USF Libraries
USF Digital Collections

Application of luminescence sensors in oxygen diffusion measurement and study of luminescence enhancement/quenching by m...

MISSING IMAGE

Material Information

Title:
Application of luminescence sensors in oxygen diffusion measurement and study of luminescence enhancement/quenching by metallic nanoparticles
Physical Description:
Book
Language:
English
Creator:
Chowdhury, Sanchari
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Quantum efficiency
Oxygen diffusion
Stern-Volmer plot
Surface plasmon resonance
Alloy nanoparticles
Optical properties
Dielectric constant
Dissertations, Academic -- Chemical Engineering -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The first part of this dissertation deals with the application of a luminescence quenching method to measure diffusion and permeation coefficients of oxygen in polymers. Most luminescence oxygen sensors do not follow linearity of the Stern-Volmer (SV) equation due to heterogeneity of luminophore in the polymer matrix, thus the complexity of data analysis is increased. To circumvent this limitation, inverted fluorescence microscopy is utilized in this work to investigate the SV response of the sensors at the micron-scale. In these diffusion experiments, oxygen concentration is measured by luminescence changes in regions with high SV constants and good linearity. Thus, we avoid numerical complexity of combining nonlinear SV equation with a diffusion model. This technique allows us to measure oxygen diffusion properties in different type of polymers like transparent, opaque, free-standing polymers and polymers that cannot be cast into free standing films and polymer composites. In the second part of this thesis, we have explored the effect of Ag-Cu alloy nanoparticles on the emission intensity of luminophores at their close proximity. Alloy nanoparticles offer additional degrees of freedom for tuning their optical properties by altering atomic composition and atomic arrangement and thus can be an attractive option for manipulating signal of a wide range of luminophores. In this work, surface plasmon resonance spectrum of Ag-Cu alloy nanoparticles deposited by sputtering was easily tuned in wide wavelength range by varying one experimental condition- annealing temperature. Large metal enhanced luminescence for different luminophores viz Alexa Fluor 594 and Alexa Fluor 488 were achieved at the vicinity of Ag-Cu nanoparticles when maximum spectral overlap between SPR spectra of Ag-Cu nanoparticles and the emission and absorption spectra of the luminophores occur. We also studied the effect of composition of Ag-Cu nanoparticles synthesized by the polyol process on the luminescence of low quantum yield dye Cy3. In the third part of this thesis, quenching effect of Cu nanoparticles on CdSe/ZnS nanocrystal quantum dots has been explored. As Cu nanoparticles have comparable dielectric properties with gold nanoparticles, they are expected to show similar quenching effects. It was found that Cu is an efficient quencher of fluorescence from CdSe/ZnS quantum dots and the quenching effect is due to resonance energy transfer from quantum dots to Cu nanoparticles.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Sanchari Chowdhury.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X 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:
usfldc doi - E14-SFE0003313
usfldc handle - e14.3313
System ID:
SFS0027629:00001


This item is only available as the following downloads:


Full Text
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 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003313
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Chowdhury, Sanchari.
0 245
Application of luminescence sensors in oxygen diffusion measurement and study of luminescence enhancement/quenching by metallic nanoparticles
h [electronic resource] /
by Sanchari Chowdhury.
260
[Tampa, Fla] :
b University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: The first part of this dissertation deals with the application of a luminescence quenching method to measure diffusion and permeation coefficients of oxygen in polymers. Most luminescence oxygen sensors do not follow linearity of the Stern-Volmer (SV) equation due to heterogeneity of luminophore in the polymer matrix, thus the complexity of data analysis is increased. To circumvent this limitation, inverted fluorescence microscopy is utilized in this work to investigate the SV response of the sensors at the micron-scale. In these diffusion experiments, oxygen concentration is measured by luminescence changes in regions with high SV constants and good linearity. Thus, we avoid numerical complexity of combining nonlinear SV equation with a diffusion model. This technique allows us to measure oxygen diffusion properties in different type of polymers like transparent, opaque, free-standing polymers and polymers that cannot be cast into free standing films and polymer composites. In the second part of this thesis, we have explored the effect of Ag-Cu alloy nanoparticles on the emission intensity of luminophores at their close proximity. Alloy nanoparticles offer additional degrees of freedom for tuning their optical properties by altering atomic composition and atomic arrangement and thus can be an attractive option for manipulating signal of a wide range of luminophores. In this work, surface plasmon resonance spectrum of Ag-Cu alloy nanoparticles deposited by sputtering was easily tuned in wide wavelength range by varying one experimental condition- annealing temperature. Large metal enhanced luminescence for different luminophores viz Alexa Fluor 594 and Alexa Fluor 488 were achieved at the vicinity of Ag-Cu nanoparticles when maximum spectral overlap between SPR spectra of Ag-Cu nanoparticles and the emission and absorption spectra of the luminophores occur. We also studied the effect of composition of Ag-Cu nanoparticles synthesized by the polyol process on the luminescence of low quantum yield dye Cy3. In the third part of this thesis, quenching effect of Cu nanoparticles on CdSe/ZnS nanocrystal quantum dots has been explored. As Cu nanoparticles have comparable dielectric properties with gold nanoparticles, they are expected to show similar quenching effects. It was found that Cu is an efficient quencher of fluorescence from CdSe/ZnS quantum dots and the quenching effect is due to resonance energy transfer from quantum dots to Cu nanoparticles.
590
Advisor: Venkat R. Bhethanabotla, Ph.D.
653
Quantum efficiency
Oxygen diffusion
Stern-Volmer plot
Surface plasmon resonance
Alloy nanoparticles
Optical properties
Dielectric constant
690
Dissertations, Academic
z USF
x Chemical Engineering
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3313



PAGE 1

Application of Luminescence Sensors in Oxygen Diffusion Measurement and Study of L uminescence Enhancement/Quenching by M etallic N anoparticles by Sanchari Cho w dhury A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemical and Biomedical Engineering College of Engineering University of South Florida Co Major Professor: Venkat R. Bhethanabotla, Ph.D. Co Major Professor: Rajan Sen Ph.D. Randy W. Larsen Ph.D. Vinay K. Gupta, P h.D. Ryan Toomey Ph.D. Date of Approval: March 24, 2010 Keywords: quantum efficiency, oxygen diffusion, Stern Volmer plot, surface plasmon resonance, alloy nanoparticles, optical properties, dielectric constant Copyright 2010, Sanchari Cho w dhury

PAGE 2

Dedication T o my lovely son Soham

PAGE 3

Acknowledgements First, I would like to sincerely thank my advisors Prof. Venkat R Bhethanabotla and Prof. Rajan Sen, for their excellent guidance, constant encouragement and support throughout my PhD studies. This dissertation would not be possible without their intellectual support. I thank Prof. Vinay Gupta, Prof. Randy Larsen and Prof. Ryan Toomey for agreeing to be in my Ph.D. defense committee and for giving me some valuable suggestions during my proposal defense. M y appreciation also extends to the former and present group members of my lab, Dr. Stefan Cular, Dr. Subramanian San karanarayanan, Dr. Reetu Singh, Ni anthrini Balakrishnan, Ayse Gul Yavuz Pedro Villalba and Chandra Khoe for their advice and help through these last four years. I must give particular praise to Dr. Stefan Cular for instructing me in the initial techniques for my project. I also thank al l those who directly collaborated with me on my thesis projects. As one of the users of their systems, I would also like to acknowledge the expe rimental resources provided by Nanomaterial and Nanomanufacturing Research C enter (NNRC) of USF without which, this dissertation would be possible. Great support and technical guidance provided by staff members of NNRC are highly appreciated. Fi nally I would like to thank my husband who listened to my ideas, offered suggestions, and always gave me energy, and en couragement. M y parents deserve my sincerest love and appreciation for always being there for me with all their unconditional love and encouragement throughout the ups and downs of my life.

PAGE 4

i Table of Contents Table of Contents ............................................................................................................ i List of Tables ................................................................................................................. vi List of Figures .............................................................................................................. vii ABSTRACT ....................................................................................................................x Chapter 1 Introduction ...................................................................................................1 1.1. Introduction to Fluorescence .........................................................................1 1.2. Motivation and Objectives .............................................................................2 1.3. Organization of the Dissertation ....................................................................4 Chapter 2 Background ...................................................................................................5 2.1. Luminescence ...............................................................................................5 2.2. Luminescence Quenching ..............................................................................7 2.2.1. Frster Resonance Energy Transfer .................................................7 2.2.2. Collisional Quenching and Static Quenching ..................................8 2.3. Metal Enhanced Luminescence ................................................................... 11 2.3.1. Distance Dependence .................................................................... 13 2.3.2. Effect of Surface Plasmon Resonance of Metal Nanoparticles on Luminescence ................................................... 14

PAGE 5

ii 2.3.3. Metal Nanoparticles Used for Metal Enhanced L uminescence and Their Synthesis ............................................... 15 2.3.4. Metal Nanoparticles Quenched Luminescence .............................. 18 2.4. Theoretical Modeling .................................................................................. 19 2.4.1. Theoretical Investigation of Surface Plasmon Resonance of Nanoparticles ............................................................................... 20 2.4.2. Modeling of Plasmon Enhanced Luminescence ............................. 25 2.4.2.1 Calculation of Excitation Enhancement Factor ................ 26 2.4.3. Modeling of Effect of Metal Sphere on Excited State Decay Rate .............................................................................................. 27 2.4.3.1 Exact Electrodynamic Theory .......................................... 27 2.4. 3.2 Gersten Nitzan (GN) Model ............................................ 29 2.5. Bimetallic Nanoparticles ............................................................................. 31 2.5.1. Plasmonic Properties ..................................................................... 31 2.5. 2. Synthesis ...................................................................................... 33 2.5.2.1 Synthesis of Bimetallic Alloy Nanoparticles .................... 34 2.6. Characterization Techniques ........................................................................ 35 2.6.1. Transmission Electron Microscopy (TEM) .................................... 35 2.6.2. Scanning Electron Microscopy (SEM) .......................................... 36 2.6.3. Atomic Force Microscopy (AFM) ................................................. 36 2.6.4. Energy Dispersive X ray Spectroscopy (EDS) .............................. 37 2.6.5. UV Vis Absorption Spectroscopy ................................................. 37 2.6.6. Fluorescence Microscopy .............................................................. 37

PAGE 6

iii 2.6.7. Fluorescence Spectroscopy ........................................................... 38 Chapter 3 Measurement of Oxygen Diffusivity and Permeability in Polymers Using Fluorescence Microscopy ................................................................... 39 3.1. Introduction ................................................................................................. 39 3.2. Materials and Methods ................................................................................ 44 3.2.1. Sensor Films and Polymers ........................................................... 44 3.2.2. Instrumentation and Software ........................................................ 46 3.2.3. Image Analysis ............................................................................. 48 3.3. Analytical Models ....................................................................................... 49 3.3.1. Fil m Separated from The Luminescent Sensor by A Small Volume (Accumulation in volume Case) ...................................... 49 3.3.1.1 Ficks Equation Combined with Th e SV Equation ........... 49 3.3.1.2 Quasi steady State Model ................................................ 51 3.3.2. Film on sensor Model .................................................................. 52 3.4. Results and Discussion ................................................................................ 52 3.4. 1. Characterization of Sensors ........................................................... 53 3.4.2. Measurement of Diffusion Using Fluorescence Microscopy .......... 58 3.5. Conclusions ................................................................................................. 64 Chapter 4 Effect of Ag Cu Alloy Nanoparticle Composition on Luminescence Enhancement/Quenching .............................................................................. 67 4.1. Introduction ................................................................................................. 67 4.2. Experimental ............................................................................................... 72 4.3. Results and Discussion ................................................................................ 75

PAGE 7

iv 4.4. Conclusions ................................................................................................. 84 Chapter 5 Silver Copper Alloy Nanoparticles for Metal Enhanced Luminescence ....... 86 5.1. Introduction ................................................................................................. 86 5.2. Experimental Method .................................................................................. 89 5.3. Results and Discussion ................................................................................ 93 5.4. Conclusions ............................................................................................... 104 Chapter 6 Quenching of Fluorescence from CdSe/ZnS Nanocrystals near Copper Nanoparticles in Aqueous Solution ............................................................. 106 6.1. Introduction ............................................................................................... 106 6.2. Experimental ............................................................................................. 109 6.2.1. PVP Coa ted Cu Nanoparticles Synthesis ..................................... 109 6.2.2. CTAB Coated Cu Nanoparticle Synthesis ................................... 110 6.2.3. Nanoparticles Characterization .................................................... 110 6.2.4. Fluor escence Quenching Experiment .......................................... 111 6.3. Results and Discussion .............................................................................. 111 6.3. 1. Characterization of PVP Coated Copper Nanoparticles ............... 111 6.3.2. Collisional Quenching by PVP Coated Copper Nanoparticles ..... 112 6.3.3. Characterization of CTAB Coated Cu Nanoparticles ................... 116 6.3.4. Quenching Effect of CTAB Coated Cu Nanoparticles on CdSe/ZnS Nanocrystals .............................................................. 119 6.3.5. Quenching Mechanism and Effect of Size of Cu Nanoparticles on Quenching Efficiency ...................................... 123 6.4. Summary and Conclusions ........................................................................ 126

PAGE 8

v Chapter 7 Summary, Conclusions and Rec ommendations .......................................... 128 7.1. Introduction ............................................................................................... 128 7.2. Measurement of O2 Diffusion Properties Using Inverted Fluorescence Microscopy ............................................................................................. 128 7.3. AgCu Nanoparticles for Enhanced Luminescence .................................... 129 7.4. Fluorescence Quench ing by Cu Nanoparticles ........................................... 130 7.5. Major Contributions .................................................................................. 131 7.6. Future Directions ....................................................................................... 132 7.6.1. F luorescence Microscopy for Simultaneous Imaging and O2 Diffusion Measurement .............................................................. 132 7.6.2. Exploration of Other Alloy Nanoparticles fo r Metal Enhanced Luminescence ............................................................ 133 7.6.3. Application of Alloy Nanoparticles for Enhancement of Photovoltaic Cells ...................................................................... 133 7.6.4. Development of Sensors Based on the Quenching Property of Cu Nanoparticles .................................................................... 134 7.6.5. Theoretical and Computational Modeling of Optical Properties of Alloy Nanoparticles ............................................... 134 References .. ..................................................................................................... 136 About The Author ............................................................................................... End Page

PAGE 9

vi List of Tables Table 3 1 SV constants of different microscopic regions of luminescence sensors ........ 56 Table 3 2 Oxygen diffusion coefficients for various polymers ...................................... 61 Table 5 1 Fluorescence enhancements of Alexa Fluor 488 an d Alexa Fluor 594 on the Ag and Ag Cu nanoparticles. ........................................................... 102 Table 6 1 Concentration of reactants and characteristics of the synthesized Cu nanoparticles .............................................................................................. 118 Table 6 2 Summary of SV equation and quenching constants for different size CTAB coated Cu nanoparticles .................................................................. 122

PAGE 10

vii List of Figures Figure 21 Jablonski diagram ........................................................................................5 Figure 22 Schematic of Frster resonance energy transfer (FRET) ...............................8 Figure 23 Quenching by intersystem crossing ..............................................................9 Figure 24 Dexter interaction ...................................................................................... 10 Figure 25 Modified Jablonski diagram in the presence of metal ................................. 13 Figure 26 Distance dependence on the effect of metal on luminescence ..................... 13 Figure 31 Molecular Structure of PtOEP and PtTFPP ................................................ 43 Figure 32 Schematic diagram s of diffusion cells for (a) film on sensor e xperiment and (b) accumul ation in volume experiment ........................... 45 Figure 33 Pseudo colored microscopic fluorescence intensity images (1.64 mm X 2.19 mm) of two luminescence sensors (PtTFPP/PS). ............................ 54 Figure 34 S V plot for different regions of sensor a .. 55 Figure 35 Experimental and fitted data for the 0.025 mm thick Teflon film (* experimental data ............................................. 59 Figure 36 Experimental and fitted data (* experimental data model) for the 0.8 mm thick PDMS film. ................................................... 59 Figure 37 Experimental and fitted data (* experimental data model) for 0.55 mm 36265 HP polymer film (silicone elastomer) ............ 60 Figure 38 Data for the 0.65 mm thick PDMS fi lm containing 10% zeolite (* e xpe rimental data model). ............................................. 63 Figure 41 2Figure 42 Molecular structure of Cy3.. ...................................................................... 71 of 10 nm Ag and Cu nanoparticles. ........................................................ 71 Figure 43 Normalized absorption spectra for Ag Cu alloy nanoparticles. Dotted line is for Ag Cu nanoparticles with 33% Cu on glass slides . .................. 75

PAGE 11

viii Figure 44 T EM image of Ag Cu np synthesized from different composition feeding solution (A) Ag/Cu ( 1/ 1) and (B) Ag/ Cu (3/7). STEM EDS spectra for (C)Ag/Cu (1/ 1) (d) Ag/ Cu (2/1) ............................................... 77 Figure 45 SEM images of (A) Ag nanoparticles (B) 2:1 AgCu nanoparticles (C) 1:1 AgCu nanoparticles coated on glass substrate. ............................. 78 Figure 46 AFM images of (D) Ag nanoparticles (E) 2:1 AgCu nanoparticles (F) 1:1 AgCu nanoparticles coated on glass substrates. ............................. 78 F igure 47 Pseudo colored image of Cy3 coated on (A) glass (B) Ag (C) 1:1 Ag Cu and (D) Cu nanoparticles. ............................................................... 79 F igure 48 Experimentally observed luminescence enhancement ration of Cy3. (B) Inset shows the calculated overall quentum efficiency ratio. ............... 81 F igure 49 Calculated overall quentum efficiency measurement factor due to emission enhancement (B) and excitaion enhancement factor ................... 81 F igure 410 Quatntum efficiency enhancement ratio of Cy3 in the proximity of different diamater Ag Cu nanoparticles at different composition. ............. 83 Figure 51 Picture of DC magne tron sputterer with Ag Cu target ................................ 90 Figure 52 Luminophore on Ag Cu nanoparticl es platform ......................................... 92 Figure 53 High resolution TEM image of Ag nanoparticles ....................................... 94 Figure 54 (A) (B) HRTEM image of Ag Cu .............................................................. 95 Figure 55 STEM EDS spectra for Ag Cu alloy nanoparticles ..................................... 96 Figure 56 Absorption spectra of Annealed Ag Cu nanoparticles (su rface ratio of Cu in sputter target is 7.5%) .................................................................. 97 Figure 57 Calculated extinction spectra for the Ag Cu core shell (Ag in core and Cu in shell) materials at different shell layer thickness.. ...................... 98 Figure 58 SPR spectrum of Ag Cu and Ag nanoparticles and absorption and emission spectrum of Alexa Fluor 594 and Alexa Fluor 488. ..................... 99 Figure 59 I mage of Alexa Fluor 488 coated on (A) glass (B ) 448 K annealed Ag Cu Alexa Fluor 594 coated on (C) glass (D) 298 K AgCu. ............. 100 Figure 510 C alculated extinction coefficient (black), and overall quantum efficiency enhancement ratio for (blue) Ag (dotted line) and 1:1 Ag Cu nanospheres (solid line). ............................................................... 103

PAGE 12

ix Figure 511 C alculated extinction coefficient (black), emission enhancement factor (green) and excitation rate enhancement factor (red) for Ag (dotted line) and 1:1 AgCu nanospheres (solid line) .............................. 104 Figure 61 H RTEM micrograph of (A) Cu nanoparticles (B) Red CdSe/ ZnS nanocrystals, and (C) STEM EDS spectra of the Cu nanoparticles. .......... 112 Figure 62 Normalized a bsorbance spectrum of coppe r nanoparticles and emission spectrum of CdSe/ ZnS nanocrystals. ......................................... 113 Figure 63 (A) The emission spectra of yellow nc at different concentration of Cu nanoparticles (B) Queneching efficiency measured at 580 nm ........... 114 F igure 64 Stern Volmer plot of I I0 for 500 nanomolar concentration of CdSe/ ZnS nanocrystals vs. concentration of copper nanoparticles ........... 115 Figure 65 High resolution TEM images of different sizes CTAB coated copper nanoparticles (A) sample a (B) sample b (C) sample c ........................... 117 Figure 66 Normalized a bsorbance spectra of different size Cu nanoparticles in aqueous solution ...................................................................................... 117 Figure 67 Absorbance spectra of 500 micromol and diluted (1 micromol) copper nanoparticles. ............................................................................... 119 Figure 68 Effect of sample a, sampl e b and sample c copper nanoparticles concentration on the 500 nanomol red CdSe/ ZnS nanocrystals. ............... 120 Figure 69 S V plot for 500 nanomolar concentration of red CdSe/ ZnS n c for different size Cu nanoparticles ........... 122 Figure 610 Normalized a bsorbance spectra ( -) of different size Cu nanoparticles and the excitation ( ) and emission spectra () red CdSe/ ZnS nc ....... 124 Figure 611 Relative dynamic quenching constants (KD) ( constants (KSFigure 612 Ratio of theoretically calculated luminescence quantum yields of a dipole emitter with and without copper metal nanosphere . .................... 126 ) ( nanoparticles ................. 126

PAGE 13

x Application of Luminescence Sensors in Oxygen Diffusion Measurement and Study of Luminescence Enhancement/Quenching by Metallic N anoparticles Sanchari Cho w dhury A BSTRACT The first part of this dissertation deals with the application of a luminescence quenching method to measure diffusion and permeation coefficients of oxygen in polymer s Most luminescence oxygen sensors do not follow linearity of the Stern Volmer (SV) equation due to heterogeneity of luminophore in the polymer matrix thus the complexity of data analysis is increased To cir cumvent this limitation, inverted fluorescence microscopy is utilized in this work to investigate the SV response of the sensors at the micron scale. In these diffusion experiments, oxygen concentration is measured by luminescence changes in regions with high SV constants and good linearity. Thus, we avoid numerical complexity of combining nonlinear SV equation with a diffusion model. This technique allows us to measure oxygen diffusion properties in different type of polymers like transparent opaque f reestanding polymers and polymers that cannot be cast into free standing films and polymer composites. In the second part of this thesis, we have explored the effect of Ag Cu alloy nanoparticles on the emission intensity of luminophores at their close pro ximity. Alloy nanoparticles offer additional degrees of freedom for tuning their optical properties by altering atomic composition and atomic arrangement and thus can be an attractive option for manipulating sign al of a wide range of luminophores. In this work s urface plasmon

PAGE 14

xi resonance spectrum of Ag Cu alloy nanoparticles deposited by sputtering was easily tuned in wide wavelength range by varying one experimental conditionannealing temperature. Large metal enhanced luminescence for different lumi nophores viz Alexa Fluor 594 and Alexa Fluor 488 were achieved at the vicinity of AgCu nanoparticles when maximum spectral overlap between SPR spectra of Ag Cu nanoparticles and the emission and absorption spectra of the luminophores occur. We also studi ed the effect of composition of Ag Cu nanoparticles synthesized by the polyol process on the luminescence of low quantum yield dye Cy3. In the third part of this thesis quenching effect of Cu nanoparticles on CdSe/ ZnS nanocrystal quantum dots has been exp lored As Cu nanoparticles have comparable dielectric properties with gold nanoparticles, th ey are expected to show similar quenching effects. It was found that Cu is an efficient que ncher of fluorescence from CdSe/ZnS quantum dots and the quenching effe ct is due to resonance energy transfer from quantum dots to Cu nanoparticles.

PAGE 15

1 Chapter 1 Introduction 1.1. Introduction to Fluorescence Fluorescence is an extensively used method in the fields of biotechnology, sensors, cellular imaging, medical dia gn ostics, immunoassay, flow cytometry, and DNA sequencing, to name a few.1 3 All the observables including quantum yields, anisotropies, spectral shifts and lifetimes, have been used in wide ranging applications of fluorescence.1 There are many facto rs which can influence fluorescence and can result in enhancement or quenching of emission. The change of emission intensity has profound implications in most fluorescence applications. For example, fluorescence quenching by differe nt elements like O2, NO, and heavy metal ions can be used to detect those elements in the environment as well as in biological samples.1 On the other hand, fluorescence enhancement is one of the most important design properties for luminophores in applica tions like improved surface immunoassay cellular imaging, DNA detection, and enhanced wavelength ratiometric sensing, and amplified assay detection.2 Appropriately desi gn ed nanostructured platforms of some conducting metals like Ag, Au, Cu and Al can res ult in strong emission and can reduce the lifetime, thus increasing photostability of vicinal luminophores.3

PAGE 16

2 1.2. Motivation and O bjectives The objective for the first part of this dissertation is to develop an efficient fluores cence quenching based technique for the measurement of oxygen diffusion in polymers using inverted fluorescence microscopy. The motivation behind the first objective is as follows: Luminescence sensors have increasingly found promising applications for measuring oxygen diffu sion properties of polymers as a result of their simplicity and high sensitivity to oxygen concentration changes. Frequently, these methods use the specific assumption that luminescence quenching which occurs in the sensor film in response to O2 concentrat ion follows the linear Stern Volmer (SV) equation.4 This do es not lead to satisfactory results in man y cases as for many luminophore molecules average intensity change with oxygen concentration does not follow the linearity of Stern Volmer equation due t o the heterogeneity of dye dispersed in the polymer matrix. Though several models were developed for describing the nonlinear response of the sensors, all sensors do not follow the same nonlinear model.5The focus of the second part of this dissertation is on establishing scientific principles that exploit the unique and intense optical properties of metal alloy nanoparticles for optimum luminescence enhancement of vi cinal luminophores. The f ollowing motivate this focus: t he most important properties of metallic nanoparticles on It is complicated to derive analytical models comb ining different nonlinear SV models with the Ficks law subjected to different sets of boundary conditions. This nonlinearity issue can be addressed by the proposed fluorescence microscopy technique which would allow one to investigate SV response of luminescence sensors at the micron scale.

PAGE 17

3 which luminescence enhancement depends are the surface plasmon resonance spectra, scattering and ohmic losses of nanoparticles.3 ,7Another objective is to develop a theoretical approach for predicting suitable nanostructures for metal enhanced luminescence and interpreting experiment ally observed phenomena. Application of reliable theoretical models for the effect of metal nanostructures on luminescence would reduce the number of experimental trials and serve as a guideline for producing suitable nanoparticles for both metal enhanced and quenched luminescence. So, a fundamental understanding of the mechanism of influence of different materials and their properties is expected to result from this research. It is expected that this improved unde rstanding will lead to optimum metal nan ostructure platforms for most efficient luminescence applications. Underst anding the effect of these properties thoroughly and the ability to tune these properties to maximize the spectral overlap between emission and excitation spectra of luminophore molecules and surface plasmon resonance spectrum of nanoparticles enable the design of an effective nanoparticle platform which can enhance the intensity of particular luminophores the most. Alloy nanoparticles offer additional degrees of freedom for tuning their above properties by altering atomic composition and atomic arrangemen t, and can be an attractive option for enhancing emission intensity of a wide range of luminophores. The third part of this dissertation deals with the study the quenching effects of Cu nanoparticles on luminescence emission. This is motivated by the fact that luminescence quenching of lu minophores is most ly studied on gold nanoparticle platforms.8 The imaginary component of the dielectric constant of copper is comparable to that of gold in the wavelength range of 400 nm to 500 nm, and almost twice in the wavelength range of

PAGE 18

4 500 nm to 625 nm. Hence, it is expected that Cu nanoparticles will be better and less expensive alternative to gold for luminescence quenching. 1.3. Organization of t he D issertation This disse rtation is organized in seven chapters. Chapter 2 provides the basic concepts of luminescence and the effect of metallic nanoparticles on luminescence are discussed. The details of luminescence quenching and then those of luminescence enhancement by metallic nanoparticles are presented A brief overview of plasmonic proper ties of bimetallic nanoparticles and their synthesis are given after this. Characterization techniques used in this dissertation are described at the end of this chapter. Chapter 3 describes the fluorescence quenching based method for the measurement of o xygen diffusivity and permeability in polymers using fluorescence m icroscopy Chapter 4 discusses study of the e ffect of Ag Cu alloy nanoparticle composition on luminescence enhancement/quenching Chapter 5 studies the manipulation of surface plasmon resonance spectra of s ilver copper alloy nanoparticles and its application in metal enhanced luminescence. Chapter 6 describes fluorescence quenching effect of Cu nanoparticles on CdSe/ ZnS quantum dots in aqueous solution. Chapter 7 summarizes the conten ts of this dissertation and suggests possible future research directions.

PAGE 19

5 Chapter 2 Background 2.1. Luminescence Photoluminescence is a molecular level process which can be described as an excitation to a higher energy state due to absorption of photons which then return a to lower energy state accompanied by the emission of photon s with longer wavelength.1 This phenomenon can be described nicely by the Jablonski diagram (Figure 21) Figure 2-1 Jablonski diagram 1 In the ground state or the singlet state fluorophores can exist in a number of vibrational energy le v els Following light absorption fluorophore molecules are typically excited to some higher vibrational level of S1 or S2. In most cases fluorophore molecules rapidly relax to the lower vibrational energy level of singlet state from where these molecules emit energy as radiative or non radiative decay. This relaxation time is 1012 second or less where as fluorescence lifetime is typically near 1012 second. Absorption Fluorescence Phosphorescence Excited vibrational state T Triplet state S Singlet state ICInternal conversion ISC Intersystem crossing

PAGE 20

6 T he emission energy is less than the excitation energy. This phenomenon was first observed by Sir G.G. Stokes in 1852 in Cambridge.1 The luminescence lifetime and quantum yield are two very important characteristics of luminophores. If populations of luminophores are excited, the lifetime is the time it takes for the number of excited molecules to decay to 1/e or 36.8% of the original population. The quantum yield can be defined as the ratio of number of emitted photons to the number of absorbed photons. A fraction of the energy from the photons at excited state is emitted as non radiative decay. Hence, the quantum yi eld is less than 1. Quantum yield (Q) can be given by Hence, this wavelength shift is called Stokes shift. Photoluminescence can be of two types: phosphorescence and fluorescence. If the emission occurs from excited singlet states then it is called fluorescence. In this case the electron i n the excited state is paired with the electron in the ground state orbital so the return to ground state is allowed. As a result, the fluorescence life time is very short, o f the order of nanoseconds. In case of phosphoresc ence, absorbed photons undergo intersystem crossing into a state of higher spin multiplicity, usual ly a triplet state, and emit photons which return back to the ground state. As this transition is forbidden emission rate is very slow and lifetime is usually in the range of millisecon ds to seconds. nrk Q 21 where is radiative decay rate and nrk is nonradiative decay rate.

PAGE 21

7 2.2. Luminescence Q uenching A number of processes can lead to a quenching in luminescence intensity. These processes can occur during the excited state lifetime for example, collisional quenching, energy transfer, charge transfer reactions or photochemistry, or they may occur due to formation of complexes in the ground state. Quenching due to collisional encounters betwee n luminophore and quencher molecule is called dynamic or collisional quenching. In case of static quenc hing luminophore molecules bind with quencher molecules and form nonfluorescent complex es Resonance energy transfer from luminophore molecule to the acceptor molecule also results in the quenching of fluorescence. In the following sections these quenching pr ocesses are discussed in detail 2.2.1. F rster Resonance Energy T ransfer Resonance energy transfer occurs from excited fluorophore molecule (donor molecu le) to an acceptor molecule. The acceptor molecule can be fluorescent or nonfluorescent. In both cases quenching of fluorescence of donor molecule occurs. If the acceptor is fluorescent it may emit otherwise it will lose acquired energy as heat. Reson ance energy transfer does not require molecular contact as this happens through a space interaction and there is no direct interaction between the electron clouds in the molecules.

PAGE 22

8 Figure 2-2 Schematic of Frster resonance energy transfer (FRET) The distance dependence of quenching rate due to resonance energy transfer can be given by the following equation (2 2) where is the donor lifetime in the absence of acceptor, r is the center to center distance between donor and acceptor molecule, and R 0 is the Frster distance. 2.2.2. Collisional Q uenching and Static Q uenching For both collisional and static quenching, molecular contact between luminophore molecule and quencher molecule is required so that the electron clouds of b oth molecules can interact. There are at least three mechanisms for these quenching processes i.e. intersystem crossing or the heavy atom effect electron exchange or Dexter interactions and photoinduced electron transfer. Quenching can occur by any combination of these mechanisms. In case of intersystem crossing (Figure 2 4) due to encounter with some quencher molecules excited fluorophore molecules (F*) transfers to excited triplet state (FT*) from

PAGE 23

9 excited singlet state. As the emission from e xcited tr iplet states is usually delayed, the se molecules are likely to be quenched to the ground state by same quencher molecule or result in more loss of energy by nonradiative decay. Quenching by heavy halogen atoms and oxygen are the example s of this kind of quenching. Figure 2-3 Quenching by intersystem crossing In case of electron exchange quenching or Dexter interaction luminophore molecules act as donor molecules and transfer the electron to acceptor molecules. Electron transfer first occurs from excited donor mole in LU orbital to acceptor molecule. Then acceptor molecule transfers back the electron to donor molecule f rom HO orbital. Quenching by this process is similar as resonance energy transfer and also it depends on spectral overlap. However it is a short distance process (15 20 A) in contrast to resonance energy transfer.

PAGE 24

10 Figure 2-4 Dexter interaction Quenching due to photo induced electron transfer also results in electron exchange between donor molecule and acceptor molecule. But in this case a nonfluorescent complex is formed between donor and acceptor molecule and the luminophore molecule can be do nor or acceptor molecule. For quenching by any of above mechanisms, both luminophore molecule and quencher molecule need to be in contact as electron clouds are strongly localized and quenching requires molecular contact at the van der Walls radii. In t his case the distance dependence can be expressed as follow s (2 3) where r is the center to center distance between fluorophore and quencher molecule, and r are constants. The collisional fluorescence quenching follow s the of Stern Volmer (SV) equation given bellow.

PAGE 25

11 Q DC K I I 10 24 Where I0 QC is the luminescence intensity in the absence of quencher molecules, Q represents the quencher and is the concentration of quencher molecules. DK In case of static quenching the dependence of I is the Stern Volmer constant 0/I on quencher concentration [CQ ) 1 )( 1 (0 Q S Q DC K C K I I ] is also linear similar to dynamic quenching. So the linear dependence of intensity ratio to quencher concentration does not confirm type of quenching. In many cases both static and dynamic quenching occur together. In such case s the Stern Volmer plot wil l have an upward curvature. The f ollowing modified form of SV equation represents both static and dynamic quenching together 25 2.3. Metal E nhanced L uminescence Though the phenomena of metal enhanced luminesce n ce was known from the 1980s, the application and demonstration of metal enhanced luminescence is mostly new. Different applications of metal enhanced luminescence from diff erent metallic nanoparticles have been successfully demonstrated by the Lakowicz and the Geddes groups 1 3 9 17 C onducting metallic particles, colloids, or surfaces are known to significantly influence the emission of vicinal luminophores. The mechanism of metal enhanced fluorescence is still not fully understood. Geddes and coworkers suggested that m etal nanoparticles influence the luminescence by three known mechanisms 9. First t he

PAGE 26

12 presence of nanoparticles close to the luminophores can create new nonradiative channels due to light absorption inside the metal or Frster energy transfer thus increas ing the non radiative decay rate.9 Second metallic nanoparticles are expected to increase the local incident field at molecular location which enhance of the rate of excitation of luminophore molecules. The third mechanism is the increase of radiative decay rate of luminophore molecules in the presence of metal nanoparticles. Geddes and co workers recently suggested a unified plasmon fluorophore description for explaining the third mechanism According to this theory, non radiative energy transfer occurs from excited state of luminophore molecule to the surface plasmon resonance of vicinal metal nanostructures and luminophore induces mirror dipole in the metal. As a result surface plasmons radiate the p hotophysical properties of luminophore molecules which adds up with the radiative emission of luminophore molecule rate thus increas ing the overall radiative rate. This can be represented by following equation (2 6) 18 w here is the unmodified system radiative decay rate, is metal modified system r adiative decay rate and is the nonradiative decay rate. In case of metal enhanced luminescence the lifetime decreases as a result photobleaching effect also reduces. The metal modified lifetime can be expressed as following (2 7) Metallic platforms can enhance the radiative decay rate by coupling the emission of luminophores with surface plasmon resonance or scattering of nanoparticles So it can be inferred that the influence of metal nanoparticles on luminescence is strongly

PAGE 27

13 dependent on the surface plasmon resonance and the scattering efficiency of nanoparticles and the nanoparticles luminophore separation distance. Figure 2-5 Modified Jablonski diagram in the presence of metal. 9 2.3.1. Distance Dependence Figure 2-6 Distance dependence on the effect of metal on luminescence. 9 If the probe molecules are very near to nanoparticles, luminescence emission from the probe molecules directly gets a b sorbed onto the surface of metallic nanoparticles and is strongly quenched. Similarly if the probes are too far from the nanopar ticles platform effects of nanoparticles get diminished. Hence it is important to optimize the distance between the lum inophores and nanoparticles. It has been reported in the literature that for the fluorophores positioned less than 50 0 E Em m knr kmnr S0S1 A from the surface the lumines cence m Enhancement of radiative decay rate by coupling emission with SPR and scattering Enhancement of absorption by concentrating incident field Increasing Non radiative decay rate by Forster energy transferMetallic surface, island or colloids*

PAGE 28

14 intensity quenches with d3 dependence9. Recently some researc h work has been devoted to study of distancedependent metal enhanced luminescence13,1923. The investigation of dependence of the luminescence enhancement on luminophore metal separation distance has been done using various spacer design s Due to the extremely rough topology of metal surf ace, it is difficult to accurately control the distance. In some cases the luminophores are first dispersed in polymer binder then by coating the different thickness film of the polymer containing luminophores the average distance between luminophore and metal surface is varied21,22. Using this kind of spacer one can only meaningfully study the effect of average distance as the luminophore is distributed throughout the polymer so the distance is not precisely controlled. To overcome this limitation in r ecen t work luminophore molecules have been attached at a fixed distance using biological linker DNA as a spacer24. Alternating monolayers of biotinylated bovine serum albumin (BSA) and avidin is also used to investigate distance dependence 19. Core shell nano composites with metallic core and silica shell of various thickness have also been used for metal enhanced luminescence 23. Here the silica shell acts as a spacer. The distance is optimized by investigating metal core/ SiO2 spacer / luminophore syst em by varying shell thickness thus varying the distance. 2.3.2. Effect of S urface Plasmon Resonance of Metal Nanoparticles on Luminescence Plasmons are quantized and collective oscillation of electron gas density. When the plasmons are confined to the surface a nd interact with the incident light, then these are called surface plasmons. They usually occur at the metal and dielectric interface. Surface plasmon resonance (SPR) of nanoparticles is dependent on several properties of

PAGE 29

15 nanoparticles such as size, shape composition, conductivity and inter particles distance. The intensity of incident optical wave is enhanced in the near field of nanoparticles at the plasmon resonance wavelength. SPR of metallic nanoparticles play an important role in the luminescence e nhancement. There are few studies reported in the literature on the relationship between SPR of nanoparticles and luminescence enhancement. Tam et al.25 found that the enhancement is optimal when the nanoparticles plasmon resonance is tuned to the emission wavelength of the fluorophores. Recently, some theoretical and experimental studies have suggested that luminescence enhancement is highest when emission wavelength is red shifted from the plasmon resonance24,26 In all these cases, emissive enhancement o f luminophore is considered. It is still unknown what the effect of surface plasmon resonance of wavelength will be when the luminescence enhancement occurs due to absorption enhancement. Knowledge of the exact relationship between surface plasmon resonance and luminescence enhancement can lead us to design ing efficient nanoparticle luminophore assemblies with maximum luminescence. To obtain the information about the relation between surface plasmon resonance and luminescence enhancement it is important t o prepare nanoparticles with different surface plasmon resonance wavelengths. 2.3.3. Metal Nanoparticles Used for Metal Enhanced Luminescence and Their Synthesis Silver nanoparticles have been known to enhance luminescence2 5,7 16,2022,24,25,2738 due to their strong surface plasmon resonance. Metal enhanced luminescence has been studied for various silver nanostructures like silver colloids8, silver islands39, silver nanotriangles40, fractal like silvered surfaces41 and silver nanorods5. Silver

PAGE 30

16 nanostructures are reported to enhanc e the luminescence from six to 3000 fold42. Gold nanoparticles are known to both quench and enhance luminescence depending on the fluorophore particle separation distance, molecular dipole orientation with respect to particle surf ace, and size of the nanoparticles 22,29,43 Recently other metals such as copper17, aluminum44, nickel18 ,chromium45 and zinc46 have been reported to enhance luminescence 17,44. However, the enhancement effect of these metal nanostructures is not as pr onounced as for silver nanostructures due to higher ohmic losses. Z inc oxide (ZnO) nanorod platforms also have been reported to enhance luminescence intensity significantly from commonly utilized fluorophores in immunoassays 47 49. Zinc nanostructures e nhance the luminescence emission but do not influence the excited state lifetimes of luminophores like other metallic nanoparticles. This implies that the enhanced luminescence observed near zinc nanostructures is mostly due to electric field enhancement effect50. Silver, gold and copper nanoparticles are used for metal enhanced luminescence mainly in the visible region where aluminum, zinc and chromium nanostructured films are shown to enhance lumines cence of luminophore emitting in the ultraviolet and b lue region8,17,29,4446. Nickel nanoparticles can enhance the emission intensity of vicinal luminophores at broad wavelength range (500 800 nm)18Different techniques have been suggested in the literature for the synthesis of anisotropic metal structures for applications in metal enhanced luminescence The selection of luminophores which can be enhanced by metal nanoparticles is limited by the choice of meta ls due to the effect of surface plasmon resonance spectra of metals on metal enhanced luminescence. 3,5, 10,12,1417,2022,24,27,29,34,35,3841,51. Some researchers followed the simple wet chemical synthesis

PAGE 31

17 method for depositing spherical metal (mainly silver and gold) nanoparticles on glass slides. They prepare gold or silver colloids in suspension se parately then 3aminopropyl) trimethoxysilan e (APS) treated glass slides were immersed in the suspension to deposit colloidal nanoparticles on them29. Silver nanoparticles are also deposited on glass slides in a random fashion by using Tollens reaction21, 39. P hotodeposition technique has been used to prepare patterned silver nanostructures to facilitate its application to microfluidic devices10. Shang et al. reported a simple and fast electrochemical technique to deposit silver nanostructure on planar s ubstrates for luminescence enhancement application35. These silver nano structures have relatively homogeneous morphol ogy. Vapor deposition method has been also used for the deposition of both silver and gold nanostructures16,32,51. The morphology of vapor deposited nanostructures can easily be controlled by changing thickness and deposition rate. Vapor deposition method has recently been used for the preparation of copper nanostructures for its application to luminescence enhancem ent17Silver fractal like nanostructures were prepared by passing a current between silver electrodes in deionized water and these are found to show better enhancement than spherical nanoparticles 41. Similar to fractals, rods and triangles are also expected to show better enhancement5,40. Aslan et al. suggested simple wet chemical synthesis method for silver nanorod and triangular nanoplate deposition5,40. They suggested two methods for synthesis of nanorods5. In the first method, they deposited nanor ods by immersing APS treated glass slides in silver nanorods solution. In the second method, spherical silver seeds were first chemically attached to the planar substrate then the substrate was immersed into a solution containing a cationic surfactant and silver ions where the silver

PAGE 32

18 seeds were subsequently converted and grown into silver nanorods. They used the same method for the growth of silver triangular nanoplates40. But, by using these methods it is not possible to obtain well defined nicely arrayed structures of nanoparticles. For this sophisticated lithography techniques are necessary High resolution lithography techniques such as E beam lithography (EBL) have been used to produce highly regular cylindrical and triangular nanopatterns of gold for the application to luminescence enhancement of quantum dots22. Use of EBL allows tuning the surface plasmon resonance of nanoparticles over a wide range of wavelengths and may enable very strong enhancement. It can also help to localize the enhancem ent process with high spatial control, thus facilitating high emission intensity of luminescence. But the high cost and time involved limit applicability of the EBL technique. A relatively simpler and less expensive technique is nanosphere lithography de veloped by Van Duyne and co workers5254 by which triangular or hexagonal nanostructures can be deposited. 2.3.4. Metal Nanoparticles Quenched Luminescence Metallic nanoparticles can quench or enhance luminescence depending on the fluorophore particle separation distance, molecular dipole orientation with respect to particle surface, and size of the nanoparticles 22,29,43 The presence of nanoparticles close to the luminophores can create new nonradiative channels due to light absorption inside the metal quenching the emission of luminophores. 30 If the probe molecules are very close to the nanoparticles (typically less than 5 nm), luminescence emission is quenched due to F rster resonance energy transfer (FRET) from the excited state of the luminophore molecule (donor) to the surface plasmons of the metal nanoparticles

PAGE 33

19 (acceptor) The FRET efficiency depends on the spectral overlap of the acceptors absorption with the donors emission, and sensitivity depends o n the separation distance between acceptor and donor.55 Quenching effect due to Frster energy transfer decreases with the cube of separation distance.56Luminescence quenching by metal nanoparticles has been studied mostly using gold nanoparticles The relative orientation of luminophores molecular dipole moment with respect to metallic nanoparti cles surfa ce decides the influence of metallic nanoparticles on radiative rate The radiative rate is decreased for tangentially oriented dipole as the molecular dipole and the dipole induced on the metallic nanoparticles radiate out of phase. On the oth er hand, radiative rate is increased if the molecular dipole is oriented radially towards metallic nanoparticles. 43,55,5759 Dulkeith et al.55 studied the que nching of the fluorescence of lissamine dye molecules attached to several sizes of gold nanoparticles. They investigated the effect of gold nanoparticles on both radiative and nonradiative decay rates responsible for quenching using time resolved fluoresc ence experiments. Horimoto et al.58 studied the effect of shape of gold nanoparticles on luminescence quenching and Ghosh et al.59 studied the size dependence of luminescence quenching. 2.4. Theoretical Modeling In the following sections, the basic concepts o f theoretical approaches for the study of metal enhanced luminescence are presented. The effect of surface plasmon resonance of metal nanoparticles on metal enhanced luminescence is also studied theoretically in this work. Firstly, the calculation of the surface plasmon resonance spectra of alloy nanospheres is discussed, and then the calculation of quantum efficiency

PAGE 34

20 modification of luminophore molecule in the presence of metal nanosphere is discussed in detail. 2.4.1. Theoretical Investigation of Surface Plasmon Resonance of Nanoparticles The surface plasmon resonance spectra of metal particles have been studied for many years60 65. Mie was the first to suggest a theory to study absorption spectra for spherical particles by solving Maxwells equation His theory is based on classical electrodynamics and by 65Mies theory is valid for any s ize particles but is limited to system where inter particle separation distance is much larger than the wavelength of incident light. According to Mies theory, the to tal transmittance through films containing spherical metal particles is ) exp(2d Q a N Text tot (2 8) where N is number concentration of spheres per unit volume, a is sphere radius and d is film thickness. The extinction coefficient is given as 1 2) Re( ) 1 2 ( 2n n n extb a n x Q (2 9) where 02 an x (2 10) n0 is refractive index of the host medium and wavelength of the incident light in vacuum. an and bn are Mie scattering coefficients.

PAGE 35

21 ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( x y m x y x y m x y an n n n n n n n n (2 11) ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( x y x y m x y x y m bn n n n n n n n n (2 12) 0~ n n mm (2 13) m m mik n n ~ (2 14) nm is real refractive index of metal and km mn ay ~ 2 is absorption coefficient (2 15) ) ( 2 ) (2 / 1 2 / 1z J z zn n (2 16) ) ( 2 ) (2 / 1 ) 2 ( 2 / 1z H z zn n (2 17) w here Jn is the Bessel function and Hn 2For calculating absorption spectra for spherical particles using Mies theory, one needs to know the ef fective refractive index or die lectric constant for the system. The complex dielectric constant of metals can be calculated using Drude theory and Lorentz theory is second order Hankel function and Z is equal to x or y. 66 According to D rudes theory, the complex dielectric constants of a metal should be calculated using the following formula: 2 2 2 2 2 2 21 ) ( 1 1 ) ( p f p f (2 18)

PAGE 36

22 where is the bulk relaxation time of electrons, and is the frequency of light hitting the materials. p is plasma frequency of metal which can be calculated using the following formula: m nep 0 2/ (2 19) where n is electron density, e is the charge of electron and 0 is the vacuum permittivity and m is the mass of electrons. For optical frequency = 2 c/ is very high so ( )2 3 2 3 2 2 2) ( 1 ) ( p p f p f >> 1 under this approximation, we can write (2 20) But for metal s, the Drude model alone is in sufficient to predict dielectric constants as it implies that only plasma frequency dictates the dielectric constant. Though this works for some metals such a s Zn, for most of the metals such as Ag an d Cu, plasma frequency cannot by itself account for the dielectric constant. For these metals, the combined effects of the freeelectrons (Drude model) and the bound d electrons (Lorentz model) influence the refle ctance properties of the metal. So, for these metals, the dielectric constant can be calculated by the formula b f r (2 21) Where f is described by the Drude model ( 0 = 0) (equation 1), and b is described by the Lorentz model. ( 0 = [EF Ed 2 2 2 2 2 0 2 2 0 2/ ) ( ) ( 4 1 ) ( m neb ]/ .) (2 22)

PAGE 37

23 2 2 2 2 2 0 2/ ) ( / 4 ) ( m neb (2 23) There are many studies devoted to calculating effective dielectric constants for composite materials60,62,63,6772. Maxwell Garnett (M G) and Bergman theory are mostly used to calculate effective dielectric constant for metal d ielectric composite60,62,63,6772. These theories are valid for only spherical or ellipsoidal metal nanoparticles in dielectric media. These theories are developed considering the interaction of the external electric field with metal particles acting as interacting dipoles, with an effective polarizability given by the Drude relation, while the dielectric constant of the composite material was obtained through the Clausius Mossotti relation60,72. M G theory is based on the assumptions that the percentage of metal (fa) in dielectric media is very small and interparticle separation is very small compared to the wavelength of light. According to M G theory, effective dielectric constant of metal dielectric composite is given by h a h a a h eff h effk f k 72 (2 24) ef f is effective dielectric constant of composite h is dielectric constant of host matrix mFor a random mixture of two dissimilar materials, the effective dielectric constant can be calculated using Bergmans theory is d ielectric constant of metal k is screening parameter determined by the shape as well as the orientation of the nanoparticles with respect to the external e lectric field. 60eff. According to this theory, can be calculated using the following equation

PAGE 38

24 0 ) 1 ( eff h eff h a eff a eff a ak f k f (2 25) Gao et al. incorporated shape distribution of the components in both M G and Bergman theory67. Garcia et al. developed a self consistent technique based on mixing rules to predict the effective dielectric constants, and thus SPR spectra, for multi component mixtures68. They presented a model to correct the imaginary component of dielectric component of metal to account for the enhanced rate of electron scattering due to size dependent effect for nanoparticles (2 26) 68 where, is imaginary component of dielectric constant of bulk metal, d is diameter of the nanoparticle, is bulk relaxation time of the electron and is the speed of the electrons close to the Fermi surface. The above mentioned theories are only capable of predicting SPR spectra for spherical particles. With the development of computational resources there are some studies devoted to studying the problem of determining the scattering properties of particles of arbitrary shape and composition64,70,7385. There are two approaches most used for calculating spectra for arbitrary shaped particles. The first approach is the discrete dipole approximations (DDA) method 8185. In this method, the particl e is assumed to be composed of an equi valent volume filled by a lattice with a cubic cell whose sites are oc cupied by elementary scatterers electric dipoles. The number of dipoles considered decides the size of problem. Draine et al. have developed a FORTRAN program based on DDA approach to ca lculate scattering and absorption spectra for arbitrary shaped particles81. Another approach is approximation of N spheres where the random shaped

PAGE 39

25 parti cle is assumed to be composed of number of elementary spheres78,8688 The problem becomes more computational ly intensive with the increase in the size of spheres. Both of these methods provide a good approximation about the SPR spectra for arbitrary shape particles in the region of forward scattering. 2.4.2. Modeling of Plasmon Enhanced Luminescence The intensity of the luminophore at the proximity of metal nanoparticles can be written as 26,30 T I K C Iabs exc abs abs abs flu) ( ) ( ) ( ) (2 : ( 227) Here, abs is absorption frequency of the molecule flu ) (flu is e mission frequency of the molecule is quantum yield of emission, abs ( abs ) is absorption cross section of the molecule in vacuum ) (absexcI is excit ing intensity in vacuum, C is a constant, T is integration time of the detector, and absK is local field vector. From the above expression, it can be seen that by changing the local field for absorption absk and/or quantum yield ) (flu we can change the intensity of luminescence. The absorption rate of the luminophore can be enhanced by increasing both the absorption coefficient of the luminophore itself and the local field intensity. On the other hand, the quantum yield of the luminophore can be influenced by varying the radiative and nonradiative decay rates Kmmerlen et al. 33 suggested that the quantum efficiency enhancement factor Y (ratio of quantum efficiencies in the presence of metal nanoparticles an d without nanoparticles) can be calculated using the following equation:

PAGE 40

26 Y L (abs) 2Z (flu) (2 28) The first term represents the enhancement of local electric field at the excitation frequency (abs) The second term describes the change in quantum efficiency due to radiative and nonradiative decay rate enhancements at the emission frequency flu In the following sections calculation of both excitation enhancement factor and quantum efficiency enhancement factor are discussed. 2.4.2.1 Calculation of Excitation Enhancement Fac tor The integrated near field scattering cross section ( Qnf 2) (absL ) at the excitation wavelength divided by the surface area of the spherical particle is a good measure of average 89. The near field scattering cross section can be calculated using the following equation 1 2 ) 1 ( 2 2 ) 1 ( 1 2 ) 1 ( 1 2 2 2) ( ) 1 2 ( ) ( ) ( 1 2n n n n n n nfka h b n ka h n ka h n a a r Q 90 ( 229 ) w here r is the distance from the center of the spherical nanoparticle and a is the radius of the nanoparticle. c km/ is the optical frequency (radian per second), m is the dielectric constant of the media and c is the velocity of light in vacuum. The term hn (1) is the spherical Henkel function of the first kind. an and bn are well known scattering coefficients.

PAGE 41

27 2.4.3. Model ing of Effect of Metal Sphere on Excited State Decay Rate Quantum efficiency is calculated as the ratio of radiative decay rate to total decay rate. Spontaneous emission can be modified by resonant coupling with electromagnetic environment.91 Both the model based on exact electrodynami cal theory 92 94 and the Gersten Nitzan ( GN) model 93,95,96 can be used to provide insight into the influence of metal nanospheres on radiative and nonradiative decay rates of luminophore molecules at their close proximity, thus can be used to calculate l uminescence quantum efficiency modification of a luminophore molecule in the presence of a noble metal nanosphere. In both of these model s, and the luminophore molecule is modeled as a classical dipole with a dipole moment. Using these models, excited sta te decay rate for a dipole located outside the metallic sphere can be obtained for both radial and tangential orientation of dipoles with respect to metallic surface. In the following s ection the exact electrodynamic theory developed by Ruppins and by Kim et al. is discussed. After that the Gersten and Nitzan improved by Mertens et al.93,95,96 is described. 2.4.3.1 Exact Electrodynamic T heory The radiative and non radiative decay rate of an excited luminophore molecule in the proximity of metallic nanosphere is modeled using classical electromagnetic theory.92,94 The radiative decay rate is calculated considering the energy flow (Poynting vector) at large distances and nonradiative decay rate is obtained directly from ohmic losses inside the metallic sphere. In the presence of the metal sphere the total decay rate of emitter molecule in absorbing dielectric can be obtained by comparing the work done on a source in the presence of the sphere to the work done on the same source in the bulk

PAGE 42

28 dielectric. The radiative decay rate in the presence of metallic sphere can be derived by comparing the energy flux through a surface enclosing both source dipole and sphere to the radiated power of source dipole in the bulk dielectric. The nonradiative decay rate is the differenc e between total decay rate and radiative decay rate.92,94 The expressions for radiative decay rate ( and total decay rate ( for a luminophore molecule derived from exact electrodynamics are given below. 93,94 (2 30) These equations were developed considering luminophore molecule as dipole with dipole moment placed at the distance d from metal nanosphere with radius a and dielectric constant For radial orientation of dipole with respect to metallic sphere surface, the exp ressions are (2 31) For tangential orientation of dipole with respect to metallic sphere surface, the expressions are (2 32) (2 33) where is the radiative decay rate for the dipole located in the nonabsorb ing embedding medium in the absence of sphere, j l and hl are the ordinary spherical Bessel and Henkel functions, an and bn are the Mie scattering coefficients of the sphere, r=a+d, , is the dielectric constant of embedding medium, is the optical frequency (rad/sec), c is the speed of light in

PAGE 43

29 vacuum and l is the angular mode number, The derivatives of and are derivatives to kr. In above expressions and are total decay rate of emitter with 100% quantum efficiency in absence of sphere. 2.4.3.2 Gersten Nitzan ( GN) M odel Using the model95 R the modifications of the radiative decay rate ( ) and total decay rate (tot ) of luminophore in proximity to metal nanoparticles can be calculated According to this model excited state decay rate is calculated in two steps. First the quasistatic approximation is used to analyze the electromagn etic interaction between source dipole and metal sphere. The analysis is done based on electrostatics, as the retardation effect is neglected assuming the sizes of nanoparticles to be much smaller than the wavelength Electrostatic potential is derived from the superposition of the source dipole potential and the induced multipoles of sph e re. In the second step, radiative power is calculated from the effective dipole moment comprised of a vectorial superposition of the source dipole moment and the induced dipole moment. Radiative rate modification is obtained by normalizing to the power r adiated by an uncoupled source with identical dipole moment. The nonradiative decay rate is calculated by calculating the power dissipated in the metal sphere by the Joule heating law. This model does not consider multipole radiation and the interference between source dipole and induced dipole is neglected. The key advantage of model over exact e lectrodynamical theory is that model can be generalized to spheroidally shaped particle s Mertens et al.93,96 introduced a correction factor for radiative reaction and dynamic depolarization in the GN model to modify the quasistatic polarizability of the nanoparticles to account the

PAGE 44

30 retardation effect. This corrected model can accurately describe decay rate enhancement near larger nanoparticles (several 100 nanometers). In this model the luminophore molecule is modeled as a classical dipole with dipole moment For the radial dipole orientation, the expressions for R and tot for the luminophore molecule positioned at distance d from the surface of sphere wit h radius a and dielectric constant i located in the medium of dielectric constant m is as follow s l l m m n ref R totd a a l l C l l ka4 2 31 Im ) 1 ( 4 3 1 96 ; ( 234) 2 3 12 2 1 d a a Cm m ref R R ( 235) For the tangential dipole orientation, the expressions for R and totare l l m m l ref R totd a a l l C l ka4 2 2 3 //1 Im ) 1 ( 2 3 1 ; (2 36) 2 3 1 //2 1 d a a Cm m ref R R (2 37) In the above expression s, l is the angular mode number, and ref R is the radiative decay rate of luminophore in the absence of nanoparticles. 1C is the correction factor for radiation dumping and dynamic depolarization : a k ik C 4 6 1 12 3 1 ( 238)

PAGE 45

31 is the quasistatic polarizability m ma 2 43 ( 239) For 1 l ClIn the present work a quantum efficiency enhancement factor is calculated using the corrected GN model as suggested by Mertens et al. is assumed to be 1. 93 For better representation of experimental condition s, the source dipole orientation was averaged over all solid angles This was achieved by averaging the results for decay rates obtained for radial and tangential orientations. 2.5. Bimetallic N anoparticles Bimetallic nanoparticles constituting various combinat ions of noble metals have been attracting much attention as they can combine the advantages of two pure metals. They offer many unique properties and advantages over pure nanoparticles, for example, enhanced maetism97, electrochemical properties98, cataly tic activity99 and fine tuning of optical properties100,101 In this study the unique plasmonic property of alloy nanoparticles is of main interest. In the following sections a brief overview of plasmonic property of different bimetallic nanoparticles a nd their synthesis methods are given. 2.5.1. Plasmonic P roperties Plasmonic properties of nanoparticles are si gni ficantly influenced by dielectric constant, shape, size and structure of nanoparticles. Tunable surface plasmon resonance in wide range is the most interesting property of bimetallic nanoparticles. Bimetallic

PAGE 46

32 nanoparticles can be core shell, random and separated structures depending on their synthesis method D ielectric constant can be chan ged by alloying or mixing two metals with different dielectric constant. Mie scattering theory predicts that surface plasmon resonance of coreshell nanoparticles can be shifted between ultraviolet to mid infrared range. For Ag Cu alloy nanoparticles sur face plasmon resonance can be shifted from near infrared to ultraviolet region by changing only one experimental condition annealin g temperature. This shifting is due to reorientation of Ag and Cu atoms in AgCu nanoparticles.101,102Theoretical modeling of SPR spectra of alloy nanoparticles requires knowledge of their dielectric constants. For Ag Pt hollow nanoparticles, SPR can be redshifted by increasing Pt concentration and once the Pt. concentration exceed a maximum value the peak broadens and is blue shifted and eventually diminished. For Ag Au alloy nanoparticles SPR can be shifted by changing the compo sition. 61,103,104 Dielectric constants for alloy nanoparticles of different compositions are not available and have to be calculated using semi empirical models such as those based on Drude theory and experimental data for pure, bulk metals.105 (2 40) In most of this existing work, semi empirical models are developed based on the assumption of homogeneous distribution of metallic atoms in their alloys. For the core shell structure the dielectric constant is given as follows: (2 41) w here and are the surfaceinduced contributions to the damping.

PAGE 47

33 For alloy nanoparticles the dielectric constant is obtained by modeling the nanoparticle as homogeneous material with physical properties obtained by averaging those of pure metals. Plasmonic freq uency at bimetallic surface is given by (2 42) where is the classical plasma frequency and represents a different plasma frequency based off the dielectric constant at the interface. 2.5.2. Synthesis Bimetallic nanoparticles have been synthesized as alloys or core shell structure s using different synthesis methods like solution synthesis and physical deposition technique s In mos t cases alloy nanoparticles are synthesized in solution phase S imultaneous reduction of corresponding metal ions or metal complexes results in the formation of alloy nanoparticles. Co reduction of two metal ions also result s in bimetallic nanoparticles. Bimetallic nanoparticles can also be prepared by laser radiation or heat treatment of mixtures of monometallic nanoparticl es In all cases the morphology and the size of bimetallic nanoparticles can be controlled by controlling experimental parameters l ike temperature, ratio of precursors and stabilizing agents. Bimetallic nanoparticles synthesized by different methods will have different plasmonic characteresitics as the atomic distribution in bimetallic nanoparticles is different for different synthesis method.

PAGE 48

34 2.5.2.1 Synthesis of Bimetallic Alloy N anoparticles Cored uction is one of the important methods used for synthesizing alloy nanoparticles. Bimetallic colloids are prepared by chemical reduction, photochemical reduction or thermal decomposition. Au Ag alloy nanoparticles were produced by the coreduction of Ag sa lt ( AO3) and gold salt (HAuCl4) by reducing agent like sodium citrate. For these Au Ag nanopartic les, the SPR peak blue shift ed by increasing percentage of silver in alloy nanoparticles. This resonance shift is suggested to be due to a modification in t he band structure of these alloys, which is different from pure metal.106 Various composition Ag/Au alloy nanoparticles we re synthesized in microimulsion by the co reduction of HAuCl4 and A gN O3 with hydrazine .107 Au Cu colloidal nanoparticles we re synthesized in methanol by coreducing HAuCl4 and CuCl2 by NaBH4 and the polymer poly(N vinyl 2pyrrolidone) (PVP) is used as stabilizing agent.108 Au Cu nanoparticles were also prepared in reverse micelles by coreduction of their salts.109 Silver copp er alloy nanoparticles were synthesized via the polyol process by coreducing Ao3 and Cu(HCOO3)2,H2O110Several other interesting methods are suggested in the literature for synthesizing bimetallic alloy nanoparticles. Smetana et al. suggested low temperature digestive ripening procedure for synthesizing AgAu and Au Cu nanoparticles 111. In this method bimetallic alloy nanoparticles are synthesized by heating colloids of two different pure metal nanoparticles in the presence of alkan ethiol under reflux. Haverkamp et al. suggested a biosynthetic method using plant Brassica junc ea for sy n thesizing Ag Au and Au Cu alloy nanoparticles.112 Bimetallic nanoparticles of Ag, Cu and Au are prepared by

PAGE 49

35 photo chemical reduction of their salts u sing in ethanol by UV irradiation using benzoin.113 Bimetallic nanoparticles of Co and Cu were prepared by successive reduction of their salts in hydrazine solution with the aid of sonication.114Physical vapor deposition is also frequently used for the synthesis of bimetallic nanoparticles. Simultaneous sputter deposition of Ag and Au in ionic liquids were used o synthesize Au Ag nanoparticles in solution. 115 Co s puttering deposition was also used to deposit bimetallic nanoparticles like Ag Cu and a Ag Au on solid substrate .116,117 Pulsed laser deposition was used to synthesize bimetallic Ag Cu nanoparticles on glass substrate102 2.6. Characterization T echniques In the present work first type of characterization techniques are used to characterize the nanoparticles like imaging, composition analysis and their optical property measurement. The second type of characterization techniques are used to study the fluorescence property of luminophores. Characterization tools used in this work are briefly described below. 2.6.1. Transmission Electron M icroscopy (TEM) Transmission electron microscopy is the most useful imaging techniques for nanoparticles (specifically for less than 10 nm size). In case of t ransmission electron micro scopy ( TEM ) a beam of electrons is transmitted through a electronically transparent specimen interacting with the atoms to produce one image. Due to the small de Broglie

PAGE 50

36 wavelength of electron image with atomic resolution is possible to be captured by TEM. For TEM image sample is required to be dispersed on TEM grids (for example carbon coated copper grid, molybdenum grid). 2.6.2. Scanning Electron M icroscopy (SEM) In case of SEM the area of the sample to be analyzed is targeted by a narrowly focused electron beam which can be swept across the surface of specimen to form image or may target one place only to analyze particular position. The image is produced due to the interaction of the electron beam with atoms at or near the surface of the samples. SEM can also produce very high resolution image (1 to 5 nm). SEM specimens required to be conductive at the surface to avoid accumulation of electrostatic charge at the surface. For imaging non conductive specimens, the specimen surface is coated with a thin film of conducting metal like gold. 2.6.3. Atomic Force M icroscopy (AFM) AFM is a high resolution scanning probe microscopy technique in which a microcantilever with a sharp tip i s used to scan the surface of sample. The advantage of AFM over SEM is that AFM can provide true three dimensional image of a sample and do es not require sample to be conductive and can operate in ambient air or even in liquid. However AFM can only provid e image of area an order of 10 micrometers.

PAGE 51

37 2.6.4. Energy Dispersive X ray S pectroscopy ( ED S) Chemical characterization and elemental analysis of nanometer scale particles can be done by EDS. This analysis is based on the analysis of xrays emitted by the matte r in response to interaction between electromagnetic radiation and matter. As each element has unique atomic structure and can emit unique x rays, elemental composition can be detected by analyzing the emitted x rays. EDS for compositional characterizati on of nanoparticles is usually integrated with TEM or SEM. 2.6.5. UVVis Absorption S pectro scopy In this technique a beam of light of wavelengths in the visible and ultraviolet region pas ses through the specimen and its intensity before and after interacting of sample is measured to determine the light transmitted through or absorbed by the sample. Absorption peaks can be correlated to the surface plasmon resonance peak of nanop articles and can be indicative o f the type of bonds in a given molecule. 2.6.6. Fluorescence M icroscopy In this microscopy method images are taken based on the fluorescence property of samples. The sample is usually first tagged with a fluoresc ent molecule and excited by light with excitation energy requ ired for the fluorophores. The fluorescence emission from the specimen is collected through an emission filter to separate the emitted light from the illumination light. A single fluorophore can be imaged at a time.

PAGE 52

38 2.6.7. Fluorescence S pectroscopy In this ty pe of fluorescence electromagnetic spectroscopy fluorescence from sample is analyzed. The sample is excited using a particular wavelength of light and emitted fluorescence emission of a lower energy is detected.

PAGE 53

39 Chapter 3 Measurement of Oxygen Diffusivity and Permeability in Polymers Using Fluorescence Microscopy 3.1. I ntroduction The emission intensity of some luminophores is quenched in the presence of oxygen molecules Applications of luminescence quenching by oxygen range from the measurement of pressure distribution of air on the wing of an aircraft using pressure sensitive paint118 to the study of oxygen diffusion properties in polymers119,120,121 and biological membrane122. F or measuring diffusion coefficients of oxygen in polymers using luminescence quenching methods, t he luminophore is typically dispersed directly in the polymer and the change in the average oxygen co ncentration i s monitored by studying the avera ge intensity change or life time change of the luminophore using a spectrofluorometer4, 119,120,121. In these methods, i nitially, the polymer is equilibrated at a particular concentration of oxygen. Then, the polymer containing luminophore is exposed to h igher (diffusion in) or lower (diffusion out) concentrations of oxygen. The avera ge intensity or lifetime change in the polymer is monitored using a spectrofluorometer for determining diffusion coefficients. In mos t cases 120,123,126,141 for ease of calibration, the luminophore dispersed in the polymer is assumed to behave ideally and follow the linear Stern Volmer equation 124

PAGE 54

40 31 where I0 is the luminescence intensity in the absence of oxygen, Ksv is the Yekta et al S tern V olmer (SV) constant and is the partial pressure of oxygen over the polymer. 4 were the first to develop an appropriate model combining Ficks law of diffusion and the linear Stern Volmer (SV) equation to extract diffusion coefficients from experimental data for both diffusion in and diffusion out experiments. This model was based on the concept that the intensity change of the luminophore corresponds to the average oxygen concentration within the polymer. Additionally, this model is based on the assumption of uniform excitation of luminophore throughout the film which is only true for low optical density films. This model was extensively used later to find diffus ion coefficients for different luminophore containing polymer films which follow the linear SV equation120,123,126,141, 4However, typically, it is difficult to fabricate polymer supported luminescence oxygen sensors that exhibit linear response. Luminophore molecule s in liquid solvents almost always observe the linearity of SV plot as the temporal fluctuation s of the microenvironment are much faster than the luminescence decay rate. As a result all the luminophore molecules are expected t o be in the same e nvironment on average. In case of luminophore molecules dispersed in a polymer matrix, different luminescent molecules experience different influences from their respective microenvironment s due to micron scale irregularities in polymer morphology Heterogeneity of luminescence sensors, which occurs due to incompatibility of the polymer and luminophor e, is typically the reason for nonlinearity in response. Influence from unquenchable emission from aggregates of luminophore and also sometimes the unq uenchable background emission 21 /0 O svp K I I 2Op

PAGE 55

41 may contribute to nonlinear behavior. Influence of dual or continuous gas sorption in the polymer may also result in nonlinearity130135. The exact reasons for the non linearity are debatable. Several models were developed for describing the nonlinear response of the sensors such as the multi site quenching model ( two site model )5 the nonlinear solubility model134 and a model based on Forster type energy transfer .130,131For determining the diffusion coefficient it is quite complicated and computationally demanding to combine the nonlinear SV models with Ficks law. Kneas et al 138 suggested an improved computational scheme in which they combined a non linear gas solubility model for SV equation with Fick s la w based diffusion model and solved it numerically to interpret the data of oxygen diffusion in polymer s This model can be applied to high optical density film s, and the assumption of uniform concentration of O2 is not required. However, this model can o nly be applied to cases where the luminophore is uniformly distributed throughout the film. Schappacher and Hartmann 137 were first to develop a partial analytical model to eliminate numerical complexity. They combined the two component model (two sites model) which is mathematically equivalent to the dual sorption model for nonlinear quenching with Ficks law base d diffusion model Unfortunately, the two site model is not always sufficient to explain nonlinear behavior of real sensors and consideration of existence of dye molecules in more than two sites (with their own quenching constants) is necessary .134Some re searchers Analytical models combining a multi site model with Ficks law subjected to different sets of boundary conditions are complicated to derive. 138 140 used fluorescence microscopy to study the heterogeneity in luminescence sensors (luminophore molecules dispersed in a polymer matrix).

PAGE 56

42 Fluorescence microscopy allowed them to study the SV response of the luminescence sensors with microsco pic spatial resolution. They reported homogeneous regions of sensors show better response to oxygen concentration than regions where the dye is aggregated. In the present work, we used conventional fluorescence microscopy to study the heterogeneities of luminescence sensors and their spatial response to O2 concentration and extend its application for the measurement of oxygen diffusion properties of polymers. We investigated spatial distribution of SV response of the sensor at different oxygen concentrat ions at the microscopic level. Fluorescence microscopy allowed us to identify relatively homogeneous regions. The responses from these regions were analyzed to calculate the oxygen diffusion coefficients. This method avoids the complexity of including n onlinear SV equations in the analytical models. This method also eliminates the need f or generati ng calibration curves for non linear SV responses for each and every sensor before using it for diffusion measurement. In the present study, we used the film on sensor method and the accumulation in volume techniques141 We first chose P to investigate oxygen diffusion behavior in a variety of polymers, including transparent and opaque films and those containing additives. latinum octaethylporphyrin (PtOEP) as a probe for luminescent sensor. However, this type of the sensor was not very photostable under continuous illumination of fluoresc ence microscope. PtOEP showed decrease of intensity during the initial illumination period. This is attributed to the pho tobleaching and leaching of PtOEP from polymer matrix and deterioration of matrix itself. Hence, we replaced PtOEP with more photostable luminophore platinum(II) meso tetrakis

PAGE 57

43 (pentafluorophenyl)porphyrin (PtTFPP). The difference in photostability betwe en PtOEP and PtTFPP is mainly attributed to the differences between their side functional groups, ethyl for PtOEP and perfluorophenyl for PtTFPP. The photostability of sensors mainly depends on the size and rigidity of the side functional group and result ing efficiencies of collision with oxygen molecules. While, ethyl groups can easily move, the fluorophenyl group is large and rigid to oxidative/reductive attack. As a result, PtTFPP molecules are less reactive toward photo oxidation/reduction.142 PtTFPP also has high emission quantum efficiency and a moderately long emission lifetime which is required for application in luminescence sensing. We used polystyrene as polymer matrix for the se nsor.142 Figure 3-1 Molecular Structure of PtOEP and PtTFPP 142

PAGE 58

44 Oxygen diffusion and permeation coefficients in Teflon and Polydimethylsiloxane ( PDMS ) were measured to validate our new technique. Then, the technique was used to measure the diffusion coefficient of a high performance ( HP ) silicone elastomer (black polymer ) and PDMS containing different weight percentages of zeolite (Molecular sieves 5 It should be noted that, in this case the polymer films for which oxygen diffusion properties are measured are different from the polymer used to prepare the oxygen sensor s We combined the SV equation with Ficks law of diffusion to extract the diffusion coefficient s from experimental data. 3.2. M aterials and M ethods 3.2.1. Sensor Films and P olymers The o xygen sensing material was prepared by dispersing 1.3283x 104 mols of luminophore platinum tetrakis (pentafluorophenyl) porphyrin (PtTFPP) (Frontier Scientific, Inc., Logan UT) in 1 liter solution of polystyrene (Sigma Aldrich; Milwaukee, WI,USA, Avg. Mw 280,000 by GPC)/toluene (0.24 gl1). This solution was spin coated on 19 mm diameter glass, cut from 1 mm thick microscope slides. Before coating, the glass slides were cleaned with acetone, methanol, isopropanol and deionized water, then dried with n itrogen gas. Then, the glass pieces were put in an air plasma cleaner (Harrick PDC 32G) for 15 minutes at 6.8 watts power setting. For coating, the spin speed was maintained at 1,000 rpm for 60 seconds for each sensor. Lastly, the sensor pieces were cur ed at room temperature for 1 hour and at 120 C for 5 hours. An ellipsometer

PAGE 59

45 ( Rudolph AutoELIII) was used to measure the thickness of the resulting sensor (polystyre ne containing PtTFPP dye) film Figure 3-2 Schematic diagram s of dif fusion cell s for (a) film on sensor e xperiment and (b) accumulation in volume experiment. The oxygen diffusion and permeation coefficients were measured for different polymers with known and unknown diffusion properties to establish this technique. The permeation and diffusion coefficients of DuPonts Teflon FEP film were measured using accumu lation in volume technique. The thickness of Teflon film was 25 The diffusion coefficients of Sylgard 184, a common poly(dimethylsiloxane) (PDMS) and 36265 HP polymer (silicone elastomer) (Dow Corning) were measured by film on sensor technique. Fo r PDMS, the pre polym er mixture was first degassed under vacuum (30 in. Hg vacuum) for 30 minutes to remove any air bubbles in the mixture after which, it was Sensor Gasket Polymer film Sensor Polymer film a b

PAGE 60

46 directly drop cast on a 19 mm luminescence sensor and was pressed against a PET film to smooth t he surface. After this the PDMS film was cured at 120 C for 1 hour. The 3 6265 HP polymer film was also prepared by drop casting and was cured at 1000To disperse zeol ite (Ca C for 35 minutes. For the HP polymer, a T eflon film was used to smooth the surface. /nNa12-2n[(AlO2)12(SiO2)12] xH2 O, molecular sieves, 5 8 mesh, Sigma Aldrich, Milwaukee, WI) into PDMS, the zeolite was first ball milled in SPEX 8000 Mixer/Mill with 8 mm stainless steel balls for 2 hours. After milling, the average these were dried at 150 C for 1 hour before preparation of the film. Different weight percentages of zeolite (up to 30%) were dispersed in PDMS solution. These solutions were cast on the sensors, and cured for 1 hour at 120 C. 3.2.2. Instrumentation and S oftware The Leica DMI 4000b inverted research fluorescence microscope equipped with Leica DFC340 FX CCD Camera was used in this study. Fluorescence microscopy was carried out with a red fi lter set (Chroma Technology 41005, HQ535/50x exciter and HQ645/75m emitter). The images of luminescence sensors were taken using a 10X objective (Leica 11506228 HI Plan 10x/0.25 NA, 12.0 mm W.D) and the light source used was a tungsten halogen lamp (100W and 12V) Image Pro plus version 6 with Scope Pro version 6 (Media Cybernetics, Inc) was used for acquiring and analyzing images. Using Scope pro, the illumination intensity of light source can be controlled from 0% to 100% of the total intensity and a lso the shutter can be controlled. The specimen was only exposed to illumination while taking images and the lamp intensity

PAGE 61

47 was maintained at 10% of its total intensity. A macro was written to only have the shutter open while acquiring an image. The exp osure time for acquiring each image was set to be 300 milliseconds for each image. Thus, photobleaching of the luminescence sensors was minimized. For observing the SV behavior of a sensor, it was placed in a chamber that was flushed with different concentrations of oxygen. A circular glass dis c with sensor film coated on it was mounted on a stainless steel chamber using a viton gasket (the sensor f ilm was on the inside surface of glass disc) to make the chamber airtight. An air pump w as used to control the air pressure inside the chamber. The pressure inside the chamber was monitored using a MKS Baratron pressure tran sducer (315 BA 1000) with 1000 T orr range with a digital read out. Assuming the concentration of O2 in air is 21%, the partial pressure of O2Stainless steel cells (Figures 3.2) were constructed for diffusion measurements of the polymers such that the volume of the downstream chamber was and the exposed surface area of the polymer film was m inside the chamber was determined from the total pressure indicated by the pressure transducer. Images of the sensor were acquired at differen t concentrations of oxygen and analyzed for intensity using Image Proplus. 2For the accumulation in volume experiment, there were two chambers in the diffusion cells (Figure 3. 2 (a) ). The polymer films were placed between the chambers using viton gaskets to prevent leakage. The upstream chamber of the polymer film was continuously flushed with nitrogen (diffusion out experiment). In the downstream chamber, the luminescence sensor was mounted with the viton gasket at the opp osite side The diffusion cell was painted flat black to prevent reflection of light. 3 710 68 2 m 510 85 5

PAGE 62

48 of the polymer film. In this experimental configuration, the concentration of oxygen inside the volume of the downstream chamber changes with commensurate change in the luminescence intensity of the sensor. For the film on sensor experiment, t he polymer film attached to a sensor was placed in a stainless steel chamber (Figure 3. 2(b) ). In all cases, the polymer film was first equilibrated with air; then exposed to a zero concentration of oxygen. Luminescence intensity changes measured the changes in oxygen concentration at the sensor/polymer film boundary in the film on sensor technique. To test the cells for leakage, the same experiment was done with the cell by replacing polymer film with a stainless steel plate. The diffusion cell was mount ed on the stage of the microscope in such a way that the sensor faced the illumination source and detector. The image of the sensor film was captured through the 1 mm thick glass on which the sensor film was coated. To ensure that the sensor film was w ithin the focal length of the objective, a new insert for the microscope stage was design ed. The insert has a 2 mm deep recession to lower the sample placed on it towards the objective. 3.2.3. Image A nalysis The images were processed in the following way to minimize the er ror in intensity measurement. The intensity of the dark current image, acquired while the camera shutter was closed, was subtracted from the intensity of every raw image to correct the images for the dark current noise of the camera. The a mbient lighting image of polystyrene film, which was acquired at the same ambient light at which raw images were taken, was subtracted from the images. The mean intensities of different regions of the corrected images were measured using the intensity track function of Image Pro plus.

PAGE 63

49 3.3. Analytical M odels In this work, diffusion of gas through the polymer material is described by Ficks law, which in one dimension is written in the form 32 for constant diffusion coefficient D Here, C ( x t ) is the concentration at position x at time t and D is the diffusion coefficient of the gas in the material. The solution of this equation depends on the boundary conditions at the edges of the film. We combined the linear SV equation with the diffusion model to extract the diffusion and permeation coefficients from the experimental data. A nonlinear least square fitting method is used to fit the model to intensity vs. time data to extract the diffusion coefficient. 3.3.1. Film Separated f rom The Luminescent Sensor by A Small Volume (Accumulation in v olume Case) In this case, two different models were used to analyze the data. 3.3.1.1 Ficks Equation Combined w ith T he SV Equation When the top side ( thickness, x=0 at the film surface) of a polymer surface is continuously ( time, maintained at zero concentration of O2 gas (flushed with pure N2 gas) and at the polymer film is kept at equilibrium with air (partial pressure of oxygen pair and the concentration of oxy gen Cl0 2 2x C D t C ), the experimental condition satisfies following boundary conditions: ) 0 t 0 t

PAGE 64

50 33 34 35 w here where S is the solubility of gas in polymer, T is the temperature, A is the surface area of polymer, C is concentration of oxygen, l is the total thickness of polymer and Vcell is the volume of the cell. For these boundary conditions, Ficks second law has been solved previously143 36 : where are the roots of: 37 Combining equation 36 with the SV equation, the luminescence intensity change due to change in oxygen concentration is related to the time. 38 where Iair and I0 are the luminescence intensity at the O2 concentration in air and in the absence of O2This model is fitted with experimental data using a nonlinear least square method to extract both and D values. Then, solubility of gas is calculated f rom Thus, we respectively. airSp C C 10 0 0 t l x 00 C C 0 0 t x 0 1 t Cx C D 0 t l x cellV STAl 278 0 k l t D k k k k air airke l x p p p2 22 2 2 2sin 2 1 k tan k l t D k k k k air airke l x I I I I2 22 2 2 2 0sin 2 1 1 1

PAGE 65

51 can extract information about both diffusion and permeation coefficients by fitting this model to data. 3.3.1.2 Quasi steady State M odel In this model, time is divided into arbitrary small intervals and the diffusion process is considered to be at steady state for each interval. Steady state differential material balance for each interval is combined with Ficks law for diffusion to determ ine the accumulation of gas into the cell. The amount of oxygen accumulated into the diffusion cell at the end of time interval i is: (3 9) F is the molar flux, A is the surface area of membrane exposed to gas, P is the permeation coefficient, and are partial pressures of gas outside and ins ide of the diffusion cell, respectively, and V is the diffusion cell volume. Partial pressure inside the cell for each time interval may be calculated from the ideal gas law: (3 10) where R is universal gas constant and T is th e absolute temperature. The quasi steady state model coupled with the SV equation can be used to predict the permeation coefficient data. Luminescence intensity at any time interval ti, (3 11) The inside partial pressure of oxygen for each interval ( ph,i) is calculated from equation 310. Eliminating ksv 0p from equation 311 it can be written as, hp ) .( . .1 1 , 0 1 1 i i i h t i i i i it t A h p p P M t A F M M T R V M pi i h., i h sv ip k I I, 01 /

PAGE 66

52 ( 312) From this model, only the permeabilit y coefficient can be extracted from exp erimental data using equations 39, 310 and 312. 3.3.2. Filmon s ensor M odel For the film on sensor experiment, where the upstream of the film is maintained under pure nitrogen exposure, with the film initially conditioned with air, the solution for Ficks second law is given by Crank ( 313) 144 Here, the initial concentration C0C is the concentration of oxygen in air and 1 Combining this equation with the SV model, the final equation can be written as = 0 (pure nitrogen) ( 314) 141 3.4. Results and D iscussion In the following section the characterization of sensors and diffusion measurement are discussed. 0 2 2 2 1 0 04 ) 1 2 ( exp 1 2 ) 1 ( 4 1n nl t n D n C C C C 0 2 2 2 04 ) 1 2 ( exp 12 ) 1 ( 4 1 1 1n n air airl t n D n I I I I ) ( ) / 1 ( / 1, 0 i h air air air i airp p p I I I I

PAGE 67

53 3.4.1. Characterization of S ensor s As mentioned in the experimental section, thin film of polystyrene containing PtTFPP dye coated on glass slide was used as a sensor The diffusion coefficient of oxygen in polystyrene was reported in the literature to be of the order of 1011m2/sec 138 Ellipsometry of sensor film indicated that a verage thickne ss of the sensor film was 300 nm to 400 nm. So the diffusion time of oxygen in the sensor film is negligible. Photobleaching of the sensor was studied by exposing the sensor to continuous illumination for 10 minutes. It was found that the intensity decrea sed by only 3% of the initial intensity. As the total exposure time during experiment was approximately 1 minute, photobleaching was minimal during the experiment so the photobleaching effect was neglected. In the present work, we performed diffusion ou t experiments to distinguish the quenching effect from the photobleaching effect.

PAGE 68

54 Figure 3-3 Pseudo -colored microscopic fluorescence intensity images (1.64 mm X 2.19 mm) of two luminescence sensors (PtTFPP/PS). A B 0 3450

PAGE 69

55 Figure 3 -4 S V plot for different regions of sensor a Spatial distribution of the SV response of fluorophore PtTFPP in the luminescence sensor was determined using a conventional fluorescence microscope. Figure 3.A and 3.B show pseudo colored 100X ma gni fied images of portions of two sensors at 0% oxygen concentration. Both of these sensors were fabricated by the same procedure and at the same time. From these images, it can be seen that there are some bright fluorescent spots (blue and green regions) in a nearly homogeneous background region (red and yellow regions) of low intensity. This shows the dye was not homogeneously dispersed in the sensor. Bright spots (blue) are due to micro crystal

PAGE 70

56 formation of the luminophore due to its incompatibility with the polystyrene matrix. The image here represents a 1.64 mm 2.19 mm region of the sensor. Four different regions of these two images were investigated for their SV response at nine different oxygen pressures. These regions were chosen from four different intensity regions. SV constants calculated for different regions of the two images (A and B) are given in Table 3. 1. Coefficient of determination R2Table 3-1 SV constants of different microscopic regions of luminescence sensors gives information about t he goodness of fit of the data to the SV model. Region Sensor A Sensor B After 30 minutes photobleaching of Sensor A K (psi SV 1 R ) K 2 SV (psi 1 R ) K 2 SV (psi 1 R ) 1 (Red) 2 0.85 0.999 0.46 0.961 0.77 0.998 2 (yellow) 0.90 0.998 0.53 0.974 0.79 0.998 3 (Green) 0.61 0.999 0.45 0.967 0.54 0.996 4 (Blue) 0.33 0.986 0.03 0.527 0.25 0.958 It can be seen from these results (Table 31) that SV constants and coefficient of determination values for both sensors are higher for nearly homogeneous low intensity regions (red and yellow) in comparison to highintensity microcrystal rich regions. Past investigations on heterogeneity of different sensors also have shown that mic rocrystalline areas show greater intensity but less quenching by oxygen.139. In contrast, Bedlek Anslow et al .140 observed the opposite effect for their sensors (tris(4,7 diphenyl 1,10phenanthroline)ruthenium(II) dichloride (Rudpp) dispersed in PDMS. They found that regions where the luminophore was aggregated showed less intensity due to self quenching.

PAGE 71

57 Though both the sensors were prepared by the same procedure, comparing the SV response from sensor A and sensor B, it can be seen that the SV response of sensor B is poorer. SV constants and R2 value are higher for lower intensity region of image A ( Ksv =0.9 psi1 R2= 0.998) in comparison to image B ( Ksv =0.53 psi1 R2= 0.974). This can be attributed to the fact that t he microscopic visual heterogeneity exhibited by i mage B is more than that of image A. Therefore, it is possible to obtain high KSV The effect of photobleaching on SV response of sensor A was also studied. The sensor was first exposed to continuous illumination for 30 minutes then SV analyses of t he same regions of image A were done again. The results are summarized in Table 31. Photobleaching of fluorophore adversely affects its oxygen sensing performance. Specifically, photobleaching effect is very much pronounced in the microcrystalline regi on. In case of diffusion measurement experiments, t he specimen s were only exposed to illumination while taking images The exposure time for acquiring each image was set to be 300 milliseconds Total exposure time of sample to light while taking data w as less than about 1 minute. Thus, photobleaching of the luminescence sensors can be considered negligible. values with good linearity depending on the sensor and as well as the image field chosen to study. Based on the SV analysis of different regions of the sensor film, the intensity change of regions which follow the linearity of SV equation and have high SV constants were examined for evaluation of oxygen diffusion parameters.

PAGE 72

58 3.4.2. Measurement of Diffusion Using Fluorescence M icroscopy Diffusion coefficient of oxygen in Teflon was measured using the accumulation in volume technique. The Teflon film was placed in the diffusion cell and a tight seal was achieved. The polymer film was first equilibrated with air, then the upstream section of the polymer was flushed with pure nitrogen and images of the sensor mounted in the downstream cell were taken simultaneously. A background image was acquired for the same experimental setup without luminophore. These images were processed according t o the procedure described in the experimental section. R esponses from the nearly homogeneous low intensity regions (which follow linear SV equation with high SV constants) of the sensor were analyzed for intensity change with respect to time as oxygen dif fused out from the downstream chamber across the polymer film. Partitioning of the si gn al to select areas of uniform intensity was performed using microscopesoftware generated intensity heterogeneity information. This approach allowed for identification of regions of uniform intensity. T he sizes of the different regions for which mean int ensities were measured were in the range of 0.04 mm2 to 0.25 mm2 Using the quasi steady state model, an approximate value of the permeation coefficient of oxygen in polymer film was first determined. Then, the diffusion model (Equation 38) was fit to experimental data to extract both permeation and diffusion coefficients. and at least 5 different regions of the sensor were sampled to represent the bulk response. Figure 34 shows intensity ratio vs time data for Teflon. The solid line represents the best fit to the diffusion model (Equation 38).

PAGE 73

59 Figure 3-5 Experimental and fitted data for the 0.025 mm thick Teflon film (* experimental data fitted data from model). Figure 3-6 Experimental and fitted data (* experimental data mm thick PDMS film. 0 0.5 1 1.5 2 2.5 x 104 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0 200 400 600 800 1000 0 0.2 0.4 0.6 0.8 1 1.2 Intensity ratio Time (sec) Intensity ratio Time (sec)

PAGE 74

60 Figure 3-7 Experimental and fitted data (* experimental data m 3-6265 HP polymer film (silicone elastomer) This experiment was repeated on the sam e polymer film as well as on different films of the same polymer. For T eflon, from the measurement on the same film, the diffusion and permeation coefficients were 1.99 E 11 2.2 E 13 and 2.7e 10 7.1E 12 molm2/m3secatm respectively and from measurement s on different films the extracted diffusion and permeation coefficients were 1.7 E 11 4.99E 12 m2/sec and 2.68E 10 9.7E 11 molm2/m3secatm, respectively. The uncertainty in the data measured for the same film may be due to photobleaching, and the uncer tainty in data measured from different film s is due to differences in the polyme r samples and sensors. These data compare wi th reported values of 1.84 1011 m2/sec and 1.62 1010 molm2/m3secatm145. These values for oxygen diffusion and permeation coefficients in Teflon were measured by Koros et al.145 in the temperature range of 40 to 850 0 50 100 150 200 250 300 0 0.2 0.4 0.6 0.8 1 C using a continuous permeation cell connected with a gas chromatograph. The differences between literature reported values and the value s reported by us, could have arisen from Intensity ratio Time (sec)

PAGE 75

61 differences in the polymer samples, and from the slight extrapolation in te mperature from the literature reported values to our experimental temperature of 25 C. In our lab electrochemical sensor technique was used by other researcher s to measure the oxygen permeation coefficient in same Teflon sample.146Table 3-2 Oxygen diffusion coefficients for various polymers The value for permeat ion coefficient in the Teflon sample obtained using electrochemical sen s or technique is comparable with value obtained using the fluorescence method presented here. Polymer type Zeolite weight percentage (%) Diffusion coefficient (m 2 3 6265 HP polymer (silicone elastomer). /sec) 0.0 4.52e 09 8.71E 10 PDMS 0.0 9.75E 10 1.25e 10 PDMS + zeolite 2.5 1.09E 09 1.09E 10 PDMS + zeolite 10.0 1.17E 09 1.47E 11 PDMS + zeolite 20.0 1.23E 09 4.04E 11 PDMS + zeolite 30.0 1.32E 09 5.29E 11 For the accumulation in volume technique, there are two challenges involved. One is to control the leakage of gas from the diffusion cell, and the second is the longer time associated with the experiment. Leakages were managed by constructing the diffusion cell from stainless steel and using viton gaskets. No change of intensity of sensor was noticed during the control experiment, indicating no leakage. Inherently, accumulation in volume technique is a lo nger experiment. For thicker films with smaller diffusion coefficients, it can take si gn ificant experimental times of several months to see any si gni ficant change of oxygen concentration in the downstream cell. Though by reducing the volume of the cell ex perimental time can be reduced, it is still a lengthier experiment and the technique is suitable for thin films only.

PAGE 76

62 In film on sensor technique, the test polymer was directly cast onto the sensor. In this case also, the polymer surface which was initia lly in equilibrium with air was flushed with nitrogen. As oxygen diffused out from the polymer its concentration change at the polymer sensor boundary was sensed by luminophore. The thickness of polymer film on sensor was measured using digital Vernier calipers. The thicknesses of PDMS and the HP polymer films were in the range of 0.6 to 0.8 mm and 0.48 to 0.50 mm respectively. Experimental data were fitted with the diffusion model (Equation 314) to determine the diffusion coefficient of oxygen in th e polymer. This method was first used to measure the diffusion coefficient of oxygen in PDMS for validating the technique. Experimental data and the fitted diffusion model are shown in Figure 3. 5. Oxygen diffusion coefficient for pure PDMS obtained from this fit is 9.75E 10 1.24E 10 m2/sec. This is within the range of values reported in the literature (0.54 to 3.4 m2/sec) .147As the sensor was monitored in the reflectance mode using an inverted fluorescence microscope, it is also possible to measure diffusion coefficient of oxygen in opaque films using this technique. To demonstrate this, we measured the oxygen diffusion coeff icient in 3 6265 HP polymer, which is black in color. The diffusion model fit the experimental data well (Figure 3. 6). New data for this polymer are given in Table 3. 2 which is of the same order of ma gn itude as known data for silicone elastomers. As for the film on sensor experiment, the polymer film attached to a sensor was placed in a stainless steel chamber and the surface of polymer film, initially equilibrated with air, was continuously flushed with N 2. There is a small, unavoidable time differe nce between manual opening of N2 cylinder to flush the chamber with N2 and the start of 910 910

PAGE 77

63 image acquisition. This may be the reason for the poor fit at short time shown in Figures 3. 5 and 3. 6. Figure 3-8 Data for the 0.65 mm thick PDMS film containing 10% zeolite (* experimental data fitted data from model). Fluorescence microscopy was also used for the measurement of oxygen diffusion in PDMS containing zeolite. The diffusion coefficients of oxygen in the polymer containing zeolites were measured using the film on sensor technique. It is shown in Figure 3. 7 that the oxygen desorption experimental data fit well to the Fickian diffusion model as described by eqn. 317. However, t he presence of zeolite causes a little deviation between the simulated data and experimental data. Zeolites in polymers affect gas diffusion in several ways148 0 200 400 600 800 1000 0 0.2 0.4 0.6 0.8 1 These particles can adsorb gas molecules and act as gas reservoirs, thus decr easing the diffusion coefficient and affecting dynamic behavior of the membranes cast from these polymer composites. Zeolites can hinder or can facilitate gas Intensity ratio Time (sec)

PAGE 78

64 diffusion depending on the kinetic diameter of gas molecules147. Gas diffusion in polymers also depends on available free volume of the polymers. Free volume of the polymer at the proximity of polymer zeolite boundary can either be reduced or enhanced149 In contrast to the previous cases reported in literature On the other hand the packing density in unoccupied zones may increase which may cause the d ecrease in oxygen diffusion coefficient. As the zeolite content increases, the void spaces formed around the zeolite also increase enhancing the oxygen permeability. On the other hand the packing density in unoccupied zones may increase, which may dec rease the oxygen diffusion coefficient. 150, timelag in the diffusion in this case was reduced. Diffusion coefficients for zeolite free and zeolite filled PDMS are given in Table 3. 2. Diffusion coefficients reduce as the weight percentage of zeolite increases in PDMS. This trend agrees with literature148. Hence, it can be concluded that presence of zeolite in PDMS introduces more free volume as well as more pores which enhance oxygen (kinetic diameter 3.46 Applicability of the methods developed here is subjected to the condition that no component present in the polymer interferes with the response of the fluorescence sensor to oxygen concentration. For example this method was not successful to measure oxyg en diffusion coefficient in epoxy polymer, as this polymer shows fluorescent property in the emission wavelength range of PtTFPP dye. 3.5. C onclusions We demonstrated the application of conventional fluorescence microscopy in studying the relationship betwee n microscopic heterogeneity and the nonlinearity of SV

PAGE 79

65 responses of luminescence sensors. Based on this study a fluorescence microscopy technique was developed to measure diffusion and permeation coefficients of oxygen in polymers. Using this microscopy technique, microscopic level SV resp onses of heterogeneous sensors are measurable. This technique allows for the distinction of the responses of background region (nearly homogeneous regions) from the regions of aggregated luminophores. As the nearly ho mogeneous regions show better response to oxygen concentration and follow the linearity of the SV equation, by studying the response of these, one can eliminate the complexity of combining the nonlinear SV equation with a diffusion model. We also found th at the sensors prepared by the same procedure behave differently in term of SV responses. The sensors with less visual microscopic heterogeneity show better responses. Fluorescence microscopy allowed us to visually inspect and chose better sensors for th e application. With this method, diffusion data for Teflon and PDMS were obtained, which compared well with literature values. New data for 3 6265 HP polymer (a silicone elastomer) an d PDMS containing zeolite are of an expected order of magnitude with co mparable materials. We developed a new, simple quasi steady model for describing diffusion phenomena for the accumulation in volume technique. Photostable luminophore is essential for this technique to be successful. Minimizing photobleaching of the luminophore is a challe nge for this method that was overcome by shuttering techniques. The methods developed here can be applied for measuring oxygen diffusion properties in polymers ranging from transparent to opaque, subjected to the condition that no component is present in the polymer which interferes with the response of the sensor to oxygen concentration. The technique is suitable for polymers that cannot be cast into

PAGE 80

66 free standing films, and yields reliable data in reasonable experimental timeframes. This method is also suitable for polymer composites. We expect this fluorescence microscopy technique will be very useful for measuring O2 diffusion coefficient in biological samples simultaneously with imaging these samples.

PAGE 81

67 Chapter 4 E ffect of Ag Cu Alloy N anoparticle Composition on Luminescence Enhancement/Q uenching 4.1. Introduction The emission of luminophores is significantly influenced in close proximity of conducting metallic nanostructures. Using nanoparticle platforms, it is possible to increase the quantum yield of weakly luminescent probes This increase results from a modification of the radiative decay rate by coupling the emissio n with surface plasmon resonance (SPR), and by coupling emission at far field with nanoparticle scattering These nanostructures can also enhance the excitation intensity experienced by vicinal luminophore molecules by enhancing the incident optical field by increasing the local field at the molecular location 11,28,39,151 The presence of nanoparticles close to the luminophores can create new nonradiative channels due to light absorption inside the metal thus quenching the emission of luminophores. 30 If the probe molecules are very close to the nanoparticles (typically less than 5 nm), luminescence emission is quenched due to Frster transfer of energy from the excited state of the molecule to the surface plasmons of the metal surface. This quenchi ng effect decreases with the cube of separation distance. 56 If the probes are too far from the nanoparticle s the influence of the nanoparticles is diminished. Hence, there exists an optimum separation distance for maximum emission enhancement /quenching. 13,21,23,152,153

PAGE 82

68 Metal enhanced luminescence (MEL) has been studied mostly using silver nanoparticles 3,11,16,19,33,39,151,154 due to their intense and narrow SPR peaks. Gold nanoparticles are known to both quench and enhance luminescence depending on t he fluorophore particle separation distance, molecular dipole orientation with respect to particle surface, and size of the nanoparticles 22,29,43 Relatively smaller (typically less than 30 nm) gold nanoparticles quench fluorescence emission due to non r adiative transfer from the excited states of luminophore molecules to the gold nanoparticles 43 Larger gold nanoparticles can enhance luminescence due to the increased contribution of nanoparticle scattering 22,155 Other metals such as copper and alumi num have been reported to enhance luminescence. 17,44 Recently zinc oxide (ZnO) nanorod platforms have been reported to enhance luminescence intensity significantly from commonly utilized fluorophores in immunoassays 4749 Both enhancement and quenchi ng of luminescence due to the proximity of nanoparticles are efficiently utilized for many different applications. Enhanced signal and photostability of luminophores, improved surface immunoassay and DNA detection, enhanced wavelength ratiometric sensing, and amplified assay detection are few examples of the applications of MEL. On the other hand quenching result ing from metallic nanoparticles has been successfully utilized for the improvement of homoge neous and competitive fluorescence immunoassay 156,157 optical detection of DNA hybridization 158 competitive hybridization assay 159 and in optoelectronics 160There are some theoretical models explaining the influence of metal nanostructures on lu minescence of dye s in the literature. Model s based on exact electrodynamical theory 92,93 and the Gersten Nitzan ( GN) model 93,95,96 provide insight

PAGE 83

69 i n to the influence of metal nanospheres on radiative and nonradiative decay rates of luminophore molecule s at close proximity. These theories explain that electroma gn etic interaction between luminophore and metal nanostructures result s in the increase of both radiative and nonradiative decay rates depending on luminophore nanoparticles separation distance a nd the properties of nanoparticles (size, shape, and dielectric constant) which decide the scattering and surface plasmon resonance behavior of the nanospheres. Based on these theories it can be concluded that both radiative and nonradiative decay rates can be manipulated to result in luminescence enhancement or quenching by design ing nanostructured platforms of particular shape, size, and composition. Mertens et al. 93,96 have corrected the GN model to account for radiation da mping and dynamic depolari zation and have show n that result s obtained using this corrected GN model compare well with a model based on exact electrodynamics This corrected GN model is suitable for a larger particle size range than the original version. Kmmerlen et al. 33 presen ted a model that is based on the GN model and includes both excitation enhancement by local field effects and the change in emission intensity due to radiative and nonradiative decay rate enhancement. In our study, we used a theoretical model based on th eory proposed by K mmerlen et al. 33 and Mertens et al. 93SPR wavelength and scattering efficiency, the most important properties of nanostructures which dictate the enhancement/quenching of luminophore molecules to study the effect of composition of alloy nanoparticles on quantum efficiency enhancement. 24,154, can be manipulated by controlling any of the parameters of particle size, aspect ratio shape, particleto particle distance and surrounding dielectric medium 54,154,161 Alloy nanopartic les offer additional degrees of freedom for tuning their optical properties by

PAGE 84

70 altering atomic composition and atomic arrangement 162, thus can be an attractive option for manipulating the luminophore signal. Herein, we report the use of alloy nanoparticl es for MEL. We demonstrate that by tuning the composition of alloy nanoparticles the signal of vicinal luminophore can be manipulated. Due to their interesting optical properties, we chose silver copper alloy nanoparticles as a material for our study 1 01,102,116,163 Figure 41 shows imaginary component s 2) for 10 nm Ag and Cu nanoparticles in the wavelength range of 200 nm to 800 nm. The imaginary components of dielectric constants of bulk metal are modified using the model suggested by Garcia et al68 (Equation 226) From this F igure 41, we can see t he imaginary component of the dielectric constant of copper is significantly larger (more than twice) than that of silver in the wavel ength range of 300 nm to 600 nm Hence, it is expected that in this wavelength range due to higher ohmic losses, Cu nanoparticles will mostly quench the luminescence at close proximity 17 Further, t he SPR spectrum of Ag is more intense and narrower than that of Cu nanoparticles. The absorption peak attributed to SPR occurs at shorter wavelengths for Ag. Hence, by modifying the composition and atomic arrangement we can tune both breadth and location of the peak of the SPR spectrum of Ag Cu alloy nanoparticles 163 We observed the effect s of Ag Cu alloy nanoparticles on the fluorescence emission from Cy3, a commonly used luminophore in biological applications. We chose Cy3 due t o its low quantum yield (< .04) Cy3 is a reactive water soluble fluorescent dye of the cyanine dye family with excitation peak at 550 nm and emission pe ak at 570 nm ( see Figure 42 for molecular structure ).164 We found that the composition of alloy nanoparticles has a strong effect on

PAGE 85

71 MEL. We establish simple and straightforward routes for manipulating the brightness of emission from luminophore by changing the composition of the alloy nanoparticles Figure 4 -2 of 10 nm Ag and Cu nanoparticles. Figure 4-2 Molecular structure of Cy3. 0 2 4 6 8 10 12 0 200 400 600 800 1000 Ag Cu 164 2 Wavelength (nm)

PAGE 86

72 4.2. Experimental In this study, Ag Cu nanoparticles of five different compositions were synthesized using the polyol process as described in reference. 110 Silver nitrate (>99%), copper (II) acetate hydrate (98%) and polyvinylpyrrolidone (PVP, 55000 molecular weight) were obtained from SigmaAldrich Co. MO and used as received. Same volume of solution of PVP ( 1.0634 g in 20 ml ethylene glycol ) was first added to ethylene glycol solution of copper salt ( 0.016 moles ) and was then de aerated by bubbling with nitrogen for 30 minutes. The solution was then held at 1750 C for 20 minutes under nitrogen atmosphere, and a certain amount of AO3G lass substrates were silanized to immobilize silver copper nanopar ticles on these ethylene glycol solution was added to it. T he reaction was then allowed to continue for another 5 minutes before bringing the system down to room temperature. Alloy nanoparticles of different compositions were synthesized by varying the molar ratio of silver and copper salt s in the reaction mixture. With an increase in copper percentage, the color of the colloidal solution changed fr om yellowish to more reddish. Copper nanoparticles were synthesized following the same procedure except that the silver nitrate solution was replaced by the reducing agent ascorbic acid. 40. Glass slides were first cleaned with piranha solution for 30 minutes (1:3 30% hydrogen peroxide/concentrated sulfuric acid ) ; ( CAUTION! Piranha solution reacts v iolently with most organic materials and should be handled with extreme car e) The cleaned glass substrates were silanized by immersing them in 2% 3 (aminopropyl)triethoxysilane (APS) solution in methanol for 2 hours 40 After th is, the slides were thoroughly cleaned with methanol followed by water to remove any excess

PAGE 87

73 APS. Ag Cu nanoparticles were deposited on the APS coated glass slides by soaking the m in freshly prepared solutions for specific time s Copper nanoparticles were immobilized on the glass slides following the procedure given by Male et al. .165Silver nanoparticles were synthesized using the well know n Tollens reaction Piranha cleaned glass slides were immersed in 20% poly (diallyldimethylammonium chloride), (PDDA, MW 200 000350 000, Aldrich) aqueous solution for 16 h ours. Then, these slides were thoroughly rinsed with deionized water and dried in a nitrogen stream. These polymer coated glass slides were incubated in Cu nanoparticle solution for 3 hours. Finally, Cu nanoparticles coated glass slides were rinsed with deionized water and dried with nitrogen. 21 In summary, 10% a mmonium hydroxide was added to 10 ml of aqueous AO3 (0.1 M) while stirring. Once the initially formed brown prec ipitate dissolved, a 0.8 mole solution of NaOH in water was added to the solution. Preparation of Toll ens reagent was completed by adding NH4 OH drop wise to the solution until the brown precipitate dissolved. The Tollens reagent was stored in a refrigerator for 30 minutes to reduce its temperature to 40 C. For deposition of silver nanoparticles on glass substrates, equal amounts of the Tollens reagent and 0.5 M dextrose solution were mixed together and immediately drop cast on a piranhaclean ed glass substrate followed by rinsing with de ionized water after 1 minute. The s urface morphology of the nanost ructures was observed and characterized by transmission electron microscopy (FEI Morgai 268D) atomic force microscopy ( Digital Instruments, Nanoscope IIIa ) and scanning electron microscopy (Hitachi S 800) A UV vis spectrometer (JASCO, V 530) was used for measuring the light extinction spectra attributed to the SPR of these nanoparticles. TEM samples were prepared by

PAGE 88

74 dispersing a few drops of AgCu alloy nanoparticle solution on a carbon film supported by molybdenum grids. Luminophore coatings on the nanoparticles and glass substrates were accomplished by dispersing Cy3labeled streptavidin in 0.25% poly (vinyl alcohol) (PVA MW 15000) aqueous solution by sonicating and then coating the solution on the substrates by spin coating (1500 rpm speed). The r esulting polymer thickness was approximately 26 nm. Hence, the average distance between the substrate and a luminophore mole cule was approximated by 13 nm. A s the luminophores were coated following the same procedure for all samples, the separation distance between luminophore molecules and nanoparticles and the coverage of the luminophore molecules on nanoparticles are assumed to be the same for all samples. A Leica DMI 4000b inverted fluorescence microscope equipped with a Leica DFC340 FX CCD camera was utilized for all luminescence measurements. This allowed inspection of a large area in a single view frame. Fluorescence m icroscopy was carried out with customized filter sets (Chroma Technology) for Cy3. To avoid photobleaching, the specimen was exposed to illumination only while taking images. Image Pro plus version 6 with Scope Pro version 6 (Media Cybernetics, Inc) was used for acquiring and analyzing images. We obtained fluorescence intensities for each sample by analyzing a 1.64 mm 2.19 mm image section of each substrate. Background images were obtained from an uncoated substrate and unmodified glass cover slips at the same conditions. Images from the experimental samples were corrected for uneven illumination with the help of these background images. Image s of nanoparticle coated glass coverslip w ere captured and compared with the image of a bare glass coverslip t o test for the possibility

PAGE 89

75 of scattered light from metal particles. These images showed that the emission filters effectively removed the scattered light so its contribution is negligible. The l uminescence intensity of each sample was determined by measuring the mean intensity and subtracting the mean value of the background image. 4.3. Result s and D iscussion The UV Vis absorbance spectra attributed to surface plasmon resonance of colloidal Ag Cu nanoparticles show a single peak in the visible range. Wi th increasing copper percentage, this SPR peak shifts to longer w avelengths (Figure 43). This result confirms that the nanoparticles are a bimetallic form of silver and copper and not a mixture of silver nanoparticles and copper nanoparticles 166 The r ed shifts of the SPR peaks with increasing copper concentration are attributed to the decrease in conductivity 101 There is no visible difference between the position of absorbance peaks o f Ag Cu nanoparticles in solution and on APS coated slides. Figure 4-3 Normalized Absorption spectra for Ag -Cu alloy nanoparticles Dotted line is for Ag -Cu nanoparticles with 33% Cu on APS coated glass slides. 400 450 500 550 600 650 700 750 0.7 0.75 0.8 0.85 0.9 0.95 1 1.05 W avelength (nm) Absorbance (arb. units.) 33% 50% 70% 100 %

PAGE 90

76 Transmission electron microscopy (Figure 44) of the colloidal Ag Cu alloy nanoparticles indicated the particle size to be in the range of 130 nm to 200 nm (derived from a population of 100 particles). STEM EDS data (Figure 44(C) and 4 4 (D) ), confirms t hat the nanoparticles comprise both Ag and Cu. The energy dispersive X ray analysis on the single particle showed that the composition for each particle was roughly consistent with that of feeding solution (Figure 44(C ) and (D)). Estimation of exact com position of Ag Cu nanoparticles, which can be measured by using method like inductively coupled plasma mass spectroscopy (ICP MS) analysis167The concentration of nanoparticles increased w ith increase in immersion time of APS coated glass slides in Ag Cu colloid al solution s. As the copper percentage increased, the time required to attach the Ag Cu colloids on gl ass slides also increase d. For comparison APS coated glass slides were allowed to soak in different composition Ag Cu colloid al solution s un til the concentration s of nanoparticles on glass slides were approximately the same. The sizes of different compo sition Ag Cu nanoparticles coated on glass slides were also found to be approximately the same. The siz e of nanoparticles and particle density were measured using Image j software. From the SEM images of the Ag Cu nanoparticles (Figures 45(B) and 45(C) ) the average size of these nanoparticles on the glass slides was measured to be approximately 150 nm (derived from a population of 800 nanoparticles). SEM images of the Ag nanoparticles (Figure 45(A)) indicate their average size to be approximately 80 n m. AFM images of the Ag and AgCu nanoparticles on glass slides are given in Figures 4 6( A ), 46( B ) and 46( C ). The particle is beyond the scope of this study. However, lack of information about exact composition should not affect the conclusions of this study

PAGE 91

77 density for Ag nanoparticles was estimated to be 38 particles/square microns. Particle density for AgCu nanoparticles was estimated to be 20 particles/square microns. It is difficult to obtain the same size and particle density for silver and silver copper nanoparticles due to limitations of the synthesis techniques. Figure 4-4 TEM image s of Ag -Cu n p synthesized from different composition feeding solution (A) Ag/ Cu (1/1) and (B) Ag/ Cu (3/7) STEM EDS spectra for (C) Ag/ Cu (1/1) and (D) Ag/ Cu (2/1) A B C Energy (keV) Counts Counts Energy (keV) D

PAGE 92

78 Figure 4-5 SEM images of (A) Ag nanoparticles (B) 2:1 Ag-Cu (C) 1:1 Ag-Cu nanoparticles coated on glass substrate Figure 4 -6 AFM images of (A ) Ag nanoparticles ( B ) 2:1 Ag-Cu nanoparticles ( C ) 1:1 Ag -Cu nanoparticles coated on glass substrates. Luminescence intensity of Cy3 was observed to increase significantly in the vicinity of both Ag and Ag Cu nanoparticles (Figure 47). The e nhancement ratio for Ag and Ag Cu nanoparticles was calculated by comparing luminescence intensity of the sample with the luminescence intensity of the luminophore coated on an APS coated glass substrate Please note average fluorescence intensity of dye coated on glass is not zero here. In the case of copper nanoparticles the enhancement ratio was calculated A B C A B C

PAGE 93

79 comparing the luminescence intensity with luminophore coated on PDDA coated glass slides. The Ag nanoparticles platform result ed in very strong en hancement ( 90 19 times) for Cy3. As the quantum efficiency of dye Cy3 is very small, the enhancement effect is high. The Ag Cu nanoparticles also showed enhancement ( 55 15 times for 2:1 Ag Cu, 30 6 times for 1:1 Ag Cu) but as the copper percentage in nan oparticles increased, the enhancement decreased. Finally instead of enhancing, the Cu nanoparticles quench ed ( 7 5 times) the luminescence of Cy3. This may be due to the fact that in the vicinity of metal nanoparticles both the radiative decay rate and the nonradiative decay rates increase, and as the percentage of Cu increases the nonradiative decay rate also increases eventually surpassing the radiative decay rate. Figure 4-7 Pseudo colored image of Cy3 coated on (A) glass (B) Ag nanoparticles (C) 1:1 Ag -Cu nanoparticles and (D) Cu nanoparticles. We calculated the modified overall quantum efficiency at the proximity of different compositions of Ag Cu nanoparticles based on the model suggested by Kmmerlen et al. 33 which includes both excitation and emission enhancement factors as discussed in Sect ion III. The absorption enhancement factor was calculated based on the 3450 (a.u.) A B C D 0 (a.u.)

PAGE 94

80 enhancement of local electric field at the excitation frequency ) (abs The corrected GN model 93,96 model was used to calculate the quantum efficiency change due to radiative and the nonr adiative decay rate enhancements. For better representation of experimental condition s, the source dipole orientation was averaged over all solid angles This was achieved by averaging the results for decay rates obtained for radial and tangential orient ations. Dielectric constants for Ag Cu nanoparticles of different compositions were calculated following the procedure described by Bruzzone. 105 The dielectric function was calculated using the semi empirical model based on Drude theory and experimental data. The experimental data used for this calculation were obtained by averaging the values for pure metals over the volume. 168 Drude contributions for nanostructure and bulk were calculated using the values of pure metal averaged over volumes. Though Ag Cu cannot form a solid solution at room temperature as does Ag Au, the surface plasmon resonance spectr um resembles that of alloy nanoparticles 113 This is due to the fact that both silver and copper exist in the surface of Ag Cu nanoparticles and surface plasmon resonance is a surface phenomenon 113

PAGE 95

81 Figure 4-8 (A) Experimentally observed luminescence enhancement ratio of Cy3 (B) Inset shows theoretically calculated overall luminescence quantum efficiency enhancement ratio. Figure 4-9 (A ) Calculated quantum efficiency enhancement factor due to emission enhancement. (B) and excitation enhancement factor Calculations were done to corroborate experimental results and to establish the optimum size of nanoparticles. Figure 48 A and 48B show the theoretically calculated modified overall q uantum efficiency and the experimentally observed luminescence -0.2 0 0.2 0.4 0.6 0.8 1 1.2 0 20 40 60 80 100 120 0 0.2 0.4 0.6 0.8 1 0.5 1 1.5 2 2.5 3 3.5 4 0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5 Calculated overall quantum efficiency enhancement Mole fraction of copper Mole fraction of copper in AgCu nanoparticles Experimentally observed luminescence intensity enhancement ratio A B 0 0.2 0.4 0.6 0.8 1 0 5 10 15 20 B Calculated quantum efficiency enhancement Mole fraction of co pper Mole fraction of copper Calculated excitation enhancement ratio C D

PAGE 96

82 enhancement ratio of luminophore Cy3 in the vicinity of different compositions of Ag Cu nanoparticles. The calculations were done assuming the size of nanoparticles to be 150 nm and the separat ion distance between nanoparticles and luminophore molecules to be 13 nm to compare with experimental results. The s urrounding dielectric medium was assumed to be poly (vinyl alcohol). The e nhancement of local electric field amplitude ( 2 absL ) was calculated at the absorption frequency of Cy3 (550 nm). The q uantum efficiency change ( fluZ ) due to radiative and nonradiative decay rate enhancement was calculated at 570 nm emission wavelength, which is the emission peak for Cy3. The quantum efficiency was calcula ted taking into account all multipole modes up to l =100. Dipole orientation was assumed to be averaged over all solid angles It can be seen from Figures 48A and 48B that both theoretical and experimental results show the same trend that with increase in copper percentage in nanoparticles, the enhancement effect decreases with pure copper quenching luminescence. The t heoretically calculated emission enhancement factor and excitation enhancement factor are separately shown in Figures 49 A and 49B res pectively. We can see that both emission and excitation have comparable effects on overall quantum efficiency change. Some reasons for the discrepancy in numerical values between theoretical and experimental results are the differences in experimental ge ometry (nanoparticles are not in a homogeneous dielectric environment, all the nanoparticles were not of spherical shape and not of same size, luminophore nanostructures separation distance is not precise) with respect to theoretical calculations, which as sumed uniformity in these parameters. It can be also because we observed the luminescence intensity of the image over the entire bandwidth of filters used for fluorescence microscopy not at any particular wavelength.

PAGE 97

83 Figure 4-10 Quantum efficiency enhancement ratio of Cy3 in the proximity of different diameter AgCu nanoparticles at different compositions. It is known that qu antum efficiency enhancement depends on the size of spherical nanoparticles 93. Figure 410 shows the dependencies of quantum efficiency enhancement on the size of Ag Cu nanoparticles at the fluorophore nanoparticles separation distance of 13 nm. The cal culation for Figure 4 10 was done considering the same emitter particle orientation and surrounding conditions as for Figure 48. It can be seen from Figure 410 that there is an optimum size of nanoparticles for which quantum efficiency enhancement is maximum. The coupling between the emission of the luminophore and the plasmon mode increases as the size of the nanoparticles decreases, and coupling efficiency of emission at far field through nanoparticle scattering increases as the size of nanoparticles increases. Both of these coupling phenomena are responsible for enhancement of quantum efficiency. Spectral overlap between the absorption and emission spectra of luminophore and surface plasmon resonance spectra of metal nanoparticles is very important f or optimum luminescence enhancement24,154 0 5 10 15 20 25 30 35 40 15 35 55 75 95 Some Calculated quantum efficiency enhancement ratio Radius of nanosphere (a) Cu mole fraction 0% 20% 40% 60% 80% 100%

PAGE 98

84 theoretical and experimental studies have suggested that luminescence enhancement is largest when the emission wavelength is slightly red shifted from that of the plasmon resonance24,26,169. When the size of the particle increases, the plasmon resonance is shifted to longer wavelength and broadened and decreases in magn itude due to dynamic polarization170 So, there exists an optimum diameter. We can see from Figure 6 that the optimum radius for Ag, Ag Cu an d also for Cu nanoparticles is approximately 60 nm. At this optimum diameter, even Cu nanoparticles show enhancement instead of quenching. So it can be inferred that if we can synthesize alloy nanoparticles of optimum diameter using advanced methods like electron beam lithography, we can elucidate the effect of composition on metalenhanced luminescence better 4.4. Conclusions In summary, in this work metalenhanced luminescence/ quenching of luminophore Cy3 is explored in the vicinity of Ag Cu alloy nanoparticles at different compositions. T he effect of composition of Ag Cu alloy nanoparticles on luminescence enhancement is studied. We have shown that strongest enhancement is observed on the Ag nanoparticles platform, and as the percentage of copper increases in the nanoparticles, the enhancement decreases. At pure copper nanoparticles platforms, the luminescence is quenched. A simple technique to tune the brightness of a luminophore by changing the composition o f alloy nanoparticles is presented. E xperimentally obtained data for luminescence change qualitatively match with theoretical calculations. We believe such manipulation in luminescence brightness of a dye will open up different applications of luminescence emission. We expect quenching effect of copper nanoparticles will

PAGE 99

85 motivate the utilization of these nanoparticles as an inexpensive alternative to gold in biological applications such as homogeneous and competitive fluorescence immunoassay detection of DNA hybridization, competitive hybri dization assay and also in optoelectronics

PAGE 100

86 Chapter 5 Silver Copper Alloy N anoparticles for M etal Enhanced L uminescence 5.1. Introduction Luminescence based measurements and devices are currently widely used methods in different fields such as biology, chemistry, materials science and medicine. Single molecule detection171, DNA sequencing172, quantum cryptography173, and LEDs174 are some examples of its numerous, diverse applications. Strong luminescence intensity is one of the most important desir ed properties of luminophores for their applications in luminescence sensors. It is possible to design and synthesize luminophores w ith desired spectral properties. However, it is difficult to design luminophores with desired luminescence intensities. Nearby conducting metallic particles, colloids, and surfaces are known to significantly influence the emission of vicinal luminophores3, ,9,11 13,16,17,2024,2729,32,33,3639,44,151,154,155,175177. Planar metal films are generally known to quench the emission from nearby fluorophores. Luminescence enhancements ranging from tens to hundreds fold in signal intensity have been reported in the literature3, 22,24,27,41,47,174,178. Though the phenomena of metal enhanced luminescence (MEL) is known from the 1980s, the demonstrations and applications of MEL are mostly new. Different applications of metal enhanced luminescence and from different metallic nanoparticles have been reported in recent literature 2 17. MEL has been studied mostly using silver and gold nanoparticles19 24,2729,33,34,40,41,51,179 due to their intense and narrow SPR peaks.

PAGE 101

87 Recently, other metals such as copper and aluminum have been reported to enhance luminescence 17,44 But, due to the higher ohmic losses, the MEL effect is not as pronounced in Cu and Al as it is in Ag or Au. Recently zinc oxide (ZnO) nanorod platforms have been reported to enhance luminescenc e intensity significantly from commonly utilized fluorophores in immunoassays47 49Luminescence enhancement phenomenon is dependent on several parameters such as material properties, size and shape of nanostructures, and luminophorenanostructure separat ion distance. Metal nanoparticles can influence vicinal luminophore molecules in several ways such as by enhancing the incident optical field, increasing the radiative decay rate and quenching the emission by increasing nonradiative decay rate 11,28,39,151. If the probe molecules are very close to the nanoparticles (typically less than 5 nm), luminescence emission is quenched due to Forster transfer of energy from the excited state of the molecule to the surface plasmons of the metal surface. This quenchi ng effect decreases with the cube of separation distance56. If the probes are too far from the nanoparticle platform, the influence of the platform is diminished. Hence, there exists an optimum separation distance for maximum emission enhancement13,21,23,152,153Using nanoparticle platforms, it is possible to increase the quantum yield of weakly luminescent probes by modifying their radiative decay rate to increase their emission efficiency, or by coupling the emission with far field scattering. The emission intensity of luminophores with nearly unit quantum yield can also be improved by enhancing their absorption by increasing the local electric field. Light intensity of nanoparticles at near field is strongly dependent on the surface plasmon resonance (SPR)

PAGE 102

88 wavelength of the metal nanostructures. SPR wavelength, one of the most important properties of nanostructures, dictates the choice of materials to be used for luminescence enhancement. Tam et al.154 found that the enhancement is optimal when the plasmon resonance wavelength of the nanoparticles is tuned to the emission wavelength of the low quantum yield luminophores. Recent theoretical and experimental studies have suggested that luminescence enhan cement is largest when the emission wavelength is slightly red shifted from that of the plasmon resonance24,26. Chen et al.24 suggested that the optimal location of the SPR peak of nanoparticles is between the excitation and emission peaks of luminophores for maximum enhancement, as both excitation and emission rates can be enhanced in such a situation. One can expect that the ability to tune the position of the SPR peak of the nanoparticles over a wide range of wavelengths will allow for extension of the MEL phenomenon to a wide range of luminophores. So far, MEL has been studied mostly on pure metal platforms. SPR wavelengths of pure metal nanoparticles can be tuned to different values by controlling several parameters such as particle size, shape, part icle to particle distance and surrounding dielectric medium24,53,54,180,181. However, it is easier to tune SPR spectra of alloy nanoparticles over a wide range of wavelengths as these offer additional degrees of freedom for tuning their optical properties by altering atomic composition and atomic arrangement162Herein, we report the use of alloy nanoparticles for MEL. We demonstrate that SPR spectra of alloy nanoparticles can be tuned by manipulating an easily controlled This could potentially enable development of specifically tailored nanoparticle platforms for MEL of a wide range of luminophores. This is the motivation for us to study alloy nanostructured plat forms for MEL.

PAGE 103

89 experimental variable to result in maximum spectral overlap of the emission and absorption spectra of the luminophores with the SPR spectrum of the nanoparticles. Also, we show that Ag Cu nanomaterials can serve as excellent candidates for MEL, due to their interesting optical properties101,102,116,163. These alloy nanoparticle s are less lossy than pure Cu ones116, hence, expected to result in better MEL. The SPR spectrum of Ag is more intense and narrower than that of Cu nanoparticles. The absorption peak attributed to SPR occurs at shorter wavelengths for Ag. Hence, by modi fying the composition and atomic arrangement we can tune both breadth and location of the peak of the SPR spectrum of Ag Cu alloy nanoparticles163. SPR peak wavelengths of AgCu alloy nanoparticles can easily be tuned in the visible and near infrared reg ion by changing only the annealing temperature101 We observed enhanced fluorescence emission from two thiol reactive dyes, Alexa Fluor 594 and Alexa Fluor 488 (obtained from Molecular Probes, Invitrogen, Portland, OR), at the proximity of these Ag Cu all oy nanoparticles. We establish simple and straightforward routes for the successful growth and fabrication of nanostructured platforms which can be effectively utilized to enhance the luminescence of any luminophore. In addition, our work also provides i nsights into the effect of SPR on MEL. 5.2. Experimental M ethod In this study, Ag or Ag Cu nanoparticles were deposited on 22 22 mm glass cover slips (Fisher finest cover glass, thickness approximately 140 microns) by using DC ma gn etron sputtering (Plasma Scie nces CRC 100 Sputter Tool). Before the depositions, the cover slips were cleaned by air plasma (Harrick PDC 32G) for 10 minutes at 6.8

PAGE 104

90 watts power setting. During deposition, the background pressure was of the order of 106 Torr, the Ar pressure was 5 mTorr and the current and voltage w ere 50 mA and 0.4 kV respectively An Ag target was utilized to deposit the Ag nanoparticles and a Cu foil attached on the Ag target was utilized for the Ag Cu nanoparticle deposition (Figure 51) Varying the ratio of the surface area of Ag to Cu exposed for sputtering allowed for changing the composition of the Ag Cu alloy nanoparticles. Surface morphology of the nanostructures was observed and characterized by transmission electron micros copy ( FEI Tecnai F20 S Twin TEM) An electrical furnace (Lindberg, Blue M) was used for annealing of the Ag Cu nanoparticles. Annealing temperature ranged from 298 K to 523 K and the annealing time was 5 minutes. Annealing was done in vacuum (30 inch Hg vacuum) to minimize oxidation of the nanoparticles. An UV vis spectrometer (JASCO, V 530) was used for measuring the light absorption spectra attributed to the SPR of these nanoparticles. Figure 5-1 Picture of DC magne tron sputterer with Ag -Cu target Mouse Immunoglobulin G (IgG), labeled with luminophores Alexa Fluor 488 and Alexa Fluor 594 was coated on samples ( F igure 52) following known methods 177 Target holder shutter Ag Cu Samples

PAGE 105

91 were first non covalently coated with mouse anti rabbit IgG (Immunopure, Pierce Biotechnolo gy) solution ( 25 g/ml) which was diluted with sodium phosphate buffer (pH 7.4). Blocking was performed using blocking solution (1% bovine serum albumin solution in sodium phosphate buffer). Protein labeling kits of both Alexa Fluor 488 and Alexa Fluor 594 were used to label goat anti mouse IgG with dye. Dye labeled anti mouse IgG was also diluted using sodium phosphate buffer. Diluted dye labeled conjugate solution was coated on the sample (already coated with mouse anti rabbit IgG). D etail s of the coating procedure are as f ollow s The samples were covered with tape containing punched holes (of size 36 mm2 As the luminophores were coated following the same procedure for all samples, the separation distances between luminophore molecules and the various nanoparticle platforms are assumed to b e the same. ) to form wells on the surface of the slides A coating solution of IgG (25 phosphate buffer ) was added sample s were incubated for 4 h at room temperatu re in a humid container. Samples were then rinsed with water. Blocking was performed by adding blocking solution per well and incubating at room temperature for 4 h in a closed humid container again. 25 l abeled conjugate dye anti mouse IgG (diluted phosphate buffer, 50 mM, pH 7.4) was added to each hole of the sample slide (coated with mouse IgG) and s amples were incubated at room temperature in a humid container for 2 h. S ampl es were then rinsed with water and were ready for the measurement.

PAGE 106

92 Figure 5-2 Luminophore s on Ag -Cu nanoparticles platform 177 The Leica DMI 4000b inverted fluorescence microscope equipped with Leica DFC340 FX CCD camera was utilized for MEL measurements. This allowed overall inspection of a large area in a single view frame. We took images of each specimen with customized filter sets for each luminophore. Fluorescence microscopy was carried out with a green filter set (Chroma Technology 31001, Exciter D480/30x, Dichroic 505 nm ,Emitter D535/40m) for Alexa Fluor 488 and red filter set (Chroma Technology 31004, Exciter D560/40x, Dichroic 595 nm ,Emitter D630/60m) for Alexa Fluor 594. To avoid photobleaching, the specimen was exposed to illumination only while taking images. Image Pro plus version 6 with Scope Pro version 6 (Media Cybernetics, Inc) was used for acquiring and analyzing images. We obtained fluorescence intensities for each sample by analyzing a 1.64 mm 2.19 mm image section of each substrate. Background images were obtained from an uncoated substrate, and unmodified glass cover slips at the same conditions. Images from the e xperimental samples were corrected for uneven illumination with the help of these background images. Image of nanoparticles coated glass coverslip was captured and compared with image of bare glass coverslip to test for Blocking agent (1% BSA) Alexa Fluor Anti Mouse Immunoglobulin G Mouse Immunoglobulin G Nanoparticles

PAGE 107

93 the possibility of scattered light f rom metal particles. These images showed that the emission filters effectively removed the scattered light so its contribution is negligible. Luminescence intensity of each sample was determined by measuring the mean intensity and subtracting the mean va lue of the background image. 5.3. Result s and D iscussion Transmission electron microscopy (FEI Tecnai F20 S Twin TEM) of the Ag Cu alloy nanoparticles indicated the average size to be 14.77 nm 5.4 nm (derived from a popula tion of 100 particles) (Figure 54 ( A ) ) and after annealing these nanoparticles at 448 K, the average size is 13.88 4.07 nm and the average size of Ag nanoparticles was 13.78 3.12 nm. From the HRTEM image (Figure 5 4 (B ), the lattice spacing was measured to be 0.210.24 nm. In the {111} latt ice plane, silver has lattice spacing of 0.24 nm whereas the lattice spacing of Cu is 0.21 nm182. This, combined with STEM EDS data (Figure 55), confirms that the nanoparticles are comprised of both Ag and Cu. In these Ag Cu nanoparticles, silver and co pper remain phase separated113 From the TEM EDS data, approximate composition of the Ag Cu nanoparticles was found to be 1:1.

PAGE 108

94 Figure 5-3 High resolution TEM image of Ag nanoparticles

PAGE 109

95 Figure 5-4 (A) -(B) HRTEM image of Ag-Cu 2.13 nm {111} (A ) (B )

PAGE 110

96 Figure 5-5 STEM EDS spectra for Ag -Cu alloy nanoparticles The absorbance spectra (taken using UV Vis spectrophotometer, JASCO, V 530), attributed to SPR of Ag Cu nanoparticles, show a single peak in the visible range. Wit h increasing copper percentage, this SPR peak shifts to higher wavelengths and becomes b roader This result confirms that the nanoparticles are a bimeta llic form of silver and copper and not a mixture of silver nanoparticles and copper nanoparticles166. The red shifts of the SPR peaks with increasing copper concentration are attributed to the decrease in conductivity101 The SPR peak of Ag Cu alloy nanoparticles blue shifts upon increasing the annealing temperature from 298 K to 523 K. With increase in annealing temperature, Cu atoms surfacesegregate, thereby increasing the concentration of Ag in the nanoparticle core as a result, and the SPR peak gradually moves nearer to the SPR peak of pure Ag nanoparticles (Figure 56).

PAGE 111

97 Figure 5-6 Absorption spectra of Annealed Ag -Cu nanoparticles (surface ratio of Cu in sputter target is 7.5%) T his can be explained as follows: for the coreshell structure, the effective dielectric constant is a function of t he dielectric constant of both core and shell materials and also the volume fraction of shell layer. The SPR absorbance spectrum peak, which can be calculated from the imaginary part of polarizability, a function of effective dielectric constant, will be nearer to that of core material for shell layer volume fraction up to approximately 0.6. D etailed calculations based on equations given in literature are shown below. The extinction coefficient of well dispersed small particles is proportional to where can be calculated from the following equation ( 51) 183 where s and c are the dielectric constants of core and shell materials respectively, R is the radius of nanosphere, m 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 350 450 550 650 750 850 950 1050 1150 300 K 423 K 448 K 473 K 523 K is the dielectric constant of medium and g is the volume Wavelength Absorbance (arb. unit)

PAGE 112

98 fraction of shell layer. Based on the above equation, extiction spectra is calculated for 20 nm AgCu core shell nanosphere (Figure 57). Figure 5-7 Calculated extinction spectra for the Ag -Cu core -shell (Ag in core and Cu in shell) materials at different shell layer thickness. Luminescence intensity of both Alexa Fluor 594 and Alexa Fluor 488 was observed to increase si gn ificantly at the vicinity of these Ag Cu nanoparticles (Figure 59 D and 59 F ). Enhancement ratio was calculated by comparing luminescence intensity of the sample with the luminescence intensity of the luminophore coated on a 3 (aminopropyl)triethoxysilane (APS) coated glass cover slip. As shown in Figure 5 8, the SPR spectrum of the 448 K annealed AgCu nanoparticles nicely overlaps both th e excitation and emission spectra of Alexa Fluor 488. This annealed Ag Cu nanoparticle platform results in very strong enhancement (141 19 times) of luminescen ce of Alexa Fluor 488 (Figures 59 C and 59 D ). The Ag Cu nanoparticles annealed at 298 K, whi ch show less spectral overlap, also result in substantial enhancement (100 10 times). The lowest enhancement (50 11) was observed at the proximity of pure Ag nanoparticles 0 20 40 60 80 100 120 140 160 400 450 500 550 600 650 700 g=0 g=0.2 g=0.4 g=0.5 g=0.7 g=0.9 g=1 Wavelength (nm)

PAGE 113

99 (deposited at the same conditions as Ag Cu nanoparticles), for which the spectral overlap is least. The effect of spectral overlap on luminescence enhancement is also pronounced for Alexafluor 594. We found 23 12 times enhancement of emission from Alexa Fluor 594 at the proximity of room temperature grown AgCu nanoparticl es (Figures 59 E and 2 F ). On the other hand, both pure Ag nanoparticles and the 448 K annealed Ag Cu nanoparticles grown at similar conditions result in lower enhancements (9 1 times for 448 K annealed AgCu nanoparticles and 6 3 times for Ag particles) because of less spectral overlap. The best case Ag Cu studied was 2.8 times better than pure Ag for Alexafluor 488 and 3.5 times better for Alexafluor 594. In both cases, the spectral overlap was largest when maximum enhancement was seen. It is possibl e to achieve this enhancement for the alloy particles because the breadth of the peak can also be tuned. Please note the average intensity of luminophore s coated on glass slide is n ear to 0 but not 0 (around 150 in the scale shown in Figure 59) Figure 5-8 SPR spectrum of Ag -Cu and Ag nanoparticles used for MEL and excitation and emission spectrum of Alexa Fluor 594 and Alexa Fluor 488. 0 0.2 0.4 0.6 0.8 1 1.2 350 450 550 650 750 298 K Ag-Cu 448 K Ag-Cu 298 K Ag Absorbance/Photoluminescence (arb.unit) Wavelength (nm) Alexafluor 594 Alexafluor 488

PAGE 114

100 Figure 5-9 I mage of Alexa Fluor 488 coated on (A) glass (B ) 448 K annealed Ag -Cu Alexa Fluor 594 coated on (C) glass (D) 298 K Ag -Cu. Possible differences in protein binding to glass and Ag and AgCu nanoparticles may lead to increased fluorescence signals. The difference in protein binding may arise due to the difference in hydrophobicity, as protein adsorption increases with hydrophobicity of surface when factors like electrostatic and hy drogen bonding are not pronounced .1 84 Where Ag and Cu surfaces are usually hydrophobic in nature, the oxide layers formed on these usually reduce their hydrophobicity.185 The glass slides were coated with APS to promote their hydrophobicity. Higher surface area available for nanoparticles also can increase the protein adsorption. In this work, both the glass and the nanoparticles samples were coated with ver y less concentration of protein (7 nanogram/mm2) and sufficient time was allowed for t he absorption of the protein. This enhances the possibility of complete immobilization of the protein on the surface and may reduce the difference in amount of protein binding in glass and nanoparticles samples. In the present work the protocol described by Matveeva et al.177 has been A B C D 50 50 0 (a.u.) 1200 (a.u.)

PAGE 115

101 followed for coating fluorophore conjugated protein on both glass surface and nanoparticles surface. They reported that the protein binding to the Ag nanoparticles surface is approximately 20 30% better than the glass surfa ce.177 We expect the difference of amount of protein binding between nanoparticles surface and glass surface should be even lower in present case, as the glass surface was coated with hydrophobic APS.2By enhancing the local field for absorption and/or quantum yield due to radiative and non radiative decay rates, we can increase the intensity of luminescence. The intensity of the incident optical wave is enhanced in the near field of the nanoparticles a t the SPR wavelength. Hence, strongest excitation should be observed when the SPR spectrum of nanoparticles overlaps the excitation peak of the luminophore Hence, this small difference in protein binding it self cannot explain the large fluorescence enhancement observed on the Ag Cu and Ag surface. However, exact estimation of differences between the proteins adsorption between the glass and the nanoparticles surface, which is beyond the scope of this study, may facilitate the more accurate prediction of enhancement factor. 24. Same as for excitation, when SPR spectrum of the nanoparticles overlaps the emission spectrum o f luminophore, emission intensity enhancement should be the highest179 However, as in this case high quantum yield luminophores were used, excitation enhancement should be more pronounced than emission enhancement. As a result, the spectral overlap with excitation spectra should be more important.

PAGE 116

102 Table 5-1 Fluorescence enhancements of Alexa Fluor 488 and Alexa Fluor 594 on the Ag and Ag -Cu nanoparticles. Size SPR peak Enhancement ratio for Alexa Fluor 488 Enhancement ratio for Alexa Fluor 594 Ag nanoparticles annealed at 298 K 13.78 3.12 nm 444 nm 50 11 6 3 Ag Cu bimetallic nanoparticles annealed at 298 K 14.77 5.4 nm 631 nm 101 10 24 12 Ag Cu bimetallic nanoparticles annealed at 448 K 13.88 4.07 nm 486 nm 142 19 10 1 Here, we present a theoretical calculation for overall quantum efficiency factors in the proximity of pure Ag nanoparticles and for the 1:1 AgCu nanoparticles, based on the model suggested by Kmmerlen et al. 33 which includes both excitation and emissio n enhancement factors (detailed computational methodology is given in the supplementary information). Exact representation of experimental conditions is not possible in theoretical calculations due to the differences in experimental geometry (nanoparticle s are not in a homogeneous dielectric environment, all the nanoparticles are not of spherical shape and not of the same size, luminophorenanostructures separation distance is not uniformly the same). Furthermore, accurate dielectric constants of room tem perature and annealed Ag Cu nanoparticles are not known, or evaluable, as they remain phase separated. However, these calculations provide some insights into the experimental findings.

PAGE 117

103 Figure 5-10 C alculated extinction coefficient (black) and overall quantum efficiency enhancement ratio for (blue) Ag (dotted line) and 1:1 Ag-Cu nanospheres (solid line) Figure 510 shows the theoretically calculated extinction coefficient (using Mie theory), and the overall quantum efficiency enhancement factor for pure Ag and 1:1 bimetallic Ag Cu nanoparticles. From Figure 5 10, the effect of spectral overlap is clearly evident. In the wavelength range of 450 nm to 555 nm, as the extinction spectrum for the Ag nanoparticles is more pronounced, overall quantum efficiency enhancement in the proximity of the Ag is better than that of the Ag Cu nanoparticles. Most importantly, in the wavelength range of 555 nm to 605 n m, the Ag Cu nanoparticles show better overall quantum efficiency enhancement than pure Ag as the spectral overlap is better for the Ag Cu nanoparticles. For both Ag nanoparticles and Ag Cu nanoparticles, the maximum overall quantum efficiency enhancement wavelengths are slightly red shifted with respect to the extinction coefficient peaks. As the calculations were done for the high quantum yield (0.5) luminophore, the excitation enhancement effect is more pronounced than the emission enhancement effect T he theoretically calculated emission 0 2 4 6 8 10 12 14 16 18 0 1 2 3 4 5 6 7 8 9 10 400 450 500 550 600 650 Calculated extinction coefficient Calculated overall quantum e fficiency enhancement factor

PAGE 118

104 enhancement factor and excitation enhancement factor are separately shown in Figure 511. These theoretical findings help in interpreting our experimental observations. Figure 5-11 C alculated extinction c oefficient (black), E mission enhancement fa ctor (green) and excitation enhancement factor (red) for Ag (dotted line) and 1:1 Ag-Cu nanospheres (solid line) 5.4. Conclusions In summary, the MEL effect of Ag Cu alloy nanoparticles has been explored in this work. AgCu alloy nanoparticle platforms were found to produce strong enhancement for the two luminophores studied, viz. Alexa Fluor 488 and Alexa Fluor 594. The effect of spectral overlap on luminescence is explored in this work. A synthesis technique to tune the SPR spectrum of alloy nanoparticles from infrared to visible region very easily by changing composition or annealing schedule is presented. Ag Cu alloy nanoparticles were observed to show even better enhancement than pure Ag nanoparticles when the SPR spectrum was tuned to result in maximum spectral overlap. For a particular luminophore, we can tune the annealing temperature of particular 0 5 10 15 20 25 0 1 2 3 4 5 6 7 8 9 10 400 500 600 700 Calculated excitati on enhancement factor Calculated extinction / emission enhancement factor Wavelength (nm)

PAGE 119

105 composition Ag Cu nanoparticles to result in maximum spectral overlap which can help in optimum luminescence enhancement. We expect our study to motivate exploration of other alloy nanoparticles for MEL based applications.

PAGE 120

106 Chapter 6 Quenching of Fluorescence from CdSe/ ZnS N anocryst als near Copper Nanoparticles in Aqueous S olution 6.1. Introduction The emission of luminescent probes is modified signi ficantly at the close proximity of metal surfaces and nanoparticles. Using nanoparticles, it is possible to both enhance and quench the emission of luminescent probes. Luminescence quenching by metal nanoparticles has been studied mostly using gold nanoparticles .43,55,5759 Gold nanoparticles can show both static and dyna mic quenching effect.189 The gold nanoparticles can quench the fluorescence of different flurophores due to different reasons like resonance energy transfer, formation of static quenching complex and internal electron transfer.189 Luminescence quenching due to F rster resonance energy transfer (FRET) from the excited state of the luminophore molecule (donor) to the surface plasmons of the metal nanoparticles (acceptor) depends on the spectral overlap of the acceptors absorption with the donors emission, and sensitivity depends on the separation distance between acceptor and donor.55 Quenching effect due to Frster energy transfer decreases with the cube of separation distance.56 The quenching effect of metal nanoparticles due to resonance energy transfer is decided by several properties of the nanoparticles like dielectric constant, size and shape.

PAGE 121

107 Q uenching of luminescence due to the proximity of nanoparticles has been utilized for many different applications. Q uenching result ing due to metallic nanopar ticles has been successfully utilized for the improvement of homogeneous and competitive fluorescence immunoassay 156,157, optical detection of DNA hybridization 158, competitive hybridization assay 159 and in optoelectronics 160. Recently, many researchers have utilized the quenching effect of gold nanoparticles on nanocrystal quantum dots for biological and solar cell applications.184186I maginary component of the dielectric constant of copper is comparable to that of gold in the wavelength range of 400 nm to 500 nm, and almost twice in the wavelength range of 500 nm to 625 nm Hence, it is expected that Cu nanoparticles will show similar or better quenching effects in comparison to gold nanoparticles in these wavelength ranges due to ohmic losses. In our previous study we reported quenching of luminophore Cy3 in the vicinity of Cu nanoparticles platform. However, the quenching effect of Cu nanoparticles on fluorophores in solution is yet to be explored. Details of the quenching mechanism are also not fully understood. The observation that Cu nanoparticles efficiently quench the emission from the fluorophore suggests that Cu nanoparticles might serve as efficient quencher of different other luminophore. We study the quenching effect s of Cu nanoparticles on the fluorescence emission from different sizes of CdSe/ ZnS nanocrystals a commonly used quantum dot in biological applications. We o bserve the effect of Cu nanoparticle concentration on quenching. However the widespread application of luminescence quenching requires exploration of cheaper metals In this work, to understand the quenching mechanism, we have studied both static and dynamic quenching effects of Cu nanoparticles. Two sets of fluorescence quenching

PAGE 122

108 experiments were perfo rmed. In the first set of experiments we have studied the dynamic quenching effect of Cu nanoparticles on the emission of CdSe/ ZnS nanocrystals. For this dynamic quenching study, Cu nanoparticles coated with PVP were synthesized. These Cu nanoparticles have no functional binding sites to bind with the mercaptoundecanoic ligands coated CdSe/ ZnS nanocrystals thus the quenching should be purely collisional quenching. In the second set of experiments, we have studied the effect of different size CTAB coated Cu nanoparticles on the luminescence of mercaptoundecanoic ligands coated CdSe/ ZnS nanocrystals. In this case, electrostatic binding between cationic Cu nanoparticles and anionic CdSe/ ZnS nanocrystals is possible, thus can result in static quenching. C u nanoparticles of variable sizes have been studied to observe the effect of size on their quenching effect on luminophores. There are few studies to see the effect of gold nanoparticles size.59,189 These studies are suggestive but in some case provide con tradictory information. For example Ghosh et al.59 studied the quenching effect of gold nanoparticles of size ranging from 8 nm to 73 nm s and suggested that with the increase in nanoparticles size the quenching effect reduces. On the other hand, Cheng et al.189 have observed the opposite effect for Au nanoparticles having core diameters from 1.3 to 3.9 nm s on the luminophore [Ru(bpy)3]2+. They found quenching eff e ct increases with the increase in size. Dulkeith et al.55found a sizedependent increase in n onradiative decay rate and a decrease in the radiative rate in case of the quenching of lissamine dye attached to a Au nanoparticle. In our study, we have used the theoretical calculation based on improved Gersten Nitzan model to provide better insight in to the size dependence of quenching by metallic nanoparticles. The Gersten Nitzan ( GN) model

PAGE 123

109 93,95,96 provide s insight s i n to the influence of proximal metal nanospheres on radiative and nonradiative decay rates of luminophore molecules The GN model can be used to interpret both luminescence enhancement and quenching effects of metallic nanoparticles.55,93 According to this model, the electromagnetic interaction between luminophore s and metal nano particles result s in the increase of both radiative and n on radiative decay rates, depending upon the luminophore nanoparticle separation distance and the properties of the nanoparticle (size, shape and dielectric constant) which decide the scattering and surface plasmon resonance behavior of the nano particle Mertens et al. 93,96 have corrected the GN model to account for radiation d a mping and dynamic depolarization 6.2. Experimental 6.2.1. PVP Coated Cu Nanoparticles S ynthesis S table Cu nanoparticle colloid solution s w ere synthesized using the process described by Wu et al. 1 90 Copper (II) acetate hydrate (98%) and polyvinylpyrrolidone (PVP of molecular weight 55, 000) were obtained from SigmaAldrich and used as received. An aqueous solution of 0.8 M PVP and 0.4 M L ascorbic acid (reagent grade, fine crystal, F isher Scientific) were mixed with an aqueous solution of 0.01 M copper salt and 0.8 M PVP in 1: 1 volume ratio under constant stirring at 45 0C without any inert gas protection. The reaction was then allowed to continue for 1 hour before bringing the system down to room temperature T he initial precursor solution of light blue color

PAGE 124

110 changes to a red colloidal slurry. Then, the solution was dil uted with ethanol and centrifuged. The supernatant was rejected to remove excess PVP, unconverted salts and by products. This centrifugation was repeated 4 times and the precipitated red Cu nanoparticles we re collected and dispersed in d ionized water at room temperature. 6.2.2. CTAB Coated Cu Nanoparticle S ynthesis CTAB coated Cu nanoparticles were synthesized using a method described in literature.1 91 Hydrazine, cupric chloride and cetyltrimethylammonium b romide ( CTAB) were obtained from SigmaAldrich. Equal volume of two aqueous solutions of CTAB, one containing hydrazine (.02.04 M) and other containing cupric chloride (.001 M ) we re mixed together at room temperature. The pH of cupric chloride and CTAB solution required to be maintained at 10 to avoid the oxidation of Cu nanoparticles. NH4 OH was added to this solution to maintain the pH. Cu nanoparticles synthesis completed after about 2 hours. By varying the concentration of hydrazine different size Cu nanoparticles were obtained. 6.2.3. Nanoparticles C haracterization Shape and size of the nano particles was observed and characterized by transmission electron microscopy (FEI Morgai 268D) An UV vis spectrometer (JASCO, V 530) was used for measuring the light absorption spectra attributed to the SPR of th ese nanoparticles. TEM samples were prepared by dispersing a few drops of the Cu nanoparticle solution on a carbon film supported by molybdenum grids.

PAGE 125

111 6.2.4. Fluorescence Quenching E xperiment Three different CdSe/ ZnS nanocrystals coated with mercaptoundecanoic ligands (green, orange and red) were purchased from NNLabs Fayetteville, AR The solution of Cu nanoparticles was added to the 500 nanomol solution of nanocrystals in a required mole ratio and the spectral ch anges were monitored immediately. Fluorescence spectra of the samples were recorded on an ISS PC1 photon counting spectrofluorimeter. 6.3. Result s and D iscussion 6.3.1. Characterization of PVP Coated Copper Nanoparticles At the time of synthesis of the Cu nanopar ticles ascorbic acid serves as both reducing agent and antioxidant to reduce copper salt precursor and prevent further oxidation of synthesized Cu nanoparticles. In the aqueous solution, the absorbance peak of the copper nanoparticles is around 588 nm. Intensity and position of this absorbance peak for Cu nanoparticles in aqueous solution did not show any significant change for at least 5 days which indicates th at these nanoparticles are stable. Transmission electron microscopy of the Cu nanoparticles i ndicated the average size to be 1 0. 11 3.6 nm (Figure 61 (A)) and the average size of red CdSe/ ZnS nanocrystals was determined to be 7.63 0.83 nm (Figure 61( B )). STEM EDS data (Figure 61 ( C )), confirm that the nanoparticles are comprised of only Cu and oxidation is negligible

PAGE 126

112 6.3.2. Collisional Quenching by PVP Coated Copper N anoparticles These Cu nanoparticles have no functional binding sites to bind with CdSe/ ZnS nanocrystals with mercaptoundecanoic ligands so the quenching should be collisional quen ching. The CdSe / ZnS nanocrystals were mixed with the PVP stabilized copper nanoparticles in aqueous solution. Figure 6-1 (A) H RTEM micrograph of (A) Cu nanoparticles (B) Red CdSe/ ZnS nanocrystals, and (C) STEM EDS spectra of the Cu nanoparticles. Figure 62 shows the absorbance spectr um attributed to SPR of Cu nanoparticles and the emission spectra of green, red and yellow CdSe/ ZnS nanocrystals. The absorbance spectrum of Cu nanopar ticles overlaps only with the emission spectr um of yellow CdSe/ ZnS nanocrystals. Fluorescence from the yellow nanocrystals displayed significant quenching upon conjugation with copper nanoparticles. (A ) (B ) (C)

PAGE 127

113 Figure 6-2 Normalized absorbance spectrum of copper nanoparticles and luminescence spectra of CdSe/ ZnS nanocrystals. 500 550 600 650 700 0 0.2 0.4 0.6 0.8 1 1.2 Absorbance/luminescence Wavelength Em ission s pect rum green nc Em ission s pectrum yellow nc Em ission spectrum red nc E xtinction spectr um cu np

PAGE 128

114 Figure 6-3 (A) The emission spectra of yellow nc at different concentration of Cu n ano p articles ( B) Q uenching efficienc y measured at 580 nm. The se experimental results revealed that quenching is sensitive to nanomol range concentration of copper nanoparticles (Figure 63(A) ). The emission wavelength rem ained the same but the intensity of CdSe/ ZnS nanocrystals decreases with the 550 560 570 580 590 600 0 2 4 6 8 10 x 104 0 5 15 25 50 100 125 150 200 250 300 0 100 200 300 400 500 0 0.1 0.2 0.3 0.4 0.5 0.6 Concentration of copper in nanomols Emission intensity Conc. of copper (nanomol) Wavelength (nm) (A) Quenching efficiency Experimental data ( B ) Wavelength (nm)

PAGE 129

115 concentration of copper nanoparticles. The quenching efficiency of yellow nanocrystals estimated on the basis of emission intensity and shown in Figure 63(A) and Figure 63(B) show s a significant increase and reach es an asymptotic value at the nanoparticle concentration of 300 nM. Most of the nanocrystals have been quenched at a molar ratio of metal nanoparticles/yellow nanocrystals of 0.6. Interestingly, in the range of 0 to 250 nanomols quenching efficiency of yellow CdSe/ ZnS nanocrystals show s almost linear behavior with copper concentration. Figure 6-4 Stern -Volmer plot of I I0 for 5 00 nanomolar concentr ation of CdSe/ ZnS nanocrystals vs. concentration of copper nanoparticles Figure 64 shows t hat the fluorescence quenching of CdSe/ ZnS nanocrystals follow s the linearity of Stern Volmer (SV) equation, Q QC K I I 10 ( 61) Where I0 QC is the luminescence intensity in the absence of quencher molecules, Q represents the quencher and is the concentration of quencher molecules. Q K 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 0 50 100 150 200 250 300 is the II0 Concentration of copper nanoparticles in nanomols at wavelength 580 nm wavelength 629 nm red at 531nm

PAGE 130

116 Stern Volmer constant This can be attributed to the fact that at dilute acceptor concentration, Forster kinetics approach the SV limit .192 We can also see form Figure 64 that the SV constant is hig her for yellow nanocrystals (compared to those of the red and green nanocrystals ) for which the emission spectrum has maximum overlap with the absorption spectrum of Cu nanoparticles This phenomenon is consistent with the FRET theory that the fluorescenc e quenching efficiency increases with the increase in spectral overlap of the donors emission with the acceptors absorption. 6.3.3. Characterization of CTAB Coated Cu N anoparticles Three different size na noparticles samples a, b and c were synthesized varying the concentration of hydrazine. With the increase in concentration of hydrazine the size of copper nanoparticles decreased. Figure 65 (A C) shows the TEM micrographs of these three different size nanoparticles. From the TEM images of the CTAB coated Cu nanoparticles the average size of these nanoparticles were obtained. The absorbance spectra, attributed to SPR of different sizes C u nanoparticles are given in figure 66.

PAGE 131

117 Figure 6-5 High resolution TEM images of different sizes CTAB coated copper nanoparticles (A) sample a (B) sample b and (C) sample c Figur e 6-6 Normalized absorbance spectra of different size Cu nanoparticles in aqueous solution. 0 0.2 0.4 0.6 0.8 1 1.2 400 500 600 700 800 a b c Wavelength (nm) Absorbance (arb. unit) (A) (B) (C)

PAGE 132

118 Table 6.1 summarizes the concentration of precursor and the reducing agents used for synthesizing samples a, b and c and the resulting sizes of nanoparticles and their absorption peaks With the increase in Cu nanoparticle size the absorbance peak redshifts. Typical absorption peak for copper oxide around 800 nm is not seen confirming the negligible formation of copper oxide.193 Table 6-1 Concentration of reactants and characteristics of the synthesized Cu nanopar ticles These copper nanoparticles w ere stable at least for 3 days. Diluting these Cu nanoparticles solution also does not oxidize the nanoparticles only the absorption intensity decreases (Figure 6 7). Samples [CuCl 2 [N ] (mole) 2 H 5 OH] (mole) [CTAB] (mole) Size of nanoparticles (nm) Absorbance peak (nm) A .001 .02 .01 6.83+/ 1.00 592 B .001 .03 .01 5.58+/ 1.16 588 C .001 .04 .01 3.71+/ 1.00 5 74

PAGE 133

119 Figure 6-7 Absorbance spectra of 500 micromol and diluted (1 micromol) copper nanoparticles. 6.3.4. Quenching Effect of CTAB Coated Cu Nanoparticles on CdSe/ ZnS N anocrystals Luminescence of red CdSe/ ZnS nanocrystals in aqueous solution quenches in the presence of CTAB coated Cu nanoparticles ( F igure 6 8). The emission intensity of 500 nanomol red CdSe/ ZnS nanocrystals solution decreases with increasing concentration of Cu nanoparticles, however the peak position of emission spectra remain same. This quenching effect is sensitive to nanomol concentration of Cu nanoparticles. The fluoresc ence spectra shown i n F igure 68 is solely that of CdSe/ ZnS nanocrystals as Cu nanoparticles do not show any luminescence. 0 0.2 0.4 0.6 0.8 1 1.2 400 500 600 700 800 900 500 micromol copper nanoparticles (sample a ) 1 micromol copper nanoparticles (sample a ) Wavelength (nm) Absorbance (arb. unit)

PAGE 134

120 Figure 6-8 Effect of sample a sample b and sample c copper nanoparticles concentration on the 500 nanomol red CdSe/ ZnS nanocrystals. 0 100000 200000 300000 400000 500000 600000 600 610 620 630 640 650 Cu np conc (nanomol) 0 10 20 30 40 50 60 70 80 90 125 175 200 0 100000 200000 300000 400000 500000 600000 600 610 620 630 640 650 Cu np conc (nanomol) 0 10 20 30 40 50 60 70 80 90 100 125 150 175 200 0 100000 200000 300000 400000 500000 600000 600 610 620 630 640 650 Cu np conc (nanomol) 0 10 20 30 40 50 60 70 80 90 125 175 200 Sample c Sample b Emission intensity Emission intensity Wavelength (nm) Sample a Emission intensity Wavelength (nm) Wavelength (nm)

PAGE 135

121 Figure 68 shows the I0/I plot for red CdSe/ ZnS nanocrystals for different concentrati on fo r sample a, b and c of Cu nanoparticles. In all the s e c a se s the quenching effect does not follow the linearity of Stern Volmer plot. I0 2 0. ) ( 1Q D S Q D SC K K C K K I I /I vs. Cu nanoparticle concentration plots show an upward curvature towards the y axis indicating the quenching may be due to the combination of both collisional and static quenching. In this case the modified SV equation is given below ( 62) w here Cq is the concentration of quencher elements and Ks and Kd are static and dynamic quenching constants respectively. The above equation is fitted to the I0/I vs. Cu nanoparticles concentration data for different size Cu nanoparticles (Figure 69). KD and KS are obtained from this fitted equation. The lower value is assied to the dynamic quenching constant as probability of static quenching due to formation of electrostatic complex is more than probability of dynamic quenching. A s econd set of experiments (discussed in later sections) which deal with only dynamic quenching also give the dynamic quenching constant of same o rder. Quenching constants for different size Cu n anoparticles are summarized in T able 6 2.

PAGE 136

122 Figure 6-9 S V plot for 500 nanomolar concentration of red CdSe/ ZnS nanocrystals for different size Cu nanoparticles Table 6-2 Summary of SV equation and quenching constants for different size CTAB coated Cu nanoparticles Samples SV equation K ( / nanom ol conc .) D K ( / nanomo l conc ) S Relative K (/ Cu n p number D Relative K (/ Cu n p number ) S A 0.0006 0.01364 1.00 1 B 0.009 0.033288 8.18 0.75 C 0.006 0.06813 1.60 1.2 5 0 5 10 15 20 25 30 35 0 50 100 150 200 250 Concentration of copper in nanomols I 0 /I

PAGE 137

123 6.3.5. Quenching Mechanism and Effect of Size of Cu Nanoparticles on Quenching Efficiency The quenching of luminescence of CdSe/ ZnS by CTAB coated Cu nanoparticles may be due to the combined effect of resonance energy transfer and the formation of static quenching complexes via attractive electrostatic inte ractions. Resonance energy transfer from luminophores to nanoparticles requires good overlap between the emission and excitation spectra of luminophores and the absorbance spectra of nanoparticles. It is seen that in case of pure dynamic quenching (quenchi ng of CdSe/ ZnS nanocrystals by PVP coated Cu nanoparticles) the effect of spectral overlap is pronounced on the quenching efficiency confirming the possibility of resonance energy transfer. In the case of this static quenching experiment, though there is less spectral overlap between emission spectrum of red CdSe/ ZnS nanocrystals and the absorbance spectra of the CTAB coated Cu nanoparticles is not large enough large spectral overlap between the excitation spectrum of red CdSe/ ZnS nanocrystals and sample C of Cu nanoparticles (showing maximum quenching effect) exists (Figure 6 9)

PAGE 138

124 Figure 6-10 Normalized a bsorbance spectra (------) of different size Cu nanoparticles and the excitation (. ) and emission spectra (.) red CdSe/ ZnS n c. The results in Tables 62 and the quenching efficiency vs, Cu nanoparticles diameter plotted in F igure 611 show interesting effects of nanoparticle size on quenching. Where static quenching constants increases with the decrease in size of Cu nanoparticles, dynamic quenching constants first increase then decrease with increase in size. As Energy transfer is the most likely dominant mode of quenching in these experiments t he presence of nanoparticles not only influences the nonradiative decay rate of vicinal luminophores due to Frster energy transfer (from luminophore molecules to nanoparticles), but also affects the radiative decay rate.55 This observation can be exp lained based on the calculation using improved GN model. Using the improved model, 93,95,96 R we calculated the modifications of the radiative decay rate ( ) and total decay rate ( Tot ) of the luminophores at the proximity of metal nanoparticles. The corrected GN model was used to calculate quantum efficiency change due to radiative 0 0 2 0 4 0 6 0.8 1 1.2 450 550 650 750 850 Wavelength (nm) Absorbance/luminescence

PAGE 139

125 and nonradiative decay rate change. The calculations were done assuming the separation distance between nanoparticles and luminophore molecules to be 2.7 nm the length of CTAB molecule as suggested in literature192. Theoretically calculated quenching of quantum efficiency of luminophore molecule due to Cu nanosphere is plotted against the size of the nan osphere in F igure 68. It can be seen from Figure 6 8 that there is an optimum size of nanoparticles for which quantum efficiency quenching is maximum. Spectral overlap between the absorption and emission spectra of luminophore and surface plasmon reso nance spectra of metal nanoparticles is very important for resonance energy transfer24,154,168. When the size of the particle increases, the plasmon resonance is shifted to longer wavelength and broadened and decreases in magnitude due to dynamic polarizat ion170 So, there exists an optimum diameter. Below this optimum diameter, the quenching efficiency should increase with increase in diameter and above this diameter the quenching efficiency should decrease with decrease in diameter. This explains our e xperimental finding that the static quenching coefficient decreases with the increase in diameter. In case of dynamic quenching, collision probability between the luminophore molecule and the nanoparticles also is an important factor. The collision proba bility between the nanoparticle and luminophore increases with the increase in size of nanoparticles. Since dynamic quenching efficiency depends on both effective coupling to the plasmon mode and also collisional efficiency, give rise to a n optimum diamete r

PAGE 140

126 Figure 6-11 Relative dynamic quenching constants (KD) ( ) and static quenching constants (KS ) ( vs the mean diameter of Cu nanoparticles Figure 6-12 Ratio of theoretically calculated luminescence quantum yields of a dipole emitter with and without copper metal nano sphere 6.4. Summary and Conclusion s Quenching effect of Cu nanoparticles on CdSe/ ZnS nanocrystal quantum dots in aqueous solution has been explored in this work. Cu nanoparticle s were found to 0 1 2 3 4 5 6 7 8 9 10 1 3 5 7 9 0 50 100 150 200 250 300 350 400 0 5 10 15 20 25 Diameter of nanosphere (nm) Theoretically calculated quantum efficiency quenching Diameter of nanosphere (nm) Quenching coefficients

PAGE 141

127 produce quenching for th ree different CdSe/ ZnS nanocrystals (red, yellow and green) The luminescence of nanocrystals is sensitive to nanomolar concentrations of copper nanoparticles. Cu nanoparticles were observed to show better quenching effect when maximum spectral overlap between emission spectrum of nanocrystals and absorption spectrum of copper nanoparticles occurs suggesting quenching may b e due to the resonance energy transfer from luminophore to Cu nanoparticles This study also provides insight into the dependence of fluorescence quenching efficiency on the size of metallic nanoparticles. In this case static quenching constants were foun d to decrease with the i ncrease in size of nanoparticles however dynamic quenching constant did not show any definite trend. We used theoretical calculations based on the corrected GN model to explain our findings. These results on the quenching effect o f copper nanoparticles will motivate their utilization of these nanoparticles as an inexpensive alternative to gold in many quenching based applications

PAGE 142

128 Chapter 7 Summary, C onclusions and R ecommendations 7.1. Introduction The emission intensity of luminophore molecules can either be enhanced or quenched depending on the environment. Metal nanoparticles can influence the emission intensity of vicinal luminophores depending on different factors like their orientation with respect to luminophore mo lecules, luminophore molecule and nanoparticles separation distance and ohmic losses of metallic nanoparticles.1 3 This dissertation focused on both enhanced and q uenched luminescence. In the first part of this dissertation, based on the fluorescence quenching by O Both enhancement and quenching have important applications in biological and sensor field 2 molecule, we have developed a method to measure oxygen diffusion properties in polymer using inverted fluores cence microscopy. Then, we studied enhanced and quenched fluorescence in the vicinity of alloy nanoparticles. Finally we studied fluorescence quenching of CdSe/ ZnS nanocrystals in the presence of copper nanoparticles. In the following sections conclusions from these studies are presented. 7.2. Measurement of O2A fluorescence microscopy technique is developed to measure diffusion and permeation coefficients of oxygen in polymers. In this method the microscopic level S V Diffusion Properties Using Inverted Fluorescence Microscopy

PAGE 143

129 response s of het erogeneous sensors can be monitored This method allow s us to distinguish the responses of background region (nearly homogeneous regions) from the region where the luminophore is aggregated. As the nearly homogeneous regions show better response to oxygen concentration and follow the linearity of SV equation, studying the response of these, one can eliminate the complexity of combining non linear SV equation with a diffusion model. The method developed here can be applied for measuring oxygen diffusion properties in different polymers ranging from transparent to opaque and subjected to the condition that no component present in polymer interferes with the response of sensor to oxygen concentration and is also suitable for polymer composite. We also developed a new and simple quasi steady model for describing diffusion phenomena in the case of accumulation in volume technique, which can be applied for any other diffusion experiments. 7.3. Ag Cu N anoparticle s for Enhanced L uminescence In this part, we show that photoluminescence intensity can be enhanced in the vicinity to Ag Cu alloy nanoparticles. In the first case, different composition Ag Cu nanoparticles were synthesized by polyol synthesis method. The observed luminescence enhancement depends on the compos ition of A gCu nanoparticles. It was found that with the increase of Cu percentage the luminescence enhancement decreases and finally pure Cu nanoparticles quench the fluorescence. This is attributed to the fact that, t he imaginary component of the diele ctric constant of copper is si gn ificantly larger (more than twice) than that of silver in the wavelength range of 300 nm to 600 nm. I t is expected that in this wavelength range due to higher ohmic losses, Cu nanoparticles will

PAGE 144

130 mostly quench the luminesce nce at close proximity in contrast to the enhancement effect of Ag nanoparticles. In the second part, we synthesized Ag Cu nanoparticles using sputtering deposition and then tune their SPR spectra from visible to infra red region by annealing. This allows us to see the effect of SPR spectra of Ag Cu nanoparticles on the vicinal luminophores. We have found that with the spectral overlap between SPR spectra of nanoparticles and the emission and absorption spectra of luminophore s large metal enhanced lu mine scence is achieved ( order of 100). Interestingly, when the spectral overlap with Ag Cu nanoparticles is better, these nanoparticles show even better enhancement than pure Ag nanopart icles. This study establishes the importance of spectral overlap for met al enhanced luminescence. In both of the above cases the experimental findings are supported by the theoretical calculations using an improved Gersten Nitzan model. 7.4. Fluore scence Q uenc hing by Cu N anoparticles Cu nanoparticles were found to be efficient q uencher of fluorescence of CdSe/ ZnS quantum dots in aqueous solution. Cu nanoparticles can participate in both static and dynamic quenching and the nanomole concentration of the Cu nanoparticles can also r esult in quenching effect. It wa s found that the quenching efficiency of Cu nanoparticles depends on the spectral overlap between SPR spectra of Cu nanoparticles and excitation and emission spectra of quantum dots. This suggests that the fluorescence quenching by Cu nanoparticles may be due to resonance energy transfer from the quantum dots to Cu nanoparticles. Furthermore, it wa s found that the quenching effect

PAGE 145

131 by Cu nanoparticles si gn ificantly depends on the size of Cu nanoparticles. We hope our study will motivate the use of Cu nano particles in many fluorescence quenching based applications. 7.5. Major C ontributions The contributions of this dissertation to the field of luminescence sensor research are multifold. For the first time, the present work has explored the effect of alloy metal nanoparticles on the luminescence intensity of vicinal luminophores. This study finds that the tunable optical property of alloy nanoparticles sometime make them better candidates for metal enhanced luminescence in comparison to pure metal nanoparticles. This study also provides fundamental understanding of the effects of surface plasmon resonance properties of metal nanoparticles on metal enhanced luminescence. The outcome from the present research can be utilized to improve luminescence sensor design and produce sensors having enhanced si gn al to noise ratio, resolution and detection sensitivity. An o pportunity to enhance the luminescence of sensors is likely to improve a wealth of biomedical and biochemical application including single molecule detection, DNA sequencing, medical dia gn ostics, genomics. I mproved luminescence will also facilitate fabrication of improved emissive devices, such as lasers or organic light emitting di odes (OLEDs). The findings of this research are not only beneficial for metal enhanced luminescence applications, but also provide a good platform for the study of other SPR based applications. Finally, we have introduced Cu nanoparticles to quench the emission intensity of vicinal luminophores. The fluorescence of quantum dots is even sensitive to nanomol

PAGE 146

132 concentration of Cu nanoparticles. This finding should motivate the application of quenching effect of Cu nanoparticles in different biological sen sing methods. 7.6. Future D irections Based on the findings of the current research the following possibilities exist which could lead to many worthwhile and interesting projects. The details are discussed in this section 7.6.1. Fluorescence Microscopy for Simultaneous Imaging and O2Extension of f luorescence microscopy technique established in this work, to the measur ement of O Diffusion Measurement 2 diffusion coefficient in biological samples simultaneously with imaging will be a meritorious project to pur sue. This project is particularly interesting because of following reasons. Fluorescence microscopy is already very popular for imaging different biological samples like cell, tissue, microbes and biofilms. Understanding how these biological samples react to different concentration of oxygen is very essential to understand in some cases. For example, simultaneous monitoring molecular oxygen concentration and imaging of tissue is an important part of photodynamic therapy. Recent studies also address the s ignificance of oxygen concentration heterogeneities within a cell in health and disease.1 4 Simultaneous monitoring of oxygen concentration in microenvironment and their effect on metabolic activity of different microbial communities is also very important. However the

PAGE 147

133 technique for two dimensional monitoring of O2 concentration and imaging is yet to be fully established. 7.6.2. Exploration of Other A lloy Nanoparti cles for Metal Enhanced L uminescence We expect our study of metal enhanced luminescence by Ag Cu alloy nanoparticles will motivate further studies of other alloy nanoparticles for MEL based applications. For example, silver gold alloy nanoparticles can be an interesting candidate to study for MEL based application as silver gold alloy nanoparticles eliminate the oxidation problem of pure silver nanoparticles and their surface plasmon resonance property can be manipulated by tuning their composition. 7.6.3. App lication of Alloy N a noparticles for Enhancement of P hotovoltaic C ells Decreased absorbance of light and lower energy conversion efficiency are sometime major limitations of thin film solar cells for example amorphous silicon solar cells, GaAs solar cells a nd dye sensitized solar cells.5,6 Scattering from noble metal nanoparticles excited at their SPR and near field concentration of light can increase the light absorption and light trapping in the photovoltaic cell, thus can enhance the efficiency.7 Easy tunability of SPR wavelength of nanoparticles will be very design rable property of metallic nanoparticles for enhancing the efficiency photovoltaic cell. This proposed work can exploit the scientific principles of tunable SPR properties of AgCu alloy nanoparticles established in the present work for the ef ficiency enhancement of photovoltaic cells.

PAGE 148

134 7.6.4. Development of Sensors Based on the Quenching Property of Cu N anoparticles It was found in the present work that Cu nanoparticles can efficiently quench the fluorescence intensity of quantum dots and the quenc hing is nanomol concentration sensitive. This could be utilized to develop different biological sensors for detecting DNA hybridization and immunoassay. 7.6.5. Theoretical and Computational Modeling of Optical Properties of Alloy N anoparticles Theoretical investigation of SPR properties of alloy nanoparticles and their effect on vicinal luminophores require the exact knowledge of their exact dielectric constants. For the calculation of dielectric constants of alloy nanoparticles some semiempirical models developed on the basis of assumption of homogeneous distribution of metallic atoms in their alloys exist in literature However, there is one major limitation in applying this approach to AgCu nanoparticles. Ag Cu cannot form a solid solution at room temperature as does Ag Au. In Ag Cu nanoparticles, silver and copper remain phase separated.8 10 With increase in annealing temperature, Cu atoms surfacesegregate, thereby increasing the concentration of Ag in the nanoparticle core. So, the effect of metal segregation in the nano particles due to thermal annealing or from metallic interactions needs to be modeled. Knowledge of the atomic distribution profile in Ag Cu alloy nanoparticles simulated at different temperature using molecular dynamics can give useful insights to underst and the effect of annealing by computation of the atomic distribution profiles in the nanoparticles. This information about atomic distribution can

PAGE 149

135 be used to calculate accurate dielectric constant for the room temperature and annealed Ag Cu nanoparticles by constructing a statistical mechanical model

PAGE 150

136 Reference s 1 J. R. Lakowicz, Principles of Fluorescence Microscopy 2nd ed. (Plenum, NY, 1999). 2 K. Aslan, I gryczynski, J. Malicka, E. Matveeva, J. R. Lakowicz, and C. D. G eddes, Curr. Opin. Biotech. 55, 16 (2005). 3 K. Aslan, J. R. Lakowicz, and C. D. Geddes, Anal. Bioanal. Chem. 382, 926 (2005). 4 A. Yekta, Z. Masoumi, and M. A. Winnik, Can. J. Chem. 73, 2021 (1995). 5 J. N. Demas, B. A. Deora, and W. Xut, Anal. Chem. 6 7, 1377 (1995). 6 K. Aslan, Meng Wu, J. R. Lakowicz, and C. D. Geddes, J. Am. Chem. Soc. 129, 1524 (2007). 7 M. H. Chowdhury, K. Aslan, S. N. Malyn, J. R. Lakowicz, and C. D. Geddes, J. Fluoresc. 16, 295299 (2006). 8 L Wang, J Li, S Song, D Li, C Fan, J. Phys. D: Appl. Phys. 203001, 42 (2009) 9 C. D. Geddes and J. R. Lakowicz, J. Fluoresc. 12, 121 (2002). 10 C. D. Geddes, A. Parfenov, and J. R. Lakowicz, Appl. Spectrosc. 57, 526 ( 2003). 11 J. R. Lakowicz, J. Malicka, I. Gryczynski, Z. Gryczynski, and C. D. Geddes, J. Phys. D: Appl. Phys. 36, R240R249 (2003). 12 K. Ray, R. Badugu, and J. R. Lakowicz, J. Am. Chem. Soc. 128, 8998 (2006). 13 K. Ray, R. Badugu, and J. R. Lakowicz, Langmuir 22, 8374 (2006). 14 K. Ray, R. Badugu, and J. R. Lakowicz, J. Phys. Chem. B 110, 13499 (2006). 15 J. Zhang, Y. Fu, and J. R. Lakowicz, J. Phys. Chem. C 111, 1955 (2007).

PAGE 151

137 16 J. Zhang, E. Matveeva, I. Gryczynski, Z. Leonenko, and J. R. Lakowicz, J. Phys. Chem. B 109, 7969 ( 2005). 17 Y. Zhang, K. Aslan, M. J. R. Previte, and C. D. Geddes, Appl. Phys. Lett. 90, 173116 (2007). 18 Y. Zhang, A. Dragan, and C. D. Geddes, J. Phys. Chem. C in press (2009). 19 J. Malicka, I. Gryczynski, Z. Gryczynski, and J. R. Lakowicz, Anal. Biochem. 315, 57 (2003). 20 K. Sokolov, G Chumanov, and T. M. Cotton, Anal. Chem. 70, 3898 (1998). 21 S. Pan and L. J. Rothberg, J. Am. Chem. Soc. 127, 6087 ( 2005). 22 P. P. Pompa, L. Martiradonna, A. D. Torre, F. D. Sala, L. Manna, M. D. Vittorio, F. Calabi, R. Cingolani, and R. Rinaldi, Nat Nanotechnol. 1 ( 2006 ). 23 D. Cheng and Q. H. Xu, Chem. Commun., 248 (2007). 24 Y. Chen, K. Munechika, and D. S. Ginger, Nano Lett. 7, 690 (2007). 25 F. Tam, G. P. Goodrich, B. R. Johnson, and N. J. Halas, Nano Lett (2007). 26 M. Thomas, J. J. Greffet, and R. Carminati, Appl. Phys. Lett. 85, 3863 (2004). 27 K. Aslan and C. D. Geddes, J. Fluoresc. 16, 3 (2006). 28 K. Aslan, P. Holley, and C. D. Geddes, J. Immunol. Methods 312, 137 (2006). 29 K. Aslan, S. N. Malyn, and C. D. Geddes, J. Fluoresc 17, 7 (2007). 30 J. Azoulay, A. Dbarre, A. Richard, and P. Tchnio, Europhys. Lett. 51, 374 (2000). 31 P. Bharadwaj, P. Anger, and L. Novotny, Nanotechnology 4, 0957 (2007). 32 A. Ishida and T. Majima, Analyst 125, 535540 (2000). 33 J. Kmmerlen, A Leitner, H. A Brunner, F. R. A Aussenegg, and A Wokaun, Mol. Phys. 80, 1031 (1993). 34 E. G. Matveeva, I. Gryczynski, A. Barnett, Z. Leonenko, J. R. Lakowicz, and Z. Gryczynski, Anal. Biochem. 363, 239 (2007). 35 L. Shang, H. Chen, and S. Dong, J. Phys Chem. C 111, 10780 (2007).

PAGE 152

138 36 L. Yan, H. Liu, and H. Iwasaki, Chem. Phys. Lett. 433, 312 (2007). 37 Z. Yongxia, A. Kadir, J. R. P. Michael, and D. G. Chris, Appl. Phys. Lett. 90, 053107 (2007). 38 J. Zhang, Y. Fu, M. H. Chowdhury, and J. R. Lakowicz, Nano Lett. 7, 2101 (2007). 39 J. R. Lakowicz, Y. Shen, S. D'Auria, J. Malicka, J. Fang, Z. Gryczynski, and I. Gryczynski, Anal. Biochem. 301, 261 (2002). 40 K. Aslan, J. R. Lakowicz, and C. D. Geddes, J. Phys. Chem. B 109, 6247 (2005). 41 C. D. Geddes, A. Parfenov, D. Roll, I. Gryczynski, J. Malicka, and J. R. Lakowicz, J. Fluoresc. 13, 267 (2003). 42 C. D. Geddes, K. Aslan, I. Gryczynski, J. Malicka, and J. R. Lakowicz, Reviews in Fluorescence (Kluwer Academic/Plenum, New York, 2004), pp.365401. 43 E. Dulkeith, M. Ringler, T. A. Klar, J. Feldmann, A. MunozJavier, and W. J. Parak, Nano Lett. 5, 585 (2005). 44 K. Ray, M. H. Chowdhury, and J. R. Lakowicz, Anal. Chem. 79, 6480 (2007). 45 R. Pribik, K. Aslan, Y. Zhang and C D. Geddes, J. Phys. Chem. C 112, 17969 17973 (2008). 46 K. Aslan, M. J. R. Previte, Y. Zhang, and C. D. Geddes, J. Phys. Chem. C 112, 1836818375 (2008 ). 47 A. Dorfman, N. Kumar, and J.i. Hahm, Langmuir 22, 4890 (2006). 48 A. Dorfman, N. Kumar, and J.i. Hahm, Adv. Mater. 18, 2685 (2006). 49 N. Kumar, A. Dorfman, and J.i. Hahm, Nanotechnology 17, 28752881 (2006). 50 K. Aslan, M. J. R. Previte, Y. Zhang, and C. D. Geddes, J Phys Chem C 112, 18368 (2008). 51 P. J. Tarcha, J. DesajaGonzalez, S. Rodriguez Llorente, and R. Aroca, Appl. Spectrosc. 53, 43 (1999). 52 J. C. Hulteen and R. P. V. Duynea, J. Vac. Sci. Technol. A 13, 1553 (1995). 53 T. R. Jensen, M. L. Duval, K. L. Kelly, A. A. Lazarides, G. C. Schatz, and R. P. V. Duyne, J. Phys. Chem. B 103, 9846 (1999).

PAGE 153

139 54 T. R. Jensen, G. C. Schatz, and R. P. V. Duyne, J. Phys. Chem. B 103, 2394 (1999). 55 E. Dulkeith, A. C. Morteani, T. Niedereichholz, T. A. Klar, J. Feldmann, S. A. Levi, F. C. J. M. v. Veggel, D. N. Reinhoudt, M. Moller, and D. I. Gittins, Phys. Rev. Lett. 89, 203002 (2002). 56 A. Campion, A. R. Gallo, C. B. Harris, H. J. Robota, and P. M. Whitmore, Chem. Phys. Lett. 73, 447 (1980). 57 T. Huang and R. W. Murray, Langmuir 18, 7077 (2002). 58 N. N. Horimoto, K. Imura, and H. Okamo to, Chem. Phys. Lett. 467, 105109 (2008). 59 S. K. Ghosh, A. Pal, S. Kundu, S. Nath, and T. Pal, Chem. Phys. Lett. 395, 366 (2004). 60 D. J. Bergman, Phys Rev Lett 44, 1285 (1980). 61 R. M. Drake and J. E. Gordon, Am. J. Phys. 53, 955 (1985). 62 G. W. Milton, J. Appl. Phys. 52, 5266 (1981). 63 G. Mondio, F. Neri, and G. Currb, Physica Scripta. T41, 153 (1992). 64 S. B. Singham and C. F. Bohren, Optics Lett 12, 10 (1987). 65 G. Mie, Ann. Phys., Lpz 4 25, 377 (1908). 66 F. Wooten, Optical Properties of Solids 1st ed. (Academic Press, New York 1972). 67 L. Gao and J. Z. Gu, J. Phys. D: Appl. Phys. 35, 267 (2002). 68 H. Garcia, J. Trice, R. Kalyanaraman, and R. Sureshkumar, Phys. Rev. B 75, 0454391 (2007). 69 S. K. Mandal, R. K. Roy, and A. K. Pal, J. Phys. D: Appl. Phys. 36, 261265 ( 2003). 70 Z. Shaoli, L. Xiangang, D. Chunlei, L. Fei, Y. Shaoyun, D. Qiling, and F. Yongqi, J Appl Phys 101, 064701 (2007). 71 G. Smith, J. Phys. D: Appl. Phys. 10, L39 (1977).

PAGE 154

140 72 J. C. Maxwell Garnett Phil. Trans. R. Soc. London A 203, 385 (1904). 73 J. Chen, F. Saeki, B. J. Wiley, H. Cang, M. J. Cobb, Z. Y. Li, L. Au, H. Zhang, M. B. Kimmey, X. D. Li, and Y. Xia, Nano Lett. 5, 473 (2005). 74 E. Prodan and P. Nordlander, Nano Lett. 3, 543 (2003). 75 E. Prodan, P. Nordlander, and N. J. Halas, Nano Lett. 3, 1411 (2003). 76 E. Prodan, C. Radloff, N. J. Halas, and P. Nordlander, Science 302, 419 (2003). 77 B. Sareni, L. Krahenbuhl, A. Beroual, and C. Brosseau, J Appl Phys 81, 2375 (1997). 78 Y. l. Xu and B. A. S. Gustafson, The astrophys J 513, 894 (1999). 79 Q. Xue, Physica B: Condensed Matter 344, 129 (2004). 80 E. Zubko, Y. Shkuratov, and K. Muinonen, Opt Spectrosc. 91, 273 (2001). 81 B. T. Draine and P J. Flatau, J. Opt. Soc. Am. A 11, 1491(1994) 82 B. T. Draine, The Astrophy s. J 333, 848 (1988). 83 F. Nordi n, A. Jean, and L. Georges, J Chem Phys 111, 1195 (1999). 84 D. A. Smith and K. L. Stokes, Opt. Express 14 5746 (2006). 85 Y. Wen Hui, C. S. George, and P. V. D. Ri chard, J Chem Phys 103, 869 (1995). 86 M. I. Mishchenko and D. W. Mackowski, Opt. Lett. 19, 1604 (1994). 87 Y. l. Xu, Appl Opt 34, 4573 (1995). 88 Y. l. Xu and B. S. Gustafson, Appl. Opt. 36, 8026 (1997). 89 T. Nakamura and S. Hayashi, Jpn. J. Appl. Phys. 44, 68336837 (2005). 90 B. J. Messinger, K. U. v. Raben, R. K. Chang, and P. W. Barber, Phys. Rev. B 24, 649 (1981). 91 E. M. Purcell, Phys. Rev. 69, 681 (1946). 92 R. Ruppin, J Chem Phys 76, 1681 (1982). 93 H. Mertens, A. F. Koenderink, and A. Polman, Phys. Rev. B 76, 115123 (2007).

PAGE 155

141 94 Y. S. Kim, P. T. Leung, and T. F. George, Surf Sci 195, 1 (1988). 95 J. Gersten and A. Nitzan, J.Chem. Phys. 75, 1139 (1981). 96 H. Mertens and A. Polman, in ArXiv e prints (2007), Vol. 711. 97 C. H. Su, H. S. Sheu, C. Y. Lin, C.C. Huang, Y. W. Lo, Y.C. Pu, J.C. Weng, D. B. Shieh, J.H. Chen, and C. S. Yeh, J. Am. Chem. Soc. 129, 2139 (2007). 98 H. Cai, N. Zhu, Y. Jiang, P. He, and Y. Fang, Biosens Bioelectron 18, 1311 (2003). 99 M. S. Lang HG, Stevenson KJ, Chandler BD J Am Chem Soc 126, 294912956 (2004). 100 P. Mulvaney, Langmuir 12, 788 (1996). 101 M. Hirai and A. Kumar, J. Appl. Phys. 100, 0143091 (2006). 102 M. Hirai and A. Kumar, J Electron Mater 36, 1574 (2007). 103 F. Hao and P. Nordlander, Chem Phys Lett 446 115, (2007). 104 J. Zhu, Physica E: Low dimensional Systems and Nanostructures 27, 296 (2005). 105 S. Bruzzone, M. Malvaldi, G. P. Arrighini, and C. Guidotti, J. Phys. Chem. B 110, 11050 (2006). 106 S. Link, Z. L. Wang, and M. El Sayed, J Phys Chem B 103, 35293533. (1999). 107 D. H. Chen and C. J. Chen, J. M ater. Chem. 12, 15571562 (2002). 108 U. Pal, J. F. S. ramirez, H. B. Liu, A. Medina, and J. A. Ascencio, Appl. Phys. A 79, 79 (2004). 109 C. San gregorio, M. Galeotti, U. Bardi, and P. Baglioni, Langmuir 12, 5800 ( 1996). 110 J. Hongjin, M. Kyoung sik, and C. P. Wong, presented at the Advanced Packaging Materials: Processes, Properties and Interfaces, 2005. Proceedings. International Symposium on, 2005 (unpublished). 111 A. B. Smetana, K. J. Klabunde, C. M. Sorensen, A. A. Ponce, and B. Mwale, J. Phys. Chem. B 110, 2155 (2006).

PAGE 156

142 112 R. G. Haverkamp, A. T. Marshall, and D. v. Agterveld, J. Nanopart. Res. 9, 697 700 (2007). 113 T. Itakura, K. Torigoe, and K. Esumi, Langmuir 11, 4129 (1995). 114 Y. Jia, H. Niu, M. Wu, M. Ning, H. Zhu, and Q. Chen, Mater. Res. B ull. 40, 1623 (2005). 115 K. i. Okazaki, T. Kiyama, K. Hirahara, N. Tanaka, S. Kuwabata, and T. Torimoto, Chem. Commun., 691 (2008). 116 J. Song, H. Li, J. Li, S. Wang, and S. Zhou, Appl. Opt. 41, 5413 (2002). 117 G. S. Anderson, J. Apll. Phys. 40, 2884 (1969). 118 J. H. Bell, E. T. Schairer, Lawrence A Hand, and R. D. Mehta, Ann. Rev. Fluid Mech. 33 (2001). 119 W. J. Bowyer, W. Xu, and J. N. Demas, Anal. Chem. 76, 4374 (2004). 120 C. N. Jayarajah, A. Yekta, I. Manners, and M. A. Winnik, Macromolecules 33, 5693 (2000). 121 K. A. Kneas, J. N. Demas, B. Nguyen, A. Lockhart, W. Xu, and B. A. DeGraff, Anal. Chem. 74, 1111 (2002). 122 S. Fischkoff and J. M. Vanderkooi, J. Gen. Physiol 65, 663 (1975). 123 M. E. Cox, J polymer Sci A 24, 621 (1986). 124 O. Stern and M. Volmer, M. Phys. Z 20, 183 (1919). 125 R. Ruffolo, C. E. B. Evans, X.H. Liu, Y. Ni, Z. Pang, P. Park, A. R. McWilliams, X. Gu, X. Lu, A. Yekta, M. A. Winnik, and I. Manners, Anal. Chem. 72, 1894 (2000). 126 127 X. Lu and M. A. Winnik, Chem. Mater. 13, 3449 (2001). 128 X. Lu, I. Manners, and M. A. Winnik, Macromolecules 34, 1917 (2001). 129 C. N. Jayarajah, University of Toronto, 1998. 130 S. Draxler, M. E. Lippitsch, I. Klimant, M. Kraus, and O. S. Wolfbeis, J. Phys. Chem. 99, 3162 (1995).

PAGE 157

143 131 S. Draxler and M. E. Lippitsch, Anal. Chem. 68, 753 (1996). 132 J. R. Bacon and J. N. Demas, Anal Chem 59, 2780 (1987). 133 E. R. Carraway, J. N. Demas, B. A. DeGraff, and J. R. Bacon, Anal. Chem. 63, 337 (1991). 134 E. R. Carraway, J. N. Demas, and B. A. DeGraff, Langmuir 7, 2991 (1991). 135 E. R. Carraway, J. N. Demas, and B. A. DeGraff, Anal. Chem. 63, 332 (1991). 136 J. Hubner and B. Carroll, AIAA J. 35, 1790 (1997). 137 G. Schappacher and P. Hartmann, Anal. Chem. 75, 4319 ( 2003 ). 138 K. A. Kneas, J. N. Demas, B. A. DeGraff, and A. Periasamy, Microsc Microanal. 6, 551 (2000). 139 R. D. Bowman, K. A. Kneas, J. N. Demas, and A. Periasamy, J. Microsc. 211, 112 (2003). 140 J. M. Bedlek Anslow, J. P. Hubner, B. F. Carroll, and K. S. Schanze, Langmuir 16, 9137 (2000). 141 Y. Rharbi, A. Yekta, and M. A. Winnik, Anal. Chem. 71, 5045 ( 1999). 142 X Lu, B H Han, and M. A. Winnik J. Phys. Chem. B 13349, 107, ( 2003) 143 D. R. Paul and A. T. Dibenedetto, J. Polym. Sci.C 10, 17 (1965). 144 J. Crank, The Mathematics of Diffusion second ed. (Oxford University Press, 1975). 145 W. J. Koros, J. Wang, and R. M. Felder, J. Appl. Polym. Sci. 26, 2805 (1981). 146 C. Khoe, S. Chowdhury, V. Bhethanabotla, and R. Sen, ACI Materials J, 107, 138, (2010). 147 H. Shiku, T. Saito, C. Wu, T. Yasukawa, M. Yokoo, H. Abe, T. Matsue, and H. Yamada, Chemistry Letters 35, 234 (2006). 148 M. G. Siler, N. Bac, and L. Yilmaz, J. Membr. Sci. 91, 77 (1994). 149 D. Long and F. Lequeux, Eur. Phys. J. E 4, 371 (2001). 150 D. R. Paul and D. R. Kemp, J. Polym. Sci. S ymp. 41, 79 (1973).

PAGE 158

144 151 J. R. Lakowicz, Anal. Biochem. 298, 1 (2001). 152 H. Mertens, Utrecht University, 2007. 153 G. Schneider, G. Decher, N. Nerambourg, R. Praho, M. H. V. Werts, and M. Blanchard Desce, Nano Lett. 6, 530 (2006). 154 F. Tam, G. P. Goo drich, B. R. Johnson, and N. J. Halas, Nano Lett. 7, 496 (2007). 155 K. Aslan, S. N. Malyn, and C. D. Geddes, Chem. Phys. Letts. 453, 222 (2008). 156 N. Kato and F. Caruso, J. Phys. Chem. B 109, 19604 (2005). 157 L. Ao, F. Gao, B. Pan, R. He, and D. Cui, Anal. Chem. 78, 1104 (2006). 158 W. Zai Sheng, J. Jian Hui, L. FU, S. Guo Li, and Y. Ru Qin, Anal. Biochem. 353, 22 (2006). 159 B. Dubertret, M. Calame, and A. J. Libchaber, Nature Biotechnol. 19, 365 (2001). 160 H. Imahori and S. Fukuzumi, Adv. Ma ter. 13, 1197 (2001). 161 W. A. Weimer and M. J. Dyer, Appl. Phys. Lett. 79, 3164 (2001). 162 Z. Y. Li, J. P. Wilcoxon, F. Yin, Y. Chen, R. E. Palmer, and R. L. Johnston, Faraday Discuss 138 363 (2008). 163 J. Zhang, H. Liu, Z. Wang, and N. Ming, J. Sol id State Chem. 180, 1291 (2007). 164 http://en.wikipedia.org/wiki/Cyanine 165 K. B. Male, S. Hrapovic, Y. Liu, D. Wang, and J. H. T. Luonga, Anal. Chim. Acta 516 3541 (2004). 166 J. Zhang, H. Liu, Z. Wang, and N. Ming, J. Solid State Chem. 180, 1291 (2007). 167 E. D. Palik, Handbook of Optical Constants of Solids Ist ed. (Academic Press Inc., Orlando, 1985). 168 G. Ren, H. Shi, and Y. Xing, Nanotechnology 18, 385604 (2007). 169 P. Bharadwaj and L. Novotny, Opt. Express, Vol. 15, Issue 21, pp. 1426614274 15 14266 (2007). 170 M. Meier and A. Wokaun, Opt. Lett. 8 (1983).

PAGE 159

145 171 S. Weiss, Science 283, 1676 (1999). 172 L. M. Smith, J. Z.Sanders, R. J. Kaiser, P. Hughes, C. Dodd, C. R. Connell, C. Heiner, S. B. H. Kent, and L. E. Hood, Nature 321, 674 (1986). 173 M. Pelton, C. Santori, J. Vuckovic, B. Zhang, G. S. Solomon, J. Plant, and Y. Yamamoto, Phys. Rev. Lett. 89, 233602 (2002). 174 (2000). 175 F. C. Gong, D. X. Wu Z. Cao, and X.C. He, Biosens. Bioelectron. 22, 423 (2006). 176 H. Liu, Y. Ie, T. Yoshinobu, Y. Aso, and H. Iwasaki, Appl. Phys. Lett. 88, 061901 (2006). 177 E. Matveeva, Z. Gryczynski, J. Malicka, I. Gryczynski, and J. R. Lakowicz, Anal. Biochem. 334, 303 (2004). 178 K. Aslan, Z. Leonenko, J. R. Lakowicz, and C. D. Geddes, J Fluoresc 15, 643 (2005). 179 F. Tam, G. P. Goodrich, B. R. Johnson, and N. J. Halas, Nano Lett. (2007). 180 G. H. Chan, J. Zhao, E. M. Hicks, G. C. Schatz, and R. P. V. Duyne, N ano Lett. 7 1947 (2007). 181 G. Xu, M. Tazawa, P. Jin, and s. Nakao, Appl. Phys. A. 80, 1535 (2005). 182 R. G. Haverkamp, A. T. Marshall, and D. v. Agterveld, J. Nanopart. Res. 9, 697 (2007). 183 M. Moskovits, I. S. Chem. Phys 116, 10435 (2002). 184 Q. Luo and J. D. Andrade, Journal of Coll oid Interface Sci 200, 104 (1998). 185 R. A. Erb, J. Phys. Chem. 69, 1306 (1965). 184 B. Nikoobakht, C. Burda, M. Braun, M. Hun, and M. A. El Sayed, Photochem and Photobiol 75, 591 (2002). 185 X. Li, J. Qian, L. Jiang, and S. He, Appl. Phys. Lett. 94, 063111 (2009).

PAGE 160

146 186 L. Dyadyusha, H. Yin, S. Jaiswal, T. Brown, J. J. Baumberg, F. P. Booye, and T. Melvin, Chem. Commun., 32013203 (2005). 187 P. P. H. Cheng D. Silvester, G. Wang, G. Kalyuzhny, A. Douglas, and R. W. Murray, J. Phys. Chem. B 110, 4637 (2006). 188 C. Wu, B. P. Mosher, and T. Zeng, presented at the Mater. Res. Soc. Symp. 2005 (unpublished). 189 S. H. Wu and D. H. Chen, J. Colloid Interfac e Sci. 273, 165 (2004). 190 J. T. Buboltz, C. Bwalya, S. Reyes, and D. Kamburov, J. Chem. Phys. 127, 215101 (2007). 191 I. Lisiecki, F. B illoudet, and M. P. Pileni, J Phys Chem 100, 4160 (1996). 192 G. Palazzo, F. Lopez, M. Giustini, G. Colafemmina, and A. Ceglie, J. Phys. Chem. B 107, 1924 (2003). 193 M. Erecinska and I. A. Silver, Respir Physiol 128, 263276 (2001 ). 194 M. Intaglietta, P. C. Johnson, and R. M. Winslow, Cardiovasc Res 32, 632 643 (1996). 195 T. Jue and S. Anderson, Ma Reson Med 13, 524 (1990). 196 C. Lochmann, T. Hansel, T. haupl, and J. Beuthan, Biomed Tech 51, 111 (2006). 197 D. Derkacs, S. H. Lim, P. Matheu, W. Mar, and E. T. Yub, Appl. Phys. Lett. 89, 093103 (2006). 198 K. Nakayama, K. Tanabe, and H. A. Atwaterb, Appl. Phy s. Lett. 93, 121904 (2008). 199 K. R. Catchpole and A. Polman, Optics Express 16, 21793 (2008). 200 S. Chowdhury, V. R. Bhethanabotla, and R. Sen, Appl. Phys. Lett. 95, 131115 (2009).

PAGE 161

147 About T he Author Sanchari Ch ow dhury was born in Kolkata in West Bengal, India. Sanchari holds a BS from National I nstitute of T echnology Durgapur, India and an MS from I ndian I nstitute of T echnology Roorkee, India, both in chemical engineering. She began pursuing her Ph.D. degree in the Che mical and Bi omedical Engineering Department at University of South Florida in 2006. Her dissertation research deals with the enhancement of luminescence sensors using metallic nanoparticles resulted in publications in high quality journal s such as Applied Physics Le tters Journal of Physical Chemistry C and Journal of Microscopy and microanalysis Her research has also resulted in 1 0+ p resentations in prestigious national and international conferences.