Piezoelectric Ty

Appendix A
Fundamentals of Piezoelectricity
Abstract The objective of this chapter is to help understand the main concepts
and working of piezoelectric sensors and transducers. Accordingly, the chapter
presents a simplified explanation of the piezoelectric phenomenon. The chapter
begins with a brief overview of some historical milestones, such as the discovery
of the piezoelectric effect and the utilization of piezoelectric materials in various
applications. Various mechanisms involved in the polarization of dielectric, ferro-
electric and piezoelectric materials are discussed. The piezoelectric effect has been
explained with basic mathematical formulations based on the intermingling of elec-
tric and elastic phenomena. In essence, the chapter describes the basic concept of
piezoelectric phenomena that one needs to know while using piezoelectric materi-
als as transducers. For detailed theory on piezoelectric phenomenon, one may refer
to standard literature on piezoelectricity (Mason in Electromechanical Transducers
and Wave Filters, 1942; Mason in Physical Acoustics and the Properties of Solids,
1958; Cady in Piezoelectricity, 1946; Nalwa in Ferroelectric Polymers: Chemistry,
Physics, and Applications, 1995; Ikeda in Fundamentals of Piezoelectricity, 1996).
Keywords Piezoelectricity · Dielectric · Ferroelectric · Piezoelectric ·

Piezoelectric effect · PVDF · PVDF-TrFE · Piezoelectric polymers · Smart
materials · Sensors · Actuators
A.1 Introduction and Historical Perspective

The name “piezo” derives from the Greek, meaning “to press”; in more modern
terminology, we say that the effect intermingles electric and elastic phenomena.
Discovered by Curie brothers [6], the piezoelectricity rapidly grew as a new field
of research in the last quarter of the nineteenth century. In 1880, Pierre and Jacques
Curie found that in certain materials such as zincblende, topaz, and quartz, mechan-
ical stresses were accompanied by macroscopic polarization and hence the produc-
tion of electric surface charges. The following year, Lippmann [7], from thermody-
namic considerations, predicted the converse effect: an imposed voltage produces
mechanical deformations or strains the material. The piezoelectric effect remained
a curiosity until the early 1920’s when its presence in quartz was utilized to realize
crystal resonators for the stabilization of oscillators, thereby launching the field of
R.S. Dahiya, M. Valle, Robotic Tactile Sensing, 195

DOI 10.1007/978-94-007-0579-1, © Springer Science+Business Media Dordrecht 2013
196 A Fundamentals of Piezoelectricity
frequency control [3]. With the introduction of quartz control, timekeeping moved
from the sun and stars to small, man-made sources that exceeded astronomy based
references in stability. Since then the advent of the man-made piezoelectric materi-
als widened the field of applications and devices based on piezoelectricity are used
in sonar, hydrophone, microphones, piezo-ignition systems, accelerometers and ul-
trasonic transducers. The discovery of a strong piezoelectric effect in polyvinylidene
fluoride (PVDF) polymer, by Kawai in 1969 [8], further added many applications
where properties such as mechanical flexibility are desired. Today, piezoelectric ap-
plications include smart materials for vibration control, aerospace and astronautical
applications of flexible surfaces and structures, sensors for robotic applications, and
novel applications for vibration reduction in sports equipment (tennis racquets and
snowboards). Recently, the newer and rapidly burgeoning areas of utilization are
the non-volatile memory and the integral incorporation of mechanical actuation and
sensing microstructures (e.g. POSFET devices) into electronic chips.
A.2 Dielectric, Ferroelectric and Piezoelectric Materials
A.2.1 Electric Polarization
One of the concepts crucial to the understanding of dielectric, ferroelectric and

piezoelectric materials is their response to an externally applied electric field. When
a solid is placed in an externally applied electric field, the medium adapts to this
perturbation by dynamically changing the positions of the nuclei and the electrons.
As a result, dipoles are created and the material is said to undergo polarization. The
process of dipole formation (or alignment of already existing permanent or induced
atomic or molecular dipoles) under the influence of an external electric field that
has an electric field strength, E, is called polarization. There are three main types or
sources of polarization: electronic, ionic, and dipolar or orientation. A fourth source
of polarization is the interfacial space charge that occurs at electrodes and at het-
erogeneities such as grain boundaries. For an alternating electric field, the degree
to which each mechanism contributes to the overall polarization of the material de-
pends on the frequency of the applied field, as shown in Fig. A.1. Each polarization
mechanism ceases to function when the applied field frequency exceeds its relax-
ation frequency.
Electronic polarization may be induced to one degree or another in all atoms. It
results from a displacement of the center of the negatively charged electron cloud
relative to the positive nucleus of an atom by the electric field, as indicated in the
bottom-right of Fig. A.1. This polarization type is found in all dielectric materi-
als, and, of course, exists only while an electric field is present. Ionic polarization
occurs only in materials that are ionic. An applied field acts to displace cations in
one direction and anions in the opposite direction, which gives rise to a net dipole
moment. This phenomenon is illustrated in right-center of Fig. A.1. The third type,
orientation polarization, is found only in substances that possess permanent dipole
A.2 Dielectric, Ferroelectric and Piezoelectric Materials 197
Fig. A.1 Frequency dependence of various polarization mechanisms. The electronic, ionic, and
orientation polarization mechanisms are indicated
moments. Polarization results from a rotation of the permanent moments into the
direction of the applied field, as represented in top-right of Fig. A.1.
The total electric polarization of a substance is equal to the sum of the electronic,
ionic, and orientation (and space-charge) polarizations. It is possible for one or more
of these contributions, to the total electric polarization, to be either absent or neg-
ligible in magnitude relative to the others. For example, ionic polarization will not
exist in covalently bonded materials in which no ions are present. The average po-
−
→
larization per unit volume, P , produced by N little electric dipoles (of the electric
−
→
dipole moment, p ) which are all aligned, is given by:
−
→ 1 −
→
N
P = pk (A.1)
V olume
k=0
A.2.2 Dielectric Materials
A dielectric material is one that is electrically insulating (nonmetallic) and exhibits

or may be made to exhibit an electric dipole structure; that is, there is a separation
of positive and negative electrically charged entities on a molecular or atomic level.
The dielectric materials ordinarily exhibit at least one of the polarization types dis-
cussed in previous section—depending on the material and also the manner of the
external field application. There are two types of dielectrics. The first type is polar
dielectrics, which are dielectrics that have permanent electric dipole moments. As
depicted in top-right of Fig. A.1, the orientation of polar molecules is random in the
absence of an external field. When an external electric field E is present, a torque
is set up and causes the molecules to align with E. However, the alignment is not
complete due to random thermal motion. The second type of dielectrics is the non-
polar dielectrics, which are dielectrics that do not possess permanent electric dipole
moment. Electric dipole moments can be induced by placing the materials in an ex-
ternally applied electric field. Figure A.1 illustrates the state of non-polar molecules
with and without an external field.
Dielectric materials are electrically insulating, yet susceptible to polarization in
the presence of an electric field. This polarization phenomenon accounts for the
ability of the dielectrics to increase the charge storing capability of capacitors, the
efficiency of which is expressed in terms of a dielectric constant.
A.2.3 Ferroelectric Materials
Ferroelectrics are the class of dielectric materials which exhibit spontaneous po-
larization (polarization associated with a spontaneously formed dipole moment)—
that is, polarization in the absence of an electric field. The spontaneous polarization
of ferroelectric materials can be switched by applying an electric field; its expres-
sion is a typical P–E hysteresis loop similar to that shown in Fig. A.5(a). They
undergo a structural phase transition from a high temperature paraelectric into a
low-temperature ferroelectric phase at Curie temperature. Ferroelectric materials are
the dielectric analogue of ferromagnetic materials, which may display permanent
magnetic behavior. There must exist permanent electric dipoles in the ferroelectric
materials. Ferroelectrics are a class of the polar piezoelectrics (e.g. polarized PZT,
PVDF, and P(VDF-TrFE)) and hence all ferroelectric materials are piezoelectric.
A.2.4 Piezoelectric Materials
Piezoelectrics are the class of dielectric materials which can be polarized, in addi-
tion to an electric field, also by application of a mechanical stress (Fig. A.2). This
unusual property exhibited by a few dielectric materials is called piezoelectricity,
or, literally, pressure electricity. Piezoelectric materials can be divided into polar
(which possess a net dipole moment) and non polar piezoelectric materials (whose
dipolar moments summed in different directions give a null total moment). A de-
tailed description of the piezoelectric effect is given in following sections.
A.3 The Piezoelectric Effect
The piezoelectric effect, explained with a simple molecular model shown in

Fig. A.3, is the generation of an electric charge as a result of a force exerted on
A.3 The Piezoelectric Effect 199
Fig. A.2
Piezoelectricity—An
intermingling of electric and
elastic phenomena
the material. Before subjecting the material to an external stress the centers of the
negative and positive charges of each molecule coincide—resulting into an elec-
trically neutral molecule as indicated in Fig. A.3(a). However, in presence of an
external mechanical stress the internal reticular can be deformed, thus causing the
separation of the positive and negative centers of the molecule and generating little
dipoles as indicated in Fig. A.3(b). As a result, the opposite facing poles inside the
material cancel each other and fixed charges appear on the surface. This is illustrated
in Fig. A.3(c). That is to say, the material is polarized and the effect called direct
piezoelectric effect. This polarization generates an electric field that can be used to
transform the mechanical energy, used in the material’s deformation, into electrical
energy.
Figure A.4(a) shows the piezoelectric material with two metal electrodes de-
posited on opposite surfaces. If the electrodes are externally short circuited, with a
Fig. A.3 Piezoelectric effect explained with a simple molecular model: (a) An unperturbed
molecule with no piezoelectric polarization (though prior electric polarization may exist); (b) The
molecule subjected to an external force (Fk ), resulting in to polarization (Pk ) as indicated; (c) The
polarizing effect on the surface when piezoelectric material is subjected to an external force
Fig. A.4 Piezoelectric phenomenon: (a) The neutralizing current flow when two terminal of piezo-
electric material, subjected to external force, are short circuited; (b) The absence of any current
through the short-circuit when material is in an unperturbed state
galvanometer connected to the short circuiting wire, and force is applied the surface
of piezoelectric material, a fixed charge density appears on the surfaces of the crystal
in contact with the electrodes. This polarization generates an electric field which in
turn causes the flow of the free charges existing in the conductor. Depending on their
sign, the free charges will move toward the ends where the fixed charges generated
by polarization are of opposite sign. This flow of free charge continues until the free
charge neutralizes the polarization effect, as indicated in Fig. A.4(a). This implies
that no charge flows in the steady state or in the unperturbed state—irrespective
of the presence of external force. When the force on the material is removed, the
polarization too disappears, the flow of free charges reverses and finally the mate-
rial comes back to its original standstill state indicated in Fig. A.4(b). This process
would be displayed in the galvanometer, which would have marked two opposite
sign current peaks. If short-circuiting wire is replaced with a resistance/load, the
current would flow through it and hence mechanical energy would be transformed
into electrical energy. This scheme is fundamental to various energy harvesting tech-
niques that tap ambient mechanical energy such as vibrations [9] and convert it into
usable electrical form.
Some materials also exhibit the reverse piezoelectric effect i.e. a mechanical de-
formation or strain is produced in the material when a voltage is applied across
the electrodes. The strain generated in this way could be used, for example, to dis-
place a coupled mechanical load. This way of transforming the electrical electric
energy into usable mechanical energy is fundamental to the applications such as
nano-positioning devices.
A.4 Piezoelectric Materials—Static Actions

The piezoelectric materials are anisotropic in nature and hence their electrical, me-
chanical, and electromechanical properties differ for the electrical/mechanical ex-
A.4 Piezoelectric Materials—Static Actions 201
Fig. A.5 Piezoelectric material in sensing and actuating applications. (a) Typical P–E hysteresis
plot (top) and the strain versus electric field plot (bottom) of a piezoelectric material. (b) The
piezoelectric material before (dotted) and after poling, albeit change in dimension is exaggerated.
The polarity of poling voltage is clearly indicated. (c) Material’s dimension when applied voltage
has polarity similar to that of poling voltage. (d) Material’s dimension when applied voltage has
polarity opposite to that of poling voltage. (e) The generated voltage with polarity similar to poling
voltage when compressive force is applied in poling direction. (f) The generated voltage with
polarity opposite to poling voltage when tensile force is applied in poling direction
citations along different directions. Using them in various sensing or actuating ap-
plications requires a systematic tabulation of their properties—for which, a stan-
dardized means for identifying directions is very important. That is to stay, once the
piezoelectric material is chosen for a particular application, it is important to set the
mechanical and electrical axes of operation. Wherever crystals are concerned, the
orthogonal axes originally assigned by crystallographers are used for this purpose.
A general practice to identify the axes is to assign them the numerals e.g. 1 corre-
sponds to x axis; 2 corresponds to y axis, and 3 corresponds to z axis. These axes are
set during “poling”; the process that induces piezoelectric properties in the piezo-
electric material. The orientation of the DC poling field determines the orientation
of the mechanical and electrical axes. The direction of the poling field is generally
identified as one of the axes. The poling field can be applied in such as way that the
material exhibits piezoelectric responses in various directions or combination of di-
rections. The poling process permanently changes the dimensions of a piezoelectric
material, as illustrated in Fig. A.5(b). The dimension between the poling electrodes
increases and the dimensions parallel to the electrodes decrease. In some materi-
als, the poling step is also needed for the introduction of piezoelectric effect. For
example, in virgin state the piezoelectric materials such as PVDF, P(VDF-TrFE),
and ceramics are isotropic and are not piezoelectric before poling. Once they are
polarized, however, they become anisotropic.
After the poling process is complete, a voltage lower than the poling voltage
changes the dimensions of the piezoelectric material for as long as the voltage is
applied. A voltage with the same polarity as the poling voltage causes additional
expansion along the poling axis and contraction perpendicular to the poling axis,
as illustrated in Fig. A.5(c). One can also notice this from the P–E and S–E plots
shown in Fig. A.5(a). When a poling field, E, is applied across a piezoelectric
material, the polarization as well as the mechanical strain curves follow the path
(i)–(ii) on the P–E and S–E plots respectively. When the poling field is removed,
the path (ii)–(iii) is followed and piezoelectric material retains certain level of po-
larization, called remanent polarization, Pr , and experiences permanent strain or
permanent change in the dimensions. From operational point of view, the poling
procedure shifts the working point from (i) to (iii). After this, whenever a volt-
age with the same polarity as the poling field is applied, the P–E and S–E plots
will follow the curve (iii)–(ii) and hence positive strain will be developed, as in-
dicated in Fig. A.5(a). In other words, there is expansion along the poling axis, as
shown in Fig. A.5(c). Similarly, when a voltage with the polarity opposite to the
poling voltage is applied the P–E and S–E plots will follow the curve (iii)–(iv),
resulting in negative strain. As a result, there is contraction along the poling axis
and expansion perpendicular to the poling axis, as indicated in Fig. A.5(d). In both
cases, however, the piezoelectric material returns to its poled dimensions on the
plots (i.e. working point (iii) on Fig. A.5(a)) when the voltage is removed from the
electrodes.
If after completion of the poling process, a compressive and tensile force is ap-
plied to the piezoelectric material, a voltage is generated as shown in Fig. A.5(d), (e).
With an argument similar to that presented in previous paragraph it can be shown
that the generated voltage will have the same polarity as the poling field when a
compressive force is applied along the poling axis or a tensile force applied perpen-
dicular to the poling axis. This is illustrated in Fig. A.5(e). Similarly, as indicated
in Fig. A.5(f), a voltage with the opposite polarity will result when a tensile force is
applied along the poling axis, or when a compressive force is applied perpendicular
to the poling axis.
The knowledge of the voltage polarities is very helpful before a piezoelectric ma-
terial is actually put to use. Generally two or more of the above mentioned actions
are present simultaneously. In some cases one type of expansion is accompanied by
another type of contraction which compensate each other resulting in no change of
volume. For example, the expansion of length of a plate may be compensated by an
equal contraction of width or thickness. In some materials, however, the compen-
sating effects are not of equal magnitude and net volume change does occur. In all
cases, the deformations are, however, very small when amplification by mechanical
resonance is not involved. The maximum displacements are on the order of a few
micro-inches.
A.5 Piezoelectric Effect—Basic Mathematical Formulation

This section presents basic mathematical formulation describing the electromechan-
ical properties of piezoelectric materials. The presentation is based on the linear the-
A.5 Piezoelectric Effect—Basic Mathematical Formulation 203
ory of piezoelectricity [10], according to which, piezoelectric materials have a linear

profile at low electric fields and at low mechanical stress levels.1 For the range of
mechanical stresses and electrical fields used in this book, the piezoelectric materi-
als exhibit the linear behavior.
As explained in previous sections, when a poled piezoelectric material is mechan-
ically strained it becomes electrically polarized, producing fixed electric charge on
the surface of the material. If electrodes are attached to the surfaces of the material,
the generated electric charge can be collected and used. Following the linear theory
of piezoelectricity [10], the density of generated fixed charge in a piezoelectric ma-
terial is proportional to the external stress. In a first mathematical formulation, this
relationship can be simply written as:
Ppe = d × T (A.2)
where Ppe is the piezoelectric polarization vector, whose magnitude is equal to the
fixed charge density produced as a result of piezoelectric effect, d is the piezoelectric
strain coefficient and T is the stress to which piezoelectric material is subjected.
For simplicity, the polarization, stress, and the strain generated by the piezoelectric
effect have been specified with the ‘pe’ subscript, while those externally applied do
not have any subscript. In a similar manner, the indirect/reverse piezoelectric effect
can be formulated as:
Spe = d × E (A.3)
where Spe is the mechanical strain produced by reverse piezoelectric effect and E is
the magnitude of the applied electric field. Considering the elastic properties of the
material, the direct and reverse piezoelectric effects can alternatively be formulated
as:
Ppe = d × T = d × c × S = e × S (A.4)
Tpe = c × Spe = c × d × E = e × E (A.5)
where c is the elastic constant relating the generated stress and the applied strain
(T = c × S), s is the compliance coefficient which relates the deformation produced
by the application of a stress (S = s × T ), and e is the piezoelectric stress constant.
A.5.1 Contribution to Elastic Constants
The piezoelectric phenomenon causes an increase of the material’s stiffness. To un-

derstand this effect, let us suppose that the piezoelectric material is subjected to a
1 At high electric field or high mechanical stress, they may show considerable nonlinearity.
strain S. This strain will have two effects: One, it will generate an elastic stress
Te , proportional to the mechanical strain (Te = c × S); and two, it will generate a
piezoelectric polarization Ppe = e × S according to Eq. (A.4). This polarization will
create an internal electric field in the material Epe given by:
Ppe e×S
Epe = = (A.6)
ε ε
where ε is the dielectric constant of the material. Assuming a compressive stress,
applied on the piezoelectric material along the poling direction, it is known from
Fig. A.5(d) that the resulting electric field of piezoelectric origin will have a di-
rection same as that of poling field. Further, it is also known from Fig. A.5(c) and
related discussion in previous section that the presence of an electric field with po-
larity same as that of poling field results in positive strain and hence the expansion
of piezoelectric material in the poling direction. That is say, the electric field (Epe
from Eq. (A.6)), of piezoelectric origin, produces a stress which opposes the applied
external stress. This is also true if the external applied stress is tensile in nature. The
stress Tpe (= e × Epe ), produced by the electric field Epe , as well as that of elastic
origin, is against the material’s deformation. Consequently, the stress generated by
the strain S is:

e2 e2
T = Te + Tpe = c × S + ×S = c+ ×S =c×S (A.7)
ε ε
Thus, piezoelectric effect results in an increased elastic constant or in other

words, the material gets stiffened in presence of piezoelectric effect. The constant
c, in Eq. (A.7), is the piezoelectrically stiffened constant.
A.5.2 Contribution to Dielectric Constants
When an external electric field E is applied between two electrodes where a mate-
rial of dielectric constant ε exists, an electric displacement is created toward those
electrodes, generating a surface charge density σ = σo + σd , as shown in Fig. A.6.
The magnitude of this electric displacement is D = ε × E.2 If the material is piezo-
electric, the electric field E also produces a strain, expressed as Spe = d × E. This
strain, of reverse piezoelectric origin, can be positive or negative depending on the
direction of the external electric field with respect to the poling field. As discussed
2 Thefree charge density which appears on the electrodes, will be the sum of the charge density
which appears in vacuum plus the one that appears induced by the dielectric effect, i.e.:
σ = σo + σd = εo × E + χ × E = (εo + χ ) × E = ε × E (A.8)
where εo is the vacuum dielectric permittivity and χ is the dielectric susceptibility of the material.
Fig. A.6 Schematic diagram

indicating different electrical
displacements associated with
a piezoelectric and dielectric
material
in previous section, if the direction of external field is same as that of poling field,
the strain is positive and material undergoes expansion along the direction of poling
field. This is illustrated in Fig. A.5(c). It is also evident from Fig. A.5(f) that the
expansion of material along poling field (or compression perpendicular to poling
field) generates a voltage having polarity opposite to that of poling field or opposite
to the external applied field. The situation is similar to the one shown in Fig. A.6.
In essence, this means the polarization, and hence the surface charge density, in-
creases when the direction of applied external field is same as that of poling field.
In fact, using Fig. A.5(d) and Fig. A.5(e), it can easily be shown that the surface
charge density increases even if the direction of applied external field is opposite
to that of the poling field. Thus, the strain, of reverse piezoelectric origin, results
in polarization and therefore the surface charge density is increased by an amount
Ppe = e × Spe = e × d × E (Fig. A.6). If the electric field is maintained constant,
the additional polarization due to piezoelectric effect increases the electric displace-
ment of free charges toward the electrodes by the same magnitude i.e. σpe = Ppe .
Therefore, the total electrical displacement is:
D = ε × E + Ppe = ε × E + e × d × E = ε × E (A.9)
where ε is the effective dielectric constant.
A.5.3 Piezoelectric Linear Constitutive Relations
So far, the individual effect of piezoelectricity on the elastic and dielectric has been
discussed. In actual practice, piezoelectricity is a cross coupling between the elastic
variables, stress T and strain S, and the dielectric variables, electric charge den-
sity D and electric field E. This coupling in piezoelectric materials is discussed
in this sub-section with help of commonly used linear electro-elastic constitutive
equations.
According to the linear theory of piezoelectricity [10], the tensor relation to iden-
tify the coupling between mechanical stress, mechanical strain, electric field and
electric displacement is given as:
Fig. A.7 Tensor directions

for defining the constitutive
relations. In PVDF,
1 corresponds to the draw
direction (indicated by dotted
line), 2 to the transverse
direction, and 3 to thickness
(also the poling axis)
Sp = spq
E
Tq + dpk Ek (A.10)
Di = diq Tq + εik
T
Ek (A.11)
E is elastic compliance tensor at constant electric field, ε T is dielectric

where, spq ik
constant tensor under constant stress, dkp is piezoelectric constant tensor, Sp is the
mechanical strain in p direction, Di is electric displacement in i direction, Tq is
mechanical stress in q direction, and Ek is the electric field in k direction. The
common practice is to label directions as depicted in Fig. A.7. In case of materials
such as PVDF, the stretch direction is denoted as “l” and the axis orthogonal to
the stretch direction in the plane of the film becomes “2”. The polarization axis
(perpendicular to the surface of the film) is denoted as “3”. The shear planes are
indicated by the subscripts “4”, “5”, “6” and are perpendicular to the directions
“1”, “2”, and “3” respectively. Using these directions, Eqs. (A.1) and (A.2) can be
displayed in matrix form as follows:
⎡ ⎤
⎡ ⎤ E
s11 E
s12 E
s13 E
s14 E
s15 E
s16 ⎡ ⎤
S1 ⎢ ⎥ T1
⎥ ⎢ ⎥⎢
E E E E E E
⎢ s21 s22 s23 s24 s25 s26 ⎥
⎢ S2 ⎥ ⎢ ⎥⎢ T2 ⎥
⎢ ⎥ ⎢ E E E E E E ⎥⎢ ⎥
⎥=⎢ ⎥⎢
S3 s31 s32 s33 s34 s35 s36 T3
⎢ ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎥ ⎢ ⎥⎢
S4 E E E E E E T4
⎢ s41 s42 s43 s44 s45 s46 ⎥
⎣ S5 ⎦ ⎢
⎢ E E E E E E
⎥⎣
⎥ T5 ⎦
S6 ⎣ s51 s52 s53 s54 s55 s56 ⎦ T6
E
s61 E
s62 E
s63 E
s64 E
s65 E
s66
⎡ ⎤
d11 d12 d13
⎢ d21 d22 d23 ⎥⎡ ⎤
⎢ ⎥ E1
⎢ d31 d32 d33 ⎥
+⎢
⎢
⎥ ⎣ E2 ⎦
⎥ (A.12)
⎢ d41 d42 d43 ⎥ E3
⎣ d51 d52 d53 ⎦
d61 d62 d63
⎡ ⎤
T1
⎡ ⎤ ⎡ ⎤⎢ T2 ⎥
D1 d11 d12 d13 d14 d15 d16 ⎢ ⎢
⎥
⎥
T3
⎣ D2 ⎦ = ⎣ d21 d22 d23 d24 d25 d26 ⎦ ⎢
⎢
⎥
⎥
T4
D3 d31 d32 d33 d34 d35 d36 ⎢ ⎣
⎥
⎦
T5
T6
⎡ T T T
⎤⎡ ⎤
ε11 ε12 ε13 E1
⎢ T ⎥⎣ ⎦
+ ⎣ ε21 T
ε22 T
ε23 ⎦ E2 (A.13)
T
ε31 T
ε32 T
ε33 E3
Another fundamental parameter used in electromechanical applications is the

electromechanical coupling factor k. The electromechanical coupling factor, which
measures the ability of a material to interconvert electrical and mechanical energy,
is expressed as:
Converted Mechanical Energy

k2 = (A.14)
Input Electrical Energy
or
Converted Electrical Energy
k2 = (A.15)
Input Mechanical Energy
In many cases, processing conditions such as extrusion and the particular crys-
tal symmetry of piezoelectric material determine which components of the dielec-
tric constant, piezoelectric, and elastic compliance tensors are non-zero and unique.
For example, for an unstretched and poled piezoelectric P(VDF-TrFE) copolymer,
having 2 mm macroscopic symmetry, the matrix form of Eqs. (A.3) and (A.4) can
written as:
⎡ ⎤
⎡ ⎤ E
s11 E
s12 E
s13 0 0 0 ⎡ ⎤
S1 ⎢ ⎥ T1
⎥ ⎢ ⎥⎢
E E E
⎢ s21 s22 s23 0 0 0 ⎥
⎢ S2 ⎥ ⎢ ⎥⎢ T2 ⎥
⎢ ⎥ ⎢ E E E ⎥⎢ ⎥
⎥=⎢ ⎥⎢
S3 s31 s32 s33 0 0 0 T3
⎢ ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥
⎥ ⎢ ⎥⎢
S4 E T4
⎢ 0 0 0 s44 0 0 ⎥
⎣ S5 ⎦ ⎢
⎢ E
⎥⎣
⎥ T5 ⎦
S6 ⎣ 0 0 0 0 s55 0 ⎦ T6
0 0 0 0 0 E
s66
⎡ ⎤
0 0 d31
⎢ 0 0 d32 ⎥⎡ ⎤
⎢ ⎥ E1
⎢ 0 0 d33 ⎥
+⎢
⎢
⎥ ⎣ E2 ⎦
⎥ (A.16)
⎢ 0 d24 0 ⎥ E3
⎣ d15 0 0 ⎦
0 0 0
Table A.1 Piezoelectric, dielectric, and elastic properties of PVDF [13–15] and P(VDF-TrFE)
copolymer with 75/25 mol% [12]
Coefficient/Parameter PVDF P(VDF-TrFE) 75/25a
Real Imaginary
d31 (pC/N) 21b 10.7 0.18

d32 (pC/N) 1.5b 10.1 0.19
d33 (pC/N) −32.5b −33.5 −0.65
d15 (pC/N) −27b −36.3 −0.32
d24 (pC/N) −23b −40.6 −0.35
E (10−10 Pa−1 )
s11 3.65c 3.32 0.1
E (10−10 Pa−1 )
s22 4.24c 3.34 0.07
E (10−10
s33 Pa−1 ) 4.72c 3.00 0.07
E (10−10
s44 Pa−1 ) – 94.0 2.50
E (10−10
s55 Pa−1 ) – 96.3 2.33
E (10−10
s66 Pa−1 ) – 14.4 –
E (10−10
s12 Pa−1 ) −1.10c −1.44 –
E (10−10
s13 Pa−1 ) −2.09c −0.89 –
E (10−10
s23 Pa−1 ) −1.92c −0.86 –
T /ε
ε11 0 6.9d 7.4 0.07
T /ε
ε22 0 8.6d 7.95 0.09
T
ε33 /ε0 7.6d 7.9 0.09
a Reference [12]. b Reference [13]. c Reference [14]. d Reference [15]
⎡ ⎤
T1
⎡ ⎤ ⎡ ⎤⎢ T2 ⎥
D1 0 0 0 0 d15 0 ⎢⎢
⎥
⎥
T3
⎣ D2 ⎦ = ⎣ 0 0 0 d24 0 0 ⎦⎢
⎢
⎥
⎥
T4
D3 d31 d32 d33 0 0 0 ⎢⎣
⎥
⎦
T5
T6
⎡ T
⎤⎡
⎤
ε11 0 0 E1
⎢ ⎥
+⎣ 0 T
ε22 0 ⎦ ⎣ E2 ⎦ (A.17)
0 0 T
ε33 E3
Due to the fact that electromechanical response depends on a number of factors,

including polarization conditions, stress/strain rates, temperatures, and hydrostatic
pressure, the reported data for the values of various coefficients in the above equa-
tions for PVDF and P(VDF-TrFE) copolymer appear to involve certain inconsisten-
cies. Nevertheless, it is possible to identify the typical values such as those listed in
Table A.1.
References 209
The above tensor relations of Eqs. (A.5) and (A.6) are used to obtain the elec-
tromechanical response of a piezoelectric material along the same or other direction
as that of stimulus. An as example, the electromechanical response of P(VDF-TrFE)
copolymer, when it is used in the thickness mode i.e. both stress and electric field
are along the 3-direction, as in Fig. A.7, can be expressed as [11]:
S3 = s33
E
T3 + d33 E3 (A.18)
D3 = d33 T3 + ε33
T
E3 (A.19)
The expression for longitudinal electromechanical coupling factor k33 , under similar
conditions is:
2
d33
2
k33 = T sE
(A.20)
ε33 33
T , and s E are frequently found in the manufacturer’s data.
The constants d33 , ε33 33
In Eqs. (A.7) and (A.8), T3 and E3 are used as independent variables. If, however,
D3 and S3 are the independent variables the above relations can also be written as:
T3 = c33
D
S3 − h33 D3 (A.21)
D3
E3 = −h33 S3 + β33
S
D3 = −h33 S3 + S
(A.22)
ε33
D , h , and ε S are related to d , ε T , and s E by following
The new constants c33 33 33 33 33 33
mathematical relations:
2
d33
S
ε33 = ε33
T
− E
(A.23)
s33
d33
h33 = E T
(A.24)
s33 ε33
1
D
c33 = h233 ε33
T
+ E
(A.25)
s33
E
D
s33 = 1 − k33
2
s33 (A.26)
Depending on the independent variables chosen to describe the piezoelectric be-

havior, there can be two more variants of the (A.9) and (A.10), which are not dis-
cussed here. For a deeper understanding of piezoelectricity one may refer to standard
literature on piezoelectricity [1–5].
References
1. W.P. Mason, Electromechanical Transducers and Wave Filters (Van Nostrand, New York,
1942)
2. W.P. Mason, Physical Acoustics and the Properties of Solids (Van Nostrand, Princeton, 1958)
3. W.G. Cady, Piezoelectricity (McGraw-Hill, New York, 1946)
4. H.S. Nalwa, Ferroelectric Polymers: Chemistry, Physics, and Applications (Marcel Dekker,
Inc., New York, 1995)
5. T. Ikeda, Fundamentals of Piezoelectricity (Oxford University Press, New York, 1996)
6. P. Curie, J. Curie, Dévelopment, par pression, de l’électricité polaire dans les cristaux hémiè-
dres à faces inclinées. C. R. Acad. Sci. 91, 294–295 (1880)
7. G. Lippmann, Principe de conservation de l’électricité. Ann. Chim. Phys. 24(5a), 145–178
(1881)
8. H. Kawai, The piezoelectricity of PVDF. Jpn. J. Appl. Phys. 8, 975–976 (1969)
9. L. Pinna, R.S. Dahiya, M. Valle, G.M. Bo, Analysis of self-powered vibration-based energy
scavenging system, in ISIE 2010: The IEEE International Symposium on Industrial Electron-
ics, Bari, Italy (2010), pp. 1–6
10. ANSI/IEEE, IEEE standard on piezoelectricity. IEEE Standard 176-1987 (1987)
11. R.S. Dahiya, M. Valle, L. Lorenzelli, SPICE model of lossy piezoelectric polymers. IEEE
Trans. Ultrason. Ferroelectr. Freq. Control 56, 387–396 (2009)
12. H. Wang, Q.M. Zhang, L.E. Cross, A.O. Sykes, Piezoelectric, dielectric, and elastic properties
of poly(vinylidene fluoride/trifluoroethylene). J. Appl. Phys. 74, 3394–3398 (1993)
13. E.L. Nix, I.M. Ward, The measurement of the shear piezoelectric coefficients of polyvinyli-
dene fluoride. Ferroelectrics 67, 137–141 (1986)
14. V.V. Varadan, Y.R. Roh, V.K. Varadan, R.H. Tancrell, Measurement of all the elastic and di-
electric constants of poled PVDF films, in Proceedings of 1989 Ultrasonics Symposium, vol. 2
(1989), pp. 727–730
15. H. Schewe, Piezoelectricity of uniaxially oriented polyvinylidene fluoride, in Proceedings of
1982 Ultrasonics Symposium (1982), pp. 519–524
Appendix B
Modeling of Piezoelectric Polymers
Abstract Piezoelectric polymers are used as transducers in many applications,

including the tactile sensing presented in this book. It is valuable to use some form
of theoretical model to assess, the performance of a transducer; the effects of de-
sign changes; electronics modifications etc. Instead of evaluating the transducer and
conditioning electronics independently, which may not result in optimized sensor
performance, it is advantageous to develop and implement the theoretical model of
transducer in such a way that overall sensor (i.e. transducer + conditioning elec-
tronics) performance can be optimized. In this context, the ease with which the
conditioning electronics can be designed with a SPICE like software tool, makes
it important to implement the theoretical model of transducer also with a similar
software tool. Moreover, with SPICE it is easier to evaluate the performance of
transducer, both, in time and frequency domains. The equivalent model of piezo-
electric polymers—that includes the mechanical, electromechanical and dielectric
losses—and SPICE implementation of the same are presented in this chapter.
Keywords Piezoelectricity · Piezoelectric effect · PVDF · PVDF-TrFE ·

Piezoelectric polymers · Smart materials · Sensors · Actuators · Simulation ·
Modeling · Transmission line model · SPICE · Lossy piezoelectric model
B.1 Introduction
Much work has been published on transducers using piezoelectric ceramics, but a
great deal of this work does not apply to the piezoelectric polymers because of their
unique electrical and mechanical properties [1]. A number of attempts to model
the behavior of piezoelectric materials fail to predict the behavior of piezoelectric
polymers because of their lossy and dispersive dielectric properties and higher vis-
coelastic losses. Starting from the Redwood’s transmission line version of Mason’s
equivalent circuit [2], a SPICE implementation of piezoelectric transducer was re-
ported by Morris et al. [3]. Usage of negative capacitance, −C0 (an unphysical elec-
trical circuit element), by Morris et al. was avoided by Leach with the controlled
source technique in an alternative SPICE implementation [4]. The models presented
in both these works were verified for piezoceramics operating in the actuating mode
i.e. with electrical input and mechanical output. The transducer was assumed to

212 B Modeling of Piezoelectric Polymers
be lossless, and hence, both these implementations are insufficient for evaluating
the performance of transducers with significant losses. Püttmer et al. [5] improved
these piezoceramics models by involving a resistor—with value equal to that at fun-
damental resonance—to represent the acoustic losses in the transmission line, un-
der the assumption that transmission line has low losses. Further, the dielectric and
electromechanical losses in the transducer were assumed negligible. These assump-
tions worked well for piezoceramics, but modeling of lossy polymers like PVDF
requires inclusion of all these losses. Keeping these facts in view, the equivalent
model of piezoelectric polymers—that includes the mechanical, electromechanical
and dielectric losses—was developed [6] and SPICE implementation of the same is
presented in this chapter.
Following sections present the theory of the lossy model of piezoelectric poly-
mers; its SPICE implementation and its evaluation vis-a-vis experimental results.
Using the model presented in this chapter, many design issues associated with the
piezoelectric polymers are also discussed.
B.2 Theory
B.2.1 Piezoelectric Linear Constitutive Relations
According to the linear theory of piezoelectricity [7], the tensor relation between
mechanical stress, mechanical strain, electric field and electric displacement is:
Sp = spq
E
Tq + dkp Ek (B.1)
Di = diq Tq + εik
T
Ek (B.2)
where, Sp is the mechanical strain in p direction, Di is electric displacement in i

direction, Tq is mechanical stress in q direction, Ek is the electric field in k direction,
E is elastic compliance at constant electric field, ε T is dielectric constant under
spq ik
constant stress, and dkp is piezoelectric constant. In the event when polymer is used
in the thickness mode, as shown in Fig. B.1, the tensor relations (B.1)–(B.2) can be
written as:
S3 = s33
E
T3 + d33 E3 (B.3)
D3 = d33 T3 + ε33
T
E3 (B.4)
T , and s E are frequently found in the manufacturer’s data
The constants d33 , ε33 33
for polarized polymers. The analysis is simpler if the variables S3 , T3 , D3 and E3
are arranged in the alternate way of writing the piezoelectric relation using D3 and
S3 as independent variables. Thus the equations representing plane compression
wave propagation in the x direction (3-direction or the direction of polarization) in
a piezoelectric medium are:
B.2 Theory 213
Fig. B.1 Piezoelectric

polymer operating in
thickness mode
T3 = c33
D
S3 − h33 D3 (B.5)
D3
E3 = −h33 S3 + β33
S
D3 = −h33 S3 + S
(B.6)
ε33
D , h and ε S are related to the previous constants by:

The new constants c33 33 33
2
d33
S
ε33 = ε33
T
− E
(B.7)
s33
d33
h33 = E εT
(B.8)
s33 33
1
D
c33 = h233 ε33
T
+ E
(B.9)
s33
Based on the choice of independent variables, there can be two other variants of
the (B.5) and (B.6) which are not given here. Inside polymer, the mechanical strain,
mechanical stress and the electrical displacement can be written as:
∂ξ
S3 = (B.10)
∂x
∂T3 ∂ 2ξ
=ρ 2 (B.11)
∂x ∂t
Div(D) = 0 (B.12)
where ξ is the displacement of the particles inside polymer, ρ is the density of the
polymer. Equation (B.11) is basically Newton’s law and (B.12) is Gauss’s law. Using
(B.5) and (B.10)–(B.12), the mechanical behavior of the particles inside polymer
can be described as a wave motion which is given by:
∂ 2ξ ∂ 2ξ
2
= ν2 2 (B.13)
∂t ∂x

where, ν = c33 D /ρ is the sound velocity in the polymer and should not be confused
with particle velocity (∂ξ/∂t).
B.2.2 Losses
In general, the piezoelectric polymers possess frequency dependent mechanical, di-

electric and electromechanical losses. These losses can be taken into account by
replacing elastic, dielectric and piezoelectric constants in (B.1)–(B.13), with their
complex values [1, 8]. In other words, the mechanical/viscoelastic, dielectric and
electromechanical losses are taken into account by using complex elastic constant,
D ∗ ; complex dielectric constant, ε S ∗ and complex electromechanical coupling co-
c33 33
∗
efficient, kt respectively. These complex constants can be written as:
∗
D
c33 = cr + j ci = c33
D
(1 + j tan δm ) (B.14)
S∗
ε33 = εr − j εi = ε33
S
(1 − j tan δe ) (B.15)
∗
kt = ktr + j kti = kt (1 + j tan δk ) (B.16)
where, the subscripts r and i stand for real and imaginary terms and tan δm , tan δe ,
tan δk are the elastic, dielectric and electromechanical coupling factor loss tangent,
∗ ∗
respectively. The complex piezoelectric constants viz: d33 and h33 , can be obtained
∗
from electromechanical coupling constant kt [7].
B.2.3 Polymer Model with Losses
It is assumed that a one-dimensional compression wave is propagating in X di-

rection of thickness-mode piezoelectric transducer, as shown in Fig. B.1. It is also
assumed that the electric field E and the electric displacement D are in the X direc-
tion. Let u (= ua − ub ) be the net particle velocity, F (= Fa − Fb ) be the force, and
lx , ly , lz are the dimensions of the polymer. Using (B.5)–(B.6) and (B.10)–(B.12),
the mathematical relations for the piezoelectric polymer can be written as:
dF
= −ρAx su (B.17)
dx
∗ dξ 1 ∗
c =− F +h D (B.18)
dx Ax
∗ dξ 1
E = −h + ∗D (B.19)
dx ε
For simplicity, the subscripts have been removed from these expressions. In these
equations, s (= j ω) is the Laplace variable and Ax (where Ax = lz × ly ) is the
B.2 Theory 215
cross-sectional area perpendicular to x axis. Complex elastic constant, piezoelectric

∗ ∗ ∗
constant and dielectric constant are represented by c , h and ε respectively. Nu-
merical values of these constants are obtained from the impedance measurements,
by using non linear regression technique [9], discussed in next section.
If the current flowing through the external circuit is i, then charge q on the elec-
trodes is i/s and the electric flux density D is equal to i/(s × Ax ). The particle
displacement is related to particle velocity by ξ = u/s. From (B.12),
∗
dD d(i/s) d(h i/s)
=0⇒ =0⇒ =0 (B.20)
dx dx dx
Using (B.20) in both sides of (B.17)–(B.18), we have:
∗
d h
F − i = ρAx su (B.21)
dx s
∗
du s h
=− ∗ F − i (B.22)
dx Ax c s
∗
h 1
Vin = [u1 − u2 ] + ∗ i (B.23)
s C0 s
∗
where, Vin is the voltage at the electrical terminals of polymer and C0 is its lossy
capacitance. Equations (B.21)–(B.22) describe the mechanical behavior and (B.23)
describe the electromechanical conversion. It can be noted that (B.21)–(B.22) are
similar to the standard telegraphist’s equations of a lossy electrical transmission line
viz:
dVt
= −(Lt s + Rt )It (B.24)
dx
dIt
= −(Ct s + Gt )Vt (B.25)
dx
where, Lt , Rt , Ct and Gt are the per unit length inductance, resistance, capacitance
and conductance of the transmission line. Vt and It are the voltage across and cur-
rent passing through the transmission line. Comparing (B.21)–(B.22) with (B.24)–
(B.25), the analogy between these two sets of equations can be observed. Thus, Vt is
∗
analogous to F − (h /s) × i; Lt is analogous to ρ × Ax ; Rt is zero; It is analogous
∗
to u; and s/(Ax × c ) = Ct + Gt . The right hand side of the last expression is a
∗
complex quantity due to complex c . By substituting s = j ω and then comparing
coefficients on both sides, the expressions of Gt and Ct can be written as:
ci ω
Gt = (B.26)
+ ci2 )Ax
(cr2
cr
Ct = (B.27)
(cr2 + ci2 )Ax
Fig. B.2 Transmission line equivalent model of piezoelectric polymer
Thus, the acoustic transmission can be represented by an analogous lossy elec-

trical transmission line. It is the general practice to represent transmission losses by
taking a non-zero R and assuming G = 0. But, in acoustics the difference between R
and G is not that distinct [10] and it can be shown mathematically that the losses can
be represented by any of the following options; (a) Both R and G are used; (b) Only
G is used and R = 0; (c) Only R is used and G = 0. In the analogy presented above,
R = 0 and G = 0 and hence the losses in the transmission can be taken into account
by having a non-zero value of G. Similarly the electromechanical loss and dielec-
tric loss are considered by using complex values of h and C0 in (B.23). Complete
analogous equivalent circuit for thickness-mode piezoelectric polymer transducer
obtained by using (B.21)–(B.23), is shown in Fig. B.2. It should be noted that the
model presented here does not include the time-dependence of losses. One may re-
fer [11], for more information on viscoelastic and electromechanical energy losses,
over a period of time.
B.3 Measurement of Complex Constants
The dispersive dielectric properties and the internal viscoelastic losses in piezoelec-
tric polymers preclude the convenient use of the classical IEEE standard techniques
[7] for determining dielectric and piezoelectric properties. This is due to the fact
that the figure of merit, M for piezoelectric polymers is approximately 2–2.5 [1].
As mentioned in the IEEE standard, the parallel frequency, fp , cannot be measured
accurately when M < 5. Therefore, one should not apply the IEEE Standard’s equa-
tions for k33 or kt assuming that fp and fs are the frequencies of maximum and
minimum impedance magnitude, respectively.
A technique for characterizing and modeling PVDF was first proposed by Ohi-
gashi [8]. His approach is based on curve fitting the equation for input admittance
B.3 Measurement of Complex Constants 217
Table B.1 Dimensions, densities of the different samples and the electrodes used
Quantity Symbol PVDF-TrFE PVDF-TrFE* PVDF* Lead Metaniobate*
Density (kg/m3 ) ρ 1880 1880 1780 6000

Thickness (m) lx 50 × 10−6 0.408 × 10−3 0.27 × 10−3 1.55 × 10−3
Width (m) ly 7 × 10−3 – – –
Length (m) lz 7 × 10−3 – – –
Diameter (m) D – 14 × 10−3 14 × 10−3 25.2 × 10−3
Type of Electrode – Al+Cr Au Al Ag
Thickness of Electrode (µm) tm 0.08 <0.1 <0.1 <10
Area of sample (m2 ) Ax 49 × 10−6 154 × 10−6 154 × 10−6 498.7 × 10−6
* From [9]
for an unloaded piezoelectric resonator to actual admittance measurements made

over a broad frequency range that includes the fundamental half-wave resonance of
the sample. To account for the lossy dispersive dielectric properties, he introduced
the dielectric loss factor, tan δe and to account for the internal viscoelastic losses of
PVDF, he introduced the mechanical loss factor, tan δm . Although Ohigashi treated
the elastic and dielectric constants as complex, the piezoelectric constant was still
considered as real. Several other workers, including Smits [12] and Sherrit et al. [13]
have introduced methods to determine the mechanical, dielectric, and piezoelectric
loss constants. Smits [12] and Sherrit et al. [13] considered all the material param-
eters as complex, and the losses are given by the corresponding imaginary parts.
A good comparison of various methods used for finding the complex constants is
presented by Kwok et al. in [9]. In this work, the material parameters of PVDF,
P(VDF-TrFE), PZT/epoxy 1-3 composite, and lead metaniobate are calculated from
impedance data by using five methods, namely, the IEEE Std. 176 method, the meth-
ods of Smits [12] and Sherrit et al. [13], a software package “PRAP” [14], and the
nonlinear regression method of authors. Kwok et al. [9] also observed that the for
high loss materials like PVDF, the IEEE standard method gives much higher value
of k33 or kt , as compared to other methods.
The software package PRAP, which combines the methods of Smits [12] and
Sherrit et al. [13], is a simple way of measuring or generating complex coefficients,
if the geometry of the sample is known. In this work, PRAP is used to obtain var-
ious complex constants from the impedance and phase measurement of one of the
P(VDF-TrFE) samples. The dimension, density, type and thickness of metal elec-
trodes on the sample are given in Table B.1 and various complex constants obtained
by PRAP analysis are given in Table B.2. For other samples in Tables B.1 and B.2,
the data has been taken from [9]. This data was obtained with non linear regression
technique, presented in [9].
Table B.2 Material constants of PVDF-TRFE, PVDF and lead metaniobate

Constant PVDF-TrFE PVDF-TrFE* PVDF* Lead Metaniobate*
kt Real 0.202 0.262 0.127 0.334

Imaginary −0.0349 0.0037 0.0055 0.0003
D∗
c33 2
(N/m ) Real 1.088 × 1010 10.1 × 109 8.7 × 109 65.8 × 109
Imaginary 5.75 × 108 5.15 × 108 1.018 × 109 4.14 × 109
S ∗
ε33 Real 4.64 × 10−11 38.78 × 10−12 55.78 × 10−12 22.84 × 10−10
Imaginary 8.45 × 10−12 4.11 × 10−12 15.618 × 10−12 2.033 × 10−11
∗
h33 (V/m) Real 3.03 × 109 4.20 × 109 1.52 × 109 1.79 × 109
Imaginary −7.25 × 108 3.90 × 108 3.69 × 108 6.58 × 107
Q 18.78 19.60 8.54 15.87
* From [9]
Fig. B.3 Schematic of the equivalent model of piezoelectric polymer implemented with PSpice.
The model has been divided into various blocks showing the mechanical/acoustic/viscoelastic,
electromechanical/piezoelectric and electric/dielectric losses
B.4 SPICE Implementation
Following the discussion above, the piezoelectric polymer model has been im-
plemented in PSPICE circuit simulator, which is commercially available from
ORCAD. Figure B.3 shows the SPICE schematic of the equivalent circuit of
Fig. B.2. The mechanical, electromechanical and electrical loss blocks are clearly
marked in Fig. B.3. The SPICE implementation of various blocks is explained be-
low. The netlist for SPICE implementation is given at the end of this Appendix.
B.4 SPICE Implementation 219
B.4.1 Mechanical Loss Block
The analogy between (B.21)–(B.22) and (B.24)–(B.25) allows the implementation

of mechanical/viscoelastic behavior during acoustic transmission in polymer with
a lossy transmission line, which is available in the PSPICE. It can also be imple-
mented with lumped ladder arrangement of Gt , Ct , Lt , Rt . But, here the lossy trans-
mission line is preferred due to the advantages—in terms of accuracy and compu-
tation time—offered by the distributed values of Gt , Ct , Lt , Rt , used in it. Further,
the lossy transmission line in PSPICE allows the use of frequency dependent ex-
pression for Gt , which is desired according to (B.26). The frequency term in Gt is
implemented in SPICE, by using the expression SQRT (−s × s), where, s (= j ω)
is the Laplace operator. The parameters of transmission line viz: Gt , Ct and Lt are
∗
obtained by using complex elastic constant i.e. c in the analogous expression ob-
tained from the analogy between acoustic wave propagation and the lossy electrical
transmission line. For various sample parameters in Tables B.1 and B.2, the trans-
mission line parameters are given in Table B.2. While Gt and Ct are obtained from
(B.26)–(B.27); Lt is given by ρ × Ax and the length, lx , of the transmission line is
equal to the thickness of the sample. As shown in Fig. B.3, the lossy transmission
line is terminated into the acoustic impedance of the mediums on two sides of the
polymer—which is given by, Zm = ρm × νm × Ax . In this expression, ρm is the den-
sity of medium, νm is the velocity of sound in medium and Ax is the area of sample.
In this work air is present on both sides of samples, hence using, ρm = 1.184 kg/m3 ,
νm = 346 m/s and Ax from Table B.1, the acoustic impedances of front and backside
are given in Table B.3. In a multilayer transducer, the acoustic impedances on both
sides can be replaced by transmission lines having parameters (acoustic impedance
and time delay etc.) corresponding to the mediums on each side.
B.4.2 Electromechanical Loss Block
The electromechanical conversion is analogous to the transformer action. It is

implemented with the behavioral modeling of controlled sources, i.e. with the
‘ELAPLACE’ function in PSPICE. The currents passing through the controlled
∗
sources E1(mechanical) and E2(electrical); are h /s times the currents passing
∗
through V2 and V1 respectively. The gain term i.e. ‘h /s’ used in the controlled
sources E1(mechanical) and E2(electrical)—of PSPICE schematic in Fig. B.3—are
given in Table B.2. In SPICE, the current in any branch can be measured by us-
ing a zero value DC voltage source in that branch. The voltage sources V1 and V2,
shown in Fig. B.3, have zero DC values and are thus used to measure the current
∗
passing through them. The complex piezoelectric constant, h ensures the inclu-
sion of piezoelectric losses in the model. The complex number operator ‘j ’ in the
∗
expression of h is implemented in PSPICE by using the expression −s/abs(s).
220
Table B.3 Value/expression of parameters used in the SPICE equivalent model

Parameter PVDF-TrFE PVDF-TrFE* PVDF* Lead Metaniobate*
Lt (H) 0.0921 0.2894 0.274 2.993

√ √ √ √
Gt 9.8958 × 10−8 × ( − s × s) 3.271 × 10−8 × ( − s × s) 8.61 × 10−8 × ( − s × s) 1.912 × 10−9 × ( − s × s)
Ct (F) 1.8705 × 10−6 6.415 × 10−7 7.39 × 10−7 3.035 × 10−8
lx (m) 50 × 10−6 0.408 × 10−3 0.27 × 10−3 1.55 × 10−3
Zm () 0.020 0.063 0.063 0.204
C0 (F) 4.545 × 10−11 14.63 × 10−12 31.81 × 10−12 735.05 × 10−12
∗
h /s (3.03 × 109 − j 7.25 × 108 )/s (4.2 × 109 + j 3.9 × 108 )/s (1.52 × 109 + j 3.69 × 108 )/s (1.79 × 109 + j 6.58 × 107 )/s
Gain of E3 1/(s × (4.54 × 10−11 + j 8.28 × 10−12 )) 1/(s × (14.63 × 10−12 + j 1.55 × 10−12 )) 1/(s × (31.81 × 10−12 + j 8.91 × 10−12 )) 1/(s × (735 × 10−12 + j 6.54 × 10−12 ))
* From [9]
B Modeling of Piezoelectric Polymers
B.5 Experiment Versus Simulation 221
Fig. B.4 Equivalent

representation of lossy
capacitor
B.4.3 Dielectric Loss Block
This block consists of the lossy capacitance of polymer connected to the external
∗
voltage source or load. The use of complex permittivity i.e. ε , ensures the inclusion
∗
of dielectric losses. The lossy capacitor obtained by using ε , is equivalent to a
lossless capacitor, C0 = εAx / lx connected in parallel with the frequency dependent
resistor R0 = 1/(ωC0 tan δe ) as shown in Fig. B.4, where, C0 is the static lossless
capacitance of polymer. The voltage across the equivalent lossy capacitor is given
as:
Ic
Vc = (B.28)
sC0 + ωC0 tan δe
The SPICE circuit simulators allow only constant values of resistors and ca-
pacitors. To implement the lossy capacitor—which has the frequency dependent
resistor—the behavior modeling of controlled voltage source is used here. The con-
trolled source E3(lossy capacitor) is implemented with the ‘ELAPLACE’ function
of PSPICE. As per (B.28), the voltage across E3(lossy capacitor), is proportional to
the current flowing through itself and measured by the zero value DC source V3, as
shown in Fig. B.3.
The expressions of the gain terms used in the controlled source E3(lossy ca-
pacitor), are given in Table B.3. Again, ω is implemented with the expression
SQRT (−s × s). The netlist for the SPICE implementation shown in Fig. B.3—
generated with PSPICE—is given at the end of Appendix B.
B.5 Experiment Versus Simulation
B.5.1 Evaluation of Lossless Model
Since it is difficult to apply forces over wide range of frequencies (∼MHz) by

any mechanical arrangement, the piezoelectric polymer model in sensing mode was
evaluated by using the same SPICE model of the polymer in actuating mode. The
arrangement, as shown in Fig. B.5, is similar to the standard pulse-echo method,
explained in [7]. The actuating mode component of the whole arrangement pro-
duces the force needed as the input in sensing component of the SPICE model of
Fig. B.5 SPICE

implementation of the
standard Pulse-echo method
for evaluating the SPICE
model of piezoelectric
polymer
piezoelectric polymer. The performance of polymer model in sensing mode was

evaluated by comparing the simulated response with experiment results reported in
[15]. For the purpose of comparison only, the electrical impulse input to the trans-
mitting stage; the load at the output terminal of sensing stage; and various constants
were kept same as that used in [15]. The input to the actuating stage is a voltage
impulse (300 V, fall time 100 ns) generated by Rin = 100 and Cin = 2 nF. The
output of the sensing stage is terminated in to a load, comprising of a resistance
(Rout = 100 ) and inductance (Lout = 4.7 µH)—connected in parallel. The losses
could not be considered in this comparative study as the constants used in [15] are
real numbers. Figure B.6 shows that the simulated outputs of the polymer, both in
sensing and actuating mode, are in good agreement with the experiment results,
presented in [15].
B.5.2 Evaluation of Lossy Model
The SPICE model presented in earlier section has been evaluated by comparing the
simulated impedance and phase plots with the corresponding plots obtained from
the measured data of PVDF, P(VDF-TrFE) and lead metaniobate. As mentioned
earlier, the approximate lossy models developed for piezoceramic transducers are
insufficient for evaluating the behavior of piezoelectric polymers. Keeping this in
view, a comparison has also been made with the impedance and phase plots obtained
by Püttmer’s approach for lossy piezoceramics [5]. Using the PSPICE schematic of
Fig. B.3, the simulated impedance, Zin , for all the samples is obtained by dividing
Fig. B.6 (top-left) Force generated by piezoelectric material (actuating component) in the
pulse-echo arrangement, reproduced from [15]; (top-right) Voltage output of piezoelectric material
(sensing component) in the pulse-echo arrangement, reproduced from [15]; (bottom-left) Simulated
force obtained from SPICE model of piezoelectric material (actuating component) (bottom-right)
Simulated voltage obtained from SPICE model of piezoelectric material (sensing component)
the voltage at node E (Fig. B.3) with the current passing through this node. For
these simulations, the transmission line was terminated into the acoustic impedance
of air on both sides and the effect of electrodes present on both sides of polymer
was assumed to be negligible due to their negligible small thickness in comparison
to that of test samples.
In the case of 50 µm thick P(VDF-TrFE) polymer film, the impedance and phase
measurements were obtained with HP4285 LCR meter. The measured impedance
and phase plots for other samples, viz: PVDF, P(VDF-TrFE) (0.408 mm thick) and
lead metaniobate, used here, are same as those used in [9]. The physical dimensions,
calculated complex constants and various parameters of SPICE model for all these
samples are given in Tables B.1–B.2.
The impedance and phase plots of lead metaniobate sample, obtained by the
SPICE model presented in this work has been compared in Fig. B.7 with those
obtained by Püttmer’s approach [5] and measured data. The lead metaniobate sam-
Fig. B.7 Comparison of Impedance (left) and Phase (right) plots of lead metaniobate sample, with
the corresponding plots obtained by the Püttmer’s approach and from the measured impedance data
of [9]
ple is chosen here because the model in [5], was evaluated for lead metaniobate.
The netlist of the SPICE model using Püttmer’s approach is given in Appendix B.
Püttmer’s approach considers the losses in the transmission line only and dielec-
tric and piezoelectric losses are assumed to be negligible. Other difference between
Püttmer’s approach and the work presented here is the use of Rt and Gt respectively,
to represent the losses in the transmission line. In Püttmer’s approach, Lt is given by
ρ × Ax ; Ct is given by 1/(cr × Ax ) and a simplified expression of Rt was obtained
by assuming the transmission line to be having low loss. The value of Rt was given
by (Lt /Q) × SQRT (−s × s), where Q is the mechanical quality factor obtained
by dividing real part of elastic constant with its imaginary part. It can be observed
from Fig. B.7, that the impedance and phase plots of piezoceramics—obtained with
the SPICE model presented in this work—are in good agreement with the measured
plots and the plots obtained Püttmer’s approach presented in [5]. The agreement be-
tween these plots is in line with the observations on the equivalence of using Rt or
Gt to represent losses in the acoustic transmission [10].
The impedance and phase plots of two different P(VDF-TrFE) samples, obtained
by the SPICE model presented in this work has been compared in Fig. B.8 and
Fig. B.9 with those obtained by Püttmer’s approach [5] and from the measured data.
The plots in Fig. B.8 correspond to the thin 50 µm film and those in Fig. B.9 corre-
spond to that of 0.408 mm thick film.
An additional plot obtained by considering the transmission losses only, in the
SPICE model presented in this work, is also shown in Fig. B.8. It can be observed
that the plots obtained with the approach presented here are in good agreement with
the measured data. The difference between the approach presented here and that of
Püttmer’s approach is more evident in case of thinner polymer film (Fig. B.8) and
more so in case of phase plots of both samples. Thus, in case of polymers, Püttmer’s
approach leads to a significant deviation from the measured plots.
The comparison of the impedance and phase plots for PVDF is given in Fig. B.10.
In case of PVDF, it can be observed that—although an improvement over past
Fig. B.8 Comparison of Impedance (left) and Phase (right) plots of 50 µm thick PVDF-TrFE
polymer film with the corresponding plots obtained by the Püttmer’s approach and from the mea-
sured impedance data. The plots obtained by taking the transmission loss only in the SPICE model
presented here, has also been shown. In principle, taking only transmission loss only in the model
presented here is same as that of Püttmer’s work
Fig. B.9 Comparison of Impedance (left) and Phase (right) plots of 0.408 mm thick PVDF-TrFE
film with the corresponding plots obtained by the Püttmer’s approach and from the measured
impedance data of [9]
approach—the plots obtained with the approach presented in this work still have
some discrepancies with the plots obtained from the measured data. The discrep-
ancies are especially higher in the region outside resonance. This can be attributed
to the frequency dependence of material parameters [9], which were assumed to be
independent of frequency for these simulations. For a piezoelectric transducer the
Fig. B.10 Comparison of Impedance (left) and Phase (right) plots PVDF film with the corre-
sponding plots obtained by the Püttmer’s approach and from the measured impedance data of [9].
S
The plots obtained by using first order frequency dependence of ε33 and tan δe are also shown
electrical impedance at thickness extensional resonance is given by:

tan(πf l D∗ )
ρ/c33
lx 2∗
x
Z(f ) = S∗ A
1 − kt (B.29)
i2πf ε33 x πf lx ρ/c33 D∗
where, f is the frequency. It can be noticed from (B.29) that outside resonance the
impedance depends on ε33 S ∗ and hence on the ε S and tan δ and hence as a first ap-
33 e
proximation, taking into account the frequency dependence of ε33 S and tan δ should
e
improve the simulation results. Following Kwok’s approach, presented in [9], we
S and tan δ at various frequencies of the measured impedance
obtained values of ε33 e
data. It was observed that these constants vary with frequency. Considering the data
S and tan δ at frequencies outside resonance, a first order frequency de-
values of ε33 e
pendence of ε33S and tan δ was obtained by linear fit of the ε S and tan δ versus
e 33 e
frequency plots. Using these frequency dependent constants in the SPICE model,
the impedance and phase plots were obtained. As shown in Fig. B.10, this improves
the match between the simulated plots and plots from the measured data. The match-
ing can be further improved—albeit, at the cost of computation power—by using a
higher order frequency dependence of ε33 S and tan δ and also the frequency depen-
e
∗
dence of other material constants viz: c33 and kt∗ .
D
B.6 Relative Contribution of Various Losses

Having presented the model with all the losses, the next question that comes to mind
is—what is their relative contribution? If the contribution of a particular loss compo-
nent far outweighs others, then such an argument can be a basis for using a simplified
model. To evaluate the role of various loss components, the SPICE model was used
to simulate impedance and phase under various combinations of losses. Impedance
B.7 Design Issues Associated with Piezoelectric Polymer Film 227
Fig. B.11 Impedance (left) and Phase (right) plots comparing the relative contribution of various
losses. Simulations were performed under assumption of different combinations of losses
and phase plots so obtained are shown in Fig. B.11. It can be noticed that inclusion
of only acoustic/transmission losses gives a first approximation of the impedance
and phase, which is further improved by the dielectric and piezoelectric losses.
B.7 Design Issues Associated with Piezoelectric Polymer Film

The model developed for piezoelectric polymers can be used to study the effect of
various parameters used in the model, which ultimately helps in optimizing their
response. Various factors that influence the response of polymer are its thickness,
material on front and back of the polymer, area of the polymer film, and type of
electrodes [1]. The open circuit voltage of a piezoelectric polymer, when a step
force is applied on top is:

hzz × Fin × (1 − Rfp ) Z Z
Vpolymer = − t − (1 + Rpb ) × t − ×u t −
Zap v v

2Z 2Z
+ Rpb × (1 + Rfp ) × t − ×u t −
v v
− Rf b × Rpb × (1 + Rpb )

3Z 3Z
× t− ×u t − + ... (B.30)
v v
where, Rfp = (Zp − Zf )/(Zp + Zf ) represent the reflection coefficient at front-
polymer interface and Rpb = (Zp − Zb )/(Zp + Zb ) at polymer-back faces. Zp ,
Zf , and Zb are the acoustic impedances of polymer, and the material on the front
and back (silicon in the present case) sides respectively. Z is the thickness of the
polymer, v is the wave velocity in polymer and u(·) is the step function. It can be
noted from (7.2) that the open circuit voltage of the polymer depends on the thick-
ness of the polymer, the reflection coefficients at the two faces, internal parameters
Fig. B.12 Simulated open

circuit voltage of 25, 50 and
100 µm thick polymer film.
A step force of 0.1 N was
applied
like piezoelectric constant and elastic constant (acoustic impedance). In addition to

these, the capacitance of polymer is also involved when the polymer is connected
with a load.
In (B.30), the term outside brackets represents the net mechanical to electrical
conversion and the terms inside the bracket are combination of incident and reflected
force waves. It can be observed from the first term inside brackets that the voltage is
a ramp function. If the thickness of the polymer is increased, the step function terms
are further delayed (delay time increases as Z is increased). Thus the open circuit
voltage keeps on increasing linearly until first delay term contributes to the net volt-
age. Thus, the open circuit voltage in general increases with the thickness of the
polymer film. This thickness effect of polymer film can also be noted from the sim-
ulation result shown in Fig. B.12. Apart from above discussed factors, the response
of polymer also depends on its area, which depends on sensor configuration.
The internal properties like piezoelectric constant depend on the way the polymer
is made. Uniform thin films of piezoelectric polymers like P(VDF-TrFE), can be
obtained by spin coating the polymer solution and annealing it at around 100° [16].
In order to have piezoelectric properties, the piezoelectric polymer needs to be poled
at a rate of (∼100 V/µm). The intrinsic properties of polymers depend on the way
the films are processed.
The materials on the front and back of the polymer offer different acoustic
impedances to the force waves. The response of polymer with different materials
under polymer, simulated with the SPICE model, is shown in Fig. B.13. It can be
noticed that the polymer has higher sensitivity when a stiff material is used on the
back side. For stiffer materials, the reflection coefficient Rpb is high and the device
operates in λ/4 mode. For piezoelectric polymer based sensors realized on silicon
wafers, the back side is made of silicon whose acoustic impedance is approximately
five times that of P(VDF-TrFE). Thus, the reflection coefficient, Rpb , is 0.66. This
means that piezoelectric polymer based sensors having silicon under the polymer,
have high sensitivity.
B.8 SPICE Netlist of Piezo-Polymer Model 229
Fig. B.13 Response of

polymer with different
backside materials. Acoustic
impedance of steel, silicon,
PVDF, teflon and wood are 9,
3.8, 0.8, 0.6 and 0.02 MRayl
respectively. A step force of
0.01 N was applied
B.8 SPICE Netlist of Piezo-Polymer Model

* source LOSSY PIEZO POLYMER
E_E3(Lossy_Capacitor) 3 4 LAPLACE{I(V_V3)} = 1/(s × (C0 − C0 × tan(δe ) ×
(s/abs(s))))
R_Acoustic_Imp_back 0 BACK Zm
T_Acoustic_Transmission_Line BACK 1 FRONT 1 LEN=lx R=0 L=Lt G=Gt
C=Ct
V_V3 4 5 DC 0 AC 0 0
V_V2 E 3 DC 0 AC 0 0
∗
E_E2(electrical) 5 0 LAPLACE I(V_V1) = h /s
V_V1 1 2 DC 0 AC 0 0
V_Vin S 0 DC 0Vdc AC 1Vac
R_Acoustic_Imp_Front 0 FRONT Zm
∗
E_E1(mechanical) 2 0 LAPLACE{I(V_V2)} = h /s
R_R 5 E 1n
* source PIEZO LOSSY MODEL OF PUTTMER
V_V1 S 0 DC 0Vdc AC 1Vac
V_V2 E 3 DC 0 AC 0 0
T_T1 Back 1 Front 1 LEN=lx R=(Lt /Q) ∗ SQRT (−1 ∗ (s) ∗ (s)) L=Lt G=0
C=Ct
E_E1 2 0 LAPLACE I(V_V2) = h/s
R_Acoustic_Imp_back 0 BACK Zm
X_F1 1 2 0 3 SCHEMATIC1_F1
R_Acoustic_Imp_Front 0 FRONT Zm
C_C0 0 3 C0
R_R S E 1n
.subckt SCHEMATIC1_F1 1 2 3 4
F_F1 3 4 VF_F1 h × C0
VF_F1 1 2 0V
.ends SCHEMATIC1_F1
B.9 Summary
The SPICE model for piezoelectric polymers presented in this chapter, for the first
time, incorporates all losses namely: viscoelastic, piezoelectric and dielectric. It has
been shown that, the model provides a good match between simulated and measured
data and that the past approaches, developed mainly for piezoceramics, are not suit-
able for piezoelectric polymers. By comparing the contribution of various losses
in the model, the viscoelastic losses have been found to have a major role. Per-
haps, this is one reason why past approaches tried to model the viscoelastic losses
only. The discussion on design issues associated with piezoelectric polymers, when
they are used as transducer, provide good insight into their behavior under various
conditions. Around low frequencies (∼1 kHz), such a model can be approximated
with a capacitor in series with a voltage source, as discussed in following chapters.
Thus, the model presented in this chapter can be used to evaluate the performance
of POSFET based tactile sensing devices, discussed in Chaps. 7 and 8, over a wide
range of frequencies and can be used to explore the utility of such devices to ap-
plications other than tactile sensing as well. The successful implementation of the
transducer model in SPICE, has made it convenient to evaluate the performance of
transducer, both, in time and frequency domains. This implementation will greatly
help in designing and evaluating the sensor system i.e., transducer and conditioning
electronics, all together.
References
1. L.F. Brown, Design consideration for piezoelectric polymer ultrasound transducers. IEEE
Trans. Ultrason. Ferroelectr. Freq. Control 47, 1377–1396 (2000)
2. M. Redwood, Transient performance of piezoelectric transducers. J. Acoust. Soc. Am. 33,
527–536 (1961)
3. S.A. Morris, C.G. Hutchens, Implementation of Mason’s model on circuit analysis programs.
IEEE Trans. Ultrason. Ferroelectr. Freq. Control 33, 295–298 (1986)
4. W.M. Leach, Controlled-source analogous circuits and SPICE models for piezoelectric trans-
ducers. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 41, 60–66 (1994)
5. A. Püttmer, P. Hauptmann, R. Lucklum, O. Krause, B. Henning, SPICE models for lossy
piezoceramics transducers. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 44, 60–66 (1997)
6. R.S. Dahiya, M. Valle, L. Lorenzelli, SPICE model of lossy piezoelectric polymers. IEEE
Trans. Ultrason. Ferroelectr. Freq. Control 56(2), 387–396 (2009)
7. ANSI/IEEE, IEEE standard on piezoelectricity. IEEE Standard 176-1987 (1987)
8. H. Ohigashi, Electromechanical properties of polarized polyvinylidene fluoride films as stud-
ied by the piezoelectric resonance method. J. Appl. Phys. 47, 949–955 (1976)
9. K.W. Kwok, H.L.W. Chan, C.L. Choy, Evaluation of the material parameters of piezoelectric
materials by various methods. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 44, 733–742
(1997)
References 231
10. J. Wu, G. Du, Analogy between the one-dimensional acoustic waveguide and the electrical
transmission line for cases with loss. J. Acoust. Soc. Am. 100, 3973–3975 (1996)
11. A.M. Vinogradov, V.H. Schmidt, G.F. Tuthill, G.W. Bohannan, Damping and electromechani-
cal energy losses in the piezoelectric polymer PVDF. Mech. Mater. 36, 1007–1016 (2004)
12. J.G. Smits, Iterative method for accurate determination of the real and imaginary parts of the
materials coefficients of piezoelectric ceramics. IEEE Trans. Sonics Ultrason. SU-23, 393–402
(1976)
13. S. Sherrit, H.D. Wiederick, B.K. Mukherjee, Non-iterative evaluation of the real and imaginary
material constants of piezoelectric resonators. Ferroelectrics 134, 111–119 (1992)
14. TASI Technical Software Inc. Ontario, Canada (2007). Available at: http://www.
tasitechnical.com/
15. G. Hayward, M.N. Jackson, Discrete-time modeling of the thickness mode piezoelectric trans-
ducer. IEEE Trans. Sonics Ultrason. SU-31(3), 137–166 (1984)
16. R.S. Dahiya, M. Valle, G. Metta, L. Lorenzelli, S. Pedrotti, Deposition processing and charac-
terization of PVDF-TrFE thin films for sensing applications, in IEEE Sensors: The 7th Inter-
national Conference on Sensors, Lecce, Italy (2008), pp. 490–493
Appendix C
Design of Charge/Voltage Amplifiers
Abstract The designs of three stage charge and voltage amplifiers, to read the
taxels on piezoelectric polymer—MEA based tactile sensing chip, are presented
here. Piezoelectric polymers is approximately represented as a voltage source in
series with the static capacitance of the polymer. Due to the presence of capaci-
tance, piezoelectric polymers have very high output impedance. Thus, to measure
the charge/voltage generated due to applied forces, an amplifier with very high input
impedance (e.g. CMOS input based) is required.
C.1 Charge Amplifier

The independence of amplifier output from the cable capacitances and the mini-
mization of the charge leakage through the stray capacitance around the sensor,
make charge amplifier a preferred choice especially when long connecting cables
are present. The schematic of the charge amplifier is given in Fig. C.1. The am-
plifier consists of three stages. The differential charge amplifier in the first stage is
followed by differential to single ended amplifier, which is then followed by a sec-
ond order Sallen–Key low pass filter and a low pass RC filter. In the first stage FET
input based OPA627 operational amplifiers are used. These are low noise op-amps
with typical input bias current of 2 pA. In the second and third stage TL082 op-amp
is used. The gain of second stage is 5 and that of third stage is 4. Overall gain of the
amplifier is set by value of feedback capacitances in the first stage with respect to
the static polymer capacitance. The circuit has been designed to have a frequency
range of 2 Hz–200 kHz.
C.2 Voltage Amplifier

The voltage amplifiers are preferred if the ambient temperature variation is high,
since they exhibit less temperature dependence and hence useful for measuring
the piezoelectric polymer film response. This is due to the fact that piezoelectric
polymer’s voltage sensitivity (g-constant) variation over temperature is smaller than
the charge sensitivity (d-constant) variation. The schematic of voltage amplifier is

234 C Design of Charge/Voltage Amplifiers
Fig. C.1 Schematic of three stage charge amplifier
Fig. C.2 Schematic of three stage voltage amplifier
shown in Fig. C.2. It also comprises of three stages. Except first stage, remaining
two stages are same as that of the charge amplifier. Since the source capacitance in
the present case is ∼1 pF, the low frequency response can be brought down to ∼5 Hz
by keeping the circuit loading effect as minimum as possible. This was obtained by
providing a resistive load of 1 G in the first stage. This value appears 3–5 times
higher to the polymer, due to the bootstrapping technique adopted by using positive
feedback. Use of resistors with resistances more than 1 G is not recommended
as higher values are prone to noise due to humidity, dust, temperature variation etc.
With the adoption of this technique the voltage amplifier has a gain of 46 dB in the
frequency range of 8 Hz–200 kHz.
Index
A Anisotropic, 200
Absolute capacitance, 86 Annealing, 163, 164, 228
Acceleration, 126 Anthropomorphic, 83, 98
Accelerometer, 196 Anti-aliasing, 63
Access time, 58–60, 62 Anticipatory control, 30, 31
Accuracy, 24 Application specific integrated circuit (ASIC),
Acoustic impedance, 96, 156, 219, 223, 227, 71
228 Artificial brain map, 69
Acoustic loss, 212 Artificial intelligence (AI), 70
Acoustic transmission, 219 Artificial limbs, 3, 10
Acoustic wave propagation, 219 Aspect ratio, 159, 188, 190
Acquisition time, 64, 190 Assistive robots, 80
Action for perception, 6, 15 Attention, 26
Action potential, 22, 26, 64 Audio, 25, 31, 71
Active device, 62 Autonomous learning, 80
Active perception, 19 Autonomous robot, 61
Active pixel, 59, 60
Active pixel sensor (APS), 59
B
Active taxel, 59, 60, 66
Band gap, 110
Active touch, 14, 29
Active transducer, 62 Band pass filter, 63
Actuator, 72, 96 Bandwidth, 52, 58–60, 64, 67, 68, 97, 117, 155
Acuity, 20, 24 Barrier width, 111
Adaptation rate, 22 Bayes tree, 69
Addressing, 58, 63, 64 Behavioral modeling, 219
Addressing scheme, 54 Bendable, 47
Algorithm, 49, 65, 67–70, 73 Bendable POSFET, 191
Algorithms, 11 Bendable tactile sensing chip, 191
Amorphous silicon, 118 Biasing, 61, 62, 190
Amplification, 65, 113 Bio-robots, 80
Amplifier, 35, 60, 63, 117, 171, 233 Biomedical, 9
Amplitude modulation, 86 Biomedical robotics, 9
Analog filter, 54 Biometric, 50
Analog resistive sensing technology, 82 Bit rate (BR), 64
Analog resistive touch sensing, 81–83 Body schema, 30, 31
Analog sensors frontend, 118 Bootstrapping, 234
Analog to digital (A/D), 26, 60–64, 66, 71, 124 Bottom–up method, 192

236 Index
C Computing power, 74
Calibration, 65 Conductance, 215
CAN, 67, 68, 71, 105 Conductive elastomer, 83
Cantilever, 112 Conductive epoxy, 114
Capacitance, 86, 104, 147, 211, 215 Conductive fiber, 83, 103, 108
Capacitance to digital converter, 89 Conductive filler, 102
Capacitance-to-digital converter, 125 Conductive fluid, 109
Capacitive, 16, 54, 57, 58, 60, 62, 63, 81, Conductive gel, 83, 101
85–87, 89, 99, 112 Conductive graphite, 105
Capacitor, 85, 198 Conductive polymer, 83, 84
Carbon black, 102, 106, 129 Conductive polymer composite (CPC), 107
Carbon black nanoparticle filler, 103 Conductive polymer composites (CPC), 102,
Carbon fiber, 102 103
Carbon microcoil filler, 103 Conductive polymer nanoparticle filler, 103
Carbon microcoils (CMCs), 104 Conductive rubber, 83
Carbon nano tube (CNT), 51, 129 Conductive yarn, 83, 103
Carbon nanotube (CNT), 103 Conformability, 79, 101
Carbon nanotubes (CNT), 106, 107 Conformable, 25, 29, 35, 36, 47, 48, 53, 92,
Carrier frequency, 64 123
CCD, 91, 111 Contact force, 57, 81
Cellular rubber, 105 Contact location, 82
Central nervous system (CNS), 13, 15, 26, 27, Contact point estimation, 46
36 Contact pressure, 81
Cerebral cortex, 26 Control, 36, 44, 46
Channel current, 120, 122 Control bandwidth, 58, 60
Channel length, 160, 162, 190 Control loop, 34, 60
Channel width, 160, 162, 190 Controlled source, 211, 219, 221
Charge amplifier, 146, 147, 233, 234 Controller, 65, 67, 74
Charge carrier mobility, 122, 162 Corona poling, 166
Charge coupled device (CCD), 54, 73 Cortical neurons, 27
Charge neutrality, 154 Coupling factor, 214
Charge sensitivity, 233 Creep, 90, 105, 107, 124, 125
Chip-on-flex, 191 Cross talk, 107
Chirping, 93 Cross-talk, 24, 60, 143, 148, 182–184
Classification, 15, 151 Curie temperature, 166, 198
CMOS, 8, 91, 161, 188, 190, 191 Current mirror, 190
Coercive field, 157 Current sink, 188
Cognition, 25, 63, 69 Curve fitting, 216
Common-source, 158, 171, 173, 189, 193 Cut-off frequency, 149
Communication, 66, 187 Cutaneous, 5, 13–16, 19–21, 23, 24, 27,
Communication bandwidth, 64 29–31, 53
Communication bus, 66–68, 71, 124 Cutaneous perception, 14
Communication interface, 73, 126 Cutaneous system, 13
Communication port, 125
Compass method, 24 D
Compensation, 65 Data acquisition, 59, 61, 63, 64
Complementary metal oxide semiconductor Data processing, 11, 64
(CMOS), 59, 60, 66, 233 Data rejection, 68
Compliance, 5, 6, 8, 16, 29, 35, 49 Data representation, 69
Compliance coefficient, 203 Data selection, 68
Composite, 104–106 Data transfer, 34, 58
Computational intelligence, 69, 151 Data transmission, 53, 66, 68
Computer aided design (CAD), 48, 74 Data wire, 54
Computer aided engineering (CAE), 74 Decimation filter, 64
Index 237
Depolarization, 163 Electro-optic (EO), 110

Dermis, 20, 21, 32, 121 Electrochemical, 99
Design hints, 33 Electrode poling, 166, 167
Detector, 90, 93 Electroluminescence, 92, 110
Dexterous manipulation, 7, 15, 30, 44, 46, 49 Electroluminescent, 111
Diaphragm, 112, 114 Electromagnetic, 93
Dielectric, 195–198, 205, 208 Electromechanical conversion, 219
Dielectric constant, 85, 120, 154, 161, 169, Electromechanical coupling constant, 214
204–207, 212, 215, 217 Electromechanical coupling factor, 207
Dielectric effect, 204 Electromechanical loss, 212, 214, 216
Dielectric loss, 212, 214, 216, 221, 224, 227, Electron beam poling, 166
230 Electron tunneling, 110
Dielectric loss factor, 217 Electronic device, 116
Dielectric permittivity, 204 Electronic polarization, 196
Dielectric susceptibility, 204 Electronic skin, 35, 37, 57, 192
Digital converter chip, 63 Electrooptical, 92
Digital counter, 64 Electrophoresis, 164
Digital signal processing (DSP), 64, 71 Electrorheologic effect, 98
Digital to analog (D/A), 61 Electrorheological (ER), 98
Digitalization, 63 Electrospinning, 164
Digitization, 58 Embedded data processing, 62, 64–66, 71, 72
Diode, 116, 140, 178, 185 Embedded electronics, 74
Dipole, 196, 199 Embedded system, 66
Dipole moment, 101, 196, 198 Embroidery, 108
Direct piezoelectric effect, 199, 203 Emitter, 90, 93
Discrete resistive touch sensing, 81 Encapsulation, 129
Displays, 85 Encoding, 25, 26, 30
Distributed computing, 65 Endoscopy, 97
Distributed tactile sensing, 177 Energy barrier, 111
Dorsal column medial lemniscal pathway, 26 Energy budget, 71
Drain, 108 Energy harvesting, 71, 200
Drain current, 122 Entropy, 109
Duty cycle, 86 Epidermis, 20, 21, 29, 32, 102, 121
Dymaxion map, 48 Epoxy, 157–159
Dynamic force, 118 Equivalent circuit, 142, 216, 218
Dynamic range, 60, 97 Equivalent model, 142
Dynamic response, 96 Etching, 163, 165, 191
Dynamic sensing, 89 Ethernet, 67
Exploration, 4–8, 14, 15, 19, 28, 36, 44–46, 80
E Extended gate, 116–118, 120, 139, 141–143,
Edge detection, 29 153, 157–159, 193
Effective bit rate (EBR), 64 Extrinsic optical sensor, 90, 92
Elastic compliance, 212 Extrinsic sensing, 15, 16
Elastic constant, 161, 204, 215, 228 Extrinsic touch sensor, 70
Elastic cover, 35, 49 Eye-in-hand, 8
Elastomer, 46, 129, 192
Elastomer matrix, 102 F
Elastoresistance, 81, 83 Fabrication, 153, 161, 163, 172, 177, 179,
Elastoresistivity, 83 191–193
Electric displacement, 119, 120, 154, 204, Fabrics, 51
212–214 Fast adapting (FA), 22, 25, 28
Electrical equivalent, 158, 169 Fatigue, 65
Electrical equivalent model, 146 Fault tolerance, 36, 72, 73
Electrical impedance tomography (EIT), 51, 84 Feature extraction, 19, 65, 67–69
238 Index
Feedback, 68 Grasp control, 15, 46

Feedforward, 31 Grasp stability, 6, 7, 30, 46, 69
FeRAM, 119, 120, 154 Grasping, 6–8, 10, 89
Ferroelectret, 118
Ferroelectric, 81, 99, 157, 195, 196, 198 H
Ferromagnetic, 198 Half wave resonance, 217
Fiber Bragg grating (FBG), 92 Half-bridge, 113
Field effect transistor (FET), 62, 189 Hall effect, 93
Field effect transistors (FET), 139, 141–143 Hall voltage, 94
Figure of merit, 216 Haptic display, 99
Filtering, 35, 64 Haptic interface, 99
Fingerprint, 29, 32, 50 Haptics, 5–7, 13–15, 31
Finite element modeling, 32, 49 Hardness, 6, 29, 30, 33, 46, 69, 96, 109, 115,
Firing rate, 28 118, 150, 151, 179
Firmware, 65 Hardware, 73, 129
Flexibility, 72, 102 Hardware requirement, 47
Flexible, 16, 36, 47, 48, 53 HDMS, 165
Flexible electronics, 191 Heterogeneous integration, 115
Flexible MEMS, 114 HEX-O-SKIN, 126
Flexible printed circuit board, 114, 124, 125, Hierarchical functional diagram, 56
129 Hierarchical structural diagram, 56
Flexible printed circuit board (PCB), 48, 55 Human machine interface, 85, 89
Flexible tactile sensing chip, 191 Human robot interaction (HRI), 73
FlexRay, 67, 68 Human sense of touch, 19, 20, 36, 43, 44, 58
Floating gate, 153, 171, 180 Human–robot interaction, 3, 6, 9, 10
Force, 34, 44, 65 Human–robot interaction (HRI), 69, 74
Force distribution, 57 Humanoid, 3, 8, 9
Force perception, 30 Humanoids, 80
Force sensing, 14 Humidity, 36, 107
Force sensing resistors (FSR), 10, 51, 83 Hybrid resistive tactile sensing, 82, 83
Force sensors, 16 Hydrophone, 196
Force/torque (F/T) sensor, 58 Hysteresis, 36, 84, 89, 90, 94, 101, 107, 110,
Force/torque sensors, 8 114, 124, 125, 157, 167, 198
Free charge, 200, 204
Frequency modulation, 86 I
Frequency response, 58 I2C, 67, 68
Friction cone, 46 II–VI semiconductor, 110
Frustrated total internal reflection, 91 Image, 60
Fuel cell, 71 Image processing, 51
Fuller map, 48 Imaging, 84
Functional block diagram, 62 Impedance, 70
Fuzzy sets, 70 Impedance adaptation, 63, 190
Impedance matching, 188
G Implant, 162
Gain, 58, 182, 189 In situ poling, 166, 167
Gain plot, 148, 172 Incipient slip, 173
Gain-phase plot, 189 Induced channel, 155
Gate, 108 Inductance, 104, 215
Gauge factor, 113, 116 Inductive, 16, 68
Glabrous skin, 20, 22 Information transfer, 25, 26
Graphite, 102 Inorganic semiconductor, 140
Graphite filler, 105, 106 Input characteristics, 179
Graphite nanosheet filler, 103 Integral sensing unit, 153
Grasp, 22, 30, 31, 68 Integral sensor unit, 192
Index 239
Integrated circuit, 192 Lithium polymethacrylate (LiPMA), 98

Integrated circuit (IC), 66, 89, 119 Lithography, 162
Integrated circuits, 140 Load cell, 145
Integrated read-out, 177 Local computation, 35, 55
Integrated smart sensor, 166 Local computing, 64
Integrated system, 139 Local data processing, 66
Integrated temperature sensor, 177, 178 Local memory, 60
Intelligent routing, 54 Local processing, 65, 187, 192, 193
Intelligent textile, 103, 108 Localization, 70
Intelligent yarn, 108 Longitudinal, 97
Interconnect, 36, 50, 53, 55, 63, 68, 116, 159, Loss factor, 217
174 Loss of light, 93
Interdigitated, 104 Loss tangent, 214
Interdigitated gate, 188 Lossy capacitance, 215
Interdigitated structure, 88 Lossy capacitor, 221
Interface circuit, 190 Lossy transmission line, 215, 216, 219
Interface electronics, 57, 58, 61–63, 65, 66, 71, Low pass filter, 63, 101, 233
72, 129 LPCVD, 162
Interfacial space charge polarization, 196 LTO, 163
Interference, 93
Intermediate ridge, 21, 24, 29, 32, 35, 121 M
Intrinsic conductive polymers (ICP), 103 Machine learning, 69, 70
Intrinsic optical sensor, 91 Magnetic, 16, 81, 93, 99
Intrinsic optical tactile sensing, 92 Magnetic coupling, 93, 94
Intrinsic sensing, 15, 16 Magnetic flux, 93, 94
Intrinsic touch sensor, 71 Magneto-electric, 16
Ionic polarization, 196, 197 Magnetoelastic, 94
Ionization, 109 Magnetoresistance, 93
IRLED, 126 Magnetorheological (MR) effect, 99
ISFET, 140 Magnetorheological (MR) fluids, 99
Maintainability, 74
K Maintenance, 74
Kinematics, 7 Manipulation, 6–8, 10, 28–30, 34–36, 45, 46,
Kinesthetic, 5, 13–15, 19, 20, 28–30, 34, 70 79, 80, 171
Kinesthetic perception, 14 Manufacturability, 74
Kinesthetic system, 13 Material classification, 70
Matrix, 83, 96, 104
L Mechanical amplifier, 32, 35
Large area skin, 92 Mechanical impedance, 170
Large scale integration (LSI), 54 Mechanoreceptor, 43, 65, 161, 174
Large-area sensors, 101 Mechanoreceptors, 4, 20–22, 24–26, 28, 29,
Large-area skin, 123 34, 35
LED, 124 Medical robots, 80
Lever arm mechanism, 32 Meissner’s corpuscles, 20, 22
Light coupling, 93 Melting point, 193
Light emitting diode (LED), 90 MEMS, 83
Light spectrum, 110 Merkel cells, 20–22, 29, 32
Light touch, 20, 61 Metal nanoparticle filler, 103
Linear theory of piezoelectricity, 203, 212 Metal oxide semiconductor (MOS), 153, 154,
Linearity, 60, 63 157–161, 163, 165, 167, 169–172, 179,
Linearization, 65 188
Liquid crystal display (LCD), 111 MFMIS, 119, 154
Liquid crystal (LC), 107 Micro bending, 93
Liquid crystalline (LC), 98 Micro tactile element, 105
240 Index
Micro-/nano-structures, 35 Nociceptors, 20
Micro-/nanowires, 35 Nocioceptor, 43
Micro-lever action, 29, 32 Non invasive, 84
Microbending, 93 Non linear regression, 217
Microcontroller, 64, 67, 68, 71, 73, 105, 124, Non-polar dielectrics, 198
126 Non-volatile memory, 196
Microelectrode array (MEA), 233 Nonferromagnetic, 99
Microelectrode (MEA), 139, 141, 143–148, Nonlinear regression, 215
151 Normal force, 86
Microelectromechanical system (MEMS), 66, Normal strain, 112
85, 88, 89, 112–116 Normal stress, 112, 113
Microelectronics, 54, 114 Nyquist rate, 59, 60, 64
Microfluidics, 51
Micromachining, 97, 112, 114–116 O
Microphone, 196 Object identification, 151
Microprocessor, 55, 65 Object recognition, 15, 19, 29, 49, 69, 79, 182
Microstructure, 196 Off-the-shelf, 123
Microstructures, 24 On-chip electronics, 158, 188, 189, 193
Miniaturization, 33, 47, 72, 139, 140, 188 On-chip read-out, 190
Minimal invasive surgery, 6, 9 On-chip signal conditioning, 188
Minimally invasive surgery, 97 Operation amplifier, 233
Mobile robots, 45 Optical, 16, 81, 90, 99, 110, 123
Model, 219, 221, 227, 230 Optical arrays, 191
Modeling, 15, 169, 211 Optical fiber, 90, 91, 93
Modular, 36, 72, 73 Optical imager, 66
Modularity, 72 Optical tactile sensor, 93
Modulating frequency, 64 Optical waveguide, 91
Module, 48, 49, 53, 65, 72, 73 Optimization, 52
Molecular model, 198 Organic field effect transistor (OFET), 58, 59,
MOS, 117, 119 120
MOSFET, 117, 140, 141, 153, 155 Organic semiconductor, 140
Motion planning, 69, 126 Organic transistor, 116
Motion strategies, 4 Orientation grating method, 24
Motor control, 14, 19, 30 Orientation polarization, 196
Moving average, 64 Oscillator, 64, 195
Multi-walled carbon nanotube (MWNT), 106 Output buffer, 188, 190
Multifunctional, 33 Oversampling, 59, 60, 64
Multimodal, 57, 70, 115, 126
Multiplexer, 63, 105 P
Multiplexing, 55, 65 Pacinian corpuscles, 20–22, 29, 32
Multiplexor, 190 Packaging, 113, 129, 148, 193
Mutual capacitance, 86–88 Pain, 5, 20
Palpation, 9
N Papillary ridge, 29, 32, 35
Nanocomposite, 106, 107 Paraelectric, 198
Nanoparticle, 92, 111 Parallel processing, 15
Nanoparticles, 110 Parasitic capacitance, 86
Nervous system, 65 Particle filter technique, 68
Network, 67 Passivation, 143
Networking, 74 Passive device, 62
Neural network, 69, 70 Passive perception, 19
Neural pathways, 26 Passive pixel, 59
Neuromorphic, 66 Passive touch, 14, 29
Newtonian fluid, 98 Passive transducer, 62
Index 241
PDMS, 49, 51, 88, 89, 107, 116, 121, 171, Polarization, 98, 101, 122, 154, 155, 168, 179,
172, 174, 182 180, 193, 195–200, 202–206, 208
PECVD, 143, 163 Polarization effect, 200
Peg-in-hole, 8 Polarization level, 167
Pentacene, 120 Polarization phenomenon, 198
Perception, 14, 15, 20, 24, 25, 27–29, 31, 32, Polarization vector, 203
70 Poling, 101, 153, 161, 163, 165, 166, 201, 204,
Perception for action, 6, 15 228
Percolation theory, 102 Poly(allylamine hydrochloride) (PAH), 110
Percolation threshold, 102, 103, 106, 107 Polyaniline (PANI), 103
Percolation transition, 105 Polyelectrolyte gels, 109
Peripheral nervous system, 14 Poly(ethylene naphthalate) (PEN), 120
Permeability, 94 Polymer gels, 108
Permittivity, 86, 161 Polymer matrix, 102, 105
Personal digital assistants (PDA), 82 Poly(methyl methacrylate) (PMMA), 105
pH, 109, 116 Polypyrrole (PPy), 103
Phase, 58 Polysilicon, 84, 112, 116
Phase plot, 148, 149, 172 Polystyrene (PS), 105
Photo-semiconductor device, 140 Poly(styrene sulfonate) (PSS), 110
Photodetector, 90 Poly[styrene-b-(ethylene-cobutylene)-b-
styrene] (SEBS),
Photodiode, 91, 109
106
Photolithography, 165
Polythiophene (PTh), 103
Photoluminescence, 110
Polyurethane (PU), 107
Photoreceptor, 66
Portability, 72
Photoreflector, 92
Portable, 36
Piecewise linear mapping, 70
Pose, 52
Piezo-ignition, 196 POSFET, 64, 66, 98, 116, 119, 151, 153–174,
Piezoceramic, 211, 212, 222, 224, 230 177–185, 187–193, 196, 230
Piezoelectric, 16, 54, 57, 58, 64, 96, 101, 109, POSFET array, 189, 190
120, 155, 157, 195, 196, 198–205, POSFET tactile sensing chip, 177
207–209 POSPET, 155
Piezoelectric constant, 97, 101, 120, 142, 149, Potentiometer, 81
154, 157, 161, 165, 206, 212, 215, 217, Power consumption, 60, 62, 71, 74, 97
219, 228 Power management, 71
Piezoelectric effect, 101, 165, 195, 196, 198, Power supply, 71, 72
201–204 Pre-processing, 15, 26, 65
Piezoelectric loss, 219, 224, 227, 230 Precise manipulation, 44, 177
Piezoelectric phenomenon, 195, 203 Precision, 9
Piezoelectric polarization, 204 Precision grip, 31
Piezoelectric polymer, 60, 116, 118, 119, 139, Prehensile, 30
141–143, 147, 151, 153–155, 159–161, Pressure, 20, 44
163, 164, 167, 179, 193, 211, 212, 214, Pressure conductive rubber, 58–60, 84, 120,
216, 221, 222, 227, 228, 230, 233 121
Piezoelectric resonator, 217 Pressure sensitive elastomer, 106
Piezoelectric strain coefficient, 203 Pressure sensitive ink, 83, 101
Piezoelectric stress constant, 203 Pressure sensitive rubber, 122
Piezoelectricity, 195, 196, 198, 203, 205, 209 Pressure sensor, 125, 140, 172
Piezoresistance, 81, 83 Pressure sensor array, 106
Piezoresistive, 54, 83, 89, 102, 105, 108, 112, Pressure threshold, 25, 51, 178
113, 120, 124, 155 Primary afferent fibers, 27
Piezoresistive effect, 120, 172 Printed circuit board (PCB), 48, 94, 105, 112,
Pixel, 66 123, 126, 146
Polar dielectrics, 197 Printing electronics, 191
242 Index
Programming tool, 53 Resistive tactile sensors, 81

Programming tools, 48 Resolution, 82, 85, 89, 92, 93, 96, 98, 106,
Propagation delay, 119 111, 116, 173
Proprioception, 5, 26, 31, 46, 70 Resonance, 225
Proprioceptor, 20 Resonant frequency, 68, 86, 96, 151
Prosthetics, 9, 10, 110 Resonator, 195
Proximity, 57, 89, 126 Response, 44, 45
Proximity sensing, 46 Response speed, 82
Proximity sensor, 126 Response time, 58–60
Proximity sensors, 45 Reverse piezoelectric effect, 200, 203
PSPICE, 170, 218, 219, 221, 222 Rheological, 99
Pulse width modulation, 124 Robot interaction, 8
Pulse-echo method, 221 Robot skin, 70
PVDF, 57, 58, 95, 97, 102, 117, 119, 139, 141, Robot-assisted touch therapy, 10
142, 154, 156, 157, 166, 170, 196, 198, Robotic skin, 6
201, 206, 208, 212, 216, 217, 222–224 Robotic surgery, 9
PVDF-TrFE, 118, 120 Robotic tactile sensing, 192
P(VDF-TrFE), 139, 141, 142, 144, 148, Robots in education, 11
153–157, 163–166, 170, 172, 173, 187, Robustness, 74
189, 193, 198, 201, 207–209, 217, Rolling, 185
222–224, 228 Roughness, 29, 31, 115
Pyroelectric, 57, 98, 155, 157, 173, 187 Roughness perception, 30
PZT, 96, 102, 154, 155, 198, 217 Routing, 49, 53, 54, 73, 115
Row–column scanning, 59
Q
Row–column scheme, 54
Quality factor, 224
Ruffini corpuscles, 20–22
Quantum size effect, 110
Quantum tunneling, 103
Quantum tunneling composites (QTC), 85, 103 S
Safe grasp, 69
R Safe interaction, 44, 126
Ramp function, 228 Sallen key low pass filter, 233
Re-configure, 73 Sampling rate, 64
Reaching, 30 Sampling time, 58
Read out, 62, 64, 115 Sawyer tower circuit, 167
Read-out, 190, 193 Scalability, 72, 88, 116
Real-time communication, 124 Scalable, 36, 92, 123
Receptive field, 15, 16, 22–24, 27, 28, 69 Scanning rate, 58–60
Redundancy, 72, 73 Scanning time, 59, 83
Redundant data, 187 Second order neurons, 27
Reflection coefficient, 227, 228 Self capacitance, 86, 87
Refractive index, 91, 110 Self diagnostics, 74
Rehabilitation, 9, 10, 80, 99 Self healing, 74
Relaxation oscillator, 86 Self-assembly, 110
Relaxation time, 62, 122 Self-organizing map (SOM), 69
Reliability, 43, 67, 68, 72, 73, 159 Semiconductive yarn, 108
Remanent polarization, 119, 154, 157, 163, Semiconductive yarns topology, 108
166, 169, 202 Semiconductor, 142, 162
Repeatability, 9, 36, 114 Semiconductor device, 139, 140
Reproducibility, 117 Semiconductor nanocrystals, 110
Resistance, 104, 215 Semiconductor sensor, 140
Resistive, 16, 62, 81, 99 Sense of touch, 3–6, 8, 13, 14, 31, 43, 159
Resistive sensors, 81 Sensitive, 155
Resistive tactile sensing, 85 Sensitive skin, 80, 85
Index 243
Sensitivity, 20, 24, 25, 28, 29, 32, 34, 51, 60, Software, 73
86, 88, 92, 94, 97, 100, 102, 104–107, Solid state optical sensor, 91
110, 116, 119, 120, 122, 124, 145, 178, Somatosensory cortex, 15
187, 189, 193 Somatosensory system, 27, 43, 69
Sensor, 72 Somatotopic alphabet, 69
Sensor distribution, 51 Somatotopic map, 26, 27, 35, 69, 70
Sensor fusion, 19, 67–70, 129 Sonar, 196
Sensor hardware, 47 Source, 108
Sensor integration, 139, 140 Source-follower, 160, 171, 173, 180, 189, 193
Sensor placement, 47, 51, 52 Spatial acuity, 21, 24, 177, 178
Serial access, 60 Spatial filtering, 49
Serial bus, 68 Spatial resolution, 16, 24, 32–34, 50, 51, 60,
Service robot, 70 92, 105, 111, 112, 116–118, 124, 125,
Shannon’s theorem, 58 129, 143, 148, 155, 159, 177, 178, 182,
Shape, 5, 6, 27–31, 34, 44, 69 184, 190, 193
Shear, 97 Spatio-temporal, 28, 51, 73, 151, 178
Shear force, 46, 91 SPICE, 211, 212, 218, 219, 221–224, 226,
Shear strain, 112 228–230
Shear stress, 112, 113 Spike, 26
Sigma–Delta modulation, 86 Spin-coating, 164
Signal conditioning, 55, 58, 61–66, 115, 126, Spinal cord, 26
188 Spinothalamic pathway, 26
Signal processing, 115, 116, 139, 159 Spontaneous polarization, 165, 198
Signal-to-noise ratio (SNR), 59, 97 Stability, 117
Signal-to-noise (S/N) ratio, 116 Step function, 227, 228
Silicon micromachining, 112, 113 Stiffness, 6, 9, 16, 32, 46, 95, 100, 112, 155,
Silicon on insulator (SOI), 191 203
Silicon technology, 88 Strain, 57, 203
Silicone matrix, 103, 104 Strain gauge, 98, 115, 116
Simulation, 169, 221 Strain sensor, 104
Simultaneous localization and mapping Stress, 57, 119, 203
(SLAM), 70 Stress rate, 57
Size, 69 Stress/strain, 107
Size perception, 31 Stretchability, 50, 129
Skin, 20, 25, 27–30, 32, 34, 35, 54, 55, 65, 68, Stretchable, 36, 47, 84
89, 101, 109, 115, 116, 120, 123, 124 Stretchable skin, 51
Skin area, 55 Substrate capacitance, 158
Skin elasticity, 34 Super resolution algorithm, 51
Skin mechanics, 29, 32, 35 Surface charge density, 205
Skin stiffness, 32 Surface parametrization, 70
Skin surface area, 32, 35 Synapse, 36
Skin thickness, 32, 35, 55, 192 Synthetic polymer, 156
Skin weight, 32, 35, 47, 55, 192 System architecture, 63
Slip, 29, 30, 46, 65, 95 System in package (SIP), 66, 193
Slow adapting (SA), 22, 25, 28, 29 System on chip (SOC), 66, 177, 188, 191, 193
Small signal equivalent, 158
Smart fabrics, 66 T
Smart material, 96, 139, 141, 155, 173 Tactile, 33, 70
Smart textile, 108 Tactile actuator, 99
Smart vision, 66 Tactile acuity, 24, 159
Social robots, 80 Tactile classification, 14
Soft touch, 126 Tactile fabric, 47
Soft-touch, 45, 46 Tactile image, 7, 51
Softness, 6, 29, 70, 96, 109, 110 Tactile imaging, 85
244 Index
Tactile perception, 14, 110 Time delay, 219

Tactile representation, 26, 27, 30, 34, 35 Time multiplexing, 62, 63
Tactile sensing, 3, 4, 6, 9–11, 14–16, 19, 20, Top–down, 47
28, 30, 45, 58, 79, 80, 99, 107, 122, Top–down method, 192
140, 230 Total internal reflection, 91
Tactile sensing array, 15, 16, 58, 64, 88, 98, Touch receptor, 109
107, 116–118, 123, 139–141, 143, 148, Touch screen, 82
151, 178, 188, 189, 192, 193 Touch sensing, 15
Tactile sensing arrays, 79 Touch sensing device, 153, 159
Tactile sensing chip, 139, 141, 145, 147, Touch sensor, 7, 84, 109, 140, 141, 154
177–182, 184–187, 189–191, 193, 233 Traction sensors array, 112
Tactile sensing structure, 112, 113, 116, 120, Traction stress, 113
123, 129, 188, 191 Trade-off, 74, 159, 160
Tactile sensing system, 43, 44, 46, 47, 56, 63, Tranducer, 192
65, 68, 69, 72–74, 105, 191 Transconductance, 159, 179, 188
Tactile sensing system on chip, 188, 191 Transduce, 57
Tactile sensing technologies, 80 Transducer, 56, 58–60, 62, 65, 66, 72, 116,
Tactile sensitivity, 50 117, 122, 126, 139, 140, 151, 153, 157,
Tactile sensor, 15, 16, 19, 24–26, 30, 33–37, 174, 195, 211, 219, 222, 230
44, 46, 57, 58, 62, 69, 74, 79, 80, Transducer bandwidth, 58, 59
83–86, 89, 93, 96, 97, 104, 105, 110, Transduction, 25, 34, 54, 57, 58, 60, 65,
114–117, 140, 142 79–81, 99, 112
Tactile sensor array, 33, 57, 61, 66 Transformation, 51
Tactile sensor sheet, 124 Transformer, 170, 219
Tactile sensor technology, 129 Transistor, 112, 116, 118, 120, 122, 141–143,
Tactile sensor types, 45 153, 157–162, 167, 171, 174, 192
Tactile sensors, 9, 85 Transmission, 25, 66, 68
Tactile skin, 35, 48, 50, 55 Transmission line, 169, 170, 212, 215, 219,
Tactual perception, 14 223, 224
Tangential force, 86 Transmission loss, 224, 227
Task classification, 45 Transversal, 97
Taxel, 52, 70, 140, 142, 143, 145–151, 171, Triangulation, 48
193 Trigeminal somatic sensory system, 26
Technology spread, 62, 179 Tuned resonator, 68
Telegraphist’s equations, 215 Tunneling effect, 106
Telemedicine, 9 Two points threshold, 24, 33, 34
Teleoperation, 9
Temperature, 5, 6, 15, 20, 25, 26, 36, 44, 45, U
57, 65, 70, 107, 116, 126, 140, 165 Ultrasonic, 16, 70, 81, 95, 99, 142, 196
Temperature diode, 178, 185–187 Ultrasonic receiver, 95
Temporal acuity, 25 Ultrasonic transmitter, 95
Tensor, 97 Unstructured environment, 6, 8, 10, 45, 46
Textile technologies, 108
Texture, 5, 6, 8, 14, 24, 29, 30, 32, 34, 35, 65, V
69, 70 Very large scale integration (VLSI), 66
Thermal, 33 Vibration, 20, 24, 26, 44, 70
Thermal conductivity, 115 Vibrotactile, 5, 24
Thermal poling, 166 Viscoelastic, 26, 32, 110, 219
Thermal resistor, 115 Viscoelastic loss, 211, 214, 216, 217, 230
Thermoceptors, 20 Visible light, 110
Thermoreceptor, 43 Vision, 25, 31, 49, 70, 71, 79
Thin film transistor (TFT), 118, 192 Vision camera, 69
Threshold voltage, 140, 141, 179 Vision imager, 60
Time constant, 159 Voltage amplifier, 146, 233
Index 245
Voltage divider, 113 Wiring, 44, 74

Voltage sensitivity, 233 Wiring complexity, 52–54, 63, 66, 68, 73, 116,
Vulcanized liquid silicone rubber, 106 151, 193
Workability, 102
W World model, 68
Wave propagation, 212
Wavefront, 96
Wavelength shift, 92 Y
Wheatstone bridge, 113 Young’s modulus, 32, 113
Whole body sensing, 45, 47, 48, 54, 69, 79, 80
Wireless communication, 53, 68 Z
Wireless transmission, 68 ZnO, 101

Piezoelectric Ty

Uploaded by

Copyright:

Available Formats

Piezoelectric Ty

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

Piezoelectric Ty

Uploaded by

Copyright:

Available Formats

Appendix A

Keywords Piezoelectricity · Dielectric · Ferroelectric · Piezoelectric ·

A.1 Introduction and Historical Perspective

R.S. Dahiya, M. Valle, Robotic Tactile Sensing, 195

A.2 Dielectric, Ferroelectric and Piezoelectric Materials

A.2.1 Electric Polarization

One of the concepts crucial to the understanding of dielectric, ferroelectric and

A.2.2 Dielectric Materials

A dielectric material is one that is electrically insulating (nonmetallic) and exhibits

A.2.3 Ferroelectric Materials

A.2.4 Piezoelectric Materials

A.3 The Piezoelectric Effect

The piezoelectric effect, explained with a simple molecular model shown in

A.4 Piezoelectric Materials—Static Actions

A.5 Piezoelectric Effect—Basic Mathematical Formulation

ory of piezoelectricity [10], according to which, piezoelectric materials have a linear

A.5.1 Contribution to Elastic Constants

The piezoelectric phenomenon causes an increase of the material’s stiffness. To un-

Thus, piezoelectric effect results in an increased elastic constant or in other

A.5.2 Contribution to Dielectric Constants

Fig. A.6 Schematic diagram

where ε is the effective dielectric constant.

A.5.3 Piezoelectric Linear Constitutive Relations

Fig. A.7 Tensor directions

E is elastic compliance tensor at constant electric field, ε T is dielectric

Another fundamental parameter used in electromechanical applications is the

Converted Mechanical Energy

d31 (pC/N) 21b 10.7 0.18

Due to the fact that electromechanical response depends on a number of factors,

Depending on the independent variables chosen to describe the piezoelectric be-

Abstract Piezoelectric polymers are used as transducers in many applications,

Keywords Piezoelectricity · Piezoelectric effect · PVDF · PVDF-TrFE ·

R.S. Dahiya, M. Valle, Robotic Tactile Sensing, 211

B.2.1 Piezoelectric Linear Constitutive Relations

where, Sp is the mechanical strain in p direction, Di is electric displacement in i

Fig. B.1 Piezoelectric

D , h and ε S are related to the previous constants by:

with particle velocity (∂ξ/∂t).

In general, the piezoelectric polymers possess frequency dependent mechanical, di-

B.2.3 Polymer Model with Losses

It is assumed that a one-dimensional compression wave is propagating in X di-

cross-sectional area perpendicular to x axis. Complex elastic constant, piezoelectric

Fig. B.2 Transmission line equivalent model of piezoelectric polymer

Thus, the acoustic transmission can be represented by an analogous lossy elec-

B.3 Measurement of Complex Constants

Density (kg/m3 ) ρ 1880 1880 1780 6000

for an unloaded piezoelectric resonator to actual admittance measurements made

Table B.2 Material constants of PVDF-TRFE, PVDF and lead metaniobate

kt Real 0.202 0.262 0.127 0.334

B.4 SPICE Implementation

B.4.1 Mechanical Loss Block

The analogy between (B.21)–(B.22) and (B.24)–(B.25) allows the implementation

B.4.2 Electromechanical Loss Block

The electromechanical conversion is analogous to the transformer action. It is

Table B.3 Value/expression of parameters used in the SPICE equivalent model

Lt (H) 0.0921 0.2894 0.274 2.993

Fig. B.4 Equivalent

B.4.3 Dielectric Loss Block

B.5 Experiment Versus Simulation

B.5.1 Evaluation of Lossless Model