RSM2013 Proc. 2013, Langkawi, Malaysia

Design and Simulation of Piezoelectric Micro Power Harvester for Capturing Acoustic Vibrations

Mohd H.S Alrashdan, Burhanuddin Yeop Majlis, member, IEEE, Azrul Azlan Hamzah and Noraini Marsi Institute of Microengineering and Nanoelectronics (IMEN);

Universiti Kebangsaan Malaysia; 43600, Bangi, Selangor, Malaysia Email: burhan@vlsi.eng.ukm.my

Abstract-Piezoelectric Micro-Power Harvester (PMPH), harvests mechanical vibration sources available in the environment and converts it to usable electric power via piezoelectric effects. The low power requirements and small device dimensions enable PMPH to supply enough power necessary to a variety of applications such as wireless sensor nodes, wrist watches and cell phone signals, thus proving to be an excellent alternative source for traditional lithium iodide battery especially in body sensor nodes. In this paper we design PMPH that is able to harvest environmental vibration sounds and convert it to usable electrical power for artificial cochlea. Spring mass damper system with single degree of freedom is used to model PMPH. COMSOL Multiphysics 4.2 is used to simulate PMPH. a linear relationship between voltage and external load for piezoelectric materials during Static analysis is observed, Eigenfrequency is used to find the resonance frequencies for six modes of operation and its deflection shape, PMPH harvest the maximum acoustic vibration at first mode of operation at 589 Hz . Simulation results using Transient analysis show that PMPH total displacement about 6 μm and output voltage at center of piezoelectric material about 4*10-15Vp-pat steady state and can harvest acoustic vibration at 598Hz and convert it to electric power about 23nW, which is sufficient for cochlear implant application.

I. INTRODUCTION

Smart materials technologies and micro-electromechanical

systems (MEMS) are mature, and they became one of the most promising fields in engineering in the last decade; because of very fast diminish of electronic device dimensions and its necessary power needed specially in remote areas and medical implants.

Normal batteries such as lithium iodide are generally used as power supply of electric energy; the short life span, and high ratio between mass and electrical power for these traditional batteries, coupled with Hazards and high replacement cost especially in medical implant applications, motivates researchers and scientist for developing alternative power harvesting devices.

At the micro level, there are more than one technique for power transformation and harvesting such as electrostatic[1, 2], electromagnetic[3-5] and piezoelectric materials[6-9]. Piezoelectric technique has received much of the attention because of both having a better electromechanical coupling and no external source needed.

PMPH harvests mechanical vibration sources available in the environment and converts it to usable electric power via piezoelectric effects. The low power requirement and small device dimensions enables PMPH to supply enough power necessary to a variety of applications such as wireless sensor nodes, wrist watches and cell phone signals, thus showing an excellent alternative source for traditional lithium iodide battery especially in body sensor nodes. In this paper we design and simulate PMPH based on composite micro-cantilever beam with top proof mass able to harvest environmental vibration sounds and convert it to usable electrical power for artificial cochlea.

II. THEORY

A. PMPH Mechanical Model.

The characteristic performance of any dynamical system such as PMPH is exclusive and can be represented by its resonance frequency , stiffness and damping constant [10]. The dynamic characteristic of the PMPH being studied in this paper is investigated by studying the spring mass system .Initially PMPH with an infinite number of coordinates are chosen where there is no constrain cause PMPH to move in free space. Normally engineering problems deal with the first few mode of operation which cause feasible results come out for six (DOF). PMPH is designed to harvest maximum output power available in environments. So we are interested in PMPH motion and its shape in one single mode where the maximum displacement and hence maximum mechanical energy harvesting and maximum output power occur. spring mass damper system with single (DOF) is feasible to describe PMPH[11]; which led to study the response of PMPH in an easier and more effective way as shown in Fig.1(a ,b).

Fig.2 shows the circuit model for PMPH with storage system necessary to convert AC to usable DC output power.

Equation of motion for first DOF spring mass system as in (1) is achieved by second Newton law and equation of energy balance.

(1) Where M is mass in (Kg) which is dominated by mass

of proof mass .d: is the net mass displacement (m). : Net mass velocity. : Net mass acceleration. C is the damping constant.

383 978-1-4799-1183-7/13/$31.00 ©2013IEEE

RSM2013 Proc. 2013, Langkawi, Malaysia

K is spring mass constant representing cantilever beam stiffness. : input acceleration.

a

b

Fig. 1 (a) Cantilever beam with tip mass, (b) equivalent spring mass damper system with first (DOF)

Fig. 2. Circuit diagram of PMPH and storage system

. (2)

Where n is resonance frequency, is damping factor

Equation (2)implies that for maximum performance and maximum PMPH output power generation. The PMPH frequency should be equal to the natural frequency, which can be written as (3).

(3) Equation (3) describes the most important parameters that

influence the output power generated from PMPH, which can be achieved by increasing the mass of our cantilever beam which is dominated in by proof mass and also increasing the amplitude of external vibration source , increasing the natural frequency by choosing suitable material with suitable dimensions and thicknesses, and finally decreasing the damping factor as much as possible by increasing the PMPH quality factor by mean of choosing good MEMS fabrication techniques with appropriate conditions .

B. Cochlear Implant

A cochlear implant (CI) is an electronic device that can help a person who is profoundly deaf for sensing sound, and it is implanted using medical surgery. Cochlear implants are frequently called bionic ears. Cochlear implants possibly will help give hearing in patients that are not deaf as a result of injure to sensory hair cells in their cochlea’s. In those patients, the implants often can permit enough hearing for enhanced understanding of speech[12].The implant is surgically sited underneath the skin at the back the ear. The necessary parts of the device contain at least one microphone which picks up sound from the surroundings, a speech processor which carefully filters sound to prioritize audible speaking, breaks up

the sound throughout channels and sends the electrical sound signals throughout a fine wire to the

transmitter, a transmitter, which is a helix shape detained in place by a magnet positioned behind the external ear, and convey power and the processed sound signals across the skin to the inner device by electromagnetic induction, a receiver and stimulator positioned in bone under the skin, which converts the signals into electric impulses and transmits them through an inner cable to electrodes ,an array of up to 22 electrodes wound through the cochlea, which sends the impulses to the nerves in the Scala Tympani and then directly to the brain through the auditory nerve system.[12]

The audible frequency range of humans is between 20-20000 Hz. We designed PMPH to resonate at 589Hz which matches with some available sources of sounds in the environment such as Rock, Blues music [13].

III. METHODOLOGY

A. PMPH Device

PMPH is MEMS energy harvesting device. It consists of a composite micro cantilever beam with a rectangular shape , which has a significant role in improving the PMPH efficiency and most frequently used structure in MEMS-power harvesting devices [14].with proof mass. First the membrane layer of silicon nitride for controlling stress and bow of cantilever structure is deposited on silicon wafer. Second layer of Pt is deposited on previous layer as bottom electrode, buffer and diffusion barrier to prevent lead particles and electrical charge to flow from piezoelectric layer to silicon wafer. Also for preventing lead migration from PZT layer to silicon wafer during poling process. Third layer of PZT material, which is the functioning material in PMPH design. Fourth layer of Pt as a top electrode to perform d31 mode of operation .and SU8 photoresist as proof mass for controlling PMPH resonance frequency. Neutral axes are the axes where the beam doesn’t show tension or compression during vibration. The location of neutral axes is important to be outside the functioning piezoelectric layer. B. Simulation

Using COMSOL Multiphysics 4.2, Free Mesh Parameter was used to set the mesh. The mesh quality and its predefined mesh size, was chosen to a Coarser, X and Y directions scale factor of 0.1, Z direction scale factor of 100. With maximum element scale factor of 1.9, with element growth rate 1.7 and mesh curvature factor of 0.8 and curvature cut off of 0.05.

COMSOL Multiphysics 4.2 software is used for PMPH.MEMS Module - Structural Mechanics –Piezo solid have been used to conduct three basic analysis types which is Static, Eigenfrequency, Transient analysis.

Initially, a statical analysis using linear system solver DIRECT SPOOL has been used, where a constant load distributed has been considered at the right end of the cantilever beam. This load represents the force induced to

384

RSM2013 Proc. 2013, Langkawi, Malaysia

PMPH due to Proof mass weight which can cause deflection of PMPH starts at its free ends without applying and external force. Initial deflection important to find the etching depth below cantilever beam at free ends.

Piezoelectric materials have linear relationships between applied voltage and total displacement (actuator mode ) and vice versa (sensor mode).The boundary conditions for the static problem were chosen such that the fixed supports at the left beam ends were assumed ideal, while the rest of the structure had no constraints (free).all piezoelectric boundary electric condition are set to zero charge symmetry except the top side which was set as electric potential and the bottom side was set as ground. In actuator mode, different loads (0-2000 N/m) with irregular step size is applied at right free end in sensor mode.

During free vibration, PMPH operates at more than one resonance frequency; the purpose of the eigenfrequency analysis is to find the lowest six resonance frequencies fn and their subsequent shape modes. In order to find the most suitable modes of operation for PMPH. During eigenfrequency analysis we use exactly the same material, load and constraints as the statics analysis. Linear system solver DIRECT SPOOL was used.

A transient analysis is used to find the voltage and displacements as functions of time. Damping is very important in this analysis. Rayleigh damping is used, indicating damping parameters proportional to the mass and stiffness , boundary conditions and loads are functions of time. The main aim of this analysis is to find the transient response to a sinusoidal load with the same amplitude as the static load during the first few hundreds of cycles.

IV. RESULTS AND DISCUSSION.

Neutral axis location is found for cantilever beam composed

of four layers of (Si3N4, Pt, PZT5H, Pt) from bottom to top. material properties for this PMPH cantilever beam is shown below in table 1 .The computation results for PMPH neutral axis shows that the =770nm. This is exactly located at the top of PZT layer within Pt layer; leading to prevent charge cancelation during vibration[15] and also a maximum efficiency and output power generation from PMPH device are expected to occur.

Meshing results shows 848 nodes and 1740 triangular elements and 3153 tetrahedral elements, with 18245 degrees of freedom, an element quality of 1.84*10-4 and an element volume ratio of 0.0024.

TABLE I. PMPHMATERIAL, DIMENSION AND PROPERTIES

Material properties Material Properties Si3N4 Pt PZT5H Pt SU8

Young Modulus (GPa) 250 168 200 168 4.02

Thickness(nm) 200 20 500 20 22000 Width(mm) 1 1 1 1 1 Length(mm) 3 3 3 3 0.2 Poisson ratio 0.23 0.38 0.33 0.38 0.22

The initial deflection due to SU8 weight is measured and is found to stand at 0.40μm a.

Simulation result in sensor mode for is shown in Fig.3.which confirms linear relationship between applied load and voltage response. For the same loads we determined the displacement on y axis of the right side of the beam Fig. 4.This has the same linear relationships. The quality factors in for PMPH in sensor and actuator Mode are almost the same which is equal to31in actuator mode and (1/0.028) which is equal to 35in sensor mode. This means the PMPH is well designed [16].

Resonance frequency for the first six modes of operations is presented below intable 2.A PMPH shape during free vibration at these frequencies is important to determine its behavior in real world. maximum total displacement at center of proof mass for PMPH occurs at first mode of operation at 589 Hz as shown in Fig.5, while the displacement at the other mode is negligible, therefore maximum PMPH output power occurs at this mode, the first mode of operation is needed not only because of the maximum displacement but also because the other modes do not show uniform bending of cantilever beam. Some parts of cantilever beam show compression while others show tension and cause charge cancellation along cantilever beam. This reduces PMPH output power and its efficiency.

Rayleigh damping parameters proportional to the mass and stiffness is 71.43(1/s), and 1.8869*10-8(s) respectively. The excitation frequency has been taken 589 Hz. The displacement, output voltage, and electric energy density during the first 30 seconds are shown in Fig.6, Fig.7,and Fig.8 respectively. PMPH reaches steady state about 7 sec from beginning. Electric energy density during transient analysis is found for one point in piezoelectric layer. We calculate it first over all piezoelectric layers and then differentiate it over time to estimate output electric power of 23nW.

Fig.3.PMPHvoltage due to applied load (sensor mode).

Fig.4.PMPHTotal Displacement Vs applied load.

y = 0.028x - 0.187

-200

20406080

0 1000 2000 3000

Vol

tage

(V)

Applied load (N/m)

-10

0

10

20

0 500 1000 1500 2000 2500

Tota

l D

icip

lacm

ent

(m)

Applied load (N/m)

385

RSM2013 Proc. 2013, Langkawi, Malaysia

TABLE II

EIGENFREQUENCIES RESULTSFOR FIRST SIX MODES

Fig.5: PMPH Total Displacement at Center of Proof Mass.

Fig.6: PMPH Total Displacement at Center of Proof Mass (Transient Analysis).

Fig.7: Electric Potential (Transient Analysis).

Fig.8: Electric Energy Density (Transient Analysis).

V. CONCLUSION.

We designed, modeled and simulated PMPH that is able to

harvest acoustic vibration normally available in music sound frequency of 589 Hz and converts it to usable electric power in cochlear implant applications. COMSOL Multiphysics is used for static, Eigenfrequency, and Transient analysis. The results

are in qualitative acceptance. PMPH power system has verified that it can generate the necessary power for cochlear implant.

ACKNOWLEDGMENT

The authors would like to thank Universiti Kebangsaan

Malaysia for supporting this project under grant UKM-GUP-NBT-08-25-084.

REFERENCES

[1] Roundy, S., P.K. Wright, and K.S.J. Pister. Micro-Electrostatic Vibration-to-Electricity Converters.in ASME International Mechanical Engineering Congress & Exposition. neworleans, louisiana , USA, 2002.

[2] Mitcheson, P.D., et al., “MEMS electrostaticmicropower generator for low frequency operation,” Sensors and Actuators,” vol115(2-3 SPEC. ISS.): p. 523-529, 2004.

[3] Williams, C.B. and R.B. Yates, “Analysis of a micro-electric generator for Microsystems,” Sensors and Actuators , Pro.,vol 52(3-1):p. 8-11,1996.

[4] Glynne-Jones, P., et al., “An electromagnetic, vibration-powered generator for intelligent sensor systems,” Sensors and Actuators,Hampshire, UK,Vol110: p. 344-349,2004.

[5] Arnold, D.P., “Review of Microscale Magnetic Power Generation” Magnetics, IEEE Transactions on, Vol43(11): p. 3940-3951, 2007.

[6] Roundy, S., P.K. Wright, and J. Rabaey, “A study of low level vibrations as a power source for wireless sensor nodes” Computer Communications,vol26(11): p. 1131-1144, 2003.

[7] Sodano, H.A., D.J. Inman, and G. Park, “Comparison of piezoelectric energy harvesting devices for recharging batteries,” Journal of Intelligent Material Systems and Structures, vol16(10): p. 799-807, 2005.

[8] Jeon, Y.B., et al., “MEMS power generator with transverse mode thinfilm PZT,” Sensors and Actuators,Vol 122(1 SPEC. ISS.): p. 16-22, 2005.

[9] Liu, H., et al., “A MEMS-based piezoelectric cantilever patterned with PZT thin film array for harvesting energy from low frequency vibrations,” Physics Procedia,Vol19 )0:( p. 129-133, 2011.

[10] Kim, H., Y. Tadesse, and S. Priya, Piezoelectric Energy Harvesting, in energy harvesting technologies ,Springer: New York ,. 2009.

[11] Laura, P.A.A., J.L. Pombo, and E.A. Susemihl, “A note on the vibrations of a clamped-free beam with a mass at the free end,” Journal of Sound and Vibration, vol37(2): p. 161-16, 1974.

[12] Fan-Gang, Z., et al., “Cochlear Implants: System Design, Integration, and Evaluation,” Biomedical Engineering, IEEE Reviews, Vol1: p. 115-142, 2008.

[13] Kim, A., T. Maleki, and B. Ziaie. A novel electromechanical interrogation scheme for implant able passive transponders. in Micro Electro Mechanical Systems (MEMS), 2012 IEEE 25th International Conference on. 2012.

[14] Saadon, S. and O. Sidek, “A review of vibration-based MEMS piezoelectric energy harvesters,” Energy Conversion and Management,Vol52(1): p .500-504 , 2011.

[15] Sood, R.K., Piezoelectric Micro Power Generator (PMPG): A MEMS-based Energy Scavenger: Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science. 2003

[16] D. Popovici, F. Constantinescu, and M.Maricaru. Modeling and Simulation of Piezoelectric Devices.I-Tech Education and Publishing .Vienna ,Austria,2008

Time (s)

Total Displacem

ent (μm)

Time (s)

Electric Potential (10-15 v)

Time (s)

Electric Energy Density

(J/m3)

Frequency (Hz)

Total Displacem

ent (10-5m

)

Resonance frequency number f1 f2 f3 f4 f5 f6

Frequency (KHz) 0.589 16.0 56.6 77.1 264.3 373.5

Total Displacement (μm)

11.8 0.0005 0.000036 0.13 0.00005 0.000042

386