17 Dielectric Materials

CHAPTER

When classifying solids on the basis of their band structure, we referred to the
group of solids, called insulators, which have an energy gap of 3 eV or more,
recall Sec. 15.1. The large magnitude of the energy gap in an ideal insulator
precludes the possibility of electrons being excited from the valence band to the
conduction band by thermal means, much less so by an externally applied
electric field. Insulators are therefore very poor conductors of electricity, recall
the resistivity range for insulators from Table 14.1. Insulators are known as
dielectrics. Dielectric materials find extensive use in the electrical industry for
insulation purposes and as capacitors.
In this chapter, polarization processes in dielectric materials and their
temperature and frequency dependence are discussed. Besides, the causes of
electric breakdown are described. In the end, the nature of ferroelectric crystals
is considered briefly.

Units

SI units
Quantity
Unit Symbol
Flux density D coulomb per square metre C m–2
123

Electric field strength E volt per metre V m–l

Dielectric strength or or
newton per coulomb N C–1
Capacitance C farad or coulomb per volt F or C V–1
Dielectric constant of farad per metre F m–l
free space e0
Relative dielectric constant er – –
Dipole moment coulomb metre Cm
123

Polarization P
Saturation polarization Ps coulomb per square metre C m–2
Electronic polarizability farad metre squared F m2
Frequency n hertz (per second) Hz (s–1)

412
 Polarization 413

Constants

Dielectric constant of free space e0 = 8.854 ´ 10–12 F m–1

Electronic charge e = 1.602 ´ 10–19 C

17.1 Polarization

The relation between the electric flux density D (charge per unit area in units of
C m–2) at a point in a material and the electric field strength E (force per unit
charge in units of N C–1 or V m–1) at that point in space is given by
D = er e0 E (17.1)
where e0 is the dielectric constant or permittivity of free space (vacuum) and er
is the relative dielectric constant or relative permittivity of the material. er is
dimensionless and is a property of the material related to its atomic structure. Its
value is 1 for free space and is greater than one for all materials. In SI units,
e0 = 8.854 ´ 10–12 farad per metre. e0 has this particular value as a result of
conversion from cgs units to SI units.

D C m –2
e0 = = F m–1
Fr E V m –1
When an electric field is applied to a solid containing positive and negative
charges, the positive charges are displaced in the direction of the field towards
the negative end, while the negative charges are displaced in the opposite
direction. This displacement produces local dipoles throughout the solid. The
dipole moment per unit volume of the solid is the sum of all the individual
dipole moments within that volume and is called the polarization P of the solid.
As the polarization measures the additional flux density arising from the
presence of the material as compared to free space, it has the same units as D
and is related to it as follows:
D = e0 E + P (17.2)
As may be seen by comparing Eq. (17.2) with Eq. (16.3), polarization P is
the electrical analog of magnetization M. Combining Eqs. (17.1) and (17.2), we
can write
P = e0 (er – 1)E (17.3)
Macroscopically, the polarization or the relative dielectric constant can be
measured, using a parallel-plate capacitor. When a voltage V is applied to the
capacitor, with the plates separated by vacuum, a charge develops on the plates.
The capacitance C of the capacitor is a measure of this charge and is defined by

F0 A
C= (17.4)
d
414 Dielectric Materials

where A is the area of the parallel plates and d is the distance of separation
between them. If a dielectric material is inserted between the plates, the charge
on the plates increases due to polarization in the material. The capacitance is
now given by
A
C = er e0 (17.5)
d
er and hence P can be determined by measuring the capacitance with and
without the dielectric.

Example 17.1 Calculate the relative dielectric constant of a barium titanate

crystal, which, when inserted in a parallel plate condenser of area 10 mm ´ 10 mm
and distance of separation of 2 mm, gives a capacitance of 10–9 F.

Solution Substituting the given values in Eq. (17.5), we obtain

10 9 – 2 – 10 3
er =
8.854 – 10 12 – 10 2 – 10 6

= 2259

Polarization occurs due to several atomic mechanisms. Electronic

polarization (Fig. 17.1a) is the result of the displacement of the positively
charged nucleus and the (negative) electrons of an atom in opposite directions on
application of an electric field. On applying a field, the electron cloud around
the nucleus readily shifts towards the positive end of the field. Such a shift
results in a dipole moment within the atom, as a certain distance now separates
the nucleus and the centre of the electron cloud. The extent of this shift is
proportional to the field strength. As the dipole moment is defined as the product
of the charge and the shift distance, it is also proportional to the field strength.
The constant of proportionality is called the electronic polarizability ae of the
atom. For the inert gases, this polarizability increases with increasing volume of
the atom, as illustrated in Table 17.1. Electronic polarizability is independent of
temperature. Monoatomic gases exhibit only this kind of polarization.

TABLE 17.1
Electronic Polarizability ae of Inert Gases

Inert gas He Ne Ar Kr Xe
ae, 10–40 F m2 0.18 0.35 1.43 2.18 3.54

During chemical bonding, the atoms may acquire an excess negative or

positive charge and form an ionic bond. When an electric field is applied to an
ionic solid, cations and anions get displaced in opposite directions, see
 Polarization 415

No field Field

(a)

(b)

(c)

(d)

Fig. 17.1 Various polarization processes: (a) electronic polarization, (b) ionic
polarization, (c) orientation polarization, and (d) space charge polarization.

Fig. 17.1b. The ionic polarizability is due to this shift of the ions relative to
other oppositely-charged neighbours. It should be distinguished from electronic
polarization, where the electron cloud of an atom shifts with reference to its own
nucleus. Ionic polarization is also independent of temperature.
In methane molecule (CH4), the centre of the negative and the positive
charges coincide, so that there is no permanent dipole moment. On the other hand,
in CH3Cl, the positive and the negative charges do not coincide. The
electronegativity of chlorine being more than that of hydrogen (recall Table 4.4),
the chlorine atom pulls the bonding electrons to itself more strongly than
hydrogen. So, this molecule carries a dipole moment even in the absence of an
electric field. When an electric field is applied on such molecules, they tend to
align themselves in the applied field, see Fig. 17.1c. (Recall that atoms with
permanent magnetic moments tend to align themselves with the applied magnetic
field giving rise to paramagnetism.) The polarization due to this alignment is
called orientation polarization and is dependent on temperature. With increasing
temperature, the thermal energy tends to randomize the alignment.
416 Dielectric Materials

The fourth type of polarization is called the space charge polarization and
occurs due to the accumulation of charges at the electrodes or at the interfaces in
a multiphase material, see Fig. 17.1d. The ions diffuse over appreciable distances
in response to the applied field, giving rise to a redistribution of charges in the
dielectric medium.
The total polarization of a material is the sum of the contributions from the
various sources described above:

Ptotal = Pe + Pi + Po + Ps (17.6)

where the subscripts on the right refer to the four types: eletronic, ionic,
orientation and space-charge polarization.

17.2 Temperature and Frequency Effects

On application of an electric field, a polarization process occurs as a function of

time. The polarization P(t) as a function of time t is given by

Ë È t ØÛ
P(t) = P Ì1  exp É  Ù Ü (17.7)
Í Ê tr Ú Ý

where P is the maximum polarization attained on prolonged application of a

static field and tr is the relaxation time for the particular polarization process.
Recall Eq. (10.15), where the relaxation arising from anelastic processes was
discussed. The relaxation time tr is a measure of the time scale of a polarization
process. It is the time taken for a polarization process to reach 0.63 of the
maximum value.
The relaxation times vary widely for different polarization processes.
Electronic polarization is extremely rapid and is essentially complete at the
instant the voltage is applied. Even when the frequency of the applied voltage is
very high in the optical range (~1015 Hz), electronic polarization occurs during
every cycle of the applied voltage.
Ionic polarization is slower than electronic polarization, as the displacement
involved here is that of the much heavier ion, as compared to the electron cloud in
the above case. The frequency with which ions can be displaced over a small
fraction of the interatomic distance will be of the same order as the lattice
vibration frequency (~1013 Hz). If an electric field of frequency in the optical
range (~1015 Hz) is now applied, the ions do not respond at all, as the time
required by an ion for one vibration is 100 times larger than the period of the
applied voltage. So, at optical frequencies, there is no ionic polarization. If the
frequency of the applied voltage is less than 1013 Hz, the ions have enough time to
 Temperature and Frequency Effects 417

respond during each cycle. This is illustrated in Fig. 17.2, where the polarization
consists of only the electronic part at optical frequencies and increases to include
the ionic part, as the frequency is reduced to 1013 Hz in the infrared range.

Space charge
Polarization

Orientation

Ionic
electronic
Power loss

Electrical frequencies Optical

Power Audio Radio Infrared Visible
Frequency
Fig. 17.2 Frequency dependence of polarization processes and peak power losses
corresponding to each process.

Orientation polarization is even slower than ionic polarization. The

relaxation time for orientation polarization in liquid propyl alcohol is about
10–10 s and in solid ice, it is 3 ´ 10–6 s. It is evidently easier for the polar
molecules to reorient themselves in a liquid as compared to a solid. Orientation
polarization occurs, when the frequency of the applied voltage is in the audio
range, Fig. 17.2.
Space charge polarization is the slowest process, as it involves the diffusion
of ions over several interatomic distances. The relaxation time for this process is
related to the frequency of successful jumps of ions under the influence of the
applied field, a typical value being 102 Hz. Correspondingly, space charge
polarization occurs at machine frequencies (50–60 Hz).
Referring to Fig. 17.2, all the four types of polarization are present at
machine frequencies. As the frequency is increased, space charge, orientation
and ionic polarization become inoperative in that order. When several
polarization processes occur in a material, it follows that the dielectric constant
will decrease with increasing frequency of the applied voltage. The dielectric
constants of some typical dielectric materials are listed in Table 17.2 at
frequencies of 60 Hz and 106 Hz.
418 Dielectric Materials

TABLE 17.2
Properties of Some Dielectric Materials

er tan d , Dielectic
Material
60 Hz 106 Hz 106 Hz strength,
106 V m–1
Electrical porcelain 6 6 0.02 5
Steatite, MgO×SiO2 6 6 0.001 12
Fused silica 4 3.8 0.0001 10
Soda-lime-glass 7 7 0.005 10
Mica 8 5 0.0005 100
Nylon 6, 6 4 3.5 0.02 15
Polyethylene 2.3 2.3 0.0004 4
Polyvinylchloride 7 3.4 0.05 2
(plasticized)
Vulcanized rubber 4 2.7 0.003 25
Bakelite 4.4 4.4 0.028 15
Transformer oil 5 2.5 0.0001 10

When the period of the applied voltage is much larger than the relaxation
time of a polarization process, polarization is essentially complete at any instant
during each cycle. The charging current is 90° advanced in relation to the voltage,
as illustrated in Fig. 17.3a, so that no electrical energy is lost during charging.

I I
I
d

V V V
(a) (b) (c)

Fig. 17.3 (a) and (b) There is no energy loss when the period of the ac voltage
does not match the relaxation time of a process. (c) When there is matching,
energy loss occurs, with the current leading the voltage by less than 90°.

When the period of the applied voltage is much shorter than the relaxation time
for a polarization process, the polarization does not occur at all. Here again, the
charging current is 90° advanced of the applied voltage, see Fig. 17.3b. When the
period is in the same range as the relaxation time, resonance occurs. Here, the
current leads the voltage by (90 – d ), as illustrated in Fig. 17.3c. d is called the
loss angle and tan d is taken as a measure of the electrical loss due to resonance.
The current can be factorized into a component at 90° to the voltage as in an ideal
 Electric Breakdown 419

capacitor and another component parallel to the voltage. The parallel component
is the real part and results in I 2R loss. The tan d values of some dielectric
materials at a frequency of 106 Hz are listed in Table 17.2.
The effect of temperature on the relative dielectric constant of a material can
be two-fold. In orientation polarization, the randomizing action of thermal
energy decreases the tendency for the permanent dipoles to align themselves in
the applied field. This results in a decrease in the dielectric constant with
increasing temperature. For example, er for solid HCl decreases inversely with
temperature between 100 K and 160 K from 19 to 14. At the melting point of
160 K, it drops abruptly to 12, as there is an expansion on melting and the
number of molecules per unit volume decreases.
The other effect of temperature is to facilitate the diffusion of ions in space
charge polarization. Thermal energy may also aid in overcoming the activation
barrier for the orientation of polar molecules in the direction of the field. For
example, the relatively large polar molecule nitrobenzene is not able to reorient
itself in the solid state under an applied field and, therefore, does not exhibit
orientation polarization. However, on melting, the molecules have sufficient
thermal energy to orient themselves in the applied field. Correspondingly, the
relative dielectric constant increases here from 3 to 37 on melting, which is in
contrast to the behaviour of HCl described above.

17.3 Electric Breakdown

In practical applications, the failure or breakdown of a dielectric material is of

great concern to the engineer. In a dielectric, the charge displacement increases
with increasing field strength. Beyond a critical value of the field strength, there
is an electric breakdown due to the physical deterioration of the material. The
dielectric strength is defined as the breakdown voltage per unit thickness of the
material. The dielectric strengths of some common insulating materials are listed
in Table 17.2. Mica has one of the highest dielectric strengths and is widely used
for insulation purposes.
The intrinsic breakdown of a dielectric material is due to the excitation of
electrons into the conduction band across the energy gap under conditions of
excessive voltage. The excited electrons moving under a high accelerating force
can excite more electrons in turn, resulting in an avalanche of conducting
electrons and consequent physical breakdown. It is difficult to see how this can
occur in an ideal dielectric, where the energy gap is large. However, impurities
in the dielectric can create additional energy levels that lie in the energy gap
and can help in the excitation of electrons into the conduction band. An increase
in temperature aids the thermal excitation of electrons and can bring about the
intrinsic breakdown.
Thermal breakdown is due to the attainment of an excessive temperature in
the dielectric. The energy loss referred to earlier has to be dissipated as heat. If
the heat dissipated is less than the heat generated, there is a progressive increase
in the temperature of the dielectric, which may melt eventually.
420 Dielectric Materials

Defect breakdown is due to cracks and pores at the surface. To decrease the
possibility of surface shorting, insulators are designed with lengthened surface
paths. Moisture from the atmosphere can collect on the surface discontinuities
and result in breakdown. Glazing is done on ceramic insulators to make the
surface nonabsorbent. Gases can collect at pores and cracks and the breakdown
can occur due to a gas discharge. A graphic case of surface breakdown of an
insulator is shown in Fig. 17.4.

Fig. 17.4 Surface breakdown of an insulator. (L.H. Van Vlack, Physical Ceramics
for Engineers, by permission from Addison-Wesley, Reading, Mass.)

17.4 Ferroelectric Materials

In materials known as ferroelectrics, the dielectric constants are some two orders
of magnitude larger than those in ordinary dielectrics. Barium titanate is a
ferroelectric with a relative dielectric constant of over 2000, compared to less
than 10 for ordinary dielectrics listed in Table 17.2. The difference in the
magnetic susceptibility between ferromagnetic and paramagnetic materials bears
a direct analogy to this difference in the values of the dielectric constants.
Following the nomenclature in magnetism, materials of very large dielectric
constants are called ferroelectrics. As in the ferromagnetic phenomenon, the
electric dipoles in a ferroelectric solid are all aligned in the same direction, even
in the absence of an electric field.
 Ferroelectric Materials 421

The ferroelectric phenomenon is discussed here with reference to the

classical example of barium titanate, BaTiO3. The cubic unit cell of barium
titanate crystal is shown in Fig. 17.5a. Barium ions are at the body corners, the

0.08 Å

0.06 Å 0.06 Å

Ti4+
O2–

(a) Ba2+ (b)

Fig. 17.5 (a) Cubic unit cell of BaTiO3 crystal. (b) The dashed circle in the
middle of the front view shows the shifting of the titanium ion, on cooling
through the Curie temperature. The shift of the oxygen anions is also shown.

oxygen anions are at the face centres and the titanium ion is in the octahedral
void at the body centre. Only one out of four octahedral voids in the unit cell is
occupied and this corresponds to the chemical formula, with one titanium for
every four species of the other kinds: one barium plus three oxygen. Above
120°C, barium titanate is a cubic crystal with the ion locations as described
above. In this state, the centres of the negative and the positive charges coincide
and there is no spontaneous dipole moment. If the crystal is cooled to below the
(ferroelectric) Curie temperature of 120°C, the titanium ion shifts to one side of
the body centre as shown dotted in the front view of Fig. 17.5b. There is also a
displacement of the neighbouring oxygen anions. The crystal transforms from a
cubic to a tetragonal structure on cooling through 120°C. The c/a ratio of the
tetragonal cell is 4.03 Å/3.98 Å = 1.012. The centres of the positive and the
negative charges do not coincide any longer and local dipoles are created
throughout the crystal. The dipoles of neighbouring unit cells are all aligned
resulting in a large polarization in the solid.

Example 17.2 Calculate the polarization of a BaTiO3 crystal. The shift of

the titanium ion from the body centre is 0.06 Å. The oxygen anions of the side
faces shift by 0.06 Å, while the oxygen anions of the top and bottom faces shift
by 0.08 Å, all in a direction opposite to that of the titanium ion.
422 Dielectric Materials

Solution The dipole moments due to the effective number of each type
of ion in the unit cell are due to
(i) two O2– on four side faces: 2 ´ 2 ´ 1.6 ´ 10–19 ´ 0.06 ´ 10–10 C m;
(ii) one O2– on top and bottom: 1 ´ 2 ´ 1.6 ´ l0–19 ´ 0.08 ´ 10–10 C m; and
(iii) one Ti4+ at body centre: 1 ´ 4 ´ 1.6 ´ 10–19 ´ 0.06 ´ 10–10 C m
Total = 1.02 ´ 10–29 C m
Polarization is the sum of the dipole moments per unit volume. Ignoring the
effect due to the barium ions, we get
1.02 – 10 29
P= C m–2 = 0.16 C m–2.
4.03 – 3.982 – 10 30

At room temperature, a BaTiO3 crystal ordinarily exhibits no net

polarization, in the absence of an external field, even though the dipoles of
adjacent unit cells are aligned. The reason for this can be understood by
visualizing ferroelectric domains in the same way as ferromagnetic domains
discussed in Sec. 16.4. The application of the electric field tends to align the
domains in the direction of the field and we observe all the phenomena
associated with the hysteresis loop such as domain rotation and domain growth.
Figure 17.6 illustrates a ferroelectric hysteresis loop. The spontaneous

Ps

–Ec
E
0

Fig. 17.6 The hysteresis loop for a ferroelectric material, depicting applied field
E versus polarization P. Ps is spontaneous polarization.

polarization Ps is obtained by extrapolating the linear region of the curve

backwards to zero electric field. The ferroelectric Curie temperatures and the
spontaneous polarization Ps below the Curie point of some ferroelectric crystals
are listed in Table 17.3.
 Summary 423

TABLE 17.3
Properties of Some Ferroelectric Crystals

Crystal Curie temperature, K Spontaneous polarization,

Ps, C m–2
BaTiO3 393 0.26
KNbO3 712 0.30
PbTiO3 763 0.50
KH2PO4 123 0.05

Ferroelectric crystals always exhibit the piezoelectric property, which is the

mechanical response of a crystal to an electric field or the electrical response to
a mechanical stress. With no external field, the centres of the positive and the
negative charges are separated by a distance d. If a compressive stress is applied
to the crystal, d decreases and a potential difference V develops between the two
ends of the crystal. V will be of the opposite sign if a tensile stress is applied. If
an external voltage is applied to a ferroelectric crystal, the separation distance d
increases (or decreases), thereby elastically straining the crystal. The most
important use of BaTiO3 is as a piezoelectric crystal in applications such as
microphones, phonograph pickups, strain gauges and sonar devices. The high
dielectric constant of ferroelectric crystals is also useful for storing energy in
small-sized capacitors in electrical circuits.

SUMMARY

1. The charge stored in a capacitor increases if a dielectric material is

inserted between the capacitor plates. This is due to the polarization in the
dielectric.
2. Several polarization processes such as electronic, ionic, orientation and
space charge polarization are known. They have widely varying relaxation
times.
3. The dielectric constant of a material is dependent on the frequency of the
alternating field applied and the temperature.
4. Electrical energy loss is a maximum when the relaxation time of a
polarization process matches the period of the alternating field.
5. Impurities, surface flaws such as cracks and pores and absorption of
moisture and gases on the surface are the main causes of electric
breakdown of a dielectric material.
6. Ferroelectric crystals are in the spontaneously polarized state and have
very large dielectric constants. They are useful as piezoelectric crystals and
as capacitors.
424 Problems

PROBLEMS

17.1 The electric field strength can be expressed as V m–1 or as N C–1. Express
these in base SI units and show that they are equivalent.
17.2 Derive the units of ae from that of E and the dipole moment.
17.3 Find the capacitance of a layer of Al2O3 that is 0.5 mm thick and
2000 mm2 of surface area. er = 8.
Answer: 2.83 ´ 10–7 F.
17.4 The electronic polarizability of helium is 0.18 ´ 10–40 F m2. Calculate its
relative dielectric constant at 0°C and 1 atm pressure.
Answer: 1.000 055.
17.5 Estimate the shift of the electron cloud with respect to the nucleus in an
argon atom, when a field of 105 V m–1 is applied.
Answer: 5 ´ 10–18 m.
17.6 The dielectric constant of polyethylene is independent of temperature,
while that of polyvinylchloride is not. Explain this difference in
behaviour on the basis of their monomer structures.
17.7 When ice melts into water, the dielectric constant increases, in contrast to
the decrease observed during the melting of HCl. Explain why this is so.
17.8 What is the variation expected in the loss factor in a sodium silicate
glass, when some of the Na+ ions are replaced by Rb+ ions? Explain.
17.9 Compare the hysteresis loops in Figs. 16.4 and 17.6. Explain why there is
a constant saturation induction Bs in Fig. 16.4, and no saturation value of
polarization in Fig. 17.6.

MULTIPLE CHOICE QUESTIONS

1. The units of relative dielectric constant is

A. dimensionless B. F m–1 C. C V–1 D. F C–1
2. The units of e0 are
A. C m–2 B. H m–l C. F m–l D. dimensionless
3. The capacitance of a capacitor with a layer of Al2O3 (er = 8) of 0.5 mm
thick and 1000 mm2 area is
A. 0.283 mF B. 0.142 mF C. 17 nF D. 3540 pF
4. Expressed in base SI units, the electronic polarizability has the units
A. F m2 B. C V –1 m–2 C. A2 kg–1 s4 D. A kg–1 m–2 s2
 Multiple Choice Questions 425

5. If the electronic polarization of W is 4 ´ 10–7 C m–2, the average

displacement of the electrons relative to the nucleus is (W : at.no. 74,
BCC, a = 3.16 Å)
A. 3.9 Å B. 5.3 ´ 10–19 m C. 10.6 ´ 10–19 m D. 3.9 ´ 10–17 m
6. The shift of the electron cloud with respect to the nucleus of a helium atom
when a field of 105 V m–1 is applied is (ae for He = 0.18 ´ 10–40 F m2)
A. 5.6 ´ 10–18 m B. 6.2 ´ 10–19 m C. l.7 ´ 10–10 m D. l.7 ´ 10–18 m
7. Ionic polarization
A. decreases with temperature
B. increases with temperature
C. may increase or decrease with temperature
D. is independent of temperature
8. With increase in temperature, the orientation polarization in general
A. increases B. decreases C. is constant D. none of these
9. Among the common dielectric materials, the highest dielectric strength is
possessed by
A. mica B. transformer oil
C. PVC D. polyethylene
10. During melting, the relative dielectric constant
A. always increases B. always decreases
C. may increase or decrease D. none of these
11. The fraction of tetrahedral voids occupied by Ti4+ ion in BaTiO3 crystal is
A. 1/2 B. 1/4 C. 1/8 D. none of these
12. In the polarization versus field strength plot for a ferroelectric crystal, Ps
stands for
A. space charge polarization B. saturation polarization
C. spontaneous polarization D. none of these
13. A piezoelectric crystal has a Young’s modulus of 130 GPa. The uniaxial
tensile stress that must be applied to increase its polarization from 550 to
555 C m–2 is
A. 1.171 GPa B. 1171 MPa C. 2600 MPa D. 1.182 GPa

Answers

1. A 2. C 3. B 4. C 5. B
6. A 7. D 8. B 9. A 10. C
11. D 12. C 13. D