USF Libraries
USF Digital Collections

Transient fluorescence spectroscopy and laser induced fluorescence lifetimes of terbium doped dipicolinic acid

MISSING IMAGE

Material Information

Title:
Transient fluorescence spectroscopy and laser induced fluorescence lifetimes of terbium doped dipicolinic acid
Physical Description:
Book
Language:
English
Creator:
Makoui, Anali
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Laser induced fluorescence spectroscopy
Transient saturation fluorescence spectroscopy
Fluorescence lifetime
Bacteria spore detection
Excitatin emission matrix spectra
Dissertations, Academic -- Physics -- Doctoral -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: We have investigated the use of deep UV laser induced fluorescence for the sensitive detection and spectroscopic lifetime studies of terbium doped dipicolinic acid (DPA-Tb) and used this to study the optical characteristics of DPA which is a chemical surrounding most bacterial spores. Background absorption spectra, fluorescence spectra, and Excitation Emission Matrix (EEM) spectra were made of the DPA-Tb complex, using both fixed 266 nm wavelength and tunable (220 nm --^ 280 nm) UV laser excitations. Of importance, the fluorescence lifetimes of the four main fluorescence peaks (488 nm, 543 nm, 581 nm, and 618 nm) of the DPA-Tb complex have been measured for the first time to our knowledge. The lifetimes of all the fluorescing lines have been measured as a function of DPA-Tb concentration, solvent pH, and solvent composition, including that for the weakest fluorescing line of DPA-Tb at 618 nm. In addition, a new spectroscopic lifetime measurement technique, which we call "Transient Fluorescence Spectroscopy" was developed. In this technique, a weak, quasi-CW, amplitude modulated UV laser (8.5 kHz) was used to measure the lifetimes of the fluorescence lines, and yields insight into energy transfer and excitation lifetimes within the system. This technique is especially useful when a high power laser is not either available or not suitable.^ In the latter case, this would be when a high power pulsed deep-UV laser could produce bleaching or destruction of the biological specimen. In addition, this technique simulated the excitation and fluorescence emission of the DPA-Tb using a 4-level energy model, and solved the dynamic transient rate equations to predict the temporal behavior of the DPA-Tb emitted fluorescence. Excellent agreement between the experiments and the simulation were found. This technique has the potential to provide a more accurate value for the fluorescence lifetime values. In addition, with the use of asymmetric excitation waveforms, the dynamic transient rate equation analysis may allow for detailed studies of selected transfer mechanisms in a wide range of other spectroscopic applications including rare-earth solid-state lasing materials and biological samples.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2007.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Anali Makoui.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 264 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001968110
oclc - 268957448
usfldc doi - E14-SFE0002183
usfldc handle - e14.2183
System ID:
SFS0026501:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Transient Fluorescence Spectroscopy And Laser Induced Fluorescence Lifetimes Of Terbium Doped Di picolinic Acid by Anali Makoui A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Physics College of Arts and Sciences University of South Florida Major Professor: Dennis Killinger, Ph.D. Wei Chen, Ph.D. Stanley R. Deans, Ph.D. Nicholas Djeu, Ph.D. Myung K. Kim, Ph.D. Date of Approval: October 15, 2007 Keywords: laser induced fluorescence spect roscopy, transient saturation fluorescence spectroscopy, fluorescence lifetime, bacteria spore detection, excitatin emission matrix spectra Copyright 2007 Anali Makoui

PAGE 2

DEDICATIONS To my parents To my brother To those whom I love and love me back

PAGE 3

ACKNOWLEDGEMENTS First and foremost I thank my advisor, Dr. Killinger, for not only having guided me and taught me during these years, but for going out of his way to help me in these last few months. Your help is greatly appr eciated and I will not forget your care and patience. To Dr. Chen, Dr. Deans, Dr. Djeu, and Dr Kim, I extend my deepest gratitude for serving in my committee, for your scientific ad vice and for your patience in the set up of this defense date. To Dr. Stefanakos for chairing the examination, thank you. I thank Yvette Mattley from Ocean Optic s Inc. for her help and time during the initial stages of this project. To Ev elyn, Mary-Ann, and Kimberly: thanks for making everything so much easier for me and for your caring. I would especially like to thank our late office manager, Ms. Sue Wolfe, who made a family out of the department. Thanks to Phil and to Sam for their technica l and engineering help during all of my time in this department. I also would like to thank Leo for his he lp with the initial computer program in the trial period of the last part of the project; Feiran with his help in th e translation of an article into English; and Anna for her help in the translat ion of another article into English. Thanks to my labmates Anna Avis and Dennis for their friendship. To my friends V, and Ed: thank you fo r your help and your friendship over the years. Chris, thank you for providing me with tremendous encouragement, and some much needed advice during this process. Fi nally, thanks to my family for your ever present support.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES vii LIST OF FIGURES ix ABSTRACT xxi CHAPTER 1. INTRODUCTION 1 CHAPTER 2. BACKGROUND INFORMATI ON ON OPTICAL ABSORPTION AND FLUORESCENCE OF BACTERIA SPORES (DPA) AND TERBIUM IONS 5 2.1 Absorption, Scattering, and Fluorescence of optical and UV light 5 2.1.1 Rayleigh and Mie Scattering of Light 6 2.1.2 Raman Scattering 6 2.1.3 Absorption of light 7 2.1.4 Emission of Optical Radiation and Energy Relaxation 8 2.1.4.1 Vibrational Relaxation 10 2.1.4.2 Internal Conversion 10 2.1.4.3 Fluorescence 10 2.1.4.3.1 Fluorescence Lifetimes 13 2.1.4.4 Intersystem Crossing 14 2.1.4.5 Phosphorescence 14 2.2 Background on Bacterial Cells and Spores 15

PAGE 5

ii 2.2.1 DPA Coating Surrounding Endospores 16 2.2.2 Absorption and Fluorescence Spectra of DPA 20 2.3 Use of Rare Earth Ions (Tb) to Enhance Fluorescence Emission 24 CHAPTER 3. INITIAL ENHANCEMENT OF Tb-DPA FLUORESCENCE USING A PULSED UV LASER 36 3.1 Tb-DPA Spectrofluorometer System Using Conventional Lamp Source 36 3.1.1 Modified Spectrofluorometer System Using Pulsed UV Laser 38 3.1.1.1 4 th Harmonic Nd:YAG Microchip Laser 38 3.2 Initial Fluorescence Measurements 42 3.2.1 Fluorescence of Terbium Ion 44 3.2.2 Fluorescence of the DPA-Tb Complex 44 3.2.3 Fluorescence of the Mixtures of Bacteria Doped With Tb 47 CHAPTER 4. DETAILED ABSORBANCE MEASUREMENTS OF DPA, TbCl 3 AND THE DPA-TbCl 3 COMPLEX 49 4.1 Experimental Set Up for detailed Absorbance Measurements 49 4.1.1 Quartz and Plastic Cuvettes 52 4.2 Absorbance Measurement Procedure 57 4.3 TbCl 3 and DPA Solution Preparation 58 4.4 General Absorbance Spectra of DPA and TbCl 3 and Their Mixtures 60 4.4.1 Measurements of Absorbances and Absorption Cross Section for Individual DPA, and TbCl 3 Solutions 60 4.4.2 Absorption Cross Section of DPA and TbCl 3 Individually 66 4.4.3 Absorbances of Mixtures of DPA and TbCl 3 69

PAGE 6

iii 4.4.4 Absorption Cross Section of the DPA-TbCl 3 Complex 74 4.5 Comparison of Absorption and Fluor escence Spectrum of DPA, Tb and the DPA-Tb Complex 83 CHAPTER 5. UV EXCITATION EMISSION MA TRIX (EEM) SPECTRA OF TERBIUM CHLORIDE AND DIPICOLINIC ACID 84 5.1 Experimental Set Up for the Excitation Emission Matrix Measurements 84 5.1.1 Tunable Laser Source 86 5.1.2 Spectrometer 89 5.2 Background Subtraction and Normalization Procedures 91 5.3 Solution Preparation 99 5.4 Experimental Results: Measured EEM Spectra for DPA, TbCl 3 and Mixed Solutions of Both 100 5.4.1 EEM of TbCl 3 in Solutions 100 5.4.2 EEM of DPA in Solutions 103 5.4.3 EEM of Mixtures of DPA and TbCl 3 in Solution 108 5.5 EEM Spectra Conclusions 113 CHAPTER 6. FLUORESCENCE LIFETIMES OF SOLUTIONS OF DPA AND TbCl 3 WHEN EXCITED WITH A LOW PR F PULSED TUNABLE DYE LASER 114 6.1 Experimental Set-Up for the Lifetime M easurements Using the Dye Laser 114 6.1.1 Monochromator 117 6.1.2 PMT Module 119 6.1.3 Fast Silicon Detector 122 6.2 Experimental Procedure 122

PAGE 7

iv 6.3 Solution Preparation 124 6.4 Lifetime Data Using Low PRF Dye Laser 125 6.4.1 Measured Fluorescence Lifetime of Mixtures of DPA with TbCl 3 in Different Solvents 125 6.4.2 Measured Fluorescence Lifetime of TbCl 3 in Distilled Water 132 6.5 Lifetime Analysis 132 CHAPTER 7. FLUORESCENCE LIFETIMES OF SOLUTIONS OF DPA AND TbCl 3 WHEN EXCITED WITH A HIGH PRF AND MODULATED PULSED UV LASER 137 7.1 Experimental Set-Up for the Lifetime Measurements Using the Microchip Laser 137 7.1.1 Optical Chopper 139 7.2 Effect of Optical Chopper on Excita tion and Fluorescence 139 7.2.1 Finite Beam Size and Geometrical Effect of Chopper Wheel on Incident Beam 141 7.2.2 Effect of Chopper Speed and Laser PRF on the DPA-TbCl 3 Fluorescence Signal 144 7.3 Experimental Measurements of DPA-TbCl 3 Lifetimes Using Chopper Technique 147 7.3.1 Lifetime Dependence on DPA and TbCl 3 Concentrations 155 CHAPTER 8. NEW FLUORESCENCE LIF ETIME MEASUREMENT TECHNIQUE USING TRANSIENT FLUORESCENCE SPECTROSCOPY (TFS) 159 8.1 Experimental Observation of the Effect of Chopper Modulation on Approach to Fluorescence Saturation 160 8.2 Theoretical Modeling of DPA and TbCl 3 Energy Level Rate Equations and Influence on Fluorescence Inte nsity 167

PAGE 8

v 8.2.1 Energy Level Model for DPA-TbCl 3 Fluorescence 168 8.3 Effect of Varying the Simulations Parameters 173 8.3.1 Detailed Study of Population Levels at Changing Chopper Frequencies 178 8.3.2 Detailed Study of Population Levels at Changing Incident Intensity 180 8.3.3 Effect of the Transition Rate 30 187 8.3.4 Effect of the Transition Rate 12 190 8.3.5 Effect of Changing th e Absorption Cross Section 192 8.4 More Complex 5-Level Rate Equation Model 194 8.4.1 The Up-conversion Model (Frster Resonance Energy Transfer) 198 CHAPTER 9. FLUORESCENCE INTENSITYS AMPLITUDE AND ITS APPROACH TO SATURATION 205 9.1 Fluorescence Amplitudes 205 9.2 Fluorescence Amplitudes: Comparisons of Simulations and Experiments 207 9.3 Comparison of TFS to Relate d Fluorescence and Spectroscopy Techniques 213 CHAPTER 10. MAXIMUM FLUORESCENCE INTENSITIES FOR DIFFERENT SOLVENTS AND SOLUTIONS OF DPA AND TbCl 3 215 10.1 Trends in Fluorescence for [50 M] Solutions 215 10.2 Patterns for [5 M] to [25 M] Solutions 219 10.3 Estimate of DPA Concentration in Spores 223 CHAPTER 11. CONCLUSIONS AND FUTURE WORK 224

PAGE 9

vi REFERENCES 226 APPENDIX A ABSORBANCE SPECTRA 234 APPENDIX B DETAILS OF LIFETIME CALCULATIONS 245 APPENDIX C FLUORESCENCE SIGNALS USING ASYMMETRIC EXCITATION WAVEFORMS 254 APPENDIX D NUMERICAL EVALUATION OF THE RATE EQUATIONS FOR 4-LEVEL MO DEL AND FOR ASYMMETRIC EXCITATION 258 ABOUT THE AUTHOR End Page

PAGE 10

vii LIST OF TABLES Table 2.1 Experimental energy levels of Tb 3+ :LaF 3 31 Table 3.1 Laser parameters of the 266 nm high PRF microchip laser (JDS Uniphase, Model NU-10110-100). 41 Table 4.1 ST-2000 spectrometers parameters. 55 Table 4.2 List of different concentr ation of solutions used in the experiments where the absorbances and the absorption cross sections of DPA and TbCl 3 were measured separately. 59 Table 5.1 Specifications for the tunable laser (Laser Science Inc.) tuned to 266 nm. 90 Table 5.2 USB-2000 spectrometers specifications (Ocean Optics Inc.). 92 Table 5.3 CCD detectors specifications. 96 Table 6.1 Specifications for the Lecroy oscilloscope (Waverunner LT344). 118 Table 6.2 Specifications for the Hamamatsu PMT. 121 Table 6.3 Measured Fluorescence li fetimes of solutions of [50 M] DPA with [50 M] TbCl 3 in the indicated solvents when excited with the dye laser, exct. = 266 nm. 130 Table 6.4 Lifetimes of TbCl 3 in distilled water. The lifetimes that we were unable to obtain are indicated with a dashed entry. 136 Table 7.1 Table of lifetim es of solutions of [50 M] DPA with [50 M] TbCl 3 in the indicated solvents when excited with the modulated microchip laser, exct. = 266 nm. 154 Table 8.1 Typical or default values used for the simulation parameters 174 Table 8.2 Typical or default values used for the simulation parameters for the 5-level quenching model. 197

PAGE 11

viii Table 8.3 Typical or default values used for the simulation parameters for the 5-level upconversion model. 202

PAGE 12

ix LIST OF FIGURES Figure 2.1 Jablonski energy level diagram of competing excitation and relaxation processes in a large molecule (adapted from Bernarth 7 ). 9 Figure 2.2 Stages of spore formation (from D. Lim 22 ) 17 Figure 2.3 a) Electron microscope cross-sectio n of a spore of Bacillus subtilis. The spore is 1.2 m across; b) scanning electron microscope image of Bacillus subtilis bacter ia cells. The cell is about 2.1 m in length (from D. Lim 22 ) 18 Figure 2.4 Structure of dipicolinic ion (DPA 2), dipicolinic acid (H 2 DPA), and calci um dipicolinate (Ca(DPA )) (from D. L. Rosen 25 ) 19 Figure 2.5 UV absorption spectrum of an aqueous suspension of intact Bacillus subtilis spores (from E. Ghiamati 28 ). 21 Figure 2.6 Absorption spectrum of [100 M] DPA in aqueous solution (from F. Peral 34 ). 22 Figure 2.7 Fluorescence emission spectra of 10 M DPA in water at pH 2.0 after 0, 5, 15, 30, and 60 minutes of UV irradiation ( ex = 300 nm), each graph corresponding respectively from the lowest fluorescent signal to the highest fluo rescing signal. (from R. Nudelman 33 ). 23 Figure 2.8 Periodic Table: Lanthanides (from Solid State Physics 39 ). 25 Figure 2.9 Radial part of th e hydrogenic wave functions for the 4f, 5d, and 6s orbitals of cerium (from Lanthanides and Actinides 40 ). 26 Figure 2.10 Electronic configuration of th e Terbium atom. 27 Figure 2.11 Energy levels of th e trivalent terbium ion (Tb 3+ ) (from G. H. Dieke 41 ). 29 Figure 2.12 Detailed diagram of selected en ergy levels of the trivalent terbium ion (Tb 3+ ) (Data points obtained from W.T. Carnall 42 ). 30

PAGE 13

Figure 2.13 Optical absorption of a solution of Tb 3+ (aquo) as a function of wavenumber (bottom axis) and wavelength (nm) (top axis) (from Carnall 43 ). 32 Figure 2.14 Structure of Tb(dpa) + and complexes with water molecules (from D. Rosen -2Tb(dpa) 25 ). 33 Figure 2.15 Fluorescence emission spectrum of 0.08-mL B. subtilis suspension in 200 M TbCl 3 both before (-) and after (s) filtering, for 200 M TbCl 3 blank (-), and for scaled 50 nM [Tb(DPA) 3 ] 3(-.-), in [50mM] TRIS buffer solution ( exc = 270nm) (from D.L. Rosen 51 ). 35 Figure 3.1 Schematic of commercial xenon flash lamp excitation fluorometer used to detect bacteria spores (Endospore Detection System, Ocean Optics, Inc.). 37 Figure 3.2 Schematic of laser fluorometer system using a high PRF 266nm (or 355nm) laser for excitation. 39 Figure 3.3 Photograph fluorometer system that could use either the 4 th harmonic Nd:YAG laser or the pulsed xenon flash lamp for excitation. 40 Figure 3.4 Schematic of 4 th harmonic Nd:YAG microchip laser (JDS Uniphase, Model NU-10110-100). 43 Figure 3.5 LIF emission spectrum of [100M] TbCl 3 in water, exc = 266nm. (Tb 3+ s electronic transitions labeled to corresponding spectral peaks.) 45 Figure 3.6 Fluorescence intensity as a function of wavelength of a solution of [50M] DPA with [50M] TbCl 3 in [0.01M] HCl, when excited with three different sources: 266nm laser, 355nm laser and a conventional pulsed xenon flash lamp. 46 Figure 3.7 Fluorescence intensity as a function of wavelength of a solution of Bacillus globigii spores with [50M] TbCl 3 in [0.01M] HCl, when excited with three different sources: 266nm laser, 355nm laser and a conventional pulsed xenon flash lamp. 48 Figure 4.1 Schematic of experimental set-up for absorbance measurements. 50 Figure 4.2 CHEM-2000s CW light source output spectrum over the range of 200 nm to 900 nm; drop-off near 900 nm due to drop-off in measuring grating efficiency. 51 x

PAGE 14

xi Figure 4.3 ST-2000 spectrometers internal sc hematic for one optical channel. 53 Figure 4.4 Relative efficiency as a function of wavelength of the grating used in the ST-2000 spectrometer. 54 Figure 4.5 Measured transmission spectra of a quartz cuvette and a "UV transparent" plastic cuvette. 56 Figure 4.6 Absorbance spectra of [15M] DPA measured in di fferent solutions. Note: The background spectrum of each solvent, which was not subtracted, has been plotted along in a fainter dotted line. 61 Figure 4.7 Absorbance spectra of [15M] TbCl 3 measured in different solutions. Note: The background spectrum of each solvent, which was not subtracted, has been plotted along in a fainter dotted line. 62 Figure 4.8 Enlarged view of the absorbance spectra of [15M] DPA in distilled water; water background has been subtracted. 64 Figure 4.9 Measured absorbance of DPA at 266 nm as a function of DPA concentration. The vertical line in side the markers at a concentration of around 15 x 10 15 molecules per cm 3 corresponds to an error of about 0.01 in the absorbance scale. E ach data point is the average of 100 runs. 65 Figure 4.10 Measured absorbance of TbCl 3 at 266nm as a function of TbCl 3 concentration. The vertical line inside the markers at a concentration of around 15 x 10 15 molecules per cm 3 corresponds to an error of about 0.01 in the absorbance scale. Each data point is the average of 100 runs. 67 Figure 4.11 Absorption cro ss section of DPA and TbCl 3 at 266nm. The vertical line inside the markers at a concentration of around 15 x 10 15 molecules per cm 3 corresponds to an error of about 0.15 x 10 -17 cm 2 /Molecule in the absorption cross section scale ( 5% error). 68 Figure 4.12 Absolute absorban ce spectra of solutions of [5M] DPA measured with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in distilled water. 70

PAGE 15

xii Figure 4.13 Enlarged view of the absolute absorbance spectra of solutions shown in Fig. 4.12. 72 Figure 4.14 Enlarged view of absorbance spectra of [15M] DPA with different concentrations of TbCl 3 in distilled water; water background has been subtracted. This graph corresponds to Figure 4.12 (c). 73 Figure 4.15 Measured absorbance as a function of DPA and TbCl 3 concentration in distilled water at a waveleng th of 266 nm. Water background spectra subtracted. 75 Figure 4.16 Measured absorbance as a function of DPA and TbCl 3 concentration in [0.01M] HCl at a wavelength of 266 nm. HCl background spectra subtracted. 76 Figure 4.17 Measured absorbance as a function of DPA and TbCl 3 concentration in [1M] SAB at a wavelength of 266 nm. SAB background spectra subtracted. 77 Figure 4.18 Absorption cross section values at 266 nm of mixtures of DPA and TbCl 3 at varying concentrations, in distilled water. 78 Figure 4.19 Absorption cross section values at 266 nm of mixtures of DPA and TbCl 3 at varying concentrations, in [0.01M] HCl. 79 Figure 4.20 Absorption cross section values at 266 nm of mixtures of DPA and TbCl 3 at varying concentrations, in [1M] SAB. 80 Figure 4.21 Average absorption cross section of the terbium doped di picolinic acid as a function of the terbium ions co ncentration, (a), and as a function of DPAs concentration, (b), in different solvents at exct = 266 nm. 82 Figure 5.1 Schematic of set up used for laboratory excitation emission (EEM) spectroscopy measurements. 85 Figure 5.2 Nitrogen pumped, frequency doubl ed UV tunable dye laser (Laser Science; Models VS L-337ND, DLMS-220 and UVS-300). 87 Figure 5.3 Tuning curves for Coumarin 440, Coumarin 481, and Coumarin 500 dyes that were used for the EEM experiments. (Graph obtained from http://www.exciton.com/ ). 88 Figure 5.4 USB-2000 gratings relative efficiency as a function of wavelength (600lines/mm blazed at 500 nm). 93

PAGE 16

xiii Figure 5.5 Attenuation curve for 600m diameter UV-VIS fiber 94 Figure 5.6 Relative sensitivity of 2048-element Sony silicon CCD array (Sony ILX511). Note: The relative sensitivity data for wavelengths below 400 nm was not available. 95 Figure 5.7 Background spectra (noise) of th e USB-2000 spectrometer with the excitation source blocked. 97 Figure 5.8 Dye laser power output normalized to the detectors responsivity. 98 Figure 5.9 EEM spectra of distilled water. 101 Figure 5.10 EEM spectra of [100M] TbCl 3 in distilled water. 102 Figure 5.11 EEM spectra of [100M] TbCl 3 in [1M] HCl. 104 Figure 5.12 EEM Spectra of [100M] TbCl 3 in [0.01 M] HCl. 105 Figure 5.13 EEM spectra of [100M] TbCl 3 in [1M] SAB pH 5.6 106 Figure 5.14 EEM spectra of [100M] DPA in distilled water. 107 Figure 5.15 EEM spectra of [50] DPA [50M] TbCl 3 in distilled water. 109 Figure 5.16 EEM spectra of [25M] DPA in [100M] TbCl 3 in [0.01 M] HCl 110 Figure 5.17 EEM spectra of [25M] DPA in [100M] TbCl 3 in [1M] HCL. 111 Figure 5.18 EEM Spectra of [25mM] DPA on [100mM] TbCl 3 in [1M] SAB. 112 Figure 6.1 Schematic of the lifetime measurement experimental apparatus using the low PRF tunable dye laser. 115 Figure 6.2 Photograph of the complete lif etime measurement experimental setup. Note: the dye laser and the 4 th Harmonic Nd:YAG laser were setup such that either experiment could be carried out without needing major rearrangements. This photogr aph shows the complete set up, the second half of which (i.e. the one that includes the 4 th Harmonic Nd:YAG) is described in the following chapter. 116 Figure 6.3 Grating spectromete rs (CVI Digikrom 240) grat ing efficiency. 118

PAGE 17

xiv Figure 6.4 Gain as a function of control voltage (a) and sensitivity as a function of wavelength (b) for the Hamamatsu PMT (Model # H6780-03). 120 Figure 6.5 Plot of responsivity as a function of wavelength of the fast silicon photodetector (New Focus, Model # 1621) used as a trigger. 123 Figure 6.6 (a) Plot of fluorescence intensity as a function of time, and (b) semilog plot of the fluorescence intensity as a function of time of a solution of [50 M] DPA with [50 M] TbCl 3 in distilled water. 126 Figure 6.7 (a) Plot of fluorescence intensity as a function of time, and (b) semilog plot of the fluorescence intensity as a function of time, of a Solution of [50 M] DPA with [50 M] TbCl 3 in [0.01M] HCl. 128 Figure 6.8 (a) Plot of fluorescence intensity as a function of time, and (b) semilog plot of the fluorescence intensity as a function of time, of a solution of [50 M] DPA with [50 M] TbCl 3 in [1M] SAB (pH 5.6). 129 Figure 6.9 (a) Fluorescence intensity as a f unction of time, and (b) semi-log plot of fluorescence intensity as a function of time of a solution of [50M] TbCl 3 in distilled water. 133 Figure 7.1 Schematic of lifetime measurem ents experimental setup using the high PRF pulsed laser and the optical chopper. 138 Figure 7.2 Optical choppers blades used to modulate the laser beam. The blade on the left hand side of the picture was used for the experiment described in this chapter. 140 Figure 7.3 Diagram of Gaussian laser beam (TEM 00 ) passing through a moving chopper blade. 142 Figure 7.4 Plot of Gaussian distributions with different vari ances. We have used the plot with a variance of 1 to represent the TEM 00 laser beam from microchip laser before going thr ough the choppers blade, i.e. the original beam. 143 Figure 7.5 Cumulative distribution function (CDF) for a Gaussi an distribution. 143 Figure 7.6 Plot of the 1 mm diameter laser beams intensity after interacting with the choppers slits. Also shown in the upper part is the turning on and off of the chopper blades window. Note the similarity between the build up and the fa ll down of this graph as compared to the CDF in Figure 7.5. 145

PAGE 18

xv Figure 7.7 Fluorescence in tensity as a function of time when using the high PRF microchip laser with the chopper turned off. 146 Figure 7.8 Fluorescence signal of sample when excited with a modulated incident beam, i.e. laser beam that has passed through the optic al choppers blade. 148 Figure 7.9 Plot of amplitude as a function of time of the high PRF 266nm 4 th harmonic Nd:YAG lasers output (a), as compared to the choppers TTL signal at 50 Hz (b), and the samples fluorescence (c). 149 Figure 7.10 (a) Fluorescence intensity as a f unction of time, and (b) semi-log plot of the normalized fluorescence decay lines from [50M] DPA with [50M] TbCl 3 in distilled water (f chopper = 50Hz). 150 Figure 7.11 (a) Fluorescence intensity as a f unction of time, and (b) semi-log plot of the normalized fluorescence decay lines from [50M] DPA with [50M] TbCl 3 in [0.01M] HCl (f chopper = 50Hz). 152 Figure 7.12 (a) Fluorescence intensity as a function of time, and (b) semi-log plot of the normalized fluorescence decay lines from [50M] DPA with [50M] TbCl 3 in [1M] SAB (f chopper = 50Hz). 153 Figure 7.13 Plots of lifetime as a function of DPA and TbCl 3 concentration in distilled water. A sample error bar is shown in each of the figures at around the [25M] DPA with [5M] TbCl 3 156 Figure 7.14 Plots of lifetime as a function of DPA and TbCl 3 concentration in [0.01M] HCl. A sample error bar is shown in each of the figures at around the [25M] DPA with [5M] TbCl 3 157 Figure 7.15 Plots of lifetime as a function of DPA and TbCl 3 concentration in [1M] SAB. A sample error bar is shown in each of the figures at around the [25M] DPA with [5M] TbCl 3 158 Figure 8.1 Experimentally measured fluores cence intensity as a function of time of a sample of [50 M] DPA with [50 M] TbCl 3 in distilled water when excited with the chopper-modulated-microchip laser, for the chopper operating at four different frequencies. 161

PAGE 19

xvi Figure 8.2 (a) Original 30-slit chopper blade on the left, and modified 15 slit blade on the right; (b) original 6-slit chopper blade on the left, and 30-slit and 6-slit blades superi mposed on the right. Note: only the outer slits in each blade were used for modulation purposes in these experiments. 163 Figure 8.3 Chopper periods, or TTL signals for the 15-slit modified blade showing 1-on and 3-off (a), and for the superimposed 30-slit and 6-slit blades showing a group of 3 on-off followed by 5-off (b); these correspond to Fig. 8.2 (a), and (b), respectively. 164 Figure 8.4 Fluorescence signal as a function of time for a DPA-Tb solution when excited with the chopper-modulated -microchip laser, for the chopper operating at two different frequencie s, with a modified blade with 15 slits (1 run). 165 Figure 8.5 Fluorescence signal as a function of time for a DPA-Tb solution when excited with the chopper-modulated -microchip laser, for the chopper operating at a frequency of 50 Hz (a verage of 100 runs). 166 Figure 8.6 Comparison of the energy levels of terbium ion 42 with the lowest lying triplet state of the DPA molecule, as measured by Latva 35 and by Lima 36 169 Figure 8.7 Simple theoretical energy level m odel representing the energy transfer between the DPA molecule and the terbium ion. 171 Figure 8.8 Simulations with all the pa rameters default values, except for 23 which is changing as indicated in e ach graph. Each graph consists of 3 plots for chopper speeds of 50 Hz, 500 Hz, and 1000 Hz. 175 Figure 8.9 Experimentally obtained fluorescen ce intensity of an aqueous solution containing DPA-Tb complex (a), as compared to the simulation signals obtained using the default values (Table 8.1) (b), for the case where the chopper speed was 50 Hz, 500 Hz, and 1000 Hz. 177 Figure 8.10 Population densities for all of the energy levels as a function of time, as calculated by the simulation, using the default values for the parameters (f Chopper 50 Hz). 179 Figure 8.11 Population densities for all of the energy levels as a function of time, as calculated by the simulation, using the default values for the parameters (f Chopper 500 Hz). 181

PAGE 20

xvii Figure 8.12 Population densities for all of the energy levels as a function of time, as calculated by the simulation, using the default values for the parameters (f Chopper 1000 Hz). 182 Figure 8.13 Simulations depicting the populatio n density of each energy level for different intensities, where I 0 is the default value 1.113 x 10 22 photons/cm 2 sec for f chopper = 50 Hz. (Section A.) 183 Figure 8.13 (Continued Section B) 184 Figure 8.13 (Continued Section C) 185 Figure 8.14 Simulations with all the parameters default values, except for 30 which is at three different values, two of which are extreme values. 188 Figure 8.15 Simulation with all the pa rameters default values, except for 30 = 0.0, over a larger time scale than shown in Fig. 8.16 (a). 189 Figure 8.16 Simulations with all the parameters default values, except for 12 which is at four different values, three of which are extreme values. 191 Figure 8.17 Simulations with all the parameters default values, except for which is at its largest possible value (a), at its default value (b), and at its smallest value (c). 193 Figure 8.18 Energy level diagram for the DP A-Tb complex including quenching rates. N 1 and N 2 depict the first singlet an tr iplet states of the DPA molecule, respectively. N 3 N 4 depict the 5 D 4 the 7 F series of the Tb 3+ ion, respectively. 195 Figure 8.19 Population density of the fluorescing level, N 3 of the 5-level quenching model, as a function of time using the typical values given in table 8.2. 199 Figure 8.20 Energy level diagram for the DPA-Tb complex including upconversion rates. N 1 and N 2 depict the first singlet an triplet states of the DPA molecule, respectively. N 3 N 4 and N up depict the 5 D 4 the 7 F series, and a second excited state of the Tb 3+ ion, respectively. 200 Figure 8.21 Population density of the fluorescing level, N 3 of the 5-level upconversion model, as a functi on of time using the typical values given in table 8.3. 203 Figure 9.1 Fluorescence signals amplitudes at 50Hz and 500Hz. 206

PAGE 21

xviii Figure 9.2 Fluorescence signals amplitude fo r the 543nm line of a solution of [50M] DPA with [50M] TbCl 3 in distilled water as a function of the choppers frequency. 208 Figure 9.3 Fluorescence signals amplitude fo r the 543nm line of a solution of [50M] DPA with [50M] TbCl 3 in distilled water as a function of the number of photons that passed through one of the choppers open windows. This number is inve rsely proportional to the choppers frequency. Note: the x-axis has been drawn to only 10 x 10 23 for clarity purposes. 209 Figure 9.4 Fluorescence intensity as a function of chopper frequency, f chopper for several different lifetime values. 211 Figure 9.5 Comparison of several numerically obtained fluorescence (i.e. population density) signals amplitude as a function of the number of cumulative excitation photons th at passed through one of the choppers open windows with respec t to the experimentally obtained fluorescence signals amplitude. Each graph has been normalized to its maximum. Note: the x-axis has been drawn to only 10 x 10 23 for clarity purposes. 212 Figure 10.1 Fluorescence maxima from solution of [50 M] DPA with [50 M] TbCl 3 in different solvents when excited with the dye laser. 216 Figure 10.2 Fluorescence maxima from solution of [50 M] DPA with [50 M] TbCl 3 in different solvents when excited with the microchip laser. 218 Figure 10.3 Points of maximum fluorescence in tensity for solution with different [DPA] and [TbCl 3 ] in [1M] SAB when excited with the combination of the microchip laser and optical chopper (f = 50 Hz). 220 Figure 10.4 Points of maximum fluorescence inte nsity for solutions with different [DPA] and [TbCl 3 ] in distilled water when excited with the combination of the microchip laser and optical chopper (f = 50 Hz). 221 Figure 10.5 Points of maximum fluorescence in tensity for solution with different [DPA] and [TbCl 3 ] in [0.01M] HCl when excited with the combination of the microchip laser and optical chopper (f = 50 Hz). 222 Figure A.1 Measured absorbance of different concentrations of DPA in distilled water; water background not subtracted. 235

PAGE 22

xix Figure A.2 Measured absorbance of different concentrations of DPA in [0.01M] HCl; HCl background not subtracted. 236 Figure A.3 Measured absorbance of differe nt concentrations of DPA in [1M] SAB; SAB background not subtracted. 237 Figure A.4 Measured absorbance of different concentrations of TbCl 3 in distilled water; water background not subtracted. 238 Figure A.5 Measured absorbance of different concentrations of TbCl 3 in [0.01M] HCl; HCl background not subtracted. 239 Figure A.6 Measured absorbance of different concentrations of TbCl 3 in [1M] SAB; SAB background not subtracted. 240 Figure A.7 Measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [0.01M] HCl. 241 Figure A.8 Measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [1M] SAB. 242 Figure A.9 Enlarged view of the measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [0.01M] HCl. 243 Figure A.10 Enlarged view of the measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [1M] SAB. 244 Figure B.1 Fluorescence intensity as a function of time without any corrections: i.e. the raw signal. 246

PAGE 23

xx Figure B.2 The raw signal with the offset (at 6ms) subtracted. 247 Figure B.3 The normalized, i.e. the raw signal with the offset subtracted, and the effects of the laser intensity, the gratings efficiency, and the PMTs sensitivity accounted for. 248 Figure B.4 Natural logarithm of the normali zed fluorescence intensity as a unction of time for a solution of [50M] DPA with [50M] TbCl 3 in distilled water. 249 Figure B.5 Enlarged view of plots of the na tural logarithm of the normalized fluorescence intensity as a function of time for a solution of [50M] DPA with [50M] TbCl 3 in distilled water. 250 Figure B.6 Enlarged view of the points used in the calculations to find the slope of the 488 nm line. 251 Figure B.7 Example of points used to find the slope of the line, the negative inverse of which is equal to the fluorescence lifetime of the samples 488 nm line. Note: the y-axis has been labeled arbitrarily, but consistently with other graphs of the same type for this sample. 252 Figure C.1 Fluorescence signals obtained usi ng the symmetric 30 slit blade at the indicated chopper frequencies. 255 Figure C.2 Fluorescence signals obtained using the asymmetric 15 slit blade at the indicated chopper frequencies. 256 Figure C.3 Fluorescence signals obtained using the asymmetric superimposed blades at the indicated chopper frequencies. 257

PAGE 24

xxi TRANSIENT FLUORESCENCE SPECTROSCOPY AND LASER INDUCED FLUORE SCENCE LIFETIMES OF TERBIUM DOPED DIPICOLINIC ACID Anali Makoui ABSTRACT We have investigated the use of deep UV laser induced fluorescence for the sensitive detection and spectroscopic lifetime studies of terbium doped dipico linic acid (DPA-Tb) and used this to study the optic al characteristics of DPA wh ich is a chemical surrounding most bacterial spores. Background abso rption spectra, fluorescence spectra, and Excitation Emission Matrix (EEM) spectra we re made of the DPA-Tb complex, using both fixed 266 nm wavelength and tunable ( 220 nm 280 nm) UV laser excitations. Of importance, the fluorescence lifetimes of the four main fluorescence peaks (488 nm, 543 nm, 581 nm, and 618 nm) of the DPA-Tb complex have been measured for the first time to our knowledge. The lifetimes of all the fluorescing lines have been measured as a function of DPA-Tb concentration, solvent pH, and solvent compos ition, including that for the weakest fluorescing line of DPA-Tb at 618 nm. In addition, a new spectroscopic lifetime measurement technique, which we call Transient Fluorescence Spectroscopy,

PAGE 25

xxii was developed. In this technique, a w eak, quasi-CW, amplitude modulated UV laser (8.5 kHz) was used to measure the lifetimes of the fluorescence lines, and yields insight into energy transfer and excitati on lifetimes within the system. This technique is especially useful when a high power laser is no t either available or not suitable. In the latter case, this would be when a high power pulsed deep-UV laser could produce bleaching or destruction of th e biological specimen. In addition, this technique simulated th e excitation and fluores cence emission of the DPA-Tb using a 4-level energy model, and solv ed the dynamic transient rate equations to predict the temporal behavior of the DPA-Tb emitted fluorescence. Excellent agreement between the experiments and the simulation were found. This technique has the potential to provide a more accurate value for the fluorescence lifetime values. In addition, with the use of asymmetric exci tation waveforms, the dynamic transient rate equation analysis may allow for detailed studi es of selected transfer mechanisms in a wide range of other spectroscopic applica tions including rare-ear th solid-state lasing materials and biological samples.

PAGE 26

1 CHAPTER 1. INTRODUCTION Optical fluorescence is a sensitive technique that is often used for the detection and identification of a wide range of chem ical compounds and biological species. The technique is often used for the detection of biological and chemical species, both in the laboratory and in the field, and offers enhanced sensitivity over other analytical techniques. Often, these measuremen ts are conducted using a commercial spectrofluorometer instrument that utili zes a conventional xenon light source and a scanning spectrometer or optical fluorescence emission channels. 1-4 Studies have shown, however, that considerab le improvement in sensitiv ity may be achieved through the use of a UV laser fo r the excitation source. 1 The studies presente d in this dissertation will quantify such an improvement for the detection of bacteria endospores and the development of a new spectroscopic technique fo r the determination of the lifetime of the fluorescence emission lines. Previously, a considerable amount of work has been conducted for the use of laser induced fluorescence for the det ection of biological specimens. 1-4 For example, the sensitive detection of several chemical and biological species using dye based reagents that attach to the bacterium or vi rus has been studied and reported. 4 Fluorescence of the dye was used for tracking and detecting the sp ecimen. Also, time-resolved fluorescence has been used to detect and discri minate between bioaerosol particles. 5 Along these lines, the pioneering work of Rosen indica ted that although the direct or natural

PAGE 27

2 fluorescence of dipicolinic acid (DPA) contai ned within a bacteria spore was weak, the rare-earth ion terbium (III) (Tb 3+ ) could be introduced into the specimens DPA complex, enhancing the fluorescence by se veral orders of magnitude. 6 Based upon these earlier results with Tb doped DPA, also referred to as Tb-DPA or DPA-Tb interchangeably from here on, we have investigated the use of deep-UV excitation for the laser-induced-fluorescence of the DPA-Tb complex, and the use of several new and unique spectroscopic tec hniques for the determination of the fluorescence lifetime of the DPA-Tb complex. In particular, our measurements yield, for the first time to our knowledge, the fluorescence lifetimes of the principal four emission lines of Tb doped DPA, and also show the va riation of these lifetimes as a function of solvent concentration and pH. In addition, we developed a new spectroscopic lifetime measurement technique, which we call Transie nt Fluorescence Spectro scopy. In this technique a detailed dynamic transient solu tion to the rate equation model for Tb doped DPA was made, and excellent agreement be tween the experiments and theory were found. This transient fluorescence technique allows for the use of a weak quasi-CW laser (8 KHz pulse repetition fr equency) that is also amplitude modulated to measure the lifetimes of selected fluorescence emission lin es when a high power laser is not either available or suitable. In the latter case, this would be when a high power pulsed deepUV laser could produce bleaching or dest ruction of the biological specimen. This dissertation research is reported in the following Chapters. Chapter 2 provides background information on basic absorp tion and fluorescence optical processes related to laser-induced-fluorescence, and basi c biological information on bacteria spores, such as Bacillus globigii, an Anthrax imitator, which produces a protective sheath of

PAGE 28

3 DPA when the anthrax goes into a dormant s pore state. In add ition, the absorption and fluorescence properties of DPA and Tb-DPA as measured previously by others are presented. Chapter 3 presents our initial experimental work on the enhancement of the Tb-DPA fluorescence using a pulsed UV la ser compared to that using a more conventional CW Xenon flash lamp fo r excitation as in a conventional spectrofluorometer. In Chapter 4 the detailed absorbance spectra and absorbance values of the DPA molecule in different solutions is presented as measured by using a UVvisible lamp and a conventional spectrometer for detection. The absorption spectra of the TbCl 3 molecule in different pH and buffer so lutions and of the mixture of the two (DPA and TbCl 3 ) in different solutions is also presented. Chapter 5 presents Excitation Emission Ma trix (EEM) spectra for mixtures of DPA and TbCl 3 in different solution backgrounds us ing a tunable (220 nm 280 nm) UV frequency-doubled dye laser for excitation. Some EEM spectra are also presented for DPA and TbCl 3 by themselves in solution and show the optimal excitation wavelengths for fluorescence emission of these mixtures. In Chapter 6 we present and discuss initial fluorescence lifetime measurements of mixtures of DPA and TbCl 3 in different solutions when excited with a low pulse-repetition-freq uency (PRF) UV dye laser. The lifetimes of some of the stronger fluorescence lines of the terbium ion by itself in solution are presented in this chapter but also the limitati ons due to the weak fluorescence of some of the lines in the low concentration solutions are shown. Chapter 7 presents a new experimental technique that is used to m easure the fluorescence lifetime of mixtures of DPA and TbCl 3 in different solutions. In this te chnique a high PRF (8 KHz) microchip laser in combination with a slow speed (50 Hz) optical chopper is used in the lifetime

PAGE 29

4 experiments. In this chapter we analyze th e same samples as were used in Chapter 6, and compare the fluorescence lifetime results. The fluorescence lifetimes as a function of concentration of DPA and TbCl 3 are also presented. In Chapter 8 a new technique for measuring the lifetime and transition rate s of the DPA-Tb fluo rescing complex is presented which involves the effect of the chopper modulation on the approach to saturation of the fluorescence emission. These experimental results are then compared to a transient numerical solution of the rate-e quations for a model of the DPA-Tb energylevels and the appropriate fluorescence and ener gy transfer rates. This new technique shows excellent agreement between experiment al measurements and that predicted from the rate-equation model. We call this te chnique Transient Fluorescence Spectroscopy (TFS). Chapter 9 presents further analysis of and comparison between the experimental fluorescence signals as a function of time and the approach to saturation of the fluorescence emission. In Chapter 10 the ma ximum fluorescence intensities for different solutions of DPA and TbCl 3 are presented. In Chapter 11 we present conclusions and possible future studies. Finally, the appendices show de tailed absorption spectra, examples of lifetime calculations, effects of asymmetrical excitation intensity waveforms, and the Matlab computer program written by the author for the so lution of the dynamic rate equation model of the DPA-Tb complex.

PAGE 30

5 CHAPTER 2. BACKGROUND INFORM ATION ON OPTICAL ABSORPTION AND FLUORESCENCE OF BACTERIA SPORES (DPA) AND TERBIUM IONS This chapter contains background info rmation regarding the absorption and emission of light by matter as reported by others. In particular, absorption and fluorescence spectra from bacteria spores, di picolinic acid (DPA), and terbium chloride. 2.1 Absorption, Scattering, and Fl uorescence of Optical and UV Light When light interacts with matter, it can be absorbed, scattered, and/or transmitted. The percent to which each of these phenom ena occurs is dependent on the optical properties of the material being illuminated, the frequency of light being used, and the overlap of the excitation li ght with the absorption spectrum of the material. In the case where the excitation light is absorbed by the sample, some the molecules or ions in the sample (the ones th at absorbed the energy) will become more energetic. That increase in energy is di splayed in different ways depending on the frequency of the excitation light. For example, nuclear spins will be flipped in a magnetic field if radio frequencies (3 MHz-3 GHz) are used, rotational transitions in molecules and electron spin flips occur if microwave frequencies (3 GHz 3,000 GHz) are used, frequencies in the infrared region (100 cm -1 13,000 cm -1 ) will induce vibrational transitions, and valence elect ron rearrangements in molecules will be stimulated if visible and ultraviolet (UV) frequencies (1000 nm-10 nm) are used. 7

PAGE 31

6 2.1.1 Rayleigh and Mie Scattering of Light The deviation of light by fine particles from the main direction of a beam is a form of scattering. 8 Depending upon the size of the part icle with respect to that of the lights wavelength, we can have different types of scatterings. 9,10 Rayleigh scattering refers to when the diameter of the partic le, d, is much smaller than the wavelength, of the incident light. This a pproximation assumes that the entire molecule is in a nearly constant or homogeneous electric field. Mie Scattering refers to the scattering of light by spherical particles whose size is comparable to the wavelength of the incident light. The angular distribution of the scattered light in tensity often exhibits strong scattering lobes or beams due to interference effects within the scattered beam, which depends upon the ratio of the diameter of the part icle and wavelength of the light. 9,10 2.1.2 Raman Scattering Raman scattering involves a t ype of scattering in which the scattered light has a different wavelength to that of the incident light. The fr equency difference corresponds to the energy difference between the rotati onal or vibrational en ergy levels of the molecule. If the scattering is of a longer wave length than that of the incident light it is referred to as Stokes Raman scattering, and if it is a shorter wavelength then it is referred to as Anti-Stokes Raman scattering. 9,10 This effect is differe nt from fluorescence in that the incident light does not coincide with th e samples absorption band and is usually not a resonance effect.

PAGE 32

2.1.3 Absorption of Light In optics, the process by which the energy of a photon is taken up by another entity is known as absorption. The photon is destroyed in the process. The absorbed energy can be lost by heat and (or) radiation. As an example, the energy of an incident beam can be absorbed by an atom, causing its electrons to make a transition between two electronic energy levels. A measure of the strength of the absorption process is the absorption cross section, with units of cm 2 /molecule. It is a measure of the probability of a collision between the incident photon and a number of the samples molecules, and represents the effective area, in units of cm 2 of the absorbing particle present to the incident beam. 8 The attenuation of the optical beam due to the absorption process is given by the Beer-Lambert law, 4 I = I o e N l (2.1) where N is the concentration of the molecules in molecules/cm 3 I o is the incident optical intensity (photons/cm 2 s) l is the optical path length in cm, and I is the optical intensity after traveling a distance l. The transmission, T, of the beam is given by I/ I o the absorption is 1-T, and the absorbance, A, is given by log 10 (T). Other useful optical measures of absorption often used by different scientific communities include the optical density, OD, or absorbance, A, which is given by A = log 10 (I o / I) = c l and ODln(T) N l (2.3) (2.2) 7

PAGE 33

where c is the concentration in moles per lite r (i.e. molar concentration, M), and is the molar extinction coefficient (M -1 cm -1 ), also known as the molar absorptivity. It is common to drop the 10 subscript in log 10 and write this as log. In the biological community it is also common to express concentration in units of Molar, M, which is defined as the number of moles per liter of a substance. Using ln(x) = 2.303 log(x), and combining Eqs. (2.2) and (2.3), one can relate to by 206.02310Ilog() () Io2.303cl (2.4) where 3 1000 3236.02310 ,moleculescmmolesNcliterlitercmmolecules 1mol (2.5) I-log() = ,Iocl (2.6) and, therefore 206.02310202.6110 2.303 (2.7) 2.1.4 Emission of Optical Radiation and Energy Relaxation Once energy has been absorbed, there are several ways the molecule (or atom) can return to its ground state. Figure 2.1 shows a schematic of the absorption and emission of radiation for a large molecule which also exhibits different spin states of the 8

PAGE 34

S0S1S2 T1T2 WhereS0isthegroundelectronicsingletstateS1isthefirstexcitedelectronicsingletstateS2isthesecondexcitedelectronicsingletstateT1isthefirstexcitedelectronictripletstateT2isthesecondexcitedelectronictripletstateAbsorptionFluorescencePhosphorescenceIntersystemCrossingVibrationalRelaxation RadiationlessprocessesProcessesinvolvingphotonsInternalConversion Figure 2.1 Jablonski energy level diagram of competing excitation and relaxation processes in a large molecule (adapted from Bernarth 7 ). 9

PAGE 35

10 molecule. As can be seen from this figure, there are often competing processes. Some of these relaxation paths are listed and discussed in the following sections. 2.1.4.1 Vibrational Relaxation Molecules have energy states that may not involve electronic tr ansitions. These states are related to the rotational and vibra tional motions of the molecules and lead to energy levels that are much more closely sp aced than electronic transitions, which are seen in atoms. 11 These vibrational levels presen t a pathway of nonrad iative relaxation for an excited molecule to return to the lowe st vibrational level within the same excited state. This energy is given up in the form of near-infrared quanta and/or kinetic energy imparted to surrounding molecules in the form of collisions. This process typically has a lifetime of about 10 -14 s to 10 -12 s. 7 2.1.4.2 Internal Conversion An excited molecule can non-radiatively de cay down to a lower electronic state of the same spin multiplicity, via vibrational relaxation. This process is known as internal conversion. 4 Internal conversion may occur when, in thermal equilibrium, the lowest lying vibrational levels of the excited stat e overlap with the highe st lying vibrational levels of the ground state. 2.1.4.3 Fluorescence The term fluorescence is used within the atomic spectroscopy field to describe an allowed electric dipole transition between energy levels of the atom which results in the

PAGE 36

emission of optical radiation. The term fluorescence is often used within the molecular spectroscopy field to designate spin allowed radiative transitions between the first excited singlet state and the ground state, where the electron in the excited orbital is paired (of opposite spin) to the second electron in the ground state orbital. 3,4,12,13 The emitted radiation is usually in the UV or visible part of the spectrum. Fluorescence emission rates are on the order of 10 8 per second, which gives a typical fluorescence lifetime of around 10 ns. 7 If the emitted photon has the same energy as the absorbed photon, it is known as resonance fluorescence. This type of fluorescence usually occurs in small, gaseous molecules. However, if the emitted photon has less energy than the absorbed photon, it is known as relaxed fluorescence; this type of fluorescence usually occurs in large, condensed phase molecules. 7 The molecule can decay down to its ground state via internal conversion or fluorescence depending upon the number of vibrational levels present, the energy gaps, and the relative strength of the optical transitions and relaxation transitions. A molecule with a high degree of freedom, and hence a high number of vibrational levels, has a higher chance of de-exciting by internal conversion, whereas more rigid molecules tend to decay through optical emission. 12,13 Under certain assumptions, i.e. where the fluorescence transition is produced by single frequency absorption, the fluorescence intensity may be given by 3,14 Fo4IIA (2.8) 11

PAGE 37

where I F is the emitted fluorescence intensity in W/cm 2 -ster, /4 is the fraction of the solid angle () seen by the collection system, I o is the incident intensity of the exciting source, A is the absorbance, and is the fluorescence quantum yield. The quantum yield is defined as the ratio of the number of photons emitted as fluorescence compared to the number of photons absorbed, as given by 15 []ffdqQ (2.9) where f is the rate of fluorescence emission, d is the sum of rate constants for the various de-excitation processes, and q is the sum of the product of rate coefficient, q and the concentration of the various quenchers present, [Q]. Substituting Eq. (2.2) and Eq. (2.9) into Eq. (2.8), the fluorescence equation can be written as Fo4IIc[]ffdqlQ (2.10) Fluorescence intensity can also be defined in terms of the number of atoms per unit volume, i.e. the population density, of the fluorescing or excited state energy level, N u as given by 14 IAFhuul N (2.11) where h is Plancks constant, is the radiation frequency (s -1 ), A ul is Einsteins coefficient for spontaneous emission (cm g -1 ), and N u is the number of atoms in the upper state per unit volume (cm -3 ). 12

PAGE 38

2.1.4.3.1 Fluorescence Lifetime When a sample is excited, a number of the molecules, N u are left in an excited electronic state. The average time these molecules remain in the excited state is known as the fluorescence lifetime, 4 The population density of the excited state decays exponentially according to 0uuul A tNNe (2.12) where is the initial population density in the upper level at the time the excitation pulse occurs and 0uN ul is the rate at which the population is being transferred from the upper to the lower level in units of 1/sec. 11 The inverse of the sum of all the possible radiative decay rates, u to lower-lying levels is equal to the lifetime of the upper energy level, u, as given by 1uuuiiA 1 (2.13) In the absence of nonradiative processes this is called the intrinsic or natural lifetime, n 4 The quantum yield, can be obtained by dividing the measured fluorescence lifetime, by the intrinsic lifetime, n 4 n (2.14) The total width of the spectral line emitted from a transition from an upper level, u, to a lower level, l, known as the natural line width, is equal to the total decay rate of the decaying system, ul 16 In turn, the total decay rate of a decaying system is equal to the addition of the individual decay rates of the upper level, u, and the lower level, l, as given by 11,17,18 13

PAGE 39

11ululul (2.15) From Eq. (2.15) it can be noted that the total decay rate depends on the lifetimes of both the initial and final states. In other words, the fluorescence lifetime is determined classically as the inverse of the normalized linewidth. 19,20 Normally, the lifetime of the lower state is infinite, so that 1/ l = 0. However, when l then ul is given as in Eq. (2.15). 2.1.4.4 Intersystem Crossing When an electron in an excited singlet state flips its spin, it leaves the molecule in an excited triplet state. 7 An example of such a transfer from the excited singlet state to the excited triplet state can occur, for example, via a molecular collision. The probability of intersystem crossing occurring in a molecule is proportional to the number of vibrational states that overlap between the excited singlet state and the triplet state. The terms singlet and triplet states are the two superposition wavefunctions of the spin for two electrons. Such states are common for molecules and multi-electron atoms. 2.1.4.5 Phosphorescence An optical transition or transfer from a higher lying energy level to a lower lying level of a different multiplicity is called phosphorescence. Because such transitions are spin forbidden, these types of radiative transitions have relatively long lifetimes, i.e. milliseconds to minutes. Phosphorescence transitions may be forbidden due to not being allowed for electric dipole transitions, but may be allowed for electric quadrupole 14

PAGE 40

15 transitions with a much lower probability of transition. Hence, the long lifetime of the radiative transition. 4 It should be noted that often the term phosphoresce nce is not used within the optical physics comm unity, but is assumed to be contained within the term fluorescence, although w ith a longer lifetime. 2.2 Background on Bacterial Cells and Spores Interest in bacterial spore detection is present in a number of different fields, the most common of which are the agricultural sector, and the military and defense sector. 21 The agricultural sectors interest in detecti on of bacterial spores is mainly due to the possibility of naturally infected bovine enteri ng the food chain. The military and defense sectors interest in bacterial spore detection is mainly related to threat of the use of biological agents in an attack. The bacterial spore threat most wild ly recognized by the public sector is, arguably, that of anthrax (i.e. Bacillus anthra cis). The disease caused by this spore is known by the same name. Anthrax can be caught in two ways: through the skin (i.e. cutaneous anthrax) and inhalati on (i.e. inhalation anthrax). A biological attack of aerosol anthrax would cause inhalation anthrax, which has a 90%-100% mortality rate. A lethal dose of anthrax is considered to be about 10,000 spores. 22 Under nutrient limiting conditions, certain types of bacterial cells have the ability to salvage the most critical components of th eir structure and store it in a resistant, dormant cell type called an endospore, also known as a bacteria spore, or simply spore. These endospores have the ability to then re turn to their original active state, as a bacterial cell, also known as vegetative cel l, when, and if, they encounter favorable

PAGE 41

16 growth conditions. 22 Figure 2.2 shows a diagram of th e typical life cycle of bacteria and stages of spore formation. Bacterial spores vary slightly in th eir size depending on their taxonomic grouping, but are mostly in the orde r of one micrometer and are composed of different coatings and sections as depicted in Fig.2.3 (a). For comparison purposes, the active bacteria of the genus B acilli, which are rod shaped with dimensions of about 1 m by 3-5 m, are shown in Fig. 2.3 (b). The Bacillus subtilis type of bacteria tends to form chains or aggregates of about 10 to 40 bacteria. 22 Usually, the active b acterial cell is made up of about 70% water. However, when the cell goes to an endospore state, water ma kes up approximately 15% of the bacterial spore. Endospores do not metabolize or re produce and can withstand drying. In fact they are 10,000 times more resist ant to heat and 100 times more resistant to UV rays than are vegetative cells, and can last for cen turies in an outsi de dry environment. 22,23 As a point of reference, the extinction coefficient of the Bacillus subtilis at an excitation wavelength of approximately 266 nm has been reported as 4.1 x 10 -8 cm 2 /spore. 24 2.2.1 DPA Coating Surrounding Endospores It is believed that the toughness of bacteria l spores is due, at least in part, to the chemical dipicolinic acid (also known as DP A, or pyridine-2,6-dica rboxylic acid), found in its cortex. This chemical can constitute from 5% up to 15% of the dry weight of a bacterial spore. 22,23 In the cortex of a bacterial spore, DPA is found to be ionically bound to calcium, forming thus, calcium dipicolinate 25 as shown in Fig. 2.4. In fluorescence based

PAGE 42

Figure 2.2 Stages of spore formation (from D. Lim 22 ). 17

PAGE 43

(a) (b) Figure 2.3 a) Electron microscope cross-section of a spore of Bacillus subtilis. The spore is 1.2 m across.(from http://www.berkeley.edu/news/media/releases/2003/02/10_spores.shtml ); b) Scanning electron microscope image of Bacillus subtilis bacteria cells. The cell is about 2.1 m in length (from D. Lim 22 ). 18

PAGE 44

Figure 2.4 Structure of dipicolinic ion (DPA 2), dipicolinic acid (H 2 DPA), and calcium dipicolinate (Ca(DPA)) (from D. L. Rosen 25 ; reprinted with permission from Applied Spectroscopy). 19

PAGE 45

20 bacteria spore detection systems it is common to extract the dipicolinic acid from the sample for better sensitivity. 26 Dipicolinic acid is present naturally only in bacterial spores. 27 2.2.2 Absorption and Fluorescence Spectra of DPA There has been some previous but lim ited work on the optical spectra of DPA. 28-34 A UV absorption spectrum from an aqueous su spension of Bacillus subtilis is shown 28 in Fig. 2.5. For comparison purposes, a gr aph of the absolute absorbance of [100M] DPA in aqueous solution as a functi on of wavelength is also shown 34 in Fig. 2.6. As can be seen in these figures, DPA has an absorption peak near 250 to 280 nm. Dipicolinic acids intrinsic fl uorescence is very weak a nd it has been believed to be masked by the fluorescence of other compone nts of the bacteria spores. However, some authors have detected, after minutes of UV irradiation, fluorescence from what is believed to be DPA which may have been ra diationally changed or damaged as the UV intensity is increased in time. 29,33 This is shown in Fig 2.7 which shows measured changes in the DPA fluorescence near 400 nm being increased as the exposure to 300 nm UV excitation is increased. 33 The energy level for the lowest lying triplet state of the DPA molecule has been pub lished in several articles 35,36 as ranging between 24,272 cm -1 and 27,050 cm -1 The molar absorptivity, of the DPA molecule, has also been previously published. In the UV Atlas of Organic Compounds 37 it has been reported as 36,500 (M -1 cm -1 ) for ext = 193 nm, as 6,400 (M -1 cm -1 ) for ext = 217 nm, and as 7,100 (M -1 cm -1 ) ext = 274 nm.

PAGE 46

Figure 2.5 UV absorption spectrum of an aqueous suspension of intact Bacillus subtilis spores (from E. Ghiamati 28 ; reprinted with permission from Applied Spectroscopy). 21

PAGE 47

Figure 2.6 Absorption spectrum of [100M] DPA in aqueous solution (from F. Peral 34 Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, Copyright 2000, with permission from Elsevier). 22

PAGE 48

Figure 2.7 Fluorescence emission spectra of 10 M DPA in water at pH of 2.0 after 0, 5, 15, 30, and 60 minutes of UV irradiation ( ex = 300 nm), corresponding respectively from the lowest fluorescent signal to the highest fluorescing signal. (from R. Nudelman 33 ; reprinted with permission from Applied Spectroscopy). 23

PAGE 49

24 The fluorescence strength of pure DPA is very weak, thus not lending itself to indepth determination of energy levels and exci tation transfer rates within the DPA energy levels or those associated with collisional de -excitation with other s ubstances. However, in 1997 a method was presented and patented by D. Rosen in which it was determined that the addition of the terbium ion (Tb 3+ ) to the DPA can significantly enhance the fluorescence emission from the DPA-Tb complex, 6 thus allowing for easier detection of the presence or absence of b acteria endospores via fluorescence. Characteristics of the terbium ion as well as those of the DPA-Tb complex are covered in the following sections. 2.3 Use of Rare Earth Ions (Tb) to Enhance Fluorescence Emission Terbium III (Tb 3+ ) is a rare earth ion which retains its luminescent properties both under biological condi tions and in solution media when acting as a dopant for biomolecular systems. 38 Its placement in the periodic table of elements 39 is shown in Fig. 2.8. The involvement of the f-orbitals in bonding in lanthanides has been a controversial issue. 40 This is due to the belief that the regions of greatest 4f electron concentration in lanthanides does not extend far enough out from the nucleus as shown in Fig. 2.9. The 4f and 5d electrons are of comparable energies. The terbium atom has the [Xe].4f 9 .6s 2 electronic configurat ion as shown in Fig. 2.10. 39 Upon excitation, which can be either di rect (in the near UV or visible region) or indirect (via a resonant coupling between a donor molecule, e.g. DPA, and the Tb 3+ ), the terbium atom will lose electrons and will have the [Xe].4f 8 configuration. 39

PAGE 50

Figure 2.8 Periodic Table: Lanthanides (from Solid State Physics 39 1 st edition by Aschcroft/Mermin. 1976. Reprinted with permission of Brooks/Cole, a division of Thomson Learning: www.thompsonrights.com Fax 800 730-2215). 25

PAGE 51

Figure 2.9 Radial part of the hydrogenic wave functions for the 4f, 5d, and 6s orbitals of cerium (from Lanthanides and Actinides 40 ; used with permission from the Journal of Chemical Education, vol. 41, Iss. 7, p. 354 (1964)). 26

PAGE 52

1s 2p2s 3s3p3d 4s4p4d4f 5s5p5d5f 6s6p6dn=1n=2n=3n=4n=5n=6 Figure 2.10 Electronic configuration of the terbium atom (adapted from WebElements, http://www.webelements.com/ ) Mark Winter The University of Sheffield and WebElements Ltd, UK). 27

PAGE 53

The energy level diagram for the lanthanides is shown 41 in Fig. 2.11 and that of the terbium ion (Tb 3+ ) by itself 42 in Fig. 2.12. The numerical values for the terbium ions energy levels are listed in Table 2.1. The absorption spectrum as a function of wavenumber of an aqueous solution of Tb 3+ in either HClO 4 or DClO 4 is shown 43 in Fig. 2.13. The molar absorptivity of the terbium ion in solution has been reported as 0.25 for exct = 284 nm, as 0.31 for exct = 350 nm, as 0.34 for exct = 368 nm, and as 0.21 for exct = 377 nm. 43 The terbium ion can substitute the calcium ion in the calcium dipicolinate molecule partly due to their similar size (radius of Tb 3+ is approximately 1.12 and radius of Ca is approximately 0.99 +2 ) and terbiums strong affinity towards negatively charged donor groups. When the terbium ion substitutes the calcium ion, the terbium dipicolinate complex is formed. This DPA complex is shown 25 in Fig. 2.14. It is interesting to note that there are different complexes that can be formed between the terbium ion and the dipicolinate ion. Although this is more relevant to the chemical properties of the DPA-Tb complex, some initial studies were done by D. Rosen where he was able to determine that the lifetime of the DPA-Tb complex changes depending on which type of complex has formed. 25 The terbium dipicolinate complex can form where n = 1, 2, or 3. When DPA is in great excess, i.e. n = 3, tends to form. When Tb 32Tb(dpa)nn 33Tb(dpa) 3+ is in great excess, i.e. n = 1, tends to form. Tb(dpa) 25 The terbium ion has a total of 9 coordination sites, i.e. sites of possible bond formation, whereas each dipicolinate anion has three. In situations where the DPA is not in excess, i.e. n < 3, the remaining coordination sites of the terbium ion are occupied by 28

PAGE 54

29 Figure 2.11 Energy levels of the trivalent terbium ion (Tb 3+ ) (from G. H. Dieke 41 ; reprinted with permission from Optical Society of America).

PAGE 55

0246810121416182022242628303234363840Energy (1000 x cm-1) 5D4 7F47F57F27F07F17F37F6 5D3 5G65L105G55D25G4, 5L95G35L8, 5L7, 5G2, 5L6 5H75D15D05H65H5 5F45H45F5, 5H3, 5I85F3,5I75K9, 5D25F15F2 Figure 2.12 Detailed diagram of selected energy levels of the trivalent terbium ion (Tb 3+ ) (Data points obtained from W.T. Carnall 42 ). 30

PAGE 56

31 SLJ State Observed Energy Levels (cm -1 ) 7 F 6 0 7 F 5 2,035 7 F 4 3,281 7 F 3 4,329 7 F 2 5,016 7 F 1 5,502 7 F 0 5,819 5 D 4 20,504 5 D 3 26,270 5 G 6 26,405 5 L 10 26,946 5 G 5 27,829 5 D 2 28,197 5 G 4, 5 L 9 28,336 5 G 3 29,030 5 L 8 5 L 7 5 G 2 5 L 6 29,216 5 D 1 30,765 5 D 0 31,391 5 H 7 31,402 5 H 6 32,889 5 H 5 33,114 5 H 4 34,435 5 F 5, 5 H 3, 5 I 8 34,958 5 F 4 35,479 5 F 3, 5 I 7 36,587 5 F 2 37,226 5 F 1 37,527 5 K 9, 5 D 2 39,210 Table 2.1 Experiment al energy levels of Tb 3+ :LaF 3 (adapted from W.T. Carnall 42 ).

PAGE 57

Figure 2.13 Optical absorption of a solution of Tb 3+ (aquo) as a function of wavenumber (bottom axis) and wavelength (nm) (top axis) (from Carnall 43 ). 32

PAGE 58

Figure 2.14 Structure of Tb(DPA) + and complexes with water molecules (from D. L. Rosen -2Tb(dpa) 25 ; reprinted with permission from Applied Spectroscopy.) 33

PAGE 59

34 water molecules, whose OH vibrational modes interact with the terbium ion. 25,38,44-50 The OH oscillators dominate the relaxation of energy in the cation and shorten the lifetime of the excited state. 25 The photoluminescence lifetime of the DPA-Tb complex varies greatly with chelation number, i.e. the numbe r of DPA molecules attached to the terbium ion. For example, the photolumines cence lifetime of Tb(DPA) + = 0.66 ms, of Tb(DPA) 3 3= 2.0 ms, and of Tb(DPA) 2= 1.4 ms. 25 However, ligand exchange between the three chelation states results in their mixture having a monoexponential lifetime that is a weighed average of the lifetimes of the three states of the DPA-Tb complex. 25 Although the terbium ion fluoresces in so lution, this fluorescence is not very strong. However, when terbium is bound to DPA, its fluorescence enhances significantly due to the nonradiative transfer of energy that occurs between the DPA molecule (energy acceptor) and the terbium ion (the fluorescing species). 25 This is seen in initial work 51 by Rosen who measured the fluore scence spectrum of a solution of [200M] TbCl 3 blank, a solution of 50 nM [Tb(DPA) 3 ] 3, and a suspension of 0.08 mL B. Subtilis in 200 M TbCl 3 all in a [50mM] TRIS buffer solu tion. These results are shown in Fig. 2.15. In this work Rosen investigat ed the effect of the di fferent structures the DPA-Tb complex formed and their effect on the combined fluorescence lifetime of the complex. The fluorescence lines were not resolved. The fluorescence intensity of this complex is highly sensitive to the chemical make up of its surrounding environment. 52-55 The molar absorptivity of the DPA-Tb co mplex has been previously reported by Latva et al. 35 as 12,800 (M -1 cm -1 ) for an excitation wa velength of 275 nm.

PAGE 60

200MTbCl3 Figure 2.15 Fluorescence emission spectrum of 0.08-mL B. subtilis suspension in 200 M TbCl 3 both before (-) and after (s) filtering, for 200 M TbCl 3 blank (-), and for scaled 50 nM [Tb(DPA) 3 ] 3(-.-), in [50mM] TRIS buffer solution ( exc = 270nm) (from D.L. Rosen 51 ; reprinted with permission from Analytical Chemistry, Copyright 1996 American Chemical Society.) 35

PAGE 61

36 CHAPTER 3. INITIAL ENHANCEM ENT OF Tb-DPA FLUORESCENCE USING A PULSED UV LASER In this chapter the initial measuremen ts of the fluorescence from the DPA-Tb complex are discussed. Spectral fluorescence measurements were made using a conventional optical excitation source (X enon lamp) and then the fluorescence was measured using a UV pulsed laser. An e nhancement was observed using the pulsed UV laser for excitation. 3.1 Tb-DPA Spectrofluorometer System Using Conventional Lamp Source Fluorescence spectroscopy has been us ed in the past to detect various concentrations of biological endospores, such as bacillus globigii spores. 1,4,38,51,56-60 In traditional fluorescence based detection systems, a xenon flash lamp is usually used as the excitation source. The flash lamp can be used in combination with a monochromator and/or optical filters for narrow-band excitation, or by its elf for broad-band excitation. To measure the fluorescence spectrum of Tb-DPA, we used a conventional spectrofluorometer system (Ocean Optics, Inc.) as shown in Fig. 3.1. It consisted of a pulsed xenon flash lamp (w ith a pulse width of 5 s) which was used to excite the contents of a quartz cuvette and a spectromet er to measure the fluorescence emission. The fluorescence was collected 90 o to the excitation source beam. In addition, a

PAGE 62

PX-2XenonFlashLamp(200nm-870nm,1Hz) USB2000Spectrometer(2048elementCCD)350-nm-850nm La p t o p Com p u t erTriggerCable Mirror(Multi-pass)Sample(QuartzCuvette)CollectionLens(FiberCoupled)VIS/NIROpticalFiber600mm Figure 3.1 Schematic of commercial xenon flash lamp excitation fluorometer used to detect bacteria spores (Endospore Detection System, Ocean Optics, Inc.). 37

PAGE 63

38 multipass mirror was used in the excitation b eam to increase the excitation and a mirror was used in the fluorescence collection geomet ry. The fluorescence was collected via an end-of-fiber collection lens. The collected light then passed through a VIS/NIR fiber which was attached to a spectrometer (O cean Optics Inc Model USB 2000-FLG.) The spectrometers entrance slit was 200 m wide The spectrometer had a 600 lines per millimeter grating, blazed at 500 nm, a 2048element linear silicon CCD array with a mounted collection lens and a wa velength range of 380nm 1050nm. 3.1.1 Modified Spectrofluorometer System Using Pulsed UV Laser To increase the detection sensitivity, the pulsed xenon flash lamp was replaced by either a pulsed 4 th harmonic Nd:YAG laser with a 266nm output or a 3 rd harmonic Nd:YAG laser with a 355nm output. A schematic diagram for this setup is shown in Fig. 3.2. Both lasers were high-PRF (Pulse -repetition-frequency) diode-laser pumped Nd:YAG microchip lasers (JDS Uniphase, Model NU-10110-100). Figure 3.3 shows a photograph of the conventiona l set up along with the 4 th harmonic Nd:YAG laser, which could be substituted for the xenon flash lamp. 3.1.1.1 4 th Harmonic Nd:YAG Microchip Laser The microchip laser was a 4 th harmonic of a 1064 m Nd:YAG laser, with an output wavelength of 266 nm, an output pulse energy of 0.25 J/pulse at a pulserepetition-rate of 8.5 kHz, and a pulse widt h of 0.4 ns. The output of the laser was vertically polarized (100:1). The aver age power of this laser was 2 mW. The specifications for this laser are listed in Table 3.1.

PAGE 64

39 Pulsed4thHarmonicofNd:YAGLaser(orPulsed3rdHarmonicofNd:YAGLPRF:8.5kHz (266nm)aser(355nm)) USB2000Spectrometer(2048elementCCD)350-nm-850nm Mirror(Multi-pass)Sample(QuartzCuvette)CollectionLens(FiberCoupled) VIS/NIROpticalFiber600mm La p t o p Com p u t er Figure 3.2 Schematic of laser fluorometer system using a high PRF 266nm (or 355nm) laser for excitation.

PAGE 65

40 OceanOpticsUSB2000Spectrometer SampleHolder(withtwointernalmirrors) PulsedXenonFlashLamp 3rdHarmonicNd:YAGLaser(266nm,8.5kHz) OpticalFiber(withEnd-of-FiberLens) Figure 3.3 Photograph of fluorometer system that could use either the 4 th harmonic Nd:YAG laser or the pulsed xenon flash lamp for excitation.

PAGE 66

41 Microchip Laser Parameters Wavelength 266 nm Pulse Width 0.4 ns Pulse Energy J Repetition Rate 8 KHz Peak Power 590 W Average Power 2 mW Polarization (H:V) 100 to 1 Beam Size 1 mm Table 3.1 Laser parameters of the 266 nm high PRF microchip laser (JDS Uniphase, Model NU-10110-100).

PAGE 67

42 This laser consisted of Nd 3 :YAG crystal bonded to a thin layer of Cr:YAG saturable absorber (passive Q-switching), whic h 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 prevented lasing unt il a critical amount of energy had been absorbed. When the absorber reached its threshold, it saturated and lasing began. The pulse repetition frequency was determined by the time constant of the absorber and the amount of energy being input to the gain medi um. Pulse energy was independent of the pump power as the Q-switching was always at the same intracavity energy, and was directly proportional to the thickness of th e absorber medium. The pulse width is partially determined by the cavitys roundtrip time, so that the compactness of the cavity resulted in pulse widths shorter than a na nosecond, about 0.4 ns (Uniphase manual, NU10110-100). The output of the laser was frequency doubl ed by using a KTP (Potassium Titanyl Phosphate) crystal. The doubled output at 532 nm was frequency doubled again by a BBO crystal to 266 nm. The residual lig ht at 1064 nm and 532 nm were reduced by using optical blocking filters. 61 A schematic of the microchip laser is shown in Fig. 3.4. 3.2 Initial Fluorescence Measurements In this section we present some initial fluorescence measurements that will serve as an introduction to the fluorescence proper ties and behavior of the terbium ion and the Tb-DPA complex.

PAGE 68

43 LaserDiode GainMediumNd:YAG KTPCrystal(Doubling) BBOCrystal(Mixer) WavelengthFilteringOpticsLaserOutput:266nm1064nm 532nm266nm InputMirror SaturableAbsorberCr:YAGOutputMirror FocusingOptics Figure 3.4 Schematic of 4 th harmonic Nd:YAG microchip laser (JDS Uniphase, Model NU-10110-100).

PAGE 69

44 3.2.1 Fluorescence of Terbium Ion The laser induced fluorescence system shown in Fig. 3.2 was used to measure the fluorescence spectrum from TbCl 3 alone. Figure 3.5 shows the fluorescence intensity as a function of wavelength for a solution of [100 M] TbCl 3 in distilled water when excited with the 4 th harmonic Nd:YAG laser at 266 nm. The fluorescence spectrum of the terbium ion has four clearly visible peaks in the visible, which arise from the following electronic transitions, starting with the strongest fluorescence intensity: 5 D 4 to 7 F 5 ( emission 543 nm), 5 D 4 to 7 F 6 ( emission 488 nm), 5 D 4 to 7 F 4 ( emission 581 nm), and 5 D 4 to 7 F 3 ( emission 618 nm). The spectrum was measured over a time period of 100 ms. Note the weakness of the intensity and the low S/N ratio of the data. This shows that the terbium ion by itself in water has a relatively weak fluorescence. 3.2.2 Fluorescence of the DPA-Tb Complex As indicated earlier in Chapter 2 (Fig 2.15), the fluorescence of the terbium ion can be greatly enhanced with the additi on of the DPA molecule. As such, a measurement of the fluorescence emission from the DPA-Tb using the laser source was made. Figure 3.6 shows a comparison of the measured 100 ms integrated fluorescence intensities as a function of wa velength of a mixture of [50 M] DPA with [50 M] TbCl 3 in [0.01M] HCl, when excited with the three different excitation sour ces. From this figure it can be seen that th e excitation source that induced the highest fluorescence emission was the 266 nm 4 th harmonic Nd:YAG laser, followed by the pulsed xenon flash lamp used in the commercial setups, and the 355 nm 3 rd harmonic Nd:YAG laser. It is

PAGE 70

5D47F6Peak488nm 5D47F5Peak=543nm 5D47F4Peak=581nm 5D47F3Peak=618nm Figure 3.5 LIF emission spectrum of [100M] TbCl 3 in water, exc = 266nm. (Tb 3+ s electronic transitions labeled to corresponding spectral peaks.) 45

PAGE 71

01000200030004000470510550590630Wavelength (nm)Intensity (Counts) 266 nm Laser355 nm Laser Xn Flash Lamp Figure 3.6 Fluorescence intensity as a function of wavelength of a solution of [50M] DPA with [50M] TbCl 3 in [0.01M] HCl, when excited with three different sources: 266nm laser, 355nm laser and a conventional pulsed xenon flash lamp. 46

PAGE 72

47 also quite evident that th e latter excitation source at 355 nm fails to produce any considerable amount of fluorescence from this mixture of pure dipicolinic acid and terbium in weak hydrochloric acid. Also, it is important to note that the shape of the fluorescence of the DPA-Tb complex (Fig. 3.6) is the same as the shape of the fluorescence of the terbium ion alone (Fig. 3.5), which is consistent with earlier re ports that the terbium ion is the fluorescing species. 3.2.3 Fluorescence of the Mixtu res of Bacteria Doped Tb For comparison purposes, a solution of Bacillus globigii spores with [50 M] TbCl 3 in [0.01M] HCl was also excited with the same three different excitation sources. The results are shown in Fig. 3.7 and indi cate similar enhancement of the fluorescence emission using the 266nm laser source compared to the xenon flash lamp excitation. However, as can be seen, the overall fluores cence intensity has decreased significantly, possibly due to the lack of readily available DPA, which is now deep in the cortex of the spores. Also, the 355 nm 3 rd harmonic Nd:YAG laser seems to have produced a fluorescence emission from the sample. This emission is believed to be mostly due to the nicotinamide adenine dincleotide (NADH) wh ich is associated with cell metabolism, although this remains a cause of debate since by their nature bacter ia spores are in a dormant state. 1

PAGE 73

012345470510550590630Wavelength (nm)Intensity (Cnts) Xenon Flash Lamp266nm Laser355nm Laser Figure 3.7 Fluorescence intensity as a function of wavelength of a solution of Bacillus globigii spores with [50M] TbCl 3 in [0.01M] HCl, when excited with three different sources: 266nm laser, 355nm laser and a conventional pulsed xenon flash lamp. 48

PAGE 74

49 CHAPTER 4. DETAILED ABSORBA NCE MEASUREMENTS OF DPA, TbCl 3 AND THE DPA-Tb COMPLEX Previous absorption measurements of DPA and DPA-TbCl 3 were made under a limited number of conditions. As such, to expand upon these past studies, an investigation was made to de termine the detailed absorption spectrum under a controlled set of background conditions. This chapte r describes our experimental absorbance measurements and the measured absorption cross sections obtained when terbium chloride and dipicolinic acid, as well as th e mixtures of the two in different proportions, were examined in a number of different solvents. 4.1 Experimental Setup for Detailed Absorbance Measurements This section presents the experimental setup used to measure the absorbance spectrum of the different DPA and TbCl 3 solutions. The general setup for the absorbance experiments is shown in Fig. 4. 1. This setup consisted of a stable, continuous wave, deuterium tungsten ha logen UV-VIS lamp (CHEM 2000, Ocean Optics, Inc) that had output from 200 nm to 2062 nm. The output intensity as a function of wavelength of this light source was measur ed using a grating spectrometer (with range of 200 nm to 900 nm) and the results are show n in Fig. 4.2 to indicate the relative UVvisible portion of the output. A cuvette holder was attached to the aperture of the light

PAGE 75

CWCHEM2000UV-VISLamp(200nm-2062nm)ST-2000Spectrometer(2048elementCCD)200-nm-1100nmDesktop Sample(QuartzCuvette)CollectionLens(FiberCoupled)VIS/NIROpticalFiber400m Figure 4.1 Schematic of experimental set-up for absorbance measurements. 50

PAGE 76

01000200030004000200300400500600700800900Wavelength (nm)Intensity (Cnts) Figure 4.2 CHEM-2000s CW light source output spectrum over the range of 200 nm to 900 nm; drop-off near 900 nm due to drop-off in measuring grating efficiency. 51

PAGE 77

52 source and the signal was collected via a fibe r coupled collection lens positioned at 180 o of the said aperture. The signal was analyzed by the firs t channel (Channel #1) of an ST-2000 spectrometer (Ocean Optics, Inc.), and was sent to a desktop co mputer for storage. The spectrometer had a crossed Czerny-Turner design, as shown in Fig. 4.3. The entrance to the channel used in the spectrometer was a 25 m slit. This channel had a grating with 600 lines/mm, which was blazed at 400 nm. The relative efficiency as a function of wavelength for this grating is shown in Fig. 4.4. The 2048-element linear silicon CCD array detector was equipped with a collection lens. The spectrometers specifications are listed in Table 4.1. 4.1.1 Quartz and Plastic Cuvettes It was possible to use both quartz and plas tic cuvettes for our experiments. The transmission spectra of a UV transparent plastic cuvette (Brandtech Scientific, Inc, Model UV Plastic Cuvette) was measured and compared with that of a Suprasil quartz cuvette (Starna Cells, Inc Model # 23-Q-10,.). As can be seen from the results which are shown in Fig. 4.5, the pe rcent transmission of the quartz cuvette far surpasses that of the plastic cuvette across all wavelengths, especially the UV region. Due to this, the quartz (fused silica) cuvettes we re used in order to maximize the incident intensity and to minimize any extraneous fluorescence. The dimensions of the quartz cuvette were 12.5mm 12.5mm 45mm, with an inner path length of 10mm. These cuvettes were used throughout all the experiments pr esented in this dissertation.

PAGE 78

EntranceSlit CollimatingMirrorGratingFocusingMirrorL2CollectionLensCCDDetector Figure 4.3 ST-2000 spectrometers internal schematic for one optical channel. 53

PAGE 79

Figure 4.4 Relative efficiency as a function of wavelength of the grating used in the ST-2000 spectrometer. 54

PAGE 80

55 Parameters Channel #1 & Channel #2 Channel #3 Grating Spacing 600 lines/mm 600 lines/mm Grating Blazing 400 nm 750 nm Entrance Slit 25 m No Entrance Slit Order Sorting Filter No Yes Fiber Diameter 400 m (UV/VIS) 50 m (VIS/NIR) Detector Collection Lens Yes No 2048-element linear silicon 2048-element linearsilicon Detector CCD array CCD array Sensitivity (estimated) 86 photons/count (for 1-second integration) Integration Time 3 ms to 60 s Table 4.1 ST-2000 spectrometers parameters.

PAGE 81

050100200300400500600700800900Wavelength (nm)% Transmission Quartz CuvettePlastic Cuvette Figure 4.5 Measured transmission spectra of a quartz cuvette and a "UV transparent" plastic cuvette. 56

PAGE 82

4.2 Absorbance Measurement Procedure This section describes in detail the procedure that was followed in order to obtain the absorbance spectra, as well as the theory behind the calculations necessary to calculate the absorption cross sections of the different compounds in solution. Before spectral data was taken, the CHEM 2000 CW light source was warmed up for 15 minutes. In a darkened room, and after the output spectra of the lamp had stabilized, the output light was blocked and a dark or background spectrum, D, was taken. The light was then unblocked and its spectra was saved as a reference spectrum, R. Next, the sample was placed in the cuvette holder, i.e. in the path of the light, and the samples spectrum, S, was taken. All of these spectra were normalized to the grating efficiency, and detector responsivity. The absorbance, A, of the sample was then obtained by applying SDAlog(10RD ) (4.1) at each wavelength . Unless stated otherwise, all the spectra presented are the average of 100 runs. All the absorption spectra have been normalized to the fiber attenuation, the spectrometers grating efficiency, and the CCDs sensitivity. The absorption cross sections, were calculated from the absorbance measurements using 10AN log() el (4.2) 57

PAGE 83

58 4.3 TbCl 3 and DPA Solution Preparation It was determined that a wide range of DPA and TbCl 3 concentrations were required for our studies, includi ng the use of different solvents. The different solvents were guided by past studies into the chemical and spectral nature of DPA. Toward this end, four different basic solvent concentr ations were prepared in order to produce different proportions of terbium and dipicolin ic acid. [0.01M] HCl and [1M] HCl were prepared by diluting [17M] HCl (F isher Scientific) with distilled water. Solutions of [1M] sodium acetate buffer (SAB), pH 5. 6, were created by mixing [1M] sodium hydroxide with [1M] acetic acid until a pH of approximately 5.6 was obtained. Distilled water by itself was also used as a solvent. It should be noted that one of the reasons we used SAB pH 5.6 as a solvent was because previous work 52 had indicated that this buffer yielded an optimum environmen t for maximum fluorescence. Once the solvents were prepared, terbium chloride hexahydrate (TbCl 3 6H 2 O) (Sigma-Aldrich, 99.9% purity), was added to each of the solvents in order to produce the following concentrations: [5M], [10M], [15M], [20M], and [25M]. Also, DPA (99.5%, Sigma Aldrich, Inc.) was added separately to each of the solvents to form the following concentrations: [5M], [10M], [15M], [20M], and [25M]. These solutions were used to obtain absorbance data for each compound by itself, without the interaction effect between the two. The resultant list of different starter solutions are given in Table 4.2. In order to see the effect of the conc entration on the absorbance spectra once the two compounds were mixed, terbium chloride hexahydrate (TbCl 3 6H 2 O) (SigmaAldrich, 99.9% purity) and pyrid ine-2,6-dicarboxylic acid, or DPA, (Sigma Aldrich, Inc.)

PAGE 84

59 Solvents H2O [1M] HCL [0.01M] HCL [1M] SAB [5M] [5M] [5M] [5M] [10M] [10M] [10M] [10M] [15M] [15M] [15M] [15M] [20M] [20M] [20M] [20M] DPA [25M] [25M] [25M] [25M] [5M] [5M] [5M] [5M] [10M] [10M] [10M] [10M] [15M] [15M] [15M] [15M] [20M] [20M] [20M] [20M] Solutes TbCl 3 [25M] [25M] [25M] [25M] Table 4.2 List of different concentration of starting solu tions used in the experiments where the absorbances and the absorption cross sections of DPA and TbCl 3 were measured separately.

PAGE 85

60 were mixed in the same solution in order to produce a matrix of di fferent concentrations in each of the solvents described above. For each of the solvents a total of 25 mixed solutions were formed and analyzed. 4.4 General Absorbance Spectra of DPA and TbCl 3 and Their Mixtures In this section we present the expe rimental results of the absorbance measurements that were conducted on solutio ns of different concentrations of DPA by itself in different solvents, of TbCl 3 by itself in different solvents, and of mixtures of DPA and TbCl 3 in different solvents. The absorpti on cross sections were also calculated using the absorbance values at 266 nm obtained for each compound and their mixtures. 4.4.1 Measurements of Absorbances and Abso rption Cross Sections for Individual DPA and TbCl 3 Solutions Absorbance spectra were taken from solutions of [5M] DPA, [10M] DPA, [15M] DPA, [20M] DPA, and [25M] DPA, each in distil led water (DW), [0.01M] HCl, and in [1M] sodium acetate buffer (S AB) pH 5.6. The same was done with solutions of TbCl 3 Figure 4.6 shows the measured absorbance values as a function of wavelength of [15M] DPA in distilled water (DW) (Fig. 4.6 (a)), in [0.01M] HCl (Fig. 4.6 (b)), and in [1M] sodium acetate buffer (SAB), pH 5.6 (Fig. 4.6 (c)). Figure 4.7 shows the absorbance as a function of wavelength of [15M] TbCl 3 in distilled water (DW) in Fig. 4.7 (a), [0.01M] HCl in Fig. 4.7 (b), and [1M] sodium acetate buffer (SAB), pH 5.6 in Fig. 4.7 (c). The background of the solvents, i.e. distilled water, [0.01M] HCl, and [1M]

PAGE 86

0 0.1 0.2 Absorbance(a) 0 0.1 0.2 Absorbance(b) 250 450 650 850 0 0.1 0.2 Wavelength (nm)Absorbance(c)In Distilled WaterIn [0.01M] HClIn [1M] SAB (pH 5.6) Figure 4.6 Absorbance spectra of [15M] DPA measured in different solutions. Note: The background spectrum of each solvent, which was not subtracted, has been plotted along in a fainter dotted line. 61

PAGE 87

0 0.1 0.2 Absorbance(a) 0 0.1 0.2 Absorbance(b) 250 450 650 850 0 0.1 0.2 Wavelength (nm)Absorbance(c)In Distilled WaterIn [0.01M] HClIn [1M] SAB (pH 5.6) Figure 4.7 Absorbance spectra of [15M] TbCl 3 measured in different solutions. Note: The background spectrum of each solvent, which was not subtracted, has been plotted along in a fainter dotted line. 62

PAGE 88

63 SAB, have not been subtracted from these sp ectra, but rather, they are shown along side of them, in a fainter dotted line. The spectra for the remaining concentrations as given in Table 4.2 are presented in Appendix A. Upon visual evaluation of the solvents in the graphs, it is notic eable that sodium acetate buffer has a distinct absorption band in the 440 nm to 487 nm region, with the peak absorbance occurring near 469 nm. The absorption for distilled water and [0.01M] HCl is uniformly low across the shown spectral region. As can be seen in Fig. 4.6 (a), (b), an d (c), dipicolinic acid has an absorption band from 218 nm to 296 nm, with three distinct pe aks near 221 nm, 231 nm, and 271 nm. There is a much weaker absorption band fr om around 430 nm to 532 nm. This is in agreement with measurements done by previous investigators. 34 A magnified region of the absorption gra ph of DPA in distilled water (with the water background already subtracted) is s hown in Fig. 4.8 where the individual absorption peaks are seen more easily. The absorbance spectrum of TbCl 3 in Fig. 4.7 shows negligible absorption for the terbium ion in solution in this wavelength region. This is in agreement with previously published measurements. 43 Figure 4.9 shows the absorbance values at 266 nm as a function of the DPA molecules concentration in distilled water, in [0.01M] HCl, and in [1M] SAB. The background absorbance values for each of the solvents have been subtracted, and hence, the absorbance values shown in this graph are, in theory, independent of the effect of the solvent. For the sake of completeness, how ever, it has been indicated in the graph which points belong to which solvent. From this figure it can be seen that the absorption of

PAGE 89

Figure 4.8 Enlarged view of the absorbance spectra of [15M] DPA in distilled water; water background has been subtracted. 64

PAGE 90

0 5 10 15x 1015 0 0.1 0.2 Concentration (Molecules/cm3)Absorbance Distilled Water [0.01M] HCl [1M] SAB Figure 4.9 Measured absorbance of DPA at 266 nm as a function of DPA concentration. The vertical line inside the markers at a concentration of around 15 x 10 15 molecules per cm 3 corresponds to an error of about 0.01 in the absorbance scale. Each data point is the average of 100 runs. 65

PAGE 91

66 DPA at 266nm increases as the concentration incr eases. This is consistent with Eq. (2.2) where A is proportional to the concentration. However, some chemical products such as clusters or other dimers or molecules may produce a non-linear effect. Figure 4.10 shows the absorbance values at 266 nm as a function of the terbium ions concentration in distilled water, in [0.01M] HCl, and in [1M] SAB. The absorbance values for each of the solvents have also been subtracted in this instance. The absorbance of the terbium ion seems negligible at 266 nm. 4.4.2 Absorption Cross Section of DPA and TbCl 3 Individually Using Eq.(4.2), and the absorbance values presented in Figs. 4.9 and 4.10, the absorption cross section values fo r the DPA molecule and for the Tb 3+ ion at 266 nm were calculated. A graphical presentation of the absorption cross section values for the DPA molecule and the terbium ion is shown in Figure 4.11. The averages of these values are (2.10 0.38) 10 -17 cm 2 /molecule for the DPA molecule and (0.18 0.22) 10 -17 cm 2 /molecule for the Tb 3+ ion. It is interesting to note that the terbium ions absorption cross section in solution at 266nm is about twelve times smaller than that of the DPA molecule. The value obtained for the absorption cro ss section of the DPA molecule is very close to data obtained by the UV Atlas of Organic Compounds, 37 which found the absorption cross section of the DPA mo lecule at 274nm to be around 2.7 x 10 -17 cm 2 /molecule.

PAGE 92

0 5 10 15x 1015 0 0.1 0.2 Concentration (Molecules/cm3)Absorbance Distilled Water [0.01M] HCl [1M] SAB Figure 4.10 Measured absorbance of TbCl 3 at 266nm as a function of TbCl 3 concentration. The vertical line inside the markers at a concentration of around 15 x 10 15 molecules per cm 3 corresponds to an error of about 0.01 in the absorbance scale. Each data point is the average of 100 runs. 67

PAGE 93

Figure 4.11 Absorption cross section of DPA and TbCl 3 at 266nm. The vertical line inside the markers at a concentration of around 15 x 10 15 molecules per cm 3 corresponds to an error of about 0.15 x 10 -17 cm 2 /Molecule in the absorption cross section scale ( 5% error). 68

PAGE 94

69 The molar absorptivity of the terbium ion in solution has been reported as 9.6 10 -22 cm 2 /molecule for exct = 284 nm. 43 It must be noted that even though the numerical values of this published data and our findings are quite far from each other, they are both much smaller than the absorption cross sect ion value obtained for the DPA molecule. Also, it should be noted that there is an al most 20 nm difference between the excitation wavelength used in our experiment a nd that used in this reference. 4.4.3 Absorbances of Mixtures of DPA and TbCl 3 As stated earlier, the absorbance values for a mixture of DPA and TbCl 3 is much greater than that for TbCl 3 alone or for DPA alone. As such, absorbances were also measured for solutions that contained varyi ng concentrations of mixtures of DPA and TbCl 3 in different solvents, i.e. distilled water, HCl, and SAB. Our results are shown in the following figures. Figure 4.12 shows the measured absorbance spectra of solutions of varying TbCl 3 concentrations, i.e. [5M], [10M], [15M], [20M], and [25M], mixed with different DPA concentrations. Figure 4.12 (a) shows the absorbance spectra as a function of wavelength of mixtures of the various TbCl 3 concentrations with [5M] DPA in distilled water. Figure 4.12 (b) s hows the absorbance spectra of mixtures of the various TbCl 3 concentrations with [10M] DPA in distilled water. Figure 4.12 (c) shows the absorbance spectra of mixtures of the various TbCl 3 concentrations with [15M] DPA in distilled water. Figure 4.12 (d) shows the ab sorbance spectra of mixtures of the various TbCl 3 concentrations with [20M] DPA in distilled water. In Fig. 4.12 (e) one can see

PAGE 95

Figure 4.12 Absolute absorbance spectra of solutions of [5M] DPA measured with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in distilled water. 70

PAGE 96

the absorbance spectra of mixtures of the various TbCl 3 concentrations with [25M] DPA in distilled water. The distilled water background has been subtracted from each one of these spectra. For completeness, the detailed spectra of mixtures of TbCl 3 and DPA in solutions of [0.01M] HCl, and [1M] SAB (pH 5.6) were also measured. These results are presented in Appendix A. As can be seen in Fig. 4.12, the absorbances of the solutions increase in their peak values with an increasing DPA concentration. It is also interesting to note from these figures that as the DPA concentration increases a slight splitting in the main absorption peak near 271 nm becomes more prominent. This effect can be seen better in the enlarged graphs of Fig. 4.13. This phenomenon has been observed before by Cable and seems to be due to the effect of the orbital interaction in the DPA. 58 Also, a weaker secondary absorption band (~ 430 nm to 532 nm), that was also visible in the spectra of the DPA by itself (Figs. 4.6(a) 4.6(c)), becomes more visible. Figure 4.14 shows an enlarged view of the absorbance of mixtures of [15M] DPA with [5M], [10M], [15M], [20M], and [25M] TbCl 3 in distilled water, as a function of wavelength. It is clear from this figure that the absorbance spectra of the mixtures do not seem to be affected by the concentration of the TbCl 3 in any systematic way. Next, the absorbance value for each different solution at 266 nm was extracted, and a graph of absorbance as a function of varying DPA and TbCl 3 concentrations was plotted for each solvent. The corresponding solvents background absorbance value has been subtracted from each data point. 71

PAGE 97

Figure 4.13 Enlarged view of the absolute absorbance spectra of solutions shown in Fig. 4.12. 72

PAGE 98

Figure 4.14 Enlarged view of absorbance spectra of [15M] DPA with different concentrations of TbCl 3 in distilled water; water background has been subtracted. This graph corresponds to Figure 4.12 (c). 73

PAGE 99

Figure 4.15 shows the measured absorbance values at 266 nm as a function of concentration for mixtures of DPA and TbCl 3 in distilled water. When there is no DPA present, the absorbance of TbCl 3 at 266 nm is essentially zero. As noted before, the absorbance of TbCl 3 does not seem to change considerably for a fixed concentration of DPA, regardless of the TbCl 3 concentration. This suggests an almost linear relationship between the absorbance of light at 266nm and the concentration of the DPA molecules, for the DPA-TbCl 3 complex, indicating that DPA is the main, and possibly only, absorbing species in the DPA-TbCl 3 complex. The same pattern seems to hold true for mixtures of DPA and TbCl 3 in [0.01M] HCl and in [1M] SAB (pH 5.6), within experimental uncertainty, as shown in Fig. 4.16 and Fig. 4.17. The indentations in the graphs are thought to be due to experimental error. 4.4.4 Absorption Cross Section of the DPA-TbCl 3 Complex Next, the absorption cross section values for the DPA-Tb complex were calculated. The absorbance values presented in Figs. 4.15 4.17, were used in these calculations. Since the DPA molecules were shown to be the main absorbers of the incident light at 266 nm, their concentration alone was used for the concentration variable, N, in Eq. (4.3). In other words: DPA-TbCl3DPA-TbDPA10AN log() el (4.4) The graphical presentation of the calculated absorption cross section values are presented in Fig. 4.18 for the DPA-Tb complex in distilled water, in Fig. 4.19 for the DPA-Tb complex in [0.01M] HCl, and in Fig. 4.20 for the DPA-Tb complex in [1M] SAB 74

PAGE 100

0 5 10 15x 1015 0 5 10 15x 1015 0 0.05 0.1 0.15 0.2 [TbCl3] (Molecules/cm3) [DPA] (Molecules/cm3) Absorbance Figure 4.15 Measured absorbance as a function of DPA and TbCl 3 concentration in distilled water at a wavelength of 266 nm. Water background spectra subtracted. 75

PAGE 101

0 5 10 15x 1015 0 5 10 15x 1015 0 0.05 0.1 0.15 0.2 [TbCl3] (Molecules/cm3) [DPA] (Molecules/cm3) Absorbance Figure 4.16 Measured absorbance as a function of DPA and TbCl 3 concentration in [0.01M] HCl at a wavelength of 266 nm. HCl background spectra subtracted. 76

PAGE 102

0 5 10 15x 1015 0 5 10 15x 1015 0 0.05 0.1 0.15 0.2 [TbCl3](Molecules/cm3) [DPA](Molecules/cm3) Absorbance Figure 4.17 Measured absorbance as a function of DPA and TbCl 3 concentration in [1M] SAB at a wavelength of 266 nm. SAB background spectra subtracted. 77

PAGE 103

Figure 4.18 Absorption cross section values at 266 nm of mixtures of DPA and TbCl 3 at varying concentrations, in distilled water. 78

PAGE 104

Figure 4.19 Absorption cross section values at 266 nm of mixtures of DPA and TbCl 3 at varying concentrations, in [0.01M] HCl. 79

PAGE 105

Figure 4.20 Absorption cross section values at 266 nm of mixtures of DPA and TbCl 3 at varying concentrations, in [1M] SAB. 80

PAGE 106

81 (pH 5.6). Except for small discrepancies in th e data, it is clear that the absorption cross section of the terbium doped di picolinate molecule is the same as that for the DPA alone, within experimental error, which is signifi cantly larger than that for the terbium ion alone. The absorption cross section value for the DPA-Tb complex was calculated to be about (1.9 0.46) 10 -17 cm 2 /molecule. This value falls within the uncertainty of the absorption cross section value for the DP A molecule, which was (2.1 0.38) 10 -17 cm 2 /molecule. The obtained value for the absorption cr oss section of the DPA-Tb complex is within a factor of 2 of similar data obtained by Latva, 35 which was around 4.9 x 10 -17 cm 2 /molecule at 275 nm. To better see the dependence of the ab sorption cross section of the DPA-Tb complex on the concentration of the individual compone nts, two graphs were plotted. A plot of the average absorption cr oss section as a function of TbCl 3 concentration, shown in Fig. 4.21 (a), shows a rela tively invariant dependence of DPA-Tb with respect to the terbium chloride concentration. A plot of the average DPA-Tb as a function of DPA concentration shows a stronger dependence of the absorption cross section on the amount of DPA present in the mixture, with the value dramatically decreasing (to Tb ) when there is no DPA present, Fig. 4.21 (b).

PAGE 107

01E-172E-1704E+158E+151.2E+161.6E+16[TbCl3] in Molecules/cm3DPA-Tb in cm2/Molecule SAB HCl DW (a) 01E-172E-173E-174E-1704E+158E+151.2E+161.6E+16[DPA] in Molecules/cm3DPA-Tb in cm2/Molecule SAB HCL DW (b) Figure 4.21 Average absorption cross section of the terbium doped dipicolinic acid as a function of the terbium ions concentration, (a), and as a function of DPAs concentration, (b), in different solvents at exct = 266 nm. 82

PAGE 108

83 4.5 Comparison of Absorption and Fluorescence Spectrum of DPA, Tb, and the DPA-Tb Complex It is important to compare the absorpti on and fluorescence spectrum of DPA, Tb, and the DPA-Tb complex. For absorption, the DPA absorption spectrum was given in Fig. 4.6, and showed almost the same absorption spectrum as that for the DPA-Tb complex (Fig. 4.12). Note that the absorption spectrum of terbium alone gi ven in Fig. 4.7 was essentially zero. As a result, the absorption spectrum of DPA-Tb is essentially the same as that for the DPA. For fluorescence, Fig. 3.5 shows the fluorescence spectrum emitted by terbium alone and Fig. 3.6 shows the spectrum emitted by the DPA-Tb complex. Figure 2.7 shows negligible fluorescence emission from DPA alone for the case of no UV irradiation; these results are also confirmed la ter in Chapter 5 of this thesis which shows the negligible fluorescence emission from DPA al one (see Fig. 5.14). As a result, the fluorescence emission spectrum from the DPA-Tb complex is essentially the same as that for terbium alone. Of importance, though, is that whil e the above absorption and fluorescence spectrum are similar in shape, they di ffer in magnitude and intensity. Based upon this, we conclude that the ab sorption of light by the DPA-Tb complex is mostly due to the DPA, and the fluorescence emission of the DPA-Tb complex is mostly associated with Tb emission. As such, for simplicity, we can use the same descriptions of the energy levels for DPA and Tb 3+ as that for the DPA-Tb complex.

PAGE 109

84 CHAPTER 5. UV EXCITATION EMI SSION MATRIX (EEM) SPECTRA OF TERBIUM CHLORIDE AN D DIPICOLINIC ACID In order to study the absorption spectrum of DPA and DPA-TbCl 3 an investigation was made to determine the opt imized laser wavelength to be used for excitation. Toward this end, an Exci tation Emission Matrix (EEM) spectral study was made to vary the laser excitation wavelength and detect the emitted fluorescence light. This chapter describes the experimental setup and the spectral results obtained when different proportions of terbium chlori de and dipicolinic acid were mixed in a number of different solvents, and their fluorescence spectra were recorded in the form of excitation emission matrices (EEM), after being excited with a low pulse repetition frequency (PRF) UV tunable dye laser. 5.1 Experimental Setup for the Excita tion Emission Matrix Measurements This section presents the experimental setup used to obtain the EEM spectra of DPA and TbCl 3 in different solutions. The overall EEM setup is shown in Fig. 5.1, and consisted of a pulsed nitrogen pumped, fr equency doubled tunable dye laser whose UV output was focused into the center of a quartz cell containing about 3 ml of each solution. With the help of an end-of-fiber lens, and a UV/VIS fiber, the fluorescence spectra were collected at right angles to the quartz cuvette containing the samples. The fluorescence was then analyzed using a USB-2000 sp ectrometer (Ocean Optics Inc.).

PAGE 110

Det.#1UVEnhancedSiDetectorFocusingLens Det.#2UVEnhancedSiDetectorFocusingLensNitrogenPumpedFreq.DoubledTunableDyeLaser220nm-280nm1mJ,10Hz OceanOpticsUSB2000 Spectrometer2048elementCCD350nm-850nmCollectionLens-FiberCoupledFocusingLensMirrorBeamSplitter NDFilter NDFilterQuartzCell Figure 5.1 Schematic of set up used for laboratory excitation emission (EEM) spectroscopy measurements. 85

PAGE 111

86 The incident laser light was partially deviated to two UV enhanced silicon detectors, before and after the cuvette, by using two beam splitters. These detected laser intensities were used to normalization the spec tra with respect to the incident laser power. 5.1.1 Tunable Laser Source A schematic of the tunable dye laser is shown in Fig. 5.2. This laser consisted of three separate modules. The first module wa s a pulsed nitrogen laser (Laser Science Inc., Model VSL-337ND-S), with an output wavelength of 337 nm, output pulse energy of 300J/pulse, a repetition rate of 10 Hz, and a pulse width of 4 ns. 62 The output of the nitrogen laser pumped the second module, which consisted of the dye laser (Laser Science Inc., Model DL MS-220). The nitrogen lasers output was focused onto a quartz cuvette containing an or ganic dye by means of a cylindrical lens. This dye laser gain medium in the dye cuvett e was inside a linear mirror to mirror cavity, with the end mirror being a 2400 lines/mm grating arranged in a Littrow configuration. The dye laser was tunable from 360 nm to 700 nm, depending on the emission band of the dye used. For each dye used, different output wavelengths could be chosen by changing the grating angle, which was calibrated to read th e emission wavelength. The line width of the output was about 0.3 nm. The energy of the outgoing laser pulse was about 50 J/pulse at 500nm. 63 The output of the dye laser was polarized 8:1, vertical to horizontal, respectively. The dyes that were used in these experiments were Coumarin 440, Coumarin 481, and Coumarin 500 (Exciton Inc.). The rela tive spectral outputs of these dyes are shown in Fig. 5.3. The lifetime of the dyes we re prolonged by using a magnetic stirrer to

PAGE 112

87 Figure 5.2 Nitrogen pumped, frequency doubled UV tunable dye laser (Laser Science; Models VSL-337ND, DLMS-220 and UVS-300).

PAGE 113

Figure 5.3 Tuning curves for Coumarin 440, Coumarin 481, and Coumarin 500 dyes that were used for the EEM experiments. (Graph obtained from http://www.exciton.com/). 88

PAGE 114

89 circulate the dyes and, therefore, minimize photo-bleaching. Coumarin 440 was used for the 440-470 nm output range. Coumarin 481 was used for the 470-490 nm range. Coumarin 500 was used to obtain an output wavelength range of 490-555 nm. The output of the dye laser was sent into a frequency doubling module (Laser Science Inc., Model UVS-300). The vertica lly and horizontally po larized parts of the dye lasers output were separated by a polarizing beam splitter, which directly transmitted the horizontally polarized part of the beam out of the module, and it reflected the vertically polarized compone nt of the beam into the rema inder of the module. This reflected part of the beam was focused into the center of a nonlinear frequency doubling -barium borate (BBO) crystal, which was oriented to frequency double vertically, polarized light. The frequency doubled light was then separated from the fundamental beam by using a Pellin-Broca prism. 64 Two separate BBO crystals were used in the doubling module for the EEM measurements, and were cut and polished to cover the wavelength ranges between 220 nm to 250 nm and from 240 nm to 285 nm, respectively. The laser output from the doubling module had a pulse width of 3 ns and a line width of 0.2 nm. At a repetition rate of 10 Hz, the pulse energy varied from about 0.5J/pulse to 5J/pulse, depending on the wavelength of the laser. The parameters of this laser are listed in Table 5.1. 5.1.2 Spectrometer The fluorescence spectra of the samples was detected by a spectrometer (Ocean Optics USB-2000) with an entrance slit of 200 m and a 2048-element silicon CCD

PAGE 115

90 Tunable Laser Parameters Wavelength 220 nm 280 nm Pulse Width 3 ns Pulse Energy 1 J Repetition Rate 10 Hz Peak Power 333 W Average Power 10 W Polarization (H:V) 100 to 1 Beam Size 7 mm x 2 mm Table 5.1 Specifications for the tunable la ser (Laser Science In c.) tuned to 266nm.

PAGE 116

91 detector. The design of the optical layout was a crossed Czerny-Turner, as shown in the previous chapter (Fig. 4.3). The spectrom eters effective det ection range was 380 nm1050 nm. The grating used in the spectrometer ha d 600 lines/mm and was blazed at 500 nm. Details about the spectrometers specifications are listed in Table 5.2. The grating efficiency as a function of wavelength is shown in Fig. 5.4. A UV-VIS optical fiber, whos e attenuation as a function of wavelength is shown in Fig. 5.5, was attached to the entrance slit, and at the other end of the fiber, i.e. the collection side, there was a fused silica collect ion lens suitable for the 200 nm 2000 nm wavelength range. The spectrometer had a Sony ILX511 linear silicon CCD array detector with 2048 pixels and a 250:1 signal-to-noi se ratio at full signal. Th e detectors sensitivity as a function of wavelength is shown in Fig. 5.6, a nd its specifications are given in Table 5.3. 5.2 Background Subtraction and Normalization Procedures Before fluorescence spectra were obtain ed, a background or dark spectrum was obtained while the illumination source was blocked. This background spectrum was then subtracted from each fluorescence spectral data. A sample background spectrum is shown in Fig. 5.7. All the obtained data that are presented in this chapter were normalized to the incident lasers power, shown in Fig. 5.8, the spectrometers grating efficiency (Fig. 5.4) and to the spectrometers detectors sens itivity (Fig. 5.6) for the corresponding wavelengths. Specifically, the normaliza tion process for the spectral data used

PAGE 117

92 Parameters Specifications Entrance Slit 200 m Detector 2048-element Linear Silicon CCD Array Detector Collection Lens Yes Sensitivity (estimated) 86 photons/count (for 1second Integration) Effective Range 380 nm 1050 nm Grating Spacing 600 lines/mm Grating Blazing 500 nm Grating Efficiency 350 nm 850 nm Optical Resolution 10 nm FWHM Integration Time 3 ms to 65 s Fiber Diameter 600 m (VIS/NIR) Table 5.2 USB-2000 spectrometers specifications (Ocean Optics Inc.).

PAGE 118

Figure 5.4 USB-2000 gratings relative efficiency as a function of wavelength (600lines/mm blazed at 500 nm). 93

PAGE 119

Figure 5.5 Attenuation curve for 600m diameter UV-VIS fiber. 94

PAGE 120

Figure 5.6 Relative sensitivity of 2048-element Sony silicon CCD array (Sony ILX511). Note: The relative sensitivity data for wavelengths below 400 nm was not available. 95

PAGE 121

96 Detector: Sony ILX511 linear silicon CCD array Number of elements: 2048 pixels Pixel size: 14 m x 200 m Pixel well depth: 62,500 electrons Signal-to-noise ratio: 250:1 (at full signal) A/D resolution: 12 bit Dark noise: 2.5 RMS counts Corrected linearity: >99.8% Table 5.3 CCD detectors specifications.

PAGE 122

97 05001000150020003004005006007008009001000Wavelength (nm)Intensity (Counts) Figure 5.7 Background spectra (noise) of the USB-2000 spectrometer with the excitation source blocked.

PAGE 123

050100150200250220230240250260270280Output Wavelength (nm)Power (Relative) Figure 5.8 Dye laser power output normalized to the detectors responsivity. 98

PAGE 124

iNormalizediGratingCCDFDFPES (5.1) where F Normalized represents the normalized fluorescence spectrum, F i represents the (raw) fluorescence spectrum corresponding to an incident wavelength i, D represents the dark spectra, P i is the dye lasers power at wavelength i, is the efficiency of the grating at wavelength and is the sensitivity of the silicon CCD detector at wavelength GratingE CCDS 5.3 Solution Preparation Four different solvents were prepared in order to dilute different proportions of terbium and dipicolinic acid. [0.01M] HCl and [1M] HCl were prepared by diluting [17M] HCl (Fisher Scientific) with distilled water. Solutions of [1M] sodium acetate buffer (SAB), pH 5.6, were created by mixing [1M] sodium hydroxide with [1M] acetic acid until the desired pH was obtained. Distilled water by itself was also used as a solvent. Once the solvents were prepared, terbium chloride hexahydrate (TbCl 3 6H 2 O) (Sigma-Aldrich, 99.9% purity) was added to each solvent in order to produce [100M] TbCl 3 Next, dipicolinic acid (Sigma Aldrich, Inc.) was added to each solvent in order to make [25 M] DPA. 99

PAGE 125

100 The final set of solutions consiste d of mixtures of DPA with TbCl 3 in each solvent, to form [25M] DPA with [100M] TbCl 3 mixed in each of the four different solvents described above. 5.4 Experimental Results: Meas ured EEM Spectra for DPA, TbCl 3 and Mixed Solutions of Both EEM spectra were obtained from a variety of solutions and background solvents. The laser wavelength was tuned over 6 different wavelengths from 225 nm to 275 nm, and the fluorescence spectrum recorded for each sample. As an example, Fig. 5.9 shows the normalized fluorescence intensity of a sample of distilled water as function of the emission wavelength, emsn with respect to the incident excitation wavelength, exct As can be seen, for the emission wavelength range of 470 nm-650 nm distilled waters fluorescence is undete ctable for excitation wavelengt hs of 225 nm to 275 nm. Additional EEM spectra were obtained fo r a series of solutions of TbCl 3 in distilled water, TbCl 3 in [0.01M] HCl, TbCl 3 in [1M] HCl, and TbCl 3 in [1M] SAB (pH 5.6). In addition, EEM spectra were obtained for DPA in distilled water; and mixtures of DPA with TbCl 3 in distilled water, in [0.01M] HCl, in [1M] HCl, and in [1M] SAB (pH 5.6). These EEM results are discusse d in the following sections. 5.4.1 EEM of TbCl 3 in Solutions The normalized fluorescence spectrum of a solution of [100M] TbCl 3 in distilled water was measured for excitation wavelength s of 225 nm to 275 nm in increments of 10 nm, and the data is shown in Fig. 5.10. As can be seen in this and most of the EEM

PAGE 126

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.9 EEM spectra of distilled water. 101

PAGE 127

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.10 EEM spectra of [100M] TbCl 3 in distilled water. 102

PAGE 128

103 graphs, there is more noise present in the spectra corresponding to excitation wavelengths 225 nm, 235 nm, and 275 nm; we attribute this to the low laser power at these excitation wavelengths. In Fig. 5.10, there is a very weak fluorescence peak at around 540 nm for the 235 nm, 255 nm, and 265 nm excitation wavelengths The strongest fluorescence signal, however, occurred when the sample was excite d with 225 nm light. At this excitation wavelength, there are four fluorescence peaks correspondi ng to the four strongest (electronic) transitions in the terbium ion (Tb 3+ ), as described in chapter 2. No fluorescence was detected when measuring the EEM spectra of [100M] TbCl 3 in [1 M] HCl, and this is shown in Fig. 5.11. Although the same is true for the fluorescence from [100M] TbCl 3 in [0.01 M] HCl (Fig. 5.12), a small but negligible peak was consistently present at arou nd 540nm. In the case of the [100M] TbCl 3 in [1 M] SAB (pH 5.6) solution, no fluorescence was observed (Fig. 5.13). 5.4.2 EEM of DPA in Solutions Figure 5.14 shows the measured EEM spectra of [100M] DPA in distilled water, when excited from 225 nm to 275 nm in incr ements of 10nm. As can be seen, the fluorescence emission for this chem ical is negligible. These results are consistent with previous data. Dipicolinic acid, which is thought to offer protect ion from UV radiation to bacteria spores, has a very weak fluorescence. 29,33 The data presented in this section is in accordance with these studies, in that the fluorescence of DPA in solution is negligible, when UV-irradiated for less than at least 1 minute.

PAGE 129

225 235 245 255 265 275 480 520 560 600 640 0 0.005 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.11 EEM spectra of [100M] TbCl 3 in [1M] HCl. 104

PAGE 130

225 235 245 255 265 275 480 520 560 600 640 0 0.05 0.1 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.12 EEM Spectra of [100M] TbCl 3 in [0.01 M] HCl. 105

PAGE 131

225 235 245 255 265 275 480 520 560 600 640 0 0.005 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.13 EEM spectra of [100M] TbCl 3 in [1M] SAB pH 5.6. 106

PAGE 132

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.14 EEM spectra of [100M] DPA in distilled water. 107

PAGE 133

108 5.4.3 EEM of Mixtures of DPA and TbCl 3 in Solution Figure 5.15 shows the measured EEM spectra of [50M] DPA mixed with [50M] TbCl 3 in distilled water. As can be seen all four fluorescence peaks are present in all of the emission spectra, regardless of the excitation wavelength, although the fourth peak (i.e. 618 nm peak) is less pronounced There does not seem to be a strong dependence between the excitation waveleng th and the emission intensity for this solution. Figure 5.16 shows the EEM spectra of [25M] DPA mixed with [100M] TbCl 3 in [0.01M] HCl, for excitation wavelengths of 225 nm to 275 nm. For this case of DPATb in a weak acid solution, all four fluor escence peaks are clearly present in each emission spectrum. The strongest emission at 543 nm occurs with excitation at 225 nm. Figure 5.17shows the EEM spectra ofM] DPA mixed with [100 M] TbCl 3 in [1M] HCl, for excitation wavelengths of 225 nm to 275 nm. In the case of DPA-Tb in a strong acid solution, no fluorescence is detect ed for any of the excitation wavelengths. This, we believe, is due to the high acidity of the solvent, i.e. [1M] HCl, and is consistent with previous pH dependent fluoresce nce quantum yield experimental data. 65 Figure 5.18 shows the EEM spectra ofM] DPA mixed with [100M] TbCl 3 in [1M] SAB, pH ~ 5.6, for excitation wavelengths of 225 nm to 275 nm. All four fluorescence peaks are clearly present in each emission spectrum, and once again, no strong or direct relation is evident between the excitation wavele ngth and the emission intensity near 543 nm.

PAGE 134

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.15 EEM spectra of [50] DPA [50M] TbCl 3 in distilled water. 109

PAGE 135

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.16 EEM spectra of [25M] DPA in [100M] TbCl 3 in [0.01 M] HCl. 110

PAGE 136

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.17 EEM spectra of [25M] DPA in [100M] TbCl 3 in [1M] HCl. 111

PAGE 137

225 235 245 255 265 275 480 520 560 600 640 0 0.002 0.004 0.006 0.008 0.01 exct(nm) emsn(nm) Intensity (Cnts) Figure 5.18 EEM Spectra of [25M] DPA in [100M] TbCl 3 in [1M] SAB. 112

PAGE 138

113 5.5 EEM Spectra Conclusions The above results indicate that the wave length for excitation can be within the range of 225 nm to 270 nm for the mixture of DPA and TbCl 3 The EEMs did not follow a pattern between the excitation wavelengths and the emission fluorescence intensities for any of the solutions. As a whole, 235 nm excitation wavelength seemed to produce some of the strongest fluorescence of all of the wavelengths in most of the solutions, i.e. in [0.01M] HCl, in distille d water, and in [1M] SAB, for the DPA-Tb complex. The solvent that seemed to provide th e most fluorescence-inducing environment with most of the excitation wavelengths seemed to be the [0.01M] HCl, followed by distilled water, which was slightly better th an [1M] SAB pH 5.6. [1M] HCl did not enable the generation of fluorescence in the solutions. The EEMs of terbium ion in solution do not show a detectable fluorescence for most excitation wavelengths in most of the solvents, except for exct = 225 nm in distilled water, where terbiums four distinct fluoresce nce peaks are clearly visible. Despite the fact that the OH vibrational modes in wa ter tend to bond to the terbium ion and are effective quenchers, 25,38,44-50 the deep UV excitation wavelength is able to excite the ion. This is in agreement with previously published 43 absorption data for the terbium ion in solution where the ion has a relatively hi gh absorbance in the deep UV region as was shown in Fig. 2.13.

PAGE 139

114 CHAPTER 6. FLUORESCENCE LIFETIMES OF SOLUTIONS OF DPA AND TbCl 3 WHEN EXCITED WITH A LOW PRF PULSED TUNABLE UV DYE LASER This chapter describes the experimental setup and the fluorescence lifetimes obtained when mixtures of DPA and TbCl 3 in different solutions were excited with a low PRF tunable UV dye laser (266 nm). Usi ng a monochromator, the fluorescence signal was resolved into its four different lines or components. The lifetimes of each of these lines were measured separately, and to the best of our knowledge, this is the first time that these measurements have been reported. Previous measurements had determined the lifetime of the combined fluorescence si gnal, but not of the individual lines. However, while some of the stronger fluores cence lines were measured using the low PRF tunable dye laser, several of the weaker lines could not be measured with higher accuracy. 6.1 Experimental Set-Up for the Life time Measurements Using the Dye Laser The initial fluorescence lifetime of the DPA-TbCl 3 complex was measured using a low PRF UV laser as described earlier. Fi gure 6.1 shows the schematic diagram of the setup used to measure the fluorescence lifetimes of the DPA samples, and Fig. 6.2 shows a photograph of the system. A pulsed nitr ogen pumped, frequency doubled tunable dye

PAGE 140

115 Sample Mirror GratingSpectrometer(CVIDigikrom240Monochromator)PMT Os cilloscope SiDetector LSILaser NitrogenPumpedFrequencyDoubledDyeLaser(13Hz;266nm,3nsPulseLength) NLaserDyeLaserf-doubling Module PMTsPowerSupply Aperture FrequencyDoubledBeam FundamentalBeam Aperture Figure 6.1 Schematic of the lifetime measurement experimental apparatus using the low PRF tunable dye laser.

PAGE 141

116 Figure 6.2 Photograph of the complete lifetime measurement experimental setup. Note: the dye laser and the 4 th Harmonic Nd:YAG laser were setup such that either experiment could be carried out without needing major rearrangements. This photograph shows the complete set up, the second half of which (i.e. the one that includes the 4 th Harmonic Nd:YAG) is described in the following chapter.

PAGE 142

117 laser (Chapter 5, Section 5.1.1) was used as the excitation source, at a frequency of approximately 13 Hz and an output wavelengt h of 266 nm. The UV output from the dye laser was focused into the center of a quart z cell, using a quartz lens, which contained about 3 ml of the sample. Perpendicular to the incident beam side of the cuvette, a pair of collection lenses gathered the fluorescence signal from the sample and focused it onto the entrance slit of a grating spectrometer (C VI Digikrom 240 Monochromator). A PMT (Photo Multiplier Tube) detector module (H amamatsu Model # H6780-03) was placed at the exit slit of the monochromator. The signal from the PMT was then sent to a 500 MHz oscilloscope (Lecroy, Waverunner LT 344), w hose specifications are in Table 6.1. The trigger signal for the oscilloscope was the fundamental beam from the dye laser, which was focused onto a fast silicon photo-de tector (New Focus, Model # 1621). The data was then transferred to a PC for furthe r analysis and storage via a GPIB interface It is important to note that the dye laser was tuned so that the frequency doubled beam had a wavelength of 266 nm, in order to later compare these lifetimes with those obtained when using the high PRF 4 th harmonic (266 nm) Nd:YAG laser (Chapter 7). 6.1.1 Monochromator The monochromator that was used to resolve the fluorescence signal of the sample into its different components was a CVI Digikrom 240 Monochromator. This monochromator could be used in a manual or a computer-controlled mode. The control electronics were internal and calibration was automatic. The monochromator had a modified Czerny-Turner configuratio n, with three different gratings mounted on a rotating platform. The efficiency of the gr ating that was used in our experiments is

PAGE 143

Bandwidth (-3dB): 500 MHz Sampling Rate: 500 MS/s Vertical Resolution: 8 bits Sensitivity: 2 mV 10 V/div Processor: 160 MHz PowerPC Memory: 16 Mbytes Table 6.1 Specifications for the Lecroy oscilloscope (Waverunner LT344). Figure 6.3 Grating spectrometers (CVI Digikrom 240) grating efficiency. 118

PAGE 144

119 shown in Fig. 6.3. Although it is not shown in the graph, the range of wavelengths that was covered by this grating was about 300 nm to 900 nm. Since spatial resolution was not a priority in this experiment, and the si gnal from some of the transitions were quite weak, the entrance slit and the exit slits were set to a width of 2 mm each to increase the transmitted light. 6.1.2 PMT Module To collect and temporally resolve the fluorescence signals, a PMT module (Hamamatsu, Model # H6780-03) was positione d flat against the exit slit of the monochromator. The module consisted of a PMT detector and a high voltage power supply, both enclosed in a small package (5cm 2cm 2cm). A supply voltage of 12 V was provided to the module. A control volta ge (0-1 V) was applied to the module which varied the output of the output of the high voltage of the pow er supply, thus controlling the gain of the detector. Fig. 6.4 (a) and (b) show the gain curve and the spectral response curve for this PMT module, respectivel y. The typical control voltage used was around 0.8 V (0.75 V to 0.85 V), which correspo nds to an applied high voltage of about 800 V and a PMT gain of approximately 7 10 5 Table 6.2 lists the specifications for this PMT detector. For a maximum anode current of 100 A and using the PMT gain value, the current at the photocathode can be calculated to be 0.14 nA (100 5 ) or 8.9 8 electrons/s. At 266 nm the energy per photon can be calculated to be 7.47 -19 J. The maximum average power that can impinge on the PMT photocathode can be calculated to be 6.65 -10 W (7.47 10 -19 J 8.9 10 8 electrons/s). For 10 Hz repetition rate of the

PAGE 145

(a) (b) Figure 6.4 Gain as a function of control voltage (a) and sensitivity as a function of wavelength (b) for the Hamamatsu PMT (Model # H6780-03). Note: the curves marked as -03 correspond to the module used in our setup. 120

PAGE 146

121 Effective Area 8mm Supply Voltage 11.5 V 15.5 V Control Voltage 0.25 V 0.95 V Min Wavelength 185 nm Max Wavelength 650 nm Rise Time 0.78 ns Table 6.2 Specifications for the Hamamatsu PMT.

PAGE 147

dye laser, the maximum signal energy that can be detected by the photocathode safely was about 66.5 pJ, i.e. 6.65 -10 W/10 Hz. For the 4 th harmonic Nd:YAG laser, with an output of 266 nm and a repetition rate of 8.5 KHz, which was used in a similar setup (see Chapter 7) the maximum signal energy per excitation pulse that could be safely detected by the photocathode was about 0.08 pJ. 6.1.3 Fast Silicon Detector A fast silicon photodetector (New Focus, Model # 1621) was used to detect the laser pulse and was used as a trigger signal. The detectors diameter was 0.8 mm. The typical rise time for this detector was 1 nanosecond (for 50 load resistance setting). A plot of responsivity for this detector is shown in Fig. 6.5. 6.2 Experimental Procedure The fluorescence lifetime, of the emission lines was measured by recording the fluorescence intensity as a function of time and determining the decay time to the 1/e intensity point. To determine the value of the natural logarithm of the fluorescence intensity as a function of time was plotted using an Excel or Matlab program. The fluorescence lifetime of the sample was then calculated by taking the negative inverse of the slope of the semi-log plot, or, 1Slope of semi-log plot (6.1) 122

PAGE 148

Figure 6.5 Plot of responsivity as a function of wavelength of the fast silicon photodetector (New Focus, Model # 1621) used as a trigger. 123

PAGE 149

124 The process of lifetime calculations is expl ained in more detail in Appendix B. In order to find the fluorescen ce lifetime of the samples, the average of 100 runs of the fluorescence intensity as a function of time was often used. All the data presented here have been normalized to the lasers power, the monochromators grating effi ciency at the relevant wa velengths, and the PMTs sensitivity at the relevant wavelengths. The measurements were also performed in a darkened room in order to minimize background noise from ambient light. 6.3 Solution Preparation Three different solvents were used in order to dilute diffe rent proportions of terbium and dipicolinic acid. [0.01M] HCl was prepared by diluting [12N] HCl (Fisher Scientific, 34%-37%) with distilled water. A solution of [1M] sodium acetate buffer (SAB), pH 5.6, was created by mixing [1M] sodium hydroxide with [1M] acetic acid until a pH of approximately 5.6 was obtained. Distilled water by itself was also used as a solvent. Once the solvents were prepared, terbium chloride hexahydrate (TbCl 3 .6H 2 O) (Sigma-Aldrich, 99.9% purity), was added to each of the solvents in order to make a solution with a final concentration of [50 M] TbCl 3 in it. Next, DPA (Sigma Aldrich, Inc.) was added to each of the previous solutions to form final solutions that consisted of [50M] DPA with [50M] TbCl 3 in distilled water; [50M] DPA with [50M] TbCl 3 in [0.01M] HCl; and [50M] DPA with [50M] TbCl 3 in [1M] SAB (pH 5.6). These solutions were then evaluated in order to obtain their fluorescence lifetime. The pHs of

PAGE 150

125 the solvents that were used were as follows: distilled water pH ~ 7, [1M] SAB pH ~ 5.6, and [0.01M] HCl pH ~ 2. 6.4 Lifetime Data Using Low PRF Dye Laser The experimental apparatus given in the pr evious section was used to measure the fluorescence lifetime of DPA and TbCl 3 in different solutions. It is important to note that the pulse width of the dye laser used in this experiment, which was about 3 ns, was negligible in comparison with the time scale of the fluores cence decay of the samples, which was on the order of milliseconds. In addition, the speed of the oscilloscope and PMT detector response is also much fa ster than the fluorescence decay time. 6.4.1 Measured Fluorescence Lifetime of Mixtures of DPA with TbCl 3 in Different Solvents The first sample that was analyzed was a solution of [50 M] DPA with [50 M] TbCl 3 in distilled water. The measured fluor escence intensity as a function of time is presented in Fig. 6.6 (a). As can be seen fr om the figure, all four individual lines and the combined fluorescence signal show lifetimes of about 1 ms. The 618 nm fluorescence line had the weakest intensity. The lifetime is more easily seen as in Fig. 6.6 (b), which shows a plot of the natural logarithm of the fluorescence intensity as a function of time for the same solution. The lifetime for the st rongest lines is easily seen, but digitization noise (finite resolution) of the oscilloscope A/D converter and low signal-to-noise is evident for the weakest lines or at the end of the fluorescence tail.

PAGE 151

0 1 2 3 4 5 6 0 2 4 6 8 Time (ms)Fluorescence Intensity(Arbitrary Units) All fluorscence lines mixed together 543 nm line 488 nm line 581 nm line 618 nm line (a) 0 1 2 3 4 5 6 2 4 6 8 10 Time (ms)Ln (Fluorescence Intensity) 618 nm line 488 nm line 543 nm line All fluorescing lines mixed together 581 nm line (b) Figure 6.6 (a) Plot of fluorescence intensity as a function of time, and (b) semi-log plot of the fluorescence intensity as a function of time of a solution of [50 M] DPA with [50 M] TbCl 3 in distilled water. 126

PAGE 152

127 Similar measurements for the [50 M] DPA with [50 M] TbCl 3 in [0.01M] HCl are presented in Fig. 6.7, and that for [50 M] DPA with [50 M] TbCl 3 in [1M] SAB are shown in Fig. 6.8. As seen in the figures, the 543 nm line and the 488 nm lines are the strongest lines and exhibit a decay close to that of a str aight line, indicative of a single exponential decay in the semi-log plots. 4,25 The two weaker fluores cence signals, i.e. those corresponding to the 581 nm and the 618 nm lines, have significantly more noise and exhibit less of a straight line in their semi log plots. This can be indicative of a multiexponential decay, but in this case it is most probably due to the low S/N ratio which in turn is due to the relatively low average laser power. The semi-log plot from all the fluorescence signals mixed together also has a slight curvature; this, however, is probably due to the additive influence of the two weakest fluorescence lines. The fluorescence lifetimes of all of the lines mixed together, as well as for each fluorescence line by itself, were calculated and are listed in Table 6.3; a curve fitting routine in the Matlab software was used to estimate the slope and uncertainty value in the middle of the curve. For each of the samples, the lifetimes of the first three individual fluorescence lines are within uncertainty of each other. This is consistent with these electronic transitions emanati ng from the same upper level, i.e. 5 D 4 The lifetime measured for the weakest line is within this va lue considering the larger uncertainty in the measurement. It is interesting to compare the magnitude of the lifetimes of these samples with respect to the solvents properties. The lif etimes for these samples have the following trend: SAB > DW > HCl with the numerical value for all of the fluorescence lines mixed

PAGE 153

0 1 2 3 4 5 6 0 2 4 6 8 Time (ms)Fluorescence Intensity(Arbitrary Units) All fluorescence lines mixed together 543 nm line 488 nm line 581 nm line 618 nm line (a) 0 1 2 3 4 5 6 2 4 6 8 10 Time (ms)Ln (Fluorescence Intensity) 618 nm line 581 nm line 488 nm line 543 nm line All fluorescence lines mixed together (b) Figure 6.7 (a) Plot of fluorescence intensity as a function of time, and (b) semi-log plot of the fluorescence intensity as a function of time, of a solution of [50 M] DPA with [50 M] TbCl 3 in [0.01M] HCl. 128

PAGE 154

0 1 2 3 4 5 6 0 2 4 6 8 Time (ms)Fluorescence Intensity(Arbitrary Units) All fluorescence lines mixed together 543 nm line 488 nm line 618 nm line 581 nm line (a) 0 1 2 3 4 5 6 2 4 6 8 10 Time (ms)Ln (Fluorescence Intensity) All fluorescence lines mixed together 543 nm line 488 nm line 581 nm line 618 nm line (b) Figure 6.8 (a) Plot of fluorescence intensity as a function of time, and (b) semi-log plot of the fluorescence intensity as a function of time, of a solution of [50 M] DPA with [50 M] TbCl 3 in [1M] SAB (pH 5.6). 129

PAGE 155

130 (ms) Distilled Water [0.01M] HCl [1M] SAB All 0.89 0.12 0.75 0.01 1.11 0.01 488 nm 0.88 0.01 0.81 0.01 1.14 0.01 543 nm 0.89 0.01 0.76 0.01 1.08 0.01 581 nm 0.87 0.01 0.70 0.01 1.16 0.01 618 nm 1.43 0.08 1.09 0.13 5.37 0.57 Table 6.3 Measured Fluorescence lifetimes of solutions of [50 M] DPA with [50 M] TbCl 3 in the indicated solvents when excited with the dye laser, exct. = 266 nm.

PAGE 156

131 together for each of these solution being 1.11 ms for the solution in [1M] SAB, 0.89 ms for the solution in distilled water, and 0.75 ms for the solution in [0.01M] HCl. As stated earlier, lifetimes of terbium doped DPA complexes are very sensitive to the nature of their environment, especially to the number of water molecules occupying the inner coordination sites. 38,66 Based on this, we would expect to see the lowest lifetime values, as well as the lowest fluorescence intensities, for the solutions mixed in distilled water. In these experiments the lowest lifetime values belong to the solutions mixed in [0.01M] HCl. This could be an effect of the low pH of the [0.01M] HCl, in addition to that of the OH vibrational oscillators. According to previously published results, 65 the quantum yield of [Tb(DPA) 3 ] 3in water, which is proportional to the fluorescence lifetime, stea dily increases as a function of pH until it reaches a maximum around pH ~ 7.5. In this case, the order of the lifetimes magnitudes does not follow that of the solvents pHs, which is pH DW > pH SAB > pH HCl This could be due to the fact that the DPA to TbCl 3 ratio is 1:1 in these solutions of [50M] DPA with [50M] TbCl 3 and therefore, the trisspecies, i.e. [Tb(DPA) 3 ] 3, which tends to form in the excess of DPA, as described earlier in Chapter 2, was not dominantly present. Lastly, as can be seen from all of the figures, there exist a distinctive build up time before the start of the fluorescence decay. This may be due to the transfer of energy from the DPA to the Tb 3+ ion. A visual inspection of Fi g. 6.6 yields an estimate of about 0.4 ms for this transfer time. This transfer time has been previously suggested to be on the order of 2.5 ms, but these results were obtained under different solvent conditions. 25

PAGE 157

132 6.4.2 Measured Fluorescence Lifetime of TbCl 3 in Distilled Water A sample of [50 M] TbCl 3 in distilled water was excited using the low PRF 266 nm laser pulse from the dye laser. The fluorescence intensity as a function of time for this sample is plotted in Fig. 6.9 (a), and its semi-logarithmic plot is shown in Fig. 6.9 (b). As can be seen in the figures, the 488 nm fluorescence line was extremely weak and was not able to be measured. In addi tion, the 618 nm line was also very weak, and the PMT voltage was increased to help increase this signal. This resulted in an increase in the detected fluorescence signal of the 618 nm line which was not accounted for, i.e. normalized, in the data presented here. While this influenced the relative strength of the 618 line in the Fig 6.9 (a), it does not affect the slope of the cu rve in Fig. 6.9 (b) which is the logarithmic plot of the intensity. The lifetimes of each of th ese lines were calculated using the inverse slope of the semi-log plot; the lifetim e of all the lines mixed together was found to be (0.36 0.02) ms. The lifetimes of the weaker 488 nm fl uorescent line in distil led water could not be detected. Due to the weak signals with TbCl 3 in HCl and in SAB, further experiments to measure the lifetimes of TbCl 3 in the other solvents were not carried out. 6.5 Lifetime Analysis Table 6.4 shows the compilation of lifetim es measured using the tunable UV dye laser system. As can be seen only lifetimes of TbCl 3 in distilled water are presented.

PAGE 158

0 1 2 3 4 5 6 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Time (ms)Fluorescence Intensity(Arbitrary Units) 581 nm Line All fluorescence lines mixed together 543 nm Line 618 nm Line 488 nm Line (a) 0 1 2 3 4 5 6 0 2 4 6 8 Time (ms)Ln (Fluorescence Intensity) All fluorescence lines mixed together 618 nm line 581 nm line 543 nm line (b) Figure 6.9 (a) Fluorescence intensity as a function of time, and (b) semi-log plot of fluorescence intensity as a function of time of a solution of [50M] TbCl 3 in distilled water. 133

PAGE 159

134 Similar measurements were initially tried for TbCl 3 in HCl and SAB solutions, but experimental difficulties precl uded obtaining these results. However, later experiments presente d in this thesis yielded good value for for the mixture of DPA and TbCl 3 in all of these solutions. As mentioned previously, the fluorescence signal from aqueous terbium was very difficult to obtain. This is because of the relatively good radiationl ess energy transfer from the terbium ion to its aqueous environment, i.e. the OH oscillators, and its low absorption cross section. The fluorescent lines lifetimes that we were able to detect do fall within uncertainty of each other (this is consiste nt with all of the transitions in the Tb 3+ ion starting at the same upper excited level). These lifetime values are within the ranges of previously published data for which the lifetime of the 5 D 4 emitting state of the terbium (III) ion falls in the 0.4 ms to 5.0 ms range. 44,67,68 Figure 6.9 for TbCl 3 alone shows almost zero buildup time for the fluorescence to start. This is different than the lifetime data for the DPA-TbCl 3 complex (Fig. 6.6 (a)), due to this being a direct excitation of the Tb 3+ ions 5 F 1 level. It may be added that the above measured lifetimes were probably not modified due to radiation trapping. For example, all of the DPA-Tb complex absorption values near the fluorescence emission wavelengths ha d absorbance (A) values of less than 0.02 for the 1 cm path length (see Fig. 4.12, and Fig. A.7). A value of A = 0.02 yields a transmission of 96% or absorption of 4%; Fi g. A.8 for SAB yields values about 2 to 3 times greater. Radiation trapping only occu rs when the absorption is much stronger, approaching 90% or more for the optical path.

PAGE 160

135 It is important to mention that the slight variations in the fluorescence lifetimes of the four different fluorescence lines from the DPA-Tb complex, as well as from the terbium ion by itself, could be due to slight variations in the lifetime of the termination level (see Section 2.1.4.3.1) and the possibili ty that the different lines originate or terminate on similar levels but which may ha ve different degeneracy or parity which could then influence the overall transiti on rate or lifetime of the transitions. In these data we averaged 100 runs and then obtained the lifetime. The results maybe slightly different if the reverse or der was made, but this would depend upon the correlation properties of the noise.

PAGE 161

136 (ms) TbCl 3 in: Distilled Water [0.01M] HCl [1M] SAB All 0.36 0.02 --488 nm ---543 nm 0.34 0.02 --581 nm 0.43 0.07 --618 nm 0.48 0.15 --Table 6.4 Lifetimes of TbCl 3 in distilled water. The lifetimes that we were unable to obtain are indicated with a dashed entry.

PAGE 162

137 CHAPTER 7. FLUORESCENCE LIFETIMES OF SOLUTIONS OF DPA AND TbCl 3 WHEN EXCITED WITH A HIGH PRF AND MODULATED PULSED UV LASER In order to detect the weaker fluorescen ce lines, especially the 618 nm line, we needed a higher average power UV laser with which to excite the sample. As such, a high-PRF UV 266 nm microchip laser was used which had about several orders of magnitude greater average power than that of the pulsed low-PRF dye laser. This chapter describes the experi mental setup and the measured fluorescence lifetimes obtained when mixtures of DPA and TbCl 3 in different solutions were excited with a high PRF microchip 4 th harmonic Nd:YAG laser (266 nm). However, because of the quasiCW nature of the high-PRF (8.5 kHz) 266 nm laser, additional slower modulation of the laser excitation pulse train was required to al low detection of the 1 ms fluorescence decay times. These experiments and results are presented in the following sections. 7.1 Experimental Setup for the Lifetime Measurements Using the Microchip Laser A high pulse repetition frequency, 4 th harmonic Nd:YAG microchip laser (JDS Uniphase) with an output wavelength of 266 nm in combination with an optical chopper, replaced the nitrogen pumped dye laser as the excitation source in the setup described in Chapter 6, and is shown in Fig. 7.1.

PAGE 163

GratingSpectro-meterPMT4thHarmonicNd:YAGLaser(266nm,8.5KHz) OscilloscopeQuartzCuvette(Sample) OpticalChopper(0Hz-4000Hz) PMTsPowerSupply BeamStopper SiDetector Figure 7.1 Schematic of lifetime measurements experimental setup using the high PRF pulsed laser and the optical chopper. 138

PAGE 164

139 The laser beam was routed using two alignment mirrors, passed through an optical choppers blades, and then focused in to the center of the sample cuvette using a fused silica lens. Part of the beam was split off to trigger the fast silicon detector. Except for the excitation source and chopper, th e rest of the setup remained as described in Section 6.1. 7.1.1 Optical Chopper The optical chopper (Stanford Research Sy stems, Model# SR-540) was used to modulate the excitation beam and used one of the two following chopper blades: one with 6 outer slits (slit width of approximately 2 cm) for frequencies up to 400 Hz, and one with 30 outer slits (slit width of approximate ly 4 mm), for frequencies of up to 3.7 kHz (Fig. 7.2). For the lifetime experiments desc ribed in this chapter, the blade with 30 outer slits was used at the indicated frequencies. The laser beam passed through the center of the outer slits. The optical chopper also produces an optically generated TTL reference signal which was used for triggering the oscilloscope; th e fast silicon detector was also used to trigger the oscilloscope. 7.2 Effect of Optical Chopper on Excitation and Fluorescence The laser beam was modulated using the optical chopper at di fferent frequencies in order to modulate the optical excitation at a rate closer to that of the expected 1 ms fluorescence decay time. However, the physic al act of chopping has an effect on the

PAGE 165

Figure 7.2 Optical choppers blades used to modulate the laser beam. The blade on the left hand side of the picture was used for the experiment described in this chapter. (From Stanford Research Systems website at http://www.thinksrs.com/products/SR540.htm). 140

PAGE 166

141 geometry of the incident beam, which conseque ntly affects the fluorescence signal. This geometrical effect is explained in following section. 7.2.1 Finite Beam Size and Geometrical Effect of Chopper Wheel on Incident Beam The geometrical effect of the finite size of the laser beam and the speed of the chopper wheel was modeled to determine its effect upon the fluorescence intensity time dependence. The beam was usually focu sed down to a small circular spot size ( 1 mm), before going through the chopper. While th e exact beam distribution was not measured, it appeared to be approximately close to a TEM 00 Gaussian mode with little if any side lobes within the pattern. As such, we m odeled the laser beam before passing through the chopper by a Gaussian distribution as shown in Fig. 7.3. In order to explore this effect, different parameters were chosen to descri be the Gaussian beam size and placement using a variance, 2 and a possible offset, Figure 7.4 shows different Gaussian distributions in terms of the size compared to 1 mm and a possible offset. For our experiments, the beams diameter was approxi mately 1 mm, and the widths of each of the chopper blades slits were approximately 4 mm The beams geom etrical distribution after passing through the choppers blades was graphed by taking into account the interaction of the Gaussian di stribution of the original beam with the effect that the blades slits had on it by integrating the spatial Gaussian beam profile with that of the moving slit function. The resultant function, called the Cumulative Distribution Function (CDF), 69 has been calculated for Gaussian beams before, and is shown in Fig. 7.5. The resultant effect of using the CDF function is to modify the rising and fall-off

PAGE 167

LaserBeamOpticalChopper Figure 7.3 Diagram of Gaussian laser beam (TEM 00 ) passing through a moving chopper blade. 142

PAGE 168

Figure 7.4 Plot of Gaussian distributions with different variances. We have used the plot with a variance of 1 to represent the TEM 00 laser beam from microchip laser before going through the choppers blade, i.e. the original beam (from http://en.wikipedia.org/wiki/Normal_distribution). Figure 7.5 Cumulative distribution function (CDF) for a Gaussian distribution (from http://en.wikipedia.org/wiki/Normal_distribution). 143

PAGE 169

144 patterns of the beam afte r passing through the blades. This effect is shown in Fig. 7.6 for our experimental case of 1 mm beam size ( of 1 mm) and slit width of 4 mm In the experiments presented in this ch apter, the choppers frequency was kept at 50 Hz. At this frequency, it took approxi mately 20 ms for one period of the chopper, defined as the start of one open slit to the end of next closed slit (8 mm), to pass an infinitesimally small point. This means that every millimeter of the chopper blades width takes about 2.5 ms to pass an in finitesimally small point at 50 Hz. From Fig. 7.6 it can be seen that from the time the chopper allows the beam to first pass through until the time the full intensity of the beam is able to go through the slit, it takes approximately 2.5 ms. Since this is a symmetric effect, it also takes approximately 2.5 ms from the time the chopper first blocks off the beam until the time when the intensity of the beam is zero after the blade. This analysis deals only with the spatial geometry of the beam and the blad e, as well as the times involved in this geometric interaction which are much longer than the pulse width or the PRF period of the laser. 7.2.2 Effect of Chopper Speed and Laser PRF on the DPA-TbCl 3 Fluorescence Signal When the laser excitation was turn ed on (without the optical chopper) the fluorescence emission was steady, as seen in Fi g. 7.7. As can be seen, the fluorescence signal did not increase or decay over this 45 ms time period. In this data, the sample used was a solution of [50 M] DPA with [50 M] TbCl 3 in SAB.

PAGE 170

CHOPPERSWINDOW:CLOSED CHOPPERSWINDOW:OPEN 0.11.00.90.80.70.60.50.40.30.20.0 1.00.0 Figure 7.6 Plot of the 1 mm diameter laser beams intensity after interacting with the choppers slits. Also shown in the upper part is the turning on and off of the chopper blades window. Note the similarity between the build up and the fall down of this graph as compared to the CDF in Figure 7.5. 145

PAGE 171

0 5 10 15 20 25 30 35 40 0 2 4 6 8 10 12 14 Time (ms)Fluorescence Intensity Figure 7.7 Fluorescence intensity as a function of time when using the high PRF microchip laser with the chopper turned off. 146

PAGE 172

147 However, when the chopper was turned on and was operating at a frequency of 50 Hz, the incident beam was blocked every 10 ms for a duration of 10 ms. This can be seen in Figure 7.8 which shows two periods of the fluorescence signal, i.e. two periods of the chopper blade, when the sample had been excited with the chopped incident laser beam. A graphical summary of this process is shown in the experimental data shown in Fig. 7.9, where in part (a) the pulses from the microchip laser before they have passed through the chopper are shown, in part (b ) a TTL reference signal from the 50 Hz chopper driver is shown, and in part (c) the measured fluorescence signal obtained from the excitation of the sample is shown. 7.3 Experimental Meas urements of DPA-TbCl 3 Lifetimes Using Chopper Technique The optical chopper UV laser excitation just described was used to stimulate the fluorescence emission from different solutions containing DPA-TbCl 3 Figure 7.10 (a) shows the fluorescence intensity signal and Fi g. 7.10 (b) shows the semi-log plot as a function of time for a sample of [50M] DPA with [50M] TbCl 3 in distilled water for all the fluorescence lines mixed together and for each of the fluorescence lines resolved separately. As can be seen, after the chopper slits are completely closed then the fluorescence decay appears to be linear. For clarity, this figure shows marks indicating the corresponding positions on the signals wh ere the choppers blade was open, i.e. letting the beam through, and where the choppers blade was closed, i.e. when the

PAGE 173

148 Figure 7.8 Fluorescence signal of sample when excited with a modulated incident beam, i.e. laser beam that has passed through the optical choppers blade.

PAGE 174

149 Figure 7.9 Plot of amplitude as a function of time of the high PRF 266nm 4 th harmonic Nd:YAG lasers output (a), as compared to the choppers TTL signal at 50 Hz (b), and the samples fluorescence (c).

PAGE 175

(a) (b) Figure 7.10 (a) Fluorescence intensity as a function of time, and (b) semi-log plot of the normalized fluorescence decay lines from [50M] DPA with [50M] TbCl 3 in distilled water (f chopper = 50Hz). 150

PAGE 176

151 blade was blocking the beam. Each graph has also been marked with the time that it takes for the beam to be completely blocked off. It should be noted that the intensity of th e weakest fluorescence line in this figure, which is approximately 50 a.u., is about 100 times stronger than compared to the fluorescence intensity of the same sample when excited with the low PRF dye laser, shown in Fig. 6.6 (a), which was about 0.5 a.u. This increase in the fluorescence intensity is reflected in the greater accuracy of the lifetime measur ement for the weaker lines. Similar fluorescence intensity measurements as a function of time are shown in Figs. 7.11 and 7.12 for a sample of [50M] DPA with [50M] TbCl 3 in [0.01M] HCl for and for [50M] DPA with [50M] TbCl 3 in [1M] SAB, respectively. The measured lifetime values of these sa mples, for each of the fluorescent lines and all of them mixed together are listed in Table 7.1. The uncertainties shown in Table 7.1 were calculated using variations of the slope due to a 2-po int analysis of the data (see Appendix A). A more accurate technique using a curve fitting routine in Matlab yielded uncertainties about a factor of 3 lowe r for the strongest lines. These values are similar to those obtained earlier using the lowPRF pulsed dye laser as given in Table 6.3. As was the case for the lifetime of the sample s when excited with the dye laser (Chapter 6), the magnitudes of the lifetimes when exci ted with the microchip laser (in combination with the chopper) have the same order, i.e. SAB > DW > HCl This will be discussed in Chapter 10.

PAGE 177

(a) (b) Figure 7.11 (a) Fluorescence intensity as a function of time, and (b) semi-log plot of the normalized fluorescence decay lines from [50M] DPA with [50M] TbCl 3 in [0.01M] HCl (f chopper = 50Hz). 152

PAGE 178

(a) (b) Figure 7.12 (a) Fluorescence intensity as a function of time, and (b) semi-log plot of the normalized fluorescence decay lines from [50M] DPA with [50M] TbCl 3 in [1M] SAB (f chopper = 50Hz). 153

PAGE 179

154 (ms) Distilled Water [0.01M] HCl [1M] SAB All 0.93 0.04 0.77 0.05 1.07 0.03 488 nm 0.89 0.04 0.75 0.05 1.04 0.05 543 nm 0.95 0.04 0.78 0.05 1.05 0.04 581 nm 0.91 0.05 0.77 0.05 1.07 0.04 618 nm 1.04 0.05 0.83 0.06 1.04 0.03 Table 7.1 Table of life times of solutions of [50M] DPA with [50M] TbCl 3 in the indicated solvents when excited with the modulated microchip laser laser, exct. = 266 nm.

PAGE 180

155 7.3.1 Lifetime Dependence on DPA and TbCl 3 Concentrations In order to better view any possible trend in the lifetime values, a three dimensional plot of the lifetimes of the fl uorescence of the DPA-Tb complex was made for each of the lines for different concentrations of [DPA] and [TbCl 3 ] in distilled water. This data is shown in Fig. 7.13. Figure 7.13 (a) through (e) shows the lifetimes from all of the samples fluorescence signals mi xed together, from the samples 488 nm fluorescence line, the 543 nm fluorescence li ne, the 581 nm fluorescence line, and the 618 nm fluorescence line, respectively. As can be seen, a general upwards trend in seems to exist as we approach an increasing DPA concentration and a decreasing TbCl 3 concentration at the same time. Similar plots were generated for the DPA-TbCl 3 samples in the low concentration [0.01M] HCl solvent (Fig. 7.14), and for those in the [1M] SAB solvent (Fig. 7.15). In these solvents the values of the lifetimes do not seem to be strongly dependent upon the DPA and the TbCl 3 concentrations. This seems to be different than that observed for the DPA-TbCl 3 solution in distilled water shown in Fig. 7.13. However, further work would be required to better quantify this effect and to reduce the measurement uncertainty.

PAGE 181

(a) (b) (c) (d) (e) Figure 7.13 Plots of lifetime as a function of DPA and TbCl 3 concentration in distilled water. A sample error bar is shown in each of the figures at around the [25M] DPA with [5M] TbCl 3 156

PAGE 182

(a) (b) (c) (d) (e) Figure 7.14 Plots of lifetime as a function of DPA and TbCl 3 concentration in [0.01M] HCl. A sample error bar is shown in each of the figures at around the [25M] DPA with [5M] TbCl 3 157

PAGE 183

(a) (b) (c) (d) (e) Figure 7.15 Plots of lifetime as a function of DPA and TbCl 3 concentration in [1M] SAB. A sample error bar is shown in each of the figures at around the [25M] DPA with [5M] TbCl 3 158

PAGE 184

159 CHAPTER 8. NEW FLUORESCENCE LIFETIME MEASUREMENT TECHNIQUE USING TRANSIENT FLUO RESCENCE SPECTROSCOPY (TFS) As explained in the previous chapter, th e fluorescence signals that were obtained when using the microchip laser as the excita tion source were modulated due to the optical chopper. During these studies, it was no ticed that the fluor escence intensity and modulation depth varied with th e speed of the chopper. An in-depth analysis of the fluorescence energy levels and excitation transitions led us to investigate this effect more. As such, we investigated the relation be tween the choppers frequency and the lifetime signals behavior experimentally and theore tically by numerically computing a simple energy level model of the DPA-Tb complex. We have found excellent agreement between our numerical solutions of the transient energy rate equations for the DPA-Tb energy levels and experimental measurements. In addition, these numerical simulations have provided curve fitting to the experiment al data which yields a fitted fluorescence lifetime that is very close to that measured in the previous chapter. This new lifetime technique is called Transient Fluorescence Spectroscopy (TFS), which is the dynamic, time dependent solution of the transient ra te equations, and may provide some new advantages over other lifetime t echniques or provide another way to measure the lifetime. Details of this are presented in the following sections.

PAGE 185

160 8.1 Experimental Observation of the E ffect of Chopper Modulation on Approach to Fluorescence Saturation When a sample containing a mixture of DPA and TbCl 3 was excited with the modulated microchip laser, th e fluorescence signal was also modulated. For the purpose of lifetime measurements, when the frequenc y of the chopper was maintained at around 50 Hz the signal was able to fully decay to zero intensity. However, as the choppers frequency was increased, the amplitude of the fluorescence signal decreased significantly. This is shown in Fig. 8.1 where the fluores cence intensity as a function of time is shown for several different chopper speeds. As can be seen, as the chopper speed increased, the maximum extent of the modulated fluorescence decreased. One explanation was that due to the rela tively long fluorescence lifetime of the samples, which are in the millisecond range, the higher chopper frequencies were not allowing enough time for the fluorescence to build up to its maximum or saturated intensity or to fully decay. The time available for the laser beam to reach the sample, i.e. when the choppers blade was in an open position, varied depending on the choppers frequency of rotation. Due to the symmetric pattern of the blade, this time was the same as when the choppers window was closed. The open/close (or on/off) times for the frequencies shown in Fig. 8.1 are as follows: in Fig. 8.1 (a) the blade was open for 10 ms for f chopper = 50 Hz (also, the blade was closed for 10 ms); this left enough time for the signal to fully decay to the zero intensity level. In Fig. 8.1 (b) the blade was open and closed for 5 ms in each case, for f chopper = 100 Hz; in this case the signal was able to decay down almost completely. In Fig. 8.1 (c) the blade was open and closed for 1ms in each case, for f chopper = 500 Hz; and for f chopper = 1000 Hz, the blade was open and closed

PAGE 186

(a) 50 Hz (c) 500 Hz 161 (b) 100 Hz (d) 1000 Hz Figure 8.1 Experimentally measured fluorescence intensity as a function of time of a sample of [50 M] DPA with [50 M] TbCl 3 in distilled water when excited with the chopper-modulated-microchip laser, for the chopper operating at four different frequencies.

PAGE 187

162 for 0.5 ms in each case. The signal appears to be reduced symmetrically, possibly due to the symmetry of the excitation waveform. However, this will be discussed more fully later for the asymmetr ical excitation case. To better understand the be havior of the maximum fluorescence level and the minimum fluorescence level observed, the chopp ers blades were modified to be nonsymmetrical with regard to the on-time and offtime. In one case every other slit in the 30 slit blade was taped off, thus cutting the number of open slits by half to 15, and, therefore, tripling the available decay time by blocking off the beam. This modification is shown in Fig. 8.2 (a). In another instan ce the 30-slit and 6-slit blades were physically superimposed so that the blade pattern shown in Fig. 8.2 (b) was created. The resulting modified 15 slit blade was open for 1 mm and closed for 3 mm, Fig. 8.2 (a). In the case of the superimposed blades the blades were open for 1 mm, closed for 1 mm, open for 1 mm, closed for 1 mm, open for 1 mm, and closed for 5 mm, Fig. 8.2 (b). The graphical representations of these new m odified blades for the intensit y as a function of time for 1 period of each blade are show n in Fig. 8.3 (a) and (b). By using these modified blades we were ab le to control or influence the peak and the trough of the fluorescence signal, as shown in Figs. 8.4 and 8.5. Figure 8.4 shows the resulting fluorescen ce signals obtained at 50 Hz and 500 Hz using the modified blade with 15 slits. One can see that even at 500 Hz, the beam was blocked off for long enough and thus the signal was able to decay to the same level as when the choppers frequency was at 50 Hz, while the build ups maximum still remained at a lower level at 500 Hz than when the chopper was at 50 Hz.

PAGE 188

Figure 8.2 (a) Original 30-slit chopper blade on the left, and modified 15 slit blade on the right; (b) original 6-slit chopper blade on the left, and 30-slit and 6-slit blades superimposed on the right. Note: only the outer slits in each blade were used for modulation purposes in these experiments. 163

PAGE 189

IntensityTime (a) IntensityTime (b) Figure 8.3 Chopper periods, or TTL signals, for the 15-slit modified blade showing 1-on and 3-off (a), and for the superimposed 30-slit and 6-slit blades showing a group of 3 on-off followed by 5-off (b); these correspond to Fig. 8.2 (a), and (b), respectively. 164

PAGE 190

0 10 20 30 40 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (ms)Signal Strength fChopper = 500 Hz fChopper = 50 Hz Figure 8.4 Fluorescence signal as a function of time for a DPA-Tb solution when excited with the chopper-modulated-microchip laser, for the chopper operating at two different frequencies, with a modified blade with 15 slits (1 run). 165

PAGE 191

0 10 20 30 40 0 200 400 600 Time (ms)Signal Strength 15 Slit blade Superimposed:6 slit balde & 30 slit blade Figure 8.5 Fluorescence signal as a function of time for a DPA-Tb solution when excited with the chopper-modulated-microchip laser, for the chopper operating at a frequency of 50 Hz (average of 100 runs). 166

PAGE 192

167 As a second test, the 30-slit blade was supe rimposed on the 6-slit blade, as in Fig. 8.2 (b), and the combination of the two was used to modulate the excitation beam. The resulting fluorescence signal is shown in Fig. 8.5 The fluorescence signal using the 15slit modified blade is also overplotted in the same graph. It may be noted that the solution used for Fig. 8.4 was a mixture of lo w DPA-Tb concentration in water, while the solution used for Fig. 8.5 was [250M] DPA with [250M] TbCl 3 in SAB. The intensity of the fluorescence signal whose excitation beam was modulated with the 15-slit modulated blade is slightly higher than the intensity of the fluorescence signal whose excitation beam was modulated with the supe rimposed 6-slit and 30-slit blades. At 50 Hz each slit in the 15-slit m odulated blade takes about 4.7 ms to pass an infinitesimally small point, while it takes one slit in the superimposed blade combination about 1.9 ms to pass an infinitesimally small point, for the slits width of 4 mm. This means that the sample whose excitation is being modulated with the superimposed modified blade has less than half the bu ild up time as compared to the one whose excitation is modulated using the 15-slit modifi ed blade. Further measurements using the asymmetrical chopper patterns at different frequencies ar e presented in Appendix C. 8.2 Theoretical Modeling of DPA and TbCl 3 Energy Level Rate Equations and Influence on Fluorescence Intensity Several previous studies have been made regarding the saturation effects of laser induced fluorescence usi ng pulsed laser excitation 70-73 However, nothing has been reported, as far as we know, on the use of a simulation as a possible means of predicting the lifetime of fluorescing samples. With the ultimate goal of predicting the lifetime of

PAGE 193

168 the samples, we set out to try to model th e fluorescence behavior of the samples when they were excited with a modulated laser beam. To this e nd a theoretical model made of a simple energy level diagram representing the DPA molecule, the Tb 3+ ion and the energy transfer between them was developed, and the transient rate equations for this model were solved. 8.2.1 Energy Level Model for DPA-TbCl 3 Fluorescence As it was shown in Chapter 4, the DP A molecule in solution has a strong absorption peak in the UV, while the Tb 3+ in solution is practically lacking the presence of any absorption bands. It was also shown in the same chapter that the absorption cross section for the DPA-Tb complex in solution is within experimental uncertainty of the value for the absorption cross section of th e DPA molecule, providi ng further proof that in this complex the DPA molecule seems to be the energy donor and the Tb 3+ ion the energy acceptor, as well as the fluorescing spec ies, as shown by its signature fluorescence spectrum. As stated in Chapter 2, the DPA molecule s lowest lying triplet state has been reported as 24,272 cm -1 by Lima and at 27,050 cm -1 by Latva. 35,36 Figure 8.6 shows the energy levels for the triplet state of the DPA molecule and for the Tb 3+ ion. As can be seen from this figure, de pending on the true location of the DPAs triplet state, the transfer of energy between the DPA mol ecule and the terbium ion may take different paths. If DPAs triplet state is indeed at around 27,000 cm -1 then this transfer most probably takes place via the 5 L 10 or 5 G 6 terbium levels, which sit at 26,946 cm -1 and at

PAGE 194

1.4x1041.5x1041.6x1041.7x1041.8x1041.9x1042.0x1042.1x1042.2x1042.3x1042.4x1042.5x1042.6x1042.7x104Energy(cm-1)DPAsTripletStateTb3+ExcitedStates 5G65L10LatvaLima 5D4 Figure 8.6 Comparison of the energy levels of terbium ion 42 with the lowest lying triplet state of the DPA molecule, as measured by Latva 35 and by Lima 36 169

PAGE 195

26,405 cm -1 respectively. 42 From the 5 G 6 level in the terbium ion, the electrons are then transferred to the 5 D 4 fluorescing level, which sits at around 20,500 cm -1 from where they decay down to the ground state (the 7 F series, not shown in Fig. 8.6) in the form of fluorescence. However, if the state is at around 24,000 cm -1 then this could be a direct transfer between the DPAs triplet state and the terbium ions 5 D 4 The energy transfer between the DPA molecule and the terbium ion has been usually assumed to be nonradiative. 25 Taking these levels and the assumed transfer processes noted, a simple energy level diagram for the DPA and Tb complex (DPA-Tb, or in Section 2.3) was developed, and is shown in Fig. 8.7. 3-2nnTb(dpa) Figure 8.7 shows a relatively simple energy level model for the DPA-Tb complex and the assumed excitation and de-excitation rates as shown. In this model, the population density of the DPA-Tb levels is given by N 0 N 1 N 2 and N 3 The DPA-Tb molecule is excited (I) to its lowest lying triplet level, N 1 with excitation intensity I and absorption cross section ; stimulated emission I back down to the N 0 is also shown. The energy is then transferred nonradiatively to the 5 G 6 energy level in the terbium ion, also represented here as N 1 from where it relaxes down to the 5 D 4 terbium ion energy level, N 2 with a transfer rate of 12 The electrons then fluoresce from this level to one of the highest energy levels of the terbium ions 7 F series, N 3 at a rate of 23 From this level they decay back down to the ground state, N 0 at a rate of 30 The inverse of the 23 transfer rate is equivalent to the fluorescent lifetime, of the sample. The fluorescence 170

PAGE 196

N0N1N2N3DP A T b II(Fluorescence) Figure 8.7 Simple theoretical energy level model representing the energy levels of the DPA-Tb complex. 171

PAGE 197

intensity of the sample is proportional to the population density of the fluorescing level, N 2 14 The rate equations for this system are given by 0103031 011212 1212323 232303dN=ININNdtdN=ININNdtdN=NNdtdN=NNdt ,,,, (8.1) (8.2) (8.3) (8.4) and 0123N=N+N+N+N (8.5) The approximate values for some of these parameters can be obtained from previous experimental measurements. The value used for 23 was 1053/s, which is the inverse of the average of the experimentally obtained lifetime of 0.95 ms for a solution of [50 M] DPA with [50 M] TbCl 3 in distilled water, when using the microchip laser and chopper combination. The value used for the number of ground state molecules is 3 x 10 16 per cm 3 which corresponds to a concentration of [50 M] of DPA since this is the only species that contributes to the absorption of the incident light. For simplicity, we can assume that this also represents the concentration of the DPA-Tb complex molecules; this is valid if there is one DPA molecule bound to one Tb atom (see Section 2.3). The initial value for the remainder of the energy levels was zero, since at room temperature the molecule is in its ground state. The intensity of the microchip laser was experimentally measured to be 1.113 x 10 22 photons/cm 2 sec. The assumed value for the absorption cross section was found experimentally to be 1.94 x 10 -17 cm 2 /molecule, as 172

PAGE 198

173 explained in Chapter 4, Section 4.4.4. The values for the transfer rates, 12 and 30 were obtained by trial and error (i.e. curve fitting). 8.3 Effect of Varying th e Simulations Parameters The rate equations were numerically solved for dN 0 dN 1 dN 2 and dN 3 using the initial assumed or default values listed in Table 8.1. The transient solution was obtained numerically using a computer progra m written by the author and explained in more detail in Appendix D. The time incremen t, dt, was set for a value several orders of magnitude smaller than the other temporal processes in the fluorescence model (i.e. dt 0.04 ns). The transient solution to the rate equations was solved as a function of time to obtain the population density of each of the levels as a function of time. Using the numerical simulation, all of the parameters listed in Table 8.1 were kept at their default values except for the transition rate, which is related to the fluorescence lifetime, i.e. 23 The resulting population density for the fluorescing level, N 2 was calculated and plotted as a function of time (in milliseconds). As explained earlier, 23 is inversely proportional to the samples fluorescence lifetime. Figure 8.8 shows the calculated fluor escence signals for a wide range of fluorescence decay rate values obtained as a result of changing 23 from its default value of 1053 sec -1 to a range of values that span from 500 sec -1 which corresponds to a fluorescence lifetime of 2 ms, to 1400 sec -1 which corresponds to a fluorescence lifetime of ~0.71 ms, in steps of 100 sec -1 These graphs are a plot of the population density of the fluorescing level, N 2 as a function of time for the chopper operating at approximately 50 Hz, 500 Hz, and 1000 Hz. These gr aphs are plotted on the same scale.

PAGE 199

174 Parameters Typical Values 12 (sec -1 ) 5 x 10 3 23 (sec -1 ) 1053 30 (sec -1 ) 10 4 N 0 (Molecules) at t = 0 3 x 10 16 N 1 (Molecules) at t = 0 0 N 2 (Molecules) at t = 0 0 N 3 (Molecules) at t = 0 0 I 0 (Photons/cm 2 .sec) 1.113 x 10 22 cm 2 /Molecule) 1.94 x 10 -17 Table 8.1 Typical or default valu es used for the simulation parameters.

PAGE 200

0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 1000 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 500 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 1100 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 600 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 1200 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 700 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 1300 0 1 2 3 4 5x 1013 Population Density of N2(# Molecules)23 = 800 0 10 20 30 40 0 1 2 3 4 5x 1013 Time (ms)Population Density of N2(# Molecules)23 = 1400 0 10 20 30 40 0 1 2 3 4 5x 1013 Time (ms)Population Density of N2(# Molecules)23 = 900 175 Figure 8.8 Simulations with all the parameters default values, except for 23 which is changing as indicated in each graph. Each graph consists of 3 plots for chopper speeds of 50 Hz, 500 Hz, and 1000 Hz.

PAGE 201

176 As 23 became larger, i.e. as the fluorescence lifetime became smaller, the fluorescence signals at 50 Hz reached a steady state faster. This clearly influenced the shape of the signal, which wa s more squared (for higher 23 ) as compared to more pointed (for smaller 23 ) at the top of the signal. The signals maximum peak was also influenced by this change, becoming smaller as 23 became larger. At 23 = 500 the signal was not able to fully decay, because the slow tran sition rate did not allow the population in N 2 to empty out fast enough before the choppers window opened again allowing the laser to pump the system again. It is interest ing to note that the effect of changing 23 was not as noticeable in the signals that were excited wi th the chopper going at 500 Hz and 1000 Hz. Neither their shape nor their amplitudes cha nged considerably across the simulations in Fig. 8.8. It is interesting to compare the simula tion results to that of our experimental results. In effect, picking the curve in Fig. 8.8 which most nearly agrees with the waveforms measured experimentally. For example, Fig. 8.9 shows a comparison between the measured fluorescence intensity signals for a sample of [50 M] DPA with [50 M] TbCl 3 in distilled water as a function of time, when excited with the chopper modulated microchip laser (c hopper frequency ~ 50 Hz, 500 Hz, 1000 Hz), and the population density of the N 2 energy level as a function of time obtained using the simulation with the default values listed in Table 8.1 for chopper frequencies of approximately 50Hz, 500Hz, and 1000 Hz. As can be seen, there is excellent agreement between the experimental and simulated waveforms for a set of lifetime parameters chosen judiciously. Details of these simu lations are given in the following sections.

PAGE 202

(a) (b) Figure 8.9 Experimentally obtained fluorescence intensity of an aqueous solution containing DPA-Tb complex (a), as compared to the simulation signals obtained using the default values (Table 8.1) (b), for the case where the chopper speed was 50 Hz, 500 Hz, and 1000 Hz. 177

PAGE 203

178 8.3.1 Detailed Study of Population Leve ls at Changing Chopper Frequencies Although the energy level of most inte rest to us is the fluorescing level, N 2 it was important to see the behavior of each of the energy levels of the diagram shown in Fig. 8.7 for a better understanding of the overall system. To this end the population densities of each of the energy levels was plotted as a function of time for three chopper frequencies: 50 Hz, 500 Hz, and 1000 Hz. Figure 8.10 shows the population densities of each of the levels in the system, i.e. N 0 N 1 N 2 and N 3, as a function of time for a chopper frequency of 50 Hz. The plot of N 0 as a function of time starts at the maximum population density, 3 x 10 16 since all of the molecules are in the ground state. As the incident beam pumps the electron population into the next energy level, N 1 the population density of N 0 starts to decrease; when the chopper blocks the incident laser b eam, thus allowing for the electrons to decay back to ground state, the population of N 0 starts to build up again. With the exception of the population density of N 0 which is opposite to the pattern of the choppers TTL signal, the population densities of all the other levels follow the same pattern as the choppers TTL signal. The periodic behavior of the population density of N 1 is in phase with that of the choppers TTL signal. The population density of N 1 mimics the laser pulses strongly in the top part of the peaks in this figure. Th is is mainly due to the transition rate between the N 1 and N 2 energy levels, i.e. 12 which allows a fast enough transition at its default value of 5,000/sec.

PAGE 204

0 10 20 30 40 2.997 2.998 2.999 3x 1016 Time (ms)N0 0 10 20 30 40 1 2 3 4 5 6 7x 1012 Time (ms)N1 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)N2 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1012 Time (ms)N3 Figure 8.10 Population densities for all of the energy levels as a function of time, as calculated by the simulation, using the default values for the parameters (f Chopper 50 Hz). 179

PAGE 205

180 The graph for the population density of N 2 as a function of time, which is in good agreement with the experimentally obtained fl uorescence behavior as a function of time, also shows small laser pulse patterns on the peak of the signals. These patterns were also observed in the experimentally obtained fluore scence intensity as a function of time in Fig. 8.1 (a). The population density as functi on of time of N 3 is the only graph that does not show the individual high-PRF lase r pulse excitation pattern, unless 12 is very large. As the choppers frequency was change d to 500 Hz in the simulation, the amplitudes of the population densities diminished for all of the energy levels, except for N 1 This is seen in Fig. 8.11 where the p opulations are observed to not return to their low intensity or non-saturated levels. As the frequency increased, the signals peaks and troughs were decreased relatively symmetrically from both their top half and their bottom half, and as a result, a smaller periodic signal was left. In accordance to the experimental results, the amplitudes of the signals were reduced even further when the choppers frequency was increased to 1000 Hz in the simulation, as shown in Fig. 8.12, for all of the energy levels except N 1 When the value of 12 was decreased as compared to the default value, however, the amplitude of the population density as a function of time of N 1 did decrease. 8.3.2 Detailed Study of Population Leve ls at Changing Incident Intensity The incident laser intensity, I, was varied in the simulations to determine its effect on the population levels, and thus fluorescence intensity from N 2 Figure 8.13 shows the effect that different incident intensities have on the population densities of each of the energy levels as a function of time, for a ch opper frequency of 50 Hz. In these figures,

PAGE 206

0 10 20 30 40 2.997 2.998 2.999 3x 1016 Time (ms) N0 0 10 20 30 40 1 2 3 4 5 6 7x 1012 Time (ms) N1 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)N2 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1012 Time (ms) N3 Figure 8.11 Population densities for all of the energy levels as a function of time, as calculated by the simulation, using the default values for the parameters (f Chopper 500 Hz). 181

PAGE 207

182 0 10 20 30 40 2.997 2.998 2.999 3x 1016 TiN0 me (ms) 0 10 20 30 40 0 1 2 3 4 5 6 7x 1012 Time (ms)N1 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 TiN2 me (ms) 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1012 Time (ms)N3 Figure 8.12 Population densities for all of the energy levels as a function of time, as calculated by the simulation, using the default values for the parameters (f Chopper 1000 Hz).

PAGE 208

N 0 N 1 N 2 N 3 0.5 1 1.5 2 2.5 3x 1016 0 2 4 6 8 10 12x 1015 0 5 10 15x 1015 0 0.5 1 1.5 2x 1015 I = I 0 10 5 1.5 2 2.5 3x 1016 0 1 2 3x 1015 0 2 4 6 8 10x 1015 0 5 10 15x 1014 I = I 0 10 3 0 10 20 30 40 2.7 2.75 2.8 2.85 2.9 2.95 3x 1016 0 10 20 30 40 0 1 2 3 4 5 6x 1014 0 10 20 30 40 0 0.5 1 1.5 2x 1015 0 10 20 30 40 0 0.5 1 1.5 2x 1014 I = I 0 10 2 Time (ms) Time (ms) Time (ms) Time (ms) Figure 8.13 Simulations depicting the population density of each energy level for different intensities, where I 0 is the default value 1.113 x 10 22 photons/cm 2 sec for f chopper = 50 Hz. (Section A.) 183

PAGE 209

N 0 N 1 N 2 N 3 2.97 2.98 2.99 3x 1016 0 2 4 6x 1013 0 0.5 1 1.5 2 2.5x 1014 0 0.5 1 1.5 2 2.5x 1013 I = I 0 10 2.9965 2.997 2.9975 2.998 2.9985 2.999 2.9995 3x 1016 0 1 2 3 4 5 6 7x 1012 0 0.5 1 1.5 2 2.5x 1013 0 0.5 1 1.5 2 2.5x 1012 I = I 0 0 10 20 30 40 2.9997 2.9998 2.9998 2.9998 2.9999 3 3x 1016 0 10 20 30 40 0 2 4 6 8x 1011 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1012 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1011 I = I 0 10 -1 Time (ms) Time (ms) Time (ms) Time (ms) Figure 8.13 (Continued Section B) 184

PAGE 210

185 N 0 N 1 N 2 N 3 3 3 3 3x 1016 0 1 2 3 4 5 6 7x 1010 0 0.5 1 1.5 2 2.5x 1011 0 0.5 1 1.5 2 2.5x 1010 3 3 3 3 3x 1016 0 2 4 6 8x 109 0 0.5 1 1.5 2 2.5x 1010 0 0.5 1 1.5 2 2.5x 109 0 10 20 30 40 0 2 4 6x 1016 0 10 20 30 40 0 0.2 0.4 0.6 0.8 1 0 10 20 30 40 0 0.2 0.4 0.6 0.8 1 0 10 20 30 40 0 0.2 0.4 0.6 0.8 1 Time (ms) Time (ms) Figure 8.13 (Continued Section C) Time (ms) Time (ms) I = I 0 10 -2 I = I 0 10 -3 I = 0.0

PAGE 211

186 the laser intensity was varied by 8 orders of magnitude. Figure 8.13-A shows this for incident intensities of two, three, and five orders of magnitude higher than the default intensity of 1.1 x 10 22 photons/cm 2 sec. As can be seen from the figure, the intensity that achieve d a long term (> 0.1 ms) saturation effect of the fluorescing level, N 2 was five orders of magnitude hi gher than the default value, at which point the population density of the fluorescing level became filled up with almost two thirds of the total number of electrons th at the ground level originally had. Saturation effects in the population levels of N 0 and N 1 are seen within the 0.4 ns laser pulse at about 10 x I 0 It maybe added that there may be different uses for the term saturation. The term saturation is often used to describe signi ficant deviation in th e ground state population which is associated with saturation of the ab sorption transmission. However, the term saturation can also be applied to population build up within energy levels in a 4-level laser type system, as in the term saturat ion of the fluorescence intensity where the upper population state (and fluorescence signal) has become saturated or no longer linear with excitation intensity. The shape of the signal for the N 0 level did not change considerably throughout the different incident intensities except for when the intensity decreased by three orders of magnitude (Fig. 8.13-C), at which point a stro ng step pattern along with a lack of the laser beam imprint could be detected. Also, as the intensity decreased, the variation in the population density was very small, e.g. for I = I 0 x 10 -3 the change in the y-scale is practically negligible.

PAGE 212

187 The maximum number of molecules that N 1 N 2 and N 3 reached as a function of the incident intensity decreased as the intens ity decreased. As expected, this maxim for N 0 was its ground state population, i.e. 3 x 10 16 molecules. Except for this, the population densities as a function of time of N 3 did not change considerably with changing intensity. 8.3.3 Effect of the Transition Rate 30 Due to the closeness of the 7 F energy levels, the rate of transition, 30 between the level where the fluorescence terminates, N 3 and the ground state, N 0 is assumed to be very fast. Figure 8.14 shows the effect that changing this rate ha d on the signal depicting the population density of the fluorescing level N 2 as a function of time for chopper frequencies of 50 Hz, 500 Hz, and 1000 Hz (overplotted on top of each other). As is shown in this figure, when the defau lt value of 10,000/sec was changed from 0/sec to 2 x 10 7 /sec no considerable difference can be detected during 40 ms in the shape of these signals with respect to one another. This was not surprising when using higher transition rate values for 30 since this facilitates the return of electrons to the ground state. However, at very low transition rate values of 30 the population density of N 2 should in theory start to decline, since N 0 would not be getting repopulated quickly enough. Figure 8.15 is a plot of the data shown in Fig. 8.14 with 30 = 0. In this graph where the time axis covers from 0 sec to 1sec, one can see that indeed the population density of the fluorescing level N 2 starts to diminish over tim e, due to the depletion of

PAGE 213

0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)Population Density of N2(# Molecules)30 = 0.0 (a) 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)Population Density of N2(# Molecules)30 = 1000 (b) 0 10 20 30 40 0 1 2 3x 1013 Time (ms)Population Density of N2(# Molecules)30 = 2 x 107 (c) Figure 8.14 Simulations with all the parameters default values, except for 30 which is at three different values, two of which are extreme values. 188

PAGE 214

0 0.2 0.4 0.6 0.8 1 0 0.5 1 1.5 2 2.5x 1013 Time (s)Population Density of N2(# Molecules)30 = 0.0 Figure 8.15 Simulation with all the parameters default values, except for 30 = 0.0, over a larger time scale than shown in Fig. 8.14 (a). 189

PAGE 215

190 electrons in the ground state N 0 which in turn is caused by an extremely small transition rate back down to the ground state, 30 8.3.4 Effect of the Transition Rate 12 The transition rate, 12 between the first upper level, N 1 and the upper fluorescing level, N 2 was changed to a number of extreme values in order to see its effect on the population density of N 2 for chopper frequencies of 50 Hz, 500 Hz, and 1000 Hz. Figure 8.16 shows the simulations obtained wh en changing this transition rate from 0/sec to 2 x 10 7 /sec over several steps. For 12 =0, the population density as a functio n of time for the fluorescing level is zero. However, for a small transition rate of 0.1/sec, a signal starts to build slowly over the 40 ms time shown in the graph. As the rate increases to 1000/sec the signal resembles the experimental signals obtained at the respective chopper frequencies. However, unlike the case of 30 for a very large value of 12 = 2 x 10 7 /sec the simulation breaks down, leaving a noisy signal. This is because the time increment used in the simulations was not fast enough to capture th e correct transient effects within the rate equations. The time increment could have been reduced, but at the expense of significant increased computation time. Another effect of 12 is on the amplitude of the signals at chopper frequencies of 500 Hz and 1000 Hz. In our original trials, the default value that was used for 12 was 10,000/sec. After a few trial and er rors, we found that as the value of 12 decreased from this number, so did the amplitudes of the signal at 500 Hz and 1000 Hz. The amplitude

PAGE 216

0 10 20 30 40 0 0.2 0.4 0.6 0.8 1 Time (ms)Population Density N2(# Molecules) 12 = 0.0 0 10 20 30 40 0 2 4 6x 1010 Time (ms)Population Density of N2(# Molecules)12 = 0.1 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)Population Density of N2(# Molecules)12 = 1000 0 10 20 30 40 -3 -2 -1 0 1 2 3x 1014 Time (ms)Population Density of N2(# Molecules)12 = 2 x 107 Figure 8.16 Simulations with all the parameters default values, except for 12 which is at four different values, three of which are extreme values. 191

PAGE 217

192 of the signal at these higher chopper freque ncies for the experimental data were about 30%, and about 10% of the of the signals amp litude at 50 Hz, respectiv ely. In order to better match the percentage decrease in the am plitude of the signals in the simulation to those found experimentally, 12 was set to a new default value 5,000/sec, which brought the amplitudes of the signals at a chopper frequency of 500 Hz and 1000 Hz down to 35% and 12% of the amplitude of the signal at 50 Hz respectively. 8.3.5 Effect of Changing the Absorption Cross Section The absorption cross sections value was al so varied and its values close to its thresholds were found using the numerical solu tion to the rate equations. Figure 8.17 shows three plots of the population density of N 2 as a function of time for chopper frequencies of 50 Hz, 500 Hz, and 1000 Hz as the value for the abso rption cross section was changed. In Fig. 8.17 (a) the population density N 2 was plotted as a function of time for the highest value of the absorption cross section, i.e. = 2 x 10 -12 cm 2 /molecule. At this value, which was five orders of magn itude larger than its default value of 1.94 x 10 -17 cm 2 /molecule, the population density of N 2 reached saturation at around two thirds of the total number of the electron population. This graph is very similar to the steady state that was reached when the intensity of the incident beam was increased in the simulation by five orders of magnitude as show n in Fig. 8.13. The effect of increasing the intensity of the incident beam by five orders of magnitude is therefore similar to the effect of increasing the absorption cross section by the same amount. Figure 8.17 (c) shows the signal at an extremely small absorption cross section value, i.e. = 1 x 10 -320 cm 2 /molecule. Although the behavior of the population density

PAGE 218

0 10 20 30 40 0 0.5 1 1.5 2x 1016 Time (ms)Population Density of N2(# Molecules) = 2 x 10-12 (a) 0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)Population Density of N2(# Molecules) = 1.9 x 10-17 (b) 0 10 20 30 40 0 0.5 1 1.5x 10-290 Time (ms)Population Density of N2(# Molecules) = 2 x 10-320 (c) Figure 8.17 Simulations with all the parameters default values, except for which is at its largest possible value (a), at its default value (b), and at its smallest value (c). 193

PAGE 219

194 of N 2 is in good agreement with the experiment al data, it must be noted that such a decrease in the size of the ab sorption cross section transl ated into a maximum population density of approximately 1.3 x 10 -290 molecules. Figure 8.17 (b) shows the simulations obtained when using th e default value of for comparison purposes. 8.4 More Complex 5-Level Rate Equation Model In order to more closely imitate and control the transition of energy between the DPA molecules lowest lyi ng triplet state and the Tb 3+ s 5 L 10 state, one more level was added to the simple energy level system pr esented earlier. The transfer of energy between these two levels has been assumed to be a very fast and radiationless process. In order to take into account other possible processes, more tr ansfer routes were added to this new 5-level system. One of the main processes known to ha ppen in the fluorescence of the DPA-Tb complex in solution is quenching 38 This quenching tends to occur mainly due to the interaction between the terbium ion and the wa ter molecules in aqueous solutions. In order to take this into account, an energy level model was developed which included quenching rates at every possibl e level. Figure 8.18 shows a diagram of this new 5-level quenching model. The rate equations related to this system are given below.

PAGE 220

N0N1N2N3N4 II30(Fluorescence)D P A -Tb Figure 8.18 Energy level diagram for the DPA-Tb complex including quenching rates. N 1 and N 2 depict the first singlet an triplet states of the DPA molecule, respectively. N 3 N 4 depict the 5 D 4 the 7 F series of the Tb 3+ ion, respectively. 195

PAGE 221

0011012023034041011011212 12120223232323033434343404dN=-IN+IN+N+N+N+N dtdN=IN-IN-N+N dtdN=N-N+NdtdN=N-N-N dtdN=N-N dt,, ,,, (8.6) (8.7) (8.8) (8.9) (8.10) and 01234N=N+N+N+NN+ (8.11) The values found that best managed to imitate the experimental results obtained are given in Table 8.2. The values for N 0 I 0 , 40 and the fluorescing transfer rate, in this model called 34 remained the same. The transfer rate between the DPA molecules lowest lying triplet and the terbium ions 5 G 6 level, 12 was chosen to be 5 x 10 5 /sec. The transfer rate from the energy level immediately preceding the upper fluorescing level to the upper fluorescing level, 23 was found to be best kept at the same value as its counterpart in the simple model, i.e. 5,000/sec. It is important to note that using the default values described so far, and setting the quenching rates to zero, the signals that were obtained were identical to those obtained using the simple model described earlier with its default values. At this point the quenching transfer rates were set to 15/sec. The only effect these quenching rates seemed to have had, was in slightly reducing the maximum peak values reached in the graphs of the population density of the fluorescing level, N 3 as a function of time. A 196

PAGE 222

197 Parameters Typical Values 10 (sec -1 ) 15 12 (sec -1 ) 5 x 10 5 20 (sec -1 ) 15 23 (sec -1 ) 5 x 10 3 30 (sec -1 ) 15 34 (sec -1 ) 1053 40 (sec -1 ) 104 N 0 (Molecules) at t = 0 3 x 10 16 N 1 (Molecules) at t = 0 0 N 2 (Molecules) at t = 0 0 N 3 (Molecules) at t = 0 0 N 4 (Molecules) at t = 0 0 I 0 (Photons/cm 2 .sec) 1.113 x 10 22 (cm 2 /Molecule) 1.94 x 10 -17 Table 8.2 Typical or default values used for the simulation parameters for the 5-level quenching model.

PAGE 223

plot of the population density of N 3 as a function of time for chopper frequencies of 50 Hz, 500 Hz, and 1000 Hz using the default values given in Table 8.2 is shown in Fig. 8.19. 8.4.1 The Up-conversion Model (Frster Resonance Energy Transfer) When a distance-related transfer of energy occurs between two molecules nonradiatively due to long-range dipole-dipole coupling mechanisms, an excited molecule(the donor) can transfer its energy to another molecule (the acceptor) leaving the acceptor molecule in an excited state. This phenomenon is commonly known as up-conversion in the optics and laser communities or Frster resonance energy transfer (FRET) in the biology and chemistry community. 4 Upconversion was added as an additional process to the quenching model described earlier in Section 8.4. The energy level diagram for this model is shown in Fig. 8.20. The rate equations for this system are 2001101202404up3up-0up1 011011212 12120223223232up33434343404dN=-IN+IN+N+N+N+N+NdtdN=IN-IN-N+NdtdN=N-N+NdtdNdtdNdt ,,=N-2(N)-N ,=N-N up2up3404up-0updNdt ,=N-NN (8.12) (8.13) (8.14) (8.15) (8.16) (8.17) and 198

PAGE 224

0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)Population Density of N3(# Molecules) Figure 8.19 Population density of the fluorescing level, N 3 of the 5-level quenching model, as a function of time using the typical values given in table 8.2. 199

PAGE 225

N0N1N2N3N4NUP IIupupup-0N32N32(Fluorescence)DPA-Tb Figure 8.20 Energy level diagram for the DPA-Tb complex including upconversion rates. N 1 and N 2 depict the first singlet an triplet states of the DPA molecule, respectively. N 3 N 4 and N up depict the 5 D 4 the 7 F series, and a second excited state of the Tb 3+ ion, respectively. 200

PAGE 226

01234upN=N+N+N+NNN++ (8.18) The quenching model presented in Fig. 8.18 is retained in this upconversion model, with the exception of the quenching transfer rate, 30 which for simplicity was omitted from this system. The default values used for the parameters in this model remained the same as those that were used for the quenching model, and are listed in Table 8.3. For the two new transfer rates up and up-0 a very small rate value (10 -15 /sec) and larger rate value (250/sec) were chosen respectively, in order to minimize their effect. A graph of the population density of the fluorescing level, N 3 as a function of time is plotted in Fig. 8.21. From this figure we can see that the existence of this new upper level does not significantly affect the shape or peak value of the new signal. From these different models we can therefore conclude that the simple 4energy level model is not only suitable but also sufficient for these analyses. However, it is important to note that the 5-level and up-conversion model may be appropriate for other fluorescing systems, such as bio-luminescence cells and even solid-state laser models. 74-76 We have also studied the case of a more complex system where N 0 of DPA is separate from N 0 of Tb 3+ In this system the transfer rate between the DPA and the terbium ion has been assumed to be very slow, i.e. in the order of 3 ms. The results, which are not shown here, have not been satisfactory so far, although further modeling is being carried out to improve the outcome. Finally, we plan to use an asymmetric excitation waveform in our simulation model and then compare it to our experimental data. Some of these initial results are shown in Appendix C, and the initial results are encouraging. The use of asymmetric 201

PAGE 227

202 Parameters Typical Values 10 (sec -1 ) 15 12 (sec -1 ) 5 x 10 5 20 (sec -1 ) 15 23 (sec -1 ) 5 x 10 3 34 (sec -1 ) 1053 up (sec -1 ) 15-Oct up-0 (sec -1 ) 250 40 (sec -1 ) 104 N 0 (Molecules) at t = 0 3 x 10 16 N 1 (Molecules) at t = 0 0 N 2 (Molecules) at t = 0 0 N 3 (Molecules) at t = 0 0 N 4 (Molecules) at t = 0 0 N up (Molecules) at t = 0 0 I 0 (Photons/cm 2 .sec) 1.113 x 10 22 (cm 2 /Molecule) 1.94 x 10 -17 Table 8.3 Typical or default values used for the simulation parameters for the 5-level upconversion model.

PAGE 228

0 10 20 30 40 0 0.5 1 1.5 2 2.5x 1013 Time (ms)Population Density of N3(# Molecules) Figure 8.21 Population density of the fluorescing level, N 3 of the 5-level upconversion model, as a function of time using the typical values given in table 8.3. 203

PAGE 229

204 excitation waveforms may be important in ot her spectroscopic fields according to the systems being studied and the complexity of the saturation and decay process.

PAGE 230

205 CHAPTER 9. FLUORESCENCE INTENSITYS AMPLITUDE AND ITS APPROACH TO SATURATION In the previous chapter, it was shown that the experimental data could be fitted to a transient solution of the rate equation energy level model of the DPA-Tb complex. The agreement was obtained by varying th e assumed fluorescence decay rate (i.e. fluorescence lifetime) to produce the best fit to the experimental data. This procedure yielded results that were within about 5% to 10% of the values obtained by the low-PRF pulsed laser technique used for th e strongest of the DPA-Tb emi ssion lines. It should be noted that this technique is essentially a be st fit guess by visual observation as to the selection of the fitted fluorescence lifetime value. A more mathematically rigorous "curve f itting" technique would be to fit all of the varying chopper amplitude data and th e change in the amplitudes as the chopper frequency is varied. In this case, the ch opper frequency was varied smoothly from 1 Hz to above 1000 Hz, and the simulated results for the predicted fluorescence amplitude was fitted to that of the experimental data. This technique is shown in the following section and yielded a parameter fit to the data with an accuracy on the order of 5%. 9.1 Fluorescence Amplitudes We define the amplitude of the fluorescence emission as the extent from the lowest intensity measured to the highest intensity measured as shown in Fig. 9.1. Using

PAGE 231

0 10 20 30 40 0 200 400 600 800 1000 Time(ms)Fluorescence Intensity AmplitudefChopper 50 Hz 0 10 20 30 40 0 200 400 600 800 1000 Time(ms)Fluorescence Intensity fChopper 500 Hz Amplitude Figure 9.1 Fluorescence signals amplitudes at 50Hz and 500Hz. 206

PAGE 232

207 this definition Fig. 9.2 is a plot of the fl uorescence signals amplitude as a function of the chopper frequency, for the 543 nm line of a solution of [50M] DPA with [50M] TbCl 3 in distilled water. We can see that at very low frequencie s, the sample was saturated and the fluorescence intensity remained constant. However, as the choppers frequency started to increase, the signals amplitudes started to decrease in an exponential manner. Figure 9.3 shows the amplitudes of the signals from the same sample plotted as a function of the cumulative excitation photons. In order to find the number of cumulative photons that were hitting the sample, the fo llowing steps were taken: the amount of time it took for an infinitesimally small point to pass an open window of the choppers blade was calculated by dividing the inverse of th e choppers frequency by two. Next, this number was multiplied by the microchip lasers pulse repetition frequency, i.e. 8,500 Hz. This is the number of laser pulses that ma de it through one open window of the choppers blade at the set frequency. Lastly we multiplied this number by 1.11 x 10 22 photons/cm 2 sec, which was the intensity of the laser. 9.2 Fluorescence Amplitudes: Comparis ons of Simulation s and Experiments As shown in Chapter 8, the signals obta ined when using the numerical solutions of the rate equations were in good agreement with those obtained ex perimentally. In order to test the goodness of fit, the am plitudes of the signals using the numerical simulations were plotted along side the amplitudes of the signals obtained experimentally. From here, the fluorescence lifetime of the simple energy level model presented in Chapter 8, i.e. 1/ 23 was changed to different values in order to compare the behavior of these curves with the experimental one.

PAGE 233

0 200 400 600 800 1000 0 100 200 300 400 500 600 700 800 fChopper (Hz)Fluorescence Intensity Signal's Amplitude Figure 9.2 Fluorescence signals amplitude for the 543nm line of a solution of [50M] DPA with [50M] TbCl 3 in distilled water as a function of the choppers frequency. 208

PAGE 234

0 2 4 6 8 10x 1023 0 100 200 300 400 500 600 700 800 Number of Cumulative Excitation Photons( 1/fChopper)Fluorescence Intensity Signal's Amplitude Figure 9.3 Fluorescence signals amplitude for the 543nm line of a solution of [50M] DPA with [50M] TbCl 3 in distilled water as a function of the number of photons that passed through one of the choppers open windows. This number is inversely proportional to the choppers frequency. Note: the x-axis has been drawn to only 10 x 10 23 for clarity purposes. 209

PAGE 235

210 Figure 9.4 shows the signals amplitudes as a function of the choppers frequency for the experimental data as well as severa l numerically obtained graphs. Since the fluorescence intensity of the experimental si gnals cannot be numerically compared to the population densities of the graphs obtained using the simulation, all of the graphs were normalized to their individual maxima. As can be seen from this figure, all of the plots follow the same general pattern for the lower chopper frequencies, followed by an exponential declin e in their signals amplitudes. The graphs that follow the e xperimental results closest are those whose 23 transition rates were 952.4/sec and 869.6/sec, corresponding to fl uorescence lifetimes of 1.05 ms and 1.15 ms, respectively. These lifetime values, however, are slightly further from the experimentally obtained 0.95 ms for the 543 nm fluorescence line for this solution (Table 7.1). In Fig. 9.5, which shows the same simulations as a function of the cumulative excitation photons that were incident on the sample, the same trend is present. This method provides an estimate of the fluorescence lifetime to about 5%, and gives a good sense of the accuracy of the chosen numerical values for the different parameters of the energy level system used to mimic the DPA-Tb complex.

PAGE 236

0 200 400 600 800 1000 0 0.2 0.4 0.6 0.8 1 fChopper (Hz)Fluorescence Intensity Signal's Amplitude = 0.75ms = 0.85ms = 0.95ms = 1.15ms = 1.05ms Experimental( 0.95ms) Figure 9.4 Fluorescence intensity as a function of chopper frequency, f chopper for several different lifetime values. 211

PAGE 237

0 2 4 6 8 10x 1023 0 0.2 0.4 0.6 0.8 1 Number of Cumulative Photons( 1/fChopper)Fluorescence Intensity Signal's Amplitude = 1.05ms = 1.15ms Experimental ( 0.95ms) =0.85ms = 0.75ms = 0.95ms Figure 9.5 Comparison of several numerically obtained fluorescence (i.e. population density) signals amplitude as a function of the number of cumulative excitation photons that passed through one of the choppers open windows with respect to the experimentally obtained fluorescence signals amplitude. Each graph has been normalized to its maximum. Note: the x-axis has been drawn to only 10 x 10 23 for clarity purposes. 212

PAGE 238

213 9.3 Comparison of TFS to Related Fluorescence and Spectroscopy Techniques A similar term, Transient Saturation Absorp tion Spectroscopy, has been used in a publication where the differential transmission spectra of GaAs thin films have been obtained. 77 The technique described in their paper, however, appears to be the technique commonly known as saturation absorption spectroscopy. Even though saturation absorption spectroscopy is a co mmonly used procedure, it has been suggested that the term used to describe it is a misnomer and the term pump-probe spectroscopy would be a mo re appropriate choice. 78 Although a number of articles have been published where the saturation behavior of laser induced fluorescence has been studi ed, including modeling techniques, most of these are studies related to gases or flame spectroscopy. 79-80 Perhaps the most similar of these, is a study in which the rate equations for a two level system have been solved using an excitation of one square laser pulse. 71 A somewhat conceptually similar study has been published, in which a Laser Induced Fluorescence Transient (LIFT) technique has been developed to remotely measure photosynthetic activity in vegetation. 81 Laser induced saturation of molecular fluorescence, along with the spontaneous Raman scattering signal of the solvent, has b een reported to have been used to measure the fluorescence lifetime of samples, even w ith laser pulses longer than the fluorescence lifetime of the species. 70 It should be noted that our TFS look s similar to phase spectroscopy or modulation spectroscopy where the phase and/or amplitude of the modulated fluorescence emission is compared to that of the modulated excitation in order extract the fluorescence lifetime value. 82 However, the phase or modulation fluorescence technique

PAGE 239

214 was initially developed for use with weak CW (xenon lamp) excitation and later using CW laser excitation. In addition, a simp lified two level energy system was used with only excitation and fluorescence emission mo deled, sinusoidal excitation assumed, and no population or stimulated emission taken into account. 83 Our technique of TFS represents an extension of this technique, an d includes the transient excitation of a pulsed laser and modulated amplitude waveform, asymmetric excitation, and transfer rates between other energy levels. Also, the ph ase and modulation met hod does not allow for the more detailed study of the complex s energy levels and transfer rates. It is interesting to note that our TFS technique is essentia lly an outgrowth or extension of the simpler modulation spectro scopy and a detailed energy-level rate equation analysis, as often used in solid-state laser modeling. 84

PAGE 240

215 CHAPTER 10. MAXIMUM FLUORESCENCE INTENSITIES FOR DIFFERENT SOLVENTS AND SOLUTIONS OF DPA AND TbCl 3 In this chapter we show the relationship between the fluorescence lifetimes of the different samples, and their respective maxi mum fluorescence intensities as a function of the solvents used. These results show that under some conditions the maximum fluorescence intensity is dependent upon the background solvent and concentration of the DPA and TbCl 3 These results are important to those in the biological and chemical fields to help understand the formation of DPA-Tb complexes under different pH and solvent conditions. 10.1 Trends in Fluorescence for [50 M] Solutions Figure 10.1 is a three dimensional representation of the maximum fluorescence intensities as a function of the detected fluorescence line(s) and the solvents that were used for samples of [50 M] DPA with [50 M] TbCl 3 when excited with the dye laser ( exct. =266nm). These intensities have been normalized to the lasers power, to the gratings efficiency and to the detectors sensit ivity. As can be seen from the figure, for any particular fluorescence line(s) detected using this technique, there was no strong trend as a function of the different types of solvents, i.e., SAB vs. DW vs. HCl, for the maximum fluorescence intensity the samples. For example, for all the fluorescence signals mixed together, the maximum fluorescence intensity seems to be the same for the

PAGE 241

Mixed 488 nm 543 nm 581 nm 618 nm HCl DW SAB 0 2 4 6 8 Fluorescing Line SolventMaximum Fluorescence Intensity Figure 10.1 Fluorescence maxima from solution of [50 M] DPA with [50 M] TbCl 3 in different solvents when excited with the dye laser. 216

PAGE 242

sample that was dissolved in distilled water or dissolved in SAB or dissolved in [0.01M] HCl. This trend (or lack thereof) seems to hold true for the 488 nm lines (in the different solvents), and also approximately for the 543 nm line. The maximum intensities in the different solvents for the two weakest lines seem to be roughly the same, but in these cases the fluorescence intensity was very weak and the signal-to-noise was close to 1. In summary, the maximum fluorescence intensities between the different solvents seem to be the same for each of the fluorescence lines for the samples that were excited with the dye laser. For the same DPA-Tb samples that were measured with the optical chopper/microchip laser combination, in place of the dye laser, there was, in fact, a stronger pattern or trend present in their maximum fluorescence intensities, as shown in Fig. 10.2. Here, it is clear that the fluorescence intensity was at its highest for the samples that were in SAB, followed by those that were in distilled water, and finally by those that were in HCl, i.e. SABDWHClIII The patterns for the lifetime values for [50 M] DPA with [50 M] TbCl 3 solutions that were excited with the dye laser and with the microchip laser were both SABDWHCl as discussed in Chapters 6 and 7, respectively. One can conclude that there is a direct proportionality between the order of the lifetimes and their fluorescence intensities, for the [50 M] DPA with [50 M] TbCl 3 solutions as a function of their solvents, for those solutions that were excited with the microchip laser, as suggested by Eqs. (2.8) and (2.14). 217

PAGE 243

Mixed 488 nm 543 nm 581 nm 618 nm HCl DW SAB 0 200 400 600 800 1000 1200 Fluorescence Lines SolventMaximumFluorescence Intensity Figure 10.2 Fluorescence maxima from solution of [50 M] DPA with [50 M] TbCl 3 in different solvents when excited with the microchip laser. 218

PAGE 244

10.2 Patterns for [5 M] to [25 M] Solutions The normalized maximum fluorescence intensity points for solutions of different DPA and TbCl 3 concentrations in [1M] SAB (pH ~ 5.6) are plotted in Fig. 10.3. Although the fluorescence intensity increases when both the DPA and the TbCl 3 increase, the effect is more pronounced when the DPA concentration is increased (for a fixed TbCl 3 concentration) as compared to when the TbCl 3 concentration is increased for a fixed DPA concentration. According to theory 25,65 both photoluminescence intensity and lifetime increase as the ratio of DPA to TbCl 3 increases, since this leaves fewer unoccupied binding sites for the terbium ion, hence decreasing the possibility for OH binding (and for its oscillators quenching effect). This pattern holds true for all of the measured fluorescence signals, whether resolved separately, or detected as a combination of all. The maximum fluorescence intensity points for solutions of different DPA and TbCl 3 concentrations in distilled water and in [0.01M] HCl are plotted in Figs 10.4 and 10.5, respectively. Although the fluorescence intensity seems to also be increasing with an increasing concentration of DPA more so than with an increasing concentration of TbCl 3 the changes seem more sudden for the solutions in distilled water, as compared to the solutions in SAB or HCl. Once again the slope of these graphs may be significantly different because of the effect of their solvents pHs. The patterns for the lifetime values for solutions of changing concentrations of DPA and TbCl 3 which were excited with the microchip laser was DWSABHCl The pattern for the maximum fluorescence intensity points for these same solutions follows the same order, i.e. Again there is a direct proportionality between the DWSABHClIII 219

PAGE 245

(a) All Lines Mixed (b) 488nm Line (c) 543nm Line (d) 581nm Line (e) 618nm Line Figure 10.3 Points of maximum fluorescence intensity for solution with different [DPA] and [TbCl 3 ] in [1M] SAB when excited with the combination of the microchip laser and optical chopper (f = 50 Hz). 220

PAGE 246

(a) All Lines Mixed (b) 488nm Line (c) 543nm Line (d) 581nm Line (e) 618nm Line Figure 10.4 Points of maximum fluorescence intensity for solutions with different [DPA] and [TbCl 3 ] in distilled water when excited with the combination of the microchip laser and optical chopper (f = 50 Hz). 221

PAGE 247

(a) All Lines Mixed (b) 488nm Line (c) 543nm Line (d) 581nm Line (e) 618nm Line Figure 10.5 Points of maximum fluorescence intensity for solution with different [DPA] and [TbCl 3 ] in [0.01M] HCl when excited with the combination of the microchip laser and optical chopper (f = 50 Hz). 222

PAGE 248

223 pattern of the lifetime values of thes e solutions and their maximum fluorescence intensities. The difference in the patterns of the [50 M] solutions discussed in section 10.1 and the solutions of varying concentrations, up to [25 M] solutions, discussed in this section, could be due to concentration effects. At higher concentrations, i.e. the [50 M] solutions, the [1M] SAB (pH 5.6) seems to induce a higher lifetime and fluorescence intensity, as compared to distilled water. At lower concentrations, however, the opposite is true. In both cases, [0.01M] HCl seems to provide the suitable environment for both the lowest lifetime and the lowest fluorescence intensity. 10.3 Estimate of DPA Concentration in Spores The above results are dependent upon the ex act concentration of DPA, and in the above experiments pure DPA was used. However, if spores are used then an estimate of the concentration of DPA contained in spores is required. As a point of comparison it is important to note that one B. subtilis spore contains approximately 3.65 x 10 -16 moles of DPA. 52 This translates into a concentration of [0.4 pM] for a suspension of 1 B. subtilis spore in 1mL of solution.

PAGE 249

224 CHAPTER 11. CONCLUSIO NS AND FUTURE WORK The EEM spectra of the different solutions showed that for the DPA-Tb complex, as well as for the terbium ion by itself in distilled wa ter, the strongest fluorescence signal was obtained using deep UV excitation wa velengths, i.e. 235 nm for the DPA-Tb complex and 225 nm for the Tb 3+ Although the optimum conditions for maximum fluorescence intensity depends on the environmental conditions as well as the ratio of DPA to TbCl 3 these preliminary data seem to suggest that at lower concentrations, i.e. up to 25M of each DPA and TbCl 3 the most fluorescence inducing solvent for the readily available 266 nm wavelength (i.e. 4 th harmonic Nd:YAG laser) may be di stilled water, followed by [1M] SAB pH 5.7, and [0.01M] HCl. For higher concentrations, i.e. [50M] of each DPA and TbCl 3 distilled water and [1M] SAB produce equally suitabl e environments, followed by [0.01M] HCl. The absorption cross sections for the terb ium ion, dipicolinic acid molecule, and the terbium dipicolinate complex were measured at 266 nm. Except for the terbium ion, whose absorption cross section was found to be 3 orders of magn itude larger than a previously published value at a close but not equal excitation wavelength ( 284nm = 0.96 x 10 -21 cm 2 /molecule from Carnall 43 and 266nm = 0.18 x 10 -17 cm 2 /molecule from our experiments), the absorption cross sections for DPA and DPA-Tb were found to be very similar to other published va lues, i.e. for the DPA the UV Atlas of Organic Compunds 37

PAGE 250

225 found an 274nm = 2.7 x 10 -17 cm 2 /molecule, while our expe rimental values were 266nm = 2.1 x 10 -17 cm 2 /molecule; for the DPA-Tb complex Latva 35 found 275nm = 4.9 x 10 -17 cm 2 /molecule, and our experimental values were 266nm = 1.9 x 10 -17 cm 2 /molecule. Finally, a new technique was developed to simulate the dynamic behavior of the transient fluorescence signal of the terbiu m doped dipicolinic acid complex using the numerical solutions to the systems transiti on rate equations in conjunction with a modulated pulsed UV excitation source. This technique has the potential to help more accurately measure the fluorescence lifetime. It can also be used as a tool in saturation spectroscopy, as well as the study of solid stat e lasers. The technique has an accuracy of % in water, but may have larger error in other pH solution. It is felt that this technique may have usefulness in other spectros copic fields. Material presented in this dissertation has been published and presente d in a number of technical conferences. 85-91 Future plans include further modifying the modeling program in order to control the exposure time of the sample to the incident excitation light, thus simulating asymmetrical excitation sources, and investiga ting the use of multiple rare earth ions in the DPA complex. The program can also be automated to obtain real time simulations which would closely match the real time fl uorescence signal. Finally, these results maybe important for the detection of bacteria spores. It is anticipated that a library of fluorescence lifetimes of the DPA-Tb complex in different solvents in concentrations close to those found in natural occurring contamina tions, i.e. less than 10 M DPA, can be compiled in the aid of such a detection system.

PAGE 251

226 REFERENCES 1. V. Sivaprakasam, D. Killinger, Multiple UV wavelength excitation and fluorescence of bioaerosols, Appl. Opt. 42, 6739 (2003). 2. D. L. Andrews, Applied Laser Spectroscopy, VCH Publishers, Inc., New York, 1992. 3. G.G. Guilbault, Practical Fluorescence, Marcel Dekker, Inc., New York, 1973. 4. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3 rd Edition, Springer, Singapore, 2006. 5. W. S. Barney, System and method fo r bioaerosol discrimination by timeresolved fluorescence. U.S. Patent 7,060,992 (2006). 6. D.L. Rosen, Bacterial Spore Detec tion and Quantification Methods. U.S. Patent 5,876,960 (1999). 7. P. F. Bernath, Spectra of Atoms and Molecules, Oxford University Press, USA, 1995. 8. V. Illingworth, The Penguin Dictionary of Physics, Penguin, 1999. 9. F. A. Jenkins, H. E. White, Fundamentals of Optics, McGraw-Hill Book Company, 3 rd Edition, 1957. 10. M. Born, E.Wolf, Principles of Optics: Electromagnetic Theory of Propagation, Interference, and Diffraction of Light, Cambridge University Press, 1999. 11. W. T. Silfvast, Laser Fundamentals, Cambridge University Press, 1996. 12. G. Herzberg, Molecular spectra and molecular structure, Krieger Pub Co, 1991. 13. G. G. Guilbault, Fluorescence: theory, inst rumentation, and practice, Marcel Dekker, Inc., 1967. 14. G. F. Kirkbright, M. Sargent, Atomic Absorption and Fluorescence Spectroscopy, Academic Press, London, 1974.

PAGE 252

227 15. A. Sharma, S.G. Schulman, Fluorescence Spectroscopy, John Wiley & Sons, Inc. New York, 1999. 16. A. Bhm, Quantum Mechanics, Springer-Verlag, New York, 1979. 17. F. Mandl, Quantum Mechanics, John Wiley & Sons, We st Sussex, England, 1992. 18. G. Herzberg, Electronic Spectra of Polyatomic Molecules, D. Van Nostrand Company, Inc., New Jersey, 1966. 19. Klimov V. V., Ducloy M., Letokhov V. S., Lebedev P. N. Radiative frequency shift and linewidth of an atom dipole in the vicinity of a dielectric microsphere, J. Mod. Opt. 43 (11), 2251-2268 (1996). 20. Klimov V. V., Ducloy M., Letokhov V. S., Spontaneous emission rate and level shift of an atom insi de a dielectric microsphere , J. Mod. Opt. 43, 549563 (1996). 21. I. E. Alcamo, Fundamentals of Microbiology, Addison-Wesley, Reading, MA, 1984. 22. D.V. Lim, Microbiology, WCB/McGraw-Hill, Boston, MA, 1998. 23. G. W. Gould, A. Hurst, The Bacterial Spore, Academic Press, New York, 1969. 24. G.W. Faris, R. A. Copeland, K. Mortel mans, B.V. Bronk, Spectrally resolved absolute fluorescence cross sections for bacillus spores, Appl. Opt. 36 (4), 958-967 (1997). 25. D. L. Rosen, S. Niles, Chelation Numb er of Terbium Dipicolinate: Effects on Photoluminescence Lifetime and Intensity, Appl. Spectrosc. 55 (2), 208-216 (2001). 26. P. M. Pellegrino, N. F. Fell Jr., J. B. Gillespie, Enhanced spore detection using dipicolinate extraction techniques, Anal. Chim. Acta 455, 167 (2002). 27. A. Ponce, Live/Dead Bacterial Sp ore Assay Using DPA-Triggered Tb Luminescence, NASA Tech Brief 27 (3), NPO30444 (2003). 28. E. Ghiamati,, R. Manoharan, W. H. Nelson, J. F. Sperry, UV Resonance Raman Spectra of Bacillus Spores, Appl. Spectrosc. 46 (2), 357 (1992).

PAGE 253

228 29. S. Sarasanandarajah, J. Kunnil, B.V. Bronk, L. Reinisch, Two-dimensional multiwavelength fluorescence spectra of dipicolinic acid and calcium dipicolinate, Appl. Opt. 44 (7), 1182 (2005). 30. G. F.Bailey, S. Karp, L. E. Sacks, Ultraviolet-Absorption Spectra of Dry Bacterial Spores, J. Bacteriol. 80 (4), 984 (1965). 31. V. H. Holsinger, L. C. Blankenship, M. J. Pallansch, Ultraviolet Absorption by Dipicolinic Acid in Model Systems an d Bacterial Spores, Arch. Biochem. Biophys. 119, 282-287 (1967). 32. H. F. Hameka, J. O. Jensen, J. L. Jensen, C. N. Merrow, C. P. Vlahacos, Theoretical studies of the fluorescence of dipicolinic acid and its anions, J. Mol. Struc-Theochem. 365, 131-141 (1996). 33. R. Nudelman, B. V. Bronk, S. Efrima Fluorescence Emission Derived from Dipicolinic Acid, its Sodium Salts, an d its Calcium Salts, Appl. Spectrosc. 54 (3), 445-449 (2001). 34. F. Peral, E. Gallego, Self-association of pyridine-2,6-dicarboxylic acid in aqueous solution as determined from ultraviolet hypochromic and hyperchromic effects, Spectrochim. Acta A 56, 2149 (2000). 35. M. Latva, H. Takalo, V. Mukkala, C. Matachescu, J. C. Rodriquez-Ubis, J. Kankare, Correlation between the lowe st triplet state energy level of the ligand and the lanthanide(III) luminescence quantum yield, J. Lumin. 75, 149-169 (1997). 36. P. Lima, O. L. Malta, S. Alves Junior, Estudo Espectroscpico de Complexos de Eu 3+ Tb 3+ e Gd 3+ com Ligantes Deri vados de cidos Dicarboxlicos, Quim. Nova 28 (5), 805-808 (2005). 37. UV Atlas of Organic Compounds, Vol. V, Plenum Press, New York. 38. F. S. Richardson, Terbium(III) and Europium(III) Ions as Luminescent Probes and Stains for Biomolecular Systems, Chem. Rev. 82, 541-552 (1982). 39. N.W. Aschroft, D. Mermin, Solid State Physics, 1 st Edition. Brooks Cole (1976). 40. S. Cotton, Lanthanides and Actinides, Oxford University Press, New York, 1991.

PAGE 254

229 41. G. H. Dieke, H. M. Crosswhite, The Spectra of the Doubly and Triply Ionized Rare Earths, Appl. Opt. 2 (7), 675-696 (1963). 42. Physics Auxiliary Publication Servic e, JCPSA-90-3443-63. J. Chem. Phys., April 1, 1989, Vol. 90, No. 7, p. 3443 W. T. Carnall, G. L. Goodman, K. Rajnak & R. S. Rana, American Inst itute of Physics (Physics Microform Reference: 8904A 0719). 43. W. Carnall, Handbook on the Physics and Chemistry of Rare Earths, vol. 3 Non-Metallic Compounds I, Chap ter 24, North-Holland Publishing Company, 1979. 44. G. Stein, E. Wrzberg, Energy gap law in the solvent isotope effect on radiationless transitions of rare earth ions, J. Chem. Phys. 62 (1), 208-213 (1975). 45. Y. Haas, G. Stein, Pathways of Radi ative and Radiationless Transitions in Europium(III) Solutions: the Role of High Energy Vibrations, J. Phys. Chem. 75 (24), 3677-3681 (1971). 46. Y. Haas, G. Stein, Radiative and Nonradiative Pathways in Solution: Excited States of the Europium(III) Ion, J. Phys. Chem. 76 (8), 1093-1104 (1972). 47. D. N. Terpilovskii, Afterglow Lifeti mes of Aqueous Solutions of Europium and Terbium Salts, Optics and Spectroscopy-USSR 24 (4), 313-316 (1968). 48. P. K. Gallagher, Fluorescence of Trivalent Europium in D 2 O-H 2 O Mixtures, J. Chem. Phys. 43 (5), 1742-1744 (1965). 49. A. Heller, Formation of Hot OH Bonds in the Radiationless Relaxations of Excited Rare Earth Ions in Aqueou s Solutions, J. Am. Chem. Soc. 88 (9), 2058-2059 (1966). 50. J. L. Kropp, M. W. Windsor, L uminescence and Energy Transfer in Solutions of Rare-Earth Complexes. I. Enhancement of Fluorescence by Deuterium Substitution, J. Chem. Phys. 42, 1599 (1965). 51. D.L. Rosen, C. Sharpless, L. B. McGown, Bacterial Spore Detection and Determination by Use of Terbium Dipicolinate Photoluminescence, Anal. Chem. 69, 1082-1085 (1997). 52. A. A. Hindle, E. A. H. Hall, Dipicolinic acid (DPA) assay revisited and appraised for spore detection, Analyst 124, 1599-1604 (1999).

PAGE 255

230 53. G. Jones II, V. I. Vullev, Medium E ffects on the Photophysical Properties of Terbium (III) Complexes With Pyridi ne-2,6-dicarboxylate, Photochem. Photobiol. Sci. 1, 1-10 (2002). 54. G. Jones II, V. I. Vullev, Medium Effects on the Stability of Terbium (III) Complexes with Pyridine-2,6-di carboxylate, J. Phys. Chem. 106, 8213-8222 (2002) 55. P. Pellegrino, N. F. Fell, Jr., D. L. Rosen, J. B. Gillespie, Bacterial Endospore Detection Using Terbium Dipicolinate Photoluminescence in the Presence of Chemical and Biological Materials, Anal. Chem. 70, 1755 (1998). 56. P. Jonsson, F. Kullander, M. Nordstrand, T. Tjrnhage, P. Wsterby, M.l Lindgren, Development of fluorescence-based point detector for biological sensing, in Proceedings of Optically Based Biological and Chemical Sensing for Defence, J. C. Carrano, A. Zukauskas, eds., Proc. SPIE 5617, 60-74 (2004). 57. Q. Li, Purnedu K. Dasgupta, H. Temkin, M. H. Crawford, A. J. Fischer, A. A. Allerman, K. H. A. Bogart, and S. R. Lee, Mid-ultraviolet Light-Emitting Diode Detects Dipicolinic Acid, Appl. Spectrosc. 58 (11), 1360-1363 (2004). 58. M. L. Cable, J. P. Kirby, K. Sorasaenee, H. B. Gray, A. Ponce, Bacterial Spore Detection by [Tb 3+ (macrocycle)(dipicolinate)] Luminescence, J. Am. Chem. Soc. 129, 1474-1475 (2007). 59. D. L. Rosen, Airborne bacterial endos pores detected by use of an impinger containing aqueous terbium chloride, Appl. Opt. 45 (13), 3152-3157 (2006). 60. Y. S. Cheng, E. B. Barr, B. J. Fan, P. J. Hargis, Jr., D. J. Rader, T. J. OHern, J. R. Torczynski, G. C. Tisone, B. L. Preppernau, S. A. Young, and R. J. Radloff, Detection of Bioaerosols Usi ng Multiwavelength UV Fluorescence Spectroscopy, Aerosol Sci. Tech. 30, 186-201 (1999). 61. Uniphase Corporation, NU-10110-100 Laser Operators Manual (1999). 62. Laser Science, Inc., DLMS-220 Dye stirred, Dye Laser Modules Operation Manual (1997). 63. Laser Science, Inc., UVS-300 Frequency-Doubling Module Operation Manual (1997). 64. Laser Science, Inc., VSL-337ND-s Nitrogen Laser Operators Manual (1997).

PAGE 256

231 65. A. S. Chauvin, F. Gumy, D. Imbert, J. G. Bnzli, Europium and Terbium tris(Dipicolinates) as Secondary Standards for Quantum Yield Determination, Spectrosc. Lett. 37 (5), 517-532 (2004). 66. W. Horrocks, Jr., D. R. Sudnick, Lanthanide Ion Probes of Structure in Biology. Laser-Induced Luminescence Decay Constants Provide a Direct Measure of the Number of Metal-Co ordinated Water Molecules, J. Am. Chem. Soc. 101 (2), 334-340 (1979). 67. W. R. Dawson, J. L. Kropp, M. W. Windsor, Internal-Energy-Transfer Efficiencies in Eu 3+ and Tb 3+ Chelates Using Excitation to Selected Ion Levels, J. Chem. Phys. 45 (7), 2410-2418 (1996). 68. N. Arnaud, J. Georges, Comprehensive study of the lumi nescent properties and lifetimes of Eu 3+ and Tb 3+ chelated with various ligands in aqueous solutions: influence of the synergic ag ent, the surfactant and the energy level of the ligand triplet, Spechtrochim. Acta A 59, 1829-1840 (2003). 69. M. Evans, N. Hastings, B. Peacock, Statistical Distributions, 3 rd Edition. New York: Wiley, 2000. 70. A. M. Chekalyk, V. V. Fadeev, G. M. Georgiev, Zh. S. Nickolov, Application of Laser Induced Satura tion of Molecular Fluorescence for Lifetime Measurements, Opt. Commun. 38 (3), 177-181 (1981). 71. V.I. Yuzhakov, K.G. Yevsyukhina, S.V. Patsayeva, Laser induced saturation of fluorescence for complex organic molecules, in Proceedings of ALT'97 International Conference on Laser Surface Processing, V. I. Pustovoy, ed., Proc. SPIE 3404, 388-396 (1998). 72. I. Gregor, D. Patra, J. Enderlein, Optical Saturation in Fluorescence Correlation Spectroscopy under Conti nuous-Wave and Pulsed Excitation, ChemPhysChem, 6, 164-170 (2005). 73. J. Daily, Saturation effects in lase r induced fluorescence spectroscopy, Appl. Opt. 16 (3), 568-571 (1977). 74. O. Toma, S. Georgescu, Dynamics of an upconversion Er:YAG laser with reabsorption losses, JOSA B 21 (9), 1630-1637 (2004). 75. R. A. McFarlane, Dual wavelength visible upconversion laser, Appl. Phys. Lett. 54, 2301 (1989). 76. E. Lacot, F. Stoeckel, M. Chenevier, Dynamics of an erbium-doped fiber laser, Phys. Rev. A 49, 3997 (1994).

PAGE 257

232 77. W. Song-Jiang, W. Dan-Ling, J. Hong-Bing, Y. Hong, G. Qi-Huang, J. YaLin, L. Wei, Transient saturation ab sorption spectroscopy excited near the band gap at high excitation carrier density in GaAs, Chinese Phys. 13 (1), 111-114 (2004). 78. D. A. Smith, I. G. Hughes, The ro le of hyperfine pumping in multilevel systems exhibiting saturated absorption, Am. J. Phys. 72 (5), 631-637 (2004). 79. D. R. Olivares, G. M. Hieftje, Saturati on of energy levels in analytical atomic fluorescence spectroscopy I. Theory, Spectrochim. Acta 33B, 79-99 (1978). 80. K. Okuda, K. Furuto, S. Kishimoto, A. Kono, T. Goto, Determination of Transition Probabilities Using the Sa turation Behavior of Laser-Induced Fluorescence, J. Quant. Spectrosc. Ra. 48 (3), 273-278 (1992). 81. Z. Kolber, D. Klimov, G. Ananyev, U. Rascher, J. Berry, B. Osmond, Measuring photosynthetic parameters at a distance: laser induced fluorescence transient (LIFT) method for remote measurements of photosynthesis in terrestrial vegetation, Photosynth. Res. 84, 121-129 (2005). 82. B. Valeur, Molecular Fluorescence: Principles and Applications, Wiley-VCH Verlag GmbH, 2001. 83. D. M. Jameson, E. Gratton, R. D. Hall, The Measurement and Analysis of Heterogeneous Emissions by Multifrequency Phase and Modulation Fluorometry, Appl. Spectrosc. Rev. 20 (1), 55-106 (1984). 84. S. Bjurshagen, J. E. Hellstrm, V. Pasiskevicius, M. C. Pujol, M. Aguil, F. Daz, Fluorescence dynamics and rate equation analysis in Er 3+ and Yb 3+ doped double tungstates, Appl. Opt. 45 (19), 4715-4725 (2006). 85. A. Makoui, D. Killinger, Tunable UV LIF of terbium-doped bacteria endospores, in Proceedings of Optically Based Biological and Chemical Sensing for Defence, J. C. Carrano, A. Zukauskas, eds., Proc. SPIE 5617, 202208 (2004). 86. A. Makoui, D. Killinger, Laser Induced Fluorescence (LIF) Lifetime Studies of Terbium-doped DPA, in Proceedings of Chemical and Biological Sensors for Industrial and Environmental Security, A. J. Sedlacek, S. D. Christesen, R. J. Combs, T. Vo-Dinh, eds., Proc. SPIE 5994, 59940Q (2005).

PAGE 258

233 87. A. Makoui, D. Killinger, Laser Enhanced Spectrsocopic Detection of Biological Endospores, OSA: Optics in the Southeast, Orlando, FL (November 2003). 88. A. Makoui, D. Killinger, Tunable UV LIF of Terbium-doped Bacteria Endospores, Bacterial Spores: Challenges and Future Directions for Biodefense, Argonne National Laboratory, Argonne, IL (November 2004). 89. A. Makoui, A. Sharikova, D. Killinge r, Detection of Trace Organics and Bacteria Endospores in Water Using UV Laser Induced Fluorescence (LIF), P12, Conference on Homeland Security, Washington, D.C. (December 2005). 90. A. Makoui, D. Killinger, Deep UV Laser Induced Fluorescence for Sensitive Detection of DPA in Bacterial Spores, USF Interdisciplinary Graduate Student Research Symposium (April 2006). 91. A. Makoui, D. Killinger, Fluorescence Lifetime Studies of Terbium Doped Dipicolinic Acid (DPA) Using a High PRF UV Laser, JThC83, OSA: CLEO/QELS, Long Beach, CA (May 2006).

PAGE 259

234 APPENDIX A ABSORBANCE SPECTRA This appendix contains the graphs of the measured absorbance as a function of wavelength for different concentrations of each DPA and TbCl 3 separately, in distilled water, in [0.01M] HCl, and in [1M] SAB. The spectra of the absorbance of solutions containing mixtures of different concentrations of DPA and TbCl 3 as a function of wavelength, in [0.01M] HCl and in [1M] SAB can also be found here. The last two graphs are an enlarged view, i.e. 200 nm-400 nm, of these same spectra.

PAGE 260

Figure A.1 Measured absorbance of different concentrations of DPA in distilled water; water background not subtracted. 235

PAGE 261

Figure A.2 Measured absorbance of different concentrations of DPA in [0.01M] HCl; HCl background not subtracted. 236

PAGE 262

Figure A.3 Measured absorbance of different concentrations of DPA in [1M] SAB; SAB background not subtracted. 237

PAGE 263

Figure A.4 Measured absorbance of different concentrations of TbCl 3 in distilled water; water background not subtracted. 238

PAGE 264

Figure A.5 Measured absorbance of different concentrations of TbCl 3 in [0.01M] HCl; HCl background not subtracted. 239

PAGE 265

Figure A.6 Measured absorbance of different concentrations of TbCl 3 in [1M] SAB; SAB background not subtracted. 240

PAGE 266

Figure A.7 Measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [0.01M] HCl. 241

PAGE 267

Figure A.8 Measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [1M] SAB. 242

PAGE 268

Figure A.9 Enlarged view of the measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [0.01M] HCl. 243

PAGE 269

Figure A.10. Enlarged view of the measured absorbance spectra of solutions of [5M] DPA with varying [TbCl 3 ] (i.e. [5M], [10M], [15M], [20M], and [25M]), (a), of [10M] DPA with varying [TbCl 3 ], (b), of [15M] DPA with varying [TbCl 3 ], (c), of [20M] DPA with varying [TbCl 3 ], (d), and of [25M] DPA with varying [TbCl 3 ], (e), in [1M] SAB. 244

PAGE 270

245 APPENDIX B DETAILS OF LIFETIME CALCULATIONS The method used to obtain the fluorescen ce lifetimes from the plot of fluorescence intensity as a function of time is presented in this appendix. All of the signals presented here are from a solution of [50M] DPA with [50M] TbCl 3 in distilled water.

PAGE 271

0 1 2 3 4 5 6 7 8 9 10 -0.02 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Time (ms)Fluorescence Intensity(Arbitrary Units) 488 nm line 543 nm line All fluorescence lines mixed together 581 nm line 618 nm line Figure B.1 Fluorescence intensity as a function of time without any corrections: i.e. the raw signal. 246

PAGE 272

0 1 2 3 4 5 6 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Time (ms)Fluorescence Intensity (Arbitrary Units) 488 nm line 543 nm line All fluorescence lines mixed together 581 nm line 618 nm line Figure B.2. The raw signal with the offset (at 6ms) subtracted. 247

PAGE 273

0 1 2 3 4 5 6 0 2 4 6 8 Time (ms)Fluorescence Intensity(Arbitrary Units) All fluorescence lines mixed together 543 nm line 488 nm line 581 nm line 618 nm line Figure B.3 The normalized, i.e. the raw signal with the offset subtracted, and the effects of the laser intensity, the gratings efficiency, and the PMTs sensitivity accounted for. 248

PAGE 274

0 1 2 3 4 5 6 1 2 3 4 5 6 7 8 9 10 Time (ms)Ln (Fluorescence Intensity) All fluorescence signals mixed together 618 nm line 581 nm line 488 nm line 543 nm line Figure B.4 Natural logarithm of the normalized fluorescence intensity as a function of time for a solution of [50M] DPA with [50M] TbCl 3 in distilled water. 249

PAGE 275

1 2 3 2 3 4 5 6 7 8 9 Time (ms)Ln (Fluorescence Intensity) All fluorescence signals mixed together 488 nm line 581 nm line 618 nm line 543 nm line Segment ofCurvesUsed for Slope Calculations Figure B.5 Enlarged view of plots of the natural logarithm of the normalized fluorescence intensity as a function of time for a solution of [50M] DPA with [50M] TbCl 3 in distilled water. 250

PAGE 276

1 2 3 4 5 6 7 Time (ms)Ln (Fluorescence Intensity) Point # 1(0.001793,-0.03105) Point # 2(0.002428,-0.7619) Figure B.6 Enlarged view of the points used in the calculations to find the slope of the 488 nm line. 251

PAGE 277

0 1 2 3 4 5 6 1 2 3 4 5 6 7 8 9 10 Time (ms)Ln (Fluorescence Intensity) Point # 1(0.001793,-0.03105) Point # 2(0.002428,-0.7619) 488 nm line Figure B.7 Example of points used to find the slope of the line, the negative inverse of which is equal to the fluorescence lifetime of the samples 488 nm line. Note: the y-axis has been labeled arbitrarily, but consistently with other graphs of the same type for this sample. 252

PAGE 278

In order to find the error associated with the slope of the line, i.e. with the lifetime value, alternate lines with the maximum possible slope and the smallest possible slope were also plotted, as they have been shown in the above figure in dotted lines. Next, these slopes were used in order to find the smallest possible and the largest possible lifetime values, using the negative of their inverse, as described previously. From these values the errors for the lifetime were found. It is important to note that the point of reference used to plot the maximum and minimum slopes was the upper part of the graphs slope, since in this area the signal to noise ratio was significantly smaller than that of the bottom part of the slope, which due to digitization had significantly larger S/N ratio. A sample calculation for the most reasonable value of the slope for the data presented in this appendix is given below. 331Slope(0.76)(0.03)Slope1.15110/sec0.00240.001810.8710sec-1151 253

PAGE 279

254 APPENDIX C FLUORESCENCE SIGNALS USING ASYMMETRIC EXCITATION WAVEFORMS The initial study of the effect of asymmetric excitation waveforms on our measurements is shown in this appendix. The asymmetric waveform is when the positive (on) excitation intensity patterns are not equal or symmetric to the off portions of the waveform. Such asymmetries can alter the symmetry of the fluorescence emission, so that the long-term average fluorescence inte nsity level changes as the chopper speed is varied. It is anticipated that such asymmetric excitation waveform may be suitable for more complex fluorescing systems, and that the effect of exc itation and fluorescence decay processes can be tailored to accentuat e selected transfer processes amongst the energy levels. As can be seen, the interplay of the excitation waveform and the transfer/excitation process within the DPATb complex can be controlled as to the resultant maximum fluorescence peak, the average fluorescence level, and the minimum fluorescence peak.

PAGE 280

T imeIntensity 30 slit blade at 50 Hz 0 200 400 600 Signal IntensityfChopper 50 Hz 0 200 400 600 fChopper 100 Hz 0 10 20 30 40 0 200 400 600 Time (ms)Signal IntensityfChopper 250 Hz 0 10 20 30 40 0 200 400 600 Time (ms)fChopper 500 Hz Figure C.1 Fluorescence signals obtained using the symmetric 30 slit blade at the indicated chopper frequencies. 255

PAGE 281

Time I n t ens i t y 15 slit blade at 50 Hz 0 200 400 600 Signal IntensityfChopper 50 Hz 0 200 400 600 fChopper 100 Hz 0 10 20 30 40 0 200 400 600 Time (ms)Signal IntensityfChopper 250 Hz 0 10 20 30 40 0 200 400 600 Time (ms)fChopper 500 Hz Figure C.2 Fluorescence signals obtained using the asymmetric 15 slit blade at the indicated chopper frequencies. 256

PAGE 282

Superimposed blade at 50 Hz 0 200 400 600 Signal IntensityfChopper 16 Hz 0 200 400 600 fChopper 50 Hz 0 20 40 60 0 200 400 600 Time (ms)Signal IntensityfChopper 80 Hz 0 20 40 60 0 200 400 600 Time (ms)fChopper 160 Hz Figure C.3 Fluorescence signals obtained using the asymmetric superimposed blades at the indicated chopper frequencies. 257

PAGE 283

258 APPENDIX D NUMERICAL EVALUATION OF THE RATE EQUATIONS FOR 4-LEVEL MODEL AND FOR ASYMMETRIC EXCITATION The following program was written and used in Matlab to numerically calculate the solution to the set of coupled differential equations which describe the rate equations of the DPA-Tb energy level system. In this program a laser pulse which consists of an on time of 0.4 ns (in increments of dt = 0.04ns) and an off time of 118 s (in increments of dt = 1.2s), with a PRF of 8.5 KHz, is passed through a chopper blade which runs at different frequencies (or range of frequencies based on user input). The new intensity profile of the laser beam, which is very similar to that shown in Fig. 7.6, is then used as the incident intensity in the rate equations of the simple model, i.e. Eq. (8.18.5). The population densities of each of th e energy levels as a function of time are then obtained numerically by solving for dN 1 at time t = i th dt, and calculating N 1 [t = (i th + 1) dt] = N 1 [t = i th dt + dN 1 (t = i th dt), for example]. The resultant values for the population densities were then analyzed and plotted using Matlab. Similar programs have been used for the 5-Level (quenching) model and the upconversion model. Lastly, an initial computer program for asymmetric excitation is also presented, for completeness.

PAGE 284

259 % SIMPLE FOUR LEVEL SYSTEM clear; chop_freq = 50; % Operating chopper frequency. run_time = 0.040; % Time (in seconds) you would like to simulate the system for. N0 = 3.0e16; % Population densities at t=0sec. In molecules/cm3. N1 = 0.0; N2 = 0.0; N3 = 0.0; I = 1.1e22; % Laser intensity in photons/cm2.s sigma = 1.9e-17; % Experimentally measured absorption cross section of the DPA-Tb complex. In cm2/molecule. g12 = 5000; % Transfer rates. In 1/sec. g23 = 1053; g30 = 10000; dt_on = 0.04e-9; % Time step (in sec) when the laser intensity is nonezero in one period of the laser. This is 1/10 of the pulse width, i.e. 0.4ns/10. dt_off = 1.18e-6; % Time step (in sec) when the laser intensity is zero in one period of the laser. This is 1/100 of the laser period, i.e. 118us/100. num_laser_periods = run_time/118e-6; % Numer of laser periods that "fit" in the specified "run_time". chop_per = 1/chop_freq; % The chopper period (in sec.) num_chop_per = run_time*chop_freq; % Number of chopper periods that "fit" in the specified "run_time". chop_pulses = round((0.5*chop_per)/118e-6) %Number of laser pulses that pass through the open window of the symmetric chopper blade, %i.e. half period. n=0; %Initial value of the counter. t = 0.0; % Initial time. fid = fopen( 'E:\Chop_Sim\output.txt' 'a+'); % Opens (or creates) file and appends data to it. Need to delete old file before Need to delete old file before running new simulation. for k=1:num_laser_periods % The program runs for the total number of laser periods that fit in the "run-time". if mod(k,chop_pulses)==0 % Every half chopper period, the counter adds one to n. n = n+1; end if mod(n,2)==0 % The chopper window is "on". for i=1:10 % The "on" part of one laser period when chopper window is open. dN0 = (I sigma (N1-N0) + g30 N3) dt_on; dN1 = (I sigma (N0-N1) g12 N1) dt_on;

PAGE 285

260 dN2 = (g12 N1 g23 N2) dt_on; dN3 = (g23 N2 g30 N3) dt_on; t = t + dt_on; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end for j=1:100 % The "off" part of one laser period when chopper window is open. dN0 = g30 N3 dt_off; dN1 = g12 N1 dt_off; dN2 = (g12 N1 g23 N2) dt_off; dN3 = (g23 N2 g30 N3) dt_off; t = t + dt_off ; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end else if mod(n,2)==1 % The chopper window is "off" for i=1:10 % The closed chopper window is killing the "on" part of one laser period... dN0 = g30 N3 dt_on; dN1 = g12 N1 dt_on; dN2 = (g12 N1 g23 N2) dt_on; dN3 = (g23 N2 g30 N3) dt_on; t = t + dt_on; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end for j=1:100 % The "off" part of one laser period when the chopper window is closed. dN0 = g30 N3 dt_off; dN1 = g12 N1 dt_off; dN2 = (g12 N1 g23 N2) dt_off; dN3 = (g23 N2 g30 N3) dt_off;

PAGE 286

261 t = t + dt_off; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end end nd e end fclose(fid)

PAGE 287

262 % SIMPLE MODEL (ASYMMETRIC CHOPPER) clear; chop_freq = 50; % Operating chopper frequency chop_per_width = 16; % Width of one chopper period. In mm. chop_open_width = 4; % Width of open window/slit in one chopper period. In mm. chop_close_width = 12; % Width of closed window/slit in one chopper period. In mm. run_time = 0.040; % Time (in seconds) you would like to simulate the system for. N0 = 3.0e16; % Population densities at t=0sec. In molecules/cm3. N1 = 0.0; N2 = 0.0; N3 = 0.0; I = 1.1e22; % Laser intensity in photons/cm2.s sigma = 1.9e-17; % Experimentally measured absorption cross section of the DPA-Tb complex. In cm2/molecule. g12 = 5000; % Transfer rates. In 1/sec. g23 = 1053; g30 = 10000; dt_on = 0.04e-9; % Time step (in sec) when the laser intensity is nonezero in one period of the laser. This is 1/10 of the pulse width, i.e. 0.4ns/10. dt_off = 1.18e-6; % Time step (in sec) when the laser intensity is zero in one period of the laser. This is 1/100 of the laser period, i.e. 118us/100. num_laser_periods = run_time/118e-6; % Numer of laser periods that "fit" in the specified "run_time". chop_per_t = 1/chop_freq; % Time of one chopper period (in sec). t_chop_open = (chop_open_width/chop_per_width)*chop_per_t % Time it takes the open window of the chopper to pass a small point. t_chop_closed = chop_per_t t_chop_open % Time it takes the closed window of the chopper to pass a small point. num_chop_per = run_time*chop_freq % Numer of chopper periods that fit into the "run-time". chop_pulses_open = round(t_chop_open/118e-6) % Number of laser pulses that fit in the open window of the chopper (at the specified chop. freq.) chop_pulses_closed = round(t_chop_closed/118e-6) % Number of laser pulses that fit in the closed window of the chopper (at the specified chop. freq.) chop_per_pulses = chop_pulses_open + chop_pulses_closed % % Total number of laser pulses that "fit" during one chopper period. n=0; %Initial value of the counter. t = 0.0; % Initial time.

PAGE 288

263 fid = fopen( 'E:\Chop_Sim\output_asymt.txt' 'a+'); % Opens (or creates) file and appends data to it. Need to delete old file before Need to delete old file before running new simulation. for k=1:num_laser_periods % The program runs for the total number of laser periods that fit in the "run-time". if mod(k,chop_per_pulses)==0 % Keeps track of numer of chopper periods. n = n+1; end if (k>=n*chop_per_pulses) && (k<(n*chop_per_pulses+chop_pulses_open)) % Chopper window is open for i=1:10 % The "on" part of one laser period when chopper window is open dN0 = (I sigma (N1-N0) + g30 N3) dt_on; dN1 = (I sigma (N0-N1) g12 N1) dt_on; dN2 = (g12 N1 g23 N2) dt_on; dN3 = (g23 N2 g30 N3) dt_on; t = t + dt_on; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end for j=1:100 % The "off" part of one laser period when chopper window is open dN0 = g30 N3 dt_off; dN1 = g12 N1 dt_off; dN2 = (g12 N1 g23 N2) dt_off; dN3 = (g23 N2 g30 N3) dt_off; t = t + dt_off; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end else % Chopper window is closed for i=1:10 % The closed chopper window is killing the "on" part of one laser period... dN0 = g30 N3 dt_on; dN1 = g12 N1 dt_on; dN2 = (g12 N1 g23 N2) dt_on; dN3 = (g23 N2 g30 N3) dt_on;

PAGE 289

264 t = t + dt_on; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end for j=1:100 % The "off" part of one laser period when the chopper window is closed. dN0 = g30 N3 dt_off; dN1 = g12 N1 dt_off; dN2 = (g12 N1 g23 N2) dt_off; dN3 = (g23 N2 g30 N3) dt_off; t = t + dt_off; N0 = N0 + dN0; N1 = N1 + dN1; N2 = N2 + dN2; N3 = N3 + dN3; fprintf(fid, '%e %e %e %e %e \n' t, N0, N1, N2, N3); end end end fclose(fid)

PAGE 290

ABOUT THE AUTHOR Anali Makoui Roohy obtained a Bachelor of Science degree in Physics from the University of South Florida, Fl orida, U.S.A. in 2000, and a Masters of Science degree in Physics from the University of South Florida, Florida, U.S.A. in 2002 under the guidance of Prof. Dennis K. Killinger. She entere d the Ph.D. program in Applied Physics at the University of South Florida in Fall 2002. She completed an industrial practicum at Ocean Optics Inc., Dunedin, Florida as part of the Applied Physics practical training. She has a joint patent in Differential Spectroscopic Imaging of the Human Retina (Patent number: US 6,709,109 B1), and has presented papers at severa l technical conferences incl uding SPIE, OSA, and CLEO.


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 Ka
controlfield tag 001 001968110
003 fts
005 20081105095209.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 081105s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002183
035
(OCoLC)268957448
040
FHM
c FHM
049
FHMM
090
QC21.2 (ONLINE)
1 100
Makoui, Anali.
0 245
Transient fluorescence spectroscopy and laser induced fluorescence lifetimes of terbium doped dipicolinic acid
h [electronic resource] /
by Anali Makoui.
260
[Tampa, Fla.] :
b University of South Florida,
2007.
3 520
ABSTRACT: We have investigated the use of deep UV laser induced fluorescence for the sensitive detection and spectroscopic lifetime studies of terbium doped dipicolinic acid (DPA-Tb) and used this to study the optical characteristics of DPA which is a chemical surrounding most bacterial spores. Background absorption spectra, fluorescence spectra, and Excitation Emission Matrix (EEM) spectra were made of the DPA-Tb complex, using both fixed 266 nm wavelength and tunable (220 nm --^ 280 nm) UV laser excitations. Of importance, the fluorescence lifetimes of the four main fluorescence peaks (488 nm, 543 nm, 581 nm, and 618 nm) of the DPA-Tb complex have been measured for the first time to our knowledge. The lifetimes of all the fluorescing lines have been measured as a function of DPA-Tb concentration, solvent pH, and solvent composition, including that for the weakest fluorescing line of DPA-Tb at 618 nm. In addition, a new spectroscopic lifetime measurement technique, which we call "Transient Fluorescence Spectroscopy" was developed. In this technique, a weak, quasi-CW, amplitude modulated UV laser (8.5 kHz) was used to measure the lifetimes of the fluorescence lines, and yields insight into energy transfer and excitation lifetimes within the system. This technique is especially useful when a high power laser is not either available or not suitable.^ In the latter case, this would be when a high power pulsed deep-UV laser could produce bleaching or destruction of the biological specimen. In addition, this technique simulated the excitation and fluorescence emission of the DPA-Tb using a 4-level energy model, and solved the dynamic transient rate equations to predict the temporal behavior of the DPA-Tb emitted fluorescence. Excellent agreement between the experiments and the simulation were found. This technique has the potential to provide a more accurate value for the fluorescence lifetime values. In addition, with the use of asymmetric excitation waveforms, the dynamic transient rate equation analysis may allow for detailed studies of selected transfer mechanisms in a wide range of other spectroscopic applications including rare-earth solid-state lasing materials and biological samples.
502
Dissertation (Ph.D.)--University of South Florida, 2007.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 264 pages.
Includes vita.
590
Adviser: Dennis Killinger, Ph.D.
653
Laser induced fluorescence spectroscopy.
Transient saturation fluorescence spectroscopy.
Fluorescence lifetime.
Bacteria spore detection.
Excitatin emission matrix spectra.
690
Dissertations, Academic
z USF
x Physics
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2183