xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 001709545
006 m d s
007 cr mnu uuuuu
008 060522s2005 flua sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001329
Reliability of liquid core optical waveguides for sensitive optical absorption measurements of trace species in water
h [electronic resource] /
by Avishekh Pal.
[Tampa, Fla.] :
b University of South Florida,
Thesis (M.S.)--University of South Florida, 2005.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 121 pages.
ABSTRACT: Long path optical waveguides can be used in optical absorption measurements to increase the optical path length and, thus, the overall absorption of a sample. Recently, 1m long coiled Liquid Waveguide Capillary Cells (LWCC) have been used by analytical spectroscopists to measure the absorption strength of weakly absorbing liquids. However, most of these measurements have used conventional light sources such as Xenon or Halogen lamps and not spectroscopic laser sources. In this thesis study, we used a LWCC absorption waveguide and a laser light source to measure, for the first time to our knowledge, the optical transmission through several water or liquid samples. It was found upon using the LWCC waveguide, the coherent laser light source tended to produce larger variability (>15%) in the measurements of transmission readings than that for a conventional absorption cell or a conventional light source.This was especially evident when the LWCC waveguide was chemically cleaned with an acid and a base solution between each sample run as directed by the manufacturer. The non-coherent optical sources, Halogen lamp and Xenon arc lamp, produced more stable (3%) transmission measurements. Finally, using a Helium Neon laser scattered off a diffuse reflecting surface was found to produce moderate variability (7%), but this was much less than the coherent Helium Neon laser alone. It was concluded that the use of the coherent source was more susceptible than the non-coherent source to small changes in the reflectivity or index of refraction along the wall of the coiled LWCC waveguide. Our results are consistent with recent work by Barwicz and Haus, and by Lytle and Splawn who saw a large dependence of the transmission through a hollow straight waveguide upon changes in the polarization and input angle of the laser beam directed into the waveguide.
Adviser: Dennis Killinger.
t USF Electronic Theses and Dissertations.
Reliability of Liquid Core Optical Waveguides for Sensitive Optical Absorption Measurements of Trace Species in water by Avishekh Pal A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Physics College of Arts and Sciences University of South Florida Major Professor: Dennis Killinger, Ph.D. Sarath Witanachchi,Ph.D. Myung K. Kim Ph.D. Wei Chen, Ph.D. Chun Min Lo, Ph.D. Date of Approval: July 27, 2005 Keywords: Laser absorption,spectroscopy, optical waveguide Copyright 2005 Avishekh Pal
ACKNOWLEDGEMENTS I would like to express my deepest gr atitude to Dr. Killinger for being an invaluable advisor and his support, guidance and patience during this project. I would also like to express my thanks to my committee members Dr. Wei Chen, Dr. M. K. Kim, Dr. S. Witanachi and Dr. C. M. Lo for serving on my dissertation committee. I would like to thank Edward Navarro and Anali Makoui for their friendship and help. I would like to thank the staff of the Physics Department for their support: Sam Valenti, Phil Bergeron, Evelyn Keeton-William and especially Sue Wolfe. Finally I want to express my heartfelt appreciation for my parents especially Minakshi Pal, my mother for always believing in me and enabling me to reach this stage of my life. Thank you.
i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES v ABSTRACT xiii CHAPTER 1. INTRODUCTION 1 CHAPTER 2. THEORY OF WAVEGUIDES. 5 2.1 Beer Lambert Law 5 CHAPTER 3. BASIC THEORY OF RECTANGULAR WAVEGUIDES. 9 3.1 Waveguide Mode 9 3.2 Optical Propagation Constants 14 3.3 Group Velocities 15 CHAPTER 4. UNDERSTANDING TRANSMISSION OF LIGHT THROUGH CYLINDRICAL WAVEGUIDE 16 4.1 Optical Ray Transmission 16 4.2 Guided Waves 22 4.3 The V Parameter 28 4.4 Attenuation and Dispersion 28 4.5 Combined Material and Waveguide Dispersion 31 4.6 The GoosHanchen Shift 32
ii CHAPTER 5. LIQUID CORE CAPILLARY WAVEGUIDES 35 5.1 Introduction 35 5.2 Commercial LWCC from World Precision Instruments, Inc 38 5.3 Waveguide Properties 42 5.4 Manufacturers cleaning procedure 43 CHAPTER 6. PRELIMINARY ABSORPTION MEASUREMENTS OF SELECTED LIQUIDS USING 266nm AND 355nm LASER SOURCES AND 1cm ABSORPTION CELL 45 6.1 Microchip Laser Source 49 6.2 Absorption measurements using a 1cm Quartz cuvette 6.3 Double pass absorption measurements 53 CHAPTER 7. ABSORPTION MEASUREMENT OF SELECTED LIQUIDS USING 355nm LASER SOURCE AND 1m LWCC 58 CHAPTER 8. VARIABILITY OF LASER TRANSMISSION MEASUREMENTSUSING THE LWCC WAVEGUIDE 63 8.1 Experimental set up 63 CHAPTER 9. VARIABILITY OF 1m LWCC FOR NONCOHERENT LIGHT SOURCES 74 9.1 Experimental set up 74 9.2 Measurements with Xenon Arc light source 77 9.3 Measurements with Xenon Arc light source 82 CHAPTER 10. VARIABILITY OF SCATTERED LASER SOURCE USING THE 1m LWCC WAVEGUID 87 10.1 Transmission Experiment with Scattered HeNe laser 87
iii CHAPTER 11. COMPARISON OF VARIATION IN TRANSMITTED LWCC WAVEGUIDE VALUES AND POSSIBLE LINK 93 11.1 Comparison With Recent Studies 93 11.2 Additional LWCC Measurements Using Focused HeNe Laser Into Fiber 97 11.3 General Conclusions 97 CHAPTER 12. CONCLUSION AND FUTURE WORK 102 REFERENCES 104
iv LIST OF TABLES Table 5.1 Specifications of Type I and Type II LWCC waveguide. 41 Table 8.1 Specifications of the Ocean Optics ST 2000 spectromete r 66
LIST OF FIGURES Figure 1.1 Attenuation coefficient of water (adapted from Tyler and Presisendorfer, 1962). 2 Figure 2.1. Schematic of apparatus to measure optical absorption and transmission. 6 Figure 2.2 Optical waveguides: (a) slab geometry;(b) strip geometry;(c) fiber or capillary geometry. 8 Figure 3.1 Optical plane waveguide. 10 Figure 3.2 Condition of self-consistency: as the wave reflects twice it duplicates itself 12 Figure 3.3 At angles for which self-consistency is satisfied, the two waves interfere and create a pattern that does not change with z. 12 Figure 4.1 Meridional ray entering waveguide and being guided by total internal reflection where n 2 < n 1 17 Figure 4.2 Skew rays in core of fiber. 17 Figure 4.3 End on view of skew ray. 17 Figure 4.4 Cylindrical dielectric waveguide. 21 Figure 4.5 Cylindrical coordinate sysytem for waveguide 23 Figure 4.6 Examples of the radial distribution u(r) given by equation (4.7) for (a) 1=0and (b) I = 3. The shaded areas represent the fiber core and the unshaded areas the cladding. The parameters k T and and v
the two proportionality constants in equation (4.7) have been selected such that u(r) is continuous and has a continuous derivative at r = a. Larger values of k T and lead to a greater number of oscillations in u(r). 27 Figure 4.7 The GoosHanchen shift. When n 1 > n 2 33 Figure 5.1 Cross sectional view of light transmission through the liquid capillary waveguide cell. The refractive index of seawater quartz glass and the amorphous cladding are shown as n 1 n 2 37 Figure 5.2 Typical LWCC waveguide experimental setup 39 Figure 5.3 Photo of LWCC waveguide. 40 Figure 6.1 Schematic of optical absorption experiment using 1cm quartz absorption cell. 45 Figure 6.2 Photograph of optical absorption experiments using 1cm quartz absorption cell. 46 Figure 6.3 Schematic of the 4 Nd:YAG Microchip laser operating at 266 nm (JDS Uniphase, Model NU-10110-100). 48 Figure 6.4. Transmission as a function of concentration plot for diluted coffee through a 1cm quartz cell 355nm laser source. 50 Figure 6.5. Optical transmission as a function of concentration for diluted coffee through a 1cm quartz cell and 266nm laser source. 50 vi
vii Figure 6.6 Transmission as a function of concentration plot for diluted cola through a 1cm quartz cell using 355nm laser source. 51 Figure 6.7 Optical transmission as a function of concentrated for diluted cola through a 1cm quartz cell and 266nm laser source. 51 Figure 6.8 Transmission as a function of concentration plot for diluted orange juice through a 1cm quartz cell using 266nm laser source. 52 Figure 6.9 Optical transmission as a function of concentrated for diluted orange juice through a 1cm quartz cell and 266nm laser source. 52 Figure 6.10 Absorption coefficient as a function of concentration for diluted coffee using 355nm laser source in 1cm cell. 54 Figure 6.11 Absorption coefficient as a function of concentration for diluted coffee using 266nm laser source in 1cm cell. 54 Figure 6.12 Absorption coefficient as a function of concentration for diluted cola using 355nm laser source in 1cm cell. 55 Figure 6.13 Absorption coefficient as a function of concentration for diluted cola using 266nm laser source in 1cm cell. 55 Figure 6.14 Absorption coefficient as a function of concentration for diluted orange juice using 355nm laser source in 1cm cell. 56
viii Figure 6.15 Absorption coefficient as a function of concentration for diluted orange juice using 266nm laser source in 1cm cell. 56 Figure 6.16 Absorption coefficient as a function of concentration for diluted coffee using 266nm laser source in 1cm cell. 57 Figure 7.1 Schematic diagram of optical absorption experiment using the 1m optical liquid core waveguide. 59 Figure 7.2 Absorption spectrum of coffee at different concentration through the LWCC waveguide by using 355nm laser. Maximum concentration was Already diluted by a factor of twenty compared to data given previously in Figure 6.11. 60 Figure 7.3 Transmission of 355nm laser through the LWCC waveguide USF fountain water, USF tap water and USF distilled water through the waveguide over a period of time. 62 Figure 8.1 Schematic of experimental setup of transmission system with liquid core capillary waveguide. 64 Figure 8.2 (a) Spectrometer detectors sensitivity as function of (b) Grating efficiency as a function of wavelength (channel #1 and channel #2).(Ocean Opticss ST 2000 Spectrometer Operating Manual). 65 Figure 8.3. Typical spectrum for 266nm laser source. 67
ix Figure 8.4. Input intensity of 266nm laser source over an hour when the LWCC waveguide was not cleaned. 69 Figure 8.5. Output intensity of 266nm laser source through LWCC filled with distilled water (Zephyrhills brand) over an hour when the waveguide was not cleaned. 6 9 Figure 8.6. Incident intensity spectrum of 266nm laser source for different trials when the waveguide was cleaned. 70 Figure 8.7 Transmitted intensity spectrum of 266nm laser source through the LWCC waveguide filled with distilled water for different trials when it was chemically cleaned. 70 Figure 8.8. Input intensity of Helium Neon laser source over a period when the LWCC was not cleaned. 72 Figure 8.9. Output intensity of Helium Neon laser source through LWCC Waveguide filled with distilled water (Zephyrhills brand) over an hour when the waveguide was not cleaned. 72 Figure 8.10. Incident intensity of Helium Neon laser source for different trials when the waveguide was chemically cleaned. 73 Figure 8.11. Output intensity of Helium Neon laser source through the LWCC, waveguide which is filled with distilled water for different trials when the waveguide was chemically cleaned. 73 Figure 9.1 Measured emission spectrum of halogen light source. 75 Figure 9.2 Measured emission spectrum of Xenon Arc light source. 76 Figure 9.3 Output spectrum of Xenon Arc lamp as a function a function of
x time over a period of time. 78 Figure 9.4 Transmitted spectrum of distilled water through the LWCC waveguide as a function of time using a Xenon Arc lamp. 79 Figure 9.5 Transmitted spectrum of Xenon Arc lamp as a function of different sample runs through the LWCC waveguide(filled with distilled water, Zephyrhills brand) when it was cleaned between each sample runs. 80 Figure 9.6. Transmission spectrum of distilled water through the LWCC waveguide using Xenon Arc lamp as the light source for Eight different sample trials when the waveguide was chemically cleaned. 81 Figure 9.7 Output spectrum of Halogen light source as a function of time. 83 Figure 9.8 Transmission spectrum of distilled water through the LWCC waveguide as a function of time using a Halogen light source. 84 Figure 9.9 Transmission spectrum of distilled water through the waveguide using Halogen li ght source for different trails when the waveguide was chemically cleaned between runs. 85 Figure 9.10. Relative transmission spectrum of distilled water through the LWCC wave guide using Halogen light source for different runs when the LWCC wavguide was chemically cleaned. 86
xi Figure 10.1 Schematic of experimental setup of transmission system with liquid core capillary waveguide for scattered Helium Neon laser as light source. 88 Figure 10.2 Emission spectrum of scattered Helium Neon laser as function of time. 89 Figure 10.3. Transmitted emission spectrum as a function of time of scattered HeNe laser through the LWCC waveguide that was filled with distilled water ( Zephyhills brand). 89 Figure 10.4. Emission spectrum of scattered Helium Neon laser source for different sample runs. 91 Figure 10.5. Transmitted intensity of scattered Heluim Neon laser through the LWCC waveguide when the LWCC waveguide was chemically cleaned between each sample run. 91 Figure 10.6 Relative transmission ratio of scattered Helium Neon laser through the LWCC waveguide when the LWCC was cleaned between each sample run. 92 Figure 11.1 Transmission spectrum of various light sources through the LWCC waveguide for various trial runs during which the LWCC waveguide was chemically cleaned between each run. 94 Figure 11.2 (a) perspective view and (b) top view of a rough rectangular waveguide The top and bottom walls are considered smooth and only the sidewalls are believed rough. A uniform surrounds the waveguide-core cladding of refractive index n clad (c) Spherical coordinate system used. (from Barwicz, 2005). 95
xii Figure 11.3 Scattering losses in dB/cm normalized to the roughness variance in nm 2 for (a) the TE like mode and (b) the TM like mode of Si waveguides.(from Barwicz, 2005) 98 Figure 11.4 Propagation losses when a 488nm laser is focused to a 20 m radius at the entrance to a 100 m x 100m square waveguide. The lines represent various insertion angles between the Gaussian beam and the waveguide axis.(from Lytle,2003) 99 Figure 11.5 Measured transmission values for the focused He-Ne (into fiber and LWCC) and LWCC cleaned between runs. 100
xiii Reliability Of Liquid Core Optical Waveguides For Sensitive Optical Absorption Measurements Of Trace Species In Water Avishekh Pal ABSTRACT Long path optical waveguides can be used in optical absorption measurements to increase the optical path length and, thus, th e overall absorption of a sample. Recently, 1m long coiled Liquid Waveguide Capillary Cells (LWCC) have been used by analytical spectroscopists to measure the absorption strength of weakly absorbing liquids. However, most of these measurements have used conventional light sources such as Xenon or Halogen lamps and not spectroscopic laser sources. In this thesis study, we used a LWCC absorption waveguide and a laser light source to measure, for the first time to our knowledge, the optical transmission through several water or liquid samples. It was found upon using the LWCC waveguide, the coherent laser light source tended to produce larger variability (> 15%) in the measurements of transmission readings than that for a conventional absorption cell or a conventional light source. This was especi ally evident when the LWCC waveguide was
xiv chemically cleaned with an acid and a base solution between each sample run as directed by the manufacturer. The non-coherent opti cal sources, Halogen lamp and Xenon arc lamp, produced more stable (3%) transmission measurements. Finally, using a Helium Neon laser scattered off a diffuse reflecting surface was found to produce moderate variability ( 7%), but this was much less than the coherent Helium Neon laser alone. It was concluded that the use of the coherent source was more susceptible than the noncoherent source to small changes in the refl ectivity or index of refraction along the wall of the coiled LWCC waveguide. Our results ar e consistent with recent work by Barwicz and Haus, and by Lytle and Splawn who sa w a large dependence of the transmission through a hollow straight waveguide upon change s in the polarization and input angle of the laser beam directed into the waveguide.
1 CHAPTER 1. INTRODUCTION Optical absorption is a very well known and tested form of spectroscopy. The absorption of visible and ultraviolet radiati on by different chemical compounds is due to the excitation of electrons in the absorb ing molecule. Usually optical absorption measurements are performed in the infrared spectral region in order to investigate rotational or vibrational-rotational transiti ons and in the UV-visible to study electronic energy levels. In the past, optical absorption has been employed for detecting trace elements in samples 1,2,3,4,5 and to determine real-time water purity analysis. 6,7 Also of importance is the attenuation of different bodi es of water and oceans since this effects optical propagation and transmission of s unlight and underwater laser communication systems. This can be seen in Figure 1.1 that shows typical attenuation values in water for different water samples. As can be seen, differences in water quality and levels of organic matter (often called turbidity) affect the attenuation coefficient. For the absorption spectroscopy measurements to be sensitive we need to make sure the absorption of the sample is substantial. For a highly absorbing material this is not a problem, but for a weakly absorbing samp le one needs to increas e the path length of the absorption cell to increase the overall absorption. To achieve this, various methods have been used including multipassing the li ght beam through an absorption cell by using parallel mirrors, using a long optical fiber, and more recently using hollow waveguides or liquid core capillary waveguides.
Figure 1.1 Attenuation coefficient of water (adapted from Tyler and Presisendorfer, 1962) 2
3 In the work performed in this thesis we measured the absorption of various water samples with a conventional 1 cm quartz abso rption cell and with a 1 m long liquid core capillary waveguide cell that was coiled into a smaller 20 cm diameter package. In addition, we used various light sources like a halogen lamp, Xenon arc lamp, Nd:YAG lasers at 266 nm and 355 nm and a Heliuim Neon laser at 632.8 nm wavelength for the excitation source. Our results were inte resting in that we found upon using the LWCC (Liquid Waveguide Capillary Cell) the cohere nt light sources tended to produce larger variability in the measurements of transmissi on readings than that for the conventional absorption cell. This was especially evid ent when the LWCC waveguide was flushed and cleaned between each sample runs. The non-c oherent sources seemed more stable and thus were more reliable for sensitive measurements. Finally, using a Helium Neon laser scattered off a diffuse reflecting surface al so was found to produce variability, but this was much less than using the direct coherent Helium Neon laser alone. It was concluded that the use of the coherent source was more susceptible than the non-coherent source to small changes in the reflectivity, index of refraction, or angle of propagation along the wall of the LWCC waveguide. These results are important because they show, for the first time to our knowledge, the measurement error using a laser spectroscopic source with the coiled LWCC waveguide. The organization of the thesis is as follows. Chapter 2 describes optical absorption, Beer Lambert Law, and the need for a long pa th optical waveguide. Chapter 3 describes
4 the basic theory of optical waveguide modes. Chapter 4 gives an overview of the theory of transmission of light through cylindrical waveguides. Chapter 5 describes the commercial LWCC waveguide that was used in this thesis and the cleaning procedure used after every measurement. Chapter 6 desc ribes the experimental results obtained for transmission through a conventional 1 cm quart z cell for different water samples using the 266 nm and 355 nm laser source. Chapter 7 describes and explains the experimental setup using the LWCC waveguide and the results obtained for various water samples using the 355 nm laser source. In Chap ter 8 the variability in LWCC waveguide transmission measurements for various water samples using a coherent light sources are given. In Chapter 9 the results for the variability in LWCC waveguide transmission measurements for various water samples using non-coherent light sources are shown. In Chapter 10 the variability in LWCC waveguide transmission measurements for various water samples using a scattered coherent light source is shown. Comparison of the variability of LWCC waveguide with different light sources and possible link to the surface changes or propagation changes are discussed in Chapter 11. Our conclusions and areas of future work are presented in Chapter 12. Finally it should be added that the motivation for this work was the need to measure, with precision, the absorption or a ttenuation of different ocean water samples during research cruises. 8 During previous work, it wa s found that precision absorption measurements would be helpful in interp reting changes in laser-induced fluorescence measurements observed as different bodies of water were probed.
CHAPTER 2. OPTICAL ABSORPTION AND NEED FOR LONG PATH OPTICAL WAVEGUIDE Optical absorption is due to the electronic transitions between the allowed energy levels of a molecule usually involving either the rotational or vibrational or electronic levels. In order to measure the absorption strength or attenuation, an optical absorption setup is often used. Figure 2.1 shows the basic schematic of an absorption measurement setup. It uses a laser as a light source to transmit the light through an absorption cell containing the sample to be measured. A beam splitter and a detector is used before the light enters the cell, to sample the incident beam intensity and the transmitted light beam is then detected by detector number two. The transmission of the sample is calculated from the two readings at the two detectors using the Beer Lambert Law. 2.1 Beer Lambert Law The Beer Lambert law describes the transmission (and absorption) of an optical beam by an absorbing species and is given by 9 (2.1) zNzeIeII00 where = absorption coefficient (cm -1 ) 5 z = optical path length ( cm )
Figure 2.1. Schematic of apparatus to measure optical absorption and transmission. 6
7 = absorption cross section ( cm 2 ) and N = concentration of molecule (molecules / cm 3 ) The transmission and absorption of the beam is given by Transmission T = I / I 0 Absorption A = 1 T and Absorbance = log 10 ( T ) As can be seen in Equation (2.1), the attenuation, z is the product of the absorption of the substance and the optical path length. For sensitive spectroscopic measurements we want the sample to have moderate to high absorption, but for weakly absorbing substances, where is small, we have to increase the pathlength of the cell. This can often be accomplished by making multipasses of the beam through the cell or by using long optical waveguides. An optical waveguide is a li ght conduit consisting of a sl ab, strip or cylinder of transparent material surrounded by another mate rial of lower refractive index. The light is transported through the inner medium without radiating into the surrounding medium because of total internal reflection. Figure 2.2 shows several different geometries of optical waveguide. The most widely used of these waveguide geometries is an optical fiber, as shown in Figure 2.2 (c) which is made of an inner core and an outer cladding of low-loss dielectric material such as glass.
(a) (b) (c) Fig 2.2 Optical waveguides : (a) slab geometry ; (b) strip geometry ; (c) fiber or capillary geometry 8
9 CHAPTER 3. BASIC THEORY OF RECTANGULAR OPTICAL WAVEGUIDES In this chapter the theory behind the transmission of electromagnetic radiation through a rectangular waveguide is given. Th e equations outlined were obtained from Fundamentals of Photonics by Saleh and Laser and Electro-Optics by Davis. 10,11 The equations are given here for completeness and will be used later to show that the surface roughness and small changes in index of refrac tion or reflection angle inside the core (especially due to surface films) can greatly influence the overall transmission of the waveguide. 3 .1 Wave Guide Modes There are a number of important effects in waveguide propagation that require the use of electromagnetic theory. A simple a pproach to carrying out an electromagnetic analysis is to associate with each optical ray a transverse electromagnetic (TEM) plane wave. The total electromagnetic field is the sum of these plane waves. Consider a propagating monochromatic TEM plane wave of wavelength = 0 /n, wave number k=nk 0 and phase velocity c = c 0 / n, where n is the refractive index of the medium between two mirrors as shown in Figur e 3.1. The wave is polarized in the x direction and its wave vector lies in the y z plane at an angle with the z-axis. Like the
Fig.3.1. Optical plane waveguide 10
optical ray, the wave reflects from the upper mirror, travels at an angle -, reflects from the lower mirror, and travels once more at an angle and so on. Since the electric field is parallel to the mirror, each reflection is accompanied by a phase shift but the amplitude and polarization are not changed. The phase shift ensures that the sum of each wave and its own reflection vanishes so that the total field is zero at the mirrors. At each point within the wave-guide we have TEM waves traveling in the upward direction at an angle and others traveling in the downward direction at an angle ; all waves are polarized in the x direction. On imposing a self-consistency condition by requiring that as the wave reflects twice, it reproduce itself as shown in Figure 3.2, so that we have only two distinct plane waves. Fields that satisfy this condition are called eigenmodes or simply modes of the waveguide. Modes are fields that maintain the same transverse distribution and polarization at all distances along the waveguide axis. Self-consistency guarantees this shape invariance as shown in Figure 3.3. In reference to Figure 3.2, the phase shift encountered by the original wave in traveling from A to B must be equal to, or different by an integer multiple of 2, from that encountered when the wave reflects, travels from A to C, and reflects once more. Accounting for a phase shift of at each reflection, we have 2 AC / 2 2 AB / = 2 q where q = 0,1,2,... Since AC AB = 2d sin where d is the distance between the mirrors, 2 (2d sin )/ = 2 (q + 1), and md 2sin22 (3.1) 11
where m =1,2, and m = q +1 The self consistency condition is therefore satisfied only for certain bounce angles, = m satisfying dm2sin (3.2) where each integer m corresponds to a bounce angle, m and the corresponding field is called the m th mode. The m = 1 mode has the smallest angle where 1 = sin (/2d); modes with larger m are composed of more oblique plane wave components. When the self consistency condition is satisfied, the phases of the upward and downward plane waves at points on the Z axis differ by half the round trip phase shift q q = 0,1,.., or (m ), m = 1,2,, so that they add for odd m and subtract for even m. Since the y component of the propagation constant is k y = nk 0 sin, it is quantized to the values k ym = nk 0 sin m = (2/) sin m Using Equation (3.2) we obtain dmkym (3.3) 13
As can be seen, the allowed values for k ym are spaced by /d. Equation (3.3) states that the phase shift encountered when a wave travels a distance 2d (one round trip) in the y direction, with propagation constant k ym must be a multiple of 2. 3.2. Optical Propagation Constants The guided wave is composed of two distinct plane waves traveling in the y-z plane at angles with the z-axis. Their wavevectors have components (0, k y k z ) and (0, k y k z ). Their sum or difference therefore varies with z as exp (-jk z z), so that the propagation constant of the guided wave is = k z = k cos Thus is quantized to the values m = k cos m from which 2 m = k 2 (1 sin 2 m ). Using Equation (3.2), we obtain 22222dmkm (3.4) As can be seen in Equation (3.4), higher order (more oblique) modes travel with smaller propagation constants. 14
3.3 Group Velocities A pulse of light (wavepacket) of angular frequency centered at and propagation constant travels with a velocity dd/ known as the group velocity. The propagation constant of mode m is given by Equation (3.4) from which This is an explicit relation between 22222/)/(dmcm m and known as the dispersion relation. Taking the derivative and assuming that c is independent of (i.e., ignoring dispersion in the waveguide material), we obtain so that from which the group velocity of mode m is given by 2/2/2cdddmm mmmmckccddcos/cos//22 mmc cos (3.5) Thus, different modes have different group velocities. More oblique modes travel with a smaller group velocity since the longer path of the zigzagging process delays them. 15
16 CHAPTER 4. UNDERSTANDING TRANSMISSION OF LIGHT THROUGH CYLINDRICAL WAVEGUIDES In this chapter the theory behind the transmission of electromagnetic radiation through a cylindrical waveguide along with attenuation, dispersion and other surface effects are discussed. The equations outlined were obtained from Fundamentals of Photonics by Saleh and Laser and Electro-Optics by Davis. 10,11 The equations are given here for completeness to help interpret our experimental results. 4.1 Optical Ray Transmission A step-index fiber or an optical cylindrical waveguide has a central core of index of refraction n 1 surrounded by cladding of index n 2 where n 2 < n 1 When a ray of light enters such a optical fiber or waveguide, as shown in Figure 4.1, it will be guided along inside the core of the fiber if the angle of incidence between core and cladding is greater than the critical angle. Two distinct types of rays can travel along inside the fiber in this way: meridional rays travel in a plane that contains the fiber axis, skew rays travel in a nonplanar zig-zag path and never cross the fiber axis, as illustrated in Figure 4.2. For the meridional ray in Figure 4.1, total in ternal reflection (TIR) occurs within the core if i > c or sin i > n 2 / n 1 From Snell's law, applied to the ray entering the fiber, sin = sin 0 / n 1 (4.1)
Here is the angle of incidence at the fiber entrance, i is the angle of incidence at the first reflection inside the fiber, 0 is the angle of refraction, and c is the critical angle of the fiber. Since + i = 90, the condition for total internal reflection is 2122210][sinnn (4.2) Defining = (n 1 -n 2 ) / n 1 and for the case where is small, Equation (4.2) can be written as 2121210)])([(sinnnnn (4.3) or sin 0 < n 1  0.5 (4.4) If a lens is used to focus light from a point source into a fiber, as shown in Figure 4.3, then there is a maximum aperture size D that can be used. When the end of the fiber is a distance d from the lens, light rays outside the crosshatched region enter the fiber at angles too great to allow total internal reflection. The numerical aperture, NA, is given by NA = sin 0 D/2d = sin 0 = 212221][nn = n 1  0.5 (4.5) 18
If the lens is chosen to be no larger than necessary, then the lens diameter will be D. The ratio of focal length to diameter, f / D is a measure of the focusing / light collecting properties of the lens and is called the F-number. To match a distant source to the fiber the F number should be equal to (2 NA) -1 It should be noted that if the size of the input optical beam is smaller than the aperture size of the lens, D, then the F-number is calculated using the beam size. The above analysis of a step-index cylindrical optical fiber by means of simple ray theory considered only the paths of meridional rays, whose trajectories lie within the plane containing the fiber axis. We have also tacitly assumed that Snell's law strictly delineates between those rays that totally internally reflect, and are bound within the fiber, and those that escape or refract into the cladding. However, if one considers the trajectories of skew rays, a third class of rays, so-called tunneling or leaky rays exists. These are rays that appear at first glance to satisfy the Snell's law requirement for total internal reflection. However, because the refraction occurs at a curved surface they can in certain circumstances leak propagating energy into the cladding. The actual length of the light traveled inside the waveguide is related to the index of refraction of the core by the relation. 11 ]][2[cos12221222corecorecoreznlnnL (4.6) where 19 z = the angle the ray makes with the z axis
= radius of the core, = screw angle, .cossin,coszcorezcorenln and n core = refractive index of the core. As can be seen in Equation (4.6), the actual path length depends upon the index of refraction of the core as well as small variations in the angle of the rays hitting the wall surfaces. As such, small variations in the surface properties (roughness or n) could alter the L value. The LWCC waveguide is just like an optical fiber. In this section we cover the basic analysis of optical propagation in a cylindrical waveguide made of low loss materials. It has a central core in which the light is guided, embedded in an outer cladding of slightly lower refractive index Figure 4.4. Light rays incident on the core-cladding boundary at angles greater than the critical angle undergo total internal reflection and are guided through the core without refraction. Rays of greater inclination to the fiber axis lose part of their power into the cladding at each reflection and are not guided. In a waveguide light propagates in the form of modes. Each optical mode travels along the axis of the waveguide with a distinct propagation constant and group velocity, maintaining its transverse spatial distribution and its polarization. In planar waveguides, each mode was the sum of the multiple reflections of a TEM wave bouncing within the 20
slab in the direction of an optical ray at a certain bounce angle. This approach is approximately applicable to cylindrical waveguides as well. When the core diameter is Figure 4.4 Cylindrical dielectric waveguide 21
small, only a single mode is permitted and the fiber is said to be a single-mode fiber. Fibers with large core diameters are multimode fibers. One of the difficulties associated with light propagation arises from the differences among the group velocities of the modes. This results in a variety of travel times so that light pulses are broadened as they travel through the fiber. This effect, called modal dispersion, limits the speed at which adjacent pulses can be sent without overlapping and therefore the speed at which a fiber-optic communication system can operate. 4.2 Guided Waves The propagation of monochromatic light in step-index fibers can be analyzed using electromagnetic theory and determining the electric and magnetic fields of guided waves that satisfy Maxwell's equations and the boundary conditions imposed by the cylindrical dielectric core and cladding. As in all waveguides, there are certain special solutions, called modes, and two independent polarization states. Each of the components of the electric and magnetic fields must satisfy the Helmholtz equation, where n = n ,02022UknU 1 in the core (r < a) and n = n 2 in the cladding (r > a) and k o = 2 / 0 We assume that the radius b of the cladding is sufficiently large that it can safely be assumed to be infinite when examining guided light in the core and near the core-cladding boundary. In a cylindrical coordinate system (see Figure 4.5) the Helmholtz equation is 22
Figure 4.5 Cylindrical coordinate sysytem for waveguide 23
0112022222222UknzUUrrUrrU (4.7) where the complex amplitude U = U(r, z) represents any of the cartesian components of the electric or magnetic fields or the axial components E z and H z in cylindrical coordinates. We are interested in solutions that take the form of waves traveling in the z direction with propagation constant so that the z dependence of U is of the form e -jz Since U must be a periodic function of the angle with period 2, we assume that the dependence on is harmonic, e -jl where l is an integer. Substituting U = U(r, z) = u(r) e jl e -jz ( 4.8) where l = ..-2,-1,0, 1,2 into Equation (4.7), an ordinary differential equation is obtained, 0)(122220222u r lkndrdurd r ud (4.9) The wave is guided (or bound) if the propagation constant is smaller than the wave number in the core ( < n 1 k o ) and greater than the wave number in the cladding ( > n 2 k 0 ). It is therefore convenient to define 24
, (4.10) 220212knkT and (4.11) 202222kn For guided waves, and are positive and k 2Tk 2 T and are real. Equation (4.9) can be written in core and cladding separately. 0)(122222u r lkdrdurd r udT r < a ( core), (4.12a) and ,0)(122222u r ldrdurd r ud r > a (cladding ). (4.12b) Equations (4.12) are well-known differential equations whose solutions are the family of Bessel functions. Excluding functions that approach at r = 0 in the core or at r ~ in the cladding, we obtain the bounded solutions u ( r ) J l (k T r), r < a ( core ) (4.13a) and u ( r ) J l (k T r), r > a ( clading ) (4.13b) 25
where J l (x) is the Bessel function of the first kind and order l, and K l (x) is the modified Bessel function of the second kind and order l. The function J l (x) oscillates like the sine or cosine functions but with a decaying amplitude. In the limit x >> 1, ]2)21(cos[)2()(21lxxxJl (4.14) In the same limit, K l (x) decays with increasing x at an exponential rate, )exp()8141()2(221)(xxlxKxl (4.15) The examples of the radial distribution u(r) given in Equation (4.13) are shown in Figure 4.6. The parameters k T and determine the rate of change of u(r) in the core and in the cladding, respectively. A large value of k T means faster oscillation of the radial distribution in the core. A large value of means faster decay and smaller penetration of the wave into the cladding. As can be seen from Equation (4.13) and Equation (4.14) the sum of the squares of k T and is a constant (4.16) 20220222122.)(kNAknnkT So that as k T increases, decreases and the field penetrates deeper into the cladding. As k T exceeds NA times k o becomes imaginary and the wave ceases to be bound to the core. 26
Figure 4.6 Examples of the radial distribution u(r) given by Equation (4.7) for (a) 1=0 and (b) I = 3. The shaded area represent the fiber core and the unshaded areas the cladding. The parameters k T and and the two proportionality constants in (7.7) have been selected such that u(r) is continuous and has a continuous derivative at r = a. Larger values of k T and lead to a greater number of oscillations in u(r). 27
4.3 The V Parameter It is convenient to normalize k T and by defining X = k T a Y= a (4.17) Using Equations( 4.16) and (4.17) X 2 + Y 2 =V 2 (4.18) where V=NA k 0 a and NAaV02 (4.19) V is an important parameter that governs the number of allowed modes within the fiber and their propagation constants 4.3 Attenuation and Dispersion The attenuation of the optical beam follows from the Beer Lambert Law Equation (2.1). Rayleigh scattering is another intrinsic effect that contributes to the attenuation of light. The random localized variations of the molecular positions in glass create random inhomogeneities of the refractive index that act as tiny scattering centers. The amplitude of the scattered field is proportional to 2 The scattered intensity is therefore proportional to 4 or to 1 / 4 so that short wavelengths are scattered more than long wavelengths. Thus blue light is scattered more than red. 28
When a short pulse of light travels through an optically fiber/waveguide its power is dispersed in time so that the pulse spreads into a wider time interval. There are four sources of dispersion in optical fibers: modal dispersion, material dispersion, waveguide dispersion, and nonlinear dispersion (Fundamentals of Photonics; Saleh, 1991). Modal dispersion occurs as a result of the differences in the group velocities of the modes. A single impulse of light entering an M-mode fiber at z = 0 spreads into M pulses with the differential delay increasing as a function of z. For a fiber/wave guide of length L, the time delays encountered by the different modes are q = L / v q q = 1,..., M, where v q is the group velocity of mode q. If v min and v max are the smallest and largest group velocities, the received pulse spreads over a time interval L / v min L / v nax Since the modes are generally not excited equally, the overall shape of the received pulse is a smooth profile. An estimate of the overall rms pulse width is T = (L / v min L / v nax ). This width represents the response time Refractive index is a function of wavelength for any dispersive medium. An optical pulse travels in a dispersive medium of refractive index n with a group velocity v = c 0 /N, where N = n 0 d /d 0 Since the pulse is a wavepacket, composed of a spectrum of components of different wavelengths each traveling at a different group velocity, its width spreads. The temporal width of an optical impulse of spectral width (nm), after traveling a distance L, is |)/)(/(||)/)(/(|000cLNddvLddT from which LDT || (4.20) is the response time and where 29
20200dndcD (4.21) is the material dispersion coefficient. As can be seen in Equation (4.20) the response time increases linearly with the distance L. The group velocities of the modes may depend on the wavelength even if material dispersion is negligible. This dependence, known as wave guide dispersion, results from the dependence of the field distribution in the fiber on the ratio between the core radius and the wavelength (a / 0 ). If this ratio is altered, by altering 0 the relative portions of optical power in the core and cladding are modified. Since the phase velocities in the core and cladding are different, the group velocity of the mode is altered. The group velocity v = (d/d) -1 and the propagation constant are determined from the characteristic equation, which is governed by the fiber V parameter V = 2( a / 0 ) NA = (a NA/c 0 ). In the absence of material dispersion (i.e., when NA is independent of ), V is directly proportional to so that dVdcNAaddVdVdddv 01 (4.22) The pulse broadening associated with a source of spectral width is related to to the time delay, L / v, by T =| (d / d 0 )(L / v)| Thus 30
T = |D w | L ( 4.23 ) where )1()1(00vddvddDw (4.24) is the waveguide dispersion coefficient. Substituting Equation(4.22) into Equation (4.23) we obtain 2220)21(dVdVcDw (4.25) Thus the group velocity is inversely proportional to d/dV and the dispersion coefficient is proportional to V 2 d 2 / dV 2 4.5 Combined Material and Waveguide Dispersion The combined effects of material dispersion and wave guide dispersion (referred to here as chromatic dispersion) may be determined by including the wavelength depen-dence of the refractive indices, n 1 and n 2 and therefore NA, when determining d/dw from the characteristic equation. Although generally smaller than material dispersion, wave guide dispersion does shift the wavelength at which the total chromatic dispersion is minimum. Since chromatic dispersion limits the performance of single-mode fibers, more advanced fiber designs aim at reducing this effect by using graded-index cores with 31
refractive-index profiles selected such that the wavelength at which wave guide disper-sion compensates material dispersion is shifted to the wavelength at which the fiber is to be used. Dispersion shifted fibers have been successfully made by using a linearly tapered core refractive index and a reduced core radius. This technique can be used to shift the zero-chromatic-dispersion wavelength from 1.3 m to 1.55 m, where the fiber has its lowest attenuation. Note, however, that the process of index grading itself introduces losses since dopants are used. Other grading profiles have been developed for which the chromatic dispersion vanishes at two wavelengths and is reduced for wavelengths between. These fibers, called dispersion flattened, have been implemented by using a quadruple-clad layered grading. 4.9 The Goos-Hanchen Shift. For a cylindrical waveguide Goos Hanchen showed that the effective pathlength of the ray of the light might be much different than predicted by following the simple approach as described above. 11 The shift as shown in Figure 4.7 and can be given by 2211zazsnZ (4.26) where z = /2 1 Here, a = /2 c where 1 is the incident angle and c the critical angle. This shift in axial position makes the distance traveled by the ray in propagating a distance l along the waveguide shorter than it would be without the shift. However, the 32
Figure 4.7 The GoosHanchen shift. When n 1 > n 2 33
34 shift is very small unless the ray angle is close to the critical angle, in which case the evanescent portion of the associated wave penetrates very far into the cladding.
35 CHAPTER 5. LIQUID CORE CAPILLARY WAVEGUIDES Recently liquid-core waveguides have been shown to offer advantages over conventional cells in absorbance spectroscopy, 1,2 fluorescence spectroscopy, and Raman spectroscopy. 12,13 In absorption spectroscopy the guiding of light in the liquid core makes possible a long path length. 5.1 Introduction Liquid-core waveguides based on capillary tubes have been investigated for spectroscopic applications for several decades. 1,2,3,4,6 The earliest work studied glass (refractive index n = 1.52 ) or silica (n = 1.46) capillary tubes, but the choice of liquids was very limited because of the requirement th at the refractive index of the liquid core exceed that of the capillary cladding. In pa rticular, light could not be guided in water (n=1.33) or in aqueous solutions using th ese capillaries. The development of the lowrefractive-index amorphous fluoropolymers TeflonAF 1600 (n=1.31) and Teflon AF 2400 (n= 1.29) made it possible to develop capillary waveguides with aqueous cores. 14 That the presence of the TeflonAF layer pr omotes waveguiding has been demonstrated by comparing coated channels to similar cha nnels made without the Teflon AF coating. Teflon AF has long been the only option for a low-refractive-index cladding for an
36 aqueous-core waveguide; however, it is not an ideal material for this application. Teflon AF has poor adhesion to commonly us ed substrates, requiring that additional adhesion-promoting steps be added in fabri cation. In addition, the refraction index contrast between core and cladding is at most n = n core n cladding 0.04. Figure 5.1 shows a schematic of a Teflon coated capillary waveguide. The total reflection occurs at the outside surface of the tubing, which holds the liquid. As demonstrated by Tsunoda, 15 if a layer of the medium external to the capillary has a refractive index lower than the liquid core, and the external medium is optically clear, light launched into the liquid core within an acceptance angle will be totally reflected at the tubing/medium interface, independent of the refractive index of the wall material. In the important application of collec tion and transport of fluorescence generated within the aqueous core, the fraction of fluorescent photons that are captured by total internal reflection and transported along the waveguide increases as the cladding index decreases according to 16 = (2/ ) cos ( n clad / n core ) (5.1) for TeflonAF 2400 with n clad 1.31, = 0.11. For TeflonAF 1600 with n clad 1.29, = 0.16. In the limiting case of a cladding with n clad 1 (air), = 0.46. Hence, cladding materials with refractive indices lowe r than that of TeflonAF could provide a substantial increase in the efficiency.
Figure 5.1 Cross sectional view of light transmission through the liquid capillary waveguide cell. 37
38 5.2 Commercial LWCC from World Precision Instruments,Inc A commercial company, WPI (World Prece ssion Instruments), has developed the LWCC (liquid wavegide capillary cell), i.e. li quid core optical waveguide, which offers an increased optical pathlength compared to a standard cuvette and a small volume for spectroscopic applications. 14 A schematic of the unit is shown in Figure 5.2 and a photograph of the unit is shown in Figure 5.3. As can be seen in Figure 5.3 the flexible LWCC optical waveguide is coiled into one and a half loops inside the container and has a diamter of about 20 cm. The LWCC as developed by WPI was a second-generation unit termed as Type II, which is made from fused silica tubing with an outer coating of Teflon AF. Type II offers an improved signal stability and easier removal of any air bubbles trapped at its hydrophilic cell wall. Type I (the earlier mode l) required pH buffering of the sample due to CO 2 penetration. Baseline shifts were also observed due to trapped organic vapor. These effects, which were supposedly only observed in Type I, are due to the gas permeable nature of Teflon AF. This effect is eliminated in Type II, since the fused silica wall is impermeable to gases. Differences are shown in Table 5.1.
Figure 5.2 Schematic of commercial 1 m long Liquid Waveguide Capillary Cell from WPI, Inc showing Input/Output for liquid sample flow and Input/Output for light source. 39
Figure 5.3 Photo of LWCC waveguide 40
Table 5.1 Specifications of Type I and Type II LWCC waveguide. 41
42 5.3 Waveguide Properties The LWCC waveguide has two ST-ma ting sleeves for light input and output. Two fiber optic cables with a fiber core diameter of 400 m are necessary to connect LWCC waveguide to a light source and a detector. The LWCC waveguide can be connected to a pump, to a chromatography co lumn, or can even be filled manually by a syringe. Input and output connectors are 1/16" Compression fittings are made of PEEK (a type of plastic). 14 Pressures of approximately 1.5 PSI are n ecessary to maintain a water flow rate of 1 ml/min. Cells have been operated at 100 to 200 PSI without observing malfunctions. The Teflon AF tubing and sili ca tubing are reported to withstand pressures of 1000 PSI and 2000 PSI, respectively. The effective pathlength as provided by the manufacturer for 1 m LWCC waveguide is (0.94+ 0.01) m. This is caused by the fact that light is partially traveling in the fused silica wall and is consistent with the earlier analysis of the Goos-Hanschen shift. The effective pathlength was provided by the ma nufacturer and was probably determined from precession absorption measurements of a known absorbing solution.
43 The LWCC manufacturer has indicated that it is essential to clean the walls of the capillary tubes between each sample run to rem ove any left over bio or chemical film on the tube walls, which may effect the re flective properties and the overall optical pathlength of the unit. Thus the cleaning procedure is important and the following method was suggested by the manufacturer. 14 The Standard Procedure incl udes flushing the LWCC waveguide before and after each usage with the following reagents, in sequence, for 1-2 minutes each: 1. Organic solvent such as acetonitrile 2. 1 M NaOH (Sodium Hydroxide) 3. 1 M HCl ( Hydrocloric Acid ) 4. Distilled water when the standard cleaning procedure fails to stabilize the transmission of light through the LWCC waveguide Liquid Drano diluted 1: 1 in water maybe be used. Although its is a very effective chemical for removing most contaminants, its toxicity and the difficulty of rinsing it out of the waveguide make it the last choice. 5.4 Manufacturers Cleaning Procedure
44 CHAPTER 6. PRELIMINARY ABSORPTION MEASUREMENT S OF SELECTED LIQUIDS USING 266 nm AND 355 nm LASER SOURCES AND 1 cm ABSORPTION CELL Preliminary absorption measurements were made using a short pulse 266 nm and 355 nm laser beam and 1cm long absorption cell. A schematic of the apparatus is shown in Figure 6.1 and a photo is s hown in Figure 6.2. The output from one of the selected lasers was sampled by a beam splitter a nd then transmitted through a 1 cm quartz absorption cell that contained the water samp le. The transmitted beam intensity after going through the cell was measured by the s econd silicon detector (UDT Sensors Inc, Model No UV-013E). To reduce extraneous fluorescence and scattering, the optics and sample cells were made of fused silica or qua rtz. A quartz cuvette (Starna cells, Model IQ-10) was used as the sample holder. The cuvette was made of Suprasil quartz to reduce its fluorescence. The dimension of the cell was 12.5 mm 12.5 mm 45 mm, and had an inner pathlength of 10 mm. 6.1 Microchip Laser Source A microchip laser (JDS Uniphase, Model NU-10110-100) operating at 266 nm and another microchip laser (JDS Uni phase, Model NV-21401-100) operating at 355 nm were used as the light source.
Figure 6.1 Schematic of optical absorption experiment using 1cm quartz absorption cell. 45
Figure 6.2 Photograph of optical absorption experiments using 1cm quartz absorption cell. 46
47 The microchip laser was a compact laser, 3 cm 3.5 cm 15 cms, in dimension, and operated at 266 nm, the 4-th harmonic of the 1064 nm Nd:YAG laser. The output pulse energy was about 1 J/pulse at a repetition rate of 8 KHz and a pulse width of 0.4 ns. 17 The output of the laser was vertically polarized 100:1. The laser was mounted in the system so that the polarization of the laser was in the horizontal direction. A schematic of the laser is shown in Fig. 6.3. The laser consisted of a Nd 3 :YAG crystal bonded to a thin layer of Cr 4 :YAG saturable absorber (passive Q-switching), which was coated with a thin film mirror to form the laser cavity. The crystal was end pumped by a CW diode laser. The Cr:YAG was opaque to 1064 nm and prevents lasing until a critical amount of energy had been ab sorbed. When the absorber reached its threshold, it saturates and lasing began. Th e pulse repetition rate was determined by the time constant of the absorber and the amount of energy being input to the gain medium. Pulse energy was independent of the pump pow er as the Q-switching was always at the same intracavity energy, and was directly pr oportional to the thickness of the absorber medium. The pulse width was determined by the cavitys roundtrip time, so that the
49 compactness of the cavity resulted in pulse widths shorter than a nanosecond, about 0.4 ns. 17 The output of the laser was frequency doubled by using a KTP (Potassium Titanyl Phosphate) crystal. The doubled output at 532 nm was frequency doubled once again by a BBO crystal to 266 nm. The residual light at 1064 nm and 532 nm were reduced by using optical blocking filters. 6.2 Absorption Measurements Using a 1 cm Quartz Cuvette The cuvette was filled with various wate r samples and absorption was measured by comparing the signal intensity of the laser intensity before and after passing through the cell. Clean distilled water (Zephyrhills bra nd) was used in the cell to establish the full transmission level (i.e, T =100%). As a result of this normalization, the 4% Fresnel reflection loss at each cell window interf ace was normalized out. It was found that absorption due to the pure water was too sma ll for us to measure using the 1cm optical cell for both the 266 nm and 355 nm laser source. This was to be expected since the absorption coefficient is very small for pure water as indicated in Figure 1.1. As a result we measured the absorption due to several other liquids of choice with high turbidity including coffee, orange ju ice, and cola. Figures 6.4 6.9 show the measured transmission of the laser radiati on through the sample as a function of the relative concentration of the samples. The concentration of the samples was reduced by diluting it with distilled water. As can be seen the transmission is reduced as the
Cola (355 nm)00.10.20.30.126.96.36.199.80.9188.8.131.52.811.2Concentration(arbitrary units)Transmission Figure 6.6 Transmission as a function of concentration for diluted cola through a 1 cm quartz cell using 355 nm laser source. Cola (266 nm)00.10.20.30.184.108.40.206.80.9220.127.116.11.811.2Concentration(arbitrary units)Transmission Figure 6.7 Optical transmission as a function of concentration for diluted cola through a 1 cm quartz cell and 266 nm laser source. 51
Orange Juice (355 nm)00.10.20.30.18.104.22.168.80.922.214.171.124.811.2Concentration(arbitrary units)Transmission Figure 6.8 Transmission as a function of concentration for diluted orange juice through a 1 cm quartz cell using 266 nm laser source. Orange Juice (266 nm)00.10.20.30.126.96.36.199.80.9188.8.131.52.811.2Concentration(arbitrary units)Transmission Figure 6.9 Optical transmission as a function of concentration for diluted orange juice using a 1 cm quartz cell and 266 nm laser source. 52
53 concentration increases for all samples. Also shown in the figure is an exponential (least squares) fit to the data. According to the Beer Lambert law the absorbance should be a linear function of the concentration while the transmission is an exponential function. In order to see this better the measured absorption coefficient, was plotted as a function of concentration and is given in Figures 6.10 6.15. As s hown in the Figures the measured absorption coefficient was found to be linear with the concentration. 6.3 Double Pass Absorption Measurements The removable double pass mirror shown in Figure 6.1 was used to double the optical path through the cell. The resultant absorption coefficient measurements are shown in Figure 6.16. It can be observed that the value of, is half that measured for the 1cm path length in Figure 6.11, as expected. The above experiments established that our basic absorption measurement system ha d accuracy on the order of few percent over a range of magnitude in the transmission readings.
Cola (355 nm) 00.10.20.30.184.108.40.206.80.9220.127.116.11.811.2Concentration(arbitrary units)Absorption coefficient (1/cm) Figure 6.12 Absorption coefficient as a function of concentration for diluted cola using 355 nm laser source in 1 cm cell. Cola (266 nm) 00.10.20.30.18.104.22.168.80.922.214.171.124.811.2Concentration(arbitrary units)Absorption coefficient (1/cm) Figure 6.13 Absorption coefficient as a function of concentration for diluted cola using 266 nm laser source in 1 cm cell. 55
Orange Juice (355 nm)00.10.20.30.126.96.36.199.80.9188.8.131.52.811.2Concentration(arbitrary units)Absorption coefficient(1/cm) Figure 6.14 Absorption coefficient as a function of concentration for diluted orange juice using 355 nm laser source in 1 cm cell. Orange Juice (266 nm)00.10.20.30.184.108.40.206.80.9220.127.116.11.811.2Concentration(arbitrary units)Absorption coefficient(1/cm) Figure 6.15 Absorption coefficient as a function of concentration for diluted orange juice using 266 nm laser source in 1 cm cell. 56
Coffee (266 nm)00.10.20.30.18.104.22.168.80.922.214.171.124.811.2Concentration(arbitrary units)Absorption Coefficient (1/cm) Figure 6.16 Absorption coefficient as a function of concentration for diluted coffee using 266 nm laser source in 2 cm cell. 57
58 CHAPTER 7. ABSORPTION MEASUREMENT OF SELECTED LIQUIDS USING 355 nm LASER SOURCE AND 1 m LWCC The LWCC waveguide was used in the absorption experiments as a way to increase the overall pathlength in order to obs erve weak absorption features of diluted water samples or weakly absorbing solutions. The LWCC waveguide was added to the absorption setup as shown in Figure 7.1. It can be seen the liquid waveguide was used in the absorption arm and the optical beam was gui ded by the liquid core flow. The rest of the apparatus is same as given earlier in Figur e 6.1. It should be noted that the laser beams were directed into the fiber optics w ith no focusing lens; this will be discussed further in a later section. The LWCC waveguide absorption setup wa s used to measure the absorption of several liquids including various water sample s and coffee. It was found that the 266 nm laser did not propagate well through the waveguide for liquids with high turbidity and was absorbed completely. This was probabl y caused by the fact that its wavelength was near the lower cut-off of the LWCC wave guide, which had a bandpass from 230 nm to 850 nm. The 355 nm laser on the other hand was transmitted through the LWCC waveguide with high transmission. The absorption due to diluted coffee was measured and our results are shown in Figure 7.2. It can be seen from this figure that the transmitted intensity of the 355 nm laser followed the Beer Lambert law as expected. It should be noted that the maximum
Figure 7.1 Schematic diagram of optical absorption experiment using the 1m optical liquid core waveguide. 59
Figure 7.2 Absorption spectrum of coffee at different concentration through the LWCC waveguide by using 355nm laser. Maximum concentration was Already diluted by a factor of twenty compared to data given previously in Figure 6.11. 60
61 concentration for the coffee shown in Figur e 7.2 was diluted by a factor of twenty compared to that used in Figure 6.11, to comp ensate for the increased optical path of the 1m long LWCC compared to the 1 cm long quartz cell. The 355 nm and LWCC absorption syst em was also used to measure the transmission of several types of water and th e results are given in Figure 7.3 for several sequential measurements over a four-minute period. As can be seen the Physics Department distilled water has much higher absorption at 355 nm than the fountain or USF tap water samples. This may have been due to plastic or organic contaminants and is consistent with previous laser induced fluorescence measurements performed by V.Sivaprakaram who saw contaminants in the Physics Department distilled water 18 Of interest in the data shown in Figure 7.3 is th at the measured transmission values are fairly constant (3%) over the 30-minute data run. This helps to establish that the basic absorption setup is stable, at least to this 3% levels. However, additional variability was observed when the LWCC waveguide was chemica lly cleaned between data runs; this is discussed in the following sections.
Transmission curve through LWCC00.20.40.60.8105101520253035Time (minutes)Transmission Fountain waterUSF tapwaterUSF distilled Figure 7.3 Transmission of 355nm laser through the LWCC for USF fountain water, USF tap water, USF distilled water over a period of time. 62
63 CHAPTER 8. VARIABILITY OF LASER TRANSMISSION MEASURENTS USING THE LWCC WAVEGUIDE It was found that the transmission measur ements using a laser source would often change in value between different water sa mple trials especially when the LWCC was cleaned (as suggested by the manufacture r) between each sample run following the procedure as discussed in Section 5.4. Thus we investigated the variability of the transmitted light through the LWCC using several different laser sources. 8.1 Experimental Set Up We replaced the two silicon detectors from the experimental setup used in the Figure 7.1 with a dual channel spectrometer and a computer to record the data as shown in Figure 8.1. The spectrometer (Ocean Optics ST 2000) used a 600 lines/mm grating (blazed at 400 nm) and enabled us to look at the inte nsity readings of the full spectrum of wavelengths from 220 nm to 900 nm for the di fferent light sources. It had an optical resolution of about 1.3 nm. The sensitivity of the silicon array detector in the spectrometer is shown in Figure 8.2(a), while the efficiency of the grating is given in Figure 8.2(b). The specifications of the spectrometer 19 are given in Table 8.1. Figure 8.3 shows the typical measured spectrum of the 266 nm laser source; Note: the 266 nm laser was used for this portion of the experiments instead of the 355 nm laser because the
Figure 8.1. Schematic of experimental setup of transmission system with liquid core capillary waveguide. 64
Figure 8.2 (a) Spectrometer detectors sensitivity as function of (b) Grating efficiency as a function of wavelength (channel #1 and channel #2).(Ocean Opticss ST 2000 Spectrometer Operating Manual) 65
Table 8.1 Specifications of the Ocean Optics ST 2000 spectrometer. 66
266 nm Wavelength spectrum02004006008001000200220240260280300Wavelength(nm)Intensity(arbitrary units) Figure 8.3 Typical spectrum for 266 nm laser source. 67
68 355 nm had broken and could not be repaired. The peak value of the laser source was used as the reference for the laser intensity. To check the drift in laser intensity we recorded the input and output power intensity over a period of time when the waveguide was filled with distilled water (Zephyhills brand). The results are shown in Figure 8.4 and Figure 8.5. As can be seen the power re mained fairly constant over the recorded time period of 50 minutes. The transmission of the LWCC using the 266nm laser was measured for several sample runs for pure distilled water (Zephyrhills brand) but the LWCC was chemically cleaned between each run by following the proce dure described in Section 5.4. Eight trial runs were made over a one-hour period and the measured input power levels and output power values are presented in Figure 8.6 and Figure 8.7. As can be seen the output power intensity had a great variability. This effect was consistently noticed on subsequent experiments. The cause of the variability was theorized to be due to several potential phenomena. At first we checked for alignment errors, vibration or movement of the laser output and input fiber, but this did not seem to be a major contributor to the discrepancy; the error was on the order of a few percent fr om these factors. After careful alignment and making sure that the apparatus did not m ove, it was interpreted that the transmission of the LWCC waveguide may change si gnificantly as the LWCC waveguide was chemically cleaned (i.e., flushed with solvents) between each sample runs.
Input intensity (266 nm)0200400600800100012001400010203040506Time (minutes)Intensity (arbitrary units) 0 Figure 8.4 Input intensity of 266 nm laser source over an hour. Output intensity (266 nm)0200400600800100012001400010203040506Time (minutes)Intensity (arbitrary units) 0 Figure 8.5 Output intensity of 266 nm laser source through LWCC filled with distilled water (Zephyrhills brand) over an hour. 69
Incident (266 nm) 01002003004005000246810Sample runsIntensity (arbitrary units) Figure 8.6 Incident intensity spectrum of 266 nm laser source for different trials when the waveguide was cleaned between sample runs. Output (266 nm) 01002003004005000246810Sample runsIntensity (arbitrary units) Figure 8.7 Transmitted intensity spectrum of 266 nm laser source through the LWCC filled with distilled water for different trials when it was chemically cleaned between runs. 70
71 In order to check on this hypothesis, the 266 nm laser (polarized) was replaced with a 632.8 nm Helium Neon laser (unpolaris ed). The Helium Neon laser was in the middle of the LWCC waveguide band pass while 266 nm is at the edge of the LWCC waveguides 230 nm 850 nm band pass. The input and output power of the tr ansmitted LWCC beam using the Helium Neon laser is shown in Figure 8.8 and Figure 8.9. As can be seen, the input and output is fairly constant. However when we chemi cally cleaned the LWCC waveguide between eight different sample runs, the variability was observed to be quite large and is shown in Figure 8.10 and Figure 8.11. The large vari ability seen with the Helium Neon laser is consistent with that observed with the 266 nm laser. It was theorized that the variability in the transmission observed above might have been due to the small changes in th e surface optical characteristics of the walls inside the LWCC waveguide. This will be discussed later.
Incident He-Ne laser05001000150020002500300035000510152025303540Time (minutes)Intensity (arbitrary units) Figure 8.8 Input intensity of Helium Neon laser source over a period of time Output He-Ne laser05001000150020002500300035000510152025303540Time (minutes)Intensity (arbitrary units) Figure 8.9 Output intensity of Helium Neon laser source through LWCC filled with distilled water( Zephyrhills brand) over an hour. 72
Incident He-Ne laser050010001500200025000123456789Sample runsIntensity( arbitrary units) Figure 8.10 Incident intensity of Helium Neon laser source for different trials when the waveguide was chemically cleaned between runs. Output He-Ne laser050010001500200025000123456789Sample runsIntensity (arbitrary units) Figure 8.11 Output intensity of Helium Neon laser source through the LWCC, which is filled with distilled water for different trials when the waveguide was chemically cleaned between runs. 73
74 CHAPTER 9. VARIABILITY OF 1m LWCC FOR NON COHERENT LIGHT SOURCES The previous chapter showed the measured variability in the transmission when a laser source was used with the LWCC. It was thought that maybe the use of a noncoherent light source could reduce the variab ility observed. This is because the LWCC wall reflectivity and optical path length for the coherent laser could have a greater dependence upon the individual propagating laser modes and polarization than that of a non coherent source. In order to study this we used a halogen light source and a Xenon arc lamp. 9.1 Experimental Set Up The absorption setup shown in Figure 8. 1 was modified by replacing the laser source with either a halogen lamp (General Electric Halogen Automotive Light H3 100) or a Xenon arc lamp (Model No. 790 Newport). The halogen lamp was powered by a HP DC power supply (Model No. 6264B operating at 12 V and 6A) and the Xenon arc lamp operated at a power level of about 75 W. The measured emission spectrum of the halogen light source was obtained using the Ocean Optics spectrometer and is shown in Figure 9.1. As can be seen it has a smooth emission spectrum from 475 nm to 875 nm, with a peak near 650 nm. The Xenon arc lamps measured emission spectru m is shown in Figure 9.2 and shows a smooth emission spectrum from 300 nm to 800 nm.
Halogen Lamp0500100015002000250030003500400002004006008001000Wavelength(nm)Intensity(arbitrary units) Figure 9.1 Measured emission spectrum of halogen light source. 75
Xenon Arc Lamp0500100015002000250030003500400002004006008001000Wavelength(nm)Intensity(arbitrary units) Figure 9.2 Measured emission spectrum of Xenon arc light source. 76
77 9.2 Measurements with Xenon arc Light Source The Xenon arc lamps measured emission spectrum was found to be quite stable as can be seen from Figure 9.3, which is the measured emission spectrum of Xenon arc lamp as a function of time. Figure 9.4 shows the spectrum measured after passing through the LWCC waveguide when it was filled with distilled water (Zephyrhills brand) over a period of 60 minutes. As can be seen both spectra are reasonably stable. To repeat the same variability measurements as done in the previous chapter using the laser sources, we measured the transm itted spectrum of the Xenon arc light source through the LWCC waveguide but where the LWCC waveguide was chemically cleaned between each sample run. The transmitted spectrum through the LWCC waveguide which was filled with distilled water (Zephyrh ills brand) was recorded over a of two hour period and is plotted in Figure 9.5. As can be seen, the output spect rum is fairy constant. It should be noted that the observed increase in sample run No. 2 wa s unusual, but was also proportionately observed in the direct output spectrum of th e Xenon arc lamp. The relative transmission spectrum of the eight sample runs was calculate d and is shown in Figur e 9.6 As can be seen, there was some variability in the transm ission spectrum, but overall the values were more consistent and displayed less variability than that seen earlier with the laser sources.
Figure 9.3 Output spectrum of Xenon Arc lamp as a function a function of time over a period of time. 78
Figure 9.4 Transmitted spectrum of distilled water through the LWCC waveguide as a function of time using a Xenon Arc lamp. 79
Figure 9.5 Transmitted spectrum of Xenon Arc lamp as a function of different sample runs through the LWCC waveguide(filled with distilled water, Zephyrhills brand) when it was cleaned between each sample runs. 80
Xenon arc transmission Ratio00.20.40.60.811.21.4300400500600700800900WavelengthTransmission ratio Figure 9.6. Transmission spectrum of distilled water through the waveguide using Xenon arc lamp as the light source for eight different sample runs when the waveguide was chemically cleaned between runs. 81
82 9.3 Measurements with Halogen Light Source The Halogen light output spectrum was measured as a function of time and the data is shown in Figure 9.7. As can be seen, the output was stable over a period of 100 minutes. Figure 9.8 shows the measured spectral intensity after transmission through the LWCC waveguide and its spectrum is observed to be stable also. The variability measurements were conducted over a period of two hours when the LWCC waveguide was chemically clean ed. The transmitted spectrum through the LWCC waveguide that was filled with distilled water (Zephyrhills brand) was recorded and is plotted in Figure 9.9. From the figure it can be interpreted that the LWCC waveguide readings remained constant. Using the incident and output spectra we calculated the relative transmission spectrum of eight sample runs, which were chemically cleaned between each run. The data is shown in Figure 9.10. As can be seen, the values were very consistent and displayed less variability than that seen earlier with the laser sources.
Figure 9.7 Output spectrum of Halogen light source as a function of time. 83
Figure 9.8 Transmission spectrum of distilled water through the LWCC waveguide as a function of time using a Halogen light source. 84
Figure 9.9 Transmission spectrum of distilled water through the waveguide using Halogen light source for different trails when the waveguide was chemically cleaned between runs. 85
Halogen transmission ratio00.10.20.30.126.96.36.19900500600700800Wavelength(nm)Transmission ratio Figure 9.10 Relative transmission spectrum of distilled water through the LWCC waveguide using Halogen light source for different runs when the LWCC wavguide was chemically cleaned between runs. 86
87 CHAPTER 10. VARIABILITY OF SCATTERED LASER SOURCE USING THE 1 m LWCC WAVEGUIDE The previous measurements indicated that a non-coherent source might exhibit less variability than that due to a coherent light source. Following this line of thought we used a laser source that had been backscattered off a white card thinking this would be more like a non-coherent light source. The experimental setup is the same as shown in Figure 8.1 with the addition of a white scattering card, and is shown in Figure 10.1. 10.1 Transmission Experiments With Scattered HeNe Laser The intensity of the scattered Helium Neon laser over a period of time was measured and is shown in Figure 10.2. For the same period of time, the intensity of the scattered Helium Neon transmitted through the LWCC waveguide that was filled with distilled water (Zephyrhills brand) was record ed over a period of time and is shown in Figure 10.3. It was observed that there is some variation of about 7 %. To check the variability of the LWCC waveguides transmission associated with cleaning process, the scattered Helium Neon laser source transmission was measured for several runs where the LWCC was chemically cleaned between each run. Eight runs were made over an hour period and input and output intensities were recorded. The data
Figure 10.1 Schematic of experimental setup of Transmission system with Liquid core Capillary waveguide for Scattered HeNe as the light source 88
Incident scattred He-Ne 05001000150020002500020406080100120140Time (minutes)intensity( arbitrary units) Figure 10.2 Emission spectrum of scattered HeNi laser over as a function of time Output scattered He-Ne (LWCC) 05001000150020002500020406080100120140Time (minutes)Intensity(arbitrary units) Figure 10.3 Transmitted emission spectrum as a function of time of scattered HeNe laser through the LWCC waveguide which was filled with distilled water ( Zephyhills brand) as a function of time. 89
90 is presented in Figure 10.5. It was observed that there was about the same level of variability on the order of about 7 % between each trial run. The ratio of the output to input is shown in Figure 10.6 and had a variability of about 7 %. The above results are somewhat surprising since we had thought that a more non-coherent aspect would have prevented this. It is possible that the fiber optics input was only sampling an individual speckle lobe emitted by the white scattering card and could be changing in intensity for small angular changes. This was even seen in the input data as shown in Figure 10.2. and Figure 10.3.
Scattered He-Ne0500100015002000250002468Sample runsIntensity(arbitrary units) 10 Figure 10.4 Emission spectrum of scattered Helium Neon laser source for different sample runs. Output scattered He-Ne (LWCC cleaned between runs)0500100015002000250002468Sample runIntensity(arbitrary units) 10 Figure 10.5 Transmitted intensity of scattered He-Ne laser through the LWCC waveguide when the LWCC waveguide was chemically cleaned between each sample run. 91
Relative transmission for scattered He-Ne00.511.522.502468Sample runsRelative transmission ratio 10 Figure 10.6 Relative transmission ratio of scattered HeNe laser through the LWCC waveguide when the LWCC was cleaned between each sample run. 92
93 CHAPTER 11. COMPARISON OF VARIATION IN TRANSMITTED LWCC WAVEGUIDE VALUES AND POSSIBLE LINK TO SURFACE ROUGHNESS OR WALL REFLECTANCE ANGLE VARIATION The previous transmission intensity values for the LWCC waveguide as a function of sample runs is s hown in a combined plot in Figure 11.1. As can be seen, there was a relatively little variability ( 3 %) using the non-coherent Halogen or Xenon light source, large variability ( 50 %) using the He-Ne laser source or 266 nm Nd:YAG laser ( 20 %) and moderate variability ( 7 %) using a scattered He-Ne laser source. 11.1 Comparison With Recent Studies There may be several different explanati ons for the above results, including potential changes in the LWCC waveguides surface physical/optical characteristics between sample runs, the development of a bio or chemical film along the waveguides surface, and changes in the laser propagation modes or polarization if there were small alignment changes in the optical fiber leading to the LWCC. It is not yet known if these are responsible for the variability seen, but it is in teresting to look at recent studies performed by Barwicz and H. Haus who studied the effect of surface roughness in microphotonic waveguides and their influence on scattering losses and transmission 19 In their paper, they used a surface geometry as shown in Figure 11.2 and assumed a linear
Figure 11.2 (a) perspective view and (b) top view of a rough rectangular waveguide The top and bottom walls are considered smooth and only the sidewalls are believed rough. A uniform surrounds the waveguide-core cladding of refractive index n clad (c) Spherical coordinate system used. (from Barwicz, 2005) 95
96 straight waveguide. As can be seen, the roughness and index of refraction varied across the waveguide. They calculated the scatte r due to such a surface as a laser beam propagated along the waveguide using a wavepropogation analysis including evanescent field calculations. They were able to show that the scattering losses show significant polarization dependence. This can be seen as in Figure 11.3 that shows the scattering losses for the TElike mode and a TM-like mode from their paper. As can be seen, they are very different for the two different polarizations. In another important paper by Lytle, they recognized the problem that curved hollow waveguides might be very inefficient because light no longer strikes the walls at a highly reflective grazing angle. 20 In addition, Propagation losses can increase when the focused laser beam is not inserted parallel to the waveguide axis. As a result they conducted a theoretical study of shallow reflectance angle effects in a linear straight waveguide. Figure 11.4 shows their propagati on losses when light is focused to a 20 m radius, inputted into a hollow square waveguide and the input angle is changed from 0.1 0 up to 1.0 0 As can be seen the loss (and transmi ssion) varies significantly as the input angle is changed slightly. It should be noted that in a paper by Song, a 785 nm excitation laser was used to enhance the Raman spectra of organic so lvents like benzene in a Type I LWCC waveguide. 12 The author saw significant differe nce in the measured or deduced absorption coefficient of water and benzene a nd these values were in quantitative but not
97 qualitative agreement with the previous theore tical work. These results are consistent with our measurements in that, their discrepancy is similar to our readings when a coherent laser source was used with the LWCC waveguide. 11.2 Additional LWCC Measurements Using Focused HeNe Laser Into Fiber Using the papers by Lytle as a guide (a s to the importance of input angle) we repeated some of our He-Ne laser source expe riments, and used a short focal length lens to input the light into the fibers. Unfortunately, the variability was observed to be of same order as observed without the focusing le ns and is shown in Figure 11.5. Further careful work is required to study this further. 1.3 General Conclusions While the above analysis does not directly answer our variability data, it does support the observation that the scatter of the polarized laser modes may depend significantly upon changes in the surface characteristics or changes in the reflection angle along the curved or coiled wall. In additi on, for the non-coherent sources (and possibly scattered He-Ne source), a summation over a ll polarizations and modes would produce a more homogeneous transmission value and thus less variability between sample runs.
Figure 11.4 Propagation losses when a 488nm laser is focused to a 20 -m radius at the entrance to a 100m x 100m square waveguide. The lines represent various insertion angles between the Gaussian beam and the waveguide axis.(from Lytle,2003) 99
Focussed He-Ne into Fiber(LWCC cleaned between runs)00.10.20.30.188.8.131.52.80.9102468Sample runsRelative transmission 10 Figure 11.5 Measured transmission values for the focused He-Ne (into fiber and LWCC) and LWCC cleaned between runs. 100
101 These results are important because they show the utility, for the first time, of using the LWCC waveguide with a spectroscopic laser source as opposed to a more conventional non-coherent light source. It s hould be added that anal ytical spectroscopists that use the flow injection analysis are just starting to use the LWCC waveguide as a standard instrument for the process control, but most often using a conventional non coherent light sources. 21,22 Our results have shown that the use of the LWCC waveguide with a laser source needs to be conducted with caution until further calibration and studies are conducted.
102 CHAPTER 12. CONCLUSION AND FUTURE WORK Laser spectroscopic absorption measurements were conducted on various samples of water using a long wave liquid capillary cell in order to study its potential for the measurement of trace species in water samples. An increase in the absorption cell length from one cm to one meter for the coiled LWCC waveguide increased the sensitivity of our measurement by two orders of magnitude, which enabled us to measure the absorption of clean water samples. The transmission intensity or spectrum was measured using the coiled liquid core capillary waveguide (LWCC waveguide), whic h was chemically cleaned between each sample run. The readings showed good c onsistency when the light source was noncoherent like a Xenon arc lamp or a Halogen lamp. However upon using a coherent laser source such as a Helium Neon laser operati ng at 632 nm or a Nd:YAG microchip laser operating at 266 nm and 355 nm, the transmi ssion readings were found to vary. In addition it was found that using scattered He lium Neon laser light as the source produced much less variability in the subsequent runs.
103 Our results seemed to be consistent with previous published work regarding the effect of thin films and surface roughness on waveguide walls (Barwicz, 2005). Their studies showed the variability in transmi ssion due to changes in the surface roughness (and index of refraction), especially between different polarizati on modes (TE or TM) propagating down the waveguide. Such an eff ect could be present in our experiments using the coiled LWCC waveguide especially since the surfaces were chemically cleaned between each sample run lending them susceptib le to changes in the films or coatings along the wall and possible slight changes in stability of the coil. In addition our results are consistent with studies by Lytle who showed the large sensitivity due to slight changes in the input direction into the waveguide. This thesis shows that further work needs to be conducted to study the use of a LWCC waveguide for laser spectroscopy applica tions, including the use of graded index fibers, different cleaning proce dures using a neutral pH solution to limit the formation of films inside the waveguide, and using a conti nuous water flow system to avoid the due to formation of water bubbles. In addition, detailed optical modeling of laser propagation through a curved (i.e., coiled) waveguide s hould be conducted as well as the possible changes in the coiled waveguide geometry (i .e. diameter or position) when the coil is cleaned using flowing chemicals. It may be possible that if the coiled LWCC waveguide is fixed or glued in place to limit its move ment, the variability observed with the laser source may be much smaller. Further studies are needed to verify these ideas.
104 REFERENCES 1. Yao, Wensheng., Byrne, H. Robert. Talanta 48, 27-282, 1988. 2. Callahan, R. Michael.,Rose, B. Joan., Byrne, H Robert .Talanta 58, 891-898, 2002. 3. Zhang, Jia-Zhong Deep Sea Research 47, 1157-1171, 2000. 4. Byrne, H. Robert., Waterbury, D. Robert., Kaltenbacher, Eric. Talanta. 50, 1307-1312, 2000. 5. Deng,Gang., Wei, Lily., Collins, E. Greg. Talanta 60, 9-16, 2003. 6. Kim, Young., Lee, Won. Bull. Korean chem. Soc. 8, Vol 22, 821-826, 2001 7. Li, Qian Perry,Hansell, A. Dennis., Millero, J. Frank., Marine chemistry. 2005 8. Sivaprakasam, Vasanthi. UV laser induced fluorescence spectroscopic studies and trace detection of dissolved plastics( Bisphenol-A) and organic compounds in water,(PhD Thesis USF, 2002). 9. Bernath, Peter F. Spectra of Atoms and Molecules (Oxford University Press, New York, 1995). 10. Saleh, Bahaa A. Fundamentals of Photonics ( Wiley Interscience Publication, 1991) 11. Davis, Christopher C. Lasers and Electro-Optics (Cambridge University Press, 1996) 12. Song, Li, Zhelyaskov, Valenti. Applied Spectroscopy Vol 52. 1364 -1367, No 10(1998). 13. Li, Song., Zhelyaskov, Valentin., Applied Spectroscopy 52, 1364 -1367, 1998 14. World Precession Instruments Inc Instruction Manual LWCC . 15. Tsunoda, K., Nomura, J., Nishi, S., Applied Spectroscopy 43, 11341137, 49 (1989) 16. Temkin, H., Gangopadhyay S. Optical Express Vol 12,No 26 p.6446 6455(2004).
105 17. Fein, Harry., Liu, Su-Yi. World Precision Instruments, Inc (Sarasota, FL) Chemical sensing techniques employing liquid core optical fibers . United States Patent 6016372. 18. Uniphase Corporation. NU-10110100 Laser Operators Manual, (1999) 19. Barwicz, T., Haus Hermann A. Threedimensional Analysis of scaterring losses due to sidewall roughness in microphotonic waveguides. Submitted to Journal of lightwave Technology (2005). 20.Lytle, Fred., Splawn, Bryan G. Applied Optics Vol.41, No. 31 p.6660-6665 (2002). 21.Trojanowicz, Marek, Flow Injection Analysis Instrumentation and applications, (World Scientific).2000. 22.Pappas, Alexandros.,Stalikas D. Constantine., Analytica Chimica Acta 455, 305-313, 2002.