Semiconductor Diodes

1
CHAPTER OBJECTIVES
●
● Become aware of the general characteristics of three important semiconductor
materials: Si, Ge, GaAs.
● Understand conduction using electron and hole theory.
● Be able to describe the difference between n- and p-type materials.
● Develop a clear understanding of the basic operation and characteristics of a diode in
the no-bias, forward-bias, and reverse-bias regions.
● Be able to calculate the dc, ac, and average ac resistance of a diode from the
characteristics.
● Understand the impact of an equivalent circuit whether it is ideal or practical.
● Become familiar with the operation and characteristics of a Zener diode and
light-emitting diode.

1.1 INTRODUCTION
●
One of the noteworthy things about this field, as in many other areas of technology, is how
little the fundamental principles change over time. Systems are incredibly smaller, current
speeds of operation are truly remarkable, and new gadgets surface every day, leaving us to
wonder where technology is taking us. However, if we take a moment to consider that the
majority of all the devices in use were invented decades ago and that design techniques
appearing in texts as far back as the 1930s are still in use, we realize that most of what we
see is primarily a steady improvement in construction techniques, general characteristics,
and application techniques rather than the development of new elements and fundamen-
tally new designs. The result is that most of the devices discussed in this text have been
around for some time, and that texts on the subject written a decade ago are still good ref-
erences with content that has not changed very much. The major changes have been in the
understanding of how these devices work and their full range of capabilities, and in
improved methods of teaching the fundamentals associated with them. The benefit of all
this to the new student of the subject is that the material in this text will, we hope, have
reached a level where it is relatively easy to grasp and the information will have applica-
tion for years to come.
The miniaturization that has occurred in recent years leaves us to wonder about its limits.
Complete systems now appear on wafers thousands of times smaller than the single element
of earlier networks. The first integrated circuit (IC) was developed by Jack Kilby while
working at Texas Instruments in 1958 (Fig. 1.1). Today, the Intel® CoreTM i7 Extreme

1
2 SEMICONDUCTOR Edition Processor of Fig. 1.2 has 731 million transistors in a package that is only slightly
DIODES larger than a 1.67 sq. inches. In 1965, Dr. Gordon E. Moore presented a paper predicting that
the transistor count in a single IC chip would double every two years. Now, more than
45 years, later we find that his prediction is amazingly accurate and expected to continue
for the next few decades. We have obviously reached a point where the primary purpose
of the container is simply to provide some means for handling the device or system and to
provide a mechanism for attachment to the remainder of the network. Further miniaturiza-
tion appears to be limited by four factors: the quality of the semiconductor material, the
network design technique, the limits of the manufacturing and processing equipment, and
the strength of the innovative spirit in the semiconductor industry.
The first device to be introduced here is the simplest of all electronic devices, yet has a
range of applications that seems endless. We devote two chapters to the device to introduce
the materials commonly used in solid-state devices and review some fundamental laws of
electric circuits.

1.2 SEMICONDUCTOR MATERIALS: Ge, Si, AND GaAs

●
Jack St. Clair Kilby, inventor of the The construction of every discrete (individual) solid-state (hard crystal structure) electronic
integrated circuit and co-inventor of device or integrated circuit begins with a semiconductor material of the highest quality.
the electronic handheld calculator. Semiconductors are a special class of elements having a conductivity between that of a
(Courtesy of Texas Instruments.) good conductor and that of an insulator.
Born: Jefferson City, Missouri,1923. In general, semiconductor materials fall into one of two classes: single-crystal and
MS, University of Wisconsin. compound. Single-crystal semiconductors such as germanium (Ge) and silicon (Si) have a
Director of Engineering and Tech- repetitive crystal structure, whereas compound semiconductors such as gallium arsenide
nology, Components Group, Texas (GaAs), cadmium sulfide (CdS), gallium nitride (GaN), and gallium arsenide phosphide
Instruments. Fellow of the IEEE. (GaAsP) are constructed of two or more semiconductor materials of different atomic
Holds more than 60 U.S. patents. structures.
The three semiconductors used most frequently in the construction of electronic
devices are Ge, Si, and GaAs.
In the first few decades following the discovery of the diode in 1939 and the transis-
tor in 1947 germanium was used almost exclusively because it was relatively easy to
find and was available in fairly large quantities. It was also relatively easy to refine to
obtain very high levels of purity, an important aspect in the fabrication process. How-
ever, it was discovered in the early years that diodes and transistors constructed using
germanium as the base material suffered from low levels of reliability due primarily to
its sensitivity to changes in temperature. At the time, scientists were aware that another
material, silicon, had improved temperature sensitivities, but the refining process for
manufacturing silicon of very high levels of purity was still in the development stages.
The first integrated circuit, a phase-
Finally, however, in 1954 the first silicon transistor was introduced, and silicon quickly
shift oscillator, invented by Jack S.
became the semiconductor material of choice. Not only is silicon less temperature sensi-
Kilby in 1958. (Courtesy of Texas
Instruments.) tive, but it is one of the most abundant materials on earth, removing any concerns about
availability. The flood gates now opened to this new material, and the manufacturing
FIG. 1.1 and design technology improved steadily through the following years to the current high
Jack St. Clair Kilby. level of sophistication.
As time moved on, however, the field of electronics became increasingly sensitive to
issues of speed. Computers were operating at higher and higher speeds, and communica-
tion systems were operating at higher levels of performance. A semiconductor material
capable of meeting these new needs had to be found. The result was the development of
the first GaAs transistor in the early 1970s. This new transistor had speeds of operation
up to five times that of Si. The problem, however, was that because of the years of intense
design efforts and manufacturing improvements using Si, Si transistor networks for most
applications were cheaper to manufacture and had the advantage of highly efficient design
strategies. GaAs was more difficult to manufacture at high levels of purity, was more ex-
pensive, and had little design support in the early years of development. However, in time
the demand for increased speed resulted in more funding for GaAs research, to the point that
today it is often used as the base material for new high-speed, very large scale integrated
(VLSI) circuit designs.
 This brief review of the history of semiconductor materials is not meant to imply that COVALENT BONDING 3
GaAs will soon be the only material appropriate for solid-state construction. Germanium AND INTRINSIC
MATERIALS
devices are still being manufactured, although for a limited range of applications. Even
though it is a temperature-sensitive semiconductor, it does have characteristics that find
application in a limited number of areas. Given its availability and low manufacturing costs,
it will continue to find its place in product catalogs. As noted earlier, Si has the benefit of
years of development, and is the leading semiconductor material for electronic components
and ICs. In fact, Si is still the fundamental building block for Intel’s new line of processors.

1.3 COVALENT BONDING AND INTRINSIC MATERIALS

●
To fully appreciate why Si, Ge, and GaAs are the semiconductors of choice for the elec-
tronics industry requires some understanding of the atomic structure of each and how the
atoms are bound together to form a crystalline structure. The fundamental components of
an atom are the electron, proton, and neutron. In the lattice structure, neutrons and protons
form the nucleus and electrons appear in fixed orbits around the nucleus. The Bohr model
for the three materials is provided in Fig. 1.3.

Valence shell (Four valence electrons) Valence electron

FIG. 1.2
Shells Intel® Core™ i7 Extreme Edition
Processor.
+ +
Orbiting
electrons
Nucleus

Silicon Germanium

(a) (b)

Three valence Five valence

electrons electrons

+ +

Gallium Arsenic

(c)

FIG. 1.3
Atomic structure of (a) silicon; (b) germanium; and
(c) gallium and arsenic.

As indicated in Fig. 1.3, silicon has 14 orbiting electrons, germanium has 32 electrons,
gallium has 31 electrons, and arsenic has 33 orbiting electrons (the same arsenic that is
a very poisonous chemical agent). For germanium and silicon there are four electrons in
the outermost shell, which are referred to as valence electrons. Gallium has three valence
electrons and arsenic has five valence electrons. Atoms that have four valence electrons
are called tetravalent, those with three are called trivalent, and those with five are called
pentavalent. The term valence is used to indicate that the potential (ionization potential)
required to remove any one of these electrons from the atomic structure is significantly
lower than that required for any other electron in the structure.
4 SEMICONDUCTOR
DIODES – – –
– Si – – Si – – Si –
– – –
Sharing of electrons
– – –
– Si – – Si – – Si –
– – –
Valence electrons
– – –
– Si – – Si – – Si –
– – –

FIG. 1.4
Covalent bonding of the silicon atom.

In a pure silicon or germanium crystal the four valence electrons of one atom form a
bonding arrangement with four adjoining atoms, as shown in Fig. 1.4.
This bonding of atoms, strengthened by the sharing of electrons, is called covalent
bonding.
Because GaAs is a compound semiconductor, there is sharing between the two different
atoms, as shown in Fig. 1.5. Each atom, gallium or arsenic, is surrounded by atoms of the
complementary type. There is still a sharing of electrons similar in structure to that of Ge
and Si, but now five electrons are provided by the As atom and three by the Ga atom.

– – – –
– As – – As –
– – Ga – –
–
– –
Ga
– – – Ga
– – As – –
– – – – – –
– As – – – – As –
– – Ga Ga
– –
– –
– –
– As –
–

FIG. 1.5
Covalent bonding of the GaAs crystal.

Although the covalent bond will result in a stronger bond between the valence electrons
and their parent atom, it is still possible for the valence electrons to absorb sufficient kinetic
energy from external natural causes to break the covalent bond and assume the “free” state.
The term free is applied to any electron that has separated from the fixed lattice structure and
is very sensitive to any applied electric fields such as established by voltage sources or any
difference in potential. The external causes include effects such as light energy in the form
of photons and thermal energy (heat) from the surrounding medium. At room temperature
there are approximately 1.5 : 1010 free carriers in 1 cm3 of intrinsic silicon material, that
is, 15,000,000,000 (15 billion) electrons in a space smaller than a small sugar cube—an
enormous number.
The term intrinsic is applied to any semiconductor material that has been carefully ENERGY LEVELS 5
refined to reduce the number of impurities to a very low level—essentially as pure as
can be made available through modern technology.
The free electrons in a material due only to external causes are referred to as intrinsic car-
riers. Table 1.1 compares the number of intrinsic carriers per cubic centimeter (abbreviated ni)
for Ge, Si, and GaAs. It is interesting to note that Ge has the highest number and GaAs the
lowest. In fact, Ge has more than twice the number as GaAs. The number of carriers in the
intrinsic form is important, but other characteristics of the material are more significant
in determining its use in the field. One such factor is the relative mobility (mn) of the free
carriers in the material, that is, the ability of the free carriers to move throughout the mate-
rial. Table 1.2 clearly reveals that the free carriers in GaAs have more than five times the
mobility of free carriers in Si, a factor that results in response times using GaAs electronic
devices that can be up to five times those of the same devices made from Si. Note also that
free carriers in Ge have more than twice the mobility of electrons in Si, a factor that results
in the continued use of Ge in high-speed radio frequency applications.

TABLE 1.1
Intrinsic Carriers ni TABLE 1.2
Relative Mobility Factor mn
Intrinsic Carriers
Semiconductor (per cubic centimeter) Semiconductor Mn (cm2/V·s)

GaAs 1.7 : 106 Si 1500

Si 1.5 : 1010 Ge 3900
Ge 2.5 : 1013 GaAs 8500

One of the most important technological advances of recent decades has been the abil-
ity to produce semiconductor materials of very high purity. Recall that this was one of the
problems encountered in the early use of silicon—it was easier to produce germanium of
the required purity levels. Impurity levels of 1 part in 10 billion are common today, with
higher levels attainable for large-scale integrated circuits. One might ask whether these
extremely high levels of purity are necessary. They certainly are if one considers that the
addition of one part of impurity (of the proper type) per million in a wafer of silicon material
can change that material from a relatively poor conductor to a good conductor of electricity.
We obviously have to deal with a whole new level of comparison when we deal with the
semiconductor medium. The ability to change the characteristics of a material through this
process is called doping, something that germanium, silicon, and gallium arsenide readily
and easily accept. The doping process is discussed in detail in Sections 1.5 and 1.6.
One important and interesting difference between semiconductors and conductors is their
reaction to the application of heat. For conductors, the resistance increases with an increase
in heat. This is because the numbers of carriers in a conductor do not increase significantly
with temperature, but their vibration pattern about a relatively fixed location makes it in-
creasingly difficult for a sustained flow of carriers through the material. Materials that react
in this manner are said to have a positive temperature coefficient. Semiconductor materials,
however, exhibit an increased level of conductivity with the application of heat. As the tem-
perature rises, an increasing number of valence electrons absorb sufficient thermal energy to
break the covalent bond and to contribute to the number of free carriers. Therefore:
Semiconductor materials have a negative temperature coefficient.

1.4 ENERGY LEVELS

●
Within the atomic structure of each and every isolated atom there are specific energy levels
associated with each shell and orbiting electron, as shown in Fig. 1.6. The energy levels
associated with each shell will be different for every element. However, in general:
The farther an electron is from the nucleus, the higher is the energy state, and any
electron that has left its parent atom has a higher energy state than any electron in
the atomic structure.
Note in Fig. 1.6a that only specific energy levels can exist for the electrons in the atomic
structure of an isolated atom. The result is a series of gaps between allowed energy levels
6 SEMICONDUCTOR Energy
DIODES Valence level (outermost shell)
Energy gap
Second level (next inner shell)
Energy gap
Third level (etc.)
etc.

Nucleus

(a)

Energy
Electrons Energy Energy
Conduction band "free" to
establish
conduction Conduction band
– – – – The bands Conduction band
overlap – – – –
E g > 5 eV Eg
Unable to reach Valence band
conduction level – – Valence – – – –
– –
electrons
Valence band Conductor
– – – – bound to
Valence band the atomic
stucture
Insulator
E g = 0.67 eV (Ge)
E g = 1.1 eV (Si)
E g = 1.43 eV (GaAs)
Semiconductor
(b)

FIG. 1.6
Energy levels: (a) discrete levels in isolated atomic structures; (b) conduction and valence bands of an insulator,
a semiconductor, and a conductor.

where carriers are not permitted. However, as the atoms of a material are brought closer
together to form the crystal lattice structure, there is an interaction between atoms, which
will result in the electrons of a particular shell of an atom having slightly different energy
levels from electrons in the same orbit of an adjoining atom. The result is an expansion
of the fixed, discrete energy levels of the valence electrons of Fig. 1.6a to bands as shown
in Fig. 1.6b. In other words, the valence electrons in a silicon material can have varying
energy levels as long as they fall within the band of Fig. 1.6b. Figure l.6b clearly reveals
that there is a minimum energy level associated with electrons in the conduction band and
a maximum energy level of electrons bound to the valence shell of the atom. Between the
two is an energy gap that the electron in the valence band must overcome to become a free
carrier. That energy gap is different for Ge, Si, and GaAs; Ge has the smallest gap and GaAs
the largest gap. In total, this simply means that:
An electron in the valence band of silicon must absorb more energy than one in the
valence band of germanium to become a free carrier. Similarly, an electron in the
valence band of gallium arsenide must gain more energy than one in silicon or
germanium to enter the conduction band.
This difference in energy gap requirements reveals the sensitivity of each type of
semiconductor to changes in temperature. For instance, as the temperature of a Ge sample
increases, the number of electrons that can pick up thermal energy and enter the conduction
band will increase quite rapidly because the energy gap is quite small. However, the number
of electrons entering the conduction band for Si or GaAs would be a great deal less. This
sensitivity to changes in energy level can have positive and negative effects. The design of
photodetectors sensitive to light and security systems sensitive to heat would appear to be
an excellent area of application for Ge devices. However, for transistor networks, where
stability is a high priority, this sensitivity to temperature or light can be a detrimental factor.
 The energy gap also reveals which elements are useful in the construction of light-emitting n-TYPE AND p-TYPE 7
devices such as light-emitting diodes (LEDs), which will be introduced shortly. The wider MATERIALS
the energy gap, the greater is the possibility of energy being released in the form of visible
or invisible (infrared) light waves. For conductors, the overlapping of valence and conduc-
tion bands essentially results in all the additional energy picked up by the electrons being
dissipated in the form of heat. Similarly, for Ge and Si, because the energy gap is so small,
most of the electrons that pick up sufficient energy to leave the valence band end up in the
conduction band, and the energy is dissipated in the form of heat. However, for GaAs the
gap is sufficiently large to result in significant light radiation. For LEDs (Section 1.9) the
level of doping and the materials chosen determine the resulting color.
Before we leave this subject, it is important to underscore the importance of understand-
ing the units used for a quantity. In Fig. 1.6 the units of measurement are electron volts (eV).
The unit of measure is appropriate because W (energy) = QV (as derived from the defining
equation for voltage: V = W/Q). Substituting the charge of one electron and a potential dif-
ference of 1 V results in an energy level referred to as one electron volt.
That is,
W = QV
= (1.6 * 10-19 C)(1 V)
= 1.6 * 10-19 J
and

1 eV = 1.6 * 10-19 J (1.1)

1.5 n-TYPE AND p-TYPE MATERIALS

●
Because Si is the material used most frequently as the base (substrate) material in the con-
struction of solid-state electronic devices, the discussion to follow in this and the next few
sections deals solely with Si semiconductors. Because Ge, Si, and GaAs share a similar
covalent bonding, the discussion can easily be extended to include the use of the other
materials in the manufacturing process.
As indicated earlier, the characteristics of a semiconductor material can be altered sig-
nificantly by the addition of specific impurity atoms to the relatively pure semiconductor
material. These impurities, although only added at 1 part in 10 million, can alter the band
structure sufficiently to totally change the electrical properties of the material.
A semiconductor material that has been subjected to the doping process is called an
extrinsic material.
There are two extrinsic materials of immeasureable importance to semiconductor device
fabrication: n-type and p-type materials. Each is described in some detail in the following
subsections.

n-Type Material
Both n-type and p-type materials are formed by adding a predetermined number of impurity
atoms to a silicon base. An n-type material is created by introducing impurity elements that
have five valence electrons ( pentavalent), such as antimony, arsenic, and phosphorus. Each is
a member of a subset group of elements in the Periodic Table of Elements referred to as Group
V because each has five valence electrons. The effect of such impurity elements is indicated in
Fig. 1.7 (using antimony as the impurity in a silicon base). Note that the four covalent bonds
are still present. There is, however, an additional fifth electron due to the impurity atom, which
is unassociated with any particular covalent bond. This remaining electron, loosely bound to
its parent (antimony) atom, is relatively free to move within the newly formed n-type material.
Since the inserted impurity atom has donated a relatively “free” electron to the structure:
Diffused impurities with five valence electrons are called donor atoms.
It is important to realize that even though a large number of free carriers have been estab-
lished in the n-type material, it is still electrically neutral since ideally the number of posi-
tively charged protons in the nuclei is still equal to the number of free and orbiting negatively
charged electrons in the structure.
8 SEMICONDUCTOR
DIODES – – –
– Si – – Si – – Si –
– – –
Fifth valence electron
of antimony
– – – –
– Si – – Sb – – Si –
– – –
Antimony (Sb)
impurity
– – –
– Si – – Si – – Si –
– – –

FIG. 1.7
Antimony impurity in n-type material.

The effect of this doping process on the relative conductivity can best be described
through the use of the energy-band diagram of Fig. 1.8. Note that a discrete energy level
(called the donor level ) appears in the forbidden band with an Eg significantly less than that
of the intrinsic material. Those free electrons due to the added impurity sit at this energy
level and have less difficulty absorbing a sufficient measure of thermal energy to move into
the conduction band at room temperature. The result is that at room temperature, there are a
large number of carriers (electrons) in the conduction level, and the conductivity of the ma-
terial increases significantly. At room temperature in an intrinsic Si material there is about
one free electron for every 1012 atoms. If the dosage level is 1 in 10 million (107), the ratio
1012>107 ⫽ 105 indicates that the carrier concentration has increased by a ratio of 100,000:1.

Energy

Conduction band
E g = considerably less than in Fig. 1.6(b) for semiconductors
– – – –
Eg for intrinsic Donor energy level
materials – – – –
Valence band

FIG. 1.8
Effect of donor impurities on the energy band structure.

p-Type Material
The p-type material is formed by doping a pure germanium or silicon crystal with impurity
atoms having three valence electrons. The elements most frequently used for this purpose
are boron, gallium, and indium. Each is a member of a subset group of elements in the Peri-
odic Table of Elements referred to as Group III because each has three valence electrons.
The effect of one of these elements, boron, on a base of silicon is indicated in Fig. 1.9.
Note that there is now an insufficient number of electrons to complete the covalent bonds
of the newly formed lattice. The resulting vacancy is called a hole and is represented by a
small circle or a plus sign, indicating the absence of a negative charge. Since the resulting
vacancy will readily accept a free electron:
The diffused impurities with three valence electrons are called acceptor atoms.
The resulting p-type material is electrically neutral, for the same reasons described for
the n-type material.
 n-TYPE AND p-TYPE 9
– – – MATERIALS
– Si – – Si – – Si –
– – Void –
(O or +)
– – –
– Si – – B – Si –
– – Boron (B) –
impurity
– – –
– Si – – Si – – Si –
– – –

FIG. 1.9
Boron impurity in p-type material.

Electron versus Hole Flow

The effect of the hole on conduction is shown in Fig. 1.10. If a valence electron acquires
sufficient kinetic energy to break its covalent bond and fills the void created by a hole, then
a vacancy, or hole, will be created in the covalent bond that released the electron. There is,
therefore, a transfer of holes to the left and electrons to the right, as shown in Fig. 1.10.
The direction to be used in this text is that of conventional flow, which is indicated by the
direction of hole flow.

– –
–
– Si – – B – – Si – – B – – Si – – B –
– – – – – –

(a) (c)
Hole flow
Electron flow
(b)

FIG. 1.10
Electron versus hole flow.

Majority and Minority Carriers

In the intrinsic state, the number of free electrons in Ge or Si is due only to those few elec-
trons in the valence band that have acquired sufficient energy from thermal or light sources
to break the covalent bond or to the few impurities that could not be removed. The vacan-
cies left behind in the covalent bonding structure represent our very limited supply of
holes. In an n-type material, the number of holes has not changed significantly from this
intrinsic level. The net result, therefore, is that the number of electrons far outweighs the
number of holes. For this reason:
In an n-type material (Fig. 1.11a) the electron is called the majority carrier and the
hole the minority carrier.
For the p-type material the number of holes far outweighs the number of electrons, as
shown in Fig. 1.11b. Therefore:
In a p-type material the hole is the majority carrier and the electron is the minority carrier.
When the fifth electron of a donor atom leaves the parent atom, the atom remaining ac-
quires a net positive charge: hence the plus sign in the donor-ion representation. For similar
reasons, the minus sign appears in the acceptor ion.
10 SEMICONDUCTOR Donor ions Acceptor ions
DIODES
Majority
+ –– + – + – – –
– –
carriers – –
+
– + – – + – – – –
–
+ –
– + + Minority Majority – –
– + – – + carrier carriers – –
Minority
n-type p-type carrier

(a) (b)

FIG. 1.11
(a) n-type material; (b) p-type material.

The n- and p-type materials represent the basic building blocks of semiconductor devices.
We will find in the next section that the “joining” of a single n-type material with a p-type ma-
terial will result in a semiconductor element of considerable importance in electronic systems.

1.6 SEMICONDUCTOR DIODE

●
Now that both n- and p-type materials are available, we can construct our first solid-state
electronic device: The semiconductor diode, with applications too numerous to mention, is
created by simply joining an n-type and a p-type material together, nothing more, just the
joining of one material with a majority carrier of electrons to one with a majority carrier of
holes. The basic simplicity of its construction simply reinforces the importance of the
development of this solid-state era.

No Applied Bias (V ⴝ 0 V)
At the instant the two materials are “joined” the electrons and the holes in the region of the
junction will combine, resulting in a lack of free carriers in the region near the junction, as
shown in Fig. 1.12a. Note in Fig. 1.12a that the only particles displayed in this region are
the positive and the negative ions remaining once the free carriers have been absorbed.
This region of uncovered positive and negative ions is called the depletion region due
to the “depletion” of free carriers in the region.
If leads are connected to the ends of each material, a two-terminal device results, as
shown in Figs. 1.12a and 1.12b. Three options then become available: no bias, forward
bias, and reverse bias. The term bias refers to the application of an external voltage across
the two terminals of the device to extract a response. The condition shown in Figs. 1.12a
and 1.12b is the no-bias situation because there is no external voltage applied. It is simply
a diode with two leads sitting isolated on a laboratory bench. In Fig. 1.12b the symbol for
a semiconductor diode is provided to show its correspondence with the p–n junction. In
each figure it is clear that the applied voltage is 0 V (no bias) and the resulting current is
0 A, much like an isolated resistor. The absence of a voltage across a resistor results in
zero current through it. Even at this early point in the discussion it is important to note the
polarity of the voltage across the diode in Fig. 1.12b and the direction given to the current.
Those polarities will be recognized as the defined polarities for the semiconductor diode.
If a voltage applied across the diode has the same polarity across the diode as in Fig. 1.12b,
it will be considered a positive voltage. If the reverse, it is a negative voltage. The same
standards can be applied to the defined direction of current in Fig. 1.12b.
Under no-bias conditions, any minority carriers (holes) in the n-type material that find
themselves within the depletion region for any reason whatsoever will pass quickly into the
p-type material. The closer the minority carrier is to the junction, the greater is the attraction
for the layer of negative ions and the less is the opposition offered by the positive ions in
the depletion region of the n-type material. We will conclude, therefore, for future discus-
sions, that any minority carriers of the n-type material that find themselves in the depletion
region will pass directly into the p-type material. This carrier flow is indicated at the top of
Fig. 1.12c for the minority carriers of each material.
 Depletion region SEMICONDUCTOR DIODE 11

–– – ++
– – – + + + – – +–
– – – –– –– + + – + –
– ++
– – ++ + – + –
– – – ++ +
– – – – ++ – –
– – – + + +
– – – ++ – – + – Metal contact

p n

ID = 0 mA ID = 0 mA
+ VD = 0 V –
(no bias)

(a)

Minority carrier flow

+ VD = 0 V –
Ielectron Ihole
(no bias)

ID = 0 mA
Ihole Ielectron

p n
Majority carrier flow

(b) (c)

FIG. 1.12
A p–n junction with no external bias: (a) an internal distribution of charge; (b) a diode symbol,
with the defined polarity and the current direction; (c) demonstration that the net carrier
flow is zero at the external terminal of the device when VD ⫽ 0 V.

The majority carriers (electrons) of the n-type material must overcome the attractive
forces of the layer of positive ions in the n-type material and the shield of negative ions in
the p-type material to migrate into the area beyond the depletion region of the p-type mate-
rial. However, the number of majority carriers is so large in the n-type material that there
will invariably be a small number of majority carriers with sufficient kinetic energy to pass
through the depletion region into the p-type material. Again, the same type of discussion
can be applied to the majority carriers (holes) of the p-type material. The resulting flow due
to the majority carriers is shown at the bottom of Fig. 1.12c.
A close examination of Fig. 1.12c will reveal that the relative magnitudes of the flow
vectors are such that the net flow in either direction is zero. This cancellation of vectors
for each type of carrier flow is indicated by the crossed lines. The length of the vector
representing hole flow is drawn longer than that of electron flow to demonstrate that the
two magnitudes need not be the same for cancellation and that the doping levels for each
material may result in an unequal carrier flow of holes and electrons. In summary, therefore:
In the absence of an applied bias across a semiconductor diode, the net flow of charge
in one direction is zero.
In other words, the current under no-bias conditions is zero, as shown in Figs. 1.12a
and 1.12b.

Reverse-Bias Condition (VD * 0 V)

If an external potential of V volts is applied across the p–n junction such that the positive
terminal is connected to the n-type material and the negative terminal is connected to the
p-type material as shown in Fig. 1.13, the number of uncovered positive ions in the deple-
tion region of the n-type material will increase due to the large number of free electrons
drawn to the positive potential of the applied voltage. For similar reasons, the number of
uncovered negative ions will increase in the p-type material. The net effect, therefore, is a
12 SEMICONDUCTOR Is Minority-carrier flow
DIODES Imajority ⬵ 0A

– – – –+++ + –
– – – – + + + –+ + –
–
–
VD
+
– – – – –+++ – +
– – – – – + + + –+ –+ Is
–– – –+++
p n
Depletion region
p n
Is Is
– +
(Opposite)
–V +
D

(a) (b)

FIG. 1.13
Reverse-biased p–n junction: (a) internal distribution of charge under
reverse-bias conditions; (b) reverse-bias polarity and direction of reverse
saturation current.

widening of the depletion region. This widening of the depletion region will establish too
great a barrier for the majority carriers to overcome, effectively reducing the majority car-
rier flow to zero, as shown in Fig. 1.13a.
The number of minority carriers, however, entering the depletion region will not change,
resulting in minority-carrier flow vectors of the same magnitude indicated in Fig. 1.12c
with no applied voltage.
The current that exists under reverse-bias conditions is called the reverse saturation
current and is represented by Is.
The reverse saturation current is seldom more than a few microamperes and typically in
nA, except for high-power devices. The term saturation comes from the fact that it reaches its
maximum level quickly and does not change significantly with increases in the reverse-bias
potential, as shown on the diode characteristics of Fig. 1.15 for VD ⬍ 0 V. The reverse-biased
conditions are depicted in Fig. 1.13b for the diode symbol and p–n junction. Note, in particu-
lar, that the direction of Is is against the arrow of the symbol. Note also that the negative side of
the applied voltage is connected to the p-type material and the positive side to the n-type ma-
terial, the difference in underlined letters for each region revealing a reverse-bias condition.

Forward-Bias Condition (VD + 0 V)

A forward-bias or “on” condition is established by applying the positive potential to the
p-type material and the negative potential to the n-type material as shown in Fig. 1.14.
The application of a forward-bias potential VD will “pressure” electrons in the n-type mate-
rial and holes in the p-type material to recombine with the ions near the boundary and reduce
the width of the depletion region as shown in Fig. 1.14a. The resulting minority-carrier flow

– – – –+ + +
–+ + +
– – –+
–+ + + +
– – – –+

+ –

(a) (b)

FIG. 1.14
Forward-biased p–n junction: (a) internal distribution of charge under forward-bias
conditions; (b) forward-bias polarity and direction of resulting current.
of electrons from the p-type material to the n-type material (and of holes from the n-type SEMICONDUCTOR DIODE 13
material to the p-type material) has not changed in magnitude (since the conduction level is
controlled primarily by the limited number of impurities in the material), but the reduction
in the width of the depletion region has resulted in a heavy majority flow across the junc-
tion. An electron of the n-type material now “sees” a reduced barrier at the junction due to
the reduced depletion region and a strong attraction for the positive potential applied to the
p-type material. As the applied bias increases in magnitude, the depletion region will con-
tinue to decrease in width until a flood of electrons can pass through the junction, resulting
in an exponential rise in current as shown in the forward-bias region of the characteristics
of Fig. 1.15. Note that the vertical scale of Fig. 1.15 is measured in milliamperes (although
some semiconductor diodes have a vertical scale measured in amperes), and the horizontal
scale in the forward-bias region has a maximum of 1 V. Typically, therefore, the voltage
across a forward-biased diode will be less than 1 V. Note also how quickly the current rises
beyond the knee of the curve.
It can be demonstrated through the use of solid-state physics that the general charac-
teristics of a semiconductor diode can be defined by the following equation, referred to as
Shockley’s equation, for the forward- and reverse-bias regions:

ID = Is(eVD>nVT - 1) (A) (1.2)

where Is is the reverse saturation current

VD is the applied forward-bias voltage across the diode
n is an ideality factor, which is a function of the operating conditions and physi-
cal construction; it has a range between 1 and 2 depending on a wide variety of
factors (n ⫽ 1 will be assumed throughout this text unless otherwise noted).
The voltage VT in Eq. (1.1) is called the thermal voltage and is determined by

kTK
VT = (V) (1.3)
q

where k is Boltzmann’s constant ⫽ 1.38 ⫻ 10⫺23 J/K

TK is the absolute temperature in kelvins ⫽ 273 ⫹ the temperature in °C
q is the magnitude of electronic charge ⫽ 1.6 ⫻ 10⫺19 C

EXAMPLE 1.1 At a temperature of 27°C (common temperature for components in an

enclosed operating system), determine the thermal voltage VT.
Solution: Substituting into Eq. (1.3), we obtain
T = 273 + ⬚C = 273 + 27 = 300 K
kTK (1.38 * 10-23 J/K)(30 K)
VT = =
q 1.6 * 10-19 C
= 25.875 mV ⬵ 26 mV
The thermal voltage will become an important parameter in the analysis to follow in this
chapter and a number of those to follow.

Initially, Eq. (1.2) with all its defined quantities may appear somewhat complex. How-
ever, it will not be used extensively in the analysis to follow. It is simply important at this
point to understand the source of the diode characteristics and which factors affect its shape.
A plot of Eq. (1.2) with Is ⫽ 10 pA is provided in Fig. 1.15 as the dashed line. If we
expand Eq. (1.2) into the following form, the contributing component for each region of
Fig. 1.15 can be described with increased clarity:
ID = IseVD>nVT - Is
For positive values of VD the first term of the above equation will grow very quickly and
totally overpower the effect of the second term. The result is the following equation, which
only has positive values and takes on the exponential format ex appearing in Fig. 1.16:
ID ⬵ IseVD>nVT (VD positive)
14 SEMICONDUCTOR ID (mA)
DIODES
20
19
Actual commercially
18
available unit
17
16 Eq. (1.1)
15
14
13
12
Defined polarity and
11 direction for graph
10 VD
+ –
9
8 ID
7 Forward-bias region
6 (V
VD > 0 V, II D > 0 mA)
5
4
3
2
1

–40 –30 –20 –10 0 0.3 0.5 0.7 1 V D (V)

– 10 pA
Reverse-bias region No-bias
– 20 pA
(VD < 0 V, ID = –Is ) (VD = 0 V, ID = 0 mA)
– 30 pA
– 40 pA
– 50 pA

FIG. 1.15
Silicon semiconductor diode characteristics.

The exponential curve of Fig. 1.16 increases very rapidly with increasing values of x.
At x ⫽ 0, e0 ⫽ 1, whereas at x ⫽ 5, it jumps to greater than 148. If we continued to x ⫽ 10,
the curve jumps to greater than 22,000. Clearly, therefore, as the value of x increases, the
curve becomes almost vertical, an important conclusion to keep in mind when we examine
the change in current with increasing values of applied voltage.

ex ex
e5.5 ⬵ 244.7

200
5

150 e5 ⬵ 148.4
e1 ⫽ e ⬵ 2.718

1 100

0 1 2 x
50 e4 ⬵ 54.6
e0 ⫽ 1
e3 ⬵ 20.1

0 1 2 3 4 5 6 7 x

FIG. 1.16
Plot of ex.
 For negative values of VD the exponential term drops very quickly below the level of I, SEMICONDUCTOR DIODE 15
and the resulting equation for ID is simply
ID ⬵ -Is (VD negative)
Note in Fig. 1.15 that for negative values of VD the current is essentially horizontal at
the level of ⫺Is.
At V ⫽ 0 V, Eq. (1.2) becomes
ID = Is(e0 - 1) = Is(1 - 1) = 0 mA
as confirmed by Fig. 1.15.
The sharp change in direction of the curve at VD ⫽ 0 V is simply due to the change in
current scales from above the axis to below the axis. Note that above the axis the scale is in
milliamperes (mA), whereas below the axis it is in picoamperes (pA).
Theoretically, with all things perfect, the characteristics of a silicon diode should appear
as shown by the dashed line of Fig. 1.15. However, commercially available silicon diodes
deviate from the ideal for a variety of reasons including the internal “body” resistance and the
external “contact” resistance of a diode. Each contributes to an additional voltage at the same
current level, as determined by Ohm’s law, causing the shift to the right witnessed in Fig. 1.15.
The change in current scales between the upper and lower regions of the graph was noted
earlier. For the voltage VD there is also a measurable change in scale between the right-hand
region of the graph and the left-hand region. For positive values of VD the scale is in tenths
of volts, and for the negative region it is in tens of volts.
It is important to note in Fig. 1.14b how:
The defined direction of conventional current for the positive voltage region matches
the arrowhead in the diode symbol.
This will always be the case for a forward-biased diode. It may also help to note that the
forward-bias condition is established when the bar representing the negative side of the
applied voltage matches the side of the symbol with the vertical bar.
Going back a step further by looking at Fig. 1.14b, we find a forward-bias condition is
established across a p–n junction when the positive side of the applied voltage is applied to
the p-type material (noting the correspondence in the letter p) and the negative side of the
applied voltage is applied to the n-type material (noting the same correspondence).
It is particularly interesting to note that the reverse saturation current of the commercial
unit is significantly larger than that of Is in Shockley’s equation. In fact,
The actual reverse saturation current of a commercially available diode will normally
be measurably larger than that appearing as the reverse saturation current in
Shockley’s equation.
This increase in level is due to a wide range of factors that include
– leakage currents
– generation of carriers in the depletion region
– higher doping levels that result in increased levels of reverse current
– sensitivity to the intrinsic level of carriers in the component materials by a squared
factor—double the intrinsic level, and the contribution to the reverse current could
increase by a factor of four.
– a direct relationship with the junction area—double the area of the junction, and
the contribution to the reverse current could double. High-power devices that have
larger junction areas typically have much higher levels of reverse current.
– temperature sensitivity—for every 5°C increase in current, the level of reverse sat-
uration current in Eq. 1.2 will double, whereas a 10°C increase in current will result
in doubling of the actual reverse current of a diode.
Note in the above the use of the terms reverse saturation current and reverse current. The
former is simply due to the physics of the situation, whereas the latter includes all the other
possible effects that can increase the level of current.
We will find in the discussions to follow that the ideal situation is for Is to be 0 A in the
reverse-bias region. The fact that it is typically in the range of 0.01 pA to 10 pA today as
compared to 0.l mA to 1 mA a few decades ago is a credit to the manufacturing industry.
Comparing the common value of 1 nA to the 1-mA level of years past shows an improve-
ment factor of 100,000.
16 SEMICONDUCTOR Breakdown Region
DIODES
Even though the scale of Fig. 1.15 is in tens of volts in the negative region, there is a point
where the application of too negative a voltage with the reverse polarity will result in a
sharp change in the characteristics, as shown in Fig. 1.17. The current increases at a very
rapid rate in a direction opposite to that of the positive voltage region. The reverse-bias
potential that results in this dramatic change in characteristics is called the breakdown
potential and is given the label VBV.

ID

Is
VBV
0 VD

Zener
region

FIG. 1.17
Breakdown region.

As the voltage across the diode increases in the reverse-bias region, the velocity of the
minority carriers responsible for the reverse saturation current Is will also increase. Eventu-
ally, their velocity and associated kinetic energy (WK = 21 mv2) will be sufficient to release
additional carriers through collisions with otherwise stable atomic structures. That is, an
ionization process will result whereby valence electrons absorb sufficient energy to leave the
parent atom. These additional carriers can then aid the ionization process to the point where
a high avalanche current is established and the avalanche breakdown region determined.
The avalanche region (VBV) can be brought closer to the vertical axis by increasing the
doping levels in the p- and n-type materials. However, as VBV decreases to very low levels,
such as ⫺5 V, another mechanism, called Zener breakdown, will contribute to the sharp
change in the characteristic. It occurs because there is a strong electric field in the region
of the junction that can disrupt the bonding forces within the atom and “generate” carriers.
Although the Zener breakdown mechanism is a significant contributor only at lower levels
of VBV, this sharp change in the characteristic at any level is called the Zener region, and
diodes employing this unique portion of the characteristic of a p–n junction are called Zener
diodes. They are described in detail in Section 1.15.
The breakdown region of the semiconductor diode described must be avoided if the
response of a system is not to be completely altered by the sharp change in characteristics
in this reverse-voltage region.
The maximum reverse-bias potential that can be applied before entering the break-
down region is called the peak inverse voltage (referred to simply as the PIV rating) or
the peak reverse voltage (denoted the PRV rating).
If an application requires a PIV rating greater than that of a single unit, a number of
diodes of the same characteristics can be connected in series. Diodes are also connected in
parallel to increase the current-carrying capacity.
In general, the breakdown voltage of GaAs diodes is about 10% higher those for silicon
diodes but after 200% higher than levels for Ge diodes.

Ge, Si, and GaAs

The discussion thus far has solely used Si as the base semiconductor material. It is now impor-
tant to compare it to the other two materials of importance: GaAs and Ge. A plot comparing
the characteristics of Si, GaAs, and Ge diodes is provided in Fig. 1.18. The curves are not
 SEMICONDUCTOR DIODE 17
ID (mA)

30

25

20
Ge Si GaAs
15

10

VBV (GaAs) 100 V 50 V

0.3 0.7 1.0 1.2 VD (V)
Is (GaAs)
VK (Ge) VK (Si) VK (GaAs)
5 ␳A
VBV (Si)
10 ␳A
Is (Si)

VBV (Ge)
1 µA
Is (Ge)

FIG. 1.18
Comparison of Ge, Si, and GaAs commercial diodes.

simply plots of Eq. 1.2 but the actual response of commercially available units. The total reverse TABLE 1.3
current is shown and not simply the reverse saturation current. It is immediately obvious that Knee Voltages VK
the point of vertical rise in the characteristics is different for each material, although the general
Semiconductor VK (V)
shape of each characteristic is quite similar. Germanium is closest to the vertical axis and GaAs
is the most distant. As noted on the curves, the center of the knee (hence the K is the notation Ge 0.3
VK) of the curve is about 0.3 V for Ge, 0.7 V for Si, and 1.2 V for GaAs (see Table 1.3). Si 0.7
The shape of the curve in the reverse-bias region is also quite similar for each material, GaAs 1.2
but notice the measurable difference in the magnitudes of the typical reverse saturation
currents. For GaAs, the reverse saturation current is typically about 1 pA, compared to 10 pA
for Si and 1 mA for Ge, a significant difference in levels.
Also note the relative magnitudes of the reverse breakdown voltages for each material.
GaAs typically has maximum breakdown levels that exceed those of Si devices of the same
power level by about 10%, with both having breakdown voltages that typically extend be-
tween 50 V and 1 kV. There are Si power diodes with breakdown voltages as high as 20 kV.
Germanium typically has breakdown voltages of less than 100 V, with maximums around
400 V. The curves of Fig. 1.18 are simply designed to reflect relative breakdown voltages
for the three materials. When one considers the levels of reverse saturation currents and
breakdown voltages, Ge certainly sticks out as having the least desirable characteristics.
A factor not appearing in Fig. 1.18 is the operating speed for each material—an impor- TABLE 1.4
tant factor in today’s market. For each material, the electron mobility factor is provided Electron Mobility mn
in Table 1.4. It provides an indication of how fast the carriers can progress through the
material and therefore the operating speed of any device made using the materials. Quite Semiconductor Mn(cm2/V~ s)
obviously, GaAs stands out, with a mobility factor more than five times that of silicon and Ge 3900
twice that of germanium. The result is that GaAs and Ge are often used in high-speed ap- Si 1500
plications. However, through proper design, careful control of doping levels, and so on, GaAs 8500
silicon is also found in systems operating in the gigahertz range. Research today is also
looking at compounds in groups III–V that have even higher mobility factors to ensure that
industry can meet the demands of future high-speed requirements.
18 SEMICONDUCTOR
DIODES EXAMPLE 1.2 Using the curves of Fig 1.18:
a. Determine the voltage across each diode at a current of 1 mA.
b. Repeat for a current of 4 mA.
c. Repeat for a current of 30 mA.
d. Determine the average value of the diode voltage for the range of currents listed above.
e. How do the average values compare to the knee voltages listed in Table 1.3?
Solution:
a. VD(Ge) ⫽ 0.2 V, VD(Si) ⫽ 0.6 V, VD (GaAs) ⫽ 1.1 V
b. VD(Ge) ⫽ 0.3 V, VD(Si) ⫽ 0.7 V, VD (GaAs) ⫽ 1.2 V
c. VD(Ge) ⫽ 0.42 V, VD(Si) ⫽ 0.82 V, VD (GaAs) ⫽ 1.33 V
d. Ge: Vav ⫽ (0.2 V ⫹ 0.3 V ⫹ 0.42 V)>3 ⫽ 0.307 V
Si: Vav ⫽ (0.6 V ⫹ 0.7 V ⫹ 0.82 V)>3 ⫽ 0.707 V
GaAs: Vav ⫽ (1.1 V ⫹ 1.2 V ⫹ 1.33 V)>3 ⫽ 1.21 V
e. Very close correspondence. Ge: 0.307 V vs. 0.3, V, Si: 0.707 V vs. 0.7 V, GaAs: 1.21 V
vs. 1.2 V.

Temperature Effects
Temperature can have a marked effect on the characteristics of a semiconductor diode, as
demonstrated by the characteristics of a silicon diode shown in Fig. 1.19:
In the forward-bias region the characteristics of a silicon diode shift to the left at a rate
of 2.5 mV per centigrade degree increase in temperature.

ID (mA)
Shift to left = (100°C)(–2.5 mV/°C) = –0.35 V

30
–75°C
125°C
25°C

25

20
Increasing Decreasing
temperature temperature
15

10

Silicon diode at
Is ⬵ 0.01 ␳A 5 room temperature

40 30 20 10
0.7 V VD (V)
1 ␳A
Silicon diode at
room temperature Increasing
temperature
Increasing
temperature

1 µA
–75°C

25°C

125°C

FIG. 1.19
Variation in Si diode characteristics with temperature change.
 An increase from room temperature (20°C) to 100°C (the boiling point of water) results SEMICONDUCTOR DIODE 19
in a drop of 80(2.5 mV) ⫽ 200 mV, or 0.2 V, which is significant on a graph scaled in
tenths of volts. A decrease in temperature has the reverse effect, as also shown in the figure:
In the reverse-bias region the reverse current of a silicon diode doubles for every 10°C
rise in temperature.
For a change from 20°C to 100°C, the level of Is increases from 10 nA to a value of
2.56 mA, which is a significant, 256-fold increase. Continuing to 200°C would result in a
monstrous reverse saturation current of 2.62 mA. For high-temperature applications one
would therefore look for Si diodes with room-temperature Is closer to 10 pA, a level com-
monly available today, which would limit the current to 2.62 μA. It is indeed fortunate that
both Si and GaAs have relatively small reverse saturation currents at room temperature.
GaAs devices are available that work very well in the ⫺200°C to ⫹200°C temperature
range, with some having maximum temperatures approaching 400°C. Consider, for a mo-
ment, how huge the reverse saturation current would be if we started with a Ge diode with
a saturation current of 1 mA and applied the same doubling factor.
Finally, it is important to note from Fig. 1.19 that:
The reverse breakdown voltage of a semiconductor diode will increase or Russell Ohl (1898–1987)
decrease with temperature. American (Allentown, PA;
Holmdel, NJ; Vista, CA) Army
However, if the initial breakdown voltage is less than 5 V, the breakdown voltage may
Signal Corps, University of
actually decrease with temperature. The sensitivity of the breakdown potential to changes Colorado, Westinghouse, AT&T,
of temperature will be examined in more detail in Section 1.15. Bell Labs Fellow, Institute of
Radio Engineers—1955
Summary (Courtesy of AT&T Archives
History Center.)
A great deal has been introduced in the foregoing paragraphs about the construction of a Although vacuum tubes were
semiconductor diode and the materials employed. The characteristics have now been pre- used in all forms of communication
sented and the important differences between the response of the materials discussed. It is in the 1930s, Russell Ohl was deter-
now time to compare the p–n junction response to the desired response and reveal the pri- mined to demonstrate that the future
mary functions of a semiconductor diode. of the field was defined by semicon-
Table 1.5 provides a synopsis of material regarding the three most frequently used semi- ductor crystals. Germanium was not
conductor materials. Figure 1.20 includes a short biography of the first research scientist to immediately available for his
discover the p–n junction in a semiconductor material. research, so he turned to silicon, and
found a way to raise its level of
purity to 99.8%, for which he
received a patent. The actual discov-
TABLE 1.5 ery of the p–n junction, as often
The Current Commercial Use of Ge, Si, and GaAs happens in scientific research, was
the result of a set of circumstances
Ge: Germanium is in limited production due to its temperature sensitivity and high that were not planned. On February
reverse saturation current. It is still commercially available but is limited to 23, 1940, Ohl found that a silicon
some high-speed applications (due to a relatively high mobility factor) and crystal with a crack down the mid-
applications that use its sensitivity to light and heat such as photodetectors dle would produce a significant rise
and security systems. in current when placed near a source
Si: Without question the semiconductor used most frequently for the full range of of light. This discovery led to fur-
electronic devices. It has the advantage of being readily available at low cost ther research, which revealed that
and has relatively low reverse saturation currents, good temperature character- the purity levels on each side of the
istics, and excellent breakdown voltage levels. It also benefits from decades of crack were different and that a
enormous attention to the design of large-scale integrated circuits and process- barrier was formed at the junction
ing technology. that allowed the passage of current
GaAs: Since the early 1990s the interest in GaAs has grown in leaps and bounds, and it in only one direction—the first
will eventually take a good share of the development from silicon devices, solid-state diode had been identified
especially in very large scale integrated circuits. Its high-speed characteristics and explained. In addition, this sen-
are in more demand every day, with the added features of low reverse satura- sitivity to light was the beginning of
tion currents, excellent temperature sensitivities, and high breakdown voltages. the development of solar cells. The
More than 80% of its applications are in optoelectronics with the development results were quite instrumental in
of light-emitting diodes, solar cells, and other photodetector devices, but that the development of the transistor in
will probably change dramatically as its manufacturing costs drop and its use 1945 by three individuals also work-
in integrated circuit design continues to grow; perhaps the semiconductor ing at Bell Labs.
material of the future.
FIG. 1.20
20 SEMICONDUCTOR 1.7 IDEAL VERSUS PRACTICAL
DIODES ●
In the previous section we found that a p–n junction will permit a generous flow of charge
when forward-biased and a very small level of current when reverse-biased. Both condi-
tions are reviewed in Fig. 1.21, with the heavy current vector in Fig. 1.21a matching the
direction of the arrow in the diode symbol and the significantly smaller vector in the oppo-
site direction in Fig. 1.21b representing the reverse saturation current.
An analogy often used to describe the behavior of a semiconductor diode is a mechanical
switch. In Fig. 1.21a the diode is acting like a closed switch permitting a generous flow of
charge in the direction indicated. In Fig. 1.21b the level of current is so small in most cases
that it can be approximated as 0 A and represented by an open switch.

VD VD
+ – – +

ID Is

(a) (b)

FIG. 1.21
Ideal semiconductor diode: (a) forward-
biased; (b) reverse-biased.

In other words:
The semiconductor diode behaves in a manner similar to a mechanical switch in that it
can control whether current will flow between its two terminals.
However, it is important to also be aware that:
The semiconductor diode is different from a mechanical switch in the sense that when
the switch is closed it will only permit current to flow in one direction.
Ideally, if the semiconductor diode is to behave like a closed switch in the forward-bias
region, the resistance of the diode should be 0 ⍀. In the reverse-bias region its resistance
should be ⬁⍀ to represent the open-circuit equivalent. Such levels of resistance in the forward-
and reverse-bias regions result in the characteristics of Fig. 1.22.

ID

Ideal characteristics

10 mA

ID

⫺20 V

0.7 V VD

Actual characteristics
Is ⬵ 0 mA

FIG. 1.22
Ideal versus actual semiconductor characteristics.
 The characteristics have been superimposed to compare the ideal Si diode to a real-world RESISTANCE LEVELS 21
Si diode. First impressions might suggest that the commercial unit is a poor impression of
the ideal switch. However, when one considers that the only major difference is that the
commercial diode rises at a level of 0.7 V rather than 0 V, there are a number of similarities
between the two plots.
When a switch is closed the resistance between the contacts is assumed to be 0 ⍀. At
the plot point chosen on the vertical axis the diode current is 5 mA and the voltage across
the diode is 0 V. Substituting into Ohm’s law results in
VD 0V
RF = = = 0⍀ (short@circuit equivalent)
ID 5 mA
In fact:
At any current level on the vertical line, the voltage across the ideal diode is 0 V and
the resistance is 0 ⍀.
For the horizontal section, if we again apply Ohm’s law, we find
VD 20 V
RR = = ⬵ ⬁⍀ (open@circuit equivalent)
ID 0 mA
Again:
Because the current is 0 mA anywhere on the horizontal line, the resistance is
considered to be infinite ohms (an open-circuit) at any point on the axis.
Due to the shape and the location of the curve for the commercial unit in the forward-bias
region there will be a resistance associated with the diode that is greater than 0 ⍀. However,
if that resistance is small enough compared to other resistors of the network in series with
the diode, it is often a good approximation to simply assume the resistance of the com-
mercial unit is 0 ⍀. In the reverse-bias region, if we assume the reverse saturation current
is so small it can be approximated as 0 mA, we have the same open-circuit equivalence
provided by the open switch.
The result, therefore, is that there are sufficient similarities between the ideal switch and
the semiconductor diode to make it an effective electronic device. In the next section the
various resistance levels of importance are determined for use in the next chapter, where
the response of diodes in an actual network is examined.

1.8 RESISTANCE LEVELS

●
As the operating point of a diode moves from one region to another the resistance of the
diode will also change due to the nonlinear shape of the characteristic curve. It will be dem-
onstrated in the next few paragraphs that the type of applied voltage or signal will define the
resistance level of interest. Three different levels will be introduced in this section, which
will appear again as we examine other devices. It is therefore paramount that their determi-
nation be clearly understood.

DC or Static Resistance
The application of a dc voltage to a circuit containing a semiconductor diode will result in
an operating point on the characteristic curve that will not change with time. The resistance
of the diode at the operating point can be found simply by finding the corresponding levels
of VD and ID as shown in Fig. 1.23 and applying the following equation:

VD
RD = (1.4)
ID

The dc resistance levels at the knee and below will be greater than the resistance levels
obtained for the vertical rise section of the characteristics. The resistance levels in the
reverse-bias region will naturally be quite high. Since ohmmeters typically employ a rela-
tively constant-current source, the resistance determined will be at a preset current level
(typically, a few milliamperes).
22 SEMICONDUCTOR
DIODES

FIG. 1.23
Determining the dc resistance of a diode at a
particular operating point.

In general, therefore, the higher the current through a diode, the lower is the dc resis-
tance level.
Typically, the dc resistance of a diode in the active (most utilized) will range from about
10 ⍀ to 80 ⍀.

EXAMPLE 1.3 Determine the dc resistance levels for the diode of Fig. 1.24 at
a. ID ⫽ 2 mA (low level)
b. ID ⫽ 20 mA (high level)
c. VD ⫽ ⫺10 V (reverse-biased)

FIG. 1.24
Example 1.3.

Solution:
a. At ID ⫽ 2 mA, VD ⫽ 0.5 V (from the curve) and
VD 0.5 V
RD = = = 250 ⍀
ID 2 mA
b. At ID ⫽ 20 mA, VD ⫽ 0.8 V (from the curve) and
VD 0.8 V
RD = = = 40 ⍀
ID 20 mA
c. At VD ⫽ ⫺10 V, ID ⫽ ⫺Is ⫽ ⫺1 mA (from the curve) and RESISTANCE LEVELS 23
VD 10 V
RD = = = 10 M⍀
ID 1 mA
clearly supporting some of the earlier comments regarding the dc resistance levels of a
diode.

AC or Dynamic Resistance
Eq. (1.4) and Example 1.3 reveal that
the dc resistance of a diode is independent of the shape of the characteristic in the
region surrounding the point of interest.
If a sinusoidal rather than a dc input is applied, the situation will change completely. The
varying input will move the instantaneous operating point up and down a region of the char-
acteristics and thus defines a specific change in current and voltage as shown in Fig. 1.25.
With no applied varying signal, the point of operation would be the Q-point appearing on
Fig. 1.25, determined by the applied dc levels. The designation Q-point is derived from the
word quiescent, which means “still or unvarying.”

FIG. 1.25
Defining the dynamic or ac resistance.

A straight line drawn tangent to the curve through the Q-point as shown in Fig. 1.26
will define a particular change in voltage and current that can be used to determine the ac
or dynamic resistance for this region of the diode characteristics. An effort should be made
to keep the change in voltage and current as small as possible and equidistant to either side
of the Q-point. In equation form,

⌬Vd
rd = (1.5)
⌬Id

where ⌬ signifies a finite change in the quantity.

The steeper the slope, the lower is the value of ⌬Vd for the same change in ⌬Id and the FIG. 1.26
lower is the resistance. The ac resistance in the vertical-rise region of the characteristic is Determining the ac resistance at a
therefore quite small, whereas the ac resistance is much higher at low current levels. Q-point.
In general, therefore, the lower the Q-point of operation (smaller current or lower
voltage), the higher is the ac resistance.
24 SEMICONDUCTOR
DIODES EXAMPLE 1.4 For the characteristics of Fig. 1.27:
a. Determine the ac resistance at ID ⫽ 2 mA.
b. Determine the ac resistance at ID ⫽ 25 mA.
c. Compare the results of parts (a) and (b) to the dc resistances at each current level.

I D (mA)

30

25 ∆ Id

20
∆Vd

15

10

5
4
2 ∆ Id

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 VD (V)
∆Vd

FIG. 1.27
Example 1.4.

Solution:
a. For ID ⫽ 2 mA, the tangent line at ID ⫽ 2 mA was drawn as shown in Fig. 1.27 and a
swing of 2 mA above and below the specified diode current was chosen. At ID ⫽ 4 mA,
VD ⫽ 0.76 V, and at ID ⫽ 0 mA, VD ⫽ 0.65 V. The resulting changes in current and
voltage are, respectively,
⌬Id = 4 mA - 0 mA = 4 mA
and ⌬Vd = 0.76 V - 0.65 V = 0.11 V
and the ac resistance is
⌬Vd 0.11 V
rd = = = 27.5 ⍀
⌬Id 4 mA
b. For ID ⫽ 25 mA, the tangent line at ID ⫽ 25 mA was drawn as shown in Fig. 1.27 and
a swing of 5 mA above and below the specified diode current was chosen. At ID ⫽ 30 mA,
VD ⫽ 0.8 V, and at ID ⫽ 20 mA, VD ⫽ 0.78 V. The resulting changes in current and
voltage are, respectively,
⌬Id = 30 mA - 20 mA = 10 mA
and ⌬Vd = 0.8 V - 0.78 V = 0.02 V
and the ac resistance is
⌬Vd 0.02 V
rd = = = 2⍀
⌬Id 10 mA
c. For ID ⫽ 2 mA, VD ⫽ 0.7 V and
VD 0.7 V
RD = = = 350 ⍀
ID 2 mA
which far exceeds the rd of 27.5 ⍀.
 For ID ⫽ 25 mA, VD ⫽ 0.79 V and RESISTANCE LEVELS 25
VD 0.79 V
RD = = = 31.62 ⍀
ID 25 mA
which far exceeds the rd of 2 ⍀.

We have found the dynamic resistance graphically, but there is a basic definition in dif-
ferential calculus that states:
The derivative of a function at a point is equal to the slope of the tangent line drawn
at that point.
Equation (1.5), as defined by Fig. 1.26, is, therefore, essentially finding the derivative of
the function at the Q-point of operation. If we find the derivative of the general equation
(1.2) for the semiconductor diode with respect to the applied forward bias and then invert
the result, we will have an equation for the dynamic or ac resistance in that region. That is,
taking the derivative of Eq. (1.2) with respect to the applied bias will result in

3 I (eVD>nVT - 1) 4
d d
(I ) =
dVD D dVD s
and
dID 1
= (I + Is)
dVD nVT D
after we apply differential calculus. In general, ID W Is in the vertical-slope section of
the characteristics and
dID ID
⬵
dVD nVT
Flipping the result to define a resistance ratio (R ⫽ V/I) gives
dVD nVT
= rd =
dID ID
Substituting n ⫽ 1 and VT ⬵ 26 mV from Example 1.1 results in

26 mV
rd = (1.6)
ID

The significance of Eq. (1.6) must be clearly understood. It implies that

the dynamic resistance can be found simply by substituting the quiescent value of the
diode current into the equation.
There is no need to have the characteristics available or to worry about sketching tangent
lines as defined by Eq. (1.5). It is important to keep in mind, however, that Eq. (1.6) is
accurate only for values of ID in the vertical-rise section of the curve. For lesser values of
ID, n ⫽ 2 (silicon) and the value of rd obtained must be multiplied by a factor of 2. For
small values of ID below the knee of the curve, Eq. (1.6) becomes inappropriate.
All the resistance levels determined thus far have been defined by the p–n junction and
do not include the resistance of the semiconductor material itself (called body resistance)
and the resistance introduced by the connection between the semiconductor material and the
external metallic conductor (called contact resistance). These additional resistance levels
can be included in Eq. (1.6) by adding a resistance denoted rB:

26 mV
r⬘d = + rB ohms (1.7)
ID

The resistance r⬘d, therefore, includes the dynamic resistance defined by Eq. (1.6) and
the resistance rB just introduced. The factor rB can range from typically 0.1 ⍀ for high-
power devices to 2 ⍀ for some low-power, general-purpose diodes. For Example 1.4 the ac
resistance at 25 mA was calculated to be 2 ⍀. Using Eq. (1.6), we have
26 mV 26 mV
rd = = = 1.04 ⍀
ID 25 mA
26 SEMICONDUCTOR The difference of about 1 ⍀ could be treated as the contribution of rB.
DIODES For Example 1.4 the ac resistance at 2 mA was calculated to be 27.5 ⍀. Using Eq. (1.6)
but multiplying by a factor of 2 for this region (in the knee of the curve n ⫽ 2),

b = 2a b = 2(13 ⍀) = 26 ⍀
26 mV 26 mV
rd = 2a
ID 2 mA
The difference of 1.5 ⍀ could be treated as the contribution due to rB.
In reality, determining rd to a high degree of accuracy from a characteristic curve using Eq.
(1.5) is a difficult process at best and the results have to be treated with skepticism. At low lev-
els of diode current the factor rB is normally small enough compared to rd to permit ignoring
its impact on the ac diode resistance. At high levels of current the level of rB may approach that
of rd, but since there will frequently be other resistive elements of a much larger magnitude in
series with the diode, we will assume in this book that the ac resistance is determined solely
by rd, and the impact of rB will be ignored unless otherwise noted. Technological improve-
ments of recent years suggest that the level of rB will continue to decrease in magnitude and
eventually become a factor that can certainly be ignored in comparison to rd.
The discussion above centered solely on the forward-bias region. In the reverse-bias
region we will assume that the change in current along the Is line is nil from 0 V to the
Zener region and the resulting ac resistance using Eq. (1.5) is sufficiently high to permit
the open-circuit approximation.
Typically, the ac resistance of a diode in the active region will range from about 1 ⍀ to 100 ⍀.

Average AC Resistance
If the input signal is sufficiently large to produce a broad swing such as indicated in Fig.
1.28, the resistance associated with the device for this region is called the average ac resis-
tance. The average ac resistance is, by definition, the resistance determined by a straight
line drawn between the two intersections established by the maximum and minimum values
of input voltage. In equation form (note Fig. 1.28),

`
⌬Vd
rav = (1.8)
⌬Id pt. to pt.

For the situation indicated by Fig. 1.28,

⌬Id = 17 mA - 2 mA = 15 mA
I D (mA)

20

15

∆ Id 10

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 VD (V)
∆Vd

FIG. 1.28
Determining the average ac resistance between indicated limits.
and ⌬Vd = 0.725 V - 0.65 V = 0.075 V DIODE EQUIVALENT 27
⌬Vd 0.075 V CIRCUITS
with rav = = = 5⍀
⌬Id 15 mA
If the ac resistance (rd) were determined at ID ⫽ 2 mA, its value would be more than 5 ⍀,
and if determined at 17 mA, it would be less. In between, the ac resistance would make the
transition from the high value at 2 mA to the lower value at 17 mA. Equation (1.7) defines
a value that is considered the average of the ac values from 2 mA to 17 mA. The fact that
one resistance level can be used for such a wide range of the characteristics will prove quite
useful in the definition of equivalent circuits for a diode in a later section.
As with the dc and ac resistance levels, the lower the level of currents used to determine
the average resistance, the higher is the resistance level.

Summary Table
Table 1.6 was developed to reinforce the important conclusions of the last few pages and
to emphasize the differences among the various resistance levels. As indicated earlier, the
content of this section is the foundation for a number of resistance calculations to be per-
formed in later sections and chapters.

TABLE 1.6
Resistance Levels

Special Graphical
Type Equation Characteristics Determination

VD
DC or static RD = Defined as a point on the ID
ID
characteristics Qpt.

VD

⌬Vd 26 mV
AC or dynamic rd = = Defined by a tangent line ID
⌬Id ID Qpt. ⌬Id
at the Q-point

⌬Vd

`
⌬Vd
Average ac rav = Defined by a straight ⌬Id
⌬Id pt. to pt.
line between limits of
operation

⌬Vd

1.9 DIODE EQUIVALENT CIRCUITS

●
An equivalent circuit is a combination of elements properly chosen to best represent the
actual terminal characteristics of a device or system in a particular operating region.
In other words, once the equivalent circuit is defined, the device symbol can be removed
from a schematic and the equivalent circuit inserted in its place without severely affecting
the actual behavior of the system. The result is often a network that can be solved using
traditional circuit analysis techniques.
28 SEMICONDUCTOR Piecewise-Linear Equivalent Circuit
DIODES
One technique for obtaining an equivalent circuit for a diode is to approximate the charac-
teristics of the device by straight-line segments, as shown in Fig. 1.29. The resulting equiv-
alent circuit is called a piecewise-linear equivalent circuit. It should be obvious from Fig.
1.29 that the straight-line segments do not result in an exact duplication of the actual char-
acteristics, especially in the knee region. However, the resulting segments are sufficiently
close to the actual curve to establish an equivalent circuit that will provide an excellent first
approximation to the actual behavior of the device. For the sloping section of the equiva-
lence the average ac resistance as introduced in Section 1.8 is the resistance level appearing
in the equivalent circuit of Fig. 1.28 next to the actual device. In essence, it defines the resis-
tance level of the device when it is in the “on” state. The ideal diode is included to establish
that there is only one direction of conduction through the device, and a reverse-bias condi-
tion will result in the open-circuit state for the device. Since a silicon semiconductor diode
does not reach the conduction state until VD reaches 0.7 V with a forward bias (as shown in
Fig. 1.29), a battery VK opposing the conduction direction must appear in the equivalent
circuit as shown in Fig. 1.30. The battery simply specifies that the voltage across the device
must be greater than the threshold battery voltage before conduction through the device in
the direction dictated by the ideal diode can be established. When conduction is established
the resistance of the diode will be the specified value of rav.

I D (mA)

10

rav

0 0.7 V 0.8 V VD (V)

(VK)

FIG. 1.29
Defining the piecewise-linear equivalent
circuit using straight-line segments to approximate
the characteristic curve.

+ VD –
VD VK Ideal diode
+ – r av
+ – 10 Ω
0.7 V
ID ID

FIG. 1.30
Components of the piecewise-linear equivalent circuit.

Keep in mind, however, that VK in the equivalent circuit is not an independent voltage
source. If a voltmeter is placed across an isolated diode on the top of a laboratory bench, a
reading of 0.7 V will not be obtained. The battery simply represents the horizontal offset of
the characteristics that must be exceeded to establish conduction.
The approximate level of rav can usually be determined from a specified operating
point on the specification sheet (to be discussed in Section 1.10). For instance, for a sili-
con semiconductor diode, if IF ⫽ 10 mA (a forward conduction current for the diode) at
VD ⫽ 0.8 V, we know that for silicon a shift of 0.7 V is required before the characteristics DIODE EQUIVALENT 29
rise, and we obtain CIRCUITS

`
⌬Vd 0.8 V - 0.7 V 0.1 V
rav = = = = 10 ⍀
⌬Id pt. to pt. 10 mA - 0 mA 10 mA
as obtained for Fig. 1.29.
If the characteristics or specification sheet for a diode is not available the resistance rav
can be approximated by the ac resistance rd.

Simplified Equivalent Circuit

For most applications, the resistance rav is sufficiently small to be ignored in comparison to
the other elements of the network. Removing rav from the equivalent circuit is the same as
implying that the characteristics of the diode appear as shown in Fig. 1.31. Indeed, this
approximation is frequently employed in semiconductor circuit analysis as demonstrated in
Chapter 2. The reduced equivalent circuit appears in the same figure. It states that a forward-
biased silicon diode in an electronic system under dc conditions has a drop of 0.7 V across
it in the conduction state at any level of diode current (within rated values, of course).
ID
+ VD –

r av = 0 Ω VK = 0.7 V

+ –
ID Ideal diode

0 V K = 0.7 V V D

FIG. 1.31
Simplified equivalent circuit for the silicon semiconductor diode.

Ideal Equivalent Circuit

Now that rav has been removed from the equivalent circuit, let us take the analysis a step
further and establish that a 0.7-V level can often be ignored in comparison to the applied
voltage level. In this case the equivalent circuit will be reduced to that of an ideal diode as
shown in Fig. 1.32 with its characteristics. In Chapter 2 we will see that this approximation
is often made without a serious loss in accuracy.

FIG. 1.32
Ideal diode and its characteristics.

In industry a popular substitution for the phrase “diode equivalent circuit” is diode model—
a model by definition being a representation of an existing device, object, system, and so on.
In fact, this substitute terminology will be used almost exclusively in the chapters to follow.

Summary Table
For clarity, the diode models employed for the range of circuit parameters and applications
are provided in Table 1.7 with their piecewise-linear characteristics. Each will be investi-
gated in greater detail in Chapter 2. There are always exceptions to the general rule, but it
30 SEMICONDUCTOR TABLE 1.7
DIODES Diode Equivalent Circuits (Models)

Type Conditions Model Characteristics

Piecewise-linear model + –
VK
VK

Simplified model Rnetwork W rav + – VK

VK

Ideal device Rnetwork W rav

Enetwork W VK

is fairly safe to say that the simplified equivalent model will be employed most frequently
in the analysis of electronic systems, whereas the ideal diode is frequently applied in the
analysis of power supply systems where larger voltages are encountered.

1.10 TRANSITION AND DIFFUSION CAPACITANCE

●
It is important to realize that:
Every electronic or electrical device is frequency sensitive.
That is, the terminal characteristics of any device will change with frequency. Even the
resistance of a basic resistor, as of any construction, will be sensitive to the applied fre-
quency. At low to mid-frequencies most resistors can be considered fixed in value. How-
ever, as we approach high frequencies, stray capacitive and inductive effects start to play a
role and will affect the total impedance level of the element.
For the diode it is the stray capacitance levels that have the greatest effect. At low frequen-
cies and relatively small levels of capacitance the reactance of a capacitor, determined by
XC = 1>2pfC, is usually so high it can be considered infinite in magnitude, represented by
an open circuit, and ignored. At high frequencies, however, the level of XC can drop to the
point where it will introduce a low-reactance “shorting” path. If this shorting path is across
the diode, it can essentially keep the diode from affecting the response of the network.
In the p–n semiconductor diode, there are two capacitive effects to be considered. Both
types of capacitance are present in the forward- and reverse-bias regions, but one so out-
weighs the other in each region that we consider the effects of only one in each region.
Recall that the basic equation for the capacitance of a parallel-plate capacitor is defined by
C = PA>d, where P is the permittivity of the dielectric (insulator) between the plates of area A
separated by a distance d. In a diode the depletion region (free of carriers) behaves essentially
like an insulator between the layers of opposite charge. Since the depletion width (d) will in-
crease with increased reverse-bias potential, the resulting transition capacitance will decrease,
as shown in Fig. 1.33. The fact that the capacitance is dependent on the applied reverse-bias
potential has application in a number of electronic systems. In fact, in Chapter 16 the varactor
diode will be introduced whose operation is wholly dependent on this phenomenon.
This capacitance, called the transition (CT), barriers, or depletion region capacitance, is
determined by

(1 + 兩VR >VK 兩 )n
C(0)
CT = (1.9)
 C (pF) REVERSE RECOVERY 31
15 TIME

10
CT

5 CT + CD ⬵ CD

–25 –20 –15 –10 –5 0 +0.25 +0.5 )V(

FIG. 1.33
Transition and diffusion capacitance versus applied bias for a silicon diode.

where C(0) is the capacitance under no-bias conditions and VR is the applied reverse bias
potential. The power n is 1⁄2 or 1⁄3 depending on the manufacturing process for the diode.
Although the effect described above will also be present in the forward-bias region, it
is overshadowed by a capacitance effect directly dependent on the rate at which charge is
injected into the regions just outside the depletion region. The result is that increased levels
of current will result in increased levels of diffusion capacitance (CD) as demonstrated by
the following equation:

CD = a bI
t
T
(1.10)
VK D

where t is the minority carrier lifetime—the time is world take for a minority carrier such
T

as a hole to recombine with an electron in the n-type material. However, increased levels
of current result in a reduced level of associated resistance (to be demonstrated shortly),
and the resulting time constant (t ⫽ RC), which is very important in high-speed applica-
tions, does not become excessive.
In general, therefore,
the transition capacitance is the predominant capacitive effect in the reverse-bias
region whereas the diffusion capacitance is the predominant capacitive effect in the
forward-bias region.
FIG. 1.34
The capacitive effects described above are represented by capacitors in parallel with the Including the effect of the transition
ideal diode, as shown in Fig. 1.34. For low- or mid-frequency applications (except in the or diffusion capacitance on the
power area), however, the capacitor is normally not included in the diode symbol. semiconductor diode.

1.11 REVERSE RECOVERY TIME

●
There are certain pieces of data that are normally provided on diode specification sheets
provided by manufacturers. One such quantity that has not been considered yet is the
reverse recovery time, denoted by trr. In the forward-bias state it was shown earlier that
there are a large number of electrons from the n-type material progressing through the
p-type material and a large number of holes in the n-type material—a requirement for con-
duction. The electrons in the p-type material and holes progressing through the n-type
material establish a large number of minority carriers in each material. If the applied volt-
age should be reversed to establish a reverse-bias situation, we would ideally like to see the
diode change instantaneously from the conduction state to the nonconduction state. How-
ever, because of the large number of minority carriers in each material, the diode current
will simply reverse as shown in Fig. 1.35 and stay at this measurable level for the period of
time ts (storage time) required for the minority carriers to return to their majority-carrier
state in the opposite material. In essence, the diode will remain in the short-circuit state
with a current Ireverse determined by the network parameters. Eventually, when this storage
phase has passed, the current will be reduced in level to that associated with the nonconduc-
tion state. This second period of time is denoted by tt (transition interval). The reverse recov-
ery time is the sum of these two intervals: trr ⫽ ts ⫹ tt. This is an important consideration in
32 SEMICONDUCTOR ID
DIODES
Change of state (on off)
I forward applied at t = t1

Desired response
t1 t

I reverse
ts tt

t rr

FIG. 1.35
Defining the reverse recovery time.

high-speed switching applications. Most commercially available switching diodes have a

trr in the range of a few nanoseconds to 1 ms. Units are available, however, with a trr of
only a few hundred picoseconds (10⫺12 s).

1.12 DIODE SPECIFICATION SHEETS

●
Data on specific semiconductor devices are normally provided by the manufacturer in one
of two forms. Most frequently, they give a very brief description limited to perhaps one
page. At other times, they give a thorough examination of the characteristics using graphs,
artwork, tables, and so on. In either case, there are specific pieces of data that must be
included for proper use of the device. They include:
1. The forward voltage VF (at a specified current and temperature)
2. The maximum forward current IF (at a specified temperature)
3. The reverse saturation current IR (at a specified voltage and temperature)
4. The reverse-voltage rating [PIV or PRV or V(BR), where BR comes from the term
“breakdown” (at a specified temperature)]
5. The maximum power dissipation level at a particular temperature
6. Capacitance levels
7. Reverse recovery time trr
8. Operating temperature range
Depending on the type of diode being considered, additional data may also be provided,
such as frequency range, noise level, switching time, thermal resistance levels, and peak
repetitive values. For the application in mind, the significance of the data will usually be
self-apparent. If the maximum power or dissipation rating is also provided, it is understood
to be equal to the following product:

PDmax = VD ID (1.11)

where ID and VD are the diode current and voltage, respectively, at a particular point of
operation.
If we apply the simplified model for a particular application (a common occurrence), we
can substitute VD ⫽ VT ⫽ 0.7 V for a silicon diode in Eq. (1.11) and determine the resulting
power dissipation for comparison against the maximum power rating. That is,

Pdissipated ⬵ (0.7 V)ID (1.12)

The data provided for a high-voltage/low-leakage diode appear in Figs. 1.36 and 1.37. This
example would represent the expanded list of data and characteristics. The term rectifier is
applied to a diode when it is frequently used in a rectification process, described in Chapter 2.
Specific areas of the specification sheet are highlighted in blue, with letters correspond-
ing to the following description:
A The data sheet highlights the fact that the silicon high-voltage diode has a minimum
reverse-bias voltage of 125 V at a specified reverse-bias current.
B Note the wide range of temperature operation. Always be aware that data sheets typi- DIODE SPECIFICATION 33
cally use the centigrade scale, with 200°C ⫽ 392°F and ⫺65°C ⫽ ⫺85°F. SHEETS
C The maximum power dissipation level is given by PD = VD ID = 500 mW = 0.5 W.
The effect of the linear derating factor of 3.33 mW/°C is demonstrated in Fig. 1.37a.
Once the temperature exceeds 25°C the maximum power rating will drop by 3.33 mW
for each 1°C increase in temperature. At a temperature of 100°C, which is the boiling
point of water, the maximum power rating has dropped to one half of its original value.
An initial temperature of 25°C is typical inside a cabinet containing operating elec-
tronic equipment in a low-power situation.
D The maximum sustainable current is 500 mA. The plot of Fig. 1.37b reveals that the
forward current at 0.5 V is about 0.01 mA, but jumps to 1 mA (100 times greater) at
about 0.65 V. At 0.8 V the current is more than 10 mA, and just above 0.9 V it is close

DIFFUSED SILICON PLANAR

5.0

FIG. 1.36
Electrical characteristics of a high-voltage, low-leakage diode.
 Room temperature
At VF = 1 V,
IF ≅ 250 mA

Boiling water At VR increases,

At VF = 0.7 V,
IF ≅ 30 mA IR increases

(a) (b) (c)

REVERSE CURRENT VERSUS

TEMPERATURE

30 mA, RD ≅ 2 ⍀

C ⫺ Capacitance-pf
CT
1 mA, RD ≅ 50 ⍀
As TA , IR

(d) (e) (f)

FIG. 1.37
Terminal characteristics of a high-voltage diode.

to 100 mA. The curve of Fig. 1.37b certainly looks nothing like the characteristic
curves appearing in the last few sections. This is a result of using a log scale for the
current and a linear scale for the voltage.
Log scales are often used to provide a broader range of values for a variable in a
limited amount of space.
If a linear scale was used for the current, it would be impossible to show a range
of values from 0.01 mA to 1000 mA. If the vertical divisions were in 0.01-mA incre-
ments, it would take 100,000 equal intervals on the vertical axis to reach 1000 mA. For
the moment recognize that the voltage level at given levels of current can be found by
using the intersection with the curve. For vertical values above a level such as 1.0 mA,
the next level is 2 mA, followed by 3 mA, 4 mA, and 5 mA. The levels of 6 mA to 10 mA
can be determined by simply dividing the distance into equal intervals (not the true
distribution, but close enough for the provided graphs). For the next level it would be
10 mA, 20 mA, 30 mA, and so on. The graph of Fig. 1.37b is called a semi-log plot to
reflect the fact that only one axis uses a log scale. A great deal more will be said about
log scales in Chapter 9.
E The data provide a range of VF (forward-bias voltages) for each current level. The
higher the forward current, the higher is the applied forward bias. At 1 mA we find VF
can range from 0.6 V to 0.68 V, but at 200 mA it can be as high as 0.85 V to 1.00 V.
For the full range of current levels with 0.6 V at 1 mA and 0.85 V at 200 mA it is cer-
tainly a reasonable approximation to use 0.7 V as the average value.
F The data provided clearly reveal how the reverse saturation current increases with
applied reverse bias at a fixed temperature. At 25°C the maximum reverse-bias cur-
rent increases from 0.2 nA to 0.5 nA due to an increase in reverse-bias voltage by the
same factor of 5. At 125°C it jumps by a factor of 2 to the high level of 1 mA. Note the
34
 extreme change in reverse saturation current with temperature as the maximum cur- SEMICONDUCTOR DIODE 35
rent rating jumps from 0.2 nA at 25°C to 500 nA at 125°C (at a fixed reverse-bias NOTATION
voltage of 20 V). A similar increase occurs at a reverse-bias potential of 100 V. The
semi-log plots of Figs. 1.37c and 1.37d provide an indication of how the reverse satu-
ration current changes with changes in reverse voltage and temperature. At first
glance Fig. 1.37c might suggest that the reverse saturation current is fairly steady for
changes in reverse voltage. However, this can sometimes be the effect of using a log
scale for the vertical axis. The current has actually changed from a level of 0.2 nA to
a level of 0.7 nA for the range of voltages representing a change of almost 6 to 1. The
dramatic effect of temperature on the reverse saturation current is clearly displayed in
Fig. 1.37d. At a reverse-bias voltage of 125 V the reverse-bias current increases from
a level of about 1 nA at 25°C to about 1 mA at 150°C, an increase of a factor of 1000
over the initial value.
Temperature and applied reverse bias are very important factors in designs sensitive
to the reverse saturation current.
G As shown in the data listing and on Fig. 1.37e, the transition capacitance at a reverse-
bias voltage of 0 V is 5 pF at a test frequency of 1 MHz. Note the severe change in
capacitance level as the reverse-bias voltage is increased. As mentioned earlier, this
sensitive region can be put to good use in the design of a device (Varactor; Chapter 16)
whose terminal capacitance is sensitive to the applied voltage.
H The reverse recovery time is 3 ms for the test conditions shown. This is not a fast time
for some of the current high-performance systems in use today. However, for a variety
of low- and mid-frequency applications it is acceptable.
The curves of Fig. 1.37f provide an indication of the magnitude of the ac resistance of the
diode versus forward current. Section 1.8 clearly demonstrated that the dynamic resistance
of a diode decreases with increase in current. As we go up the current axis of Fig. 1.37f it
is clear that if we follow the curve, the dynamic resistance will decrease. At 0.1 mA it is
close to 1 k⍀; at 10 mA, 10 ⍀; and at 100 mA, only 1 ⍀; this clearly supports the earlier
discussion. Unless one has had experience reading log scales, the curve is challenging to
read for levels between those indicated because it is a log–log plot. Both the vertical axis
and the horizontal axis employ a log scale.
The more one is exposed to specification sheets, the “friendlier” they will become, es-
pecially when the impact of each parameter is clearly understood for the application under
investigation.

1.13 SEMICONDUCTOR DIODE NOTATION

●
The notation most frequently used for semiconductor diodes is provided in Fig. 1.38. For
most diodes any marking such as a dot or band, as shown in Fig. 1.38, appears at the cath-
ode end. The terminology anode and cathode is a carryover from vacuum-tube notation.
The anode refers to the higher or positive potential, and the cathode refers to the lower or
negative terminal. This combination of bias levels will result in a forward-bias or “on”
condition for the diode. A number of commercially available semiconductor diodes appear
in Fig. 1.39.

Anode
p
n or •, K, etc.
Cathode

FIG. 1.38
Semiconductor diode notation.
 General purpose diode Surface mount high-power PIN diode Power (stud) diode Power (planar) diode

Beam lead pin diode Flat chip surface mount diode Power diode Power (disc, puck) diode

FIG. 1.39
Various types of junction diodes.

1.14 DIODE TESTING

●
The condition of a semiconductor diode can be determined quickly using (1) a digital dis-
play meter (DDM) with a diode checking function, (2) the ohmmeter section of a multime-
ter, or (3) a curve tracer.

Diode Checking Function

A digital display meter with a diode checking capability appears in Fig. 1.40. Note the
small diode symbol at the top right of the rotating dial. When set in this position and
hooked up as shown in Fig. 1.41a, the diode should be in the “on” state and the display will
provide an indication of the forward-bias voltage such as 0.67 V (for Si). The meter has an
internal constant-current source (about 2 mA) that will define the voltage level as indicated
in Fig. 1.41b. An OL indication with the hookup of Fig. 1.41a reveals an open (defective)
diode. If the leads are reversed, an OL indication should result due to the expected open-
circuit equivalence for the diode. In general, therefore, an OL indication in both directions
is an indication of an open or defective diode.

FIG. 1.40 Red lead Black lead

Digital display meter. (Courtesy of (VΩ) (COM)
B&K Precision Corporation.)

(a) (b)

FIG. 1.41
Checking a diode in the forward-bias state.

Ohmmeter Testing
In Section 1.8 we found that the forward-bias resistance of a semiconductor diode is quite
low compared to the reverse-bias level. Therefore, if we measure the resistance of a diode
36
using the connections indicated in Fig. 1.42, we can expect a relatively low level. The result- DIODE TESTING 37
ing ohmmeter indication will be a function of the current established through the diode by the
internal battery (often 1.5 V) of the ohmmeter circuit. The higher the current, the lower is the
resistance level. For the reverse-bias situation the reading should be quite high, requiring a (Ohmmeter)
high resistance scale on the meter, as indicated in Fig. 1.42b. A high resistance reading in Red lead
Relatively low R
Black lead
both directions indicates an open (defective-device) condition, whereas a very low resis- (V⍀) (COM)
tance reading in both directions will probably indicate a shorted device. + –
(a)
Curve Tracer
Relatively high R
The curve tracer of Fig. 1.43 can display the characteristics of a host of devices, including Black lead Red lead
the semiconductor diode. By properly connecting the diode to the test panel at the bottom (COM) (V⍀)
center of the unit and adjusting the controls, one can obtain the display of Fig. 1.44. Note – +
that the vertical scaling is 1 mA/div, resulting in the levels indicated. For the horizontal axis
the scaling is 100 mV/div, resulting in the voltage levels indicated. For a 2-mA level as (b)
defined for a DDM, the resulting voltage would be about 625 mV ⫽ 0.625 V. Although the FIG. 1.42
instrument initially appears quite complex, the instruction manual and a few moments of Checking a diode with an
exposure will reveal that the desired results can usually be obtained without an excessive ohmmeter.
amount of effort and time. The display of the instrument will appear on more than one occa-
sion in the chapters to follow as we investigate the characteristics of the variety of devices.

FIG. 1.43
Curve tracer. (© Agilent Technologies, Inc. Reproduced with
Permission, Courtesy of Agilent Technologies, Inc.)

FIG. 1.44
Curve tracer response to IN4007 silicon diode.
38 SEMICONDUCTOR 1.15 ZENER DIODES
DIODES ●
The Zener region of Fig. 1.45 was discussed in some detail in Section 1.6. The characteristic
drops in an almost vertical manner at a reverse-bias potential denoted VZ. The fact that
the curve drops down and away from the horizontal axis rather than up and away for the
positive-VD region reveals that the current in the Zener region has a direction opposite to
that of a forward-biased diode. The slight slope to the curve in the Zener region reveals that
there is a level of resistance to be associated with the Zener diode in the conduction mode.
This region of unique characteristics is employed in the design of Zener diodes, which
have the graphic symbol appearing in Fig. 1.46a. The semiconductor diode and the Zener
diode are presented side by side in Fig. 1.46 to ensure that the direction of conduction of
each is clearly understood together with the required polarity of the applied voltage. For
the semiconductor diode the “on” state will support a current in the direction of the arrow
in the symbol. For the Zener diode the direction of conduction is opposite to that of the
arrow in the symbol, as pointed out in the introduction to this section. Note also that the
polarity of VD and VZ are the same as would be obtained if each were a resistive element
as shown in Fig. 1.46c.

ID

VZ

0 VD IZ ID IR
VZ VD
VR
R

(a) (b) (c)

FIG. 1.45 FIG. 1.46

Reviewing the Zener region. Conduction direction: (a) Zener diode;
(b) semiconductor diode;
(c) resistive element.

The location of the Zener region can be controlled by varying the doping levels. An in-
crease in doping that produces an increase in the number of added impurities, will decrease
the Zener potential. Zener diodes are available having Zener potentials of 1.8 V to 200 V
with power ratings from 1⁄4 W to 50 W. Because of its excellent temperature and current
capabilities, silicon is the preferred material in the manufacture of Zener diodes.
It would be nice to assume the Zener diode is ideal with a straight vertical line at the
Zener potential. However, there is a slight slope to the characteristics requiring the piece-
wise equivalent model appearing in Fig. 1.47 for that region. For most of the applications
appearing in this text the series resistive element can be ignored and the reduced equivalent
model of just a dc battery of VZ volts employed. Since some applications of Zener diodes
swing between the Zener region and the forward-bias region, it is important to understand
the operation of the Zener diode in all regions. As shown in Fig. 1.47, the equivalent model
for a Zener diode in the reverse-bias region below VZ is a very large resistor (as for the
standard diode). For most applications this resistance is so large it can be ignored and the
open-circuit equivalent employed. For the forward-bias region the piecewise equivalent is
the same as described in earlier sections.
The specification sheet for a 10-V, 500-mW, 20% Zener diode is provided as Table 1.8,
and a plot of the important parameters is given in Fig. 1.48. The term nominal used in the
specification of the Zener voltage simply indicates that it is a typical average value. Since this
is a 20% diode, the Zener potential of the unit one picks out of a lot (a term used to describe a
package of diodes) can be expected to vary as 10 V ⫹ 20%, or from 8 V to 12 V. Both 10%
and 50% diodes are also readily available. The test current IZT is the current defined by the
 IZ

rZ ⬵ +
+
– 0.7 V +
– ⬵ 0.7 V
rZ –
–

VZ VR 0.7 V

ⵑ10 μA = IR VZ
ⵑ
0.25 mA = IZK

+
+ ⵑ
ⵑ
VZ +
– ⬵ VZ
rZ – IZT = 12.5 mA
rZ = 8.5 ⍀ = ZZT
– ⵑ
ⵑ

⌬ VZ
rZ =
⌬ IZ
IZM = 32 mA

FIG. 1.47
Zener diode characteristics with the equivalent model for each region.

TABLE 1.8
Electrical Characteristics (25°C Ambient Temperature)

Zener Maximum Maximum Maximum Maximum

Voltage Test Dynamic Knee Reverse Test Regulator Typical
Nominal Current Impedance Impedance Current Voltage Current Temperature
VZ IZT ZZT at IZT ZZK at IZK IR at VR VR IZM Coefficient
(V) (mA) (⍀) (⍀) (mA) (MA) (V) (mA) (%/°C)

10 12.5 8.5 700 0.25 10 7.2 32 ⴙ0.072

Temperature coefficient (TC) Dynamic impedance (rZ)

versus Zener current versus Zener current
+0.12 1 kΩ
Temperature coefficient – TC (%/˚C)

500
Dynamic impedance, ZZ – ( Ω )

+0.08 24 V
10 V 200
+0.04 100
50
0
20 3.6 V
– 0.04 10
3.6 V 5 10 V
– 0.08
2
24 V
– 0.12 1
0.01 0.05 0.1 0.5 1 5 10 50 100 0.1 0.2 0.5 1 2 5 10 20 50 100
Zener current IZ – (mA) Zener current IZ – (mA)

(a) (b)

FIG. 1.48
Electrical characteristics for a 10-V, 500-mW Zener diode.
39
40 SEMICONDUCTOR 1
⁄4-power level. It is the current that will define the dynamic resistance ZZT and appears in
DIODES the general equation for the power rating of the device. That is,

PZmax = 4IZTVZ (1.13)

Substituting IZT into the equation with the nominal Zener voltage results in
PZmax = 4IZTVZ = 4(12.5 mA)(10 V) = 500 mW
which matches the 500-mW label appearing above. For this device the dynamic resistance
is 8.5 ⍀, which is usually small enough to be ignored in most applications. The maximum
knee impedance is defined at the center of the knee at a current of IZK ⫽ 0.25 mA. Note
that in all the above the letter T is used in subscripts to indicate test values and the letter K
to indicate knee values. For any level of current below 0.25 mA the resistance will only get
larger in the reverse-bias region. The knee value therefore reveals when the diode will start
to show very high series resistance elements that one may not be able to ignore in an appli-
cation. Certainly 500 ⍀ ⫽ 0.5 k⍀ may be a level that can come into play. At a reverse-bias
voltage the application of a test voltage of 7.2 V results in a reverse saturation current of
10 mA, a level that could be of some concern in some applications. The maximum regulator
current is the maximum continuous current one would want to support in the use of the
Zener diode in a regulator configuration. Finally, we have the temperature coefficient
(TC) in percent per degree centigrade.
The Zener potential of a Zener diode is very sensitive to the temperature of operation.
The temperature coefficient can be used to find the change in Zener potential due to a
change in temperature using the following equation:

⌬VZ >VZ
TC = * 100%/⬚C (%/⬚C) (1.14)
T1 - T0

where T1 is the new temperature level

T0 is room temperature in an enclosed cabinet (25°C)
TC is the temperature coefficient
and VZ is the nominal Zener potential at 25°C.
To demonstrate the effect of the temperature coefficient on the Zener potential, consider
the following example.

EXAMPLE 1.5 Analyze the 10-V Zener diode described by Table 1.7 if the temperature is
increased to 100°C (the boiling point of water).
Solution: Substituting into Eq. (1.14), we obtain
TCVZ
⌬VZ = (T - T0)
100% 1
(0.072%/⬚C)(10 V)
= (100⬚C - 25⬚C)
100%
and ⌬VZ = 0.54 V
The resulting Zener potential is now
VZ ⬘ = VZ + 0.54 V = 10.54 V
which is not an insignificant change.

It is important to realize that in this case the temperature coefficient was positive. For Zener
diodes with Zener potentials less than 5 V it is very common to see negative temperature
coefficients, where the Zener voltage drops with an increase in temperature. Figure 1.48a
provides a plot of T versus Zener current for three different levels of diodes. Note that the
3.6-V diode has a negative temperature coefficient, whereas the others have positive values.
The change in dynamic resistance with current for the Zener diode in its avalanche re-
gion is provided in Fig. 1.48b. Again, we have a log–log plot, which has to be carefully read.
Initially it would appear that there is an inverse linear relationship between the dynamic LIGHT-EMITTING DIODES 41
resistance because of the straight line. That would imply that if one doubles the current, one
cuts the resistance in half. However, it is only the log–log plot that gives this impression,
because if we plot the dynamic resistance for the 24-V Zener diode versus current using
linear scales we obtain the plot of Fig. 1.49, which is almost exponential in appearance.
Note on both plots that the dynamic resistance at very low currents that enter the knee of
the curve is fairly high at about 200 ⍀. However, at higher Zener currents, away from the
knee, at, say 10 mA, the dynamic resistance drops to about 5 ⍀.

FIG. 1.49
Zener terminal identification and symbols.

The terminal identification and the casing for a variety of Zener diodes appear in Fig.
1.49. Their appearance is similar in many ways to that of the standard diode. Some areas of
application for the Zener diode will be examined in Chapter 2.

1.16 LIGHT-EMITTING DIODES

●
The increasing use of digital displays in calculators, watches, and all forms of instrumenta-
tion has contributed to an extensive interest in structures that emit light when properly
biased. The two types in common use to perform this function are the light-emitting diode
(LED) and the liquid-crystal display (LCD). Since the LED falls within the family of p–n
junction devices and will appear in some of the networks of the next few chapters, it will
be introduced in this chapter. The LCD display is described in Chapter 16.
As the name implies, the light-emitting diode is a diode that gives off visible or invis-
ible (infrared) light when energized. In any forward-biased p–n junction there is, within the
structure and primarily close to the junction, a recombination of holes and electrons. This
recombination requires that the energy possessed by the unbound free electrons be trans-
ferred to another state. In all semiconductor p–n junctions some of this energy is given off
in the form of heat and some in the form of photons.
In Si and Ge diodes the greater percentage of the energy converted during recombina-
tion at the junction is dissipated in the form of heat within the structure, and the emitted
light is insignificant.
For this reason, silicon and germanium are not used in the construction of LED devices.
On the other hand:
Diodes constructed of GaAs emit light in the infrared (invisible) zone during the
recombination process at the p–n junction.
Even though the light is not visible, infrared LEDs have numerous applications where
visible light is not a desirable effect. These include security systems, industrial processing,
optical coupling, safety controls such as on garage door openers, and in home entertainment
centers, where the infrared light of the remote control is the controlling element.
Through other combinations of elements a coherent visible light can be generated. Table 1.9
provides a list of common compound semiconductors and the light they generate. In addi-
tion, the typical range of forward bias potentials for each is listed.
The basic construction of an LED appears in Fig. 1.50 with the standard symbol used
for the device. The external metallic conducting surface connected to the p-type material is
smaller to permit the emergence of the maximum number of photons of light energy when
the device is forward-biased. Note in the figure that the recombination of the injected carri-
ers due to the forward-biased junction results in emitted light at the site of the recombination.
42 SEMICONDUCTOR TABLE 1.9
DIODES Light-Emitting Diodes

Typical Forward
Color Construction Voltage (V)

Amber AlInGaP 2.1

Blue GaN 5.0
Green GaP 2.2
Orange GaAsP 2.0
Red GaAsP 1.8
White GaN 4.1
Yellow AlInGaP 2.1

There will, of course, be some absorption of the packages of photon energy in the structure
itself, but a very large percentage can leave, as shown in the figure.

(b)

(a)

FIG. 1.50
(a) Process of electroluminescence in the LED; (b) graphic symbol.

Just as different sounds have different frequency spectra (high-pitched sounds generally
have high-frequency components, and low sounds have a variety of low-frequency compo-
nents), the same is true for different light emissions.
The frequency spectrum for infrared light extends from about 100 THz (T ⴝ tera ⴝ
1012) to 400 THz, with the visible light spectrum extending from about 400 to 750 THz.
It is interesting to note that invisible light has a lower frequency spectrum than visible
light.
In general, when one talks about the response of electroluminescent devices, one refer-
ences their wavelength rather than their frequency.
The two quantities are related by the following equation:

c
l = (m) (1.15)
f

where c ⫽ 3 ⫻ 108 m/s (the speed of light in a vacuum)

f ⫽ frequency in Hertz
l ⫽ wavelength in meters.
 LIGHT-EMITTING DIODES 43
EXAMPLE 1.6 Using Eq. (1.15), find the range of wavelength for the frequency range of
visible light (400 THz–750 THz).
Solution:

c = 3 * 108 c d = 3 * 1017 nm/s

m 109 nm
s m
c 3 * 1017 nm/s 3 * 1017 nm/s
l = = = = 750 nm
f 400 THz 400 * 1012 Hz
c 3 * 1017 nm/s 3 * 1017 nm/s
l = = = = 400 nm
f 750 THz 750 * 1012 Hz
400 nm to 750 nm

Note in the above example the resulting inversion from higher frequency to smaller wave-
length. That is, the higher frequency results in the smaller wavelength. Also, most charts
use either nanometers (nm) or angstrom (Å) units. One angstrom unit is equal to 10⫺10 m.
The response of the average human eye as provided in Fig. 1.51 extends from about
350 nm to 800 nm with a peak near 550 nm.
It is interesting to note that the peak response of the eye is to the color green, with red and
blue at the lower ends of the bell curve. The curve reveals that a red or a blue LED must
have a much stronger efficiency than a green one to be visible at the same intensity. In other
words, the eye is more sensitive to the color green than to other colors. Keep in mind that
the wavelengths shown are for the peak response of each color. All the colors indicated on
the plot will have a bell-shaped curve response, so green, for example, is still visible at 600
nm, but at a lower intensity level.

Luminosity (Lm/w)

700
Green

600

500

400
ULTRAVIOLET INFRARED

300 Yellow
Amber

200

Orange
Blue
100

Red
0
0 100 400 500 600 700 800 900 ␭ (nm)

FIG. 1.51
Standard response curve of the human eye, showing the eye’s response to light energy
peaks at green and falls off for blue and red.

In Section 1.4 it was mentioned briefly that GaAs with its higher energy gap of 1.43 eV
made it suitable for electromagnetic radiation of visible light, whereas Si at 1.1 eV resulted pri-
marily in heat dissipation on recombination. The effect of this difference in energy gaps can be
44 SEMICONDUCTOR explained to some degree by realizing that to move an electron from one discrete energy level
DIODES to another requires a specific amount of energy. The amount of energy involved is given by

hc
Eg = (1.16)
l

with Eg ⫽ joules (J) [1 eV ⫽ 1.6 ⫻ 10⫺19 J]

h ⫽ Planck’s constant ⫽ 6.626 ⫻ 10⫺34 J # s.
c ⫽ 3 ⫻ 108 m/s
l ⫽ wavelength in meters
If we substitute the energy gap level of 1.43 eV for GaAs into the equation, we obtain the
following wavelength:

1.43 eV c d = 2.288 * 10-19 J

1.6 * 10-19 J
1 eV
hc (6.626 * 10-34 J # s)(3 * 108 m/s)
and l = =
Eg 2.288 * 10-19 J
= 869 nm
For silicon, with Eg = 1.1 eV
l = 1130 nm
which is well beyond the visible range of Fig. 1.51.
The wavelength of 869 nm places GaAs in the wavelength zone typically used in infrared
devices. For a compound material such as GaAsP with a band gap of 1.9 eV the resulting
wavelength is 654 nm, which is in the center of the red zone, making it an excellent com-
pound semiconductor for LED production. In general, therefore:
The wavelength and frequency of light of a specific color are directly related to the
energy band gap of the material.
A first step, therefore, in the production of a compound semiconductor that can be used
to generate light is to come up with a combination of elements that will generate the desired
energy band gap.
The appearance and characteristics of a subminiature high-efficiency red LED manufac-
tured by Hewlett-Packard are given in Fig. 1.52. Note in Fig. 1.52b that the peak forward
current is 60 mA, with 20 mA the typical average forward current. The text conditions
listed in Fig. 1.52c, however, are for a forward current of 10 mA. The level of VD under
forward-bias conditions is listed as VF and extends from 2.2 V to 3 V. In other words, one
can expect a typical operating current of about 10 mA at 2.3 V for good light emission, as
shown in Fig. 1.52e. In particular, note the typical diode characteristics for an LED, permit-
ting similar analysis techniques to be described in the next chapter.
Two quantities yet undefined appear under the heading Electrical/Optical Characteristics
at TA ⫽ 25°C. They are the axial luminous intensity (IV) and the luminous efficacy (hV). Light
intensity is measured in candelas. One candela (cd) corresponds to a light flux of 4p lumens
(lm) and is equivalent to an illumination of 1 footcandle on a 1-ft2 area 1 ft from the light
source. Even if this description may not provide a clear understanding of the candela as a unit of
measure, it should be enough to allow its level to be compared between similar devices. Figure
1.52f is a normalized plot of the relative luminous intensity versus forward current. The term
normalized is used frequently on graphs to give comparisons of response to a particular level.
A normalized plot is one where the variable of interest is plotted with a specific level
defined as the reference value with a magnitude of one.
In Fig. 1.52f the normalized level is taken at IF ⫽ 10 mA. Note that the relative lumi-
nous intensity is 1 at IF ⫽ 10 mA. The graph quickly reveals that the intensity of the light
is almost doubled at a current of 15 mA and is almost three times as much at a current of
20 mA. It is important to therefore note that:
The light intensity of an LED will increase with forward current until a point of
saturation arrives where any further increase in current will not effectively increase
the level of illumination.
 Absolute Maximum Ratings at TA ⴝ 25°C

High-Efficiency Red
Parameter 4160 Units
Power dissipation 120 mW
Average forward current 20[1] mA
Peak forward current 60 mA
Operating and storage temperature range ⫺55°C to 100°C
Lead soldering temperature 230°C for 3 s
[1.6 mm (0.063 in.) from body]
NOTE: 1. Derate from 50°C at 0.2 mV/°C.

(b)

(a)

Electrical/Optical Characteristics at TA ⴝ 25°C

High-Efficiency Red
4160
Symbol Description Min. Typ. Max. Units Test Conditions
IF = 10 mA
IV Axial luminous
intensity 1.0 3.0 mcd
2u1/2 Included angle
between half
luminous intensity
points 80 degree Note 1
lpeak Peak wavelength 635 nm Measurement
at peak
ld Dominant wavelength 628 nm Note 2
ts Speed of response 90 ns
C Capacitance 11 pF VF ⫽ 0; f ⫽ 1 Mhz
uJC Thermal resistance 120 °C/W Junction to
cathode lead at
0.79 mm (0.031
in.) from body
VF Forward voltage 2.2 3.0 V IF = 10 mA
BVR Reverse breakdown
voltage 5.0 V IR = 100 mA
hv Luminous efficacy 147 lm/W Note 3
NOTES:
1. u1/2 is the off-axis angle at which the luminous intensity is half the axial luminous intensity.
2. The dominant wavelength, ld, is derived from the CIE chromaticity diagram and represents the single
wavelength that defines the color of the device.
3. Radiant intensity, Ie, in watts/steradian, may be found from the equation Ie ⫽ Iv/hv, where Iv is the
luminous intensity in candelas and hv is the luminous efficacy in lumens/watt.
(c)

FIG. 1.52
Hewlett-Packard subminiature high-efficiency red solid-state lamp: (a) appearance; (b) absolute maximum ratings; (c) electrical/optical
characteristics; (d) relative intensity versus wavelength; (e) forward current versus forward voltage; (f) relative luminous intensity
versus forward current; (g) relative efficiency versus peak current; (h) relative luminous intensity versus angular displacement.

For instance, note in Fig. 1.52g that the increase in relative efficiency starts to level off
as the current exceeds 50 mA.
The term efficacy is, by definition, a measure of the ability of a device to produce the
desired effect. For the LED this is the ratio of the number of lumens generated per applied
watt of electrical power.
The plot of Fig. 1.52d supports the information appearing on the eye-response curve of
Fig. 1.51. As indicated above, note the bell-shaped curve for the range of wavelengths that
will result in each color. The peak value of this device is near 630 nm, very close to the
peak value of the GaAsP red LED. The curves of green and yellow are only provided for
reference purposes.
45
 (d)

(e)

(f) (g) (h)

FIG. 1.52
Continued.

Figure 1.52h is a graph of light intensity versus angle measured from 0° (head on)
to 90° (side view). Note that at 40° the intensity has already dropped to 50% of the
head-on intensity.
One of the major concerns when using an LED is the reverse-bias breakdown voltage,
which is typically between 3 V and 5 V (an occasional device has a 10-V level).
This range of values is significantly less than that of a standard commercial diode,
where it can extend to thousands of volts. As a result one has to be acutely aware of this
severe limitation in the design process. In the next chapter one protective approach will be
introduced.
In the analysis and design of networks with LEDs it is helpful to have some idea of the
voltage and current levels to be expected.
For many years the only colors available were green, yellow, orange, and red, permitting
the use of the average values of VF ⴝ 2 V and IF ⴝ 20 mA for obtaining an approximate
operating level.
However, with the introduction of blue in the early 1990s and white in the late 1990s the
magnitude of these two parameters has changed. For blue the average forward bias voltage
can be as high as 5 V, and for white about 4.1 V, although both have a typical operating
current of 20 mA or more. In general, therefore:
Assume an average forward-bias voltage of 5 V for blue and 4 V for white LEDs at
currents of 20 mA to initiate an analysis of networks with these types of LEDs.
Every once in a while a device is introduced that seems to open the door to a slue of
possibilities. Such is the case with the introduction of white LEDs. The slow start for white
LEDs is primarily due to the fact that it is not a primary color like green, blue, and red.
Every other color that one requires, such as on a TV screen, can be generated from these
three colors (as in virtually all monitors available today). Yes, the right combination of
these three colors can give white—hard to believe, but it works. The best evidence is the
46
human eye, which only has cones sensitive to red, green, and blue. The brain is responsible LIGHT-EMITTING DIODES 47
for processing the input and perceiving the “white” light and color we see in our everyday
lives. The same reasoning was used to generate some of the first white LEDs, by combining
the right proportions of a red, a green, and a blue LED in a single package. Today, however,
most white LEDs are constructed of a blue gallium nitride LED below a film of yttrium-
aluminum garnet (YAG) phosphor. When the blue light hits the phosphor, a yellow light is
generated. The mix of this yellow emission with that of the central blue LED forms a white
light—incredible, but true.
Since most of the lighting for homes and offices is white light, we now have another
option to consider versus incandescent and fluorescent lighting. The rugged characteristics
of LED white light along with lifetimes that exceed 25,000 hours, clearly suggest that
this will be a true competitor in the near future. Various companies are now providing
replacement LED bulbs for almost every possible application. Some have efficacy ratings
as high as 135.7 lumens per watt, far exceeding the 25 lumens per watt of a few years
ago. It is forecast that 7 W of power will soon be able to generate 1,000 lm of light, which
exceeds the illumination of a 60 W bulb and can run off four D cell batteries. Imagine
the same lighting with less than 1>8 the power requirement. At the present time entire of-
fices, malls, street lighting, sporting facilities, and so on are being designed using solely
LED lighting. Recently, LEDs are the common choice for flashlights and many high-end
automobiles due to the sharp intensity at lower dc power requirements. The tube light of
Fig. 1.53a replaces the standard fluorescent bulb typically found in the ceiling fixtures of
both the home and industry. Not only do they draw 20% less energy while providing 25%
additional light but they also last twice as long as a standard fluorescent bulb. The flood
light of Fig. 1.53b draws 1.7 watts for each 140 lumens of light resulting in an enormous
90% savings in energy compared to the incandescent variety. The chandelier bulbs of Fig.
1.53c have a lifetime of 50,000 hours and only draw 3 watts of power while generating
200 lumens of light.

(a) (b) (c)

FIG. 1.53
LED residential and commercial lighting.

Before leaving the subject, let us look at a seven-segment digital display housed in a
typical dual in-line integrated circuit package as shown in Fig. 1.54. By energizing the
proper pins with a typical 5-V dc level, a number of the LEDs can be energized and the
desired numeral displayed. In Fig. 1.54a the pins are defined by looking at the face of
the display and counting counterclockwise from the top left pin. Most seven-segment
displays are either common-anode or common-cathode displays, with the term anode
referring to the defined positive side of each diode and the cathode referring to the nega-
tive side. For the common-cathode option the pins have the functions listed in Fig. 1.54b
and appear as in Fig. 1.54c. In the common-cathode configuration all the cathodes are
connected together to form a common point for the negative side of each LED. Any LED
with a positive 5 V applied to the anode or numerically numbered pin side will turn on
and produce light for that segment. In Fig. 1.54c, 5 V has been applied to the terminals
that generate the numeral 5. For this particular unit the average forward turn-on voltage
is 2.1 V at a current of 10 mA.
Various LED configurations are examined in the next chapter.
48 SEMICONDUCTOR COMMON CATHODE
DIODES PIN # FUNCTION
1. Anode f
2. ANODE g
3. NO PIN
1 14
a 4. COMMON CATHODE
5. NO PIN
f b 6. ANODE e
0.630" 7. ANODE d
e g c 1.0875" 8. ANODE c
9. ANODE d
d 10. NO PIN
7 8 11. NO PIN
12. COMMON CATHODE
13. ANODE b
0.803" 14. ANODE a

(a) (b)

Computer control
5V 5V5V 5V 5V

1 14
2 13
3 12
4 11
5 10
6 9
7 8

(c)

FIG. 1.54
Seven-segment display: (a) face with pin idenfication; (b) pin function; (c) displaying the numeral 5.

1.17 SUMMARY
●
Important Conclusions and Concepts
1. The characteristics of an ideal diode are a close match with those of a simple switch
except for the important fact that an ideal diode can conduct in only one direction.
2. The ideal diode is a short in the region of conduction and an open circuit in the
region of nonconduction.
3. A semiconductor is a material that has a conductivity level somewhere between that
of a good conductor and that of an insulator.
4. A bonding of atoms, strengthened by the sharing of electrons between neighboring
atoms, is called covalent bonding.
5. Increasing temperatures can cause a significant increase in the number of free elec-
trons in a semiconductor material.
6. Most semiconductor materials used in the electronics industry have negative tem-
perature coefficients; that is, the resistance drops with an increase in temperature.
7. Intrinsic materials are those semiconductors that have a very low level of impurities,
whereas extrinsic materials are semiconductors that have been exposed to a doping
process.
8. An n-type material is formed by adding donor atoms that have five valence electrons
to establish a high level of relatively free electrons. In an n-type material, the electron
is the majority carrier and the hole is the minority carrier.
9. A p-type material is formed by adding acceptor atoms with three valence electrons to
establish a high level of holes in the material. In a p-type material, the hole is the
majority carrier and the electron is the minority carrier.
10. The region near the junction of a diode that has very few carriers is called the deple-
tion region.
11. In the absence of any externally applied bias, the diode current is zero.
12. In the forward-bias region the diode current increases exponentially with increase in
voltage across the diode.
13. In the reverse-bias region the diode current is the very small reverse saturation cur- COMPUTER ANALYSIS 49
rent until Zener breakdown is reached and current will flow in the opposite direction
through the diode.
14. The reverse saturation current Is will just about double in magnitude for every 10-fold
increase in temperature.
15. The dc resistance of a diode is determined by the ratio of the diode voltage and cur-
rent at the point of interest and is not sensitive to the shape of the curve. The dc resis-
tance decreases with increase in diode current or voltage.
16. The ac resistance of a diode is sensitive to the shape of the curve in the region of inter-
est and decreases for higher levels of diode current or voltage.
17. The threshold voltage is about 0.7 V for silicon diodes and 0.3 V for germanium diodes.
18. The maximum power dissipation level of a diode is equal to the product of the diode
voltage and current.
19. The capacitance of a diode increases exponentially with increase in the forward-bias
voltage. Its lowest levels are in the reverse-bias region.
20. The direction of conduction for a Zener diode is opposite to that of the arrow in the
symbol, and the Zener voltage has a polarity opposite to that of a forward-biased diode.
21. Light emitting diodes (LEDs) emit light under forward-bias conditions but require 2
V to 4 V for good emission.

Equations
ID = Is(eVD>nVT - 1)
kT
VT = TK = TC + 273⬚ k = 1.38 * 10-23 J>K
q
VK ⬵ 0.7 V (Si)
VK ⬵ 1.2 V (GaAs)
VK ⬵ 0.3 V (Ge)
VD
RD =
ID
⌬Vd 26 mV
rd = =
⌬Id ID

`
⌬Vd
rav =
⌬Id pt. to pt.
PDmax = VD ID

1.18 COMPUTER ANALYSIS

●
Two software packages designed to analyze electronic circuits will be introduced and applied
throughout the text. They include Cadence OrCAD, version 16.3 (Fig. 1.55), and Multi-
sim, version 11.0.1 (Fig. 1.56). The content was written with sufficient detail to ensure that
the reader will not need to reference any other computer literature to apply both programs.

FIG. 1.55 FIG. 1.56

Cadence OrCAD Design package version 16.3. Multisim 11.0.1.
(Photo by Dan Trudden/Pearson.) (Photo by Dan Trudden/Pearson.)
50 SEMICONDUCTOR Those of you who have used either program in the past will find that the changes are minor
DIODES and appear primarily in the front end and in the generation of specific data and plots.
The reason for including two programs stems from the fact that both are used throughout
the educational community. You will find that the OrCAD software has a broader area of
investigation but the Multisim software generates displays that are a better match to the
actual laboratory experience.
The demo version of OrCAD is free from Cadence Design Systems, Inc., and can be
downloaded directly from the EMA Design Automation, Inc., web site, info@[Link].
Multisim must be purchased from the National Instruments Corporation using their web
site, [Link]/multisim.
In previous editions, the OrCAD package was referred to as a PSpice program primarily
because it is a subset of a more sophisticated version used extensively in industry called
SPICE. The result is the use of the term PSpice in the descriptions to follow when initiating
an analysis using the OrCAD software.
The downloading process for each software package will now be introduced along with
the general appearance of the resulting screen.

OrCAD
Installation:
Insert the OrCAD Release 16.3 DVD into the disk drive to open the Cadence OrCAD
16.3 software screen.
Select Demo Installation and the Preparing Setup dialog box will open, followed by
the message Welcome to the Installation Wizard for OrCAD 16.3 Demo. Select
Next, and the License Agreement dialog box opens. Choose I accept and select
Next, and the Choose Destination dialog box will open with Install OrCAD 16.3
Demo Accept C:\OrCAD\OrCAD_16.3 Demo.
Select Next, and the Start Copying Files dialog box opens. Choose Select again, and
the Ready to Install Program dialog box opens. Click Install, and the Installing
Crystal Report Xii box will appear. The Setup dialog box opens with the prompt:
Setup status installs program. The Install Wizard is now installing the OrCAD
16.3 Demo.
At completion, a message will appear: Searching for and adding programs to the
Windows firewall exception list. Generating indexes for Cadence Help. This
may take some time.
When the process has completed, select Finish and the Cadence OrCAD 16.3 screen
will appear. The software has been installed.

Screen Icon: The screen icon can be established (if it does not appear automatically) by
applying the following sequence. START-All Programs-Cadence-OrCAD 16.3 Demo-
OrCAD Capture CIS Demo, followed by a right-click of the mouse to obtain a listing
where Send to is chosen, followed by Desktop (create shortcut). The OrCAD icon will
then appear on the screen and can be moved to the appropriate location.

Folder Creation: Starting with the OrCAD opening screen, right-click on the Start
option at the bottom left of the screen. Then choose Explore followed by Hard Drive
(C:). Then place the mouse on the folder listing, and a right-click will result in a listing in
which New is an option. Choose New followed by Folder, and then type in OrCAD 11.3
in the provided area of the screen, followed by a right-click of the mouse. A location for all
the files generated using OrCAD has now been established.

Multisim
Installation:
Insert the Multisim disk into the DVD disk drive to obtain the Autoplay dialog box.
Then select Always do this for software and games, followed by the selection of
Auto-run to open the NI Circuit Design Suite 11.0 dialog box.
 Enter the full name to be used and provide the serial number. (The serial number PROBLEMS 51
appears in the Certificate of Ownership document that came with the NI Circuit
Design Suite packet.)
Selecting Next will result in the Destination Directory dialog box from which one will
Accept the following: C:\Program Files(X86) National Instruments\. Select Next
to open the Features dialog box and then select NI Circuit Design Suite 11.0.1
Education.
Selecting Next will result in the Product Notification dialog box with a succeeding
Next resulting in the License Agreement dialog box. A left-click of the mouse on I
accept can then be followed by choosing Next to obtain the Start Installation dialog
box. Another left-click and the installation process begins, with the progress being
displayed. The process takes between 15 and 20 minutes.
At the conclusion of the installation, you will be asked to install the NI Elvismx driver
DVD. This time Cancel will be selected, and the NI Circuit Design Suite 11.0.1
dialog box will appear with the following message: NI Circuit Design Suite 11.0.1
has been installed. Click Finish, and the response will be to restart the computer to
complete the operation. Select Restart, and the computer will shut down and start up
again, followed by the appearance of the Multisim Screen dialog box.
Select Activate and then Activate through secure Internet connection, and the Acti-
vation Wizard dialog box will open. Enter the serial number followed by Next to
enter all the information into the NI Activation Wizard dialog box. Selecting Next
will result in the option of Send me an email confirmation of this activation. Select
this option and the message Product successfully activated will appear. Selecting
Finish will complete the process.

Screen Icon: The process described for the OrCAD program will produce the same
results for Multisim.

Folder Creation: Following the procedure introduced above for the OrCAD program, a
folder labeled OrCAD 16.3 was established for the Multisim files.
The computer section of the next chapter will cover the details of opening both the
OrCAD and Multisim analysis packages, setting up a specific circuit, and generating a
variety of results.

PROBLEMS
●
*Note: Asterisks indicate more difficult problems.
1.3 Covalent Bonding and Intrinsic Materials
1. Sketch the atomic structure of copper and discuss why it is a good conductor and how its struc-
ture is different from that of germanium, silicon, and gallium arsenide.
2. In your own words, define an intrinsic material, a negative temperature coefficient, and cova-
lent bonding.
3. Consult your reference library and list three materials that have a negative temperature coeffi-
cient and three that have a positive temperature coefficient.
1.4 Energy Levels
4. a. How much energy in joules is required to move a charge of 12 mC through a difference in
potential of 6 V?
b. For part (a), find the energy in electron-volts.
5. If 48 eV of energy is required to move a charge through a potential difference of 3.2 V, deter-
mine the charge involved.
6. Consult your reference library and determine the level of Eg for GaP, ZnS, and GaAsP, three semi-
conductor materials of practical value. In addition, determine the written name for each material.
1.5 n-Type and p-Type Materials
7. Describe the difference between n-type and p-type semiconductor materials.
8. Describe the difference between donor and acceptor impurities.
9. Describe the difference between majority and minority carriers.
52 SEMICONDUCTOR 10. Sketch the atomic structure of silicon and insert an impurity of arsenic as demonstrated for
DIODES silicon in Fig. 1.7.
11. Repeat Problem 10, but insert an impurity of indium.
12. Consult your reference library and find another explanation of hole versus electron flow. Using
both descriptions, describe in your own words the process of hole conduction.

1.6 Semiconductor Diode

13. Describe in your own words the conditions established by forward- and reverse-bias conditions
on a p–n junction diode and how the resulting current is affected.
14. Describe how you will remember the forward- and reverse-bias states of the p–n junction
diode. That is, how will you remember which potential (positive or negative) is applied to
which terminal?
15. a. Determine the thermal voltage for a diode at a temperature of 20°C.
b. For the same diode of part (a), find the diode current using Eq. 1.2 if Is ⫽ 40 nA, n ⫽ 2 (low
value of VD), and the applied bias voltage is 0.5 V.
16. Repeat Problem 15 for T ⫽ 100°C (boiling point of water). Assume that Is has increased to 5.0 mA.
17. a. Using Eq. (1.2), determine the diode current at 20°C for a silicon diode with n ⫽ 2, Is ⫽
0.1 mA at a reverse-bias potential of -10 V.
b. Is the result expected? Why?
18. Given a diode current of 8 mA and n ⫽ 1, find Is if the applied voltage is 0.5 V and the tem-
perature is room temperature (25°C).
*19. Given a diode current of 6 mA, VT ⫽ 26 mV, n ⫽ 1, and Is ⫽ 1 nA, find the applied voltage VD.
20. a. Plot the function y = ex for x from 0 to 10. Why is it difficult to plot?
b. What is the value of y = ex at x ⫽ 0?
c. Based on the results of part (b), why is the factor ⫺1 important in Eq. (1.2)?
21. In the reverse-bias region the saturation current of a silicon diode is about 0.1 mA (T ⫽ 20°C).
Determine its approximate value if the temperature is increased 40°C.
22. Compare the characteristics of a silicon and a germanium diode and determine which you would
prefer to use for most practical applications. Give some details. Refer to a manufacturer’s listing
and compare the characteristics of a germanium and a silicon diode of similar maximum ratings.
23. Determine the forward voltage drop across the diode whose characteristics appear in Fig. 1.19 at
temperatures of ⫺75°C, 25°C, 125°C and a current of 10 mA. For each temperature, determine the
level of saturation current. Compare the extremes of each and comment on the ratio of the two.

1.7 Ideal versus Practical

24. Describe in your own words the meaning of the word ideal as applied to a device or a system.
25. Describe in your own words the characteristics of the ideal diode and how they determine the
on and off states of the device. That is, describe why the short-circuit and open-circuit equiva-
lents are appropriate.
26. What is the one important difference between the characteristics of a simple switch and those
of an ideal diode?

1.8 Resistance Levels

27. Determine the static or dc resistance of the commercially available diode of Fig. 1.15 at a for-
ward current of 4 mA.
28. Repeat Problem 27 at a forward current of 15 mA and compare results.
29. Determine the static or dc resistance of the commercially available diode of Fig. 1.15 at a reverse
voltage of ⫺10 V. How does it compare to the value determined at a reverse voltage of ⫺30 V?
30. Calculate the dc and ac resistances for the diode of Fig. 1.15 at a forward current of 10 mA and
compare their magnitudes.
31. a. Determine the dynamic (ac) resistance of the commercially available diode of Fig. 1.15 at a
forward current of 10 mA using Eq. (1.5).
b. Determine the dynamic (ac) resistance of the diode of Fig. 1.15 at a forward current of 10 mA
using Eq. (1.6).
c. Compare solutions of parts (a) and (b).
32. Using Eq. (1.5), determine the ac resistance at a current of 1 mA and 15 mA for the diode of
Fig. 1.15. Compare the solutions and develop a general conclusion regarding the ac resistance
and increasing levels of diode current.
 33. Using Eq. (1.6), determine the ac resistance at a current of 1 mA and 15 mA for the diode of PROBLEMS 53
Fig. 1.15. Modify the equation as necessary for low levels of diode current. Compare to the
solutions obtained in Problem 32.
34. Determine the average ac resistance for the diode of Fig. 1.15 for the region between 0.6 V
and 0.9 V.
35. Determine the ac resistance for the diode of Fig. 1.15 at 0.75 V and compare it to the average
ac resistance obtained in Problem 34.

1.9 Diode Equivalent Circuits

36. Find the piecewise-linear equivalent circuit for the diode of Fig. 1.15. Use a straight-line seg-
ment that intersects the horizontal axis at 0.7 V and best approximates the curve for the region
greater than 0.7 V.
37. Repeat Problem 36 for the diode of Fig. 1.27.
38. Find the piecewise-linear equivalent circuit for the germanium and gallium arsenide diodes of
Fig. 1.18.

1.10 Transition and Diffusion Capacitance

*39. a. Referring to Fig. 1.33, determine the transition capacitance at reverse-bias potentials of
⫺25 V and ⫺10 V. What is the ratio of the change in capacitance to the change in voltage?
b. Repeat part (a) for reverse-bias potentials of ⫺10 V and ⫺1 V. Determine the ratio of the
change in capacitance to the change in voltage.
c. How do the ratios determined in parts (a) and (b) compare? What does this tell you about
which range may have more areas of practical application?
40. Referring to Fig. 1.33, determine the diffusion capacitance at 0 V and 0.25 V.
41. Describe in your own words how diffusion and transition capacitances differ.
42. Determine the reactance offered by a diode described by the characteristics of Fig. 1.33 at a
forward potential of 0.2 V and a reverse potential of ⫺20 V if the applied frequency is 6 MHz.
43. The no-bias transition capacitance of a silicon diode is 8 pF with VK ⫽ 0.7 V and n ⫽ 1>2.
What is the transition capacitance if the applied reverse bias potential is 5 V?
44. Find the applied reverse bias potential if the transition capacitance of a silicon diode is 4 pF but
the no-bias level is 10 pF with n ⫽ 1>3 and VK ⫽ 0.7 V.

1.11 Reverse Recovery Time

45. Sketch the waveform for i of the network of Fig. 1.57 if tt = 2ts and the total reverse recovery
time is 9 ns.

FIG. 1.57
Problem 45.

1.12 Diode Specification Sheets

*46. Plot IF versus VF using linear scales for the diode of Fig. 1.37. Note that the provided graph
employs a log scale for the vertical axis (log scales are covered in Sections 9.2 and 9.3).
47. a. Comment on the change in capacitance level with increase in reverse-bias potential for the
diode of Fig. 1.37.
b. What is the level of C(0)?
c. Using VK ⫽ 0.7 V, find the level of n in Eq. 1.9.
48. Does the reverse saturation current of the diode of Fig. 1.37 change significantly in magnitude
for reverse-bias potentials in the range ⫺25 V to ⫺100 V?
54 SEMICONDUCTOR *49. For the diode of Fig. 1.37 determine the level of IR at room temperature (25°C) and the boiling
DIODES point of water (100°C). Is the change significant? Does the level just about double for every
10°C increase in temperature?
50. For the diode of Fig. 1.37, determine the maximum ac (dynamic) resistance at a forward cur-
rent of 0.1, 1.5, and 20 mA. Compare levels and comment on whether the results support con-
clusions derived in earlier sections of this chapter.
51. Using the characteristics of Fig. 1.37, determine the maximum power dissipation levels for the
diode at room temperature (25°C) and 100°C. Assuming that VF remains fixed at 0.7 V, how
has the maximum level of IF changed between the two temperature levels?
52. Using the characteristics of Fig. 1.37, determine the temperature at which the diode current
will be 50% of its value at room temperature (25°C).
1.15 Zener Diodes
53. The following characteristics are specified for a particular Zener diode: VZ ⫽ 29 V, VR ⫽ 16.8 V,
IZT ⫽ 10 mA, IR ⫽ 20 mA, and IZM ⫽ 40 mA. Sketch the characteristic curve in the manner
displayed in Fig. 1.47.
*54. At what temperature will the 10-V Zener diode of Fig. 1.47 have a nominal voltage of 10.75 V?
(Hint: Note the data in Table 1.7.)
55. Determine the temperature coefficient of a 5-V Zener diode (rated 25°C value) if the nominal
voltage drops to 4.8 V at a temperature of 100°C.
56. Using the curves of Fig. 1.48a, what level of temperature coefficient would you expect for a
20-V diode? Repeat for a 5-V diode. Assume a linear scale between nominal voltage levels and
a current level of 0.1 mA.
57. Determine the dynamic impedance for the 24-V diode at IZ = 10 mA for Fig. 1.48b. Note that
it is a log scale.
*58. Compare the levels of dynamic impedance for the 24-V diode of Fig. 1.48b at current levels of
0.2, 1, and 10 mA. How do the results relate to the shape of the characteristics in this region?

1.16 Light-Emitting Diodes

59. Referring to Fig. 1.52e, what would appear to be an appropriate value of VK for this device?
How does it compare to the value of VK for silicon and germanium?
60. Given that Eg ⫽ 0.67 eV for germanium, find the wavelength of peak solar response for the
material. Do the photons at this wavelength have a lower or higher energy level?
61. Using the information provided in Fig. 1.52, determine the forward voltage across the diode if
the relative luminous intensity is l.5.
*62. a. What is the percentage increase in relative efficiency of the device of Fig. 1.52 if the peak
current is increased from 5 mA to 10 mA?
b. Repeat part (a) for 30 mA to 35 mA (the same increase in current).
c. Compare the percentage increase from parts (a) and (b). At what point on the curve would
you say there is little to be gained by further increasing the peak current?
63. a. If the luminous intensity at 0° angular displacement is 3.0 mcd for the device of Fig. 1.52,
at what angle will it be 0.75 mcd?
b. At what angle does the loss of luminous intensity drop below the 50% level?
*64. Sketch the current derating curve for the average forward current of the high-efficiency red
LED of Fig. 1.52 as determined by temperature. (Note the absolute maximum ratings.)