planar total internal reflection biofouling sensors · planar total internal reflection biofouling...
TRANSCRIPT
Planar Total Internal Reflection Biofouling Sensors
By
Koo Hyun Nam
A dissertation submitted in partial satisfaction of the
requirements for the degree of
Doctor of Philosophy
in
Engineering - Mechanical Engineering
in the
Graduate Division
of the
University of California at Berkeley
Committee in charge:
Professor Liwei Lin, Chair
Professor Costas P. Grigoropoulos
Professor Dorian Liepmann
Professor Michael Maharbiz
Fall 2010
Planar Total Internal Reflection Biofouling Sensors
Copyright 2010
by
Koo Hyun Nam
1
Abstract
Planar Total Internal Reflection Biofouling Sensors
by
Koo Hyun Nam
Doctor of Philosophy in Mechanical Engineering
University of California at Berkeley
Professor Liwei Lin, Chair
Planar, integrated microscale sensors utilizing prism-coupler type angular interrogation sensing
technique have been demonstrated. The main structure of the sensor consists of an optical
prism coupled to a built-in waveguide to introduce Fraunhofer diffraction when light ray comes
into the prism from the waveguide. The Fraunhofer diffraction creates spectrum of consecutive
rays over the sensing edge of the prism such that there is no need for the bulky scanning
mechanisms typically used in other macro scale sensing systems. Two types of sensors are
presented: (1) total internal reflection based critical point detection (CPD) sensor, and (2) surface
plasmon resonance (SPR) based resonance point detection (RPD) sensor.
The CPD sensor is fabricated by a simple, two-mask process which creates a right angle prism
with three sides with lengths of 1, 0.86, and 1.33 mm, respectively in the prototype design and a
waveguide with a cross sectional area of 4×0.25 μm2. The 0.25 μm-thick core and the 2.5μm-
thick cladding layers of the waveguide are made of silicon nitride and silicon dioxide,
respectively. The CPD sensing technique measures the shift of the critical point of the total
reflection as the results of change of refractive index due to biofouling. Optical simulations are
used to validate the working principle and the calculated biofouling sensitivity is comparable to
the other optical sensing methods. A baseline measurement has been conducted to verify the
operation of the sensor with an error of less than ± 0.002 R.I.U. During a 9-hour biofouling
measurement using milk as the media, a change in the refractive index as much as 0.0089 is
recorded as the result of biofouling.
The RPD sensing technique employs surface plasmon resonance as its sensing mechanism by
measuring the shift of the resonance point with respect to the change of the incident angle. The
design and fabrication process is similar to the fundamental structure of CPD sensors with an
additional deposition of a thin metal layer on the sensing edge of the prism. The theoretical
sensitivity is calculated as 90 deg RIU-1
, which is comparable with the state-of-the-art optical
sensors at 127 deg RIU-1
. The refractive index measurement for selected liquids agrees with the
values in the literature with an error range of less than ± 0.002 R.I.U. Furthermore, the
refractive index change of biofouling formation is measured to be 0.0078 for a 9-hour
experiment using milk as the testing media.
Professor Liwei Lin, Chair Date
ii
Acknowledgments
I would like to thank my advisor, Professor Liwei Lin, for giving me the opportunity to
be at the forefront of this field of research and for the support, interest, and encouragement that
he has so generously provided from the days of my undergraduate studies at U. C. Berkeley up to
the present time. I am extremely grateful for what I have learnt from you. There are many other
people I would also like to thank for their help, encouragement, and support, including all of my
friends and associates, and I very much regret not adequately expressing my very deep
appreciation for all of you.
Most of all, I cannot find words capable of describing my appreciation for my parents.
You are the reason for which I live, for why I am the person I am, and for my success. So most of
all, I would like to dedicate this doctoral study to you. To my mother, Mrs. Yeon Soon Seo: you
are the strongest and sweetest person in the world. And to my father, Mr. Ki Hong Nam: you are
my hero, and there is nothing that will ever change that. Mom and dad, you are my tears, smiles,
and everything that I can have that is of any value or is worth living for.
iii
Contents
List of Figures and Tables ........................................................................................................... vi
Chapter 1 Introduction................................................................................................................. 1
1.1 Prism-coupler Type Angular Interrogation System........................................................... 1
1.1.1 Critical Point Detection Sensing ................................................................................. 3
1.1.2 Resonance Point Detection Sensing ............................................................................ 4
1.2 Chapter Overviews ............................................................................................................ 6
References ................................................................................................................................... 7
Chapter 2 Macroscale Prism-coupler Type Sensing using Fraunhofer Diffraction ............... 9
2.1 Introduction ....................................................................................................................... 9
2.2 Theoretical Background .................................................................................................. 10
2.2.1 Total Internal Reflection ........................................................................................... 10
2.2.2 Fraunhofer Diffraction .............................................................................................. 11
2.2.3 The Convolution of Reflection and Diffraction Patterns .......................................... 12
2.2.4 Detection of Biofouling ............................................................................................ 13
2.3 Numerical Simulation ...................................................................................................... 14
2.4 Working Principle............................................................................................................ 16
2.5 Experiment Protocols and Considerations ....................................................................... 17
2.5.1 Direct Refractive Index Measurement ...................................................................... 17
2.5.2 Media of Different Optical Properties ...................................................................... 17
2.5.3 Sources of Error ........................................................................................................ 17
2.5.4 Image Processing ...................................................................................................... 18
2.6 Sensing Experiment ......................................................................................................... 18
2.6.1 Experimental Setup ................................................................................................... 18
2.6.2 Experiment Results ................................................................................................... 19
2.7 Summary .......................................................................................................................... 22
References ................................................................................................................................. 23
iv
Chapter 3 Critical-Point-Detection (CPD) Sensing ................................................................. 25
3.1 Introduction ..................................................................................................................... 25
3.2 Microscale Sensor Design and Fabrication ..................................................................... 26
3.2.1 Sensor Design ........................................................................................................... 26
3.2.2 Fabrication Procedure ............................................................................................... 27
3.2.3 Design Variations...................................................................................................... 28
3.2.4 Fabrication Results.................................................................................................... 31
3.3 Experiments and Results ................................................................................................. 32
3.3.1 Experimental Setup ................................................................................................... 32
3.3.2 Calibration................................................................................................................. 34
3.3.3 Direct Refractive Index Measurement ...................................................................... 35
3.3.4 Biofouling Sensing.................................................................................................... 37
3.4 Discussions ...................................................................................................................... 38
3.4.1 General ...................................................................................................................... 38
3.4.2 Tribology................................................................................................................... 39
3.5 Summary .......................................................................................................................... 40
References ................................................................................................................................. 42
Chapter 4 Resonance-Point-Detection (RPD) Sensing ............................................................ 44
4.1 Introduction ..................................................................................................................... 44
4.2 Theoretical Background .................................................................................................. 45
4.3 Numerical Simulation ...................................................................................................... 46
4.4 Working Principle and Design ........................................................................................ 48
4.5 Microscale Sensor Fabrication ........................................................................................ 49
4.6 Experiments and Results ................................................................................................. 50
4.6.1 Experimental Setup ................................................................................................... 50
4.6.2 Image Processing ...................................................................................................... 51
4.6.3 Effect of Polarization ................................................................................................ 51
4.6.4 Experimental Results ................................................................................................ 52
4.7 Discussion ........................................................................................................................ 54
4.7.1 Design Variations...................................................................................................... 54
4.7.2 Determination of Sensitivity ..................................................................................... 55
4.8 Summary .......................................................................................................................... 56
References ................................................................................................................................. 58
v
Chapter 5 Conclusions and Future Work ................................................................................ 60
5.1 Conclusions ..................................................................................................................... 60
5.2 Future Directions ............................................................................................................. 61
5.2.1 Self Cleaning Sensor using UV and F-IR Ray.......................................................... 61
5.2.2 Integration of Laser Diode and Photo Detector ........................................................ 62
5.2.3 Measurement of phase difference between s- and p-polarizations ........................... 63
5.2.4 Measurement in Total Internal Reflection Region .................................................... 64
5.2.5 Digitated Sensor Geometry ....................................................................................... 64
5.2.6 Thickness measurement of SPR-based sensor .......................................................... 65
5.2.7 Sensor using Fresnel diffraction ............................................................................... 65
References ................................................................................................................................. 67
vi
List of Figures and Tables
Figure 1.1: Sensitivity plotted against size of current sensors and goal of this study. ................... 2
Figure 1.2: Schematic of proposed prism-coupler type microscale sensor. ................................... 2
Figure 1.3: Critical point shift on the reflectance profile under TIR conditions with regard to
changes in refractive index. ......................................................................................... 3
Figure 1.4: Typical reflectance and diffraction patterns for light of varying incident angles and
the convolution pattern of their combination as a resulting sensor signal. .................. 4
Figure 1.5: Resonance point shift on the SPR intensity profile according to changes of refractive
index. ............................................................................................................................ 5
Figure 1.6: Typical SPR intensity profile, diffraction patterns for varying incident angle of light
and the convolution patterns as resulting sensor signals. ............................................. 5
Figure 2.1: Reflectance profile and total internal reflection. ........................................................ 10
Figure 2.2: Normalized intensity of light and Fraunhofer diffraction. ......................................... 11
Figure 2.3: Schematic diagram of convoluted light patterns induced by reflection and diffraction.
.................................................................................................................................... 13
Figure 2.4: Output signals of different refractive indices. ............................................................ 15
Figure 2.5: Refractive index change versus angle of convolution peak. ...................................... 15
Figure 2.6: Schematic illustration of prism-coupler type sensing mechanism utilizing Fraunhofer
diffraction. .................................................................................................................. 16
Figure 2.7: Image analysis process illustrating how to find the critical point. ............................. 18
Figure 2.8: Complete optical setup for macroscale experiment. .................................................. 19
Figure 2.9: Typical output signal captured by webcam and corresponding theoretical patterns of
diffracted and reflected light separated. ..................................................................... 20
Figure 2.10: Output signals captured by CCD and corresponding intensity profiles for a milk
biofouling formation. ................................................................................................. 21
Figure 2.11: Position of critical point and corresponding refractive indices with respect to time.
.................................................................................................................................... 22
Figure 3.1: Schematic illustration of CPD sensing mechanism. .................................................. 26
Figure 3.2: Schematic diagram of light paths in a prototype CPD sensor. ................................... 27
Figure 3.3: Microfabrication procedures for proposed sensors. ................................................... 28
Figure 3.4: Optical simulation results showing coupling of a single mode of light and the
propagation loss during its passage through the waveguide. ..................................... 30
Figure 3.5: SEM image of sensing edge of sensor (x3,800 and x13,000). ................................... 31
Figure 3.6: SEM pictures of exit edge and waveguide entrance (x13,000 and x370). ................. 31
Figure 3.7: Edge of sensor beyond exit edge. ............................................................................... 32
Figure 3.8: (a) Microscale sensors on a chip, (b) diffraction pattern from waveguide, and (c)
reflection pattern. ....................................................................................................... 32
Figure 3.9: Microscale CPD sensor experimental setup. .............................................................. 33
Figure 3.10: (a) PDMS microchannel on sensor chip; (b) cross-sectional view of the PDMS
channel. ...................................................................................................................... 34
vii
Figure 3.11: Segmented pattern of Fraunhofer diffraction showing the main observation area. . 34
Figure 3.12: Refractive index measurement results for air, water, ethanol and acetone. ............. 36
Figure 3.13: Experiment results of CPD sensor for different media of known refractive indices.
.................................................................................................................................... 37
Figure 3.14: Microscale sensing experiment results of TIR-based sensor for biofouling formation.
.................................................................................................................................... 38
Figure 3.15: Illustration of coupling light source and its effect at different angles of incidence. 39
Figure 3.16: Illustration of secondary diffraction at exit edge of sensor and possible noise
creation at the chip edge............................................................................................. 39
Figure 3.17: Schematic of erosive wear on (a) a CPD sensor and on (b) an SPR-based (RPD)
sensor. ........................................................................................................................ 40
Figure 4.1: Schematic of SPR-based RPD sensor. ....................................................................... 45
Figure 4.2: Results of numerical simulation with different refractive indices. ............................ 47
Figure 4.3: Refractive index change versus the angles of the convolution peaks. ....................... 47
Figure 4.4: (a) Schematic diagram illustrating RPD sensing mechanism; (b) surface plasma wave
(SPW) formation within metal, biofilm, and background. ........................................ 48
Figure 4.5: Two methods of metalizing the SPR sensor. ............................................................. 49
Figure 4.6: SEM picture of gold layer on the exposed edge of prism. ......................................... 50
Figure 4.7: Experimental setup of SPR-based sensor. .................................................................. 50
Figure 4.8: Image analysis process illustrating how to find the resonance point. ........................ 51
Figure 4.9: Image analysis process illustrating the resonance points of different polarization
states of light. ............................................................................................................. 52
Figure 4.10: Refractive index measurement results for water, ethanol, and acetone. .................. 53
Figure 4.11: Experiment results of RPD sensor for media of known refractive indices. ............. 53
Figure 4.12: Microscale sensing experimental result of the RPD sensor test for biofouling
formation. ................................................................................................................... 54
Figure 4.13: Reflectance variation with respect to the thickness of the gold (Au) layer. ............ 55
Figure 5.1: Schematic illustrating a strategy to remove biofilm deposition from a sensor using
ultraviolet and far-infrared rays as a means of cleaning. ........................................... 62
Figure 5.2: Schematic of a self-powered sensing device. ............................................................. 62
Figure 5.3: Response of reflectance in accordance with incident angle for s- and p-polarized light.
.................................................................................................................................... 63
Figure 5.4: Schematic of Three Exit Waveguide sensors and their corresponding points of
measurement on a convoluted light profile. ............................................................... 64
Figure 5.5: (a) Schematic of conceptual design of sensor using Fresnel diffraction, (b)
corresponding intensity profile. ................................................................................. 65
Table 1: Summary of Microscale Prism-coupler type Sensor for this study ................................ 29
Table 2: Summary of sensitivity study for various sensing techniques. ...................................... 56
1
Chapter 1
Introduction
1.1 Prism-coupler Type Angular Interrogation System
Total Internal Reflection (TIR) is a well known physical phenomenon that has been
researched and utilized for optical observations for several decades [1, 2, 3, 4, 5]. Used as a
sensing technique, it has been appreciated for its ability to provide label-free sensing: that is,
analyses without the use of fluorescent or radioactive materials [6, 7]. The presence of the
critical points and surface plasmonic evanescent (decaying) waves are two of the most
representative mechanisms for sensing employing TIR. Among various TIR applications, the
prism-coupler method, also known as attenuated total reflection (ATR) method, is most
advantageous [8, 9] as the angular interrogation sensors measure variations with respect to the
changes of incident angle of light for high sensitivity and wide range detections [10, 11]. Other
techniques such as the optical fiber sensors [12, 13] and the grating-coupler type sensors [14, 15,
16] have been developed, but they are not as versatile as the prism-coupler type sensor in terms
of sensitivity and detection range [9].
Conventionally, in the prism-coupler type angular interrogation sensors, the change of an
incident light angle is applied by rotating the light source which requires bulky and expensive
rotational devices [17]. The rotating light source is especially problematic for micro-scale
optical experiments and in-situ online measurements. Another problem arises from the
extended scanning time. When used in multi-point measurements, long scanning times may
lead to inaccurate results and discrepancies between the sensors. In contrast, waveguide type
sensors are typically smaller in size, while their application could be limited because of lower
sensitivity and narrower detection range of the propagating light in the guide [8, 18, 19].
2
Intensity
Measurement Type
Se
nsit
ivit
y
Capacity for Miniaturization
Waveguide
Type
Wavelength
Interrogation Type
Angular Interrogation
Type
78~90°
Breakthrough
= Goal
10-1 10-2 10-3 [m]1
10-6
([] RIU-1)
Figure 1.1: Sensitivity plotted against size of current sensors and goal of this study.
Figure 1.1 illustrates the sensitivity versus capacity for miniaturization plot, including
waveguide type, intensity measurement type, wavelength interrogation type and angular
interrogation type sensors. Clearly, the goal is to make miniaturized sensors with good
sensitivity as shown. In this study, we propose to adopt a simple optical phenomenon,
Fraunhofer diffraction, to replace the rotating light source employed in the angular interrogation
sensing technique for the scanning of the incident light angles within a prism-coupler structure.
This principle simplifies the detection mechanism such that the prototype microscale sensor
requires only a two-mask fabrication process. This microscale sensor consists of a built-in
waveguide with a prism-coupler to take advantage of Fraunhofer diffraction occurring at the end
of the waveguide. Since the proposed sensor is fully planar integrated; it can be made using an
IC fabrication process. Therefore, integration with other IC devices and micro-size fabrication
could be readily available, and a low unit cost should be achievable through mass production.
Furthermore, the angular interrogation type sensors typically have better sensitivity and we aim
to maintain the sensitivity of the proposed planar total internal reflection sensors.
Sensed Medium
Diffraction
Figure 1.2: Schematic of proposed prism-coupler type microscale sensor.
3
The proposed sensor could potentially improve the sensitivity and detection range while
maintaining the benefits of using a simple waveguide structure as illustrated in Fig. 1.2. There
are a wide range of applications that may utilize this sensing structure, including process,
motion, gas, and chemical monitoring. Specific characteristics suitable for these applications
include: non-destructive nature for material characterization [20]; highly sensitive for in-situ gas
sensing [21, 22] and artificial olfactory sensing [23, 24]; and possible benefits from scaling
effects which enables the measurement of protein conformation [25], chemical sensing [26],
polymer concentration [27], and monolayer formation (biofouling) [28]. This research
concentrates on the possible applications for biofouling to reduce the maintenance cost in clean
water technologies.
Two types of sensing methods are investigated: Critical Point Detection (CPD) and
Resonance Point Detection (RPD), including details in analysis, fabrication, and experiments.
Both sensor designs enable: (1) the reduction of sensor size and the elimination of moving parts,
(2) the elimination of scanning operation, and (3) planar integration of the entire system.
Moreover, small-sized sensors can better enable multi-point local detections for improved
sensing accuracy for broader application areas.
1.1.1 Critical Point Detection Sensing
The CPD sensor detects the movement of the critical point in order to measure the
refractive index changes utilizing the TIR scheme. The shift of critical point in accordance with
the change of refractive index of the analyte occurs as is illustrated in Figure 1.3. This sensing
method makes use of the rapid drop of intensity at a certain angle of incident light under the TIR.
Since this is a simple physical phenomenon, no additional fabrication or manipulation is required
for sensing.
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1Critical Points
Response of
Refractive Index
Change
Angle of Incidence (deg.)
No
rma
lize
d In
ten
sity
Figure 1.3: Critical point shift on the reflectance profile under TIR conditions with regard to
changes in refractive index.
4
When the reflectance change under TIR conditions is combined with the Fraunhofer
diffraction by a waveguide structure as proposed, resulting light signal becomes convoluted, as
shown in Figure 1.4. Compared to other optical sensors such as the SPR sensor which requires
a thin metal layer, the structure materials of this sensor is less vulnerable to erosive wear.
Therefore, this type of sensor is much better suited for environment that may require chemical
and mechanical polishing on the surface such as biofouling in water desalination stations.
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
=
*
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
Reflectance
Pattern
Diffraction
Pattern
Convolution
Pattern
Figure 1.4: Typical reflectance and diffraction patterns for light of varying incident angles and
the convolution pattern of their combination as a resulting sensor signal.
1.1.2 Resonance Point Detection Sensing
RPD sensors measure the refractive index of a material by detecting the movement of
resonance points through Surface Plasmon Resonance (SPR), and the shift of resonance point in
accordance with the change of the refractive index of an analyte as is illustrated in Figure 1.5.
Unlike the CPD sensor, RPD sensors require a thin metal layer to form the metal/dielectric
interface where surface plasmons resonate and propagate. Thin metal coating material and the
thickness of the layer are two key factors in the optimization of the SPR signals and gold layer of
several tens of nanometers in thickness is commonly used.
5
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
Response of
Refractive Index
Change
Angle of Incidence (deg.)
No
rma
lize
d In
ten
sity
Resonance Points
Figure 1.5: Resonance point shift on the SPR intensity profile according to changes of refractive
index.
When the intensity variations caused by SPR values with respect to the angle of the incident light
are combined with the Fraunhofer diffraction, the resulting light signal is a convolution pattern,
as shown in Figure 1.6. Compared with other optical sensing methods, this technology could
have greater sensitivity. Additionally, the sharply defined output signal offers great clarity for
signal detection in practical measurements.
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
Diffraction
Pattern
=
*
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
SPR Intensity
Profile
Convolution
Pattern
Figure 1.6: Typical SPR intensity profile, diffraction patterns for varying incident angle of light
and the convolution patterns as resulting sensor signals.
6
1.2 Chapter Overviews
Chapter 2 discusses the concept of the proposed prism-coupler type sensing method
using Fraunhofer diffraction. A numerical simulation is conducted to investigate the feasibility
of this sensor and its sensitivity which is comparable to other current sensing methods. A
macroscale experiment using commercial optical components is conducted. The technique is
found to be feasible as a refractive index monitoring sensor.
Chapter 3 presents a discussion of the theoretical background of the proposed micro
CPD sensing technique for measuring the refractive index of an analyte through the utilization of
both Fraunhofer Diffraction and the convolution of reflected and diffracted light. A distinctive
feature of the prototype sensor is an integrated waveguide which takes advantage of the
Fraunhofer diffraction occurring at the end of the waveguide. The experimental results
demonstrate that this sensing technique enables a wider application area than conventional
prism-coupler type sensors while maintaining the quality of measurement. Since the material of
the sensor structure is stiff to resist erosive wear better than other materials used on the sensing
surfaces, tribological theory and techniques are employed to prove the actual superiority of this
class of sensor.
Chapter 4 is a study of Surface Plasmon Resonance (SPR) and its application to the CPD
sensor as presented in chapter 3. The advantage of SPR technology is its ability, under certain
conditions, to measure refractive indices with high selectivity. As is the case with the CPD
sensor discussed in Chapter 3, the proposed RPD sensor takes advantage of an integrated
waveguide. The numerical simulation and experiments are conducted to investigate the
feasibility of this sensor, and the result shows better accuracy than the CPD sensors.
Finally, the chapter 5 concludes the thesis and suggestions for future study are presented.
7
References
[1] A. Otto, ―Excitation of nonradiative surface plasma waves in silver by the method
of frustrated total reflection,‖ Zeitschrift fur Physik A Hadrons and Nuclei, vol.
216, pp. 398-410, 1968.
[2] E. Kretschmann and H. Raether, ―Radiative decay of nonradiative surface plasmons
excited by light,‖ Zeitschrift Fuer Naturforschung, vol. 23A, pp. 2135-2136, 1968.
[3] A. Birks, P. J. Roberts, P. S. J. Russell, and D. M. Atkin, ―Full 2-D photonic
bandgaps in silica/air structures,‖ Electronics Letters, vol. 31, pp. 1941-1943, 2002.
[4] D. Axelrod, E. Hellen, and R. Fulbright, ―Total Internal Reflection Fluorescence,‖
Topics in Fluorescence Spectroscopy, vol. 3, pp. 289-343, 2002.
[5] J. Y. Han, Low-cost multi-touch sensing through frustrated total internal reflection,
Proceedings of the 18th annual ACM symposium on User interface software and
technology, pp. 115-118, 2005.
[6] A. Brecht and G. Gauglitz, ―Label free optical immunoprobes for pesticide
detection,‖ Analytica Chimica Acta, vol. 347, pp. 219-233, 1997.
[7] H. A. Engströma, P. O. Anderssonb, and S. Ohlson, ―A label-free continuous total-
internal-reflection-fluorescence-based immunosensor,‖ vol. 357, pp. 159-166, 2006.
[8] J. Homola, I. Koudelab and S. S. Yee, ―Surface plasmon resonance sensors based
on diffraction gratings and prism couplers: sensitivity comparison,‖ vol. 54, pp. 16-
24, 1999.
[9] J. Homola, S. S. Yee, G. Gauglitz, ―Surface plasmon resonance sensors: review,‖
Sensors and Actuators B, vol. 54, pp. 3-15, 1999.
[10] S. B. Mendes and S. Saavedra, ―On probing molecular monolayers: a spectroscopic
optical waveguide approach of ultra-sensitivity,‖ Optics Express, vol. 4, pp. 449-
456, 1999.
[11] T. G. Giallorenzi, J. A. Bucaro, A. Dandridge, G. H. Sigel, J. H. Cole, S. C.
Rashleigh, and R. G. Priest, ―Optical Fiber Sensor Technology,‖ IEEE Transactions
on Microwave Theory and Techniques, vol. 30, pp. 472-511, 1982.
[12] T. Woo-Hua, L. Yu-Chengb, T. Jiu-Kaia, and T. Yu-Chia, ―Multi-step structure of
side-polished fiber sensor to enhance SPR effect,‖ vol. 42, pp. 453-456, 2010.
[13] J. M. Lopez-Higuera, Handbook of Optical Fibre Sensing Technology, New York,
John Wiley & Sons, 2002.
[14] M. L. Dakss, L. Kuhn, P. F. Heidrich, and B. A. Scott, ―Grating coupler for efficient
excitation of optical guided waves in thin films,‖ Applied Physics Letters, vol. 16,
pp. 523-525, 1970.
[15] K. Tiefenthaler and W. Lukosz, ―Sensitivity of grating couplers as integrated-
optical chemical sensors,‖ Journal of the Optical Society of America, vol. 6, pp.
8
209-220, 1989.
[16] J. Vörös, J. J. Ramsden, G. Csucs, I. Szendr, and S. M. D. Paul, M. Textor, and N.
D. Spencer, ―Optical grating coupler biosensors,‖ vol. 23, pp. 3699-3710, 2002.
[17] H. C. Pedersen, W. Zong, M. H. Sørensen, and C. Thirstrup, ―Integrated
holographic grating chip for surface plasmon resonance sensing,‖ Optical
Engineering, vol. 43, pp. 2505-2510, 2004.
[18] C. R Lavers and J. S. Wilkinson, ―A waveguide-coupled surface-plasmon sensor for
an aqueous environment,‖ Sensors and Actuators B: Chemical, vol. 22, pp. 75-81,
1994.
[19] I. Abdulhalim, M. Zourob, and A. Lakhtaki, ―Surface Plasmon Resonance for
Biosensing: A Mini-Review,‖ Electromagnetics, vol. 28, pp. 214-242, 2008.
[20] D. Nedelkov and R. W. Nelson, ―Surface plasmon resonance mass spectrometry:
recent progress and outlooks,‖ Trends in Biotechnology, vol. 21, pp. 301-305, 2003.
[21] C. Nylander, B. Liedberg, T. Lind, ―Gas detection by means of surface plasmons
resonance,‖ Sensors and Actuators A, vol. 3, pp. 79-88, 1982.
[22] M. Niggemann, A. Katerkamp, M. Pellmann, P. Bolsmann, J. Reinbold and K.
Cammann, ―Remote sensing of tetrachloroethene with a micro-fibre optical gas
sensor based on SPR spectroscopy,‖ Sensors and Actuators B, vol. 34, pp. 328-333,
1996.
[23] J. Vidic, J. Grosclaude, R. Monnerie, M. Persuy, K. Badonnel, C. Baly, M. Caillol,
and L. Briand, ―On a chip demonstration of a functional role for odorant binding
protein in the preservation of olfactory receptor activity at high odorant
concentration,‖ Lab on a Chip, vol. 8, pp. 678-688, 2008.
[24] J. M. Vidic, J. Grosclaude, M. Persuy, J. Aioun, and R. Salesse, ―Quantitative
assessment of olfactory receptors activity in immobilizednanosomes: a novel
concept for bioelectronic nose,‖ Lab on a Chip, vol. 6, pp. 1026-1032, 2006.
[25] H. Sota and Y. Hasegawa, ―Detection of Conformational Changes in an
Immobilized Protein Using Surface Plasmon Resonance,‖ Analytical Chemistry,
vol. 70, pp. 2019-2024, 1998.
[26] J. G. Gordon II and S. Ernst, ―Surface plasmons as a probe of the electrochemical
interface,‖ Surface Science, vol. 101, pp. 499-506, 1980.
[27] G. Sakai, K. Ogata, T. Uda, N. Miura, N. Miura, and N. Yamazoe, ―A surface
plasmon resonance-based immunosensor for highly sensitive detection of
morphine,‖ Sensors and Actuators B: Chemical, vol. 49, pp. 5-12, 1998.
[28] K. A. Peterlinz and R. Georgiadis, ―In Situ Kinetics of Self-Assembly by Surface
Plasmon Resonance Spectroscopy,‖ Langmuir, vol. 12, pp. 4731-4740, 1996.
9
Chapter 2
Macroscale Prism-coupler Type Sensing
using Fraunhofer Diffraction
2.1 Introduction
A prism coupler is an instrument widely used to measure the refractive index and
thickness of materials as well as to couple light in a waveguide. Prism structure in a prism
coupler is utilized to introduce incident light from a variable angle source and to deliver the
reflected light from the prism surface to a detector or screen. For most applications, a prism
coupler characterizes the properties of an analyte using measurements of reflected light emitted
at various angles. Angle scanning devices such as goniometers are usually employed to change
the angle of incident light, and the measurement resolution of the prism coupler is dependent on
the rotational accuracy of the scanning device. For this reason, prism couplers commonly
require a bulky and expensive angle scanning device for high precision measurements.
The prism-coupler method has been researched for adoption in conventional devices
such as ellipsometry type sensors [1], and also in very high sensitivity SPR sensing techniques [2,
3, 4]. Among the prism-coupler type SPR sensors, fan-shaped light diverging schemes do not
need moving parts (for instance, a goniometer). Thus the latter design has been actively studied
in terms of its possibilities in miniaturization [5, 6]. However, due to the limitation that follows
from the fact that the divergence of light requires macro scale optical components such as lenses,
the prospects are remote for this sensor type being becoming a microscale device.
In this study, the divergence light within a diffraction pattern induced by a simple single-
slit is employed in a prism-coupler structure to create a microscale sensor. Therefore, in this
chapter, the feasibility of the sensing concept is first verified through macroscale experiments
utilizing reliable optical components such as commercial single-slits and right-angle prisms. A
precisely constituted macroscale experimental setup generates a TIR response, and a critical
10
point is detected to analyze the refractive index of an analyte. Microscale fabrication and
detailed sensing experiments of CPD sensors are discussed later in Chapter 3. Although SPR is
not incorporated into the macroscale experiment in this chapter, the fundamental concept is
basically the same. An RPD sensor utilizing SPR is discussed in Chapter 4.
2.2 Theoretical Background
2.2.1 Total Internal Reflection
When the incident angle is low with respect to the angle normal to the interface between
two media with different refractive indices, a portion of the ray of light striking the interface is
transmitted and absorbed, and the remaining portion is partially reflected [7]. However, as the
angle of incidence increases, the reflectance of the light increases and eventually reaches a value
of 1 (100% reflection) at the critical point (Fig. 2.1 and Eqn. 2.1). The incident angle at critical
point is determined by the refractive index of the medium being measured; therefore the critical
angle can be calculated to determine the refractive index of a medium to be examined.
2
2
2
121
2
2
121
2
cos1cos
cos1cos
ii
ii
n
nnn
n
nnn
rR
(2.1)
30 35 40 45 50 55 60 65 700
0.5
1
1.5TE Reflection
Angle [deg]
R-T
E
n = 1
n = 1.333
n = 1.45
Angle of Incidence (deg.)
Reflecta
nce
Figure 2.1: Reflectance profile and total internal reflection.
The reflectance profiles in Figure 2.1 show how reflectance changes with regard to the refractive
indices of three selected media, air, water, and milk. For each medium, there are three critical
11
points clearly shown on each of the curves. After experimentally determining the position of a
critical point, it is possible to redraw the profile, and thus to derive the value of refractive index.
This is the basic theoretical background of the sensing methodology employed by the CPD
sensing techniques examined in this study.
2.2.2 Fraunhofer Diffraction
Fraunhofer diffraction is a far-field optical phenomenon which can be observed when a
wave passes through an obstruction. Typically, light waves passing through slits or holes are
diffracted. Fraunhofer diffraction occurs as a result of the interference and reinforcement
interactions of incoming and diffracted waves [7, 8, 9]. In general, Fraunhofer diffraction is
approximated through the following equation:
2
0
sin
II (2.2)
is called sinc function and is defined as,
sin
d (2.3)
where d is the width of the slit and is the wavelength of the light source.
-10 -8 -6 -4 -2 0 2 4 6 8 100
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Angle of Incidence (deg.)
Norm
aliz
ed Inte
nsity
Figure 2.2: Normalized intensity of light and Fraunhofer diffraction.
When the light passes through a small opening or slit, the intensity of the spread of the light
varies with respect to the angle of diffraction, as shown in Figure 2.2. Due to the reinforcement
and destructive interference of light waves, numerous peaks and dark fringes are generated. In
particular, the region around the zero-order (central) peak is brighter than the others. This light
spreading ability of Fraunhofer diffraction can be utilized to manipulate light waves, especially
12
in situations that limit the use of optical components such as a lens. In this study, the
Fraunhofer diffraction is utilized to enable fabrication of microscale planar devices as well as to
minimize the use of complex optical components.
As the distance between the slit and the detector decreases (typically, the position at
which the light signal is observed in the CPD sensor), the Fraunhofer diffraction begins to lose
its dominance over the final pattern of the light, and the final output light pattern begins to follow
Fresnel diffraction [10]. This distance was set to be the minimum design parameter
(dimension) of the sensor so that the sensor is only affected by Fraunhofer diffraction. The
boundary condition between Fraunhofer and Fresnel diffractions is defined by the Fresnel
Number, F , which is given by the following equation [11]:
L
dF
2
(2.4)
where d is a width of the slit (the aperture) and L is the distance between the slit and the
detector. When the Fresnel Number is much smaller than 1, the Fresnel diffraction begins to
dominate.
2.2.3 The Convolution of Reflection and Diffraction Patterns
When two or more patterns of light merge, the resulting pattern is a summation of each
component, which is called a convolution, and the interactions of light waves through
reinforcement and interference cause such convolutions. By setting the peak of the zero-order
Fraunhofer diffraction at a specific point, the profile of convolution for a light wave can be
derived as follows:
2
2
121
2
2
1212
0
sin1cos
sin1cos
sin
rr
rr
d
n
nnn
n
nnn
dcI
RII
(2.5)
where shiftcritr
For example, when reflection and diffraction occur simultaneously, the pattern of light at the
detector becomes complicated as illustrated schematically in Figure 2.3.
13
15
20
25
30
35
40
45
50
55
60
65
0
0.2
0.4
0.6
0.8
1
=
*
15
20
25
30
35
40
45
50
55
60
65
0
0.2
0.4
0.6
0.81
Figure 2.3: Schematic diagram of convoluted light patterns induced by reflection and diffraction.
Convoluted by multiple light waves, the resulting pattern is affected by each individual signal.
Therefore, careful separation of these signals is necessary if there is a need to study any single
pattern contributing to the complex pattern of the convolution. In this study, only the light
reflection profile of the reflection-diffraction convolution pattern is allowed to vary, while the
diffraction pattern remains fixed. Therefore, image analysis is significantly simplified when
compared with the convolution patterns that would result from multiple varying patterns of light.
2.2.4 Detection of Biofouling
Biofouling is the consequence of an unwanted accumulation of biological substances on
exposed surfaces in aqueous environments. Such surfaces are found in pipeline systems [12,
13], heat sinks [14, 15], marine equipment [16, 17, 18, 19], the blood vessels of the human body,
[20] and teeth in the case of dental plaque [21, 22]. The substances that deposit and adhere on
these surfaces give rise to thin films of bacteria, algae, and fungi, as well as inorganic material.
Left unattended, these biofilms may result in the reduced efficiency of aquatic equipment and the
increased frequency of system maintenance. For example, thick biofilms deposited over time
result in dramatically increased maintenance costs because the usage of common antimicrobial
agents is not usually sufficient to remove the biofilms completely [23]. Today, neither the
mechanisms of the biofouling processes nor the procedures needed to control them are clearly
understood, such that the monitoring of the growth of biofilms is a fundamental and practical
issue.
As biofouling progresses to grow thicker, the optical properties of biofilms change.
When the thickness of biofilms increases, their refractive index increases accordingly [24].
However, state of the art biofouling sensors based on optical detection do not take into account
14
changes in the refractive index of biofilms, and this could be a source of measurement errors in
those characterizations [25, 26].
Changes in the refractive index are relatively stable and predictable at the base of the
biofilm because its location is away from the disturbances of the fluid medium flowing above.
Thus protection is provided by biofouling materials between the base and the biofouling surface.
Furthermore, after the initial layer of biofilm is deposited, the newly formed biolayers can have a
great variety of compositions as well chemical and physical properties. The initial deposition
can be seen as medium or substratum capable of providing opportunities for colonization and
competition by other organisms. This unpredictable variation of the biofilm-fluid interface on
top of the biofilm surface can affect the layer thickness, diffusion depth of biofoulants, nutrients,
and other particulates, coefficient of friction, growth rate of the film itself, tendencies toward the
adhesion of organic and inorganic particulates and, significantly, the refractive index. These
variables make the properties of the surface layer of a biofilm difficult to predict and, therefore,
unsuited to the requirements of contemporary sensing technologies. Consequently, the
measurement at the bottom of a biofilm constitutes a more reasonable approach to determining
the baseline property of biofilm formations.
For this study, the refractive index at the interface between the surface of a sensor and
the biofouling media is measured to study the condition of the biofilm and characterized with
respect to time to monitor the formation of biofouling processes.
2.3 Numerical Simulation
Numerical simulation allows us to predict the convolution of reflection and diffraction
patterns of light and to show the feasibility of the sensing technique proposed for this study.
Figure 2.4 shows several cases of convolution patterns with media of differing refractive indices.
In the figures below, thick solid lines represent the convolution patterns which are likely to be
observed experimentally in measurements of the refractive index. The blue and red dotted lines
in the figures illustrate the reflection and diffraction of light, respectively. For each case, the
change of reflection pattern is the only factor contributing to the shift of the peak in the
convolution curve, and this peak constitutes the signal to be detected.
Unlike the original diffraction and reflection patterns, the peak of the convolution line is
sharp, and the tip of this signal clearly shifts with respect to the change of refractive index of the
analyte. As depicted in Figure 2.5, the peak shift in the convolution curve is a possible
indicator for the measurement of a medium‘s refractive index. The result indicates that the
changes of refractive index and corresponding critical angle are linearly related; therefore the
sensitivity is the slope of the fitting line. The theoretical sensitivity of the CPD sensor used in
this simulation was 36 deg. R.I.U.-1
.
15
10 20 30 40 50 60 700
0.2
0.4
0.6
0.8
1
10 20 30 40 50 60 700
0.2
0.4
0.6
0.8
1
10 20 30 40 50 60 700
0.2
0.4
0.6
0.8
1
10 20 30 40 50 60 700
0.2
0.4
0.6
0.8
1
Re
flecta
nce
Relative Incident / Offset Angle (deg.)
Δn = 0 Δn = 0.01
Δn = 0.02 Δn = 0.03
Figure 2.4: Output signals of different refractive indices.
0 0.005 0.010 0.015 0.020 0.025 0.030
40.6
40.8
41.0
41.2
41.4
41.6
41.8
y = 36.37*x + 40.56
Cri
tical A
ngle
(d
eg
.)
Refractive Index Change (Δ R.I.U.)
Figure 2.5: Refractive index change versus angle of convolution peak.
16
2.4 Working Principle
Δθ
CCD
L
D
Single Slit
Exposed Edge
Exit Edge
Input Edge
Light Source
Figure 2.6: Schematic illustration of prism-coupler type sensing mechanism utilizing Fraunhofer
diffraction.
Figure 2.6 illustrates the sensing principle of the macroscale experiment setup for a
prism-coupler type sensor employing Fraunhofer diffraction. The setup consists of a light
source, a single-slit, a right angle prism, and a CCD camera as a detector. The single-slit which
is joined to the light source diffracts light, and the reflected light emerging from the prism is
captured by the CCD. The divergent light reflects against the interface between the edge of the
prism and the analyte, and the refractive index of the analyte determines the position of the
critical point in the resulting light pattern. Apart from the utilization of a single-slit, the
configuration of experimental setup is identical to that of conventional prism-coupler type
sensors [27, 28, 29]. The shift in refractive index of the sensed medium is calculated through
the observation of the movement of the critical point positions of the total internal reflection.
The output signals captured by the CCD can record the position of the critical point of
the reflected light, and the refractive index of the media is calculated by the derivation based on
the Fresnel equation expressed in Equation 2.1 [7]. The path that the light travels is illustrated
in Figure 2.6, where L represents the total distance of light traveling from the input edge of the
prism to the CCD, D gives the distance the light traveling from the reflection point to the CCD,
and d is the shift of the critical point on the CCD detector. The basic formulation of Snell‘s
law states:
12
1
int /sin nnpocritical
(2.6)
17
Therefore, the refractive index of medium being measured is given as follows:
Ldnnn icritprismicritprismanalyte /sinsin __ (2.7)
where crit_i and are, the critical angle at the initial stage (reference material) and the total
angle shift of the critical point due to the change of the refractive index during the experiment,
respectively. Likewise, nanalyte and nprism represent the refractive indices of analytes to be
measured and the prism, respectively.
2.5 Experiment Protocols and Considerations
2.5.1 Direct Refractive Index Measurement
The critical point of the output signal changes as soon as the prism‘s sensing facet
(exposed edge) is exposed to medium. However, since total internal reflection is a surface
phenomenon, it requires an appropriate period of time for the target medium to wet the sensing
surface. Otherwise molecules attaching at the surface will not be distributed homogenously,
resulting in an inaccurate measurement. This is a common problem for all optical sensors, and
it is an important consideration when there is need for precise measurement.
If the analytes are very viscous and adhesive, it will be difficult to remove the molecules
attaching to the sensing surface by purging with the reference material (such as water) when
cleaning the apparatus. Therefore, liquids with low viscosity are recommended for earlier
testing experiments if several analytes are to be measured sequentially by the same device. The
solubility of an analyte in the reference material is also an important factor in deciding the
sequence of measurements. Water is used as a reference material in this study because of its
good solubility with regard to many other liquids. Viscous liquids such as glycerol are tested in
the final stages measurement due to their resistance to the purging process.
2.5.2 Media of Different Optical Properties
In-situ measurements capable of observing the dynamic responses of an analyte are
possible with this sensing technique. Examples of dynamic responses include: the mixing of
liquids, cell growth and densification, as well as biological film deposition and aging. To
observe the formation and aging of biofouling, the sensing surface of the sensor is exposed to the
liquid containing the biofoulants which provide both the adhesive molecules necessary for
biofouling to occur and the feed nutrients for growth. To monitor and characterize biofouling,
the surface refractive index between the prism and analytes is measured and plotted with respect
to time.
2.5.3 Sources of Error
The improper arrangement and alignment of a sensor and optical components are
common sources of measurement error for most optical experiments. Because the intensity of
18
light signal at each phase is a main target for observation, a stable and steady light source is
required in experiments.
The refractive index is also affected by the temperature of the medium. Therefore, the
thermal inhomogeneity, or temperature variation of the analyte and the sensor components
through which the light passes could be a source of measurement error. As a result, temperature
must be controlled throughout all experiments; otherwise compensation must be added during
the data processing process to compensate possible errors coming from temperature variations.
Humidity is also a factor that influences the transmission of light, but it is regarded to be a minor
source of error.
2.5.4 Image Processing
Output signals have been analyzed by Image-pro Plus (Media Cybernetics, USA), an
image processing software. The numerical results in Figure 2.7 clearly indicate the detection of
a critical point. Here, the critical point is revealed as the position after which the intensity of
the output signal rapidly diminishes. The flat region of the curve results from a partially
saturated region of the light pattern induced by Fraunhofer diffraction, and after the critical point,
the output signal decreases with the drop in reflectance.
Critical Point
Figure 2.7: Image analysis process illustrating how to find the critical point.
2.6 Sensing Experiment
2.6.1 Experimental Setup
Macroscale experiments were conducted using optical components including a
commercial single-slit diffraction aperture (precision air slit) and right-angle prism to determine
the feasibility of the sensing concept. As shown in Figure 2.8, the setup of this macroscale
experiment consists of a 630 nm laser source, a single-slit with a 5 um gap, a right-angle fused
silica prism coupled in a container in which several media with different refractive indices for
measurement, and a detector (the CCD chip in a webcam).
19
Laser Source(λ=630nm)
Right Angle Prism(Fused Silica)
5 μm Slit
Medium Tank
(Air, Water, and Milk)
Webcam(CCD)
Figure 2.8: Complete optical setup for macroscale experiment.
In the setup, the incident light provided by the laser source is reflected from the interface
between the reflecting (sensing) facet of prism and the analyte and exits through the exit edge of
the prism. The pattern of reflection is changed in accordance with the refractive indices of the
analytes in the medium tank as discussed in previous sections of this chapter. The final optical
profile is then captured by the webcam.
2.6.2 Experiment Results
Because of the high quality of the commercial optical components used for this
macroscale experiment, the signals were clear and the image process generated precise
measurements was close to expected values. Figure 2.9 shows the first experiment on a clean
water and milk mixture. The experiment started with empty medium tank (air), and water was
added and came into contact with the sensing surface of the prism. Milk was then added and a
change in the refractive index of the mixture was observed. The figure shows captured signals
from each medium as well as the corresponding theoretical patterns of diffracted and reflected
light that have been manually separated into discrete patterns. The captured signals are the
convolution of the Fraunhofer diffraction and the Fresnel reflection, and the critical point appears
on the curve adjacent to the zero-order peak of the diffraction pattern. For the convolution, the
diffraction pattern does not move, while critical point shifts along the diffraction pattern
according to the change of refractive index of various types of media. These output signals
agree well with theoretical derivation given in Equation 2. 5.
20
Air
Wate
rW
ate
r + M
ilk
Patte
rn o
f
Diffra
ctio
n
Patte
rn o
f
Refle
cta
nce
*
Fig
ure 2
.9: T
ypical o
utp
ut sig
nal cap
tured
by w
ebcam
and co
rrespondin
g th
eoretical p
atterns o
f diffracted
and reflected
light
separated
.
21
Critical Point
Figure 2.10: Output signals captured by CCD and corresponding intensity profiles for a milk
biofouling formation.
Observations for the transient behaviors were also conducted to monitor the dynamic
responses of biofouling formation. Figure 2.10 shows the typical output signals captured by
CCD and its intensity profiles for the milk tested after 2, 4 and 8 hours into experiments,
respectively. It is observed that the critical point gradually moved leftwards and the
corresponding refractive indices variations can be calculated to quantify biofouling
characteristics.
The position of critical point was recorded with respect to time, and the refractive index
at each point monitored was calculated as shown in Figure 2.11. As illustrated in Figure 2.6,
there were two groups of setups for measurement in this experiment, and each group was defined
by a distance, D, of 5cm or 10cm, set between the sensing surface of prism and the detector.
Theoretically, the distance should not affect the calculation of the refractive index, and the final
refractive index results for both of the cases showed a reasonable tendency and were consistent
with theoretical predictions. Although temperature was carefully controlled during experiments,
compensation for unexpected temperature variation was conducted and has been displayed as a
dashed line in Figure 2.11.
22
0 2 4 6 8 10 12 14 16 180
100
200
300
Rela
tive p
ixel
Time [per 30 min.]
Pixel movement and R.I. change
0 2 4 6 8 10 12 14 16 181.345
1.35
1.355
1.36
0 2 4 6 8 10 12 14 16 181.345
1.35
1.355
1.36
0 2 4 6 8 10 12 14 16 181.345
1.35
1.355
1.36
0 2 4 6 8 10 12 14 16 181.345
1.35
1.355
1.36
Refr
active I
ndex
Po
sitio
n o
f C
ritica
l P
oin
t (p
ixe
l)
Time (every 30 min.)
■ Group 1 (D = 5 cm)
● Group 2 (D = 10 cm)
■ Pixel Number (measured)
□ Reflective Index (calculated)
--- Temperature Effect
Re
fra
ctive
In
de
x
Figure 2.11: Position of critical point and corresponding refractive indices with respect to time.
2.7 Summary
In this chapter, prism-coupler type sensing mechanism utilizing the Fraunhofer
diffraction induced by a single-slit is studied to validate the proposed architecture for biofouling
sensing. Both numerical simulations and macroscale experiments have been conducted.
Fraunhofer diffraction has been generated by using a commercial single-slit, and a portion of the
diffracted light has reached the sensing edge of the prism with a broad range of incident angles.
The total internal reflection critical point can be determined from the convoluted light with the
combined patterns of the reflection and diffraction light patterns. The shift of the critical point
in accordance with the change of the refractive index of the analyte has been calculated by
numerical simulation. Experimental results from the prototype, macro scale system from
several liquids of different refractive indices have shown good consistency with theoretical
values. The refractive index of biofouling has been measured continuously with respect to
time, and the dynamic responses of biofouling characteristics have been recorded. In
conclusion, the macroscale experiment validates the concept of the proposed sensing principle.
23
References
[1] Maven Biotechnologies website, www.mavenbiotech.com.
[2] Biacore, www.biacore.com
[3] EcoChemie website, www.ecochemie.nl
[4] K-MAC website, www.k-mac.co.kr
[5] Nomadics website, www.nomadies.com
[6] Reichert website, www.reichertai.com
[7] E. Hecht, Optics, 4th ed., Reading, Addison-Wesley, 1979.
[8] K. Singh and B. N. Gupta, "Fraunhofer diffraction in partially space coherent light
by slit aperture with amplitude filters," Optica, Acta, vol. 17, pp. 609-622, 1970.
[9] E. S. Lamar, "Fraunhofer Diffraction Patterns of Squares and Rectangles," Journal
of the Optical Society America, vol. 39, pp. 929-931, 1949.
[10] C. J, Bouwkamp, "Diffraction Theory," Reports on Progress in Physics, vol. 17, pp.
35-100 ,1954.
[11] Y. P. Kathuria, "Fresnel and far-field diffraction of a truncated elliptical Gaussian
beam due to an elliptical aperture," Journal of Modern Optics, vol. 34, pp. 1085-
1092, 1987.
[12] H. -C. Flemming, "Biofouling in water systems ? cases, causes and
countermeasures," Applied Microbiology and Biotechnology, vol. 59, pp. 629-640,
2002.
[13] K. P. H. Meesters, J. W. Van Groenestijn and J. Gerritse, "Biofouling reduction in
recirculating cooling systems through biofiltration of process water," Water
Research, vol. 37, pp. 525-532,2003.
[14] P.A. Pilavachi, "Heat exchanger R&D, a tool for energy conservation?activities
within the non-nuclear energy R&D programme of the european community," Heat
Recovery Systems and CHP, vol. 9, pp. 411-419, 1989.
[15] A. Nabi, P. Rodgers, and A. Bar-Cohen, "Prediction of Thermal Performance
Degradation of Air-Cooled Fine-Pitch Fin Array Heat Sinks due to Fouling,"
Semiconductor Thermal Measurement and Management Symposium, 2006 IEEE
Twenty-Second Annual IEEE, pp. 2-9, 2006.
[16] M. Wahl, "Marine epibiosis. I. Fouling and antifouling: some basic aspects,"
Marine Ecology Progress Series, vol. 58, pp. 175-189, 1989.
[17] S. Snyder, B. J. Zahuranec, M. Whetstone, "Biofouling research needs for the
United States Navy: Program history and goals," Biofouling, vol. 6, pp. 91-95,
1992.
[18] M. E. Callow, J. A. Callow, "Marine biofouling: a sticky problem," Biologist, vol.
24
49, pp. 1-5, 2002.
[19] L. S. Godwin, "Hull Fouling of Maritime Vessels as a Pathway for Marine Species
Invasions to the Hawaiian Islands," Biofouling, vol. 19, pp. 123-131, 2003.
[20] J. A. Schaber, W. J. Triffo, S. J. Suh, J. W. Oliver, M. C. Hastert, J. A. Griswold, M.
Auer, A. N. Hamood, and K. P. Rumbaugh, "Pseudomonas aeruginosa Forms
Biofilms in Acute Infection Independent of Cell-to-Cell Signaling", Infection and
Immunity, vol. 75, pp. 3715-3721, 2007.
[21] P. D. Marsh and D. J. Bradshaw, "Dental plaque as a biofilm," Journal of Industrial
Microbiology & Biotechnology, vol. 15, pp. 169-175, 1995.
[22] T.M. Auschillb, N.B. Artweilerb, L. Netuschil, M. Brecxb, E. Reich, A. Sculeana,
"Spartial distribution of vital and dead microorganisms in dental biofilms,"
Archives of Oral Biology, vol. 46, pp. 471-476, 2001.
[23] J. W. Costerton, P.S. Steward, and E. P. Greenberg, ―Bacterial Biofilms: A Common
Cause of Persistent Infections,‖ Science, vol. 284, pp. 1318-1322, 1999.
[24] M.Leitz, A. Tamachkiarow, H. Franke and K. T. V. Grattan, ―Monitoring of biofilm
growth using ATR-leaky mode spectroscopy,‖ Journal of Physics D: Applied
Physics, vol. 35, pp. 55-60, 2002.
[25] M. G. Trulear and W. G. Characklis, ―Dynamics of biofilm processes,‖ Water
Pollution Control Federation, vol. 54, pp. 1288-1301, 1982.
[26] R. Bakke, R. Kommedal and S. Kalvenes, ―Quantification of biofilm accumulation
by an optical approach,‖ Journal of Microbiological Methods, vol. 44, pp. 13-26,
2001.
[27] R. Ulrich and R. Torge, ―Measurement of thin film parameters with a prism
coupler,‖ Applied Optics, vol. 12, pp.2901-2908, 1973.
[28] J. Midwinter, "Evanescent field coupling into a thin-film waveguide," IEEE Journal
of Quantum Electronics, vol. 6, pp. 583-590, 1970.
[29] H. Rigneault, F. Flory, and S. Monneret, "Nonlinear totally reflecting prism
coupler: thermomechanic effects and intensity-dependent refractive index of thin
films," Applied Optics, vol. 34, pp. 4358-4369, 1995.
25
Chapter 3
Critical-Point-Detection (CPD) Sensing
3.1 Introduction
Total Internal Reflection (TIR) is a well known optical phenomenon when all incoming
energy is reflected back to the incident medium [1]. This phenomenon has been exploited in
several ways for a variety of sensing techniques and applications. TIR-based measurement
techniques have been under investigation for several decades. Notable practical studies began
in 1964 with Osterberg and Smith [2] who utilized a prism to couple a light beam into a surface
wave. Ulrich and Torge [3] utilized total internal reflection phenomena to characterize the
parameters of thin-film. Kitajima and Hieda [4] investigated the use of attenuated total
reflection (ATR) to measure the refractive indices of metal-foils. Nee and Bennett [5] studied
measurements of the refractive index of transparent materials. And recently, Patskovsky and
Meunier [6] developed a phase-sensitive silicon-based total internal reflection sensor using the
differential phase shift between p- and s- polarized components of reflected light.
The TIR-based methods have also been studied by many other researchers [7, 8, 9], but
they have not been widely utilized because of the development of the Surface Plasmon
Resonance (SPR) method which provides higher sensitivity compared with TIR approaches [10,
11, 12]. However there are drawbacks to the SPR technologies. The thin metal layers on the
surface of SPR-based sensors are vulnerable to mechanical damage; and this often leads to
unreliable measurements. In cases where surface friction could be a concern, the use of TIR-
based sensors is more desirable because of their greater structural reliability in resisting
mechanical damage including shearing and abrasion.
Complex and large-scale engineered systems could utilize the TIR technology such as
pressurized liquid transportation systems in desalination applications [13, 14, 15] and food
manufacturing plants [16, 17]. These systems operate on solutions containing solid materials
under high pressure. The improvement of the biofouling sensing can assist in the effective
26
control of the cleaning processes as well as the real-time response to operational faults under
severe shearing conditions. Furthermore, there are other demands calling for the elimination of
the moving parts in conventional TIR-based sensors, such as the rotation goniometer. Such
structural simplification also enables a significant reduction of sensor size and scanning time,
especially for multi-points and/or mobile measurements. Furthermore, the planar integration of
the TIR sensor reduces the cost in device fabrication.
This chapter investigates the microscale total internal refection sensing techniques as
well as possible applications of the CPD sensor.
3.2 Microscale Sensor Design and Fabrication
3.2.1 Sensor Design
Laser Source
Coupling
Final Output
Signal
Core
Cladding Diffraction
Critical
Point
Exposed
Edge
Input
Edge
Exit Edge
Figure 3.1: Schematic illustration of CPD sensing mechanism.
Figure 3.1 illustrates the sensing principle of the prototype sensor which consists of a
light source, an integrated waveguide-prism structure, and a detector. The waveguide is
designed to deliver light from its open end to the prism-coupler, and the reflected light can exit
through the side-facet of the prism. The surface of the prism is exposed to the analytes, and
diffraction occurs at the input edge (the interface of the waveguide and prism) to spread light in
various directions. This design is similar to conventional prism-coupler type sensors without
the need for dynamic scanning of light sources.
27
Δθ
CCD
L
θprism
D (5cm or 10 cm)
Waveguide
Exposed Edge
(1.33mm)
Exit Edge
(0.86mm)Input Edge
(1mm)
Figure 3.2: Schematic diagram of light paths in a prototype CPD sensor.
Figure 3.2 shows the schematic diagram of light paths in a prototype microscale sensor.
The waveguide and the materials used for the prism are the two major differences as compared
with the macroscale setup presented in chapter 2 while the basic sensing mechanisms are the
same. Therefore, the equation to calculate the refractive index of medium being measured is
modified as follows:
Ldnnn icritniticritnitanalyte /sinsin __ (3.1)
where nnit is the refractive indices of the silicon nitride which constitutes the core of the prism-
coupler. prism is a design parameter for the prism selected such that the initial critical point is
positioned appropriately on the CCD for ease of observation.
3.2.2 Fabrication Procedure
In this study, components of the proposed sensor were constructed separately to simplify
fabrication. First the body of sensor which consisted of the prism-coupler and waveguide was
built, and the external light source and photo-detector were assembled separately. Coupling
and alignment of the device were accomplished after all sensor components had been integrated.
As illustrated in Figure 3.3, a two-mask fabrication process was employed to construct
the built-in waveguide with integrated prism. A 2.5 μm-thick silicon dioxide (bottom cladding)
layer and 0.25 μm-thick silicon nitride (core layer) were deposited on a silicon substrate (Fig.
3.3a). The body of the prism-coupler and waveguide were defined with the first mask (Fig. 3.3b),
and the cross sectional area of the waveguide patterned in this step was 4×0.25 mm2. An
28
additional 2.5 μm-thick layer of oxide (top cladding) was deposited on the top of the structure
(Fig. 3.3c). The final structure was defined by the second mask and the use of RIE (reactive ion
etching) to create smooth sidewall surfaces and to achieve a good aspect ratio (Fig. 3.3d).
Finally, the end of the waveguide was cut-opened by wafer dicing, and an optional polishing was
applied to reduce the insertion loss in the light signal. The dimensions of the geometry shown
in this section are determined by optical simulations, and this will be discussed in next section.
(a) (b)
(c) (d)
(b’)
(c’)
(d’)
Si
Silicon Dioxide Silicon Nitride
Figure 3.3: Micro fabrication procedures for proposed sensors.
Silicon nitride was used as the core layer because its refractive index is higher than that
of silicon dioxide to guide light into the device. Standard stoichiometric silicon nitride was
deposited using LPCVD (low-pressure chemical vapor deposition), and its refractive index was
2.05.
3.2.3 Design Variations
There are fabrication parameters that must be determined for optimal sensor
performance. One of the most important parameters is the thickness of each cladding and the
thickness of the core of the waveguide, all of which will affect the quality of resultant signal and
light intensity.
The thickness of the core layer in the built-in waveguide determines the number of the
modes of light the device can accommodate. In order to minimize this number to a single mode
of light for a clear resultant signal, the thickness of the silicon nitride core is investigated. In
general, the number of modes is inversely proportional to the wavelength of light source and
proportional to the diameter of core and the numerical aperture of waveguide. As a reference,
number of mode can be simplified as: [18]:
2
/.2
NAcoreofDiaModeofNumber (3.2)
where NA is numerical aperture and is the wavelength of the transmission light source.
Generally speaking, the strength of signal is directly proportional to the thickness of the silicon
29
nitride core layer. In other words, the light delivering capacity of the waveguide significantly
depends upon the thickness of the core layer. Therefore, the core layer should be sufficiently
thick both to maintain optimal signal strength and to compensate for the insertion loss at the
entrance of the sensor where the light source is coupled with the waveguide. In this study, the
thickness of the core layer was determined by using the thickness that maximizes intensity of the
light, which can experience attenuation through insertion and propagation loss, within the
geometry of the rectangular waveguide which can allow the single mode of light. Another
fabrication constraint is that the stoichiometric silicon nitride cannot be deposited to a thickness
of more than 250 nm over silicon dioxide without risking the formation of cracks. To avoid
cracking, silicon nitride was deposited to 250 nm in thickness to minimize propagation loss.
The thickness of the cladding layer is another constraint. The cladding needs to be
thick enough to prevent excessive propagation leakage of light into the silicon substrate. Figure
3.4 shows several simulation results by using the optical software: BeamPROP (RSoft Inc.,
USA). The left side of Figure 3.4a is the cross sectional view of the waveguide with simulated
light intensity profile and the right side is the schematic three-dimensional illustration of the
waveguide with core and cladding layer. Figure 3.4b is the simulations result showing a
waveguide with insufficient cladding thickness of 1 m and the light intensity drops to about
0.025 of the original strength about 1000 m away from the entrance point of the original
intensity and continues to drop quickly. Figure 3.4c shows the same simulation with a 2.5 μm-
thick oxide layer and the light intensity is about asymptotic to 0.06 of the original strength at
about 1000 m away from the entrance and afterward. Considering the manufacturing process
that thicker cladding layer will be more difficult in the etching process, 2.5 m is selected as the
thickness of the cladding layer.
The sensor was designed to have a prism of 1, 0.86 and 1.33 mm in length, respectively,
for the exit, input, and exposed edges (as illustrated in Fig. 3.2). Two larger sensors have also
been designed with the same prism angle but larger edges (5 and 10 mm for the input edge,
respectively) to characterize the size effects. Table 1 summarizes the important parameters for
the prototype device.
Table 1: Summary of Microscale Prism-coupler type Sensor for this study
Component Dimension
Cladding thickness 2.5 m
Core thickness 0.25 m
Width of waveguide 4 m
Input edge (3 sensors) 1, 5, and 10 mm
PDMS channel height 200 and 300 m (Chapter 3.3.1)
30
Ho
rizon
tal D
irectio
n (μ
m)
Pro
pa
gatio
n D
irectio
n (μ
m)
Pro
pa
gatio
n D
irectio
n (μ
m)
IntensityVertical Direction (μm)
Sin
gle
Mo
de
Lig
ht
Co
re
(a)
(b)
(c)
Cla
dd
ing
Th
ick
ne
ss
= 1
.0 μ
m
Cla
dd
ing
Th
ick
ne
ss
= 2
.5 μ
m
Figure 3.4: Optical simulation results showing coupling of a single mode of light and the
propagation loss during its passage through the waveguide.
31
3.2.4 Fabrication Results
The quality of sensing results depends on the quality of fabrication processes including
mask alignment and the smoothness of the surface. Figures 3.5 and 3.6 show the etched
surfaces of the sensing (reflection) and exit edges of the prism and the entrance of the waveguide
into the prism-coupler. These figures show the sub-microscale roughness and the poor quality
of surface roughness at the entrance to the prism-coupler structure. These are the main sources
for the degradation of output light signal. Since the roughness of an etched surface depends
strongly on the etcher and etching recipes, using a higher quality silicon dioxide etcher could
yield a better surface than the general-purpose etcher that was used for this study. Possible
post-processing to smooth the surface could also improve the sensor performance.
Silicon
Dioxide
Silicon
Nitride
Figure 3.5: SEM images of sensing edge of sensor (x3,800 and x13,000).
Silicon
Dioxide
Silicon
Nitride
Waveguide
Entrance
Figure 3.6: SEM pictures of exit edge and waveguide entrance (x10,000 and x370).
In order to reduce light scattering, the exit edge of the prism has been set several tens of
micrometers from the edge of the sensor chip. However, as shown in Figure 3.7, damage to the
edge of the sensor chip is significant to introduce noises in the output signals.
32
Chip EdgeChip Edge
Figure 3.7: Edge of sensor beyond exit edge.
A completely fabricated microscale sensor is shown in Figure 3.8a. Figures 3.8b and
3.8c show the measured diffraction and reflection patterns of the light traveling through the built-
in waveguide and prism-coupler. The waveguide and prism-coupler in the figure were
intentionally degraded after deposition processes to make the light path more visible.
Exposed Edge(Reflection Edge)
Waveguide
Prism-coupler
Diffraction
WaveguideReflection
Exposed Edge
(a) (b)
(c)
Figure 3.8: (a) Microscale sensors on a chip, (b) diffraction pattern from waveguide, and (c)
reflection pattern.
3.3 Experiments and Results
3.3.1 Experimental Setup
Microscale sensors are relatively difficult to test compared to macroscale sensors
because the small dimensions of the waveguide entrance. Reliable but complicated optical
components including a laser source and CCD were used in these experiments. As shown in
33
Figure 3.9, a laser source was directly coupled with the opening of the built-in waveguide of the
sensor. The wavelength of the light source used was 780 nm because biofilms, which consist of
extracellular polymeric substances (EPS), sediments, and bacteria, are sensitive around this
wavelength [19]. Usually, each of the constituents of the biofilm has different optical
properties, but the mutual absorbance, which can introduce measurement errors, is relatively
small at wavelengths around 780 nm and higher [20, 21]. Furthermore, the reflectance is high
when the wavelength of light is greater than about 700 nm such that the output signal of the
sensor is stronger [22].
Laser Source(λ=780 nm)
CCD
Sensor
in Holder
Figure 3.9: Microscale CPD sensor experimental setup.
As shown in Figure 3.10, a polydimethylsiloxane (PDMS) channel was constructed and
placed on the sensor chip and the exposed edge was in contact with the liquid media. To
prevent leakage, the size of channel was determined in accordance with the viscosity of the
liquid media being measured. For the prototype experiments, a number of analytes including
milk were used as media, and the height of the microchannel was set from approximately 200 to
300 μm. The height of the channel diminishes as biofilms grow thicker during experiments;
therefore it is important to secure a sufficient initial height to prevent the blockage of the supply
of the liquid that contains the biofoulants. To minimize the accumulation of unwanted floating
substances as well as to avoid excessive shearing acting of the biofouling formation, the liquid
media was introduced at a flow rate at 0.1~0.5 ml/hr.
34
In Out
Channel
(a) (b)
Figure 3.10: (a) PDMS microchannel on sensor chip; (b) cross-sectional view of the PDMS
channel.
3.3.2 Calibration
In the case of Fraunhofer diffraction, more than 80% of the light intensity concentrates
around the center of the zero-order peak within the range of one eighth of the peak to peak
distance as illustrated in Figure 3.11. The prototype sensor was designed to have the critical
point of the reference material (water) occur within this range which is selected as a main
observation area.
The base angle of the prism, prism, previously illustrated in Figure 3.2, was set to 41°.
This angle is close to the incident light angle of the total internal reflection from the silicon
nitride to the reference material (water) at 40.56º. The main observation area remains adequate
until the refractive index of analyte reaches about 1.4, which corresponds to the incident angle of
43.07º.
15 20 25 30 35 40 45 50 55 60 650
0.2
0.4
0.6
0.8
1
Distance to First Peak
1/8 Distance to
First Peak
Critical Point Shift Direction
Main Observation
Region
Norm
aliz
ed I
nte
nsity
Incident Angle (deg.)
Figure 3.11: Segmented pattern of Fraunhofer diffraction showing the main observation area.
35
Once the sensor was tested with water, ethanol was tested, and was used for calibration
purpose. Ethanol is chosen as it can be easily purged or cleaned with water.
3.3.3 Direct Refractive Index Measurement
To illustrate the basic sensing principle as a refractive index sensor, direct refractive
index measurements were undertaken for a number of liquids, including water, ethanol, acetone,
and glycerol. Recorded signals, processed images, and critical points for each of the analytes
are displayed in Figure 3.12. Using the imaging software (Image-pro Plus), the positions of the
critical points are extracted from the output signals. As illustrated in Figure 3.13, the measured
refractive index values were in good agreement with the values in the literature. In this figure
the ―reference line‖ shows the divergence of each measurement for numbers reported in the
literature. In the case of the glycerol measurement, another sensor with a higher designed prism
base angle was used to shift the detection range for the increased refractive index. The error
range of the measurements for four selected media with different refractive indices was less than
± 0.002 R.I.U (Refractive Index Unit).
36
Air
Wa
ter
Eth
an
ol
Ac
eto
ne
No
Critica
l Poin
t
Critica
l Poin
t
Critica
l Poin
t
Critica
l Poin
t
Refe
rence
Poin
t
Figure 3.12: Refractive index measurement results for air, water, ethanol and acetone.
37
1.32 1.34 1.36 1.38 1.4 1.42 1.44 1.46 1.481.32
1.34
1.36
1.38
1.4
1.42
1.44
1.46
1.48
Refractive Index (literature)
Refr
active I
ndex (
measure
d)
Glycerol (R.I. 1.473)
Ethanol(R.I. 1.359)
Acetone (R.I. 1.360)
Water (R.I. 1.333)
Reference Line
Re
fra
ctive In
de
x (
me
asu
red)
Refractive Index (literature)
Figure 3.13: Experiment results of CPD sensor for different media of known refractive indices.
3.3.4 Biofouling Sensing
Figure 3.14 shows three independent sensor outputs in terms of pixel positions
(beginnings and ends of output profiles) with respect to time during the 9-hour experiment in the
environment of milk and water combination. The downward movements of the TIR critical
points (the top symbols) result from biofouling, and the corresponding change in the refractive
index is as much as 0.0089. Variations between each individual measurement could be due
mainly to the random quality of the biofouling accumulation process as well as fabrication
variations of different sensors [23]. The red curve at the bottom of the figure represents the
average pixel position of the reference point, the leftmost point of the output signal in the inset
picture. The positions of the reference points of individual measurements are indicated by the
symbols in the lower part of the figure. As is evident in Figure 3.14, throughout the duration of
the experiment, there was very little divergence from the reference point. Since the intensity of
the output signal before the critical point is within the region of total internal reflection, the
reference point does not shift. Therefore, this observation verifies that the shifts of critical
point were due to the change of the refractive index on the sensing surface.
38
0 2 4 6 8 10 12 14 16 1840
60
80
100
120
140
160
180
200
220
Time (every 30 min.)
Poin
ts in P
ixel
Measurement 1
Measurement 2
Measurement 3
Critical Point
Reference PointPo
sitio
n o
f C
ritica
l P
oin
t (p
ixe
l)
Time (every 30 min.) Figure 3.14: Microscale sensing experiment results of CPD sensor for biofouling formation.
3.4 Discussions
3.4.1 General
The experimentation for this research was designed to be simple and straightforward
because this study is intended to examine the feasibility of a sensing technique. Therefore, the
complicated fabrication process of an integrated light source and photo-detector was not included
in the prototype sensor design. A more elaborate and accurate installation of the sensor
components would be required for further studies. As a result, it was not possible to obtain
exact shape of the output signal from different sensors. However, the general observations as
recorded support the feasibility study from the prototype sensors.
During the calibration process, multiple efforts were made to generate the best output
signal. First of all, the variables associated with the light source including its power and
coupling angle were adjusted to change the quality of output signal. As shown in Figure 3.15, a
parallel coupled light source may cause unexpected signal noises induced by scattering against
the facets of prism body and by the propagation of light through the silicon dioxide cladding.
Therefore, tilting the light source to certain angles provides better signals even though the
intensity of light drops due to the reduction of light intensity resulting from insertion and
propagation loss.
39
Propagation through
Top Cladding
Partially Reflected
Figure 3.15: Illustration of coupling light source and its effect at different angles of incidence.
Secondly, a focusing lens was used to compensate low output signal strength by concentrating
the projected light and reducing the spot size of the light. However, manipulation of a light
source coupled with a lens was extremely difficult. Finally, the measurement direction of the
CCD also affects the quality of the output signal. As shown in Figure 3.16, the output signal
that exits the thin silicon nitride core layer is diffracted by the air at the exit edge of the sensor.
The lower parts of the light beam hit the silicon substrate, and the reflection of this light can
contain noises. Furthermore, when other parts of the light beam are reflected by the rough edge
of sensor chip, the noise becomes greater. To avoid these problems and obtain the best image,
optimal angles for positioning the detector had to be determined for the individual sensors.
Diffraction at Exposed
Edge of sensor
Light
Scattering
Figure 3.16: Illustration of secondary diffraction at exit edge of sensor and possible noise
creation at the chip edge.
3.4.2 Tribology
Wear is the process by which material is removed from a solid surface under mechanical
shearing conditions. There are several factors which cause wear, one of which is the impact of
particles of solids, liquids, and even gasses against a solid surface [24]. The rate of wear, or the
wear volume rate, is defined as:
40
Hg
vsWkV e
2/' (3.3)
Here, particleofsizeHvfke ,, ; is the effect of particle speed; is the angle factor;
is the effect of material hardness; is abrasive size effect; v is particle velocity; and g
is gravitational constant. The above equation shows that wear is inversely proportional to the
hardness of material and directly proportional to the abundance of abrasive particles on the
surface of interest.
(a) (b)Particle that
causes wear
Figure 3.17: Schematic of erosive wear on (a) a CPD sensor and on (b) an SPR-based (RPD)
sensor.
The metal layer is the essential component of SPR-based sensors; therefore once the
metal layer is partially eroded or unevenly worn, the sensor loses functionality. As shown in
Figure 3.17b, the metal layer used in the SPR sensor, is typically made of gold (Au), which has a
hardness of 150 kg/mm² [25, 26]. Consequently, such sensors are vulnerable to erosive wear
due to their low hardness value. When exposed to harsh environments, the life expectancy of
this type of sensor is shorter than that of CPD sensors, especially when many particles come into
contact with the measuring surface. (e.g. pressurized sea water in desalination plants)
However, the CPD sensor shown in Figure 3.17a, which consists of dielectric materials,
has a higher hardness value, 3.5x10³ kg/mm², than would be found in an SPR-based sensor [27].
Therefore, CPD sensors are less vulnerable to erosive wear.
3.5 Summary
The feasibility of utilizing a planar, integrated critical-point-detection type sensor for
refractive index measurement and biofouling characterization is demonstrated in this chapter.
The CPD sensor for this research presents a novel approach to the usage of Fraunhofer
diffraction and introduces the possibility for conventional prism-coupler type sensors to be
planar and in the microscale. This study details the working principles of sensing mechanism,
design, fabrication and characterizations of the refractive index sensors, including image analysis
41
and experimental verifications. The CPD sensor for this study utilizes a simple optical design
with a built-in waveguide as a prism-coupler sensor. Furthermore, the fabrication process is
simplified by the utilization of only two masks for the entire fabrication process. Optical
simulation of this sensor architecture and design/fabrication considerations are discussed. The
tribology aspects of the proposed sensor are analyzed. It is predicted this CPD sensor is better
suited for harsh environments due to the superior hardness of their exposed surfaces. Several
liquids have been tested as base-line characterizations for the sensor, and measurements of
refractive indices were in good agreement with the values given by manufacturers (literature)
with error ranges of less than ± 0.002 R.I.U. During the biofouling experiments, the sensor
measured the change of the surface refractive index of a testing liquid (milk), and a shift of as
much as 0.0089 during a 9-hour test due to biofouling has been measured. It is expected that
this sensing technique could find many potential applications, especially, where there is a need
for in-situ, multi-point local characterization of analytes.
42
References
[1] E. Hecht, Optics, 4th ed., Reading, Addison-Wesley, 1979.
[2] H. Osterberg and L. W. Smith, "Transmission of Optical Energy Along Surfaces:
Part II, Inhomogeneous Media," Journal of the Optical Society America, vol. 54,
pp. 1078-1079, 1964.
[3] R. Ulrich and R. Torge, ―Measurement of thin film parameters with a prism
coupler,‖ Applied Optics, vol. 12, pp.2901-2908, 1973.
[4] H. Kitajima, K. Hieda, and Y. Suematsu, "Use of a total absorption ATR method to
measure complex refractive indices of metal-foils," Journal of the Optical Society
America, vol. 70, pp. 1507-1513, 1980.
[5] S. F. Nee and H. E. Bennett, ―Accurate null polarimetry for measuring the
refractive index of transparent materials,‖ Journal of the Optical Society America,
vol. 10, pp. 2076-2083, 1993.
[6] S. Patskovsky, M. Meunier, and A. V. Kabashin, ―Phase-sensitive silicon-based
total internal reflection sensor,‖ Optics Express, vol. 15, pp. 12528, 2007.
[7] E. Goormaghtigh, V. Raussens, J. Ruysschaert, ―Attenuated total re£ection infrared
spectroscopy of proteins and lipids inbiological membranes,‖ Biochimica et
Biophysica Acta, vol. 1422, pp. 105-185, 1999.
[8] A. Zhang and P. S. Huang, ―Total internal reflection for precision small-angle
measurement,‖ Applied Optics, Vol. 40, pp. 1617-1622, 2001.
[9] S. Patskovsky, I. Song, M. Meunier, and A. V. Kabashin, ―Silicon based total
internal reflection bio and chemical sensing with spectral phase detection,‖ Optics
Express, vol. 17, pp. 20847-20852, 2009.
[10] J. Homola, Surfaec, ―Plasmon Resonance Based Sensor,‖ Berlin, Springer, 2006.
[11] A. Otto, ―Excitation of nonradiative surface plasma waves in silver by the method
of frustrated total reflection,‖ Zeitschrift fur Physik A Hadrons and Nuclei, vol.
216, pp. 398-410, 1968.
[12] A. S. Barker Jr., ―Direct optical coupling to surface excitations,‖ Physical Review
Letters, vol. 28, pp. 892-895, 1972.
[13] C. C. K. Liua, J. Park, R. Migitaa and G. Qina, "Experiments of a prototype wind-
driven reverse osmosis desalination system with feedback control," Desalination,
vol. 150, pp. 277-287, 2002.
[14] L. WEN-TSO, "Biofilms in water separation membrane processes: From
community structure and ecological characteristics to monitoring and control," in
Civil Engineering, vol. Doctor of Philosophy: National University of Singapore,
2007.
[15] K. Ouazzania and J. Bentama, "A promising optical technique to measure cake
43
thickness of biological particles during a filtration process," Desalination, vol. 206,
pp. 36-41, 2007.
[16] X. D. Chen, D. X. Y. Li, S. X. Q. Lin and N. Ozkan, "On-line fouling/cleaning
detection by measuring electric resistance??equipment development and
application to milk fouling detection and chemical cleaning monitoring," Journal of
Food Engineering, vol. 61, pp. 181-189, 2004.
[17] H. Bausera, H. Chmiela, N. Stroha and E. Walitza, "Control of concentration
polarization and fouling of membranes in medical, food and biotechnical
applications," Journal of Membrane Science, vol. 27, pp. 195-202, 1986.
[18] J. Crisp, Introduction to Fiber Optics, 2nd ed., Oxford, Newnes, 2001.
[19] A. W. Decho and T. Kawaguchi, ―Sediment properties influencing upwelling
spectral reflectance signatures: The ‗biofilm gel effect‘,‖ vol. 48, pp. 431-443,
2003.
[20] G. H. Meeten, A. H. North and F. M. Willmouth, "Errors in critical-angle
measurement of refractive index of optically absorbing materials," Journal of
Physics E: Scientific Instruments, vol. 17, pp. 642-643, 1984.
[21] P. R. Jarvis and G. H. Meeten, ―Critical-angle measurement of refractive index of
absorbing materials: an experimental study,‖ Journal of Physics E: Scientific
Instruments, vol. 19, pp. 296-298, 1986.
[22] S. L. Broschat, F. J. Loge, J. D. Peppin, D. White, D. R. Call, and E. Kuhn,
―Optical reflectance assay for the detection of biofilm formation,‖ Journal of
Biomedical Optics, vol. 10, pp. 044027, 2005.
[23] J. Wimpenny, ―Ecological determinants of biofilm formation,‖ Biofouling, vol. 10,
pp. 43-63, 1996.
[24] G. W. Stachowiak, and A. W. Batchelor, ―Engineering Tribology,‖ Burlington,
Elsevier Butterworth-Heinemann, 2005.
[25] J. A. Augis, C. C. LO, and M. R. Pinnel, "The hardness and ductility of sputtered
gold films," Thin Solid Films, vol. 58, pp. 357-363, 1979.
[26] N. Gane and J. M. Cox, "The Micro-hardness of Metals at very Low Loads,"
Philosophical Magazine, vol. 22, pp. 881-891, 1970.
[27] J. Z. Jiang1, F. Kragh1, D. J. Frost, K. Stahl and H. Lindelov, "Hardness and
thermal stability of cubic silicon nitride," Journal of Physics: Condensed Matter,
vol. 13, pp. 515-520, 2001.
44
Chapter 4
Resonance-Point-Detection (RPD) Sensing
4.1 Introduction
Surface plasmon resonance (SPR) is an optical phenomenon that occurs when plane-
polarized light reflects against metal under the condition of total internal reflection. Similar to
the CPD sensing technique, this phenomenon may be used for measuring the refractive index of
an analyte at the surface of the sensor. SPR has been studied extensively for several decades,
especially for chemical and biological experimentations due to its outstanding sensitivity.
Ever since the first investigations of surface plasma waves were conducted by Wood in
1902 [1], innumerable experiments have been conducted, a vast body of research has emerged,
and many kinds of optical devices using the phenomenon have been invented. Otto [2] and
Kretschmann [3] independently conducted studies on the optical excitation of surface plasmons
and developed remarkable experimentations. Pockrand et al. [4] introduced the usage of SPR
for the characterization of thin films, while Gordon and Ernst [5] employed the phenomenon to
monitor the condition of metal interfaces. Nylander and Liqedberg [6, 7, 8] conducted notable
research in several investigations for the possible utilization of SPR for gas detection and
biosensing. Biacore International AB, a life science products company, now merged into GE
Healthcare, USA [9], opened up the commercial SPR biosensor market in 1990.
There are several different technological approaches to the utilization of the SPR
phenomenon. Two of the most representative methods are angular interrogation (the
measurement of the change of the resonant angle) and wavelength interrogation (the
measurement of the change of the resonant wavelength at a fixed incident angle) [10]. Most
devices using either of these approaches are based on the prism-coupler method, and
consequently this sensing technology is the most common and widely used in SPR sensing
devices. Decladded optical fiber with a thin, metal layer on top of the exposed area is another
well-known application using the SPR technique, but its lower sensitivity and narrow detection
45
range prevent its widespread use.
Despite its exquisite sensitivity, the prism-coupler type SPR sensing technique is not
widely used for microscale applications due to its bulky structure. Instead, other kinds of
sensors such as the waveguide type are commonly used, although the sensitivity and detection
range of this particular sensor are not as good as the prism-coupler method.
This chapter proposes a new sensing technique which enables the microscale prism-
coupler type angular interrogation SPR sensing method, named ―RPD sensing.‖ This
technology employs the CPD sensor structure discussed in Chapter 3. The potential advantages
include possible good sensitivity despite its small device size and the elimination of scanning
operation and time. This simple methodology could find a broad range of applicable for a
variety of small-scale optical measurements
4.2 Theoretical Background
Surface Plasmon Resonance
Surface plasmons are electromagnetic waves on the interface shared by a thin metal film
and a dielectric material. When excited by photons, these surface plasmons propagate in the
direction parallel to such interface. This phenomenon is known as surface plasmon resonance.
The existence of surface plasmon resonance can be demonstrated in the simple procedure as
illustrated in Figure 4.1. Without the metal layer, a total reflection of light occurs, and 100% of
the light is reflected when the incident angle is equal to or greater than the critical angle. When
the interface of the prism and the analyte is coated with a thin film of a noble metal (typically
gold), the intensity of reflected light suddenly drops and then recovers as the angle of incidence
increases. This angle is defined as the Resonance Angle [11].
nmetal n1
Incident
Light Detector
Prism
Feed WaterBiofouling
Metal Layer
n2
SPR
Figure 4.1: Schematic of SPR-based RPD sensor.
This phenomenon is a consequence of the presence of surface charge density waves at
the dielectric/metal interface and has dielectric constants of different signs [12]. A surface
plasma wave (SPW) is an electromagnetic wave, specifically a transverse-magnetic (TM) wave,
that propagates along the interface. The SPW is defined with respect to its electromagnetic
46
field distribution and its propagation constant, and the propagation constant, , is expressed by
the following equation:
2
2
dm
dm
dm
dm
n
nkk
(4.1)
where k is the free space wave number, m and d are the dielectric constants of the metal
and the dielectric, respectively, and dn is the refractive index of the dielectric.
When the wave vector of the incident light, kp, matches the wave vector of the surface
charge density waves, ksp, the surface plasmons resonate, and the electromagnetic field of the
SPW is confined at the interface and decays evanescently [13]. This resonance point is usually
displayed as an upside-down peak in the intensity profile of the detector‘s output signal. The
change in the refractive index gives the change of the propagation constant of the SPW
propagating along the metal layer, and this can be detected by monitoring the intensity profile of
output signal [14].
There are several types of sensors to detect the resonance point, and the sensor design
for this study is based on the angular interrogation technique which measures optical signals with
regard to the change of the incident angle. This sensing technology is acknowledged to offer a
higher sensitivity better than other kinds of SPR sensors, and it will be discussed in Section 4.7.2.
4.3 Numerical Simulation
A numerical simulation was conducted to observe the output signal of this sensing
technique, and the light patterns that resulted from the convolution of the diffractions and
reflections of the light affected by SPR were plotted. Figure 4.2 shows the result of a numerical
simulation of analytes (e.g. water containing biofoulants) with different refractive indices. The
solid lines represent convolution patterns, dotted blue lines illustrate the SPR signal, and dotted
red lines illustrate the diffraction of light. As depicted in the these graphs, the peak of the
convolution line is sharp, and the tip of the signal‘s upside-down peak clearly shifts to the right
with respect to the changing refractive indices of the analytes.
Using the resonance angles corresponding to the peak‘s shifting positions, the refractive
index of the analyte was determined and graphed against the incident angles as shown in the
Figure 4.3. The sensitivity of the SPR sensor used in this simulation was acquired from the
slope of the fitting curve to be 90 deg. R.I.U.-1
.
47
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Re
fle
cta
nce
Relative Incident / Offset Angle (deg.)
Δn = 0 Δn = 0.01
Δn = 0.02 Δn = 0.03
Figure 4.2: Results of numerical simulation with different refractive indices.
0 0.005 0.010 0.015 0.020 0.025 0.03040.0
40.5
41.0
41.5
42.0
42.5
43.0
43.5
y = 90.07*x + 40.45
Resonance A
ngle
(deg.)
Refractive Index Change (Δ R.I.U.)
Figure 4.3: Refractive index change versus the angles of the convolution peaks.
48
4.4 Working Principle and Design
The basic structure of the RPD sensor used in this study is the same as the CPD sensor
discussed in the preceding chapter. As shown in Figure 4.4a, it differs only in the theory of its
operation and in the shape of the output signal generated. There are several types of SPR
sensors, and one of the most common setups, shown in Figure 4.1, is the Kretschmann
configuration. This type of SPR sensor was chosen for present study because of its prism-
coupler type is equivalent to that of a CPD sensor, and, consequently, such a configuration
requires no structural modification of the CPD sensor setup utilized in Chapter 3.
Laser Source
Coupling
Final
Output
Signal
Core
Cladding
Diffraction
Surface Plasmon
Resonance
Exposed
Edge
Input
Edge
Exit Edge
Thin Metal
Layer
(a) (b)
Biofilm
SPW
Metal Layer
Background Material
Prism-coupler (Sensor)
Figure 4.4: (a) Schematic diagram illustrating RPD sensing mechanism; (b) surface plasma wave
(SPW) formation within metal, biofilm, and background.
In cases where the target analyte or biofouling formation is as thick as the penetration
depth of the SPW (usually several hundreds of nanometers) [14], as shown in Figure 4.4b, the
real part of the propagation constant, , which is explained in Section 4.2.1, is changed by the
refractive index change of the analytes. The relationship between the propagation constant and
the refractive index change is approximately defined as follows:
nkRe (4.2)
where k is the free-space wave number discussed in the previous section in this chapter.
When a light is reflected at the boundary of the metal layer and the prism-coupler, the
evanescent wave propagates along the interface with a constant determined by the incident angle
of light. Therefore, the measured incident angle of the light at the event of resonance (the
resonance point or upside-down peak in Figure 4.2) can give the propagation constant, and,
therefore, the change in the refractive index of the analytes.
49
4.5 Microscale Sensor Fabrication
The basic structure of the RPD sensor used this study is the same as the CPD sensor
discussed in the previous chapter. In other words, the structure of the CPD sensor is adopted
for the construction of the RPD sensor. There is only a single additional step required, the
deposition of a layer of metal on top of the sensing (exposed) edge of the CPD sensor.
Figure 4.5 shows the additional fabrication step required for the construction of the RPD
sensor. As shown, photoresist was used to specify the area to be metalized through a lift-off
process. There are two ways to metallize the sensor, but for easy deposition setup, a
metallization of only the reflection edge was conducted and 47 nm of gold (Au) was deposited
with an e-beam evaporator along with a 3 nm titanium adhesion layer. The thickness of the
gold layer was determined by a numerical simulation that will be discussed in Section 4.7.1.
The quality of the metal layer affects the quality of the output signal because the
roughness of the prism‘s reflection edge is critical for this microscale sensor. Figure 4.6 shows
the quality of gold layer deposited on the reflection (sensing) edge of the sensor. As in the
previously fabricated CPD sensor, a sub-microscale roughness has been induced during the
etching process and this would affect the quality of output signal.
Area to be Metalized
Photoresist(a)
(b)
Photoresist
Figure 4.5: Two methods of metalizing the SPR sensor.
50
Au on
Silicon Dioxide
Au on
Silicon
Nitride
Figure 4.6: SEM picture of gold layer on the exposed edge of prism.
4.6 Experiments and Results
4.6.1 Experimental Setup
Measurements using an RPD sensor were conducted with a setup that was basically
identical to the one discussed in the preceding chapter. This sensor consisted of the optical
components used in the CPD sensor, and the experimental protocols under which the
experiments were performed were the same. However, since the surface plasma wave (SPW)
used to generate the resonance is a transverse-magnetic (TM) wave, a half-wave plate was added
to the experiment setup to filter light from the laser source in order to achieve an appropriate
polarization state of that light (Figure 4.7).
Laser Source(λ=780 nm)
CCD
Sensor
in Holder
Half-Wave
Plate
Figure 4.7: Experimental setup of RPD sensor.
51
4.6.2 Image Processing
The profiles of the measured output signals of an RPD sensor differ from those
generated by a CPD sensor, although the way to find the resonance point (the inverted peak) is
identical. Figure 4.8 shows numerical results that clearly indicate the detection of the resonance
point which is apparent as the position at which the intensity of the output signal dips suddenly.
The flatness of the curve from about 180 pixels to around 300 pixels is the result of a partially
saturated region of the light pattern induced by Fraunhofer diffraction, and the intensity profile in
the region after the resonance point shows effect of the convolution of the diffracted light and the
SPR.
Resonance Point
Figure 4.8: Image analysis process illustrating how to find the resonance point.
4.6.3 Effect of Polarization
General-purpose polarizers do not prove sufficiently precise performance. Therefore,
for accurate measurement, calibration of polarizer is necessary. Figure 4.9 shows changes in
the output signal and the corresponding polarization state of the light wave. In the figure, the
polarizer was set to find the output signal with the sharpest resonance point as a means of
calibration.
52
TM mode
TM + TE
TE mode
Resonance Point
Figure 4.9: Image analysis process illustrating the resonance points of different polarization
states of light.
4.6.4 Experimental Results
As was the case with the CPD sensing technique studied in Chapter 3, four different
media were selected and tested. Figure 4.10 shows the measurement of the resonance point of
each medium for this test rather than the critical point as was the case in the earlier examination
of CPD sensing. Figure 4.11 compares the refractive indices obtained from the experimental
results with values in the literature.
Based on its theoretical performance, the RPD sensor was expected to display better
selectivity. However, the actual result shows measurement errors in the range of ± 0.002 R.I.U.
These errors are independent of the SPR-based sensing method and arise from problems with the
alignment of the optical measurement set up as well as fabrication defects and problems in the
resolution of the measurement devices. As discussed in the preceding chapter, the alignment of
the optical devices is among the factors most likely to contribute to error. The incident light
that scatters on the side of the exit edge due to its roughness creates additional difficulties for
precise measurement.
53
Water
Ethanol
Acetone
Reference Point
Resonance
Point
Resonance
Point
Resonance
Point
Figure 4.10: Refractive index measurement results for water, ethanol, and acetone.
1.32 1.34 1.36 1.38 1.4 1.42 1.44 1.46 1.481.32
1.34
1.36
1.38
1.4
1.42
1.44
1.46
1.48
Refractive Index (literature)
Refr
active I
ndex (
measure
d)
Glycerol (R.I. 1.473)
Ethanol(R.I. 1.359)
Acetone (R.I. 1.360)
Water (R.I. 1.333)
Reference Line
Re
fra
ctive
In
de
x (
me
asu
red
)
Refractive Index (literature)
Figure 4.11: Experiment results of RPD sensor for media of known refractive indices.
The procedure employed in the CPD sensor test for biofouling discussed in Chapter 3
was utilized in this test. As shown in Figure 4.12, during the 9-hour experiment three
54
independent sensor outputs were measured in terms of pixel positions with respect to time. The
downward movements of resonance points (the symbols associated with the blue curve above the
inset) are results of biofouling. The corresponding refractive index change was 0.0078, a value
similar to the test results for the CPD sensor, and as was the case in the previous test, the
reference point did not shift significantly.
0 2 4 6 8 10 12 14 16 18
120
140
160
180
200
220
240
260
280
300
320
Time (every 30 min.)
Poin
ts in P
ixel
Measurement 1
Measurement 2
Measurement 3
Resonance Point
Reference PointPo
sitio
n o
f R
eso
na
nce
Poin
t (p
ixel)
Time (every 30 min.)
Figure 4.12: Microscale sensing experimental result of the RPD sensor test for biofouling
formation.
4.7 Discussion
4.7.1 Design Variations
In addition to the design variations for the CPD sensor discussed in the preceding
chapter, the optimal thickness of the additional metal layer requires investigations. Figure 4.13
depicts the simulation results detailing signal response as a result of SPR based on Equation 4.1
and 4.2. As shown in the figure, 47 nm was found to be the theoretical optimal value for the
thickness of the gold layer [15, 16]. Deep peaks ensure clear detection of resonance point, and
the upside-down peaks (dips in reflectance) at the resonance point observed for other thickness
values were shallower. The actual output signals measured in this study showed somewhat
shallow resonance points. The major cause for this problem was found to be thicker
depositions of gold layer induced by a non-uniform deposition process of evaporator used during
fabrication.
55
35 36 37 38 39 40 41 42 43 44 450
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
35 36 37 38 39 40 41 42 43 44 450
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Increasing Au
thickness
47 nm
27 nm
Increasing
Au thickness
47 nm
67 nmR
efle
ctia
nce
Incident Angle (deg.) Figure 4.13: Reflectance variation with respect to the thickness of the gold (Au) layer.
4.7.2 Determination of Sensitivity
The most attractive element of the SPR sensor is its high sensitivity for a wide range of
scientific measurements. Homola et al. conducted notable and practical studies on the
sensitivity of several types of SPR sensors [17, 18]. In their papers, they began with basic
equations, such as Equation 4.1, describing the physical phenomena to mathematically derive
theoretical sensitivities for each case. The theoretically derived sensitivity of prism-coupler
type angular interrogation was determined to be 127 deg. R.I.U.-1
, while other types of SPR
sensing structures such as wavelength interrogation and intensity measurement with prism or
grating coupler-based SPR sensors had relatively lower sensitivities. Under the conditions for
the usage of a goniometer with an angular resolution of 4101 deg. [7], the resolution was
calculated to be 7108 R.I.U. as expressed in following equation:
sensor
component
tmeasuremenS
CRR (4.3)
where tmeasuremenR is the resolution of the measurement, and componentCR and sensorS are the
resolution of the component such as a goniomter, a wavelength modulator or a detector, and the
sensitivity of the sensor, respectively.
For this study presented in this chapter, the theoretical sensitivity of the sensing
technique was numerically derived by simulations, as discussed in Section 4.3. As mentioned
previously, the sensing device utilized here is a prism-coupler type angular interrogation SPR-
based sensor. The sensitivity acquired by the simulation was 90 deg. R.I.U.-1
, which compares
reasonably well to the mathematically derived value from the Homola‘s work. However, unlike
the Homola‘s calculation, since angle scanning is replaced by Fraunhofer diffraction, there is no
angular resolution for this work. Therefore the resolution of the measurement instrument is the
only factor which determines the resolution of the sensing system. Therefore, a resolution of
56
the detector is required to be considered in the resolution calculation as it is able to deliver the
best performance to the system. One example of detector resolution is setting a CCD of 5 um-
pixel size placed 5cm apart from the sensor, and this gives approximately 4101 deg. of angular
resolution, which is equivalent to Homola‘s calculations. For this simulation model, the
resolution was calculated to be 6101 R.I.U using Equation 4.3. As expected, the sensitivity
of SPR-based model is comparable to the conventional prism-coupler type angular interrogation
SPR sensor, which has the highest value reported, and it is higher than that of the CPD models
discussed in Section 2.3. Table 2 summarizes the values for the sensitivity and resolution
discussed in this section.
Table 2: Summary of sensitivity study for various sensing techniques.
Sensitivity/Component
Resolution ([] RIU-1) Resolution (RIU) Reference
SPR-based sensor
Angular interrogation
Prism-coupler type
1.27×106 8×10-7
[7, 17]
SPR-based sensor
Wavelength interrogation
Prism-coupler type
4.85×105 2×10-6 [17]
SPR-based sensor
Intensity measurement
Prism-coupler type
5.75×104 2×10-5 [17, 19]
SPR-based sensor
Angular interrogation
Grating-coupler type
4.03×105 2×10-6 [7, 17]
SPR-based sensor
Waveguide type 2×104 5×10-5 [17, 20]
RPD sensor (SPR-based)
Angular interrogation Prism-coupler type
9.0×105 1×10-6
Section 4.3
CPD sensor
Angular interrogation
Prism-coupler type
3.6×105 3×10-6 Section 2.3
4.8 Summary
A prism-coupler type angular interrogation sensor was selected to demonstrate the SPR
devices and modified to remove the bulky rotating light source. This work has led to
significant benefits in size reduction to facilitate possible multi-point measurements. In order to
verify the feasibility and to calculate the theoretical sensitivity of the sensor, a numerical
simulation was conducted. The result indicates that the sensor developed in this study has
sensitivity of 90 deg. R.I.U.-1
. The body of this sensor is based on the CPD sensor in chapter 3
57
and the only modification is the deposition of a thin metal layer on the sensing edge at the
interface between the sensor and the analyte. Once the prototype sensor was fabricated,
experiments were conducted with a setup very similar to the one used previously for the
prototype CPD sensor. The experimental results were collected through image processing and
the use of a polarizer. These results were similar to those of the CPD sensing technique.
58
References
[1] R. W. Wood, ―On a remarkable case of uneven distribution of light in a diffraction
grating spectrum,‖ Philosophical Magazine, vol. 4, pp. 396-402, 1902.
[2] A. Otto, ―Excitation of nonradiative surface plasma waves in silver by the method
of frustrated total reflection,‖ Zeitschrift fur Physik A Hadrons and Nuclei, vol.
216, pp. 398-410, 1968.
[3] E. Kretschmann and H. Raether, ―Radiative decay of nonradiative surface plasmons
excited by light,‖ Zeitschrift Fuer Naturforschung, vol. 23A, pp. 2135-2136, 1968.
[4] I. Pockrand, J. D. Swalen, J. G. Gordon II, and M. R. Philpott, ―Surface plasmon
spectroscopy of organic monolayer assemblies,‖ Surface Science, vol. 74, pp. 237-
244, 1978.
[5] J. G. Gordon II and S. Ernst, ―Surface plasmons as a probe of the electrochemical
interface,‖ Surface Science, vol. 101, pp. 499-506, 1980.
[6] C. Nylander, B. Liedberg, T. Lind, ―Gas detection by means of surface plasmons
resonance,‖ Sensors and Actuators A, vol. 3, pp. 79-88, 1982.
[7] B. Liedberg, C. Nylander, and I. Lunstrom, ―Biosensing with surface plasmon
resonance-how it all started,‖ Biosensors and Bioelectronics, vol. 10, pp. i-ix,1995.
[8] B. Liedberg, C. Nylander, and I. Lunstrom, ―Surface plasmon resonance for gas
detection and biosensing,‖ Sensors and Actuators A, vol. 4, pp. 229-304, 1983.
[9] Biacore website, www.biacore.com.
[10] H. -P. Chiang, Y. -C. Wang, P. T. Leung, and W. S. Tse, ―A theoretical model for the
temperature-dependent sensitivity of the optical sensor based on surface plasmon
resonance,‖ Optics Communications, vol. 188, pp. 283-289, 2001.
[11] H. L. Lemberg, ―Surface plasmons in liquid mercury: Propagation in a nonuniform
transition layer,‖ Physical Review B, vol. 10, pp. 4079-4099, 1974.
[12] A. Ramanavieius, F. W. Herberg, S. Hutschenreiter, B. Zimmermann, I. Lapenaite,
A. Kausaite, A. Finkelsteinas, and A. Ramanavieiene, ―Biomedical application of
surface plasmon resonance biosensors (review),‖ Acta Medica Lituanica, vol. 12,
pp. 1-9, 2005.
[13] P. Pattnaik, ―Surface Plasmon Resonance,‖ Applied Biochemistry and
Biotechnology, vol. 126, pp. 79 – 92, 2005.
[14] J. Homola, ―Present and future of surface plasmon resonance biosensors, Analytical
and Bioanalytical Chemistry,‖ vol. 377, pp. 528.539, 2003.
[15] B. H. Ong, X. Yuan, S. C. Tjuin, J. Zhang, and H. M. Ng, "Optimized film
thickness for maximum field enhancement of a bimetallic surface plasmon
resonance biosensor," Sensors and Actuators B, vol. 114, pp. 1028-1034, 2006.
[16] P. Lecaruyer, M. Canva, and J. Rolland, "Metallic film optimization in a surface
59
plasmon resonance biosensor by the extended Rouard method," Applied Optics,
vol. 46, pp. 2361-2369, 2007.
[17] J. Homola, S. S. Yee, G. Gauglitz, ―Surface plasmon resonance sensors: review,‖
Sensors and Actuators B, vol. 54, pp. 3-15, 1999.
[18] J. Homola, I. Koudelab, and S. S. Yee, ―Surface plasmon resonance sensors based
on diffraction gratings and prism couplers: sensitivity comparison,‖ Sensors and
Actuators B, vol. 54, pp. 16-24, 1999.
[19] C. Ronot-Trioli, A. Trouillet, C. Veillas, and H. Gagnaire, ―Monochromatic
excitation of surface plasmon resonance in an optical-fibre refractive-index sensor,‖
Sensors and Actuators A, vol. 54, pp. 589–593, 1996.
[20] A. Trouillet, C. Ronot-Trioli, C. Veillas, and H. Gagnaire, Chemical sensing by
surface plasmon resonance in a multimode optical fibre, Pure and Applied Optics,
vol. 5, pp. 227–237, 1996.
60
Chapter 5
Conclusions and Future Work
5.1 Conclusions
The feasibility of prism-coupler type angular interrogation sensing using a planar,
integrated microscale sensor with a built-in waveguide has been investigated with two kinds of
sensing mechanisms, the CPD and RPD sensors. The feasibility of the concept of this sensing
technique was verified through macroscale optical experiments as discussed in Chapter 2. The
microscale sensors, on the other hand, have been designed to be as small as a few hundred of
micrometers, and these small structures are based on the utilization of Fraunhofer diffraction
which is observed at the end of the built-in waveguide. This diffraction is appropriately
positioned and firmly aligned within the sensor, and the built-in waveguide successfully
generates the convolution pattern of the combined diffraction and reflection at the detector
obviating the necessity of a scanner and, therefore, eliminating the time required for scanning.
As explained in Chapter 3, the prism-coupler type of angular interrogation sensing was
applied to a micro-sensor, classified as a CPD sensor. This CPD sensor has been designed and
successfully created through a simple two-mask fabrication process. The results of optical
simulations demonstrate the conceptual working principle of the sensing technique, and provide
a theoretical sensitivity comparable to that of other types of optical sensing methods.
Successful measurement has been conducted, and the results have been substituted into the
governing equations to determine the refractive index of the analytes. During the
measurements of several liquids, the error range has been evaluated at less than ± 0.002 R.I.U.
Although the error range of this prototype sensor is quite large compared to the theoretical
sensitivity, the measurement results are in good agreement with the reported values. A testing
liquid (milk) has been employed to monitor biofouling as a medium of changing refractive index.
The change measured was as much as 0.0089 R.I.U. during a 9-hour experimental duration.
This dielectric optical sensor could find applications in monitoring systems which measure
biofouling under conditions of erosive wearing.
61
As has been established in Chapter 4, the prism-coupler type of angular interrogation
sensing method can be applied to a dielectric micro-sensor with a metal layer on top, classified
as an RPD sensor. On the basis of the CPD sensor structure previously studied, an RPD sensor
was designed and fabricated. Optical simulations were conducted prior to the actual
experiments in order to predict the optical responses of the output signals (the pattern of the
convoluted light) as a reference as well as to verify the feasibility of this sensing scheme. The
theoretical sensitivity of this proposed RPD sensing technique has been evaluated, and the
sensitivity of the technique has been determined to be as good as that of the state of the art
sensing method (the goniometer-driven prism-coupler type angular interrogation sensing
method). Several common liquids with different refractive indices have been tested with the
fabricated RPD sensor, and the refractive indices measured are in agreement with the reported
values within an error range of less than ± 0.002 R.I.U. For biofouling formation
characterization, the average refractive index change has been measured as 0.0078 R.I.U. during
a 9-hour experiment. It has been determined that the error range of the sensor is similar to that
of a CPD sensor. However, because this error is mainly the result of a low quality of
manufacturing process, the problem could be solved by refining the sensor-fabrication technique.
In summary, the sensitivity of the RPD sensor could as good as the bulky conventional SPR
sensor, while additional advantages are introduced. The significant reduction of device size and
the elimination of scanning time by this RPD sensor technology could lead to a wide spectrum of
applications, especially in multi-point sensing and mobile sensing devices.
5.2 Future Directions
Throughout this research, the theoretical bases and feasibility of microscale prism-coupler
type angular interrogation sensing techniques utilizing simple prototype sensors have been
examined in order to provide preliminary foundations for future study. In this section, a variety
of advanced applications and the methods of their fabrication as well as advanced sensing
techniques are discussed.
5.2.1 Self Cleaning Sensor using UV and F-IR Ray
Ultraviolet radiation (UV) is known to kill micro-organisms including bacteria [1].
Since the majority of biofouling processes are the results of microorganisms, UV rays can be
used to eliminate the chief cause of this problem [2]. Likewise, the thermal energy of Far-
Infrared (F-IR) radiation could breakdown fouling [3, 4]. When these two forms of radiation
are applied consecutively in a localized area, a significant reduction in the biofouling within that
area can be achieved. Using the UV and F-IR rays in the CPD sensor, the accumulated
biofouling on the surface of the sensor could be removed; hence the sensor could be reset to its
initial condition without the use of removal agents which are often very difficult to apply non-
intrusively. As illustrated in Figure 5.1, the provision of those rays from the exit edge of the
prism will avoid the intensity reduction caused by the waveguide on the input edge of the prism.
62
Breakdown of Biofouling
Figure 5.1: Schematic illustrating a strategy to remove biofilm deposition from a sensor using
ultraviolet and far-infrared rays as a means of cleaning.
5.2.2 Integration of Laser Diode and Photo Detector
As discussed in the earlier chapters of this study, the construction of the prototype sensor
was simplified; therefore both the light source and the detector were excluded from integration
into the device during fabrication. To construct complete sensor device unit capable of working
independently, a laser diode [5, 6, 7] and photo-detector [8, 9] must be integrated into the sensor
chip along with the prism-coupler and built-in waveguide. This complete sensor device is
expected to be used directly for multi-point detection for characterization of inhomogeneous
materials in large-scale plants.
The integration of an energy harvester and RF transmitter into a sensor is also possible
for a stand-alone, self-powered device. A possible example is a prism-coupler type angular
interrogation sensor driven by a piezoelectric generator, which emits wireless signals used to
characterize an analyte [10]. For example, in water systems, small water-powered sensors
embedded at multiple points may be employed for monitoring the system (Figure 5.2).
Piezoelectric
Generator Rod
Figure 5.2: Schematic of a self-powered sensing device.
63
5.2.3 Measurement of phase difference between s- and p-polarizations
Along with measurement of light intensity, another method for determining the
refractive index of an analyte is offered by the measurement of the phase difference, also induced
by total internal reflection, between s and p polarizations [11]. According to the Fresnel
equation, reflectance, in fact, incorporates two components: the reflectance of the light polarized
with respect to the direction of propagation of electric field (TE, s-polarized); and the reflectance
of light polarized with respect to the magnetic field (TM, p-polarized). The reflectance given in
terms of these discrete values is usually denoted as Rs and Rp, respectively, and given by [12]:
2
2
2
121
2
2
121
sin1cos
sin1cos
ii
ii
s
n
nnn
n
nnn
R
, and
2
2
2
2
11
2
2
2
11
cossin1
cossin1
ii
ii
p
nn
nn
nn
nn
R
(5.1)
To measure the refractive index of an analyte, the phase difference between the s and p-
polarizations of the reflected light is measured by using light sources of differing polarization
states. In such a case, the refractive index can be obtained by substituting the measurement
values into the Fresnel equation. For the sensor used in this study, the comparison of two
output signals obtained at different polarization states can provide information to calculate the
change in refractive index for an analyte.
20 30 40 50 600
0.2
0.4
0.6
0.8
1.0s-polarized
p-polarized
Re
fle
cta
nce
Incident Angle [deg.] Figure 5.3: Response of reflectance in accordance with incident angle for s- and p-polarized light.
64
5.2.4 Measurement in Total Internal Reflection Region
The cloudiness of suspended particles in a liquid is referred to as turbidity and is
typically measured through the absorption and scattering of light by the suspended solids. This
sensing method has been traditionally used as a means of monitoring biofilm development [13,
14]. However, the technique is accurate only when the particles suspended in the liquid
medium are homogeneously distributed — an event which has limited the ability of this sensing
method to achieve reliable measurements.
Despite its unsuitableness for accurate measurements of biofouling, the turbidity
measurement is very simple and efficient. The region in the total internal reflection on the
reflectance profile does not change until the critical point appears. In other words, the region
after critical point conserves the original pattern of its diffracted light. Therefore, if there is any
change in the region, it must be due to a change in turbidity between measurements. This
turbidity measurement could be used as a secondary measuring tool for more precise
characterization of analytes.
5.2.5 Digitated Sensor Geometry
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.2
0.4
0.6
0.8
1
Rela
tive I
nte
nsity
Signal Pattern at Shift of 0.04
Relative Angle from Zero Order of Diffraction Pattern
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.2
0.4
0.6
0.8
1
Rela
tive I
nte
nsity
Signal Pattern at Shift of 0.02
Relative Angle from Zero Order of Diffraction Pattern
-25 -20 -15 -10 -5 0 5 10 15 20 250
0.2
0.4
0.6
0.8
1
Rela
tive I
nte
nsity
Signal Pattern at Shift of 0
Relative Angle from Zero Order of Diffraction Pattern
Detector 1
Detector 2
Detector 3
Figure 5.4: Schematic of Three Exit Waveguide sensors and their corresponding points of
measurement on a convoluted light profile.
As the output signal exits the prism, there is a great possibility that the signal will
become weaker or distorted by the external environment between the detector and the exit edge
of the prism. The solution to this problem is to build exit waveguides. As shown in Figure
5.4, three separate exit waveguides are illustrated. Each guide picks up the signals
corresponding to a fixed reflectance angle and only three points appear on the graph as an
example. The exit waveguide encases the output signal up to the detector to prevent weakening
or distortion of the signal [15]. Because only guided signals reach the photo-detector, there
should be less noises resulting from light scattered at the prism‘s exit edge or from outside light.
65
Furthermore, there is a good chance that the scattered light at the reflection edge inside the prism
will not fall within the insertion range of the exit waveguides. Therefore, the waveguide itself
could have a filtering capacity. As discussed earlier in this section, it is possible to integrate the
laser diode and photo-detector into the design and it would be straightforward to add the
additional waveguides.
5.2.6 Thickness measurement of SPR-based sensor
During the discussion of the working principle in Section 4.4 (Equation 4.2), only the
case of an analyte or biofilm of thickness comparable to or thicker than the depth of the SPW
field was considered. However, if the thickness of medium to be sensed is much thinner than
the SPW field depth, the propagation constant is expressed as follows:
ndknn
m
dr
Re
2Re
2
(5.2)
where rn is the refractive index of the reference material [16]. This suggests that it may be
possible to discover a way to measure a very small thickness change in a biofouling formation at
the initiation stage of tens of nanometers if the refractive index of the biofilm is provided by
another measuring device such as, for example, a CPD sensor [17].
5.2.7 Sensor using Fresnel diffraction
Theoretically, the smallest dimension of the sensor design used in this study is about 300
μm, the minimum scale at which Fraunhoffer diffraction becomes dominant. If the distance of
light travel is less than 300 μm for this setup, Fresnel diffraction is more significant than
Fraunhoffer diffraction. In such a case, the sensor design must be modified. In the event of
light waves incident upon a diffraction grating, the resultant image as illustrated in Figure 5.5b
(Talbot image) can be acquired when the detector is set at the proper distance from the grating
plane [18, 19]. Figure 5.5a shows the conceptual design of the sensor using Fresnel diffraction
with all aforementioned factors considered.
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
x 10-6
0
0.5
1
1.5
2
2.5
(a) (b)Multi-Angled
Reflection Edge
Figure 5.5: (a) Schematic of conceptual design of sensor using Fresnel diffraction, (b)
corresponding intensity profile.
66
In the figure, the prism has edges of multiple angles of decline, and each of the angled
edges corresponds to an individual slit of the diffraction grating. Each ray propagating through
the individual slits is reflected by the angled edge of prism, and the intensity of this reflected
light will indicate the position of the critical point. The governing equation for this principle is
derived as follows:
2
2
121
2
2
121
22
20
sin1cos
sin1cos2
cos2
coscos214
1
,
rr
rr
d
n
nnn
n
nnn
L
xm
L
x
L
zmI
RzxII
(5.3)
where the factors in first bracket specify the influence of Fresnel diffraction, and the values in the
second bracket refer to the influence of the reflected light.
67
References
[1] C. E. Bayliss and W. M. Wailtes, "The Effect of Hydrogen Peroxide and Ultraviolet
Irradiation on Non-sporing Bacteria," Journal of Applied Bacteriology, vol. 48, pp.
417-422, 1980.
[2] J. S. Patil, H. Kimoto, T. Kimoto, and T. Saino, "Ultraviolet radiation (UV-C): a
potential tool for the control of biofouling on marine optical instruments,"
Biofouling, vol.23, pp. 215-230, 2007.
[3] J. Dobson and M. Wilson, ―Sensitization of oral bacteria in biofilms to killing by
light from a low-power laser,‖ Archives of Oral Biology, vol. 37, pp. 883-887,
1992.
[4] N. Sakai and T. Hanzawa, "Application and advances in far-infrared heating in
Japan," Trends in Food Science & Technology, vol.5, pp. 357-362, 1994.
[5] T. Onishi, K. Inoue, K. Onozawa, T. Takayama, and M. Yuri, "High-Power and
High-Temperature Operation of Mg-Doped AlGaInP-Based Red Laser Diodes,"
IEEE Journal of Quantum Electronics, vol. 40, pp. 1634, 2004.
[6] P. Schlotter, J. Baur, C. Hielscher, M. Kunzer, H. Obloh, R. Schmidt, and J.
Schneider, "Fabrication and characterization of GaN/InGaN/AlGaN double
heterostructure LEDs and their application in luminescence conversion LEDs,"
Materials Science and Engineering B, vol. 59, pp. 390-394, 1999.
[7] M. A. Haase, J. Qui, J. M. De Puydt, and H. Cheng, "Blue?green laser diodes,"
Applied Physics Letter, vol. 59, pp. 1272-1274 , 1991.
[8] W. K. Chan, A. Yi-Yan, T. J. Gmitter, L. T. Florez, J. L. Jackel, D. M. Hwang, E.
Yablonovitch, R. Bhat, J. P. Harbison, "Optical coupling of GaAs photodetectors
integrated with lithium niobate waveguides," vol. 2, pp. 194-196, 1990.
[9] E Miyazaki, S Itami, and T Araki, "Using a light-emitting diode as a high-speed,
wavelength selective photodetector," Review of Scientific Instruments, vol. 69, pp.
3751, 1998.
[10] N. G. Elvin, N. Lajnef, and A. A. Elvin, ―Feasibility of structural monitoring with
vibration powered sensors,‖ Smart Materials and Structures, vol. 15, pp. 977-986,
2006.
[11] M. –H. Chiu, J. –Y. Lee, and D. –C. Su, "Refractive-index measurement based on
the effects of total internal reflection and the uses of heterodyne interferometry,"
Applied Optics, vol. 36, pp. 2936-2939, 1997.
[12] E. Hecht, Optics, 4th ed., Reading, Addison-Wesley, 1979.
[13] J. Klahre, H. C. Flemming, ―Monitoring of biofouling in papermill process waters,‖
Water Research, vol. 34, pp. 3657-3665 ,2000.
[14] K. Carlson, S. Cho, and K. Stutzman, Indirect Detection of Intentional Chemical
Contamination in the Distribution System Using Low Cost Turbidity Sensors,‖
68
Water Quality Technology, 2006.
[15] Y. Sugimoto, N. Ikeda, N. Carlsson, K. Asakawa, N. Kawai, and K. Inoue,
"Experimental verification of guided modes in 60°-bent defect waveguides in
AlGaAs-based air-bridge-type two-dimensional photonic crystal slabs," Journal of
Applied Physics, vol. 91, pp. 3477, 2002.
[16] J. Homola, ―Present and future of surface plasmon resonance biosensors, Analytical
and Bioanalytical Chemistry,‖ vol. 377, pp. 528.539, 2003.
[17] H. Kano, W. Knoll, "Locally excited surface-plasmon-polaritons for thickness
measurement of LBK films," Optics Communications, vol. 153, pp. 235-239, 1998.
[18] H. F. Talbot, "Facts relating to optical science No. IV,‖ Philosophical Magazine,
vol. 9, pp. 401-407, 1836.
[19] K. D. Mielenz, "Algorithms for Fresnel Diffraction at Rectangular and Circular
Apertures," Journal of Research of the National Institute of Standards and
Technology, vol. 103, pp. 497-509, 1998.