation and recombination, 193 GaAlAs heterojunction LED, 654
Electron in free space, 35–36 GaAs, 171, 672–673, 724
Electron inversion charge density, 381 HEMT, 605–606
Electron mobility, 158, 164, 451 heterojunctions, 354–357, 362
Electron spin, 50 ideal (See Ideal energy-band diagrams)
Electron-hole pair formation, 619 interface states (charge trapped therein), 402
Electron-hole pair generation rate, 622–624 interface states (oxide-semiconductor interface), 401
Electron-volt (eV), 648, 720–721 inversion point, 419
Elemental semiconductor, 2, 20 ionization of acceptor state, 120, 134
Elements, 17–20. See also specific elements ionization of donor state, 119, 134
group I elements, 12–13 ionized/un-ionized donors and acceptors, 136
group II elements, 2, 122 ionizing radiation-induced processes, 475
group III elements, 1–2, 119–120 MESFET, 577
group III-V elements, 19, 122 metals, 82
group IV elements, 1, 10, 13, 122–123 MOS (n-type substrate, negative applied gate bias), 384
group V elements, 1, 19, 118 MOS capacitor (accumulation mode), 395, 402
group VI elements, 2 MOS capacitor (depletion mode), 395
group VII elements, 12 MOS capacitor (flat band), 386
periodic table, 50–51, 719 MOS capacitor (inversion mode), 396, 402
work functions of, 333 MOS capacitor (midgap), 402
Emitter bandgap narrowing, 526–528 MOS capacitor (p-type substrate, large positive gate
Emitter current, 496–497 bias), 375
Emitter current crowding, 528–530, 559 MOS capacitor (p-type substrate, moderate positive
Emitter doping, 526–528, 556 gate bias), 374

nea29583_index_738-762.indd 743 12/13/10 2:24 PM

 744 Index

Energy-band diagrams—(Cont.) Epitaxial growth, 19

MOS capacitor (p-type substrate, negative Epitaxial layer, 20
gate bias), 374 Equilibrium. See Semiconductor in equilibrium
MOS capacitor (p-type substrate, zero gate bias), 374 Equivalent circuit
MOS capacitor (n-type substrate, large negative gate Ebers-Moll model, 537–540, 551
bias), 377 hybrid-pi model, 537, 541–544
MOS capacitor (n-type substrate, moderate negative JFET, 598–602
gate bias), 377 MOSFET, 422–426
MOS capacitor (n-type substrate, positive MOSFET/parasitic BJT (distributed parameters),
gate bias), 377 689–691
MOS structure (negative applied gate bias), 17 parasitic bipolar transistor, 466
MOS structure (p-type substrate), 394, 444 pn junction, 313–314
MOS structure (point x), 413 three-terminal SCR, 694
MOS structure (positive applied gate bias), 389 Equivalent fixed oxide charge, 431
MOS structure (thermal equilibrium), 382 Error function, 317, 729
MOS structure (zero gate bias), 374, 383 Esaki diode, 671
MOSFET (double-subscripted voltage variables), 419 Excess carrier generation and recombination, 194–198
MOSFET (equipotential plot), 468 Excess carrier lifetime, 221–226
MOSFET (n-channel), 419 Excess carrier pulse, 217–218
n-AlGaAs emitter and p-GaAs base junction, Excess carrier recombination rate, 226
556–557 Excess carriers, 198–201, 213, 232. See also
nonuniform donor impurity concentration, 176 Nonequilibrium excess carriers
npn bipolar transistor, 494 Excess electron concentration, 212, 504–505
npn bipolar transistor (punch-through), 532 Excess electron pulse, 212
n-semiconductor-to-metal junction, 351 Excess electrons, 194, 232
n-type semiconductor, 131, 380 Excess hole concentration, 212–213
pn junction (forward bias), 277, 280, 299, 308 Excess holes, 194, 232
pn junction (reverse biased), 251, 277 Excess minority carrier electron concentration, 504
pn junction (thermal equilibrium), 243 Excess minority carrier hole concentration, 208, 507
pn junction (zero bias), 277, 320 Excess minority carrier holes, 212
p-type semiconductor, 131, 219, 378–379 Excess minority carrier lifetime, 197, 232
reverse biased, 223, 251, 277 External quantum efficiency, 650–653, 662
Si-base transistor, 554 Extrinsic doping, 203–206, 225–226
SiGe-base transistor, 554 Extrinsic materials, 118
surface potential (p-type semiconductor), 378 Extrinsic semiconductor, 120, 123–131, 148
threshold inversion point (n-type
semiconductor), 379 F
threshold inversion point (p-type FT, 426
semiconductor), 378 Fabry-Perot cavity, 657, 659
tunnel diode, 319–321 Fabry-Perot resonator, 657
zero bias, 277, 320 Face-centered cubic structure (fcc), 4–5, 7
Energy-band splitting, 61 Fermi energy, 93–99, 109, 113, 123–124, 129–130, 146
Energy-band structure, 171 Fermi energy level, 141–147
Energy-band theory of single-crystal materials, Fermi-Dirac distribution function, 93–98
61, 63, 72, 80 Fermi-Dirac integral, 128–130
Enhancement mode Fermi-Dirac probability function, 91–93
JFET, 589–590, 592, 610 Field oxide (FOX), 428
MESFET, 577–578, 590–591 Field oxide charge, 386
MOSFET, 403–404, 406, 409, 412, 416–418, 422, Field-effect, 431
428, 431, 477 Fill factor, 627, 662
pn junction FET, 578 Fixed oxide charge, 472

nea29583_index_738-762.indd 744 12/13/10 2:24 PM

 Index 745

Fixed oxide charge effects, 400–403 E versus k diagram, 83–84

Fixed positive oxide charge, 401 effective density of states function, 113
Flat band, 374 effective mass values, 113
Flat-band capacitance, 397–398 electron affinity, 333
Flat-band condition, 385, 397 electron and hole mobilities in, 163
Flat-band voltage, 385–388, 431, 458–459, 476 electron drift velocity versus electric field, 673
Fluorine (F), 50 energy-band diagrams, 171, 672–673, 724
Forbidden energy bands, 61–62, 72, 82, 99 as group III-V compound semiconductor, 122
Forward active, 494, 559 heteroepitaxy process, 19
Forward bias, 322, 333 heterojunction, 556
Forward-active mode, 494–495, 500, 502–510, 559 impurity ionization energies, 123
Forward-active operating mode, 494 intrinsic carrier concentration, 115
Forward-bias current density, 345 JFET, 601–602
Forward-bias current-voltage (C-V) relationship, 293 LED, 653
Forward-bias recombination current, 298–301 MESFET, 576, 588, 601
Forward-bias voltage, 280, 282, 347 mobility/diffusion values, 158, 179
Forward-biased npn bipolar transistor, 496 optical devices, 621–622, 628, 630, 645, 647–648,
Forward-biased pn junction, 280, 282, 285, 299–300, 653–654, 660
303, 308 properties, 717
FOX (field oxide), 428 resistivity, 165
Free particle, 36 Schottky barrier diode, 345
Freeze-out, 133, 145, 148 Schottky diode, 337
Frenkel defect, 15 as substrate, 20
Frequency effects, 399–400 visible spectrum, 645
Frequency limitations zincblende structure, 11
bipolar transistor, 544–549 Gallium arsenide phosphide (GaAsP)
JFET, 600–602 diode brightness, 653
MOSFET, 422–430 GaAs1-xPx, 646–647, 653
Fresnel loss, 650, 662 Gallium arsenide-aluminum gallium arsenide
(GaAs-AlGaAs)
G HEMT, 604–605
GaAlAs (gallium aluminum arsenide) heterojunction, 556–557
GaAlxAs1-x, 653 junctions, 354
heterojunction LED, 653–654 Gallium phosphide (GaP), 2, 165, 653
GaAs (gallium arsenide). See Gallium arsenide (GaAs) Gamma function, 110
GaAs-AlGaAs (gallium arsenide-aluminum gallium GaP, 2, 165, 653
arsenide) Gate capacitance charging time (JFET), 600, 609
HEMT, 604–605 Gate charging time, 425
heterojunction, 556–557 Gate voltage, 452
junctions, 354 Gate-to-channel space charge regions, 573–576
GaAsP (gallium arsenide phosphide) Gate-to-drain capacitance, 424
diode brightness, 653 Gate-to-source capacitance, 424
GaAs1-xPx, 646–647, 653 Gate-to-source voltage, 410–419, 423
Gallium aluminum arsenide (GaAlAs) Gaussian-type distribution, 474
GaAlxAs1-x, 653 Gauss’s law, 411–412
heterojunction LED, 653–654 Generalized scaling, 457
Gallium arsenide (GaAs) Generalized three-dimensional unit cell, 4
barrier height, 340 Generation, 193
as compound semiconductor, 2, 11 Generation current, 322
direct bandgap material, as, 646–648 Generation rate, 232
drift velocity, 170–171 Generation-recombination currents, 295–302

nea29583_index_738-762.indd 745 12/13/10 2:24 PM

 746 Index

Germanium (Ge) Helium (He), 50

covalent bonding, 13 HEMT (high electron mobility transistor), 602–609
diamond structure of, 11 advantages/disadvantages, 609
drift velocity, 170–171 alternative names, 603
effective density of states function, 113 energy-band diagrams, 605–606
effective mass values, 113 inverted structure, 605
electron affinity, 333 multilayer, 608
electron and hole mobilities in, 163 quantum well structures, 603–604
as elemental semiconductor, 2 transistor performance, 604–609
energy bands, 728 uses, 608
as group IV element, 10 Heteroepitaxy, 19
as indirect bandgap material, 84 Heterojunction AlGaAs-GaAs bipolar transistor, 556
intrinsic carrier concentration, 122–123 Heterojunction bipolar transistor (HBT), 552,
ionization energy of, 122–123 556–559
mobility/diffusion values, 158, 179 Heterojunction solar cell, 629–630
properties, 717 Heterojunctions, 354–363
resistivity, 165 defined, 364
SiGe-base transistor, 554 electron affinity rule, 356, 364
Germer, Lester, 28 energy-band diagrams, 354–357, 362
Glossary of terms equilibrium electrostatics, 358–362
bipolar transistor, 559 I-V relationship/characteristics, 342–345, 363
carrier transport phenomenon, 183 materials, 354
crystal structure of solids, 20–21 potential well, 358
JFET (junction field-effect transistor), 609–610 two-dimension electron gas, 356–358
MOSFET (metal-oxide-semiconductor field-effect types, 355
transistor), 431–432 HEXFET, 685, 702
nonequilibrium excess carriers, 231–232 High electron mobility transistor (HEMT). See HEMT
optical devices, 662–663 High injection, 524–526
pn junction, 268 High-level injection, 302–304, 322
pn junction diode, 322 High-speed logic circuits, 608
quantum mechanics, 51–52 High-temperature coil, 17
quantum theory of solids, 98–99 (hkl) plane, 9
Schottky barrier diode, 364 Holding current, 697
semiconductor in equilibrium, 148 Hole, 78–80, 98–99, 107
semiconductor/microwave power devices, 702 Hole concentrations, 107, 113, 123–124, 135–141
Gold (Au), 16, 333 Hole conductivity effective mass, 727–728
Graded impurity distribution, 176–180 Hole density of states effective mass, 726, 728
Grain boundaries, 3 Hole diffusion coefficient, 174, 176
Grains, 2 Hole diffusion current density, 176, 302
Group III-V semiconductors, 19, 122 Hole drift, 175
Gummel-Pool model, 540–541 Hole effective mass, 99
GUNN diode, 672–675 Hole-particle flux, 198
Homoepitaxy, 19
H Homojunction, 331, 354
Hv, 619 Hot electrons, 475, 481
Hall effect, 180–183 Hot-electron charging effects, 480
Hall field, 181 Hybrid-pi equivalent circuit model, 537, 541–544
Hall voltage, 181, 183 Hydrogen (H), 13, 19, 50, 479
Haynes-Shockley experiment, 216–219 Hydrogen atom, 13
HBT (heterojunction bipolar transistor), 552, 556–559 Hydrogen chloride (HCl), 19
Heisenberg uncertainty principle, 30, 51 Hydrogen fluoride (HF), 14

nea29583_index_738-762.indd 746 12/13/10 2:24 PM

 Index 747

Hydrogen valence electrons, 13 Impurity scattering, 160–161, 183

Hydrogenic model, 122 Incident particles, 41
Hyperabrupt junction, 265–268 Incident photon illumination, 629
Hyperbolic functions, 506 Incident photon intensity, 662
Hyperbolic sine function, 505 Incident wave, 651
Incremental conductance, 305
I Incremental resistance, 305–306
ICBO, 533–535 Indirect bandgap semiconductor, 84
ICEO, 534–535 Indium phosphide (InP), 2, 621, 672
ID, 406–407, 409, 414 Induced absorption, 655
ID(sat), 418 Induced electric field, 176–178
Ideal current-voltage relationship. See I-V relationship/ Induced emission, 655
characteristics Infinite potential well, 36–40
Ideal energy-band diagrams. See also Energy-band Infinite surface recombination velocity, 231
diagrams Injection electroluminescence, 644, 648, 662
metal-n-semiconductor junction, 332, 349 InP (indium phosphide), 2, 621, 672
metal-n-type semiconductor ohmic contact, 350 Interacting pn junction, 495
metal-p-type semiconductor junction, 350 Interaction between atoms, 12
metal-semiconductor (forward bias), 342 Interdigitated bipolar transistor structure, 678
metal-semiconductor junction (forward bias), 223 Interdigitated npn bipolar transistor, 529
metal-semiconductor junction (interfacial layer and Interface charge effects, 400–403
interface states), 341 Interface states, 340–341, 367, 401–403, 431, 478–479
metal-semiconductor junction (reverse biased), 223 Interfacial layer, 340–341, 367
nN heterojunction, 357 Internal pinchoff voltage, 578–582, 610
Np heterojunction, 362 Internal quantum efficiency, 649–650, 663
nP heterojunction, 356 Internal reflection, 652
pP heterojunction, 362 International system of units, 715
Ideal intrinsic semiconductor, 107 Interstitial defects, 14–15
Ideal junction properties, 334–338 Interstitial impurity, 16
Ideal nonrectifying barriers, 349–351 Intrinsic angular momentum, 50
Ideal pn junction current, 286–290 Intrinsic carrier concentration, 113–116, 122–123, 139,
Ideal reverse-saturation current density, 288, 292, 147, 167, 376, 526, 556
298, 346 Intrinsic electron concentration, 113, 128
Ideal Richardson constant, 345 Intrinsic Fermi energy, 114
Ideal saturation drain current, 585 Intrinsic Fermi-level position, 116–118
Ideal solar cell efficiency, 628 Intrinsic hole concentration, 113, 128
Ideal-diode equation, 288 Intrinsic material, 120
Ideality factor, 302 Intrinsic semiconductor, 107
Image force-induced lowering, 338, 364 Intrinsic silicon lattice, 118
Impact ionization, 464, 475, 480 Inverse active, 500, 559
Impact ionization avalanche transit-time (IMPATT), Inverse-active mode, 499–500, 508, 559
675–677 Inversion, 394
IMPATT diode, 675–677 Inversion carrier mobility, 416
Imperfections, 14–15 Inversion charge density, 381, 452
Implant approximation, 473 Inversion charge mobility, 451
Impurities, 16–17 Inversion layer, 375, 451
Impurity atoms, 16, 118, 130 Inversion layer charge, 406, 431, 447, 450
Impurity concentration, 16, 169 Inversion layer mobility, 431
Impurity diffusion, 16 Inversion layer of electrons, 430
Impurity doping concentration, 135 Inversion layer of holes, 430
Impurity ionization energies, 123 Inversion mode, 396

nea29583_index_738-762.indd 747 12/13/10 2:24 PM

 748 Index

Inverted GaAs-AlGaAs HEMT, 605 p-channel, 573

Inverted MODFET, 604 p-channel pn JFET, 579–582
Ion implantation, 16, 20, 472–474 pn JFET, 571–576
Ionic bond, 12, 14 reading list, 616–617
Ion-implanted profile, 472 review and problems, 610–616, 736–737
Ionization effect, 133 small-signal equivalent circuit, 598–600
Ionization energy, 120, 122–123 subthreshold current/gate current effects, 596–598
Ionized impurity scattering, 160–161, 183 summary, 609
Ionizing radiation, 475–479 threshold voltage, 579
Isotype junction, 355–356, 364 transconductance, 587–588, 596, 599–600
I-V relationship/characteristics velocity saturation, 596
diode, 344–345, 364 Junction breakdown, 258–262
forward-bias, 293 Junction breakdown voltage, 470
heterojunctions, 342–345, 363 Junction capacitance, 254–256, 268
ideal bipolar transistor common-base current-voltage Junction current, 277–295
(C-V) characteristics, 497 Junction field-effect transistor (JFET). See JFET
ideal I-V characteristic of a pn junction diode,
288–289 K
JFET, 582–587 Kinetic energy, 42
MESFET, 591 Kirchoff’s voltage equation, 499
MODFET, 607 Kirchoff’s voltage law, 500
MOS capacitor, 394–399 Kronig-Penney model, 63–67, 72, 99
MOSFET, 404–418, 469 K-space diagram, 67–72, 83–84, 99
pn junction diode, 278–279 KVL equations, 498
Schottky barrier diode, 342–345
SCR, 692 L
thyristor, 695 Laplace transform technique, 551
triac, 698 Laplacian operator, 47
Lapping operation, 19
J Large-signal switching, 549–552
JFET (junction field-effect transistor), 571–617 Laser diode, 654–661, 663
capacitance charging time, 600, 609 Lasing, 655–659
channel length modulation, 594–596 Lasing modes, 658
channel transit time, 600–601 Latch-up, 429–430
cutoff frequency, 600–602, 610 Lattice, 3, 20
depletion mode, 578–580, 582–587, 610 Lattice defects, 16
drain-to-source saturation voltage, 582 Lattice planes, 6
enhancement mode JFET, 589–590, 592, 610 Lattice point, 3–4
enhancement mode MESFET, 577–578, 590–591 Lattice scattering, 160–161, 452
equivalent circuit, 598–602 Lattice vibrations, 14
frequency limitations, 600–602 LC resonant circuit, 676
GaAs, 601–602 LDD (lightly doped drain) transistor, 470–471, 481
gate capacitance charging time, 600, 609 LED (light emitting diode), 662
glossary of terms, 609–610 Light, generation of, 648–649
HEMT, 602–609 Light application by stimulated emission of radiation
high electron mobility transistor (HEMT), 602–609 (laser), 654
ideal current-voltage relationship, 582–587 Light emitting diode (LED), 648–654, 662
internal pinchoff voltage, 578–582 Light spectrum, 622
MESFET, 571, 576–578, 588–593 Lightly doped drain (LDD) transistor, 470–471, 481
n-channel pn JFET, 578–580 Lilienfeld transistor, 572
nonideal effects, 593–598 Line defects, 15

nea29583_index_738-762.indd 748 12/13/10 2:24 PM

 Index 749

Line dislocation, 15 Metals

Linearly graded junctions, 263–265, 268 characteristics, 82
Liquid-phase epitaxy, 19 energy-band diagram, 82
Lithium (Li), 50–51 work functions of, 333
Load line, 499 Metal-semiconductor diode, 332
Load resistance, 425 Metal-semiconductor field-effect transistor (MESFET).
Localized free particle, 36 See MESFET
Long diode, 322 Metal-semiconductor junction, 331–354
Long pn junction, 284 Metal-semiconductor ohmic contacts, 349–354
Long-channel MOSFET, 452 defined, 364
Low frequency, 399 forming ohmic contacts, 353–354
Low injection, 203–206, 225–226 ideal nonrectifying barriers, 349–351
Low-frequency common-base current gain, 509–521, 546 specific constant resistance, 352–354
Low-level injection, 196–197, 203, 232 tunneling barrier, 351–352
Luminescence, 643, 649, 663 Metal-semiconductor work function difference,
Luminescent efficiency, 645–646 383–385, 400, 431, 472
Microwave power devices. See Semiconductor/
M microwave power devices
Magnetic quantum number (m), 48 Midgap, 402
Majority carrier concentration, 182 Midgap energy, 108–109, 117
Majority carrier current, 291 Miller capacitance, 426, 543, 662
Majority carrier device, 348 Miller effect, 543, 643, 662
Majority carrier electron concentration, 141 Miller indices, 7–8, 21
Majority carrier hole concentration, 140 Minimum capacitance, 396–397
Majority carrier mobility, 167, 182 Minority carrier concentration, 141, 282, 312, 316
Matter waves, 28 Minority carrier diffusion current density, 295
Maximum electric field, 470 Minority carrier diffusion length, 232, 492
Maximum induced space charge width, 473 Minority carrier distribution, 283–286, 501–509
Maximum power dissipation, 679, 687 Minority carrier electron concentration, 141
Maximum rated collector current, 679 Minority carrier hole concentration, 141, 506
Maximum rated current, 702 Minority carrier hole diffusion current
Maximum rated power, 680, 702 density, 287
Maximum rated power dissipation, 701 Minority carrier hole parameters, 205, 212
Maximum rated voltage, 679, 702 Minority carrier lifetime degradation, 430
Maximum resistive cutoff frequency, 672 Mobility, 157, 183
Maxwell-Boltzmann approximation, 96, 99 Mobility effects, 159–164
Maxwell-Boltzmann probability function, 91 Mobility values, 158, 164, 168, 179
MBE (molecular beam epitaxy), 19 Mobility variation, 450–452
Melts, 17–18 Moderate inversion, 397, 399
MESFET (metal-semiconductor field-effect transistor) MODFET (modulation-doped field-effect transistor),
basic operation, 576–578 603–608
GaAs, 597, 601–603 Modified Planck’s constant, 30
high frequency, 596 Modulation-doped field-effect transistor (MODFET),
JFET, 571, 588–593 603–608
Metal work function, 332, 340 Molecular beam epitaxy (MBE), 19
Metallic bonding, 13–14 Molybdenum (Mo), 333
Metallurgical junction, 242, 268 MOS capacitor, 372–377
Metal-oxide-semiconductor capacitor (MOS capacitor), MOS gated thyristor, 700
372–377 MOS structure, two-terminal, 372–394
Metal-oxide-semiconductor field-effect transistor MOS system, 375, 382, 389
(MOSFET). See MOSFET MOS turn-off thyristor, 700–701

nea29583_index_738-762.indd 749 12/13/10 2:24 PM

 750 Index

MOSFET (metal-oxide-semiconductor field-effect subthreshold current, 445–446, 478

transistor), 371–490 summary, 430–431, 481
accumulation mode, 395 surface charge density, 380–381
avalanche breakdown, 464–465, 467 threshold voltage, 388–394, 456–457, 472, 477
ballistic transport, 453–455 threshold voltage modifications, 457–464, 472–474, 482
basic operation, 403–422 transconductance, 418–419, 427, 432, 453
breakdown voltage, 464–468 transductance, 432
capacitance charging time, 425 two-terminal MOS structure, 372–394
capacitance-voltage (C-V) characteristics, 394–403 velocity saturation, 452–453
channel length modulation, 446–450 work function differences, 382–385, 472
channel transit time, 425 Multilayer HEMT, 608
charge distribution, 387–388 Multilayer modulation-doped heterostructure, 604
CMOS, 427–431
constant-field scaling, 455–456 N
cutoff frequency, 426–427, 431, 453 Ni, 113–116
depletion layer thickness, 376–379 No equation, 109, 125
depletion mode, 394–395, 403–405, 408–409, 415, Nopo product, 127
431, 477 N-AlGaAs emitter to p-GaAs base junction, 556–557
device types, 403 Narrow-channel effects, 461–464, 481
enhancement mode, 403–404, 406, 409, 412, N-channel
416–418, 422, 428, 431, 477 MESFET, 576
equivalent circuit, 422–426, 689–691 MOSFET, 371, 423
fixed oxide/interface charge effects, 400–403 pn JFET, 572–573, 578–580
flat-band voltage, 385–388 N-channel depletion mode MOSFET, 403–404, 408, 416
frequency effects, 399–400 N-channel enhancement mode
frequency limitations, 422–430 MESFET, 578
generalized scaling, 457 MOSFET, 403–404, 406, 412, 416
glossary of terms, 431–432 Near avalanche breakdown, 465–468, 470
hot-electron charging effects, 480 Near punch-through effects, 468–470, 481
inversion mode, 396 Negative differential mobility, 171, 702
ion implantation, 472–474 Negative differential resistance, 702
I-V relationship/characteristics, 394–399, Negative effective mass, 80
402–418, 449 Negative energy, 48
lightly doped drain (LLD) transistor, 470–471, 481 Negative threshold voltage, 390
long-channel, 452 Neon (Ne), 50–51
mobility variation, 450–452 Neutrons, 30
narrow-channel effects, 461–464, 481 Newton’s laws of motion, 25
nonideal effects, 444–455 Nickel (Ni), 333
oxide breakdown, 464 Nitrogen (N), 50
oxide thickness, 397, 419 NN heterojunction, 355, 357, 362
p-channel, 371 Nondegenerate semiconductors, 130
power MOSFET, 684–689, 701 Nonequilibrium excess carriers, 192–240
radiation effects, 475–480 ambipolar transport, 198, 201–219, 231
reading list, 441–442, 489–490 ambipolar transport equation, 201–203, 206–214, 232
review and problems, 432–441, 482–488, 735–736 carrier generation and recombination, 193–198
scaling, 455–457 characteristics, 198–201
short-channel effects, 457–461, 481 continuity equations, 198–199
small-signal equivalent circuit, 422–426 dielectric relaxation time constant, 214–216
snapback breakdown, 465–468, 482 excess carrier lifetime, 221–226
substrate bias effects, 419–422 extrinsic doping, 203–206, 225–226
subthreshold conduction, 444–446, 481–482 glossary of terms, 231–232

nea29583_index_738-762.indd 750 12/13/10 2:24 PM

 Index 751

Haynes-Shockley experiment, 216–219 One-sided junctions, 256–258, 268

low injection, 203–206, 225–226 One-sided MESFET, 578
notations/symbols, 194 Open-base configuration, 534–535
quasi-Fermi energy levels, 219–221, 232 Open-base phototransistor, 642
reading list, 240 Open-circuit voltage, 625, 663
review and problems, 232–240, 732–733 Open-emitter configuration, 534–535
Shockley-Read-Hall theory of recombination, 221–225 Optical absorption, 619–624
summary, 231 Optical cavity, 657–658
surface effects, 226–231 Optical density, 658
time-dependent diffusion equations, 199–201 Optical devices, 618–669
Nonideal effects electron-hole pair generation rate, 622–624
bipolar transistor, 522–536 glossary of terms, 662–663
JFET, 593–598 laser diode, 654–661, 663
MOSFET, 444–455 LED, 648–654, 663
Nonradiative recombination rate, 650 materials, 646–648
Nonuniform absorption effects, 628–629 optical absorption, 619–624
Nonuniform base doping, 530–531 photodetectors, 633–643, 662 (See also
Nonuniform donor impurity concentrations, 176 Photodetectors)
Nonuniform doping profile, 530 photoluminescence/electroluminescence, 643–648
Nonuniform photon absorption, 628–629, 640 photon absorption coefficient, 619–622
Nonuniformly doped junctions, 262–267 reading list, 668–669
Notations/symbols review and problems, 663–668, 737
bipolar transistor, 502 solar cells, 624–632
excess carriers, 194 summary, 661–662
npn bipolar phototransistor, 642 Optoelectronics, 618
npn Darlington pair configuration, 682 Ordered region, 2
pn junction, 245 Oscillators, 670
pn junction current, 279 Output conductance, 559
Np heterojunction, 355 Output resistance, 610
NP heterojunction, 350, 355–356, 362 Overlap capacitances, 423
Npn bipolar phototransistor, 642 Oxide breakdown, 464
Npn bipolar transistor, 492 Oxide capacitance, 387, 398, 431
Npn Darlington pair configuration, 682 Oxide charge, 475–478
Npn transistor, 429 Oxide thickness, 390, 397, 419, 472
N-type compensated semiconductor, 135 Oxide-isolated npn bipolar transistor, 493
N-type semiconductor, 119, 124, 130 Oxygen (O), 16, 50
Nucleus, 46–47
N-well CMOS process, 428–429 P
Pi, 113
O Po equation, 109, 111, 125
Off state, 314 P⫹n junction, 256–257, 265–266
Ohmic contacts, 331, 364. See also Metal-semiconductor P⫹-n-i-n⫹, 675
ohmic contacts Palladium (Pd), 333
Ohm’s law, 214, 410, 583 Parabolic relationship between energy and momentum,
On resistance, 685, 702 68, 88
On state, 314 Parallel-plate capacitor, 373
One-dimensional Kronig-Penney model, 63, 72, 99 Parasitic bipolar transistor, 466–467, 470
One-electron atom, 46–51 Parasitic BJT, 689–691
(110) plane, 8 Parasitic capacitances, 423, 426, 453
[111] direction, 9 Partial ionization, 134, 139–140
(111) plane, 8 Partially filled band, 82

nea29583_index_738-762.indd 751 12/13/10 2:24 PM

 752 Index

Passive device, 491 basic structure, 242–243

Pauli exclusion principle, 50–51, 60, 131 built-in potential barrier, 243–246, 267–268
P-channel diode (See Pn junction diode)
JFET, 573 electric field, 246–249, 251–254, 267
MOSFET, 371 equivalent circuit, 313–314
pn JFET, 579–582 glossary of terms, 268
P-channel depletion mode MOSFET, 403–405 hyperabrupt junction, 265–268
P-channel enhancement mode junction breakdown, 258–262
MESFET, 578 junction capacitance, 254–256, 268
MOSFET, 403, 405, 417 linearly graded junctions, 263–265 268
Penetration depth of particle, 43 nonuniformly doped junction, 262–267
Periodic table, 50–51, 719 notation, 245
Permittivity, 47, 120, 201, 215, 372, 606 one-sided junctions, 256–258, 268
Perpendicularity, 9 reading list, 275
Phonon (lattice) scattering, 160–161 reverse applied bias, 250–258
Phosphorus (P), 16–17, 118, 122 review and problems, 268–274, 733–734
Photoconductivity, 634 space charge width, 249–254, 265, 268
Photoconductor, 633–635, 662 summary, 267
Photoconductor gain, 634–635 zero applied bias, 243–250, 267
Photocurrent, 663 P⫺n junction, 260
Photodetectors Pn junction diode, 276–330
avalanche photodiode, 641–642 boundary conditions, 279–283
photoconductor, 633–635, 662 charge storage and diode transients, 314–317
phototransistor, 642–643 diffusion resistance, 305–306, 322
PIN photodiode, 640–641 forward-bias recombination current, 298–301
pn photodiode, 635–640 generation-combination currents, 295–302
Photodiode, 633–642, 662 glossary of terms, 322
Photoelectric effect, 26–27 high-level injection, 302–304, 322
Photoluminescence/electroluminescence, 643–648 ideal pn junction current, 286–290
Photon, 27, 30, 51 ideal reverse-saturation current density, 288, 292, 298
Photon absorption coefficient, 619–622 I-V relationship/characteristics, 278–279
Photon energy, 27 junction current, 277–295
Photon flux, 620, 645 minority carrier distribution, 283–286
Photon intensity, 620 reading list, 330
Photon-semiconductor interaction mechanisms, 619 reverse-biased generation current, 296–298
Phototransistor, 642–643 review and problems, 322–330, 734
Physical constants, 716 Schottky barrier diode, compared, 345–349
Physics short diode, 293–295
crystal structure of solids, 1–24 small-signal admittance, 306–313
quantum mechanics, 25–57 small-signal equivalent circuit, 313–314
quantum theory of solids, 58–105 small-signal model, 304–314
summary of, 290–292 summary, 321–322
PIN photodiode, 640–641 temperature effects, 292–293
Pinchoff, 573–575, 610 terms/notation, 279
Pinchoff current, 584 total forward-bias current, 300–302
Pinchoff voltage, 579–580 tunnel diode, 318–321
Planck’s constant, 26, 30, 619 turn-off transient, 315–317
Platinum (Pt), 333 turn-on transient, 315–317
Pn heterojunction, 629 Pn junction FET (pn JFET), 571–576
Pn JFET, 571–576 Pn junction solar cell, 624–627
Pn junction, 241–275 Pn laser diode, 657

nea29583_index_738-762.indd 752 12/13/10 2:24 PM

 Index 753

Pn photodiode, 635–640 Quantization of particle energy, 38

P⫺n junction, 257 Quantized energies, 51, 59
Pnp bipolar transistor, 492 Quantum efficiency, 645–646, 649–652
Pnp Darlington pair, 683 Quantum mechanics, 25–57
Pnp transistor, 429 compared to classical mechanics, 33, 38, 43, 45, 80
Pnpn structure, 430 electron in free space, 35–36
Point contact diode, 332 energy quanta, 26–27
Point defect, 14–15 glossary of terms, 51–52
Poisson’s equation infinite potential well, 36–40
ambipolar transport, 201 interaction between atoms, 12
dielectric relaxation time constant, 214 one-electron atom, 46–51
electric field, 246 periodic table, 50–51, 719
ideal junction properties, 334 potential barrier, 44–46
linearly graded junctions, 263 probability functions, 31
threshold voltage, 447, 473 reading list, 57
Polishing, 19 review and problems, 51–57, 730–731
Polycrystalline, 2–3 Schrodinger’s wave equation, 31–36, 47, 51, 357,
Polysilicon emitter BJT, 552–554 722–723
Population inversion, 655, 663 step potential function, 39–43
Positive energy, 48 summary, 51
Potential, 247, 251 tunneling, 45, 51–52
Potential barrier function, 44–46, 318 uncertainty principle, 26, 30–31
Potential function, 63–64, 72, 83 wave-particle duality, 26–30
Potential well, 358 Quantum numbers, 37, 48, 50
Power bipolar transistors, 677–684, 701 Quantum states
Power devices. See Semiconductor/microwave density of, 85–90, 94, 98
power devices Pauli exclusion principle, 60, 85
Power MOSFET, 684–689, 701 Quantum theory of solids, 58–105
PP heterojunction, 362 allowed/forbidden energy bands, 60–61, 69–72,
Primitive cell, 4, 21 76–77, 79–82, 98–99
Principal quantum number (n), 48 Boltzmann approximation, 96, 98
Probability, 30 density of states function, 85–90, 98
Probability density function distribution laws, 91
free particle, 36 drift current, 74–75
incident particles, 41 electrical conduction in solids, 72–82
isolated hydrogen atom, 46 electron effective mass, 75–77, 80, 85, 98–99
Max Born, 33, 51 energy band theory (single crystal), 61, 63,
radial, 49, 59 72, 80
reflected, 42 Fermi energy, 93
Probability functions, 31 Fermi-Dirac probability function, 91–93
Problems to solve. See Review and problems glossary of terms, 98–99
Process conduction parameter, 410, 418, 432 hole, 78–80, 98–99, 107
Prompt photocurrent, 636, 640, 663 Kronig-Penney model, 63–67, 72, 99
P-type compensated semiconductor, 135 k-space diagram, 67–72, 83–84, 99
P-type semiconductor, 120, 124, 130 reading list, 104
Punch-through, 468–470, 531–532 review and problems, 99–104, 731
P-well CMOS structure, 428 statistical mechanics, 91–98
summary, 98
Q three-dimensional crystals, 83–85
Q⬘ss, 431 Quantum well structures, 603–604
Quanta, 26, 31, 51 Quasi-Fermi energy levels, 219–221, 232, 285

nea29583_index_738-762.indd 753 12/13/10 2:24 PM

 754 Index

R pn junction diode, 322–330, 734

Rc resistance, 17 quantum mechanics, 51–57, 730–731
Rd, 423 quantum theory of solids, 99–104, 731
Rds, 424 Schottky barrier diode, 364–369, 734–735
RL, 425 semiconductor in equilibrium, 148–154, 731–732
Rs, 423 semiconductor/microwave power devices,
R-f induction coil, 17 703–706, 737
Radial probability density function, 49, 59 Richardson constant, 343–345, 364
Radiation effects, 475–480
Radiation-induced interface states, 478–479 S
Radiation-induced oxide charge, 475–478 Safe operating area (SOA), 680, 702. See also SOA
Radiative efficiency, 645–646 Safety margin, 464
Radiative recombination, 650, 663 Saturation, 432, 499–500
Radiative recombination rate, 646 Saturation condition, 582
Random thermal velocity, 160 Saturation drain current, 585
Recombination, 193 Saturation mode, 499, 508
Recombination center, 221 Saturation region
Recombination current, 322 JFET, 587
Recombination factor, 511, 515–516, 518 MOSFET, 408
Recombination processes, 643–644, 648–649, 663 Saturation velocity, 453–454
Recombination rate, 225, 232, 300 Sc (simple cubic structure), 4–5, 7
Recovery phase, 317 Scaling, 455–457
Reflected probability density function, 42 Scattering, 160
Reflected wave, 651 Scattering events, 453–454
Reflection coefficient, 43, 650 Schottky barrier, 333
Refraction, 652, 660 Schottky barrier diode, 332–349
Resistivity, 81–82, 164–166, 183 barrier height, 335–336, 338–341, 364
Resistor, 168 characteristics, 332–334
Reverse applied bias, 250–258 defined, 363
Reverse biased, 268 effective Richardson constant, 343–345
Reverse saturation current, 322 glossary of terms, 364
Reverse saturation current density, 288 ideal junction properties, 334–338
Reverse-biased current density, 297 interface states, 340
Reverse-biased generation current, 296–298 interfacial layer, 340–341, 367
Reverse-biased voltage, 267 I-V relationship/characteristics, 342–345
Reverse-biased photodiode, 636 as majority carrier device, 348
Reverse-biased PIN photodiodes, 640 pn junction diode, compared, 345–349
Reverse-biased pn junction, 277, 638 reading list, 370
Reverse-saturation current density, 346 review and problems, 364–369, 734–735
Review and problems summary, 363–364
answers, 730–737 Schottky barrier height, 335–336, 338–341, 364
bipolar transistor, 559–569, 736 Schottky barrier junction, 334, 346, 363
carrier transport phenomenon, 184–191, 732 Schottky barrier lowering, 338–341
crystal structure of solids, 21–24, 730 Schottky barrier rectifying contact, 576
JFET (junction field-effect transistor), 610–616, Schottky barrier rectifying junction, 588
736–737 Schottky clamped transistor, 348
MOSFET (metal-oxide-semiconductor field-effect Schottky diode, 337, 551
transistor), 432–441, 482–488, 735–736 Schottky effect, 364
nonequilibrium excess carriers, 232–240, 732–733 Schottky-clamped transistor, 551–552
optical devices, 663–668, 737 Schrodinger, Erwin, 31
pn junction, 268–274, 733–734 Schrodinger’s wave equation, 31–36, 47, 51, 357, 722–723

nea29583_index_738-762.indd 754 12/13/10 2:24 PM

 Index 755

SCR (semiconductor controlled rectifier), 691, 702 MOSFET (metal-oxide-semiconductor field-effect

SCR turn-off, 697 transistor), 371–490
SDHT (selectively doped heterojunction field-effect nonequilibrium excess carriers, 192–240
transistor), 603 optical devices, 618–669
Second breakdown, 680, 702 pn junction diodes, 276–330
Seed, 17–18 pn junctions, 241–275
Segregation coefficient, 17 power devices (semiconductor/microwave), 670–706
Selectively doped heterojunction field-effect transistor Semiconductor/microwave power devices, 670–706
(SDHT), 603 Darlington pair configuration, 682–684
Selenium (Se), 122–123 DMOS, 684–685, 702
Semiconductor controlled rectifier (SCR), 691, 702 glossary of terms, 702
Semiconductor doping, 472 GUNN diode, 672–675
Semiconductor heterojunction. See Heterojunctions HEXFET, 685, 702
Semiconductor in equilibrium, 106–155 IMPATT diode, 675–677
carrier generation and recombination, 193–194 parasitic BJT, 689–691
charge carriers, 107–118, 148 power bipolar transistors, 677–684
charge neutrality, 135–148 power MOSFET, 684–689, 701
compensated semiconductors, 135–136, 148 reading list, 706
complete ionization, 133–134, 148 review and problems, 703–706, 737
degenerate/nondegenerate semiconductors, summary, 701–702
130–131, 148 thyristor, 691–702
donor/acceptor statistics, 151–152 tunnel diode, 671–672
dopant atoms/energy levels, 118–123 VMOS, 684–685
electron and hole concentrations, 107, 113, 123–124, Semiconductors, physics of
135–141 crystal structure of solids, 1–24
equilibrium distribution of electrons/holes, 107–109, quantum mechanics, 25–57
123–127 quantum theory of solids, 58–105
equilibrium electrostatics, 358–362 Separation-of-variables constant, 47n7, 48
extrinsic semiconductor, 120, 123–131, 148 Separation-of-variables technique, 47
Fermi energy level position, 141–147 Series resistances, 423
Fermi-Dirac integral, 128–130 Shockley-Read-Hall recombination, 221–225, 640
freeze-out, 133, 145, 148 Short channel modulation, 449
glossary of terms, 148 Short diode, 293–295, 322
group III-V semiconductors, 122 Short-channel effects, 457–461, 481
intrinsic carrier concentration, 113–116, 139, 147 Short-channel threshold voltage model, 459
intrinsic Fermi level position, 116–118 Short-circuit current, 625, 663
ionization energy, 120, 122–123 Si-base transistor, 554
no equation, 109, 125 SiGe-base transistor, 552, 554–556
nopo product, 127 Silicon (Si)
partial ionization, 134, 139–140 bandgap narrowing factor compared to donor impurity
po equation, 109, 111, 125 concentration, 527
reading list, 154 barrier height, 340
review and problems, 148–154, 731–732 conduction energy band, 724
summary, 147–148 covalent bonding, 13, 72–73
Semiconductor materials, fabrication of, 17–20 diamond structure of, 11
Semiconductor materials and devices drift velocity, 170–171
bipolar transistor, 491–570 E versus k diagram for, 83–84
carrier transport phenomenon, 156–191 effective density of states function, 113
in equilibrium, 106–155 effective mass values, 113
JFET (junction field-effect transistor), 571–617 electron affinity, 333
metal-semiconductor ohmic contacts, 349–354 electron and hole mobilities in, 162–163

nea29583_index_738-762.indd 755 12/13/10 2:24 PM

 756 Index

Silicon (Si)—(Cont.) Snapback breakdown, 465–468, 482, 690

as elemental semiconductor, 2, 20 Snell’s law, 651
energy bands, 724–725, 728 SOA
epitaxial growth, 19 defined, 680, 702
as group IV element, 10 power MOSFET, 687–688
as ideal intrinsic semiconductor, 107 power transistors, 680
impurity concentrations, 16 Sodium (Na), 13–14
intrinsic carrier concentration, 122–123 Sodium chloride (NaCl), 12
MESFET, 576 Solar cells, 624–632
mobility/diffusion values, 158, 179 amorphous silicon cells, 631–632
n-channel JFET, 586 conversion efficiency/solar concentration, 627–628
popularity of, 10 heterojunction solar cell, 629–630
properties, 717–718 nonuniform absorption effects, 628–629
resistivity, 165 pn junction solar cell, 624–627
schematic representation of, 61–62 Solar concentration, 627–628
Schottky barrier diode, 345 Solar spectral radiation, 628
Schottky diode, 337 Solids
Si-base transistor, 554 crystal structure of, 1–24
SiGe-base transistor, 552 electrical conduction, 72–82
splitting of energy states in, 80 imperfections, 14–15
two-dimensional representation of intrinsic silicon impurities, 16–17
lattice, 118–119 quantum theory, 58–105
valence energy band, 725 types, 2–3
visible spectrum, 622 Solid-state transistor, 572
Silicon controlled rectifier, 691 Source resistance, 425
Silicon tetrachloride, 19 Source-to-drain saturation voltage, 418
Silicon valence electrons, 13 Source-to-substrate pn junction, 419
Silicon wafers, 18 Space charge density, 246, 257, 263
Silicon-germanium (SiGe)-base transistor, 552, 554–556 Space charge region, 242, 267–268. See also
Silicon-silicon dioxide (Si-SiO2) interface, 471, 476–480 Depletion region
Silver (Ag), 333 Space charge width, 249–254, 265, 268, 379
Simple cubic structure (sc), 4–5, 7 Space lattices, 3–9
Simple SCR circuit, 696 Spatial dependence, 211
Simplified transistor current relation, 495–498 Specialized bipolar transistor structures, 552–559
Single crystal, 2–3 Specific contact resistance, 352–354, 364
Single-crystal regions, 2 Sphalerite (zincblende) structure, 11
Single-crystal silicon solar cells, 630 Spherically symmetric probability function, 49
Sinusoidal common-base current gain, 500–501 Spin quantum number (s), 50
Small-signal admittance, 306–313 Splitting of energy bands, 60–62, 80
Small-signal BJTs, 679 Spontaneous emission, 655
Small-signal common-base current gain, 511 Spontaneous emission curve, 658
Small-signal diffusion resistance, 305, 313 Spontaneous emission rate, 644
Small-signal equivalent circuit Staggered, 354–355
JFET, 598–600 Statistical mechanics, 91–98
MOSFET, 422–426 Steady-state diode photocurrent density, 639
pn junction, 313–314 Steady-state excess carrier concentration, 210, 228
Small-signal incremental resistance, 306 Steady-state excess carrier hole concentration, 210
Small-signal input impedance, 672 Steady-state excess electron concentration, 210
Small-signal model of pn junction, 304–314 Steady-state excess hole concentration, 228
Smearing out, 403 Steady-state excess majority carrier hole
Snapback, 467 concentration, 210

nea29583_index_738-762.indd 756 12/13/10 2:24 PM

 Index 757

Steady-state forward-bias minority carrier Thermal voltage, 245

concentration, 316 Thermal-equilibrium concentration, 107,
Steady-state minority carrier concentration, 285 123–127, 129
Step junction, 242, 473 Thermal-equilibrium density of electrons, 110
Step potential function, 39–43 Thermal-equilibrium electron concentration, 109,
Stimulated emission, 655–657, 663 135–139, 147
Storage time, 316, 322 Thermal-equilibrium hole concentration, 111, 135,
Straddling, 354–355 139–141, 147
Strong inversion, 397, 399, 432 Thermionic emission theory, 342, 365
Substitute impurity, 16 Three-dimensional crystals, 83–85
Substrate, 19, 21 Three-element (ternary) semiconductor, 2, 20–21
Substrate bias effects, 419–422 Three-terminal SCR, 694
Subthreshold conduction, 444–446, 481–482, 596 Threshold, 577
Subthreshold current, 445–446, 478 Threshold adjustment, 482
Subthreshold current/gate current effects, 596–598 Threshold current, 658–661
Surface charge density, 380–381 Threshold inversion point, 378, 388, 432
Surface density of atoms, 9 Threshold voltage
Surface effects, 226–231, 516 defined, 378, 388, 432
Surface potential, 376, 381 JFET, 579
Surface recombination velocity, 229–232, 516 MESFET, 589
Surface scattering, 450–451, 482 MOSFET, 388–394, 456–457, 472, 477
Surface states, 226–229, 232 negative, 390
Switching, 348, 364, 549–552, 558, 687, pinchoff voltage, 579–580
691, 697 short-channel effects, 457–461, 481
Symbols. See Notations/symbols Thyristor, 691–702
Symbols, list of, 707–714 avalanche breakdown, 693
Symmetry effect, 96 bilateral, 697–698
characteristics, 691–694
T device structures, 697–701
Taylor expansion, 199, 505–506 I-V characteristics, 695
TED (transferred-electron device), 672 MOS gated, 700
TEGFET (two-dimensional electron gas field-effect MOS turn-off, 700–701
transistor), 603 SCR, 691, 702
Tellurium (Te), 122–123 SCR turn-off, 697
Temperature effects triac, 698–699, 702
carrier concentration and conductivity, 167 triggering the SCR, 694–697
current gain, 679 Time behavior, 206
Fermi energy level, 133, 136, 145–146 Time dependence, 207, 211
Fermi probability function, 94, 96 Time-delay factors, 544–546
high-power MOSFETs, 687 Time-dependent diffusion equations, 199–201
intrinsic carrier concentration, 114, 116, 139 Time-independent Schrodinger’s wave equation,
ôalmostö intrinsic silicon, 162 31, 35–36
optical output versus diode current, 661 Titanium (Ti), 333
pn junction current, 292–293 Total channel current, 411
scattering, 161, 179, 452 Total charge, 412
threshold voltage, 686 Total current density, 175–176, 287
Ternary semiconductor, 2, 21 Total forward-bias current, 300–302
Tetrahedral structure, 10–11, 13 Total forward-bias current density, 301–302
Thermal annealing, 17 Total gate oxide capacitance, 398
Thermal energy, 14, 119 Total reverse-biased current density, 297
Thermal equilibrium, 12, 106, 146 Total space charge width, 252

nea29583_index_738-762.indd 757 12/13/10 2:24 PM

 758 Index

Transconductance V
enhancement mode device, 593 V⬘gs, 423
ion implants, 472–474 Vbi, 359
JFET, 587–588, 596, 599–600 VDS, 406–409, 414
MESFET, 608 VDS(sat), 408, 582
MODFET, 608 VSD, 418
MOSFET, 418–419, 427, 432, 453 VSD(sat), 418
narrow-channel effects, 461–464, 481 Vacancy defect, 14, 16
Transferred-electron device (TED), 672 Vacancy-interstitial defect, 14
Transferred-electron effect, 702 Van Allen radiation belts, 475
Transistor, 371 Van der Walls bond, 14
Transistor currents, 509–521 Varactor diode, 266, 268
Transistor cutoff frequency, 546–549 Variable reactor, 266
Transistor gain, 418 Velocity saturation
Transistor performance, 604–609 carrier transport, 169–172, 183
Transistor switching, 550 JFET, 596
Transistor types, 371 MOSFET, 452–453
Transit-time mode, 675 Vertical pn power BJT, 677
Translation, 3 Vertical power transistor structure, 677–678
Transmission coefficient, 45 Very large scale integrated (VLSI) circuits, 17
Transmitted wave, 651 V-groove MOS gated thyrsistor, 700
Transport, 156, 183. See also Ambipolar transport; V-groove MOSFET (VMOS), 684–685, 702
Carrier transport phenomenon Visible spectrum, 622, 645
Transverse electric field, 451 VLSI (very large scale integrated) circuits, 17
Trap/trapping, 221–222, 476 VMOS (V-groove MOSFET), 684–685, 702
Triac, 698–699, 702 Voltage amplifier, 500
Triggering the SCR, 694–697 Voltage gain, 500
Tungsten (W), 333, 345 Volume charge density, 164n2
Tunnel diode, 318–321, 671–672 Volume density of atoms, 5
Tunneling, 45, 51–52
Tunneling barrier, 351–352, 365 W
Turn-off transient, 315–317 Wave equation, Schrodinger’s, 31–36, 357,
Turn-off voltage, 579 722–723
Turn-on transient, 315–317 Wave function, 32–33
Twin-well CMOS process, 428–429 Wave mechanics, 25, 31
2-DEG (two-dimensional electron gas), Wave number, 35
356–358, 365 Wavelength, 27, 621–622
Two-dimensional electron gas (2-DEG), Wave-particle duality, 26–30
356–358, 365 Weak inversion, 432, 445
Two-dimensional electron gas field-effect transistor Work function differences, 382–385, 472
(TEGFET), 603 Work functions, 26–27, 333
Two-dimensional lattice, 3
Two-element (binary) semiconductor, 2, 20 Z
Two-terminal MOS structure, 372–394 Zener breakdown, 258–259
Zener effect, 258
U Zero applied bias, 243–250, 267
Uncertainty principle, 26, 30–31 Zinc (Zn), 122–123
Unipolar transistor, 572 Zincblende lattice, 21, 83
Unit cell, 3–4, 21 Zincblende (sphalerite) structure, 11
Units, international system of, 715 Zone refining, 17

nea29583_index_738-762.indd 758 12/13/10 2:24 PM

nea29583_index_738-762.indd 759 12/13/10 2:24 PM