«MICRO ELECTRET POWER GENERATORS Thesis by Justin Boland In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CALIFORNIA ...»
MICRO ELECTRET POWER GENERATORS
In Partial Fulfillment of the Requirements
for the Degree of
Doctor of Philosophy
CALIFORNIA INSTITUTE OF TECHNOLOGY
(Defended May 24, 2005)
All Rights Reserved
Yu-Chong Tai, Trevor Roper, Tanya Owen, Wen Hsieh, Ellis Meng, Tom Tsao, Mattieu
Liger, Qing He, Chi-Yuan (Victor) Shih, Scott Miserendino, Po-Jui (PJ) Chen, Nick Lo, Jayson Messenger, Svanhild (Swan) Simonson, Yuan-Heng (Denny) Chao, JR Heberle, the rest of the Caltech Micromachining Group, my candidacy committee (Yu-Chong Tai, Ken Pickar, David Rutledge, Axel Scherer, Kaushik Bhattacharya), my thesis committee (Yu-Chong Tai, Ken Pickar, David Rutledge, Melany Hunt, Changhuei Yang), funding sources (DARPA, AMRDEC, NSF), and my wife Stacey Boland.
I’ve been fortunate enough to be at the right place at the right time with enough preparation to make more accomplishments than I have time for writing this thesis. The experience has been invaluable to me, and I gained more than I thought possible. I am eternally grateful to Yu-Chong Tai, the Caltech Micromachining Laboratory and the California Institute of Technology for my time here.
For my wife, the first Dr. Boland. So few of our family and friends really understand why we like this stuff and what it is we actually do. No, I am not in school to be an electrician and my wife is not a mechanic!
v ABSTRACT Micro Electret Power Generators The taming of electricity and its widespread use allows people to see in the dark, to speak to one another instantaneously across the earth, and it allows retrieval of data from instruments sent out of the solar system. It is right to expect that the uses and demand for electricity will continue to grow, and to extend the ability to generate electricity; here two new micromachined devices for converting mechanical energy into electrical energy are presented. Aided by the wealth of micromachining process technology, generators that use an oscillatory motion to modify the physical structure of a capacitor with a built-in electric field provided by a permanent electret have been designed, built, and tested. The electret creates an electric field inside the capacitor structure, which induces mirror charge at some potential. The modification of the capacitor then generates an alternating displacement current through an external circuit, which provides useful electrical power.
The electret microphone is a similar well known device for converting pressure waves into electrical signals by varying the distance between two charged capacitive plates.
This work explores and proves feasible the ability to use mechanical forces to change the overlapping area of a charged capacitor structure and using mechanical forces to move a liquid into the gap of a charged capacitor structure, changing its permittivity to produce electricity. This work demonstrates 2.5mW of power from a 2cm diameter rotary
I.1. Scope of thesis
I.1.b. Problem Statement
I.2.a. Generating Electricity
I.2.a.ii. Photovoltaic Generation
I.2.a.iii. Chemical Generation
I.2.a.iv. Electromagnetic Generation
I.2.a.v. Electrostatic Generation
I.2.b. Alternative Energy
I.2.b.i. Portable Alternative Energy
I.3. Energy Harvesting
I.3.a. Energy Harvesting Methods
I.3.b. Survey of Kinetic Energy Harvesting Devices
I.3.b.i. Alternative Definition: Power Scavenging
I.3.b.ii. Figures of Merit
I.3.b.ii.1. Linear Energy Harvesters
I.3.b.ii.2. Rotational Energy Harvesters
I.3.b.iv. Linear Electromagnetic Power Generators
I.3.b.v. Piezoelectric Power Generators
I.3.b.vi. Charge Shuttle
I.3.b.vii. Electrostatic Power Generators
I.4. Displacement Current Power Generators
I.4.a. Origin of Displacement Current
I.4.a.i. Displacement Current in a Capacitor
I.4.b. Displacement Current For Power Generation
I.4.b.i. Variable Distance Electret Power Generators
I.4.b.ii. Variable Area Electret Power Generators
I.4.b.iii. Variable Permittivity Electret Power Generators
I.5. Physical Scaling
I.5.a. Physics-Based Definition of MEMS
I.5.b. Process-Based Definition of MEMS
I.5.c. Applying MEMS
II.1. Electret Classification
II.1.a. Heterocharge Electrets
II.1.b. Homocharge Electrets
II.2. Charging Methods
II.3. Measurement Techniques
II.3.a. Charge Density
II.3.a.i. Error in Depth of Charge
II.3.a.ii. Lateral Resolution of Charge
II.3.b. Depth Sounding Techniques
II.4.a. Floating Metal Layer Electret
II.4.a.i. Floating Metal Layer Process
III. Variable Area Rotational Electret Power Generator
III.1.a. Related Works
III.1.b. Micromachining Electrets
III.2. Theoretical Development
III.3. Design and Fabrication
III.3.a. Design Optimization
III.3.a.i. Charge Density
III.3.a.ii. Dielectric Constant
III.3.a.iii. Gap Spacing
III.3.a.iv. Number of Poles
III.3.b. Fabrication Considerations
III.3.b.i. Teflon Processing
III.3.d. REPG Version 2.0
III.3.e. REPG Version 3.0
III.3.f. REPG Version 4.0, 5.0 / Prototype Version 1.0, 2.0
III.4. Experimental Results
III.4.b.i. Rotational Speed
III.4.b.ii. Rotational Angular Misalignment
III.4.b.ii.1. Testbed Version 1
III.4.b.ii.2. Testbed Version 2
III.4.b.ii.3. Testbed Version 3
III.4.c. Power Generation Tests
III.4.c.i. REPG V1.0 on Testbed Version 1
III.4.c.ii. 32 Pole REPG V2.0
III.4.c.iii. 64 Pole REPG V2.0
IV. Liquid Rotor Electret Power Generator
IV.2.a. Using Liquid Metal Instead of Liquid Dielectric
IV.3. Design and Fabrication
IV.3.a. General Considerations
IV.3.a.ii. Cavity Material
IV.3.b.i. LEPG V1.0: Quick and Dirty
IV.3.b.ii. LEPG V2.0: PDMS Mold and Process Refinements
IV.3.c. LEPG V3.0: Multiple Channels on Single Chip
IV.4. Experimental Details
IV.4.a.i. Replacing Mercury with Steel Beads
IV.4.a.ii. Parallel Arrays
IV.4.a.iii. Serial Arrays
IV.4.a.iv. Non-Obvious Electrical Connections
V. Conclusions and Future Work
V.1. Rotary Electret Power Generator
V.2. Liquid Rotor Electret Power Generator
Figure I-1 Measured vibrations from a microwave oven
Figure I-2 Electrical power delivered over time from various sources 
Figure I-3 Generator architecture comparison
Figure I-4 Diagram of Seiko’s Kinetic line of energy harvesting watches
Figure I-5 Exploded Diagram of Seiko’s Kinetic line of energy harvesting watches.
Figure I-6 Lab test of Seiko’s Kinetic generator. Speed corresponds to the relative rotation of the magnet to the coil.
Figure I-7 Linear electromagnetic power generator developed by Perpetuum .
Figure I-8 Perpetuum’s 2-terminal power generator package 
Figure I-9 PMG0100 Evaluation Model 
Figure I-10 Piezoelectric cantilever with proof mass for converting vibrations to electricity
Figure I-11 Schematic for piezoelectric windmill power generator
Figure I-12 Cross section of a charge shuttle
Figure I-13 Drive circuitry for the charge shuttle
Figure I-14 In-plane variable gap capacitance micromachined power generator...........32 Figure I-15 Out-of-plane variable gap capacitance micromachined power generator
Figure I-16 Pulsed Chemical-Electret Generator system concept
Figure I-17 Pulsed combustion of a prototype symmetric PCTR in natural aspirating mode at three different phase angles
Figure I-18 Semi-ceramic combustion chambers
Figure I-19 Two interconnected fluidic microcavities
Figure I-20 Coupon
Figure I-21 CFD simulation of a turbine
Figure I-22 Proposed new design for pulsed combustor thermal resonator and shaker generator system.
Figure II-1 Contour plot of magnetic field from a bar electret
xii Figure II-2 Contour plot of electric potential from a sheet electret
Figure II-3 Streamline plot of electric field from a sheet electret
Figure II-4 Heterocharge by polarization
Figure II-5 Homocharge electret with implanted electrons
Figure II-6.Triboelectrically charged Teflon chip
Figure II-7 Charge density of implanted Teflon using the back lighted thyratron.........61 Figure II-8 Mean charge depth for corona charged FEP Teflon
Figure II-9 Isoprobe mounted on X-Y micropositioner
Figure II-10 Charge density measurement used to determine minimum distance between data points.
Figure II-11 Average charge density and standard deviation when dropping data points from dataset used to produce Figure II-10.
By statistically comparing neighboring data points from the scan of Figure II-10, it was determined that a measurement spacing of 1mm would allow both average measurement of charge and high contrast as shown in Figure II-12
Figure II-13 Polarization map of a 11 µm thick PVDF film poled with a Tshaped electrode. At z=1μm (top graph), the polarization is significantly lower than in the bulk. The arrow indicates the direction of the high-resolution scan [ibid].
Figure II-14 Charge implanted in a chip with floating metal layer.
Figure III-1 First electret power generator. Tada (1992)
Figure III-2. Schematic of electret generator (cross-section view).
Figure III-3. Perspective view of electret generator showing a 4-pole rotor and stator.
Figure III-4 Equivalent circuit for variable area electret power generator.
Figure III-5 Used to find the critical width w from gap distance
Figure III-6 Table of different Fluorinert solvents, which are used to dilute Teflon AF 1601-S
Figure III-7 Process flow for first version of REPG
Figure III-8 REPG V1.0 mounted on testbed version 1. Photo taken before rotor and stator are aligned
Figure III-9 REPG V1.0 mounted on testbed version 1. Photo taken after rotor and stator are aligned.
xiii Figure III-10 Process flow for bulk-etched electrets
Figure III-11 Bulk-etched Teflon with anchors before reflow step.
Figure III-12 Bulk-etched Teflon with anchors after reflow step.
Figure III-13 REPG version 4.0. Cutaway view of final assembled device including bearings.
Figure III-14 Stator for REPG version 5.0. This design incorporated the use of bulk-etched cavities
Figure III-15 Charge density measurements of a 4-pole floating metal layer electret that is triboelectrically charged
Figure III-16 Proceedure for measuring angular misalignment
Figure III-17 Law of reflections on laser trajectory used to find angular misalignment
Figure III-18 Testbed with rotor and stator mounted
Figure III-19 Side view of ball joint
Figure III-20 Inside View of ball joint
Figure III-21 Side view of testbed with rotor and stator mounted
Figure III-22 Newest testbed for REPG
Figure III-23 Power output from 3 experimental trials using different load resistances and theoretical power of a continuously load matched system
Figure III-24 Power measured and theoretical vs. rotation for the 32-pole power generator with a 600kΩ load, -5x10-4C/m2 charge implanted, 2cm diameter rotor-stator pair, and 4.25μm thick Teflon electret
Figure III-25 Load matching test of a 64 pole generator on testbed version 3 at
Figure III-26 Power vs. rotation for the 64-pole power generator with a 50.3kΩ load
Figure III-27 Comparison of power measured from an actual Seiko watch to rotational electret power generators.
Figure IV-1 Femlab modeling of spatial potential from an electret that is modified by a sphere of mercury
Figure IV-2. LEPG conceptual image
Figure IV-3. Equivalent circuit for each half of the channel.
Figure IV-4 Normalized function to describe oscillations of liquid in a channel.........122 xiv Figure IV-5 Mathematically defined capacitances over one cycle. No allowance has been made for stray capacitance.
Figure IV-6 LEPG with a small droplet of water in the channel.
Figure IV-7 Process Flow a. deposit metal on glass substrate b. pattern metal c.
spin-on Teflon AF d. mask design used.
Figure IV-8 Mold Master for Sylgard 184 and peeled PDMS
Figure IV-9 Assembled LPG Device. Clear epoxy binds the top plate to the bottom plate and prevents the mercury from leaking
Figure IV-10 Electrode pattern for 6x3 cavities with 2 top and 2 bottom electrodes per cavity.
Figure IV-11 Assembled LEPG device with cutaway to reveal bottom electrodes
Figure IV-12 Test setup for LEPG mounted on shaker
Figure IV-13 Power generated in LPG V2.1 with 100μm Teflon PTFE
Figure IV-14 Still-frame position 1 taken at 2000fps while shaking at 60Hz and 1 mm peak to peak.
Figure IV-15 Still-frames position 2 taken at 2000fps while shaking at 60Hz and 1 mm peak to peak.
Figure IV-16 Experimental values for parallel channels shaking of 2.58 mm peak-to-peak at 60Hz.
Figure IV-17 Experimental values for serial columns shaking at 1 mm peak-topeak at 60Hz
Figure IV-18 Experimental values for shaking at 2.58 mm peak-to-peak at 60 Hz and Rl of 4 MOhm
Figure IV-19 Diagram showing all connections across LEPG