Unformatted text preview: UNIVERSITY OF CINCINNATI Date: July 25, 2005 I, __________Weizhuo Li______________________________________,
hereby submit this work as part of the requirements for the degree of:
Doctorate of Phyilosphy in:
Electrical & Computer Engineering and Computer Science It is entitled:
Wavelength Multiplexing of MEMS Pressure and Temperature Sensors Using
Fiber Bragg Gratings and Arrayed Waveguide Gratings This work and its defense approved by: Chair: __________Joseph T. Boyd__________
_______Altan M. Ferendeci__________
________Howard E. Jackson_________
__________Chong H. Ahn___________ 1 Wavelength Multiplexing of MEMS Pressure and
Temperature Sensors Using Fiber Bragg Gratings and
Arrayed Waveguide Gratings A dissertation submitted to the
Division of Research and Advanced Studies
of the University of Cincinnati
in partial fulfillment of the
requirement for the degree of
DOCTORATE OF PHILOSOPHY (Ph.D.)
in the Department of Electrical and Computer Engineering and Computer Science
of the College of Engineering
B.S., Electrical Engineering, Northern Jiaotong University, Beijing, China, 1994 Committee Chair: Professor Joseph. T. Boyd Abstract
This thesis presents the design, fabrication and testing of wavelength multiplexing
of optically interrogated MEMS pressure and temperature sensors using both fiber Bragg
gratings (FBGs) and arrayed waveguide gratings (AWGs). The pressure sensors and
temperature sensors have been designed and fabricated using MEMS techniques. For the
pressure sensor, fabrication is initiated with a standard fused-silica wafer (Pyrex 7740)
polished on both sides that is patterned to form a series of cavities for the Fabry-Perot
interferometer. Positive photoresist is patterned and serves as a mask for etching the
cavities in the glass wafer with buffered HF. A silicon wafer polished on both sides is
then electrostatically bonded to the patterned glass wafer. Bulk etching techniques are
used to thin the silicon wafer down to the desired diaphragm thickness, while the other
side of the Si/glass assembly is protected. The configuration of a Fabry-Perot temperature
sensor involves a thin layer of silicon bonded to a glass wafer. The fabrication process is
similar to that of a pressure sensor, but with no air cavity.
Pressure measurements were made over the 0 to 30 psi range while temperature
measurements were made over the 24 to 100 °C range. Pressure sensor sensitivities of
0.022mv/psi for the multiplexing system using AWGs, and of 0.0072mv/psi for the
multiplexing system using FBGs were obtained. The pressure sensor were designed with
cavity diameter R0 = 300 μm, cavity depth d0 = 0.64 μm for the sensor operating at 850
nm, and d0 = 1.1 μm for the sensor operating at 1550 nm. Diaphragm thickness for the
two sensors were 14 μm and 15.5 μm. The temperature sensor was fabricated by bonding
a silicon wafer with different thickness on the glass wafer. An anti-reflective coating of
SiO was deposited on the top surface of the pressure sensor by evaporation. The purpose 2 of this anti-reflective coating was to reduce sensor’s temperature sensitivity. An oxidantresistant encapsulation scheme for the temperature sensor was proposed, fabricated and
tested, namely aluminum coated silicon nitride (Al/Si3N4).
The multiplexed sensor system using FBGs has the potential of multiplexing about
seventy sensors depending on FBGs bandwidth and the one using AWGs has the
potential of multiplexing about eight sensors depending on the number of AWGs
channels and the bandwidth of each channel. A dual- wavelength method incorporating a
tunable laser was used to interrogate either the applied pressure or temperature
experienced by the sensor, while a three-wavelength method was used to simultaneously
interrogate pressure and temperature. Experimental results, including response as a
function of pressure or temperature, were characterized by good agreement between
experimental and theoretical results. There is no observable cross-talk between the
multiplexed sensors. 3 Acknowledgements First and foremost, I would like to thank my advisor, Dr. Joseph T. Boyd. His
guidance and instruction throughout this project greatly added to my research and study
experience, and his overall believe in my ability to finish this project was also invaluable
Second, I would like to thank the committee members, Dr. Chong H. Ahn, Dr.
Altan M. Ferendeci, Dr. Peter Kosel and Dr. Howard Jackson for their review of this
Thirdly, I would also like to acknowledge the help and friendship of the
optoelectronics lab mates who have helped and encouraged me throughout my endeavor
at my doctoral program. They are Anish Saran, Jie Zhou, Pete L. Vassy and Nan Zhang.
A special acknowledgement is extended to Dr. Don Abeysinghe, a research scientist at
Taitech Inc for his involvement and guidance in my thesis project.
Finally, I would like to thank my family for their support, especially to my husband,
Jianhui Zhao. He comes along with me to the US, sacrifices his career in China for
supporting me to pursue and complete this Ph D degree. Without his support and love, I
would not succeed. So, I would like to dedicate this thesis to my beloved husband. 4 Table of Content Table of Contents
List of Figures
List of Tables 1
………………………………………………………………… …………………………………………………………………….. 1.0 INTRODUCTION …………………………………………….. 1.1 Review of Fabry-Perot Cavity-Based Pressure and Temperature Sensor
1.2 Review of Multiplexed Sensors
1.3 Introduction to Optical Fiber Telecommunications and WDM System 1.4 Summary 1.5 References
2.0 DESIGN OF SENSORS
2.1 Introduction 2.2 Plane Wave Propagation in Homogeneous Media
2.2.1 Reflection and Transmission at an Interface Between Two Materials
2.2.2 Reflection from Isotropic Layered Media 2.3 Design of Ideal Fabry-Perot Cavity-Based Pressure sensor
2.3.1 The Dynamic Properties of the Fabry-Perot Pressure Sensor
2.3.2 Calculation of the Light Loss Inside the Sensor
2.3.3 Initial Diaphragm Deflection
2.3.4 Anti-reflection Layer
2.4 Design of Ideal Fabry-Perot Temperature Sensor
2.4.1 Selection of Temperature Sensitive Material
2.4.2 Determination of Fabry-Perot Temperature Sensor’s Thickness
2.4.3 Design of Encapsulant for Temperature Sensor 2.5 Summary 2.6 References
5 3.0 MULTIPLEXED SENSORS
3.1 Introduction 3.2 Multiplexed Pressure Sensors Using Fiber Bragg Gratings (FBGs) 3.3 Multiplexed Pressure Sensors Using Arrayed Waveguide Gratings (AWGs) 3.4 Ripples Due to the Long Coherence Length of Tunable Laser
3.5 Dual-Wavelength Interrogation Technique for Multiplexed Pressure Sensors
3.6 Multiplexed Temperature Sensors and Interrogation Technique 3.7 Summary 3.8 References 4.0 FABRICATION AND CALIBRATION OF SENSORS
4.1 Introduction 4.2 Sensor Fabrication
4.2.1 Photolithographic Patterning and Wet Etching for Cavity Formation on
4.2.2 Anodic Bonding of Silicon-to-Glass
4.2.3 Silicon Wafer Thinning to Form Diaphragm
4.2.4 Polishing Silicon Diaphragm
4.2.5 Sensor Package 4.3 Sensor Characterization
4.3.1 Measuring Initial Diaphragm Deflection and Cavity Depth
4.3.2 Measuring Diaphragm Thickness 4.4 Pressure Sensor Calibration
4.4.1 Calibration of Multiplexed Pressure Sensor System
4.4.2 Comparison of the Two Multiplexed Pressure Sensor Systems
4.4.3 Temperature Sensor Calibration
4.4.4 Calibration of Multiplexed Temperature Sensor System 4.5 Summary 4.6 References 6 5.0 ANTI-RELECTION LAYER FOR PRESSURE SENSOR AND ENCAPSULATING
STRUCTURE FOR TEMPERATURE SENSOR
5.2 SiO Evaporation Process
5.2.1 Calibration of Pressure Sensor with Anti-reflective Coating
5.3 Si3N4 and Aluminum Deposition Process
5.3.1 Calibration of Temperature Sensor with Encapsulation Structure 5.4 Summary 5.5 References 6.0 MULTIPLEXED SENSOR SYSTEM FOR SIMULTANEOUS MEASUREMENT
OF PRESSURE AND TEMPERATURE
6.1 Introduction 6.2 Prerequisite for Simultaneous Measurement of Pressure and Temperature 6.3 Multiplexed Sensor System for Simultaneous Measurement of Pressure and 6.4 Summary 6.5 References 7.0 CONCLUSIONS 7 List of Figures
1.1 Schematic diagram illustrating an optical-interrogated Fabry-Perot pressure sensor
interconnected via an optical fiber.
1.2 Cross-sectional view of the sensing element.
1.3 Illustration of two configurations of fiber optically interrogated MEMS pressure
sensors. Figure 1(a) shows the usual configuration, which consists of a glass plate
with a shallow cylindrical cavity etched into one surface with the cavity covered by a
thin silicon diaphragm that has been anodically bonded to the patterned glass wafer.
Figure 1(b) shows the configuration where the cavity is formed on the end of the
optical fiber and a silicon diaphragm is bonded anodically.
1.4 Measurement system for a Fabry-Perot cavity-based pressure sensor.
1.5 Frequency multiplexing scheme.
1.6 Wavelength Division Multiplexing Bragg grating-based laser sensors.
1.7 Time-Division-Multiplexed fiber Bragg grating sensor array.
2.1 Configuration of optically interrogated MEMS pressure sensors. (a) gauge pressure
sensor; (b) absolute pressure sensor; (c) differential pressure sensor
2.2 Reflection and refraction of plane wave at a boundary between two dielectric media.
2.3 Reflection and refraction of S wave (TE).
2.4 Reflection and refraction of P wave (TM).
2.5 Transmission and reflection from one layer Fabry-Perot interferometer.
2.6 Transmission and reflection from two layer Fabry-Perot interferometer.
2.7 A multiplayer dielectric medium.
2.8 Reflectance from the Fabry-Perot cavity-based sensor as a function of cavity depth
when operating at 850nm (n0 = nglass =1.474, n1 = nair =1.0, n2 = nsi =3.46).
2.9 Reflectance from the Fabry-Perot cavity based sensor as a function of cavity depth
when operating at 1550 nm with the diaphragm thickness at t = 15.53 um
(n0 = nglass=1.473, n1 =nair =1.0, n2 = nsi =3.478, n3 = nair =1.0).
2.10 Reflectance from the Fabry-Perot cavity based sensor as a function of silicon
thickness when operating at 1550 nm with the cavity depth at d0 = 1.1 μm. 8 2.11 Reflectance from the Fabry-Perot cavity based sensor as a function of cavity depth
(0-2 μm) and silicon thickness (15.2-15.8 μm) when operating at 1550nm.
2.12 Computer simulation of diaphragm deflection of the Fabry-Perot cavity-based
sensor under the pressure using Mathematic 4.0.
2.13 Calculation of light loss in sensor.
2.14 Calculation of change of pressure inside the cavity under diaphragm bending.
2.15 Fabry-Perot cavity-based pressure sensor with a thin layer of antireflection coating
(SiO) on the top surface of silicon diaphragm.
2.16 Illustration of how an antireflection coating reduces the reflected light intensity.
2.17 Illustration of how the condition for antireflection coating is wavelength dependent
(The thickness of the antireflection coating is designed for operating at 1550 nm).
The antireflection condition is satisfied at the wavelengths of the cross points of the
two lines. Red line: Reflectivity from the three layer Fabry-Perot interferometer
(pressure sensor without antireflection layer); Blue line: Reflectivity from the one
layer Fabry-Perot interferometer (pressure sensor with antireflection layer).
2.18 Configuration of the Fabry-Perot temperature sensor.
2.19 Reflectance spectra shift with temperature increasing.
2.20 Temperature range as a function of silicon thickness at λ = 1550 nm for a FabryPerot temperature sensor.
2.21 Configuration of the Fabry-Perot temperature sensor with encapsulating layers.
2.22 Comparison of reflectance spectra shift for the sensor with encapsulant and the
sensor without encapsulant under several values of temperature.
3.1 Principle of fiber Bragg gratings (FBGs).
3.2 Schematic diagram of the multiplexed pressure sensor system using FBGs.
3.3 Simulation of the results from multiplexed pressure sensor sytem using FBGs.
3.4 Schematic diagram of the arrayed waveguide gratings (AWGs).
3.5 Schematic diagram of the multiplexed pressure sensor system using AWGs.
3.6 A cosinusoidal wave train modulated by a Gaussian envelope along with its
transform, which is also Gaussian.
3.7 Unwanted interferometric signal between the connector and FBG.
3.8 Structure of a basic ferrule connector (FC/PC and FC/APC). 9 3.9 Ratio I ( R, λ ) = R( λ1 )
from the Fabry-Perot cavity-based pressure
R( λ1 ) + R ( λ2 ) sensor as a function of cavity depth when operating at λ1 = 1548.3nm and λ 2 = 1554.1nm with the silicon diaphragm thickness at t = 15.53 μm
(n0 = nglass=1.473, n1 =nair =1.0, n2 = nsi =3.478, n3 = nair =1.0).
3.10 Reflectivity from the Fabry-Perot temperature sensor as a function of silicon
refractive index with silicon thickness at t = 20 μm
(n0 = nglass=1.473, n1 =nsi, n2 = nair =1.0).
3.11 Ratio I ( R, λ ) = R( λ1 )
from the Fabry-Perot temperature sensor as a
R( λ1 ) + R ( λ2 ) function of silicon refractive index when operating at λ1 = 1548.3nm and λ 2 = 1554.1nm and with silicon thickness at t = 20 μm.
3.12 Multiplexed temperature sensors using broad band mirror fiber gratings.
3.13 Reflection spectrum of a broad band mirror fiber grating.
3.14 Simulation of the results from multiplexed temperature sensors using broad band
mirror fiber gratings (Bandwidth ≈10 nm).
3.15 Simulation of temperature range (a) and temperature sensitivity; (b) as a function of
silicon thickness for multiplexed temperature sensors using broadband fiber gratings
(Bandwidth ≈10 nm).
4.1 MEMS fabrication processing steps.
4.2 Mask for Fabry-Perot cavity-based pressure sensors.
4.3 Cavity depth measured by Profilometer
The sensor above has the cavity depth around 1.1941 μm
The sensor below has the cavity depth around 0.9512 μm.
4.4 Glass-to-silicon anodic bonding machine in our lab.
4.5 Glass-to-silicon anodic bonding setup.
4.6 Typical current flow during anodic bonding.
4.7 A Teflon beaker for silicon diaphragm etching.
4.8 Wafer protector for etching bonded Si substrate. 10 4.9 A Fabry-Perot cavity observed from the silicon side after thinning down the silicon
4.10 The profile of a Fabry-Perot cavity after cleaving through the glass-diaphragm
4.11 Surface of the silicon diaphragm after etching: (a) before polishing; (b) after
4.12 Package configurations for housing the optical fiber and MEMS pressure sensor: (a)
differential pressure sensor; (b) absolute and gauge pressure sensor.
4.13 (a) Nikon optical microscope in our lab; (b) Place a sensor under microscope for
observation of Newton’s rings.
4.14 Illustrating the formation of Newton’s rings.
4.15 Change of Newton’s fringes with pressure increasing for a gauge pressure sensor.
4.16 Change of Newton’s fringes with pressure increasing for an absolute pressure sensor.
4.17 Change of Newton’s fringes with pressure increasing for a differential pressure
4.18 Microscope focusing can be used to determine the silicon diaphragm thickness by
cleaving through the glass-diaphragm structure and measuring its thickness on edge.
4.19 Silicon diaphragm thickness can be determined by the measured reflectance
spectrum from the sensor.
4.20 The relationship between interferometer cavity depth and cavity reflectance is
shown for LED emission at 820, 850 and 880 nm. Note the reflectance is
wavelength dependent at each cavity depth.
4.21 The static pressure response results from the pressure sensor operating at 850 nm.
4.22 Computer simulation of pressure fluctuation in wind tunnel.
4.23 Measured pressure fluctuation in wind tunnel.
4.24 Schematic diagram of single pressure sensor measurement.
4.25 Measured reflectance spectra at several values of pressures for single pressure
sensor measurement (cavity depth = 0.95 μm; diaphragm thickness = 22.95 μm).
4.26 Measured reflectance spectra at several values of pressures for single pressure
sensor measurement when using FC/PC connectors in the system instead of FC/APC
connectors. 11 4.27 Air pressure control system in our lab.
4.28 Experimental set-up for multiplexed pressure sensor system.
4.29 Results of multiplexed pressure sensor system using fiber Bragg gratings (FBGs).
4.30 Results of multiplexed pressure sensor system using arrayed waveguide gratings
4.31 Measured reflectance as a function of pressure (0 ∼ 30 psi) extracted from the
multiplexed signal for sensor #2 at two wavelengths: λ1 = 1552.1 nm and λ2 =
4.32 The ratio I ( R, λ ) = R ( λ2 )
(λ1 = 1552.1 nm and λ2 = 1557.1 nm) for sensor #2
R( λ1 ) + R ( λ2 ) as a function of pressure.
Dotted lines: theoretical curve; Solid lines: measured response.
4.33 Schematic diagram of single Fabry-Perot temperature sensor measurement.
4.34 Experimental set-up for Fabry-Perot temperature sensor measurement.
4.35 Configuration of a Fabry-Perot temperature sensor #1.
4.36 Measured reflectance spectrum shift under several values of temperature for sensor
4.37 Configuration of a Fabry-Perot temperature sensor #2.
4.38 Measured reflectance spectrum shift under several values of temperature for sensor
4.39 Results of multiplexed temperature sensor system using fiber Bragg gratings (FBGs).
4.40 Measured reflectance as a function of temperature (24 °C ∼ 35 °C) extracted from
multiplexed signal for sensor #1 at two wavelengths: λ1 = 1548.3 nm and λ2 =
4.41 Ratio of the reflectance I ( R, λ ) = R (λ2 )
measured at two wavelengths (λ1 =
R ( λ1 ) + R ( λ2 ) 1548.3 nm and λ2 = 1554.1 nm) for sensor #1.
4.42 Mathematics simulation of reflectance spectrum shift for sensor #1 using the
multiplexed temperature sensor system. (The reflected light from a broadband fiber
grating with 10 nm bandwidth was used to interrogate sensor #1 using wavelengthencoded measurement). 12 5.1 Vacuum evaporation oven.
5.2 A Fabry-Perot cavity observed from the silicon side after thinning down the silicon
diaphragm (a) without anti-reflective coating; (b) with anti-reflective coating.
5.3 (a) Theoretical reflectance spectrum for a Fabry-Perot cavity-based pressure sensor
with anti-reflective coating; (b) Measured reflectance spectrum from a Fabry-Perot
cavity-based pressure sensor with anti-reflective coating.
5.4 Measured reflectance spectra at several values of pressure for a Fabry-Perot cavitybased pressure sensor with anti-reflective coating.
5.5 Measured reflectance as a function of pressure at two wavelengths: λ1 = 1548 nm
and λ2 = 1550 nm for a Fabry-Perot cavity-based pressure sensor with anti-reflective
5.6 Measured reflectance spectra at several values of temperature from 50 ∼ 120 °C.
5.7 Sputtering machine.
5.8 Electron-beam evaporation machine.
5.9 Thickness and refractive index measurements of silicon nitride on silicon using
5.10 (a) Theoretical reflectance spectra for a Fabry-Perot temperature sensor without
encapsulating layers (silicon thickness = 132 μm); (b) Measured reflectance
spectrum from a Fabry-Perot temperature sensor without encapsulating layers
(silicon thickness = 132 μm).
5.11 (a) Theoretical reflectance spectra for a Fabry-Perot temperature sensor with 200 nm
Si3N4; (b) Measured reflectance spectrum from a Fabry-Perot temperature sensor
with 200 nm Si3N4.
5.12 (a) Theoretical reflectance spectra for a Fabry-Perot temperature sensor with 200 nm
Si3N4 + 1 μm Aluminum; (b) Measured reflectance spectrum from a Fabry-Perot
temperature sensor with 200 nm Si3N4 + 1 μm Aluminum.
5.13 Measured reflectance as a function of temperature (25 ∼ 35.5 °C) for a temperature
sensor with encapsulating layers at two wavelengths: λ1 = 1552 nm and λ2 = 1557
nm. 13 5.14 The ratio I ( R, λ ) = R ( λ2 )
(λ1 = 1552 nm and λ2 = 1557 nm) for a temperature
R( λ1 ) + R ( λ2 ) sensor with encapsulating layers as a function of temperature. Dotted lines:
theoretical curve; Solid lines: measure response.
6.1 Newton’s rings change under temperature for a Fabry-Perot cavity-based pressure
sensor without pressure applied: (a) gauge pressure sensor; (b) absolute pressure
sensor; (c) differential pressure sensor.
6.2 The ab...
View Full Document