Articles

A Review of Power Harvesting from Vibration using

Piezoelectric Materials
Henry A. Sodano, Daniel J. Inman and Gyuhae Park
ABSTRACTThe process of acquiring the energy surrounding a system and converting it into usable electrical energy is
termed power harvesting. In the last few years, there has
been a surge of research in the area of power harvesting.
This increase in research has been brought on by the modern advances in wireless technology and low-power electronics such as microelectromechanical systems. The advances
have allowed numerous doors to open for power harvesting
systems in practical real-world applications. The use of piezoelectric materials to capitalize on the ambient vibrations
surrounding a system is one method that has seen a dramatic rise in use for power harvesting. Piezoelectric materials
have a crystalline structure that provides them with the ability
to transform mechanical strain energy into electrical charge
and, vice versa, to convert an applied electrical potential into
mechanical strain. This property provides these materials with
the ability to absorb mechanical energy from their surroundings, usually ambient vibration, and transform it into electrical
energy that can be used to power other devices. While piezoelectric materials are the major method of harvesting energy,
other methods do exist; for example, one of the conventional
methods is the use of electromagnetic devices. In this paper
we discuss the research that has been performed in the area
of power harvesting and the future goals that must be
achieved for power harvesting systems to find their way into
everyday use.
KEYWORDS: power harvesting, energy scavenging, energy
generation, piezoelectric.

and replacement of the battery can become a tedious task. In

the case of wireless sensors, these devices can be placed in
very remote locations such as structural sensors on a bridge
or global positioning system (GPS) tracking devices on animals in the wild. When the battery is extinguished of all its
power, the sensor must be retrieved and the battery replaced. Because of the remote placement of these devices,
obtaining the sensor simply to replace the battery can become a very expensive task or even impossible. For instance, in civil infrastructure applications it is often desirable
to embed the sensor, making battery replacement unfeasible.
If ambient energy in the surrounding medium could be obtained, then it could be used to replace or charge the battery.
One method is to use piezoelectric materials to obtain energy lost due to vibrations of the host structure. This captured
energy could then be used to prolong the life of the power
supply or in the ideal case provide endless energy for the
electronic devices lifespan. For these reasons, the amount of
research devoted to power harvesting has been rapidly increasing. In this paper we review and detail some of the topics in power harvesting that have been receiving the most
research, including energy harvesting from mechanical vibration, biological systems, and the effects of power harvesting on the vibration of a structure.
2. Fundamentals of Power Harvesting

1. Introduction
With the recent advances in wireless and microelectromechanical systems (MEMS) technology, the demand for
portable electronics and wireless sensors is growing rapidly.
Because these devices are portable, it becomes necessary
that they carry their own power supply. In most cases this
power supply is the conventional battery; however, problems
can occur when using batteries because of their finite
lifespan. For portable electronics, replacing the battery is
problematic because the electronics could die at any time
Henry A. Sodano and Daniel J. Inman, Center for Intelligent Material Systems and Structures, Virginia Polytechnic Institute and State University,
Blacksburg, VA 24061, USA.
Gyuhae Park, Engineering Sciences and Applications, Weapon Response
Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA.
The Shock and Vibration Digest, Vol. 36, No. 3, May 2004 197205
2004 Sage Publications
DOI: 10.1177/0583102404043275

The piezoelectric effect exists in two domains: the first is

the direct piezoelectric effect that describes the materials ability to transform mechanical strain into electrical charge; the
second form is the converse effect, which is the ability to
convert an applied electrical potential into mechanical strain
energy. The direct piezoelectric effect is responsible for the
materials ability to function as a sensor and the converse piezoelectric effect is accountable for its ability to function as
an actuator. A material is deemed piezoelectric when it has
this ability to transform electrical energy into mechanical
strain energy, and likewise to transform mechanical strain
energy into electrical charge.
Piezoelectric materials belong to a larger class of materials called ferroelectrics. One of the defining traits of a ferroelectric material is that the molecular structure is oriented
such that the material exhibits a local charge separation, know
as an electric dipole. Throughout the material composition
the electric dipoles are orientated randomly, but when the material is heated above a certain point, the Curie temperature,
and a very strong electric field is applied, the electric dipoles

198

The Shock and Vibration Digest / May 2004

reorient themselves relative to the electric field; this process

is termed poling. Once the material is cooled, the dipoles
maintain their orientation and the material is then said to be
poled. After the poling process is completed the material
will exhibit the piezoelectric effect.
The mechanical and electrical behavior of a piezoelectric
material can be modeled by two linearized constitutive equations. These equations contain two mechanical and two electrical variables. The direct effect and the converse effect may
be modeled by the following matrix equations (IEEE Standard on Piezoelectricity, ANSI Standard 176-1987):
direct piezoelectric effect: 3 D 4 = 5 e 6 T 3 S 4 + 5 1 S 6 3 E 4 (1)
converse piezoelectric effect: 3 T 4 = 5 c E 6 3 S 4 5 e 6 3 E 4
(2)
Here, {D} is the electric displacement vector, {T} is the stress
E
vector, [e] is the dielectric permittivity matrix, [c ] is the matrix of elastic coefficients at constant electric field strength,
S
{S} is the strain vector, [1 ] is the dielectric matrix at constant mechanical strain, and {E} is the electric field vector.
After the material has been poled, an electric field can be
applied in order to induce an expansion or contraction of the
material. However, the electric field can be applied along any
surface of the material, each resulting in a potentially different stress and strain generation. Therefore, the piezoelectric
properties must contain a sign convention to facilitate this
ability to apply electric potential in three directions. For the
sake of keeping this discussion simple, the piezoelectric material can be generalized for two cases. The first is the stack
configuration that operates in the 33 mode and the second
is the bender, which operates in the 13 mode. The sign convention assumes that the poling direction is always in the
3 direction, with this point the two modes of operation can
be understood. In the 33 mode, the electric field is applied
in the 3 direction and the material is strained in the poling
or 3 direction; in the 31 mode, the electric field is applied in the 3 direction and the material is strained in the
1 direction or perpendicular to the poling direction. These
two modes of operation are particularly important when defining the electromechanical coupling coefficient that occurs
in two forms: the first is the actuation term d, and the second
is the sensor term g. Thus g13 refers to the sensing coefficient
for a bending element poled in the 3 direction and strained
along 1.
A full description of the piezoelectric effect and the methods used to model the behavior of these materials is beyond
the scope of this paper. However, a significant number of journal papers and conference proceedings develop accurate models and discuss the fundamentals of these materials in great
detail (Crawley and de Luis, 1987; Crawley and Anderson,
1990; Hagood et al., 1990; Smits and Choi, 1991; Smits et
al., 1991; Near, 1996; Inman and Cudney, 2000; Niezrecki et
al., 2001) as well as numerous books published on this topic
(Gandhi and Thompson, 1992; Ikeda, 1996; Banks et al.,
1996; Culshaw, 1996; Clark et al., 1998; Srinivasan and
McFarland, 2001; Worden et al., 2003).
In the following sections of the paper we break the various works on power harvesting into the following groups:
mechanical vibration, power harvesting efficiency, power
storage and circuitry, implantable and wearable power sup-

plies, and damping induced by power harvesting. In Section 3,

we discuss one paper that investigates the amount of energy
available from one power harvesting device subjected to a
vibration environment. In Section 4 we look at research that
was performed to classify the efficiency of certain methods
of power harvesting. In Section 5 we look at research into
various types of power storage mediums and different circuits
developed to maximize the electric power generated. A large
portion of work has been performed in the field of power harvesting from biological systems and, while the papers found
in section 6 may deal with various subjects, they all have a
major focus on the ability to obtain energy from human or
animal activity. Section 7 will detail work into quantifying
the effect of power harvesting on the dynamics of a vibrating structure.
3. Mechanical Vibration
One of the most effective methods of implementing a power harvesting system is to use mechanical vibration to apply
strain energy to the piezoelectric material or displace an electromagnetic coil. Power generation from mechanical vibration
usually uses ambient vibration around the power harvesting
device as an energy source, and then converts it into useful
electrical energy, in order to power other devices. The research in the following three sections has made use of mechanical vibration in order to quantify the efficiency and
amount of power capable of being generated, as well as to
power various electronic systems, ranging from digital electronics to wireless transmitters.
Williams and Yates (1996) proposed a device, which generated electricity when embedded in a vibrating environment. For their evaluation, an electromagnetic transducer was
chosen. A harmonic analysis of the generator was performed
in order to evaluate the viability of the device and to optimize the design. It was determined from the analysis that the
amount of power generated was proportional to the cube of the
vibration frequency. This illustrated that the generator was
likely to perform poorly at low frequencies. It was also determined that a low damping factor was required to maximize power generation, therefore the design must allow for
large deflections of the mass. For a typical device the predicted power generation was 1 2W at an excitation frequency of 70 Hz, and 0.1 mW at 330 Hz (assuming a deflection
of 50 2m).
4. Power Harvesting Efficiency
The two papers in this section investigate the efficiency of
a piezoelectric generator. The first paper looks at the efficiency of a piezoelectric vibrating in the 31 direction and the
second paper tests a stack that operates in the 33 direction.
It is important to quantify the efficiency of the power harvesting medium in order to allow the device to be designed to
function optimally in its intended environment.
Umeda et al. (1996) carried out an investigation concerning the fundamentals of a generator, which transformed mechanical energy to electrical energy using a piezoelectric vibrator and a steel ball. They also investigated the effect of
the various characteristics of the piezoelectric vibrator. To
simulate the generation mechanism, they introduced an electrical equivalent model. The fundamental modes of bending

Sodano et al. / A REVIEW OF POWER HARVESTING

vibration for two models were calculated: model A (the transducer with the steel ball) and model B (the transducer only).
The admittance characteristics of each model were measured
and they found that it was clear that the peak frequencies
corresponded to the vibration modes. It was seen that the
calculated waveforms of the output voltage were similar to
the measured ones; therefore, the model provided an accurate
simulation of the output voltage. An efficiency curve was
drawn for various input mechanical energies, and they determined that as the potential energy of the ball increased
the maximum efficiency decreased. A large part of the applied energy was returned to the steel ball in the form of kinetic energy causing it to bounce off the plate. It was concluded that the energy generated would be large if the steel
ball did not bounce off after an impact but rather vibrated
with the piezoelectric plate. This case was simulated and it
was determined that a maximum efficiency of 52% could be
obtained. The effects of the characteristics of the piezoelectric vibrator were investigated and it was determined that the
efficiency increased if the mechanical quality factor increased,
the electromechanical coupling coefficient increased and the
dielectric loss decreased.
Goldfarb and Jones (1999) have analyzed the efficiency of
the piezoelectric material in a stack configuration for the purpose of electric energy generation. An analytical model is
presented and suggests that the fundamental problem in generating electrical power from the piezoelectric material is
that it stores the majority of the energy produced and returns
it to the excitation source that initially caused the charge to
be generated. They state that this occurrence is particularly
problematic when the piezoceramic is placed in parallel with
a capacitor that is in series with the load. Therefore, it is suggested that the maximum efficiency of power generation can
be achieved by minimizing the amount of energy stored inside the piezoelectric material. The efficiency of the model
was determined across a spectrum of frequencies and resistive values. It was found that, at frequencies above 100 Hz,
the efficiency of the stack actuator was negligible and that the
highest efficiency was obtained at 5 Hz. This frequency is far
lower than the first mechanical and electromechanical resonances of the stack, which occur at approximately 40 and 60
kHz, respectively. The authors state that the frequency of
maximum efficiency occurs so low because of the energetic
structure of the stack. In addition, it is found that the efficiency of the stack is most strongly dependent on the frequency
of excitation, with the load resistance providing a lower effect on it.
5. Power Storage and Circuitry
When using piezoelectric materials as a means of gathering energy from the surroundings, in most cases it is a necessity that a means of storing the energy generated be used.
Without accumulating a significant amount of energy, the
power harvesting system will not be a feasible power source
for most electronics. The following research has made use of
circuitry to either store the energy generated by the piezoelectric material or to develop circuits that allow the energy to be
removed from the piezoelectric in a more efficient way allowing more power to be generated.
Umeda et al. (1997) continued their investigation with a
study into the characteristics of energy storage by a piezo-

199

generator with a bridge rectifier and capacitor. As in their

previous research, the piezo-generator consisted of a steel ball
and a piezoelectric vibrator, and with the introduction of a
bridge rectifier and capacitor they were able to determine the
energy storage characteristics both theoretically and experimentally. To simulate the generation and storage mechanism
they employed an equivalent circuit model, where the input
mechanical energy was translated into an initial electrical energy. Changing the parameters of the circuit simulated the separation of the vibrator and the ball. After examining the storage characteristics for the first impact they determined that
as the capacitance increased the electrical charge increased
due to an increased duration of oscillation. They also determined that for each value of capacitance as the initial voltage increased the stored electric charge decreased, and the
efficiency increased. When considering the overall storage
characteristics for multiple impacts they determined that, for
each value of capacitance, the first impact gave the largest
electric charge. The overall storage characteristics were observed when the initial voltage was changed; as the initial
voltage increased, the electric charge decreased for each value of capacitance, while the efficiency increased. Their prototype achieved a maximum efficiency of 35%, over three times
that of a solar cell.
Elvin et al. (2001) used a polyvinylidene fluoride (PVDF)
piezofilm sensor attached to a simply-supported Plexiglas
beam with an aspect ratio of 0.11 to generate an electrical signal. The goal of this power harvesting experiment was to generate sufficient energy from the strain induced on the piezofilm by the bending beam to power a telemetry circuit. The
energy generated from the PVDF patch was accumulated in
a capacitor. A switch was added to the circuitry to allow the
capacitor to charge to a predetermined value of 1.1 V, at
which point the switch would open and the capacitor would
discharge through the transmitter. Once the capacitor had
discharged to a value of 0.8 V, the switch would close and
the capacitor would be allowed to recharge and repeat the
process. The operation of the power harvesting system was
found to provide the required energy to power the circuitry
and transmit a signal containing information regarding the
strain of the beam a distance of 2 m.
Kasyap et al. (2002) developed a lumped element model
(LEM) using an equivalent circuit model to describe the power generated from the forced vibration of a cantilever beam
with a piezoelectric element attached. It was found that the
LEM provided results consistent with those generated using
a finite element model from excitation frequencies ranging
from DC through the first resonance of the beam. A similar
result was found during a second model validation using experimental results. The goal of the study was to use a flyback
converter to increase the efficiency of the power transfer from
the piezoelectric patch to a power storage medium. The use
of a flyback converter allows the circuit impedance to be
matched with the impedance of the piezoelectric device. It
was found that when using the flyback converter a peak power efficiency of 20% was achieved.
The previous papers in this section concentrated their efforts on the use of a capacitor as the storage medium. However, in most cases the capacitor is not an efficient method of
storing energy. Sodano et al. (2002) performed a study to investigate the amount of power generated through the vibration of a piezoelectric plate, as well as two methods of power

200

The Shock and Vibration Digest / May 2004

storage. The plate was excited using an electromagnetic shaker with both resonant and random excitation signals. It was
found that the piezoelectric could generate a maximum power
of 2 mW when excited at the resonant frequency of the
clamped-free plate. In addition, the ability of the piezoelectric plate to store its power in both a capacitor circuit and a
rechargeable battery was tested. This paper was the first to
demonstrate that the power output of piezoelectric material
was able to recharge a fully discharged battery without the
use of external energy sources. It was also shown that both
methods of power storage could be used; however, the use
of rechargeable batteries was found to possess power storage
qualities that would allow a far larger range of electronic devices to be powered than the capacitor. This is because of the
capacitors poor ability to store large amounts of power and
its fast discharge rate, which caused the output of the circuit
to switch on and off making a periodic power supply.
Following the work of Sodano et al. (2002), a second paper was published (Sodano et al., 2003) to further investigate the ability of piezoelectric materials to recharge batteries. This study compared the macro-fiber composite (MFC)
actuator with the monolithic piezoceramic material PZT for
recharging batteries. The MFC is an actuator that uses piezofibers and interdigitated electrodes to capitalize on the higher g33 piezoelectric coupling coefficient, allowing it to produce higher strain and force than typical monolithic PZT
(Sodano et al., 2004a). This property of the actuator makes
it attractive for power harvesting applications. First, the efficiency of both the MFC and PZT was determined in order to
compare their ability to generate electrical energy. It was determined that the MFC was less effective for power harvesting than the PZT because of a very low current generation by
the MFC. Reasons for the low current generation were proposed. Furthermore, because of the poor current output of the
MFC it was found to be ineffective at charging the batteries
due to their requirement for fairly significant current. However, the PZT was used to charge a variety of different capacity nickel metal hydride batteries; a typical charge cycle
of one battery is shown in Figure 1. The charge time for each
was supplied and the maximum capacity battery capable of
efficiently being charged was determined.
Another investigation into the ability to store and use the
energy generated from a power harvesting device was performed by Amirtharajah and Chandrakasan (1998). They designed and tested a chip which integrated a finite impulse response (FIR) filter, power field-effect transistors (FETs) and
pulse width modulation (PWM) control circuitry, in order to
demonstrate the possibility of running a digital system from
the power generated by vibrations in its environment. They
proposed a self-powered system consisting of a load circuit,
a generator to create voltage that could vary depending on
the environment, a voltage regulator to set the voltage to a
desired level, and a backup power source. The implementation of a backup power source was required at circuit startup
because of the need for the voltage regulator to obtain its
power from a source other than the generator, whose output
was too uncontrolled to be utilized. An inertial electromechanical generator and acoustic generator were proposed as
the power supply and a prototype of each was built to test its
ability to power the digital circuitry. It was found that the
electromagnetic generator was capable of supplying 400 2W
of power during a typical excitation that was intended to

Figure 1. Typical charge cycle of a nickel metal hydride battery; in this case an 80 mAh battery was charged (Sodano et
al., 2003)

represent that of a human walking. The electromagnetic prototype was tested and it was shown that the portable digital
system could be powered entirely from ambient environmental vibrations for a period of 23 ms. A second investigation
into the use of incident sound as a form of excitation energy
was also tested and it was found that it could generate power
sufficient to run the system; however, the acoustic energy
source was limited to very high noise environments (about
114 dB).
Rather than developing methods of accumulating the energy developed by piezoelectric materials, Ottman et al.
(2002) worked to develop a circuit that would maximize the
power flow from the piezoelectric device. A DCDC stepdown converter was implemented in coordination with a wave
rectifier, capacitor, and electrochemical battery. In addition
to the circuitry, an adaptive control technique was developed
to continuously implement optimal power transfer theory and
to maximize the power flow into the battery. This active controller varied the switching frequency of the step-down converter to maximize to the power flow from the piezoelectric
elements and to raise the current to levels more acceptable
for maintaining the charge on batteries. The circuit and controller were built and tested on a bimorph piezoelectric cantilever plate excited at resonance. It was found that when using the circuit, over four times the energy was transferred to
the battery than with direct charging alone. However, if the
power harvesting medium produced less than 10 V, then power flow into the battery was reduced because of losses in the
additional circuit components.
Hofmann et al. (2002) have continued the work of Ottman
et al. (2002) by implementing circuitry to maximize the power flow from the piezoelectric device. This study uses a similar circuit as before, but realizes that one of the drawbacks of
their previous work was that the PWM control circuitry required more power than was produced at low levels of excitation. In order to correct this problem, the authors realized
that the optimal duty cycle changes very little at higher levels of energy generation when excited with a sinusoidal sig-

Sodano et al. / A REVIEW OF POWER HARVESTING

nal. Therefore, the control circuitry was removed and a constant duty cycle was used. Furthermore, at low levels of power
generation the optimal duty cycle varies greatly, causing the
PWM generation circuitry to be ineffective; thus, this circuitry is bypassed at a certain threshold and a pulse-charging circuit is used. The optimal value of the duty cycle was investigated both analytically and experimentally, resulting in a
value of 2.8%. With this circuit, the power flow was increased
by over a factor of 3 at a peak resonant excitation level of 70
V open circuit. Additionally, their circuitry was found to reach
values as high as 70% efficiency at an optimal value of excitation.
Lesieutre et al. (2002) discuss two topics: the first is an
energy harvesting circuit and the second a measure of the
damping induced in a structure due to energy harvesting,
which will be discussed in a section 7. The goal of the first
portion of this research was to further improve upon a circuit that would maximize the energy output of the piezoelectric material through the use of a DCDC step-down converter. The energy harvesting circuit was developed to
improve on two previously constructed circuits in Ottman et
al. (2002) and Hoffman et al. (2002). The first circuit used a
controller to modify the PWM of a DCDC step-down converter, which consumed a significant amount of power. Additionally, their circuit used the controller at all times, which
means that when the piezoelectric produces a very small
amount of energy the controller would be drawing more energy than available. To correct this problem, a second circuit
was developed using a constant near-optimal duty cycle and
the control circuitry was removed. However, the circuit was
still inefficient when less than 25 V open circuit was generated. Therefore, to further correct this issue the pulse-charging circuitry that turned on below 25 V open circuit was done
away with and only direct charging was used in this range.
This circuit was found to provide a 324% increase in power
when excited at a level sufficient to produce 68 V open circuit and it alleviated many of the shortcomings of the previously used circuits.
6. Implantable and Wearable Power Supplies
In an effort to incorporate computers and digital systems
into our everyday lives, research has been carried out to investigate the possibility and practicality of imbedding them
into our clothing, or inside biological systems such as the human body. The use of power harvesting devices to capture the
energy lost during everyday human life is a captivating idea
and has been one of the main topics facilitating the rapid
growth of the power harvesting field. The following research
presented here has investigated numerous ideas of obtaining
energy from both human and animal activity.
Possibly the first investigation of power scavenging systems incorporated into a biological environment was performed in 1984 by Hausler and Stein, who published a paper
proposing the use of an implantable physiological power
supply using PVDF films (Hausler and Stein, 1984). Based
on the concept that the energy expended for respiration could
be converted into electric power, Hausler and Stein used the
relative motion of the ribs to periodically stretch a converter.
A miniaturized prototype was designed and used to conduct
an animal experiment. The converter was fixed to the ribs of
a mongrel dog and spontaneous breathing led to a peak volt-

201

age of 18 V, which corresponded to a power of about 17 2W.

However, the power generated was insufficient to power the
desired electronics, making it ineffective for use as an implanted power supply. It is speculated that optimization of
the PVDF film properties, as well as a more suitable converter attachment at the ribs would make it possible to develop
power converters with an output of 1 mW, yielding a mechanical power load of 20 mW. In addition, this study was performed at an early stage in low-power electronic and computer technology, suggesting that the use of more efficient
electronics, now available, would have resulted in significantly more promising results.
Throughout our daily activity, a significant amount of energy is expended in various forms, some of which make for
attractive power harvesting locations. Starner (1996) has performed an investigation into the amount of power expended
for a vast range of human activities. His paper explores the
possibility of eliminating bulky and inconvenient power systems by harnessing the energy expended in everyday activity and using it to generate power for a computer. The paper
contains a survey of various power generation methods ranging from body heat and breath to finger and upper limb motion. An analysis of the power available from each of the
different locations is presented. He calculates that approximately 67 W of power is lost during walking and that a piezoelectric device mounted inside a shoe with a conversion efficiency of 12.5% could achieve 8.4 W of power. Two methods
of power generation during walking are identified, piezoelectric and rotary generator, with the advantages and weaknesses of each outlined. One idea he explains is to place piezofilm patches in the joints of clothing to harvest the energy
lost during bending and he states that about 0.33 W could be
obtained. In addition to investigating the possible location and
power converters to be used, he realizes that the energy generated will never be constant and, at times, energy may not
be produced at all, making the use of a power storage medium a must. Power would be accumulated when the energy is
plentiful and would be used when insufficient energy is produced. The paper investigates two methods of power storage:
the capacitor and rechargeable batteries. He states that energy storage in a capacitor would be sufficient for low-power
areas such as blood pressure and body heat, but rechargeable
batteries are a necessity for higher power areas, such as limb
motion and walking.
The work of Starner (1996) brought the possibility of
power harvesting locations around the human body to the attention of many researchers and the work in wearable power
supplies began to grow. Post and Orth (1997) investigated
the concept of smart fabric for wearable clothing. Their research described techniques used in building circuits from
commercially available fabrics, fasteners, etc. Multiple different conductive fabrics were explored, including silk organza, constructed of silk thread wrapped in thin copper foil
running in one direction and plain silk in the other. This material was highly conductive, had a high tensile strength, and
could withstand high temperatures, allowing it to be sewn
using industrial machines. A second type of conductive yarn
is manufactured with both conductive and cloth fiber interspersed throughout the material. Post and Orth (1997) state
that by varying the amount of conductive material the resistance of the fiber can be adjusted and other components, such
as capacitors and coils, can be sewn directly into the fabric.

202

The Shock and Vibration Digest / May 2004

The use of this type of material has led to the development of

several devices constructed of fabric, including a type of fabric keyboard that can be crumpled up, thrown in the wash
and even used as a potholder without losing its ability to function. These materials would be very effective for transmitting the energy generated around the body to the storage medium in a convenient and unnoticeable way.
Kymissis et al. (1998) studied the use of piezoelectric actuators located inside the sole of a shoe for power harvesting. Their research examined three different devices that could
be built into a shoe to harvest excess energy and generate electrical power parasitically while walking. The devices that
were considered included a Thunder actuator constructed
of piezoceramic composite material located in the heel, a rotary magnetic generator also located under the heel, and a
multilayer PVDF foil laminate patch located in the sole of the
shoe. The Thunder actuator was developed by the National
Aeronautics and Space Administration (NASA) and has a
rainbow (arch) configuration that allows the high impact vertical energy of the heel to be translated into bending strain
for electrical power generation. The electromagnetic generator used the pressure of the heel to spin a flywheel and rotary generator, to extract the power from the pressure of the
heel during walking. The last device used was the laminate
of piezofilm, or stave, which was used to harness the energy lost during the bending of the sole. In order to compare
the performance of the three methods, a working prototype
was constructed for each and its performance was measured.
The peak powers were observed to approach 20 mW for the
PVDF stave and 80 mW for the PZT unimorph. However,
because of slow excitation, the average power generated from
both the PVDF stave and the Thunder actuator was significantly lower, approximately 1 and 2 mW, respectively. The
shoe mounted rotary generator resulted in a peak power of
about 1 W and averaged to about 0.25 W over a 5 s sample
period. However, the rotary generator was not easily integrated into the shoe and significantly interfered with the users gait, unlike the PVDF stave and PZT Thunder actuator.
Because of these two limitations of the rotary generator, it
was determined that it was an unrealistic method of generating energy during walking. After examining the performance of the piezoelectric generators, their ability to power a
battery-less active radio-frequency (RF) tag was tested. The
authors developed a circuit that used a capacitor to accumulate the electrical energy along with various other components to regulate the charging and discharging cycle of the
capacitor. The discharge of the capacitor was limited at 5 V
to accommodate an encoder and transmitter used to transmit
the RF tag. The circuit was found to be compatible with both
the shoe mounted piezoelectric generator systems and was
able to transmit five to six 12-bit signals every few steps.
The circuitry developed in this study has also found use in
the work of several other researchers. The research presented in this paper demonstrated the potential of piezoelectric
power harvesting devices for use as a power supply of selfpowered electronics. Further, the ability to use energy from
power harvesting for transmitting data was shown and gained
the attention of many researchers in the area of self-powered
wireless sensors.
Similar to the work of Kymissis et al. (1998), Shenck (1999)
demonstrated electrical energy generation from piezoelectric patches located in a shoe. He evaluated different regula-

tion systems for conditioning the electrical energy harnessed

by the piezoceramic source imbedded in the sole of a shoe.
A rigid bimorph piezoceramic transducer was developed
and integrated into a mass produced shoe insert. The use of
a bimorph piezoelectric device posed several advantages over
the previously used actuator. Since the vertical displacement
of the transducer was required to be very small, so the use of
a second piezoelectric patch allowed more energy to be generated. Additionally, with two piezoelectric patches present
the electrical leads could be configured as parallel energy
sources, improving the lumped impedance characteristics of
the sources. Furthermore, it was determined that a bimorph
transducer was stronger and less intrusive to the user, because it was capable of better adapting to various distributions
of body weight and footfall velocity. The piezoelectric patch
was configured similarly to the Thunder actuator previously
mentioned, allowing it to absorb the energy of a heel strike
and lift during walking, thus inducing a charge across the
capacitive PZT. The energy stored was removed at its peak
and converted into a useful form using a high-frequency
switching technique.
A design study was conducted by Ramsey and Clark
(2001), which investigated the feasibility of using a piezoelectric transducer as a power supply for an in vivo MEMS
application. The 33- and 31- modes of operation for a piezoelectric generator were analyzed and compared, and it was
determined that when using the 31- mode, or thin plate configuration, there existed a strong mechanical advantage in converting applied pressure to working stress. For very low-pressure sources, the 31- mode had a greater advantage in energy
conversion, which became important when attempting to implement this technology in a biological microsystem application. A design study was used to investigate whether or not
the 31- mode was well suited for the in vivo environment,
and it was carried out using a square thin plate driven by blood
pressure. It was shown that ample power existed from various
sources in the body to meet the requirements of their investigation, and additional calculations illustrated the feasibility of
providing intermittent power instead of continuous power.
7. Damping Effect of Power Harvesting
When a power harvesting system is integrated into a structure, energy is removed in the form of electricity. Because
energy is removed from the structure, it must see some effect on its dynamics. The subsequent papers have looked to
quantify this damping effect.
Lesieutre et al. (2002) investigated the damping added to
a structure due to the removal of electrical energy from the
system during power harvesting. The damping was first estimated using analytical methods and later verified through experimental results. It is stated that optimal power transfer is
achieved when the operating rectifier output voltage is half
the open circuit voltage; this assumption allows the effective
loss factor of the system to be dependent only on the coupling
coefficient. Using this simplification the study analytically
predicted the damping loss factor from power harvesting to
be 2.3% in the fundamental mode of vibration of a cantilever beam. The prediction was then verified to be 2.2% through
experimental results, showing excellent agreement between
theory and their experimental work. This result was found

Sodano et al. / A REVIEW OF POWER HARVESTING

Figure 2. Impulse response with a 1007 resistive load

(Sodano et al., 2004a)

203

Figure 4. Impulse response with a 100 k7 resistive load

(Sodano et al., 2004a)

three cases is shown in Figures 2, 3 and 4. It was shown that

for a small load resistance, the damping did not become much
larger than the mechanical damping of the structure, because
only a slight amount of energy was being removed from the
system. As the resistance increases, more energy is removed
from the system and the damping increases, as shown in Figure 3. At the optimal load resistance the maximum energy is
removed from the system and the damping becomes far larger
than that of the structure. As the load resistance moves past
the optimal value the damping begins to decrease again, as
shown in Figure 4. The damping begins to decrease at higher
load resistances because as the load increases in impedance
the circuit begins to look like an open circuit, thus interfering with the ability of the generated electricity to efficiently
flow out of the piezoelectric material. This simulation showed
that power harvesting works very much like a shunt damping system with the major difference being that the energy is
stored rather than dissipated.
Figure 3. Impulse response with a 15 k7 resistive load
(Sodano et al., 2004a)

to be comparable to that of resistive shunting while not having the frequency dependency that shunting does.
Sodano et al. (2004b) presented a paper that developed a
model of a power harvesting system. The model was derived
from variational principles and was used to predict the amount
of electrical energy that could be generated through the vibration of piezoelectric patches on a beam structure. To validate the accuracy of the model, a composite beam with a
complex layout of four piezoelectric patches was experimentally tested and compared to the results of the simulation. It
was shown that the model provided a very accurate estimate
of the power generated independent of the excitation frequency and load resistance. Following the validation of the model, it was used to show the effects of power harvesting on the
damping of a structure that has energy being generated from
it. The impulse response of a cantilever beam was shown for
a power harvesting system that had three different load resistances; the effect of power harvesting on each of these

8. Future of Power Harvesting

The idea of carrying electronic devices such as a portable
radio and never worrying about when the batteries will need
to be replaced could be far closer than one would think. This
thought has caused the desire for self-powered electronics to
grow quickly, leaving only one limitation before these devices can become a reality. The one issue that still needs to be
resolved is a method to generate sufficient energy to power
the necessary electronics. However, with the advances in power harvesting that have been outlined in this paper the ability
to obtain and accumulate the necessary amount of energy to
power such devices is clearly possible.
The major limitations facing researchers in the field of power harvesting revolve around the fact that the power generated by piezoelectric materials is far too small to power most
electronics. Therefore, methods of increasing the amount of
energy generated by the power harvesting device or developing new and innovative methods of accumulating the energy are the key technologies that will allow power harvest-

204

The Shock and Vibration Digest / May 2004

ing to become a source of power for portable electronics and

wireless sensors. One recent advance that shows great promise for power harvesting is the use of rechargeable batteries
as a means of accumulating the energy generated during power harvesting. Much of the early research into power harvesting looked to the capacitor as a method of storing energy and powering electronics. However, the capacitor has poor
power storage characteristics because of its quick discharge
time, causing the electrical output of such circuitry to switch
on and off as the capacitor charges and discharges. This aspect of the capacitor is not suitable for powering computational electronics. However, the rechargeable battery can be
charged and then used to run any number of electronic devices for an extended period of time while being continuously
charged by ambient motion. Innovations in power storage
such as the use of rechargeable batteries with piezoelectric
materials must be discovered before power harvesting technology will see widespread use.
Furthermore, the efficiency of the power harvesting circuitry must be maximized to allow the full amount of energy generated to be transferred to the storage medium. The
continuous advances that are being made in low-power electronics must be studied and utilized both to optimize power
flow from the piezoelectric and to minimize circuit losses.
Gains in this area are a necessity for the successful use of
piezoelectric materials as power harvesting devices. Additionally, the intended location of the power harvesting system must be identified so that its placement can be optimized
and the excitation range realized to allow for tuning of the
power harvesting device. By tuning the power harvesting medium with the structure, the excitation can be made to maximize the strain of the piezoelectric material using the concept of resonance.
Finally, practical applications for power harvesting systems such as wireless sensors and self-powered damage detection units must be clearly identified to encourage growth
in this area of research, thus allowing the contributions and
in flow of ideas to increase. With the advances in wireless
technology and low-power electronics, power harvesting is
the missing link for completely self-powered systems.
References
Amirtharajah, R., and Chandrakasan, A. P, 1998, Self-Powered Signal
Processing Using Vibration Based Power Generation, IEEE Journal
of Solid-State Circuits, Vol. 33, No. 5, 687695.
Banks, H. T., Smith, R. C., and Wang, Y., 1996, Smart Materials and Structures: Modelling, Estimation and Control, Wiley, New York.
Clark, R. L., Saunders, W. R., and Gibbs, G. P., 1998, Adaptive Structures:
Dynamics and Control, Wiley, New York.
Crawley, E., and Anderson, E., 1990, Detailed Models of Piezoceramic
Actuation of Beams, Journal of Intelligent Materials and Structures,
Vol. 1, No. 1, 425.
Crawley, E. F., and de Luis, J., 1987, Use of Piezoelectric Actuators as Elements of Intelligent Structures, AIAA Journal, Vol. 25, No. 10,
13731385.
Culshaw, B., 1996, Smart Structures and Materials, Artech House, Boston,
MA.
Elvin, N. G., Elvin, A. A., and Spector, M., 2001, A Self-Powered Mechanical Strain Energy Sensor, Smart Materials and Structures, Vol.
10, 293299.
Gandhi, M. V., and Thompson, B. S., 1992, Smart Materials and Structures, Kluwer Academic, Dordrecht.
Goldfarb, M., and Jones, L. D., 1999, On the Efficiency of Electric Power
Generation with Piezoelectric Ceramic, ASME Journal of Dynamic
Systems, Measurement, and Control, Vol. 121, 566571.

Hagood, N. W., Chung, W. H., and von Flotow, A., 1990, Modeling of Piezoelectric Actuator Dynamics for Active Structural Control, Journal of Intelligent Materials Systems and Structures, Vol. 1, 327354.
Hausler, E., and Stein, E., 1984, Implantable Physiological Power Supply
with PVDF Film, Ferroelectrics, Vol. 60, 277282.
Hofmann, H., Ottman, G. K., and Lesieutre, G. A., 2002, Optimized Piezoelectric Energy Circuit Using Step-Down Converter in Discontinuous Conduction Mode, IEEE Transactions on Power Electronics,
Vol. 18, No. 2, 696703.
Ikeda, T., 1996, Fundamentals of Piezoelectricity, Oxford University Press,
New York.
Inman, D. J., and Cudney, H. H., 2000, Structural and Machine Design
Using Piezoceramic Materials: a Guide for Structural Design Engineers, Final Report NASA Langley Grant NAG-1-1998.
Kasyap, A., Lim, J., Johnson, D., Horowitz, S., Nishida, T., Ngo, K., Sheplak, M., and Cattafesta, L., 2002, Energy Reclamation from a Vibrating Piezoceramic Composite Beam, in Proceedings of 9th International Congress on Sound and Vibration, Orlando, FL, Paper No.
271.
Kymissis, J., Kendall, C., Paradiso, J., and Gershenfeld, N., 1998, Parasitic Power Harvesting in Shoes, in Proceedings of the 2nd IEEE International Symposium on Wearable Computers, October 1920, Pittsburg, PA, 132139.
Lesieutre, G. A., Hofmann, H. F., and Ottman, G. K., 2002, Electric Power Generation from Piezoelectric Materials, in Proceedings of the
13th International Conference on Adaptive Structures and Technologies, October 79, Potsdam/Berlin, Germany.
Near, C. D., 1996, Piezoelectric Actuator Technology, in Smart Structures and Materials 1996: Smart Structures and Integrated Systems, San
Diego, CA, Proceedings of SPIE, Vol. 2717, 246258.
Niezrecki, C., Brei, D., Balakrishnam, S., and Moskalik, A., 2001, Piezoelectric Actuation: The State of the Art, Shock and Vibration Digest,
Vol. 33, No. 4, 269280.
Ottman, G. K., Hofmann, H., Bhatt A. C., and Lesieutre, G. A., 2002, Adaptive Piezoelectric Energy Harvesting Circuit for Wireless, Remote
Power Supply, IEEE Transactions on Power Electronics, Vol. 17,
No.5, 669676.
Post, E. R., and Orth, M., 1997, Smart Fabric, or Wearable Clothing, in
Proceedings of the 1st International Symposium on Wearable Computers, October 1314, Cambridge, MA, 167168.
Ramsey, M. J., and Clark, W. W., 2001, Piezoelectric Energy Harvesting
for Bio MEMS Applications, in Proceedings of the SPIE 8th Annual Smart Materials and Structures Conference, Newport Beach, CA,
Vol. 4332-2001, 429438.
Shenck, N. S., 1999, A Demonstration of Useful Electric Energy Generation from Piezoceramics in a Shoe, Masters of Science Thesis Proposal, Department of Electrical Engineering and Computer Science,
Massachusetts Institute of Technology.
Smits, J., Dalke, S., and Cooney, T. K., 1991, The Constituent Equations
of Piezoelectric Bimorphs, Sensors and Actuators, Vol. 28, 4161.
Smits, J., and Choi, W., 1991, The Constituent Equations of Heterogeneous Bimorphs, IEEE Transactions on Ultrasonic, Ferroelectrics and
Frequency Control, Vol. 38, No. 3, 256270.
Sodano, H. A., Magliula, E. A., Park, G., and Inman, D. J., 2002, Electric
Power Generation from Piezoelectric Materials, in Proceedings of
the 13th International Conference on Adaptive Structures and Technologies, October 79, Potsdam/Berlin, Germany.
Sodano, H. A., Park, G., Leo, D. J., and Inman, D. J., 2003, Use of Piezoelectric Energy Harvesting Devices for Charging Batteries, in SPIE
10th Annual International Symposium on Smart Structures and Materials, March 26, San Diego, CA, Vol. 5050, pp. 101108.
Sodano, H. A., Park, G., and Inman, D. J., 2004a, An Investigation into the
Performance of Macro-Fiber composites for Sensing and Structural
Vibration Applications, Mechanical Systems and Signal Processing,
Vol. 18, pp. 683697.
Sodano, H. A., Park, G., Leo, D. J., and Inman, D. J., 2004b, Model of Piezoelectric Power Harvesting Beam, in ASME International Mechanical Engineering Congress and Expo, November 1521, Washington, DC, Vol. 40, No. 2.
Srinivasan, A. V., and McFarland, D. M., 2001, Smart Structures: Analysis
and Design, Cambridge University Press, Cambridge.
Starner, T., 1996, Human-Powered Wearable Computing, IBM Systems
Journal, Vol. 35 No. 34, 618628.
Umeda, M., Nakamura, K., and Ueha, S., 1996, Analysis of Transformation of Mechanical Impact Energy to Electrical Energy Using a Piezoelectric Vibrator, Japanese Journal of Applied Physics, Vol. 35,
Part 1, No. 5B, 32673273.

Sodano et al. / A REVIEW OF POWER HARVESTING

Umeda, M., Nakamura, K., and Ueha, S., 1997, Energy Storage Characteristics of a Piezo-Generator using Impact Induced Vibration, Japanese Journal of Applied Physics, Vol. 36, Part 1, No. 5B, 3146
3151.

205

Williams, C. B., and Yates, R. B., 1996, Analysis of a Micro-Electric Generator for Microsystems, Sensors and Actuators, Vol. 52, No. 13,
811.
Worden, K., Bullouh, W. A., and Haywood, J., 2003, Smart Technologies,
World Scientific, River Edge, NJ.