D

All rights are reserved.
No part of this publication may be reproduced, stored in a

retrieval system, in any form in electronic, recording etc., without the prior permission of
the publisher.
Published by:
National Journal of Environment and Scientific Research
8-G-6
Jawahar Nagar
Sriganganagar-335001
Rajasthan
Mobile: 9780181083 8209583526 (Editor-in-Chief)
Email : editor@njesr.com
CONTENTS
Chapter-I
Dielectric properties of materials 1-13
Chapter-II
Dielectric properties of alternating electric fields 14-25
Chapter-III
Dielectric properties of static electric fields 26-34
Chapter-IV
Ferroelectric materials 35-55
Chapter-V
Piezoelectric materials 56-68

About the author:
Dr G. NagaRaju has 18 years experience in Education and Research. He is presently working as
Associate Professor in the Department of Physics at Acharya Nagarjuna University,
AndhraPradesh. He has obtained his Master degree in Physics with Solid State Physics as
specialization and secured Gold Medal from Acharya Nagarjuna University. He obtained M.Phil
and Ph.D. degree in Physics from the same University in the area of Glass Science. He has 53
research publications in refereed International Journals in the area of Glass Science, Material
Science and Nano-Technology. His research interest includes Glass Ceramics, Nonlinear Optical
Studies and Luminescence. He has presented more than 80 research papers in National &
International Conferences/Seminars. He has guided and awarded Four (04) M.Phil and Three
(03) Ph.D degrees. Presently he is guiding research scholars for their M.Phil. and Ph.D. degree in
the Department of Physics, Acharya Nagarjuna University in the area of Glass Science. He also
published a chapter titled “Dielectric dispersion in ZnO-ZnF2-B2O3 glasses doped with
WO3”Focus on Condensed Matter Physics Chapter 5(2005) 147-181 (Nova Science
Publishers, U.S.A). and published a Book titled “Influence of some transition metal oxides
into zinc oxy fluoro borate glasses”. (ISBN: 978-93-84502-75-1, Agri-Biovet Press, New
Delhi).
Dr. G. Naga Raju

Associate Professor
Department of Physics
Acharya Nagarjuna University
Guntur-522 510
Email: gnaga9@gmail.com
DOI - 10.53571/NJESR.2021.2.8.1-68
Chapter I Dielectric properties of materials
DIELECTRIC PROPERTIES OF MATERIALS
1. Introduction
Dielectrics are insulators i.e., non-conductors of electricity. The function of any insulator is to
prevent the flow of electricity through it when a potential difference is applied across its ends. These
materials prevent the leakage of electrical charges in electrical devices. Substances like bakelite, PVC
used in electrical wiring and pipes, polymer materials etc., come under this category. Dielectrics possess
high resistivity values in the range 106 -m to 1016-m. Under high voltage bias, they allow very little
current (10-6 A to 10-14 A). They withstand very high voltages. The conduction phenomenon in
dielectrics is mostly associated with ionic motion through defects or hopping of charges. They have no
free charges. They consist of positively and negatively charged particles bound together. The
fundamental action of the electrical field is to separate positive and negative charges of the entire volume
of the dielectric, causing what is known as the polarization of the dielectric. Fig.1. shows the effect of
polarization in a dielectric when external field E0 is applied on a dielectric. We see that the net
polarization charges produced at the faces of the dielectric, a positive charge on the right and a negative
on the left; inside the medium there is no excess charge in any given volume element. The medium as a
whole remains neutral, and the positive charge on the right is equal in magnitude to the negative charge
on the left. These induced charges create their own electric field Ep called polarization field that is
directed to the left, and thus oppose the external field E0. When we add this polarization field Ep to the
external
Eo
E Ep
+ - + -
+ - + -
+ - + -
+ - + -
+ - + -
Fig.1. Effect of polarization in a dielectric.
field E0, so as to obtain the effective field E, we find that E<E0. Therefore, effect of introducing insulating
substance (i.e., dielectric) results in reduction in applied field or reduction in surface charge density.
Thus, the polarization of the medium reduces the electric field in its interior. During the polarization the
1
charges in the dielectric are displaced from their equilibrium positions by distances that are considerably
less than atomic diameter. There is no transfer of charge over macroscopic distances such as occur when
a current is set up in a conductor.
Dielectrics: Dielectrics are the insulating materials having electric dipoles permanently or temporarily by
inducement during the application of electric field.
Electric Field Strength or Intensity (E): The space around the charged body, up to where its influence
felt is called Electric Field. Suppose an additional infinitesimal test charge q 0 is brought into the electric
field and at a certain point in it, it experiences an electrostatic force F. The electric field strength or
intensity E at the point is a vector and defined by
F
E= volt/metre
q0
Electric Field Induction (or) Flux density (or) Displacement Vector (D): Consider a charge q at the
centre of a sphere of radius r. The charge q will send q lines of force and this will be received by surface
area 4 r2.The number of electric lines of force received by a unit area is called flux density or electric
displacement D.
q q
i.e., D  ; where A is the surface area of the sphere,
4 r 2 A
The unit of electric flux density is coulomb/metre2.
Electric dipoles: The system of two equal and opposite charges separated by certain distance is called
electric dipole is shown in Fig. 2.
-q - - - - - - -p +q
Fig.2 An electric dipole
Electric Dipole moment: The product of any one of two charges of dipole and the separation between
them is called electric dipole moment.
Let the two charges are +q and –q separated by a distance r. The moment of this dipole is defined
as
p = qr
2
The dipole moment is therefore equal to the magnitude of the one of the charges times the distance
between them. The unit of electric dipole moment is esu-cm (10-18 esu-cm = 3.3 x 10-30 C-m = 1
debye)
Polarization (P): The process of producing electric dipoles out of neutral atoms and molecules
is known as polarization. Polarization P in a solid is defined as the total dipole moment per unit
volume:
q r i i
P= p
n
i  n
V
Here P is the total dipole moment (including the induced and permanent) and n is the
number of dipoles per unit volume. Polarization P has the same units as the surface
charge density (C-m-2). This equivalence is substantiated by the fact that electric field
induces charges on the surface of the dielectric and the density of charges is a measure
of the extent of polarization.
Dielectric Constant: Dielectric constant or relative permittivity is defined as the ratio of permittivity of
the substance to the permittivity of the free space,
Consider a parallel plate capacitor consisting of two plane parallel plates of area A and separation
d, charged with a surface charge density . If the space between the plates is vacuum and if d is small
compared with the dimensions of the plates.
Suppose now that the space between the plates is filled with an insulating substance the charge
on the plates being kept constant. The new potential difference V is lower than V vac and the
capacitance is increased.
The static dielectric constant is then defined by
Vvac C
= =
V C vac
Thus, the field strength is reduced from the value E vac to the value E, where
E vac
=
E
E vac
in other words, the effective surface charge density on the plates is now changed from  =
4
E
to l = .
4
3
The effect of introducing the insulating substance is thus to reduce the surface charge density by an
amount
E vac E E
 - l = - = ( - 1)
4 4 4
Since the charge on the plates is being kept constant, the positive plate thus acquires a negative induced
surface charge density ( - l) and vice versa; whole of the dielectric becomes a single dipole of moment
( - l)Ad. Under this condition, and using equation we see that ( - 1) = P. Thus, the quantity on the
left hand side of the of equation is the polarization of the dielectric and we can write
E
P = ( - 1)
4
The above explanation of the induction of charges at the surface of the dielectric is in accordance with
that considered earlier.
D = E + 4 P = E
Dielectric constant expresses the properties of the medium: all dielectric and optical properties of
the medium are contained in this constant.
Susceptibility (  ): It is defined as polarization per unit electric field.
P
  ;
E
It measures the amount of polarization a given field produces. In empty space P=0,  = 0,  = 1.
Polarizability(): The strength of the induced dipole moment an atom acquires is directly proportional
to the strength of the external applied field
i.e., p  E
p = E
where  is known as dielectric polarizability. We can relate polarizability , which is an atomic property
to the macroscopic property polarization P. It has the dimensions of volume.
1.1polarizability:
Polarization occurs due to several microscopic mechanisms. Polarization is a consequence of the
fact that when an electric field acts on a molecule/atom, its positive charges (nuclei) are displaced along
the field while the negative charges (electrons) in a direction opposite to that of the filed. The opposite
charges are thus pulled apart and the molecule is polarized. The displacements of electrical charges result
4
the formation of dipoles. Particularly in d.c. electric fields, the macroscopic polarization vector P is
created by three types of mechanisms and hence polarization can be broadly classified into three types:
1. Electronic Polarization
2. Ionic Polarization
3. Orientational Polarization
1.1. (a) Electronicpolarization: Electronic polarization is due to displacement of charge centres of
electron cloud (negative charge centre) and nucleus (positive charge centre) of an atom in the presence of
an applied electric field.
Although we are interested in the dielectric properties of solids, it will be useful to consider first
the much simpler problem of the behaviour of free atoms and molecules in an external field.
Consider an atom of a dielectric material such that its atomic number is equal to ‘Z’ and atomic
radius ‘r’. The centres of gravities of charges of electron cloud and positive nucleus are at the same point
and hence there is no displacement. Suppose if the atom is placed in a d.c. electric field of strength ‘E’,
the nucleus and the electron cloud experiences Lorentz forces of magnitude “ZeE” in opposite directions.
i.e., nucleus and electron cloud are pulled apart, therefore an attractive coulomb force develop between
them. When the Lorentz force and coulomb attractive forces are equal and opposite, there is a new
equilibrium between the nucleus and the electron cloud of the atom and hence dipole is formed. Let the
distance of separation between the centres of the displaced nucleus and electron cloud is ‘d’.
In Fig. 3 atom without any field and atom with field.
Fig. 3.Atom without any field and atom with field.

4
The negative charge enclosed in the sphere of radius ‘r’ is equal to  d 3
3
 
  Ze 
Where ‘  ’ is the charge density of electron cloud, and is equal to  
 4 r 3 
 3 
Therefore charge enclosed in the sphere of radius, d is,
5
 
4 3  - Ze    Zed 3 
 d   =  3 
3  4 r 3   r 
 3 
Therefore, Coulomb force of attraction, Fc
Ze.[charge enclosed in the sphere of radius d].


d2
 Ze.Zed 3  Z 2 e 2 d
Fc  
d 2r 3 r3
Lorentz force of repulsion experienced by the electron due to applied field ‘E’ is
FL  ZeE
In equilibrium condition,
 Z 2e 2 d
  ZeE
r3
Zed
E
r3
r3
d E
Ze
dE
i.e., the separation between the two charge centres is proportional to the applied field ‘E’.
The induced electric dipole moment,

p = Zed = r3E,
and the induced polarizability

p
e = = r3
E
Hence, e has the dimensions of a volume. It is also evident that in general atoms with many electrons
tend to have a larger polarizability than those with few electrons. Electrons in the outer electronic shells
6
will contribute more to e than do electrons in the inner shells, because the former are not so strongly
bound to the nucleus as the latter. Positive ions therefore will
have relatively small polarizabilities compared with the corresponding neutral atoms: for negative ions
the reverse is true.
1.1. (b)Ionic polarization
Ionic polarization is due to the displacement of positive ion and negative ion of a molecule in the
presence of an applied electric field and occurs in ionic crystals. One might suppose that an ionic crystal
would possess polarization even in the absence of an electric field, since each ion pair constitutes an
electric dipole. But this is not so, because the lattice symmetry ensures that these dipoles cancel each
other every where. So, the polarization in ionic crystals arises due to the fact that the ions are displaced
from their equilibrium positions by the force of the applied electric field.
Consider an ionic compound composed of positive and negative ions separated by inter atomic
distance, ro, then the dipole moment is ‘ero’ in the absence of applied field. When the field E0 is applied to
the molecule, the positive ion is displaced in the direction of field and negative ion is displaced in
opposite direction until ionic bonding forces stop the process.
Due to the ionic displacement the resultant dipole moment increases and is given by
p = e (x1 + x2)
where x1 is the shift of positive ion and x2 is the shift of negative ion with respect to their equilibrium
position.
Due to the application of static electric field E0, the force produced may be taken as F newtons
and the restoring force on positive ion is 1x1 and the restoring force on negative ion is 2x2. Here 1
and 2 are restoring force constants which depend upon the mass of ion and angular frequency of the
molecule in which ions are present.
Therefore, under equilibrium
F = 1x1 = 2x2
F eE 0
x1 = 
1 m 02
where m is the mass of the positive ion and
F = eE0 and 1 = m02
Similarly, for negative ion
eE 0
x2 =
M 02
where M is the mass of negative ion.
7
eE 0  1 1 
Therefore, (x1+x2) = 2 
 
0  m M 
dipole moment
e 2 E0  1 1 
p = e(x1+x2) =   
0  m M 
2
Ionic polarizability
p e2  1 1 
i = = 2   
E0 0  m M 
Thus ionic polarizability iis inversely proportional to the square of the natural frequency of the ionic
molecule and to its reduced mass where reduced mass
1
1 1 
  
m M 
1.1.( c) OrientationalPolarization
Orientation polarization is due to the alignment of dipoles of polar molecules in the presence of applied
electric field. Polar molecules have permanent dipole moments even in the absence of an electric field.
These polar molecular dipoles are randomly distributed in space in the absence of an electric field and
hence the net dipole moment of the dielectric is zero. But when dielectric is kept under electric field, the
field produces a torque in individual dipoles and there is a tendency for the field to align dipole with the
field and a net dipole moment per unit volume is originated in the dielectric. If the field is strong enough,
the dipoles may completely be aligned along the field direction. The polarization due to the orientation,
i.e, orientational polarizability ‘  o’. Fig. 4.Represents the orientational polarization of dipoles.
E=0 E
(a) (b)
dipoles in the dipoles in the
absence of ‘E’ presence of ‘E’
Fig. 4.Orientational polarization.
8
Consider for example, a gas containing a large number of identical molecules, each with a permanent
dipole moment p. Without an external field, the dipoles will be oriented at random and the gas as a whole
will have no resulting dipole moment. An external field E will exert a torque on each dipole and will tend
to orient the dipoles in the direction of the field. In Fig. 5torque applied by a field on a dipole.
+e
eE

E
eE
-e
Fig. 5Torque applied by a field on a dipole.
On the other hand, the thermal motion of the dipoles will counteract this ordering influence of the
external field. Therefore, an equilibrium state will reach in which different dipoles will make zero to 
radian angles with the field direction, producing a net polarization in the direction of the field. It is this
polarization that we are going to calculate.
Let us define the potential energy of a dipole making a 900 angle with the external field as zero.
The potential energy corresponding to an angle  between p and E is equal to
–p E cos = p.E
According to statistical mechanics, the probability for a dipole to make an angle between  and
+d with the electric field is then proportional to
2 sin  d exp[(pE cos )/kT]
where 2 sin  d is the solid angle between  and +d. The number of dipoles having their orientation
between  and +d is also proportional to this probability. Now a dipole of moment p making an angle
 with the field direction contributes to the polarization a component p cos . Hence the contribution
made by the above number of dipoles is
p cos . 2 sin  d exp[(pE cos )/kT]
9
and the average contribution per dipole p is given by


 p cos 2 sin  d exp ( pE cos ) / kT ]

p = 0

 2 sin  d exp[( pE cos ) / kT ]

0
( = 0 corresponds to parallel alignment and  =  corresponds to anti parallel alignment of

dipoles).
Dividing numerator and denominator by 2 and letting
pE
a= , x = a cos , dx = -a sin d,
kT
a
p  x e x dx
p = a
a
a  e x dx
a
p e a  e a 1 1
= a  = coth a - = L(a)
p e  e a a a
The function L(a) is called the Langevin function, since this first derived by Langevin in connection with
the theory of paramagnetism. As a increases, the function continues to increase, approaching the
saturation value unity as a   . This situation corresponds to complete alignment of the dipoles in the
field direction.
10
1.8 (1/3)a
1.6
1.4
1.2
L(a)
0.8
0.6
0.4
0.2
0
0 1 2 3 4 5 6 7 8
a=mE/kT
Fig. 1.7. The Langevin function L(a). For a<<1, the slope is 1/3
Fig. 6. Langevin function L(a) verses a
The above Fig. 6 presents the as long as the field strength is not too high and the temperature is not
pE
too low, the situation may be strongly simplified by making the approximation a<<1 or <<kT.
kT
Under these circumstances the Langevin function L(a) = a/3,
p a pE
= L(a) = =
p 3 3kT
p2
p E
3kT
Hence, orientational or dipolar polarizability
p p2
o = 
E 3kT
Hence, orientational polarizability o decreases with temperature. Since higher is the temperature, greater
is the thermal agitation and lower is ‘  o’
11
A large number of molecules have polarizability, yet not all the molecules. The deciding factor
for its existence is simply whether or not the molecules have a permanent moment. The existence of a
permanent moment is purely a matter of molecular geometry. For example, CO 2 has no permanent
moment at all, because its atoms are in line. On the other hand different geometry of H 2O molecule gives
p=1.87 Debye units to it.
Actually applicable to liquids and gases, because only in these substances the molecular dipole
moment may rotate as continuously and freely as has been assumed in its derivation. In solids, a dipole
may hop back and forth between certain discrete orientations in a manner which depends on the
temperature and the electric field.
12
Chapter II Dielectric properties of alternating electric fields
ALTERNATING FIELDS OF DIELECTRIC MATERIALS
2.1 Introduction
We now take up the study of the behaviour of dielectrics in alternating electric fields. Here
again we make use of the same basic atomic models used earlier and study the behaviour of this model
in alternating electric field. This study reveals that the dielectric constant under these conditions is a
complex quantity. The imaginary part of this complex dielectric constant determines the dielectric losses
of the material.
In the macroscopic theory of isotropic dielectrics under static fields, the electric flux density D is
proportional to the electric filed intensity E, so the D =  E, where  is a constant defined as the electric
permittivity and is a property of the dielectric.
When a dielectric material is subjected to an alternating field the orientation of the dipoles, and
hence the polarization, will tend to reverse every time the polarity of the field changes. As long as the
frequency remains low (<106 c/s) the polarization follows the alternations of the field without any
significant lag and the permittivity is independent of the frequency and has the same magnitude as in
static field. When the frequency is increased the dipoles will no longer be able to rotate sufficiently
rapidly so that their oscillations will begin to lag behind those of the field. As the frequency is further
raised the permanent dipoles, if present in the medium, will be completely unable to follow the field
and the contribution to the static permittivity from this molecular process, the orientation polarization
ceases; this usually occurs in the radio frequency range (106-1011 Hz) of the electromagnetic spectrum.
At still higher frequencies, usually in the infra-red (1011-1014 Hz) the relatively heavy positive and
negative ions cannot follow the field variations so that the contribution to the permittivity from the
atomic or ionic polarization ceases and only the electronic polarization remains.
The above effects lead to fall in the permittivity of a dielectric material with increasing
frequency, a phenomenon which is usually referred to as anomalous dielectric dispersion.
Dispersion arising during the transition from full atomic polarization at radio frequencies to
negligible atomic polarization at optical frequency is usually referred to as resonance absorption.
Dispersion arising during the transition from full orientational polarization at zero or low
frequencies to negligible orientational polarization at high radio frequencies is referred to as dielectric
relaxation.
13
It should be possible to explain the frequency dependence of the dielectric constant

directly in terms of the electronic structure. It is known that the refractive index varies with the
wavelength of light in the optical region the phenomenon being known as dispersion. Dispersion
can be explained on the basis of classical theory which assumes that atom contains electrons
vibrating at certain natural frequencies characteristic of the atom and that the application of an
alternating field sets such electrons into forced vibration. Since the molecules in a dielectric are
represented as dipoles on bound charges, there must be equal number of positive charges and
negative charges because the dielectric is a neutral medium. When an electromagnetic wave
impinges on this bound charge, it is caused to oscillate and therefore to radiate. If the frequency
of the wave is not equal to the natural frequency of the bound charge the forced oscillation will
have small amplitude and the radiation is very weak. This corresponds to molecular scattering. If
the frequency of the wave is equal to the natural frequency of the bound charge, there is
resonance and a much larger energy form the wave goes into the charge. In solid, liquid or gas at
high pressure there is strong intermolecular action and friction type forces cause heavy damping
with the result that the dipole energy is quickly dissipated. This corresponds to true absorption.
In a gas at low pressure there is no damping and the dipole radiate strongly. This is resonance
radiation. The absorption of an electromagnetic wave by a conducting medium is easily
explained because the conduction has a large number of free electrons. When the wave arrives its
energy makes the charge move. The moving charge constitutes current and the usual dissipation
of energy by the current explains the absorption of energy.
At optical frequencies the permittivity is almost entirely due to the electronic polarization. To
determine the dependence of the electronic polarizability on the frequency of the applied field we shall
use the classical model of an electron elastically bound to the atom.
2.2 The complex dielectric constant and Dielectric Losses

When a dielectric is kept between a capacitor plates is subjected to an alternating field the
polarization P also varies periodically with time and so does the displacement D. In general however P
and D may lag behind in phase relative to E so that for example if
E = Eo cost
we have
D = D0 cos(t-)
14
=D0 cos  cos t + D0 sin  sin t
= D1 cost + D2 sin t
where  is the phase angle,
D1 = D0 cos  and D2 = D0 sin .
For most dielectric Do is proportional to Eo but the ratio (Do/Eo) is generally frequency dependent. To
describe this situation one may thus introduce two frequency dependent dielectric constants,
D1 Do
1   cos 
Eo Eo
D2 D0
11=  sin 
E0 E0
It is frequently convenient to sum these two constants into a single complex dielectric constant,
*= 1 - i11
Thus D = * E0 eit = * E0 (cos t + i sin t)
Also we see that
 ''
tan  
'
Because both1and 11are frequency dependent the phase angle  is also frequency dependent. We
shall now show that the energy dissipated in the dielectric in form of heat is proportional to11.
d ( D)
The current density in the capacitor is equal to .
dt
d 1 dD
Thus J= ( ) 
dt 4 dt

=  D1 sin t  D2 cos t 
4
The energy dissipated or absorbed per second in the dielectric is given by
2
 

W  
 4 
 JEdt
0
15
2
   
  
4  0
W   D sin t  D cos t E cos tdt

1 2 o
 
2
   2  
  E0 D1 sin t cos tdt   
W 0 0 o 2 
2
E D cos tdt
4  
 
The value of integral containing D1 is equal to zero and we are left with
  2 11
W ( ) D2 E o  E0 
8 8
the amount of energy absorbed is proportional to sinsince11 = (D0/E0) sin . The energy so
dissipated in the dielectric medium is referred to as the dielectric loss. For this reason sin is called the
loss factor and  is the loss angle (but it is customary) to call tan as the loss factor; this is correct only
for small values of  because tan sin. The dielectric loss at low frequencies is mainly due to d.c.
resistivity. But at high frequencies the dielectric loss is mostly due to dipole rotations or to ionic
transitions from the lower energy states to higher energy states. Because of the upward transition the
energy is absorbed from the applied field. The losses associated with ions, the frequency of which fall in
the infrared region, are usually referred to as optical infrared absorption. Similarly, the losses in the
optical region, associated with the electrons, are referred to as optical absorption.
2.3 Dielectric Losses and Relaxation time
Let us consider a dielectric, for which the total polarization P s in a static field is determined by
three contributions,
Ps =Pe + Pi + Po
In general, when such a substance is suddenly exposed to an external static field, a certain length of
time is required for P to be built up to its final value. In the present section it will be assumed that the
values of Pe and Pi are attained instantaneously, i.e., we shall be concerned with frequencies appreciably
smaller than infrared frequencies. The time required for orientational polarization, Po to reach its static
value may vary between days and 10-12 second, depending on temperature, chemical constitution of the
material, and its physical state is called relaxation time.
To begin with we shall give a phenomenological description of the transient effects based on the
assumption that a relaxation time can be defined; we can then proceed to consider the case of an
16
alternating field. Let Pos denote the saturation value of Po as function of the time after the field has been
switched on is given by
Po(t) = Pos(1- e-t/)
where  is the relaxation time.
dPo/dt = (1/) [ Pos – Po (t)]
For the decay occurring after the field has been switched off, this leads to a well–known proportionality
with e-t/. In the case of an alternating field E = E0 eit, equation may employed if we make the following
change: Pos must be replaced by a function of time Pos(t) representing the saturation value which would
be obtained in static field equal to the instantaneous value E(t). Hence for alternating fields we shall
employ the differential equation
dPo/dt =(1/) [[ Pos(t) – Po (t)]
Now, our final goal is to express the real and imaginary parts of the dielectric constant in terms of the
frequency  and the relaxation time. For this purpose we shall define the “instantaneous” dielectric
constant ei by
Pe + Pi = (ei – 1)/4 E
We may then write
Pos = Ps – (Pe + Pi ) = (s - ei )/4 E
Where s is the static dielectric constant and ei is the dielectric constant arising due to electronic and
ionic polarization. Substitution of Pos
dPo/dt = (1/)[ (s - ei )/4 E0 eit- Po ]
Solving this equation, we obtain
Po(t) = Ce-t/ + 1/4 (s - ei)/(1+i) E0 eit
The first term represents a transient. The total polarization is now also a function of time and is given
by P(t) = Pe + Pi + Po(t). Hence, for the displacement one obtains
D(t)= *E(t) = E(t) + 4 P(t)
where * is the complex dielectric constant. From the last two equations and from the definition * = 1 –
i 11the following expressions result:
1() = ei + (s - ei)/(1+i22)
17
11() = (s - ei) /(1+i22)
These equations are frequently referred to as the Debye’s equations is shown in Fig. 1..
'
s

'' ea
Fig. 1 Debye curves for ’ and ’’ as function

of frequency for a dielectric with a single
relaxation time
loss, which is proportional to 11, exhibits a maximum for  = 1, i.e., for an angular frequency equal to
1/. Also, for frequencies appreciably less than 1/, the real part of the dielectric constant 1 become
equal to the static dielectric constant. In this frequency range, therefore, the losses vanish and the
dipoles contribute their full share to the polarization. On the other hand, for frequencies larger than 1/,
the dipoles are no longer able to follow the field variations and the dielectric constant 1approaches ei.
Note that for this type of mechanism the relaxation time decreases with increasing temperature
as so does the saturation polarization. It is of interest to observe that if the quantities 1 and 11are
measured at a constant frequency but at different temperatures, the curves as indicated in Fig. 2 may
be expected to result.
'

''
 T
Fig.2 The dielectric constant as a function of

temperature at a given frequency, as
predicted from the model discussed in the
text 18
2.4 The classical theory of electronic polarization and optical absorption
The concept of the static polarizability due to elastic displacements of electrons and ions was
introduced. In the present section the classical theory of this phenomenon in alternating fields will be
discussed. We have seen that restoring force determining the displacement is in first approximation
proportional to the displacement itself. The discussion is therefore based on the model of a
displacement itself. The discussions is therefore based on the model of an elastically bound particle of
charge e and mass m in an alternating field E0 eit may be written
d 2x dx
m 2
 m  m 02 x = e E0eit
dt xt
where 0 is the natural angular frequency of the particle; 0= (f/m)1/2 where f is the restoring force
constant ; the second term on the left – hand side is a damping term, which results from the fact that
the particle emits radiation as a consequence of its acceleration and  is the damping factor. The
solution for this forced damped vibration is
e E0 e it
x(t) = . 2
m  0   2  i
We first of all note that in a static field, for =0, this reduces simply to
x = eE0/m02 or s = ex/E0 = e2/m02 for  = 0
Where s is static polarizability associated with the elastically bound particle. If we take for e and m the
electronic charge and mass, this expression would correspond to the contribution of a particular
electron to the electron polarizability. The electronic polarizabilities are of the order of 10-24 cm3; this
gives a natural frequency 0 = 0/21015 per second. Thus, even for frequencies corresponding to the
visible spectrum, the electronic polarizability may be considered constant. If e and m refer to an ion, the
natural frequencies are of the order of 1013 per second, corresponding to the infrared part of the
spectrum.
The electronic polarizability is therefore
ex e 2 1
αe= =
E m  0   2  i
2
The complex dielectric constant is then given by
19
4 Ne 2 1
() = 1+ . 2
m  0   2  i
where N is the number of electrons per unit volume. This follows by using P = Nex and  = 1+4P/E.
Now, from the definition of the complex dielectric constant () = 1() – i11()
One finds
4 Ne 2  0  2 2
 ()= 1+
1
( 0   2 ) 2    2
2 2
m
4 Ne 2 
11() =
( 02   2 ) 2    2
2
m
It may be noted that 1() gives us the value of the dielectric constant and from11()we get thepower
dissipated and hence the damping loss. The variation of (1 – 1) and 11these with frequency is shown in
Fig .3.
 '

''

 

Fig.3 Behaviour of o’ and o’’ as function of
frequency in the vicinity of the resonance
frequency o
maximum at = 0. The meaning of this maximum is that the material absorbs energy at the natural
frequency; this type of absorption is called resonance absorption. In the absorption region, the
dielectric constant 1 depends on frequency and one speaks in this connection of dispersion. The region
for which 1 decreasing with frequency is referred to as the region of anomalous dispersion.
20
2.5.Total Polarizability
Let us now discuss the total polarizability α = αo +αi +αe. It has been found that the total polarizability of
a dielectric substance shows marked difference in behavior when studied as a function of frequency.
To summarize the frequency-dependence of the polarizability we have represented, in Fig.4, α() for a
dipolar substance. It is clear that as we go from the static to the optical region, the polarizabilityα
decreases by a substantial amount. Speaking in terms of dielectric constant, the dielectric constant of
water, for example is 81 at zero frequency while it is only 1.8 at optical frequencies. Moreover, the
decrease in polarizability α is not uniform –remarkable decrease occurs only in the microwave, infrared
and ultra-violet regions.

α (real part)


αe + αi + αo

 αe + αi
αe


Micro Infrared Ultra
waves violet
Fig. .4 Variation of total polarizability as a function of frequency.
The behaviour of polarizability can be understood from the various possesses and from the
concept of the relaxation time for each process. When the frequency of the applied field is much
greater than the inverse of the relaxation time for a particular polarization process, that particular
polarization process fails and so it does not contribute to polarizability. Thus, the decrease of total
polarizability with increase in frequency is due to the disappearance of αo,αi and αe successively.
2.6 Measurement of Dielectric constant
21
Dielectric constant of a given substance is usually measured by comparing the capacity C d of a
Cd
condenser filled with the substance and the capacity C 0 of a the empty condenser . The ratio = , is
C0
the dielectric constant. The capacities Cd and C0 may be measured by resonance method as shown in
the Fig .5.
L
V C Specimen
Cs
Oscillator
Fig. 5 Principle of the resonance method for measuring Co and Cd.
In the figure, Cs is a calibrated variable condenser and C is the condenser in which the given substance
which is taken in the form of a thin disc may be placed. By varying Cs so keep the resonance frequency
1
o =
[ L(C s  C )]
constant when C is empdty and then filled, we may determine C0 and Cd, and hence . The voltmeter V
measures the response of the resonant circuit.
This method is generally used to measure the dielectric constant up to frequencies 100 Mhz. At
the microwave region ( 103 to 105 Mhz) the frequencies are so high that the dimensions of the
apparatus are comparable with or greater than the wavelength, and the specimen hen can no longer be
treated as if it were in quasi-static fields. Rather, it has to be treated as a medium for the propagation
of electromagnetic waves. Here we may measure the dielectric constant of the specimen by measusring
the wavelength of the microwave radiation in the specimen and using the relation
22
vvacuum
= (µ)1/2
 specimen
where µ is the permeability; for non-magnetic materials,µ  1. For optical and infrared frequencies, 
can be measured by measuring the refractive index, n as
n2 = µ 
2.7Outline of the lesson
When a dielectric material is subjected to an alternating field the orientation of the dipoles
alter in accordance with the field changes. At higher frequencies dipoles will no longer be able to rotate
sufficiently rapidly and unable to follow the field and the permittivity of the material decreases. The
average time taken by the dipoles to orient in the field direction is known as relaxation time.
When a dielectric is subjected to an alternating field, the polarization and displacement vector
also vary periodically with time and this gives rise to complex dielectric constant. Dielectric constant
depends on the frequency of the applied electric field.When a dielectric is subjected to alternating field,
the electrical energy is absorbed by the material and dissipated in the form of heat. This dissipation of
energy is called dielectric loss. Debye’s equations relating dielectric loss and relaxation time are
1() = ei + (s - ei)/(1+i22)
11() = (s - ei) /(1+i22)
The losses associated with ions, the frequency of which fall in the infrared region, are called as
opticalinfrared absorptionand the losses in the optical region, associated with the electrons, are
referred to as optical absorption.
23
Chapter III Dielectric properties of static electric fields
STATIC ELECTRIC FIELDS OF DIELECTRIC MATERIALS
3.1 Introduction
In this chapter how the internal field influences the dielectric constant is described in detail. The
Classius –Mossotti relation that connects dielectric constant with the polarizabilities is also derived
3.2Static Dielectric Constant of Gases

We are now in a position to give an atomic interpretation of static dielectric constant of a gas. It
will be assumed that the number of molecules per unit volume is small enough so that the interaction
between them may be neglected. In that case, the field acting at the location of a particular molecule is to
a good approximation equal to the applied field E. Suppose the gas contains N molecules per unit
volume; the properties of the molecules will be characterized by an electronic polarizability e, an ionic
polarizability i, and a permanent dipole moment p. From the discussion in the preceding two sections it
follows that, as a result of the external field E, there will exist a resulting dipole moment per unit volume:
P= N(e + i + p2/3kT)E
Note that only the permanent dipole moment gives a temperature dependent contribution, because
e and i are essentially independent of T. If the gas fills the space between two capacitor plates of area A
and separation d, the total dipole moment between the plates will be equal to
M = PAd
This simple relation shows immediately that the same total dipole moment would be obtained by
assuming that the dielectric acquires an induced surface charge density P at the boundaries
facing the capacitor plates, as discussed in section 1.1.1. Hence the quantity P introduced in
moment per unit volume is identical with the quantity P introduced in section 1.1.1, where it
represented the induced surface charge density at the dielectric-plate interface. The Debye
formula for the static dielectric constant of gas.
E
P = ( - 1) = N(e + i + p2/3kT)E
4
(-1) = 4P/E = 4N(e + i + p2/3kT)
As an example of an application of this formula, the temperature dependence of some organic substances
in the gaseous state. The values of ( – 1) versus the reciprocal of absolute temperature have been plotted,
leading to straight lines. From the slope of the lines and knowledge of the number of molecules per unit
volume, the dipole moment p may be obtained. Also, form the extrapolated intercept of the lines with the
ordinate, one can calculate (e+i). The determination of dipole moments has contributed a great deal to
24
our knowledge of molecular structure. For example, CCl4 and CH4, do not possess permanent dipole
moments (indicated by zero slope), in agreement with the symmetric structure of these molecules.
Similarly, the fact that H2O has dipole moment of 1.84 Debye units, whereas CO2 has no dipole moment,
indicates that CO2 molecule has a linear structure, whereas in H2O the two OH bonds must make an angle
different from 1800 with each other.
3.3Internal Field or Local Field

In solids a molecule or atom experiences not only the external field, but the fields produced by
the dipoles as well. As a result of the long range of Coulomb forces, the later contribution
cannot be neglected. This resultant field is called the local field, and is responsible for polarizing
individual molecules or atoms of solids.
To calculate the local field, we follow the method suggested by Lorentz. According to
this method, we select a small spherical region from the dielectric with the atom for which the
local field must be calculated at the centre. The radius of the sphere is chosen large enough to
consider the region outside the sphere as a continuum whiles the region inside the
s
phere
as
the
actua
l
struct
ure
of
the
subst
ance.
Fig.1 Temperature variation of the static dielectric constant of some vapours.
We
supp
ose that, placing it in a uniform electric field between two oppositely charged parallel plates has
uniformly polarized the given dielectric.
25
+ - + -
.. .. ..
+ - - + -
+
-
+
+ +
...
-  - -
+
+ -
+ - + -
+ -
+
+ - -
Fig. 2 Illustrating the calculation of the internal field as described in the text
Now, since the part of the dielectric external to the sphere may be replaced by a system
of charges induced at the spherical surface as shown in Fig.2, the electric field at the center of
the sphere may be written as
Eloc = E0 + Ep + Es + Em
Here E0 is the primary electric field due to the charge on the plates, E p is the field due to the
polarization charges at the plate-dielectric interface, Es is the field due to the charges induced at
the spherical surface and Em due to all the dipoles of the atoms inside the spherical region.
Now we know that E0 + Ep = E, the macroscopic electric field inside dielectric. Hence,
Eloc = E + Es + Em
Further, if we are considering crystals of high symmetry (such as cubic crystals) E m = 0. This is
because Em is due to all the dipoles inside the spherical surface, and in such crystals these are
randomly distributed in position.
We may then write
Eloc = E + Es
As the assumption Em = 0 is not rue for them. To determine the E s we proceed as follows:
26
E
P cos 
+ -
+ -
+ -
r
d -
+

+ . . -
P
+ -
+ -
+ -
+ -
Fig. 3 Enlarged view of the sphere.
Fig.3represents enlarged view of the sphere shown in Fig.3. The charge element on a
surface element dS of the sphere is equal to the normal component of the polarization times the
surface element, that is, -P cos dS. According to Coulomb’s law, this charge element produces
a force, given by
qP cos  dS
dF = q1q2/r2 = 
r2
acting on a test charge q assumed at the centre of the sphere in this direction of r. Hence, the
field dE, at the centre due to this charge element is
P cos  dS
dEs = dF/q = 
r2
Now resolving dEs into components parallel and perpendicular to the direction of P, we can see a
perpendicular component will be cancelled due to an equal contribution from another
symmetrically situated surface element. Thus only the component of dE s along the direction of P
will contribute the entire surface.
27
P cos 2  dS
Thus, Es =  r2
Now the appropriate surface element dS in this case is the ring shown in Fig.3 so that
dS = 2 r sin r d = 2 r2 sin d, and the limits of integration with respect to  are from 0 to .
Thus,

P cos 2 
Es = 
0 r2
2 r2 sin d,

= 2 P  cos2 sin d,
0
This integral can be evaluated directly by making the substitution

z = cos and dz = -sin d,
so that
1
1
z3 
Es = -2 P  2
z dz = -2 P  
1  3 1
4 P
=
3
4 P
Eloc = E +
3
This equation is called Lorenz relation. This shows that E loc is indeed different from E, as it is expected.
The former field is larger than latter, so the molecules are more effectively polarized.
Substituting value of P
 2
Eloc = .
3
This field is referred as Lorentz field.
The assumption Em = 0 is valid for simple cubic lattice. It is also valid for f.c.c. and b.c.c.
lattices and for crystals such as NaCl. It does not hold for all cubic crystals. For example, in barium
titanate, which has cubic symmetry Em does not vanish.
Each type of atom in a given crystal has its own internal field because the environment of the
different atoms is generally different. Thus the internal field at the location of atoms of type 1, 2, etc.
may be written in the form
Eloc1 = E + 1P; Eloc2 = E + 2P, etc

where the ’s are the internal field constants. Only if Em = 0 do we have  = 4/3.
28
3.4The Clausius-Mossotti Relation

Now we are in a position to relate the microscopicand macroscopic quantities defined above.
The dipole moment p of a single atom is proportional to the local field, that is,
p =  Eloc
Where  is the electrical polarizability of the atom. If there are different types of atoms, the
polarizabilities are additive and the total polarization of an insulator containing N types is
 
N N
P= i 1
ni i Eloc = Eloc i 1
ni i
Where ni is the number of i atoms per unit volume having polarizabilities i and acted on by local
field Eloc .
4P

N
P = (E + ) i 1
ni i
3
or, after rearranging terms
N
P
n  i i
 i
E  4  N
1    n i  i
 3  i
P  1

E 4
N
 1 n 
i
i i
=
4  4  N
1   n i  i
 3  i
 1 4
 2
=
3
n 
i
i i
 Na
If all the atoms i are the same, then  n  = n and n =
i
i i
M
, where  =density, Na is Avogadro
number, and M is molecular weight. So, equation (12.14) can be written in this case as
 1 4  N a
 2
=

3 M
M  1 4
=
Na
  2
or
3
The Clausius-Mossotti equation. It can be used to determine the polarizabilities of the atoms if
the dielectric constant is known. Further, the dielectric constants of new materials can be predicted from
29
knowledge of the polarizabilities. This equation thus provides the necessary relation between the
microscopic and macroscopic quantities.
3.5 The static dielectric constant of solids
From the discussions in the preceding sections it is evident that in general the dielectric polarization P
may be considered the sum of three contributions,
P= Pe + Pi + Po
where the subscripts e, i and o refer, respectively, to electric, ionic and orientation polarization. This
provides a basis for the classification of dielectrics into three classes:
(i) Substances for which Pi = Po = 0 so that P =Pe
(ii) Substances for which Po= 0 and P = Pe + Pi
(iii) Substancesfor which all three contributions are different from zero.
Although the calculation of internal field is usually complicated by the fact that the Lorentz
expression does not apply, some remarks may be made about each of these classes in so far as they apply
to solids.
(i) Substances for which the static polarization is entirely due to electronic displacements are
necessarily elements, such as diamond. one obtains from the relation
Pe = NeEloc = ( –1 )E/4
The following expression for the dielectric constant:
 –1 = 4 N e/(1-Ne)
Where N represents the number of atoms per unit volume. In the particular case for which the Lorentz
expression for the internal field  =4/3. The resulting expression is then usually written in the form of
Clausius-Mossotti formula.
( –1)/ ( +2) = (4/3)Ne

It has therefore been applied mainly to gases. For solid elements one would have to vary the
temperature in order to vary N and the possible range of N values is of course very limited.
It may be noted that for this class of substances under consideration, the dielectric constant is
equal to the square of the index of refraction,  =n2. The reason is, that e is constant even for frequencies
in the visible spectrum. This relationship has been confirmed experimentally for diamond and the
dielectric constant of diamond is 5.680.03.
(ii) In general, solids containing more than one type of atom, but no permanent dipoles, exhibit
electronic as well as atomic or ionic polarization. Of particular interest in this respect are the crystals,
such as the alkali halides. Consider, for example, a NaCl crystal in an external static field E. Apart from
the electronic displacements in the ions relative to the nuclei, the positive ion lattice will tend to move as
a whole relative to the negative ion lattice. Consequently, a considerable contribution to the total
30
polarization may be expected to arise from the ionic displacements (P i). That this is indeed the case,
becomes apparent from a comparison of the values of the static dielectric constant defined by
Pe + Pa = ( – 1) E/4
and the “high-frequency dielectric constant” 0 defined by
Pe = (0 – 1) E/4
Hence Pi is about two or three times Pe in these compounds. In non-ionic compounds, on the
other hand, Pi is usually a relatively small fraction of P e.
The observed difference between the static and high- frequency dielectric constants is because of the
difficulties involved in calculating quantitatively the internal field.
It may be noted that the force constant and the masses of the positive and negative ions determine
the infrared frequency associated with the lattice vibrations. It is therefore possible to express the
difference (s - 0) in terms of infrared absorption frequency of the lattice.
(iii) In substances composed of molecules which bear permanent electric dipole moments, the
total polarization is made up of three contributions,
P =Pe + Pi +Po
Were Po corresponds to the dipolar contribution. There exists no general quantitative
theory for dipolar solids because first of all the same difficulties arises in evaluating the internal fields as
in class (ii), and further more, the dipoles in such solids may not able to rotate at all or only to some
extent. The discussion must therefore be limited to some qualitative remarks. As an example of a dipolar
solid which behaves in a relatively manner, the dielectric constant measured as function of temperature
for C6H5NO2(nitrobenzene). It is observed that at the melting point there is large increase in dielectric
constant. This is interpreted as an indication that in the solid the dipoles cannot rotate freely and Po is
essentially zero; in the liquid, alignment of the dipoles in the field direction is possible, so that the
increase in  is determined by the now freely rotating dipoles. The subsequent slow decrease in  is a
consequence of the thermal motion of particles. In other cases, the behaviour may be more complicated,
in which  versus T has been plotted for H2s. the melting point of H2S is 187.70K. in this case, the dipoles
are apparently “frozen in” at temperature below 103.50K;at this temperature the structure changes in such
a manner that the dipolar groups become mobile; as the temperature is further increased, the dielectric
constant decreases as a result of increased thermal motion. The other changes evidently affect essentially
the density of the material, i.e., N is reduced at these transition points.
31
Chapter IV Ferroelectric materials
FERROELECTRIC MATERIALS
4.1 Introduction
When the centre of a positive charge does not coincide with the centre of
negative charge in a primitive cell, the primitive cell possesses an electric dipole
moment even in the absence of applied electric field. Thus the crystal as a whole has a
polarization implying that it is spontaneously polarized The shifting of positive charge
from the centre of negative charge is exhibited in the lack of centre of symmetry in the
crystal. Out of 32 crystal point groups, 21 point groups do not have a centre of
symmetry. Except one point group, which is highly symmetric, the rest 20 point groups
represent an extremely useful class of materials, known as piezoelectrics.
Piezoelectrics: Piezoelectric crystals show electric polarization on being
externally strained and conversely, show deformation when placed under the influence
of an applied electric field. If the crystal belongs to any one of the above 20 point
groups, it can be predicted that the crystal would be piezoelectric. Ammonium
phosphate, quartz, PZT (Lead Zirconate Titanate) are some examples of piezoelectric
crystals.
Pyroelectrics: Among the class of 20 crystal point groups which lack centre of
symmetry, 10 crystal point groups are spontaneously polarized. These spontaneously
polarized dielectric crystals are called pyroelectric crystals. The polarization in
pyroelectric crystals is usually masked by surface charges that accumulate on the
surface from the atmosphere and subsequently neutralize the layers of ions. But, when
32
the temperature of the crystal is altered, the masking is no longer complete as the
polarization changes because of thermal expansion or contraction of the crystal. Owing
to the thermal effect on polarization, these crystals are named pyroelectric (pyro means
fire). The thermal effect accompanying deformation thus supports the piezoelectric
property of the crystals. This only confirms that all pyroelectric crystals are piezoelectric,
though converse is not true.
While maintaining the crystalline properties, the symmetry operations of a
pyroelectric crystal must preserve the direction of polarization P. This imposes severe
restrictions on the point group symmetries as a result of which only 10 point groups are
found to meet the conditions of pyroelectric crystals. The rotation is allowed about only
one axis that is parallel to P and there cannot exist mirror planes perpendicular to this
axis. The structural scrutiny of crystal groups reveals that only the following point
groups meet the restrictions of pyroelectric crystals:
Cn, Cnv (n =2,3,4,6), C1 and C1h
Thus the pyroelectric property too, like piezoelectricity, is solely determined by the
symmetry properties of crystals.
Ferroelectrics: Ferroelectric crystals have additional property that the
polarization in them can be changed and even reversed by an external electric field. On
the other hand, this is not possible in pyroelectrics even with the maximum electric field
that may be applied without causing electrical breakdown. The additional feature of
ferroelectrics that distinguishes them as a special class of pyroelectrics does not follow
33
from the characteristics of crystal structure. It is established only on the basis of
dielectric measurement.
Furthermore, the additional feature of ferroelectrics mentioned above converts
the usual linear relationship between polarization and applied electric field into a
hysteresis loop. Since the dielectric behaviour of these materials is in many respects
analogous to the magnetic behaviour of ferromagnetic materials, they are called
ferroelectric solids, or ferroelectrics. The ferroelectric behaviour is observed only below
a certain temperature, called the Curie point, Tc. A ferroelectric is spontaneously
polarized, i.e., it is polarized in the absence of external field; the direction of the
spontaneous polarization may be altered under influence of an applied electric field. In
general, the direction of spontaneous polarization is not the same throughout a
macroscopic crystal. Rather, the crystal consists of a number of domains; within each
domain the polarization has a specific direction, but this direction varies from one
domain to another.
4.2. Representative crystal types of ferroelectrics
In general the ferroelectric crystals may be broadly classified into four representative
groups such as i) Ilmenites and Perovskites, ii) KDP type iii) TGS type and iv) Rochelle
salt. The table gives the Curie point Tc and the spontaneous polarization Ps for a number
of common ferroelectric crystals. The electric susceptibility  in the Para electric phase
C
is related to temperature by the Curie- Weiss law: =
T  Tc
where C is the Curie constant.
34
The ferroelectric crystals are also distinguished on the basis of oscillatory nature
of the atomic displacements that destroy the ferroelectric dipole order above the Curie
temperature. In the ferroelectric phase of some crystals, the atomic displacements can
be viewed as oscillations about a polar site. In the paraelectric phase these oscillations
take place about a non-polar site. The phase transition that brings about this
transformation in the nature of oscillations is called a Displacive phase transition. These
crystals are accordingly identified as Displacive type. The well-known examples of this
class are ionic crystals with ilmenite and perovskite structures. The GeTe is the simplest
ferroelectric crystal having the ilmenite structure (i.e., NaCl structure) and BaTiO 3 is the
representative crystal of perovskites.
There is another very interesting class of crystals in whose non-ferroelectric state
the potential energy function around certain atomic sites is double-well or multiple-well
shaped. On the transition to the ferroelectric state the atomic displacements about
those sites are executed as oscillations in an ordered subset of the referred potential
wells. It involves an order-disorder type of phase transition. Common examples of these
crystals, classified as order-disorder type, are some
hydrogen bonded solids, namely KDP type crystals. The replacement of hydrogen by
deuterium in KDP type crystals raises the Curie point in an amazing proportion. Though
the increase in the molecular weight is less than 2 percent, the T c rises from 123K to
213K in the deuterated KDP and from 96K to 162K in KD2AsO4.
35
For specific description, Rochelle salt and BaTiO3 are chosen as the two
representative compounds of ferroelectrics whose properties are uniquely different.
1). Rochelle Salt
The first solid which was recognized to exhibit ferroelectric properties is Rochelle salt,
the sodium-potassium salt of tartaric acid; it has the chemical formula NaKC 4H4O6.4H2O.
It was first prepared in 1672 by a pharmacist Seignette who lived in Rochelle. It
represents the tartaric group of salts whose other well known member are lithium
ammonium tartrate and lithium tantalum tartrate. The most noteworthy characteristic
of Rochelle salt is that it is ferroelectric between two temperatures (255K and 296K). On
account of its two transition temperatures, Rochelle salt becomes a special and peculiar
example of ferroelectrics.
The crystal structure of Rochelle salt is somewhat complex. Above 296K and
bellow 255K the structure is orthorhombic (three mutually perpendicular axes a,b,c). It
has a monoclinic symmetry in the ferroelectric phase such that the angle (between the
c- and a- axes) differs from 900 and the spontaneous polarization is along the original
orthorhombic a-axis. Thus Rochelle salt has only one polar axis and two possible
polarization directions (+ and – along the a- axis).
Halblutzel has measured the dielectric constant of Rochelle salt along the three
crystal axes over the whole useful range of temperatures. The Curie-Weiss law applies
above 296 K and below
36
255 K. With the help of the experimental data it is easy to confirm that the two regions
have different values of Curie constants. The dielectric constant measured along the
polar axis a peaks at both transition temperatures, assuming a value as high as 4000.
2). BaTiO3
The BaTiO3 is the most important and most completely investigated
representative of the perovskites type ferroelectrics. In the non-ferroelectric state (i.e.
above 393 K). The Ba2+ ions are positioned at the corners, O2- ions at the centre of the
faces and the Ti4+ ion is located at the centre of the cube. It has an arrangement of
highly polarizable oxygen ions in the form of an octahedron with a small titanium ion at
the centre
The curves clearly indicate that there are three ferroelectric phases of crystal
_____________________________________________________________
Temperature Direction of Ps Crystalline symmetry
_range_______________________________________________________
278-393 K [001] Tetragonal
193-278 K [011] Orthorhombic
< 193 K [111] Rhombohedral
The dipole moment p can alternatively be estimated by multiplying Ps by the unit
cell volume. Treating the unit cell as a simple cube of edge 4 A 0 even in the ferroelectric
state, we get p = 0.3(4 x 10-10)3 = 1.92 x 10-29 C m. Thus we find this value agrees very
37
well with that obtained on the basis of the observed deformation of the unit cell. The
order of magnitude gives a measure of the ferroelectric effect in BaTiO 3.The effect,
however, is fairly large in some other perovskites (e.g. LiNbO 3).
The Ps along the [001] direction, warrants our further attention. This
implies that we must multiply by 2 and 3 to obtain the actual values in the regions
193 K < T < 278 K and T<193K, respectively, because the direction of P s in these regions
is along the [011] and [111] directions, respectively. It is then quite interesting to note
that spontaneous polarization (same as the saturation polarization) remains almost
constant below 300K.
4.3. Theory of the ferroelectric displacive transitions
The theory that gives a good account of transitions in perovskites type crystals merits a
separate treatment on account of having stood the test of vast experimental data.
These crystals generally undergo a displacive transition at the Curie point. We can
follow two approaches for finding interpretation to a displacive transition. One
approach is the polarization catastrophe and the other one is the soft mode approach.
The polarization catastrophe refers to an unusual situation in which the
polarization becomes infinitely large. In this condition the force exerted by the local
electric field is greater than the elastic restoring force. This produces an asymmetric
shift in the positions of positive and negative ions. The shift is, however, limited to a
finite displacement by the anharmonic restoring forces.
38
In the soft mode approach a transverse optical (TO) mode is frozen, i.e. its
frequency vanishes at some point in the Brillouin zone below the Curie temperature.
This TO mode is known as a soft mode. When  T = 0, the crystal becomes unstable
because of the absence of an effective restoring force.
Polarization Catastrophe
The Clausius-Mossotti relation can be rearranged in the form
3( N i i  N e e )
  1
3 0  ( N i i  N e e )
where
Ni and Ne are the density of polarizable ion pairs and electrons, respectively and
αi and αe are the ionic and electronic polarizabilities, respectively.
the dielectric constant becomes infinite , giving the state of polarization catastrophe.
Further,
P = (Niαi + Neαe) Eloc
 P 
= (Niαi + Neαe)  E  
 3 0 
for a cubic crystal ( using Lorentz expression for Eloc).
If E = 0,
 N i i  N e e 
P  1 = 0
 3 0 
39
But, when the polarization catastrophe occurs, the quantity within the brackets equals
zero
This requires that
P0
The order to understand the above situation, let us consider a highly polarizable ionic
crystal of cubic symmetry. Let α be the total polarizability and p the dipole moment of
an ion pair. Let us assume that some transient stray field starts polarizing the ion pairs.
The ion pairs will keep on polarizing until some resistance develops to stop the process.
The resistance that finally stops the process of polarization exists in the form of
anharmonic restoring forces. The dipole moment of a single ion pair with ion separation
x is
 F 
P = q.x = α Eloc =  
 q 
where F is the restoring force that tends to bring the positive and negative ions together
and q is the charge on each ion.
The work required to create N such dipoles in the unit volume of the crystal is
Nq 2 Np 2
E1 = N  F.dx =  x.dx =
 2
P2
=
N 2
On the other hand, the energy density associated with the electrical displacement due
to Eloc is
40
E2 =  Eloc. dP
 P 
=   E   . dP
 3 0 
P2
6 0 
=  E.dP
since E1 is set against E2, the net energy density of a polarized dielectric is
P 2  N 
E2 – E1 =   1   E.dP
2 N  3 0 
This shows that even when E = 0, E2>E1, provided that
Nα 30
The above condition in a general case is written in the form
N j
j j 30
where Nj stands for the density of the j th type of particles (ions/electrons) in the crystal
and αj denotes the polarizability of a single particle of this type.
N j
j j ≡ Ni Ai + Ne αe
the energy of the crystal becomes smaller in the presence of induced dipoles. The
minimum value of N 
j
j j for which the ferroelectricity would occur is 30. In any real
ferroelectric crystal the situation that exactly corresponds to the polarization
41
catastrophe is not found. However, a small deviation in the value of N 

j
j j from 30
changes the value of  by a large amount.
If we express N 
j
j j = 30-3β
with β<<1 we get
1


If we assume that β is a linear function of temperature near the Curie point and given by
T  Tc
β=

 being a constant, then
1

T  Tc
The temperature dependence of  as a given by this relation is in excellent agreement
with the observed behaviour in several perovskite crystals.
Ferroelectricity in perovskite crystal is understood in view of the following remarks
made in respect of barium titanate:
1. The titanium ion motion. The barium ions situated at the cube corners leave
a big void at the centre position. Since titanium ion is smaller than barium
ion, it is unable to fill the void and is free to rattle around in the void.
Because the ionic polarizability is a measure of the ease of displacement, its
value increased.
42
2. The non-cubic symmetry around oxygen ions. Unlike the barium and
titanium ions, the oxygen ions are in the non-cubic environment. An oxygen
ion has only two nearest neighbours in the form of titanium ions. Because of
this reason, Eloc is greater than the value given by Lorentz expression.
A larger value of α predicted under point 1 leads to a smaller value of deformation
energy E1 or the work required to create induced dipoles. Similarly, a large value of E loc
as expected under point 2 implies that the dipolar attraction will be larger. Thus, larger
values of both α and Eloc are favourable to the onset of ferroelectricity.
Soft mode approach
As mentioned earlier, a ferroelectric state can be regarded as a frozen in TO phonon.
According to Lyddane-Sachs-Teller relation (popularly known as LST relation)
 TO
2

 
 LO
2
s
where s is the static dielectric constant ,   is the dielectric constant at optical
frequencies,  TO and  LO are the transverse and longitudinal optical mode frequencies.
Above expression shows that as s increases,  TO decreases; thus, in the case of an
infinitely large s, which happens at the Curie point (Tc),  TO may even be zero. In
practice , s remains finite on approaching Tc. The TO modes in question are called soft
modes. Such TO modes have surprisingly low frequencies. For example, BaTiO 3 has a
soft mode of frequency 12 cm-1 at 297 K which is low for a TO mode.
43
We are not concerned here with LO phonons whose frequency is higher for the same
value of the wave vector. At the transition point Tc when  TO approaches the zero
value, the crystal becomes unstable and anharmonic elastic forces come into play. In
the presence of anharmonic forces,  TO may show a temperature dependence of the
form
 TO
2
 (T – Tc)
On assuming that  TO are temperature dependent, the LST relation
1
 (T – Tc)
s
Experimental results on several perovskite ferroelectrics strongly support that a large
static dielectric constant (s) is associated with a low TO phonon (the soft mode). The
temperature dependence of the energy of a low frequency TO phonon can be directly
compared with that of the inverse dielectric constant, for a KTaO3 crystal. To have a
clear idea, a schematic representation of the temperature dependence of s-1,  TO

2
and
the saturation polarization Ps is shown in Fig. 1
120
100
[h(k)]2 (meV2)
80
60
40
20
44
0 50 100 150 200 250 300

4.4. Thermodynamic theory of the ferroelectric transition
It is of interest to investigate the behaviour of a ferroelectric in the vicinity of its
transition temperature Tc on the basis of thermodynamic arguments. A thermodynamic
theory has the advantage of being independents of any particular atomic model and
thus leads to quite general conclusions. Although such a theory does not provide the
physical mechanism responsible for the ferroelectric properties of a given material, it
does point to certain features one should look for in atomic models.
Consider a crystal which is ferroelectric for temperature T < Tc. Let x denote the
relative displacement of the centres of the positive and negative ions in the crystal
during a particular mode of vibration. If F0 be the free energy of the unpolarized crystal,
the free energy of the polarized crystal F is a function of the even powers of x. That is,
F-Fo = 2 x2 + 4 x4 + 6 x6 + . . . .
The constants  are functions of all other displacements and given by their thermal
average values. They are thus functions of temperature. Since the electric polarization
P is proportional to the displacement x, we have
1 1 1
F – Fo =  2 P 2 +  4 P4 +  6 P6 + . . . .
2 4 6
The constants  are the functions of temperature. The numerical factors are introduced
to facilitate calculations.
Consider first the paraelectric phase of the crystal, i.e., for T>Tc. If a small
electric field E is applied in the absence of any external pressure, the following
thermodynamic relation holds good :
45
dF = - SdT + E dP
where S represents the entropy of the crystal.
For smaller E, P will also be smaller, and hence we retain only the first term in
 F 
E=   =  2P
 P T
The electric susceptibility P in the paraelectric phase is given by
 0  dE 
1
P =   = 0 2
P  dP 
using the Curie-Weiss law, we have
T  Tc
0 2 =
C
or 2 = C1(T-Tc)
where C1 is another constant.
shows that 2 increases linearly with increase in temperature. As a result of this
temperature dependence, 2 varies from positive values to negative values as the
temperature is lowered from above Tc to below Tc.
In the state of thermal equilibrium, the free energy is minimum which requires that
 F 
  =0
 P T
2P + 4P3 + 6P5 + . . . . . .= 0
46
The spontaneous polarization is bound to satisfy and
Ps(2 + 4Ps2 + 6Ps4 + . . . .) = 0
We find that Ps = 0 is always a root of For this solution the free energy has a
 2F 
minimum provided 2 is positive  2  2  . However, if 2, 4 and 6 are all positive
 P 
and higher order terms are neglected, the condition is satisfied only for P s = 0. Thus, Ps
=0 corresponds to the only minimum of the free energy and the paraelectric phase
exists for the positive sign of 2, 4 and 6.
When the temperature is lowered through the transition point, 2 goes from
positive do negative values while passing through 2 = 0 at the transition point. There
are two interesting situations that are identified in terms of the signs of 2, 4 and 6.
These characterize two cases of particular interest namely second-order and first-order
transitions.
Second-order Transitions: If the coefficients 4, 6, . . . . are all positive and the value of
2 varies from positive to negative as the temperature is lowered, the free energy
changes as shown in Fig. 2. Neglecting the terms beyond the second term in are
negligible, we get
2 C1 (Tc  T )
Ps2   =
4 4
Hence Ps is a continuous function of temperature and falls continuously to zero at T = Tc
47
It is useful to examine the spontaneously polarized state in terms of the
frequency of normal modes.
2 or 2 =  i2(k)
i2(k)  (T – Tc)
where  i(k) is the frequency of the normal mode i (a TO mode). The transition takes
place when  i(k)  0. This decrease in the mode frequency is called softening. This
indicates that the harmonic restoring forces are becoming very weak, permitting a large
displacement which is limited solely by anharmonic forces. When i2(k) or 2 is small
and positive then the crystal lattice becomes soft and close to instability. Below T c, 2 is
negative and hence also  i2(k); which implies that the unpolarized lattice is unstable and
the crystal is in the spontaneously polarized ferroelectric state.
The heat capacity is given by
Cv = C12T/4
The heat capacity falls discontinuously to zero at T =Tc (see Fig. 2). But there is no
latent heat at the transition. Such a transition is called a second-order transition.
Cv
T Tc
Fig. 2 Temperature dependence of specific heat showing anomaly at

a second-order phase transition
48
The transitions in Rochelle salt, KH3PO4 and LiTaO3 are some examples of the second-
order transition. The transition to the superconducting state is the most popular
example of this type of transition.
First-order Transitions
We have seen that when 2 is negative and 4 is positive, the transition is of the second-
order type. We now consider a situation where 4 is negative and 6 is positive.
Positive values of 6 are considered to restrain the free energy from going to minus
infinity. 2 varies from positive to negative as the crystal is cooled through the Curie
point.
F
The thermal equilibrium condition, = 0, in the absence of the applied
P
electric field gives
 2 Ps +  4 Ps 3 +  6 Ps 5 = 0
which implies that either Ps = 0, or
 2 +  4 Ps 2 +  6 Ps 4 = 0
At T = Tc, the free energy in the paraelectric state is equal to that in the ferroelectric
state, i.e.,
F0(Tc) = F(Tc)
1 1 1
0= 2Ps2 (Tc) + 4Ps4 (Tc) + 6Ps6(Tc) + ..
2 4 6
49
2 + 4Ps2 (Tc) + 6Ps4 (Tc) = 0
3 4  3  4
Ps2(Tc) = -  
4   6  4  6
And with
3   24 
2 =  
16   6 
3 2
Ps4(Tc) =
6
At the transition point there are two minima of free energy with equal value; one at
Ps(Tc) = 0 in the paraelectric phase and the other for the value of P s(Tc). Thus there is a
jump in the value of Ps at Tc, meaning thereby that the spontaneous polarization (the
order parameter) drops discontinuously to zero at T = Tc when a ferroelectric crystal is
heated slowly. Such transitions are called the first-order transitions. The other
important property of these transitions is that there is a latent heat at the transition. A
well known example of this type of transition is the upper transition in a BaTiO3 crystal.
Free energy as a function of polarisation as the temperature is varied near a first order
phase transition. (b) Fall of the spontaneous polarisation below the transition point Tc in
a first order phase transition.
4.5 Ferroelectric Domains
When a ferroelectric is cooled from the paraelectric phase through the Curie
temperature, the polarized phase may be nucleated at several points in the crystal.
50
These nuclei generally differ in the direction of polarization since there may be several
equivalent crystallographic directions in which the spontaneous polarization can occur.
In the case of BaTiO3, the spontaneous polarization may occur along any one of the
three edges, giving six possible directions for the spontaneous polarization. Thus, as the
nuclei grow through the crystal in the ferroelectric crystal in the ferroelectric phase,
they form several regions or domains differing in their direction of polarization. The
vector sum of these polarizations may not be always big enough to show up
macroscopically.
Polarization is accompanied by some distortion of the unit cell and the domain
walls are consequently in a state of strain; but the dimensional changes are relatively
small. Though the domain walls act as interruptions in the regularity of the crystal, they
are not regarded as grain boundaries between different crystals. A domain wall is
instead, treated as a sub-grain within a single crystal. As soon as a single nucleus of the
polarized phase is formed, the polarized phase begins to grow much faster in the
direction of polarization than in the transverse directions. Because of this reason the
growing domains are usually wedge-shaped. This was revealed by optical birefringence
studies on BaTiO3.
The ferroelectric domains are regarded as the electrical analogues of the
ferromagnetic domains despite the fact that there are some interesting differences in
their origin and growth. When the electric field is applied on a ferroelectric crystal, the
number and size of domains that are polarized in the field increase. As a result of this
51
effect, upon the reversal of the field direction a hysteresis in the P versus E curve is
observed.
4.6 Antiferroelectricity
Similar to ferroelectrics there is another group of solids, which has induced, ordered
electric dipoles below a characteristic temperature but do not show spontaneous bulk
polarization. In these crystals the neighbouring atomic lines are associated with
antiparallel polarization because of which the bulk polarization of the crystal vanishes.
Crystals exhibiting this property are called antiferroelectric crystals and the property is
known as antiferroelectricity. The structural requirement for the ferroelectrics and
antiferroelectric phases being common, a number of well-known antiferroelectric
crystals are found to be isomorphous with some ferroelectrics. For example, ammonium
dihydrogen phosphate (ADP) is isomorphous with potassium dihydrogen phosphate
(KDP).
Perovskite type crystals are known to be susceptible to several types of
deformation with almost equal energy difference between them. In many of them the
coupling through the oxygen octahedral causes adjacent lines of basic cells to be
polarized in opposite directions. Below a certain temperature the resultant deformation
is such that the total energy in the antiparallel arrangement of adjacent lines of dipoles
is lower, when compared separately to that in state of fully parallel arrangement of
dipoles and to that in the state with no induced dipoles. Lead Zirconate (PbZrO3) is a
notable example of these perovskites. It shows to
52
Antiferroelectric phases, one each ferroelectric and paraelectric phases over
different ranges of temperature.
53
Chapter V Piezoelectric Materials
PIEZOELECTRIC MATERILAS
5.1 Introduction to Piezoelectric materials
Piezoelectricity is the ability of certain dielectric crystals to produce a voltage
when subjected to mechanical stress. The word is derived from the Greekpiezein, which
means to squeeze or press. The effect is reversible; piezoelectric crystals, subject to an
externally applied voltage, can change shape by a small amount. The effect is of the
order of nanometres, but nevertheless finds useful applications such as the production
and detection of sound, generation of high voltages, electronic frequency generation,
and ultra fine focusing of optical assemblies.
Pyroelectricity, the ability of certain mineral crystals to generate electrical charge
when heated, was known of as early as the 18th century, and was named by David
Brewster in 1824. In 1880, Pierre Curie and Jacques Curie brothers predicted and
demonstrated piezoelectricity using tinfoil, glue, wire, magnets, and a jeweler's saw.
They showed that crystals of tourmaline, quartz, topaz, canesugar, and Rochelle salt
generate electrical polarization from mechanical stress. Quartz and Rochelle salt
exhibited the most piezoelectricity. Twenty natural crystal classes exhibit direct
piezoelectricity.
54
Piezoelectric materials
In addition to the materials listed above, many other materials exhibit the
piezoelectric effect, including quartz analogue crystals like berlinite (AlPO4) and gallium
orthophosphate (GaPO4), ceramics with perovskite or tungsten-bronze structures
(BaTiO3, KNbO3, LiNbO3, LiTaO3, BiFeO3, NaxWO3, Ba2NaNb5O5, Pb2KNb5O15). Polymer
materials like rubber, wool, hair, wood fiber, and silk exhibit piezoelectricity to some
extent. The polymer polyvinylidene fluoride, (-CH2-CF2-)n, exhibits piezoelectricity
several times larger than quartz. Bone exhibits some piezoelectric properties: it has
been hypothesized that this is part of the mechanism of bone remodelling in response
to stress.
Mechanism of piezoelectricity
In a piezoelectric crystal, the positive and negative electrical charges are
separated, but symmetrically distributed, so that the crystal overall is electrically
neutral.When a stress is applied, this symmetry is disturbed, and the charge asymmetry
generates a voltage. A 1 cm cube of quartz with 500 lbf (2 kN) of correctly applied force
upon it, can produce 12,500 V of electricity.
Piezoelectric materials also show the opposite effect, called converse
piezoelectricity, where application of an electrical field creates mechanical stress
(distortion) in the crystal. Because the charges inside the crystal are separated, the
applied voltage affects different points within the crystal differently, resulting in the
55
distortion.The bending forces generated by converse piezoelectricity are extremely
high, of the order of tens of meganewtons, and usually cannot be constrained. The only
reason the force is usually not noticed is because it causes a displacement of the order
of a few nanometres.
Requirements for a crystal to show piezoelectricbehaviour were discussed in the
previous Lesson. We showed earlier that all ferroelectrics are piezoelectrics and that its
converse is not true. For example, quartz is piezoelectric but it does not possess the
ferroelectric property.
The foremost condition for a crystal to piezoelectric is the absence of the centre
of symmetry. a simple two-dimensional ionic crystal with no centre of symmetry.It is
evident that a compressive force Fdecreases the electric dipole moment(hence the
polarization) and a tensile force F increases the same. This is essentially the piezoelectric
effect. We must appreciate that the displayed crystal could well be a ferroelectric
crystal.
Next we take up another example to show how the symmetry of a non-
centrosymmetric crystal controls firstly the magnitude and direction of polarization
when the crystal is stressed and secondly the crystal dimensions when the crystal is
polarized. Consider a molecule of hypothetical ionic solid which at equilibrium
has three electric dipoles of equal magnitude distributed over 3600at an interval of 1200.
The molecules belong to the point group 3m and its net dipole moment is zero. But if
the molecule together with the crystal is stressed or compressed along a direction
56
parallel or antiparallel to one of the three directions of the dipole moment, a net dipole
moment would appear. Similarly, a molecule may be distorted by an electric field
applied along one of the three arrows. The electric field produces an elongation or
contraction of the crystal along the field direction and a length change of opposite sign
in the lateral direction. An applied field that is perpendicular to one of the three dipole
directions in Fig. 1. finds itself perpendicular to a mirror plane of symmetry and,
therefore, is rendered ineffective in changing the crystal dimensions.
Because of lack of centre of symmetry and complex structure of piezoelectrics,
their electrical behaviour under strain or strain behaviour under an electric field is not
isotropic in nature. Nevertheless, a simple picture of the phenomena can be presented
in a schematic one-dimensional notation by the following equations:
P= d + 0 E :; e =  s + Ed
where P is the polarization, the stress, d the piezoelectric strain constant, 0the
permittivity of free space, E the electrical field,  the dielectric susceptibility, e the
strain and s the elastic compliance constant.
In real crystals, however, the tensile, compressional or shear strains produced by
an electrical field may develop in different directions and depend on the crystal
orientation and the field direction in view of this fact the piezoelectric strain constants,
that form a third rank tensor, are defined as
 ek 
dik=  

E
 j 
57
where i  x,y,z and k  xx,yy,zz, xy, yz, zx
P P
P=0
(a) (b) (c)
Fig. 1 (a) Directions of polarization (b) A vertical tension or a horizontal compression causing
(c) A vertical compression or a horizontal tension causing a net polarization
Depending on the application and the desired behaviour, a crystal is cut so as to
have the parallel faces of the crystal in a specific orientation. An X-cut is defined as a
section cut from the crystal such that the x-axis of the crystal is perpendicular to parallel
crystal faces. In order to obtain certain desirable properties the crystals are sometimes
given oblique cut that is cut at angles is different from 90 degrees with the principal
axes.
5.2 Electrostriction
It is appropriate to discuss a more universal phenomenon of deformation in
crystals that is caused by an applied electrical field. It refers to the deformation in ionic
crystals and the effect is commonly known electrostriction.
Electrostriction is a property of all electrical non-conductors, or dielectrics that
produces a relatively slight change of shape, or mechanical deformation, under the
58
application of an electric field. Reversal of the electric field does not reverse the
direction of the deformation.
In the first approximation the deformation of piezoelectric crystal is proportional
to applied electrical field and the stress induced polarization varies linearly with the
strain produced. But in ionic crystals, which do not have to be necessarily piezoelectrics,
the strain is much smaller and proportional to the square of electrical field. We can
understand the origin of electrostriction by appreciating that dipoles created by the
applied electrical field would interact with each other. The inline dipoles attract each
other with a repulsive poles acting perpendicular to the direction of the polarization.
Let p denote the moment of a dipole and r the separation between two inline
dipoles. The value of the electric field caused by a dipole at its in-line neighbour may be
written as
1 2p
E=
4  0 r 3
The energy of a dipole in the field U(r) and the corresponding attractive force F are
related as
dU ( r )
F=-
dr
and Ur) = - p.E
These relations yield
1 6 p2
F=-
4  0 r 4
Similarly, we can find that the repulsive force is given by
59
1 3p2
F=
4  0 r 4
Since p = αE, the attractive force can be expressed as
1  6 2  2
F=-  E
4  0  r 4 
To a first approximation the strain or deformation u may be assumed to follow
the Hooke’s law and then
F
u=-
k
where k is the usual force constant in the direction of the in-line dipoles. The above
relation, we get
1  6 2  2
U=  E
4  0  k r 4 
Thus, there will occur a compression in the field direction and an extension
perpendicular to the field direction. The above treatment holds for permanent dipoles
as well on account of the effective dipole moment being proportional to the electric
field.
5.3 Applications of piezoelectric crystals
It may be recalled all ferroelectrics are piezoelectrics, though the conversion is
not true.As a result, ferroelectric materials have been frequently used in many
applications that are based on the principle of piezoelectricity. But, because of
importance of properties such
60
as mechanical and thermal strength the use of certain piezoelectric crystals
becomes inevitable.
Piezoelectric crystals are used in numerous ways:
High-voltage sources
Direct piezoelectricity of some substances like quartz, as mentioned above, can
generate thousands of volts (known as high-voltage differentials).
 Probably the best-known application is the electric cigarette lighter: pressing the
button squeezes an piezoelectric crystal, and the high voltage thus produced
ignites the gas as the current jumps over a small spark gap. The portable
electrical sparkers used to light gas grills or stoves work the same way.
 A similar idea being researched by the Defense Advanced Research Projects
Agency (DARPA) in the USA in a project called Energy Harvesting, which includes
an attempt to power battlefield equipment by piezoelectric generators
embedded in soldiers' boots.
 A piezoelectric transformer is a type of AC voltage multiplier. Unlike a
conventional transformer, which uses magnetic coupling between input and
output, the piezoelectric transformer uses acoustic coupling. An input voltage is
applied across a short length of a bar of piezoceramic material such as PZT,
creating an alternating stress in the bar by the inverse piezoelectric effect and
61
causing the whole bar to vibrate. The vibration frequency is chosen to be the
resonant frequency of the block, typically in the 100 kilohertz to 1 megahertz
range. A higher output voltage is then generated across another section of the
bar by the piezoelectric effect. Step-up ratios of more than 1000:1 have been
demonstrated. An extra feature of this transformer is that, by operating it above
its resonant frequency, it can be made to appear as an inductive load, which is
useful in circuits that require a controlled soft start.
Sensors
 To detect sound, e.g. piezoelectric microphones (sound waves bend the
piezoelectric material, creating a changing voltage) and piezoelectric pickups for
electrically amplified guitars.
 Piezoelectric oscillators are used to convert mechanical pulses into electrical
ones and vice versa. The crystal in these devices works as a transducer. The
acoustic pulses are used in underwater search (sonars) and other applications.
The acoustic pulses are generated by the piezoelectric transducers excited by
electrical fields in almost all search cases. The generation of ultrasonic waves is
invariably accomplished by exploiting the above principle.
 Piezoelectric microbalances are used as very sensitive chemical and biological
sensors.
62
 The piezoelectric effect in synthetic poly vinyliden fluoride (PVF2) is about five
times stronger than that in quartz. Being flexible and easy to handle like
ultrasonic transducers, the PVF2 films are frequently used in applications such as
monitoring blood pressure and respiration.
 Piezoelectric elements are used in electronic drum pads to detect the impact of
the drummer's sticks.
Actuators
As very high voltages correspond to only tiny changes in the width of the crystal, this
width can be changed with better-than-micrometer precision, making piezo crystals the
most important tool for positioning objects with extreme accuracy.
 Loudspeaker: Voltages are converted to mechanical movement of a piezoelectric
polymer film.
 Piezoelectric elements can be used in laser mirror alignment, where their ability
to move a large mass (the mirror mount) over microscopic distances is exploited
to electronically align some laser mirrors. By precisely controlling the distance
between mirrors, the laser electronics can accurately maintain optical conditions
inside the laser cavity to optimize the beam output.
63
 A related application is the acousto-optic modulator, a device that vibrates a
mirror to give the light reflected off it a Doppler shift. This is useful for fine-
tuning a laser's frequency.
 Atomic force microscopes and scanning tunneling microscopes employ converse
piezoelectricity to keep the sensing needle close to the probe.
Frequency standards
 Quartz clocks employ a tuning fork made from quartz that uses a combination of
both direct and converse piezoelectricity to generate a regularly timed series of
electrical pulses that is used to mark time. The quartz crystal (like any elastic
material) has a precisely defined natural frequency (caused by its shape and size)
at which it prefers to oscillate, and this is used to stabilize the frequency of a
periodic voltage applied to the crystal.
 The same principle is critical in all radiotransmitters and receivers, and in
computers where it creates a clock pulse. Both of these usually use a frequency
multiplier to reach the megahertz and gigahertz ranges.
 Crystals shaped to have a prescribed mechanical resonance frequency are used
as narrow band electrical filters. Only those electrical signals whose frequency is
coincidence with the mechanical vibrational frequency pass through the crystal
and all other are rejected.
64
 The piezoelectric materials are used as delay lines. When an electrical signal is
converted into an acoustic one to one and of a quartz rod. The signal passes
along rod as an acoustic wave, travelling at velocity of sound. At the other end
acoustic may converted into an electrical signal. The initial signal is thus delayed.
Such an arrangement is often used in communication devices.
Piezoelectric motors
 Types of piezoelectric motor include the well-known travelling-wave motor used
for auto-focus in reflex cameras, inchworm motors for linear motion, and
rectangular four-quadrant motors with high power density (2.5 watt/cm³) and
speed ranging from 10 nm/s to 800 mm/s. All these motors work on the same
principle. Driven by dual orthogonal vibration modes with a phase shift of 90°,
the contact point between two surfaces vibrates in an elliptical path, producing a
frictional force between the surfaces. Usually, one surface is fixed causing the
other to move. In most piezoelectric motors the piezoelectric crystal is excited by
a sine wave signal at the resonant frequency of the motor. Using the resonance
effect, a much lower voltage can be used to produce a high vibration amplitude.
5.4 Applications of Piezoelectric materials
 Piezoelectricity is the ability of certain dielectric crystals to produce a voltage
when subjected to mechanical stress.
65
 Pierre Curie and Jacques Curie brothers predicted and demonstrated
piezoelectricity in 1880. They showed that crystals of tourmaline, quartz, topaz,
canesugar, and Rochelle salt generate electrical polarization from mechanical
stress.
 Piezoelectric materials also show the opposite effect, called converse
piezoelectricity, where application of an electrical field creates mechanical stress
(distortion) in the crystal.
 The foremost condition for a crystal to piezoelectric is the absence of the centre
of symmetry.
 Depending on the application and the desired behaviour, a crystal is cut so as to
have the parallel faces of the crystal in a specific orientation. An X-cut is defined
as a section cut from the crystal such that the x-axis of the crystal is perpendicular
to parallel crystal faces.
 Electrostriction is a property of all dielectrics that produces a relatively slight
change of shape, or mechanical deformation, under the application of an electric
field.
 Reversal of the electric field does not reverse the direction of the deformation.
 The compression will occur in the field direction and an extension perpendicular
to the field direction.
66
 Piezoelectric crystals are used in high-voltage sources, sensors, actuators,
frequency standards, piezoelectric motors etc.,
67
PREFACE
The present book deals with the studies on “Insulating materials”.
Present book deals with advanced insulating materials, namely dielectrics,
ferroelectrics and piezoelectric materials. I believe that the information furnished in this
book will improve the easy way of understanding the importance of insulating materials
in solid state. The main purpose of the book is link to academicians, professional and
students.
I shall feel highly satisfied and amply rewarded in case the student community is
benefitted to any substantial extent.

D

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

D

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

D

Uploaded by

Copyright:

Available Formats

All rights are reserved.

No part of this publication may be reproduced, stored in a

Dielectric properties of materials 1-13

Dielectric properties of alternating electric fields 14-25

Dielectric properties of static electric fields 26-34

Ferroelectric materials 35-55

Piezoelectric materials 56-68

Dr G. NagaRaju has 18 years experience in Education and Research. He is presently working as

Associate Professor in the Department of Physics at Acharya Nagarjuna University,

published a chapter titled “Dielectric dispersion in ZnO-ZnF2-B2O3 glasses doped with

WO3”Focus on Condensed Matter Physics Chapter 5(2005) 147-181 (Nova Science

Dr. G. Naga Raju

DIELECTRIC PROPERTIES OF MATERIALS

Fig.2 An electric dipole

Susceptibility (  ): It is defined as polarization per unit electric field.

Fig. 3.Atom without any field and atom with field.

Therefore, Coulomb force of attraction, Fc

Ze.[charge enclosed in the sphere of radius d].

The induced electric dipole moment,

and the induced polarizability

Fig. 4.Orientational polarization.

Fig. 5Torque applied by a field on a dipole.

2 sin  d exp[(pE cos )/kT]

p cos . 2 sin  d exp[(pE cos )/kT]

and the average contribution per dipole p is given by

 p cos 2 sin  d exp ( pE cos ) / kT ]

 2 sin  d exp[( pE cos ) / kT ]

( = 0 corresponds to parallel alignment and  =  corresponds to anti parallel alignment of

p=1.87 Debye units to it.

temperature and the electric field.

ALTERNATING FIELDS OF DIELECTRIC MATERIALS

It should be possible to explain the frequency dependence of the dielectric constant

2.2 The complex dielectric constant and Dielectric Losses

=D0 cos  cos t + D0 sin  sin t

where  is the phase angle,

D1 = D0 cos  and D2 = D0 sin .

Thus D = * E0 eit = * E0 (cos t + i sin t)

Also we see that

The energy dissipated or absorbed per second in the dielectric is given by

2.3 Dielectric Losses and Relaxation time

Po(t) = Pos(1- e-t/)

where  is the relaxation time.

dPo/dt = (1/) [ Pos – Po (t)]

dPo/dt =(1/) [[ Pos(t) – Po (t)]

We may then write

Pos = Ps – (Pe + Pi ) = (s - ei )/4 E

dPo/dt = (1/)[ (s - ei )/4 E0 eit- Po ]

Solving this equation, we obtain

Po(t) = Ce-t/ + 1/4 (s - ei)/(1+i) E0 eit

D(t)= *E(t) = E(t) + 4 P(t)

1() = ei + (s - ei)/(1+i22)

11() = (s - ei) /(1+i22)

Fig. 1 Debye curves for ’ and ’’ as function

Fig.2 The dielectric constant as a function of

2.4 The classical theory of electronic polarization and optical absorption

x = eE0/m02 or s = ex/E0 = e2/m02 for  = 0

The electronic polarizability is therefore

The complex dielectric constant is then given by

Fig. .4 Variation of total polarizability as a function of frequency.