Chapter 1

Introduction
1.1 General
Low-power wireless distributed sensor networks are becoming attractive for
monitoring different variables such as temperature, strain in a material, or air pressure
over a wide area. However, one drawback of these networks is the power each node
draws, though recent work has shown this can be lowered considerably. Batteries can be
used to power nodes for extended periods of time, but they have a limited life cycle and
eventually need to be replaced. As this can be a costly and time consuming procedure for
networks with many nodes, a means of powering the devices indefinitely would be a
more practical solution.
Solar power provides a considerable amount of energy per area and volume, but
unfortunately is limited to applications that are reliably sunlit. A promising alternative
takes advantage of the energy in ambient vibrations and converts it to electrical power.
This approach compares very favorably with batteries, providing equal or greater power
per unit volume. [1]
There are multiple techniques for converting vibrational energy to electrical
energy. The most prevalent three are electrostatic, electromagnetic, and piezoelectric
conversion. A majority of current research has been done on piezoelectric conversion due
to the low complexity of its analysis and fabrication.

1.2 The Piezoelectric Effect

The piezoelectric effect, in essence, is the separation of charge within a material as
a result of an applied strain. This charge separation effectively creates an electric field
within the material and is known as the direct piezoelectric effect. The converse
piezoelectric effect is the same process in reverse: the formation of stresses and strains in
a material as a result of an applied electric field. The IEEE standard on piezoelectricity
lists several different forms for the piezoelectric constitutive equations [2]. The form used
here is known as the d-form, and the equations are as follows:
S = sET + dE
D = dT +T E
These equations, known as the coupled equations, reduce to the well-known
stress-strain relationship at zero electric field, and the electric field and charge
displacement relationship at zero stress.

Figure1.1: piezoelectric effect cause crystal materials like quartz to generate

electric charge when the crystal material is compressed, twisted, or pulled. The reverse is
also true (www.cosmic-energy.org).

1.3 Piezoelectric Material

ZnO is the most attractive since it exhibits the coupling of piezoelectric and
semiconducting properties as well as the formation of a barrier at the electrode contact to
draw higher power from the ambience. The ZnO has a non-Centro symmetric structure,
which naturally exhibits a piezoelectric effect when subjected to a strain due to the
displacement of Zn+2 cations and O2 anions in its tetrahedral configuration. Self-powered
MEMS systems demand sustainable low voltage, high current characteristic inputs, and
easy integration with the MEMS device, and the ability to be controlled with reliable
transduction mechanisms. As miniaturized piezoelectric transducers have a high resonant
frequency, they cannot be used for harvesting ambient vibrations. Low resonance
frequency structures have macro dimensions that impose integration challenges on
MEMS systems.

1.4 Photosensitive polymer

SU-8 photosensitive cross-linking polymer that can be patterned for high aspect
ratio structures. Since SU-8 is highly flexible and has a very low Youngs modulus (27
GPa), it is commonly used as a structural layer in MEMS. Though this polymer is
inherently an electrical insulator with no additional functionalities. The chemical
resistance of SU-8 is quite excellent however most chemical etches it, albeit at a very
slow rate. This very slow rate creates diverging report for the etching resistance of SU8.
Adhesion of SU-8 is usually good, but depends on the material. The adhesion is worst
with gold (Au), average with silicon with native oxide, and best with silicon nitride (SiN)
and ZnO. The adhesion of SU-8 seems to be affected by the chemical and SU-8 lift-off
with immersion in KOH. [5].

1.5 Photoplastic piezoelectric Nanocomposite

Currently, there has been a growing interest in scientific community on
developing microelectromechanical-systems (MEMS)-based piezoelectric sensors and
energy-harvesting devices. The most common piezoelectric materials have been the
piezoceramics, with the leading candidates being lead zirconate titanate (PZT), barium
titanate, strontium titanate, and quartz-based structures, which possess high strain
response. There has also been an interest in developing alternative materials, and recently,
piezoelectric ZnO has attracted a lot of attention. It finds applications in MEMS due to its
unique combination of electrical, optical, and piezoelectric properties [6]. However, ZnO
is a sensitive material for wet etching and treatment by temperature, acids, bases, and
even water. Thus, for the successful fabrication of ZnO-based MEMS, a dry etching
technique is needed [6], [5]. Moreover, most of the piezoelectric materials are ceramics,
which are brittle and have low fracture toughness, posing challenges in fabrication. The
mixing of nonmaterials into polymer matrices offers the promise of developing new
polymer composite materials with extraordinary properties. Nevertheless, the
development of these materials has been constrained by difficulties in incorporation and
dispersion of these materials into polymer media because of the strong tendency of the
nanoparticles (NPs) to agglomerate.
Youngs modulus and hardness values are measured for ZnO NWs using the
nanoindentation technique. Also, the same experiment was performed on a ZnO NW
array after embedding these NWs inside the SU-8 matrix. The elastic modulus of vertical
NWs as measured by nanoindentation comes in the range 7293 GPa, which is close to

the ZnO thin-film data. After embedding these NWs inside an SU-8 matrix, the Youngs
modulus values are seen to be in the range of 519 GPa.[7]

Figure 1.2: Youngs modulus measurements of Nanocomposite thin films with different
ZnO concentration estimated using a nano-indentation method. [7]

1.7 problem statement

To increase the sensitivity (i.e. output voltage) of the energy harvesting

material by selecting proper material and geometric structure to get
maximum power output.

1.8 Objectives and scope of the project

To analyze the different structure for the material for which we get
maximum voltage for the minimum deflection by using COMSOL
Multiphysics.
Analysis of the SU 8/ZnO Nanocomposite material for the lower resonant
frequencies i.e. for the ambient vibrations by altering the structure of the
material and adding the proof mass.

1.9 Limitations

Fabrication will be difficult.

Cannot used for large power production.

Chapter 2
Literature review
This chapter addresses a literature review is presented in which previous works in
fields of vibration energy harvesting are highlighted.

2.1 S Roudy and P K Write (2004)

The focus of this is to discuss the modeling, design, and optimization of a
piezoelectric generator based on a two layer vending element, the model has been
validated and use not only to estimate power output under a given set of conditions, but
also as the basis for generator design optimization. Furthermore, an analytical model of
the generator has been developed and validated. In addition to providing intuitive design
insight, designs of 1 cm3 in size generated using the model have demonstrated a power
output of 375 w from a vibration source of 120Hz.

2.2 M. Guizzetti et al., (2009)

This paper shows the FEM simulations of piezoelectric cantilever as a
microgenarator. Piezoelectric energy converters realized in a cantilever configuration are
the most studied for this purpose, In order to improve the performances of the converter,
the geometry has to be properly designed. In this context FEM simulations have been
used in order optimize the piezoelectric mode. The electrical energy generated by the
converted under an applied acceleration is computed, finding the optimal thickness for the
piezoelectric layer. Different geometries were considered verifying that they do not affect
the optimal thickness. Geometries with different dimensions were considered, verifying
that the optimal thickness ratio tpzt/t substrate is independent from the converter
dimensions, and it is influenced only by the mechanical properties of the piezoelectric
layer and the substrate.

2.3 Suyog N Jagtap et al., (2011)

This paper gives investigation of design and modeling of piezoelectric cantilever
for energy harvesting. MEMS based energy harvesting device is designed to convert

mechanical vibration energy via piezoelectric effect. In order to improve the performance
of the device, the geometry has been optimized by using moving mesh ALE model
available in COMSOL Multiphysics. The proposed device is suitable for vibration energy
harvesting and can be uses as potential micro generator.

2.4 XiaotongGao (2009)

Traditional piezoelectric Cantilever use piezoelectric and no piezoelectric layers
of the same length. This paper shows the investigation of unequal piezoelectric and no
piezoelectric lengths namely two section piezoelectric cantilever. For step-wise tip forces
the results showed that longer non piezoelectric layer is preferred for generation a higher
induced voltage while a longer piezoelectric layer reduced the induced voltage due to
charge cancellation. With harmonic base vibrations, the results showed that there exists an
optimal no piezoelectric piezoelectric length ratio at which output voltage, current, and
power can be maximized. Theoretical analysis of two-section PCs was performed within
the framework of beam theory. The results were in good agreement with experiments.

2.5 K. Prashanthi et al., (2012)

This letter reports a Photoplastic (SU-8) piezoelectric (ZnO) Nanocomposite route
for realization of simple and low cost piezoelectric microelectromechanical systems
(MEMS). Integrating the ZnO nanoparticles into a photosensitive SU-8 polymer matrix
not only retains the highly desired piezoelectric properties of ZnO but also combines the
photopatternability and the optical transparency of the SU-8 polymer. These two aspects,
therefore, lead to exciting MEMS applications with simple photolithography-based
microfabrication. This approach opens up many new applications in the field of both
sensor and energy harvesting.

2.6 Adel Campo and CGreiner (2007)

SU-8 has become the favorite photoresist for high-aspect-ratio (HAR) and threedimensional (3D) lithographic patterning due to its excellent coating, planarization and
processing properties as well as its mechanical and chemical stability. Simple SU-8
structures (pillars and walls) with aspect ratio above 100 but a maximum lateral
resolution of 8m have been reported after UV exposure. Better lateral resolution (<1m)
maintaining high aspect ratios has been achieved using x-rays. High-energy beams may
render even smaller structures, but with limited aspect ratio.

2.7 Khaled Ramadan, Ian G. Foulds [2011]

A 1500m x1500m x 150m out-of-plane, gap closing, electrostatic energy
harvester is designed and fabricated to harvest low-frequency ambient vibrations. SU-8 is
used to fabricate the proof mass (1200m x 1200m x 150m) and the 5m springs.
Different harvesters were designed to harvest at 50, 75 and 110 Hz. At 110 Hz,
Simulations show that with an input vibration of 10m amplitude at the frequency of
resonance of the structure, the energy harvester should generate an average output power
density of 0.032W/mm3.

2.8 Matthew Hopcroft, Tobias Kramer (2003)

The mechanical properties of these cantilevers were investigated using two
microscale mechanical testing techniques: contact surface profilometer deflection, known
as MAT-Test, and static load deflection using a specially designed test machine, the
MFT2000. The Youngs modulus values from the microscale test methods are
approximately 2-3GPa. These results are compared with results from macro scale tests of
4-5GPa.

2.9 Manoj Kandpal et al.,(2012)

Photo-curable Nanocomposite material was formulated by embedding ZnO
nanoparticles into a SU-8 matrix and studied for its piezoelectric properties for low cost
fabrication of self-powered nanodevices. The piezoelectric coefficient of ZnO
nanoparticles was observed to be ranging between 15 and 23 pm/V, which is the highest
reported. These experimental studies support the recent theoretical predictions where the
piezoelectric coefficients in ZnO nanoparticles were found to be higher compared to the
thin films because of the surface relaxation induced volume reductions in the nanometer
scale.

2.10 Prasenjit Ray et al.,(2014)

The mechanical properties of piezoelectric ZnO nanowire (NW) films have been
measured using a nano indentation technique. We demonstrated a fabrication process to
embed this NW film inside a polymeric (SU-8) cantilever. Mechanical properties of this

SU-8/ZnO NW film and cantilever have been measured. The ZnO NWs have been grown
vertically using a low temperature chemical synthesis. The observed value of Youngs
modulus for ZnO NW is in the range 7293GPa. After embedding this NW film inside a
polymer matrix, the Youngs modulus value was reduced to the range 519GPa.

2.11 K. Prashanthi et al.,(2013)

Integrating piezoelectric zinc oxide (ZnO) nanoparticles (NPs) with optimum
weight fraction into a photosensitive SU-8 polymer matrix not only retains the highly
desired piezoelectric and semiconducting properties of the ZnO, but also combines the
photopatternability and the optical transparency of theSU-8. The maximum output power
produced by Nanocomposite cantilevers was 0.025W across a resistive load of 100k
with peak to peak voltage of 140mV at a resonance frequency of 4 kHz.

2.12 Mr. Ravi prakash at al.,(2012)

In this paper some basic energy harvester is modeled using piezoelectric cantilever
structure with big mass at the end of tip to lower the resonance frequency. There by
ambient vibration source (50-500 Hz) is converted into mechanical stress and finally into
electricity. These designed and simulated using the COMSOL Multiphysics. A composite
cantilever with dimension 250m x 500m x 6m is used to harvest vibration energy. A
proof mass of 0.145 mg is placed on tip of cantilever beam to lower the resonance
frequency and to maximize the out voltage of 2.1mV at 485Hz.

Chapter 3
Experimental Methodology
3.1 Introduction
3.1.1Anisotropic effect and coupling modes
Piezoelectric materials have a built-in polarization, and therefore respond
differently to stresses depending on the direction. There are two primary modes of
electromechanical coupling for piezoelectric materials: the 3-1 mode and the 3-3 mode. In
the 3-1 mode (Figure 2.1a), the electric field is produced on an axis orthogonal to the axis
of applied strain, but in the 3-3 mode (Figure 2.1b), the electric field produced is on the
same axis as the applied strain.

Figure 3.1: 3-1 mode, the electric field is produced on an axis orthogonal
to the axis of applied strain [13]

Figure 3.2: 3-3 modes, the electric field produced is on the same axis
as the applied strain. [13]
While the piezoelectric coefficient is higher in the 3-3 mode for
most materials, taking advantage of the larger coefficient requires a
much more complex design. Instead of simple planar electrodes, a
series of interdigitated electrodes (IDE) can be used to take advantage
of the 3-3 coupling mode [7].

3.1.2 Device Configuration

The vast majority of piezoelectric energy harvesting devices uses a cantilever
beam structure. A cantilever beam, by definition, is a beam with a support only one end,
and is often referred to as a fixed-free beam. When the generator is subjected to
vibrations in the vertical direction, the support structure will move up and down in sync
with the external acceleration. The vibration of the beam is induced by its own inertia;
since the beam is not perfectly rigid, it tends to deflect when the base support is moving
up and down (see Figure 2.3). Typically, a proof mass is added to the free end of the beam
to increase that deflection amount. This lowers the resonant frequency of the beam and
increases the deflection of the beam as it vibrates. The larger deflection leads to more
stress, strain, and consequently a higher output voltage and power. Electrodes
covering a portion of the cantilever beam are used to conduct the
electric charges produced to an electrical circuit, where they can be utilized to charge
a capacitor or drive a load. Different electrode lengths or shapes have been shown to
affect the output voltage, since strain is not uniform across the beam. [8]

Figure 3.3: note that strain is generated along the length of the beam,
hence the use of the 3-1 mode [13]

3.1.3 Modes of Vibration and Resonance

A cantilever beam can have many different modes of vibration, each with a
different resonant frequency. The first mode of vibration has the lowest resonant
frequency, and typically provides the most deflection and therefore electrical energy. A
lower resonant frequency is desirable, since it is closer in frequency to physical
vibration sources and generally more power is produced at lower
frequencies. Therefore, energy harvesters are generally designed to
operate in the first resonant mode.

Figure 3.4: Different mode shapes of a vibrating beam.[13]

3.2 Modeling MEMS and Piezoelectric Devices using COMSOL

Multiphysics.
MEMS and Piezoelectric devices and transducers are growing in use as their
technology becomes more sophisticated. Simulation is an important part in understanding,
designing, and optimizing them due to their size and complexity. Piezoelectric devices are
used in a wide range of applications such as actuators, sensors and ultrasonic devices. In
electronic devices, they are often deployed as precision frequency resonators in, for
example, quartz wrist watches. Direct piezoelectric behavior occurs in certain materials
where an applied mechanical stress results in the build-up of electric charge or a voltage,
while the inverse piezoelectric effect occurs when an electric potential induces
mechanical deformations. Related to this are devices based on piezoresistive and
magnetostrictive effects. [14] Energy harvester is designed and modeled in the COMSOL
Multiphysics for the anisotropic material and coupling modes, resonance frequency and
voltage for the deflection of the cantilever beam are obtained by solving it.

Chapter 4
Result and Discussion
Modeling and simulation of cantilever beam is been done in COMSOL
Multiphysics using the electrostatic analysis, Bimorph cantilever Energy harvester
consists of two layers, upper layer consist of the ZnO and lower layer is of SU 8 material
is operated in d-33 mode.

4.1 bimorph cantilever beam of ZnO and SU 8

4.1.2 Simulation :
The first six model resonance frequency analysis is carried out to find the lower
resonance frequency. The results related to first six frequencies are as shown in figure (16). The first frequency gives the lower resonance frequency.

Figure 4.1: 1st mode of resonance frequency

Figure 4.2: 2nd mode resonance frequency

Figure 4.3: 3rd mode of resonance frequency

Figure 4.4: 4th mode of resonance frequency

4.1.3 Static analysis

A force of 50N per unit area applied on the upper surface of the bimorph for the
d31 mode is selected by making electric potential for the upper surface and grounding
lower face of the ZnO layer, while all other face of piezoelectric layer are kept as zero
charge constrained.

Figure 4.5: bimorph cantilever beam

Figure 4.6: meshed model of bimorph

Figure 4.7: frequency and voltage

The transducer cantilever depends upon the length, width, thickness and the
various properties of the material used make the structure. The geometric shape of the
structure as well as the material used to build the cantilever determines the cantilever
stiffness. Various dimensions of cantilever beam simulations are carried out to get the
optimum dimensions for getting high voltage for lower resonance frequency of the
structure. Tables show the different dimensions and its frequencies, displacement and
voltage.
Dimensions(
m

Frequency(kHz)

Displacement(
m

Voltage(mV)

100 20 3

92.716

0.0282

0.55611

200 20 3

22.993

0.4582

2.154

300 20 3

10.187

2.334

4.783

400 20 3

5.725

7.391

8.44

500 20 3

3.660

18.078

13.172

Table 4.2: varying length of the cantilever beam and its frequencies, displacement,
voltage.

Dimensions(
m

Frequency(kHz)

Displacement(
m

Voltage(mV)

500 40 3

3.6783

17.902

12.126

500 60 3

3.6958

17.744

11.673

Table 4.3: varying breadth of the cantilever beam and its frequencies, displacement,
voltage.
From table (4.1) as length increases the frequency decreases and voltage
increases. After the length 500 m the edges and the face is much smaller than the
specified element size. And keeping 500 m constant, as the breadth increases the
frequency increase and voltage decreases, so for lower operating frequency and higher
voltage we assume optimum dimension as 500 m 20 m . For the thickness of 4
m , frequency is 6.701 kHz and voltage is 11.213V. i.e. the frequency increases and
voltage decreases, so the dimension we chose for the higher frequency is 500 m 20
m 3 m .

4.2 cantilever of SU 8/ZnO nano composite material

For the 20% of ZnO in the SU 8 matrix the strength of the SU 8/ZnO nano
composite material increases and the mechanical properties are as tabulated bellow.
Youngs modulus (GPa) Poissons ratio
Density(kg/m3)
SU8/ZnO nano
12
0.25
13800
composite
material
Similar simulation is carried out for the optimum dimensions 500 m 20 m 3
m Nanocomposite material, resonance frequencies are as shown in the bellow
figures, from that the lower resonance frequency is chosen to be 1.2 kW.

Figure 4.8: 1st-6th resonance frequencies of Nanocomposite material

4.2.1 Static analysis

A force of 10N per unit area applied on the upper surface of the cantilever for the
d31 mode is selected by making electric potential for the upper surface and grounding
lower face of the SU 8/ZnO Nanocomposite layer, while all other face of piezoelectric
layer are kept as zero charge constrained. Maximum displacement observed as shown in
the figure bellow, i.e. total displacement is 9.720x108 m .

Figure 4.9: total displacement of SU 8/ZnO Nanocomposite cantilever beam

4.2.2 Proof mass

Attaching the proof mass to the Nanocomposite material the resonance frequency
will get lowered and deflection increase for that voltage is going to increase.
For the optimized model, the proof mass of size 100 m 20 m 5 m
is attached to the Nanocomposite cantilever beam, by that the resonance frequency going
to reduce to a value 762.73Hz i.e. it will deflect at lower frequency and produce the more
voltage.

Figure 4.10: Resonance frequencies from 1st to 6th node frequencies.

Figure 4.11: total displacement of the Nanocomposite with proof mass.

5 Conclusions
Design and simulation of bimorph energy harvest is designed for an applied load
of 10N per unit area, static displacement and voltage as shown in table 4.2 and 4.3. the
cantilever dimension should be chose for maximum energy harvesting for lower
frequency(3.660kHz) and high voltage, for that dimension found out to be 500 m 20
m 3 m . Because of the brittleness of the ZnO material, it is added into the SU 8

matrix making Nanocomposite material results in good piezoelectric and

Photopatternable material. Its resonance frequency for the optimized dimensions is 1.2
kHz, to reduce the resonance frequency proof mass of 100 m 20 m 5 m is
attached to the Nanocomposite material, by that the resonance frequency changes to value
of 762.73Hz.

Reference
[1] S Roudy and P K write A piezoelectric vibration based generator for wireless
electronic, Institute of physics Publishing, Smart Mater. Struct. 13 (2004) 1131-1142.
[2] M. Guizzetti et al., Thickness optimaization of a piezoelectric Converter for energy
Harvesting Excerpt from the proceeding of the COMSOL conference 2009 milan.
[3] Suyog N Jagtap and Roy Paily Geometry optimization of a MEMS based Energy
Harvesting device proceeding of the 2011 IEEE students Technology Symposium 14-16
January, 2011. IIT kharagur.
[4]XiaotongGao Vibration and Flow Energy Harvesting using Piezoelectric A thesis
submitted to the Faculty of Drexel University.
[5] K. Prashanthi, M. Naresh et al.,A Novel Photoplastic Piezoelectric Nanocomposite
for MEMS Applications journal of microelectromechanical systems, vol. 21, no. 2, april
2012.

[6] A del Campo and CGreiner, SU-8: a photoresist for high-aspect-ratio and 3D
submicron lithography J. Micromech. Microeng. 17 (2007) R81R95.
[7] Khaled Ramadan, Ian G. Foulds, Fabrication of SU-8 Low Frequency Electrostatic
Energy Harvester. 978-1-4673-0465-8/11/$26.00 2011 IEEE.
[8] Matthew Hopcroft, Tobias Kramer, Micromechanical testing of SU-8 cantilevers
ATEM'03, JSME-MMD, Sep. 10-12, 2003.
[9] Manoj Kandpal, et al., Photopatternable nano-composite (SU-8/ZnO) thin films for
piezoelectric applications applied physics letters 101, 104102 (2012).
[10] K. Prashanthi et al.,Vibtrational energy harvesting using Photopatternable
piezoelectric Nanocomposite cantilevers Nano Energy (2013) 2, 923932.
[11] K. Prashanthi et al., Plastic Deformation Study of Vertical Zinc Oxide Nanowire for
Polymer Cantilever-Based Sensor Applications IEEE transactions on nanotechnology,
vol. 13, no. 4, July 2014.
[12] Mr. Ravi prakash at al., Study of Piezoelectric MEMS Transducers.
[13] Andrew Townley Vibrational energy harvesting using MEMS piezoelectric
generators Electrical Engineering, University of Pennsylvania
[14] http://www.comsol.co.in/showcase/mems#ctaframe