USF Libraries
USF Digital Collections

The design, fabrication, and characterization of polymer-carbon nanotube composites

MISSING IMAGE

Material Information

Title:
The design, fabrication, and characterization of polymer-carbon nanotube composites
Physical Description:
Book
Language:
English
Creator:
Clayton, LaNetra
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Nanotechnology
Poly(methyl methacrylate)
Poly(4-methyl-1-pentene)
Interfacial polarization
Dielectric analysis
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The design, fabrication, and characterization of polymer-carbon nanotube (CNT) composites have generated a significant amount of attention in the fields of materials science and polymer chemistry. The challenge in fabricating composites that exploit the unique properties of the CNT and the ideal processing ability and low cost of the polymer is in achieving a uniform dispersion of the filler in the polymer matrix. This body of work focuses on (1) techniques employed to disperse CNTs into a polymer matrix and (2) the effects of CNTs on the mechanical and electrical properties of the polymer. Poly (methyl methacrylate) (PMMA), an amorphous polymer, and poly (4-methyl-1-pentene) (P4M1P), a semi crystalline polymer, were chosen as the matrices. Non-functionalized single-walled carbon nanotubes and soot (unpurified carbon nanotubes) were chosen as the filler material.In the first study, single-walled carbon nanotubes (SWNTs) were sonicated in methyl methacrylate monomer and initiated via thermal energy, UV light, and gamma radiation. Composite films with increased dielectric constants and unique optical transparency were produced. Samples were characterized using differential scanning calorimetry, dielectric analysis, and dynamic mechanical analysis. Refractive Indices were obtained and correlated to the dielectric constant using Maxwells relationship. PMMA/soot composites were fabricated in the second study. Dispersion was accomplished by way of sonication and melt compounding. The PMMA/soot composites were exposed to gamma radiation, with a 137Cs gamma source, in order to investigate how the filler affects the polymers ability to resist radiation. Samples were characterized by differential scanning calorimetry, dielectric analysis, and dynamic mechanical.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by LaNetra Clayton.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 249 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001681154
usfldc doi - E14-SFE0001031
usfldc handle - e14.1031
System ID:
SFS0025352:00001


This item is only available as the following downloads:


Full Text

PAGE 1

THE DESIGN, FABRICATION, A ND CHARACTERIZATION OF POLYMER-CARBON NANOTUBE COMPOSITES by LANETRA MICHELLE CLAYTON A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Ju lie P. Harmon, Ph.D Abdul Malik, Ph.D Milton Johnston, Ph.D Ralph W. Turner, Ph.D Ashok Kumar, Ph.D Date of Approval: April 5, 2005 Keywords: Nanotechnology, poly(methyl meth acrylate), poly(4-methyl-1-pentene), Interfacial Polarization, Dielectric Analysis Copyright 2005, LaNetra Michelle Clayton

PAGE 2

DEDICATION Emmanuelle To Dad, Mom, and Karla

PAGE 3

Acknowledgments I would like to thank my Advisor, Dr. Harmon for the opportunity to pursue my degree under her direction. My committee members: Dr. Ashok Kumar, Dr. Abdul Malik, Dr. Ralph W. Turner, and Dr. Milton Johnsto n. In addition, I would like to thank the Department of Chemistry at USF for givi ng me the opportunity to pursue my degree. I would especially like to thank Dr. O. Geoffrey Okogbaa for his support and encouragement. I would also like to thank Dr. Grisselle Centeno for always being a positive role model and a friend. I am also gr ateful to Dr. Shelli Tatro (Anthony), Kadine Mohomed (Fazir), Emily Ferguson, and Souheil Zekri (Tara) for being great friends. I would like to thank Dr. Meyya Meyyappa n (Center for Nanotechnology, NASA Ames Research Center) and Martin Cinke (Eloret Corporation) for their support and supply of carbon nanotubes. I would also like to tha nk my past and present lab colleagues: Dr. Patricia Muisener, Butch Knudsen, Ke nneth Kull, Kenneth Heffner and Krystal McCann. I would especially like to thank Dr Timofey Gerasimov for being a great Post Doctoral Fellow and for all of his assistance. I would like to extend a very special th ank you to the USF/NSF STARS program for providing me with a fellowship grant.

PAGE 4

i Table of Contents List of Tables v List of Figures vii List of abbreviations and symbols xvii Abstract xix CHAPTER 1. INTRODUCTION 1 Nanotechnology 1 Polymer Nanocomposites 3 Carbon Nanotubes 4 Polymer Carbon Nanotube Composites 7 Design, Fabrication, and Characterization 10 CHAPTER 2. POLYMER RELAXATIONS AND INSTRUMENTATION THEORY 12 Temperature Dependency of Polymer Relaxations 12 Polymer Relaxations in poly (methyl methacrylate) and poly (4-methyl-1-pentene) 15 Instrumentation 20 Thermal Characterization 20 Differential Scanning Calorimetry 20 Dielectric Analysis 23 Dynamic Mechanical Analysis 32 Spectroscopic Characterization 38 Ultra-Violet Visible Spectroscopy 39 Other Characterization Techniques 40 Microhardness 40 Index of Refraction 41 Gel Permeation Chromatography 42 Scanning Electron Microscopy 43

PAGE 5

ii CHAPTER 3. TRANSPARENT POLY (METHYL METHACRYLATE)/ SINGLE WALLED CARBON NANOTUBE COMPOSITES WITH INCREASED DIELECTRIC CONSTANTS: POLYMERIZED VIA HEAT, UV RADIATION, AND IONIZING (GAMMA) R ADIATION INITIATION SOURCES 46 Introduction 46 Experimental 49 Material 49 Single-walled carbon nanotube preparation 49 Polymer-Nanotube Composite Synthesis 49 Methods of Polymerization 50 Molding 52 Differential Scanning Calorimetry 52 Gel Permeation Chromatography 52 Ultraviolet visible Spectroscopy 52 Dielectric Analysis 53 Refractive Index 53 Dynamic Mechanical Analysis 53 Microhardness 54 Results and Discussion 54 Differential Scanning Calorimetry 54 Gel Permeation Chromatography 61 Ultraviolet visible Spectroscopy 62 Scanning Electron Calorimetry 65 Dielectric Analysis 68 Refractive Index 83 Dynamic Mechanical Analysis 87 Microhardness 97 Conclusions 98 CHAPTER 4. GAMMA RADIATION EFFECTS ON PMMA/SOOT COMPOSITES 99 Introduction 99 Experimental 101 Materials 101 Composite Preparation 101 Differential Scanning Calorimetry 102 Dynamic Mechanical Analysis 102 Microhardness 102 Scanning Electron Microscopy 103

PAGE 6

iii Results and Discussion 105 Differential Scanning Calorimetry 105 Scanning Electron Microscopy 119 Microhardness 123 Dynamic Mechanical Analysis 123 Conclusions 135 CHAPTER 5. CHARACTERIZATION OF PMMA/SOOT COMPOSITES VIA DIELECTRIC ANALYSIS 136 Introduction 136 Experimental 141 Results and Discussion 141 relaxation in PMMA/soot composites 141 Dielectric Relaxation Strengths 149 Conclusion 152 CHAPTER 6. EXAMINATION OF DC CONDUCTIVITY AND INTERFACIAL POLARIZATION OF POLYMER-NANOTUBE COMPOSITES 153 CHAPTER 7. PREPARATION OF POLY (4-METHYL-1-PENTENE)/ SINGLE WALLED CARBON NANOTUBES 188 Introduction 188 Experimental 190 Materials 190 Single walled carbon nanotube Preparation 190 Polymer nanotube Composite Synthesis 190 Dynamic Mechanical Analysis 191 Microhardness 191 Differential Scanning Calorimetry 191 Optical Microscopy 192 Results and Discussion 193 Conclusions 206 CHAPTER 8. CONCLUSIONS 207 REFERENCES 209

PAGE 7

iv APPENDICES 220 Appendix A: Chapter 3 221 Appendix B: Chapter 4 223 Appendix C: Chapter 5 224 Appendix D: Chapter 7 225 ABOUT THE AUTHOR End Page

PAGE 8

v List of Tables Table 3.1. DSC Data. Glass Transition Temperatures (Tg) for PMMA and PMMA/SWNT samples polymerized via thermal, UV, and initiation sources. 54 Table 3.2. GPC Results for UV, and Heat polymerized PMMA and PMMA/SWNT composites. 61 Table 3.3 DEA Data. Activation energies of the transition (1-300 Hz). 68 Table 3.4. DEA Data. Dielectric c onstant values of PMMA and PMMA/SWNT Composites. 84 Table 3.5. Refractive Index and Dielectric constant values of PMMA and PMMA/SWNT Composites. 84 Table 3.6. DMA Data. Activation Energies of transition. 87 Table 3.7. DMA Data. Storage Modu lus (E”) values at 10 Hz and -85oC, 25oC, and 100oC. 97 Table 3.8. Microhardness Data for UV, heat, Gamma polymerized PMMA and PMMA/SWNT Nanocomposites. 97 Table 4.1. DSC Data. Glass transition temperatures (Tg) of pure PMMA and PMMA/soot composites before irradiation, immediately after irradiation and four mont hs after irra diation. 105 Table 4.2. Vickers hardness numbers of neat PMMA and PMMA/soot samples before and after irradiation. 123 Table 4.3. DMA data. Activation energies of transitions for neat PMMA and PMMA/soot composites. 124 Table 5.1. DEA data. Activation energies of transitions for neat PMMA and PMMA/soot composites. 142 Table 5.2. DEA Data. Havriliak-Negami parameters and dielectric strengths for neat PMMA, 0.25% PMMA/soot, and 1% PMMA/soot composites. 149

PAGE 9

vi Table 6.1. DEA Data. Activation ener gies of DC conductivity for neat PMMA and PMMA/SWNT samples polymerized via UV light. 157 Table 6.2. DC Conductivity values for PMMA and PMMA/SWNT composites. 158 Table 6.3. DEA Data. Activation ener gies of DC conductivity for neat PMMA and PMMA/soot samples. 171 Table 6.4. DC Conductivity values for PMMA and PMMA/soot composites. 172 Table 6.5. Havriliak-Negami values for neat PMMA and PMMA/SWNT samples polymerized via UV light at 130oC, 150oC, and 180oC. 186 Table 6.6. Havriliak-Negami values for neat PMMA and PMMA/soot samples at 130oC, 150oC and 180oC. 186 Table 7.1. Storage Modulus (E”) values at 60 Hz and -50oC, 25oC and 50oC. 199 Table 7.2. WLF shift constants for poly (4-methyl-1-pentene) and P4M1P/SWNT. 202 Table 7.3. WLF constants and calcul ated fractional free volume and expansion of thermal coefficient values. 205

PAGE 10

vii List of Figures Figure 2.1. Plot of WLF and Arrhenius behavior. 14 Figure 2.2. Structure and relaxations in poly(methyl methacrylate). (a) alpha ( ) relaxation, (b) beta ( ), and (gamma) ( ) relaxation. 18 Figure 2.3. Structure and relaxations of poly(4-methyl-1-pentene). 19 Figure 2.4. Schematic of DSC cell. 21 Figure 2.5. Polymer transition characterized via Differential Scanning Calorimetry. 22 Figure 2.6. DEA dielectric permittivity (’ ) plotted against temperature. 26 Figure 2.7. DEA dielectric loss factor ( “ ) plotted against temperature. 27 Figure 2.8. DEA dielectric loss tangent ( ) plotted against temperature. 28 Figure 2.9. DEA ionic conductivity ( ) plotted against temperature. 29 Figure 2.10. Representation of phase a ngle shift between the applied voltage and current in dielectric analysis. 30 Figure 2.11. Dielectric capacitance plotted against conductance. 31 Figure 2.12. DEA parallel plate sensor. 32 Figure 2.13. Mechanical phase angel sh ifts for an ideal liquid and elastic solid. 33 Figure 2.14. Mechanical phase angle shifts for a viscoelastic polymer. 34 Figure 2.15. Dynamic mechanical tension film clamp. 36 Figure 2.16. Storage modulus (E’) an d loss modulus (E”) plotted against temperature. 37 Figure 2.17. Illustration of the JL Sheppard 137Cs Gamma Irradiator. 45

PAGE 11

viii Figure 3.1. Illustration of UV pol ymerization process. 51 Figure 3.2. DSC data. Tg value for UV polymerized neat PMMA. 55 Figure 3.3. DSC data. Tg value for UV polymerized PMMA/SWNT composite. 56 Figure 3.4. DSC data. Tg value for gamma ( ) polymerized neat PMMA. 57 Figure 3.5. DSC data. Tg value for gamma ( ) polymerized PMMA/SWNT composite. 58 Figure 3.6. DSC data. Tg value for the thermally polymerized neat PMMA. 59 Figure 3.7. DSC data. Tg value for the thermally polymerized PMMA/SWNT composite. 60 Figure 3.8. Films (1.5 mm) of (a) neat PMMA (b) heat polymerized composite, (c) polymerized composite, (d) UV polymerized composite. 62 Figure 3.9. UV-Vis spectra of PMMA/S WNT Composites from 200-8000nm. 63 Figure 3.10. UV-Vis spectra of PMMA/SWNT/CH2Cl2 solution. 64 Figure 3.11. SEM image of UV pol ymerized PMMA/SWNT. 65 Figure 3.12. SEM image of gamma polymerized PMMA/SWNT. 66 Figure 3.13. SEM image of heat polymerized PMMA/SWNT. 67 Figure 3.14. DEA loss factor data fo r UV polymerized neat PMMA. 69 Figure 3.15. DEA loss factor data for UV polymerized PMMA/SWNT composite. 70 Figure 3.16. DEA loss factor data for polymerized neat PMMA. 71 Figure 3.17. DEA loss factor data for polymerized PMMA/SWNT composite. 72 Figure 3.18. DEA loss factor data for th e heat polymerized neat PMMA. 73 Figure 3.19. DEA loss factor data fo r heat polymerized PMMA/SWNT composite. 74

PAGE 12

ix Figure 3.20. Arrenhius Plot of transition for the UV polymerized neat PMMA sample. 75 Figure 3.21. Arrenhius Plot of transition for the UV polymerized PMMA/SWNT Composite. 75 Figure 3.22. Arrenhius Plot of transition for the gamma ( ) polymerized neat PMMA sample. 76 Figure 3.23. Arrenhius Plot of transition for the gamma( ) polymerized PMMA/SWNT Composite. 76 Figure 3.24. Arrenhius Plot of transition for the heat polymerized neat PMMA sample. 77 Figure 3.25. Arrenhius Plot of transition for the heat polymerized PMMA/SWNT Composite. 77 Figure 3.26. and relaxation of UV polymerized neat PMMA at 30 Hz, 60 Hz, and 100 Hz. 79 Figure 3.27. and relaxation of UV polymerized PMMA/SWNT at 30 Hz, 60 Hz, and 100 Hz. 79 Figure 3.28. and relaxation of gamma pol ymerized neat PMMA at 30 Hz, 60 Hz, and 100 Hz. 80 Figure 3.29. and relaxation of gamma polymerized PMMA/SWNT at 30 Hz, 60 Hz, and 100 Hz. 80 Figure 3.30. and relaxation of heat polym erized neat PMMA at 30 Hz, 60 Hz, and 100 Hz. 81 Figure 3.31. and relaxation of heat polym erized PMMA/SWNT at 30 Hz, 60 Hz, and 100 Hz. 81 Figure 3.32.a. DEA loss factor at 60 Hz of UV, gamma, and heat polymerized neat PMMA. 82 Figure 3.32.b. DEA loss factor at 60 Hz of UV, gamma, and heat polymerized PMMA/SWNT. 82

PAGE 13

x Figure 3.33. DEA data. Plot depicting enhanced relaxation of the gamma polymerized PMMA/SWNT composite compared to the neat PMMA. 85 Figure 3.34. DEA data. Plot depicting enhanced relaxation of the heat polymerized PMMA/SWNT composite compared to the neat PMMA. 86 Figure 3.35. DMA data. Loss Modulus Plot of UV polymerized neat PMMA. 88 Figure 3.36. DMA data. Loss Modulus Plot of UV polymerized PMMA/SWNT Composite. 89 Figure 3.37. DMA data. Loss Modulus Plot of heat polymerized PMMA/SWNT Composite. 90 Figure 3.38. DMA data. Loss Modulus Plot of heat polymerized PMMA/SWNT Composite. 91 Figure 3.39. DMA data. Loss Modulus Plot of Gamma ( ) polymerized neat PMMA. 92 Figure 3.40. DMA data. Loss Modulus Plot of Gamma ( ) polymerized PMMA/SWNT Composite. 93 Figure 3.41. Arrenhius Plot for the transition of the UV polymerized neat PMMA. 94 Figure 3.42. Arrenhius Plot for the transition of the UV polymerized PMMA/SWNT. 94 Figure 3.43 Arrenhius Plot for the transition of the heat polymerized neat PMMA. 95 Figure 3.44 Arrenhius Plot for the transition of the heat polymerized PMMA/SWNT. 95 Figure 3.45 Arrenhius Plot for the transition of the gamma polymerized neat PMMA. 96 Figure 3.46 Arrenhius Plot for the transition of the gamma polymerized PMMA/SWNT. 96 Figure 4.1. Illustration of a sonicator. 103

PAGE 14

xi Figure 4.2. Illstration of bandbury mixer. 104 Figure 4.3. DSC data. Glass Transiti on Temperature of neat PMMA before radiation exposure. 107 Figure 4.4. DSC data. Glass Transiti on Temperature of Neat PMMA tested immediately afte r radiation exposure. 108 Figure 4.5. DSC data. Glass Transiti on Temperature of neat PMMA tested 4 months after radiation exposure. 109 Figure 4.6. DSC data. Glass Transition Temperature of 0.25% PMMA/soot before radiation exposure. 110 Figure 4.7. DSC data. Glass Transition Temperature of 0.25% PMMA/soot tested immediately afte r radiation exposure. 111 Figure 4.8. DSC data. Glass Transition Temperature of 0.25% PMMA/soot tested 4 months after radiation exposure. 112 Figure 4.9. DSC data. Glass Transition Temperature of 0.5% PMMA/soot before radiation exposure. 113 Figure 4.10. DSC data. Glass Transition Temperature of 0.5% PMMA/soot tested immediately af ter radiation exposure. 114 Figure 4.11. DSC data. Glass Transition Temperature of 0.5% PMMA/soot tested 4 months afte r radiation exposure. 115 Figure 4.12. DSC data. Glass Tr ansition Temperature of 1% PMMA/soot tested be fore radiation exposure. 116 Figure 4.13. DSC data. Glass Tr ansition Temperature of 1% PMMA/soot tested immediat ely after radiation exposure. 117 Figure 4.14. DSC data. Glass Transition Temperature of 1% PMMA/soot tested 4 months afte r radiation exposure. 118 Figure 4.15. SEM of unpurified carbon nanotubes (soot). 119 Figure 4.16. SEM images before radia tion exposure of neat PMMA. 120 Figure 4.17. SEM images before radiati on exposure of 0.5% PMMA/soot. 121 Figure 4.18. SEM images before radiati on exposure of 1% PMMA/soot. 122

PAGE 15

xii Figure 4.19. DMA data. Loss Modulus (E”) of Neat PMMA at 30 Hz tested before, im mediately after and 4 months after radiation exposure. 125 Figure 4.20. DMA data. Loss Modulus (E”) of 0.25% PMMA/soot composite at 30 Hz tested before, immediately after and 4 months after ra diation exposure. 126 Figure 4.21. DMA data. Loss Modulus (E”) of 0.5% PMMA/soot composite at 30 Hz tested before, immediately after and 4 months after ra diation exposure. 127 Figure 4.22. DMA data. Loss Modulus (E”) of 1% PMMA/soot composite at 30 Hz tested before, immediately after and 4 months after radiation exposure. 128 Figure 4.23. Arrenhius Plot of Transition for non-ir radiated PMMA. 129 Figure 4.24. Arrenhius Plot of Transition for irradiated PMMA. 129 Figure 4.25. Arrenhius Plot of Transition for irradiated PMMA tested 4 months afte r radiation exposure. 130 Figure 4.26. Arrenhius Plot of Transition for no nirradiated 0.25%PMMA/soot. 130 Figure 4.27. Arrenhius Plot of Transition for irradiated 0.25% PMMA/soot. 131 Figure 4.28. Arrenhius Plot of Transition for 0.25%PMMA/soot tested 4 months afte r radiation exposure. 131 Figure 4.29. Arrenhius Plot of Transition for no nirradiated 0.5%PMMA/soot. 132 Figure 4.30. Arrenhius Plot of Transition for irradiated 0.5% PMMA/soot. 132 Figure 4.31. Arrenhius Plot of Transition 0.5% P MMA/soot tested 4 months after ra diation exposure. 133 Figure 4.32. Arrenhius Plot of Transition for no nirradiated 1% PMMA/soot. 133

PAGE 16

xiii Figure 4.33. Arrenhius Plot of Transition for irradiated 1% PMMA/soot. 134 Figure 4.34. Arrenhius Plot of Transition for 1% PMMA/soot tested 4 months afte r radiation exposure. 134 Figure 5.1. Argand (Cole-Cole) plot of loss factor plotted against the dielectric permittivity. 140 Figure 5.2. Complex Cole-C ole plot of loss factor plotted against the dielectric permittivity. 140 Figure 5.3. DEA loss factor pl ot for neat PMMA. 143 Figure 5.4. DEA loss factor pl ot for 0.25% PMMA/soot. 144 Figure 5.5. DEA loss factor plot for 1% PMMA/soot with enhanced region. 145 Figure 5.6. DEA loss factor plot at 60 Hz of neat PMMA, 0.25% PMMA/soot,1% PMMA/soot. 146 Figure 5.7. Arrhenius plot for neat PMMA. 147 Figure 5.8. Arrhenius plot for 0.25% PMMA/soot. 147 Figure 5.9. Arrhenius plot for 1% PMMA/soot. 147 Figure 5.10. DEA permittivity at 60 Hz for neat PMMA and PMMA/soot composites. 148 Figure 5.11. Cole-Cole plot of the relaxation for neat PMMA at 40oC. 150 Figure 5.12. Cole-Cole plot of the relaxation for 0.25% PMMA/soot at 40oC. 150 Figure 5.13. Cole-Cole plot of the relaxation for 1% PMMA/soot at 40oC. 150 Figure 5.14. Cole-Cole plot of the relaxation for neat PMMA at 100oC. 151 Figure 5.15. Cole-Cole plot of the relaxation for 0.25% PMMA/soot at 100oC. 151

PAGE 17

xiv Figure 5.16. Cole-Cole plot of the relaxation for 1 % PMMA/soot at 100oC. 151 Figure 6.1. AC conductivity plotted ag ainst frequency for neat PMMA polymerized via UV light. 159 Figure 6.2. AC conductivity plo tted against frequency for neat PMMA/SWNT polymerized via UV light. 160 Figure 6.3. AC Conductivity plotted ag ainst frequency for neat PMMA polymerized via radiation. 161 Figure 6.4. AC Conductivity plotted ag ainst frequency for PMMA/SWNT polymerized via radiation. 162 Figure 6.5. AC Conductivity plotted ag ainst frequency for neat PMMA polymerized via thermal energy. 163 Figure 6.6. AC Conductivity plotted ag ainst frequency for PMMA/SWNT polymerized via thermal energy. 164 Figure 6.7. Arrhenius plot of neat P MMA polymerized via UV light from 125oC -175oC. 165 Figure 6.8. Arrhenius plot of PMMA/SWN T polymerized via UV light from 125oC -175oC. 165 Figure 6.9. Arrhenius plot of n eat PMMA polymerized via thermal energy from 125oC -175oC. 166 Figure 6.10. Arrhenius plot of PMMA/SWNT polymerized via thermal energy from 125oC -175oC. 166 Figure 6.11. Arrhenius plot of neat PMMA polymerized via from 125oC -175oC. 167 Figure 6.12. Arrhenius plot of PMMA/SWNT polymerized via from 125oC -175oC. 167 Figure 6.13. Log dc vs. temperature of neat PMMA polymerized via UV light from 125oC – 195oC. 168 Figure 6.14. Log dc vs. temperature of P MMA/SWNT polymerized via UV light from 125oC – 195oC. 168

PAGE 18

xv Figure 6.15. Log dc vs. temperature of neat PMMA polymerized via radiation from 125oC – 195oC. 169 Figure 6.16. Log dc vs. temperature of P MMA/SWNT polymerized via radiation from 125oC – 195oC. 169 Figure 6.17. Log dc vs. temperature of neat PMMA polymerized via thermal energy from 125oC – 195oC. 170 Figure 6.18. Log dc vs. temperature of P MMA/SWNT polymerized via thermal energy from 125oC – 195oC. 170 Figure 6.19. AC conductivity plotted agai nst frequency for neat PMMA. 173 Figure 6.20. AC conductivity plo tted against frequency for 0.25% PMMA/soot. 174 Figure 6.21. AC conductivity plo tted against frequency for 1% PMMA/soot. 175 Figure 6.22. Arrhenius plot of neat PMMA from 125oC -175oC. 176 Figure 6.23. Arrhenius plot of 0.25% PMMA/soot from 125oC -175oC. 177 Figure 6.24. Arrhenius plot of 1% PMMA/soot from 125oC -175oC. 178 Figure 6.25. Log dc vs. temperature of (a) neat PMMA and (b) 0.25% PMMA/soot, and (c) 1% PMMA/soot from 125oC – 195oC. 179 Figure 6.26. DEA data. Loss factor plo tted against log frequency for neat PMMA above the gla ss transition temperature. 182 Figure 6.27. DEA data. Loss factor plo tted against log frequency for PMMA/SWNT above the glass transition temperature. 182 Figure 6.28. Electric Modulus treated lo ss factor plotted against log frequency for neat PMMA above the glass transition temperature. 183 Figure 6.29. Electric Modulus treated loss factor plotted against log frequency for PMMA/SWNT above the glass transition temperature. 183

PAGE 19

xvi Figure 6.30. Cole-Cole plot with el ectric modulus treatment of neat PMMA and PMMA/SWNT at 130oC, 150oC, 180oC. 184 Figure 6.31. Cole-Cole plot with el ectric modulus treatment of neat PMMA and PMMA/soot composites at 130oC, 150oC,180oC. 185 Figure 7.1. (a) carbon nanotubes soni cated in 1-chlorohexane, (b) carbon nanotubes in cyclohe xyl chloride pretreated with DMF,(c) carbon nanotubes/cy clohexyl chloride/polymer. 193 Figure 7.2. Optical Micrograph of (a) neat P4M1P and (b) 0.5% P4M1P/SWNT composite. 10 x 0.3 magnification. 194 Figure 7.3. DMA data. Loss Modulus (E”) plotted against temperature for neat P4M1P and P4M1P/SWNT. 196 Figure 7.4. DSC data. DSC plot of neat P4M1P. 197 Figure 7.5. DSC data. DSC plot of P4M1P/SWNT. 198 Figure 7.6. DMA data at 60 Hz of E’ and E”. 199 Figure 7.7. Arrhenius plot for neat P4M1P from 1Hz-100Hz. 201 Figure 7.8. Arrhenius plot for neat P4M1P/SWNT from 1Hz-100Hz. 201 Figure 7.9 Master Curve of neat P4 M1P and P4M1P/SWNT composite generated using WLF shift constants. 203 Figure 7.10. Master curve of reported Tg region for P4M1P using WLF shift constants. 204 Figure A.1. Syndiotactic PMMA 221

PAGE 20

xvii List of Abbreviations and symbols AC conductivity ac Activation Energy Ea Atomic mass unit amu Carbon nanotubes CNTs DC conductivity dc Dielectric Analysis DEA Dielectric Constant (permittivity) ’ Dielectric Loss Factor ” Differential Scanning Calorimetry DSC Dynamic Mechanical Analysis DMA Dynamic Loss Modulus E” Dynamic Storage Modulus E’ Gel Permeation Chromatography GPC Glass Transition Temperature Tg Maxwell’s Relationship ’=n2 Maxwell-Wagner-Sillars MWS Melting Temperature Tm Methyl methacrylate MMA Monomethyl ether hydroquinone MEHQ N,N-Dimethyl Formamide DMF National Aeronatics Space Administration NASA Number Average Molecular Weight Mn Polydispersity Mw/Mn, PD Poly (methyl methacrylate) PMMA Poly (4-methyl-1-pentene) P4M1P Refractive Index RI, n

PAGE 21

xviii Rotations per minute rpm Single –walled carbon nanotubes SWNTs Scanning Electron Microscopy SEM Ultraviolet UV Ultraviolet Visible Spectroscopy UV-Vis Vickers Hardnes Number HV Weight Average Molecular Weight Mw Williams-Landel-Ferry WLF

PAGE 22

xix THE DESIGN, FABRICATION A ND CHARACTERIZATION OF POLYMER-CARBON NANOTUBE COMPOSITES LaNetra Michelle Clayton ABSTRACT The design, fabrication, and characteriza tion of polymer-carbon nanotube (CNT) composites have generated a significant amount of attention in the fields of materials science and polymer chemistry. The challenge in fabricating composites that exploit the unique properties of the CNT and the ideal pr ocessing ability and low cost of the polymer is in achieving a uniform dispersion of th e filler in the polymer matrix. This body of work focuses on (1) techniques employed to disp erse CNTs into a polymer matrix and (2) the effects of CNTs on the mechanical and electrical properties of the polymer. Poly (methyl methacrylate) (PMM A), an amorphous polymer, and poly (4-methyl-1-pentene) (P4M1P), a semi crystalline polymer, were chosen as the matrices. Non-functionalized single-wa lled carbon nanotubes a nd soot (unpurified carbon nanotubes) were chosen as the filler material. In the first study, single-walled carbon nanotubes (SWNTs) were sonicated in methyl methacrylate monomer and initiated via thermal energy, UV light, and gamma radiation. Composite films with increased dielectric constants and unique op tical transparency were produced. Samples were characterized using differential sca nning calorimetry, dielectric analysis, and dynamic mechanical analysis. Refractive Indi ces were obtained and correlated to the dielectric constant usi ng Maxwell’s relationship. PMMA/soot composites were fabricated in the second study. Dispersion was accomplished by way of sonication and melt compounding. The PMMA/soot composites were exposed to gamma radiation, with a 137Cs gamma source, in order to investigate how the filler affects the polymers’ ability to resist radiation. Sample s were characterized by diffe rential scanning calorimetry, dielectric analysis, and dynamic mechanic al. The third study involved dispersing nonfunctionalized nanotubes into P4M1P, a polymer without polar groups.

PAGE 23

xx The polar solvent N,N dimethylformamide (D MF) is known to be an ideal dispersing agent for carbon nanotubes. However, P4M1P does not dissolve in DMF. A series of solvents that would both dissolve the polymer and disperse the nanot ubes were explored. A successful combination of pr e-treating the nanotubes with DMF and then dispersing the nanotubes and dissolving the polyme r in cyclohexyl chloride was achieved.

PAGE 24

1 CHAPTER 1 Introduction Polymeric material exists in many forms: plastics, fibers, elastomers (rubber), coatings, and adhesives (Stevens 1999). They have impacted modern society in every imaginable way. Both natural (i.e., proteins starch, and cellulose) and synthetic (i.e., nylon, polyethylene, poly (vinyl chloride)) polymers are used for applications in the medical, electrical, mechanical, and communi cation industries (Stevens 1999; National Research Council 1994). The field of polymer science has seen many advances in the development of (1) engineering plastics to replace industrial metals; (2) high – strength aromatic fibers; (3) degradable polymers for control drug release and the reduction of plastic waste; and (4) conducting polymers that have conductivities comparable to metals (Stevens 1999). Many of these polymers have been found to be light weight, easy to process, and exhibit enhanced corrosion resi stance as compared to industrial metals (Stevens 1999). During the 1980’s the use of polymeric material in industrial applications exceeded that of iron and st eel (Stevens 1999, Jones 1989). The properties of polymers have been further enhanced by incorporating fillers, that when combined with the polymer, can be used to fine t une polymer properties to better fit specific applications. Nanoscale fillers have a larg e surface-to-volume ratio, thus creating a stronger interaction at the polymer – nanopa rticle interface, re sulting in enhanced mechanical, electrical, optical, and chemical properties of polymeric materials. Nanotechnology One nanometer is defined as bei ng one billionth of a meter or 10-9 meters. The diameter of a human hair measures about 50,000 nanometers, and a bacterial cell measures about two hundred nanometers acr oss (Ratner and Ra tner 2003). The nanometer is the primary unit of measur ement used in nanotechnology. Nanotechnology

PAGE 25

2 falls under the theoretical study of nanoscience a nd is the field of study in which material measuring less than 100 nanometers is manipul ated to create a more efficient device or material (Ratner and Ratner 2003). These nanos ized materials (atoms and molecules) are known as nanoparticles and repr esent the building blocks of creating miniaturized man made devices or nanostructures with uni que and improved mechanical, electrical, chemical, and optical properties. Nanot echnology has been discussed for several decades. Nobel Laureate, Dr. Richard Feynman, coined the famous phrase There’s Plenty of Room at the Bottom in his 1959 talk at the Calif ornia Institute of Technology (Feynman 1960). Dr. Feynman presented a sc ientific address that focused on creating small devices by manipulating the basic bui lding blocks that exist in nature (atoms and molecules) (Feynman 1960). Fe ynman predicted that researchers would begin to understand and explore how the bottom up approach could be utilized in creating smart machines at the nanoscale level around the year 2000 (Feynman 1960). According to Ratner and Ratner (2003), the first nanoscal e devices and serious re search began at or near the year 2000. Forty years later, N obel Laureate, Dr. Richard Smalley, in his address to the U.S. Congress st ated, “…the next century is goi ng to be incredible. We are about to be able to build th ings that work on the smallest possible length scales, atom by atom” (Klabunde et al. 2001; Smalley 1999). Nanotechnology is still a relativ ely new field of study but is poised to affect all areas of modern life and research areas such as pharmaceuticals and biomedical devices, information storage, sensor development, computer technology, environmental devices and substances, polymer technology, and space exploration (Klabunde et al. 2001; Ratner and Ratner 2003). Recent success in the fiel d has led the U.S. Federal Government to declare Nanotechnology a national priorit y. The National Nanotechnology Initiative (NNI) under a different name star ted in 1996 as an attempt to coordinate federal work in the field of nanotechnology. NNI was officially recognized as a fede ral initiative in 2001 with its major focus being to guarantee U.S. leadership through research and development in Nanotechnology. Placed under the Nanoscale Science, Engineering and Technology (NSET) Subcommittee of the Na tional Science and Technology Council, seventeen government departments a nd agencies participate in the NNI. The Federal

PAGE 26

3 Government funding of NNI was $116 milli on in 1997 and $961 million in 2004 with a 2005 budget request of $982 million (Ratner and Ratner 2003; Davey 2004). With a solid financial backing from the U.S. Fede ral Government, researchers from several scientific and engineering disciplines are collaborating and competing for funding to explore and develop nanoscale materials a nd devices. One such field is Polymer Technology. Nanoparticles have been successfully incorporated into the matrix to create polymer nanocomposites. Despite the adva nces in the fabrication of polymer nanocomposites, research is still needed to improve the synthesis, processing and characterization techniques for these compos ites to account for the large production scale range (Krishnamoorti and Vaia 2002). Polymer Nanocomposites Polymer nanocomposites differ from other polymer composite material because the fillers exist at the na noscale level (Klabunde 2001; Krishnamoorti and Vaia 2002; Ratner and Ratner 2003). According to Kr ishnamoorti and Vaia ( 2002) a characteristic polymer nanocomposite (1) consists of partic les that measure on the nanoscale level; (2) has a polymer matrix that is interfaced with nanoparticles; and (3) consist of particles with nanoscale arrangements. When combined these materials have been shown to show superior mechanical and thermal properti es (Bandyopadhyay, Hsieh and Giannells 2002). Further, nano fillers, when dispersed uniformly throughout the polymer, will create large distances between individual pa rticles enhancing polymer-par ticle interaction and having a greater affect on the polymer structure a nd properties (Krishnamoorti and Vaia 2002). Nanoparticles are also smaller than the wave length of light, thus resulting in unique optical properties (Klabunde 2001; Petrovic and Javnis et al. 2000). Properties of polymer nanocomposites can be further manipul ated depending on the specific type of filler. Layered silicates (i.e., clay) have been dispersed in polymer matrices to create nanocomposites with improved barrier properties (Zhou and Nakamura et al. 2002). Nano boron nitride fillers have been used to improve the thermal conductivity of polymers used in microelectronic packaging (Paine and Pruss et al 2002). Iron oxide nanoparticles were incorporated into pol yvinyl alcohol to increase the magnetic

PAGE 27

4 properties (Novakova and Lanchinskaya 2003) of the polymer. Carbon nanotubes have been shown to increase the mechanical pr operties and provide increased electrical properties to the polymer (Ounaies 2003, Sandler and Sh affer et al. 1999, Zhao 2001, Muisener 2002, Tatro 2004, Clayton 2005). Carbon Nanotubes Carbon nanotubes (CNTs) can be either multi-walled (MWNTs) which were discovered in 1991 or single walled (SWNTs ) discovered in 1993 (Iijima 1991; Nalwa et al 2000). Within the last 10 years, multi -walled carbon nanotubes have found their way into field emitting devices a nd tips for Atomic Force Microscopy (AFM). Single walled nanotubes have been used in electronic devi ces and have been shown to be an ideal substrate for hydrogen storage (Nalwa et al. 2000). Single walled carbon nanotubes consist of a single layer of a graphene sheet rolled up and ca pped at either end with a five membered ring. They have diameters ranging from 1 to 2 nm (Nalwa et al. 2000) and lengths as high as 700 nm (0.9 nm diameter). Multi-walled carbon nanotubes consist of several layers of graphene sheets rolled up into cylinders capped at bot h ends with a five membered ring (half of a fullerene). The laye rs have an estimated spacing of 0.34 nm, an outer layer diameter ranging from 2 and 25 nm and an inner hollow ranging from about 1 to 8 nm (Nalwa et al. 2000). Carbon nanot ubes are known to have large aspect ratios (Dresselhaus, Dresselhaus, a nd Eklund 1996). The nanometer si ze, the helical structure, and the topology of carbon nanot ubes give them thei r excellent mechan ical (stability, strength, stiffness, and elasti c deformability), transport (ele ctron transport), and surface properties (Nalwa et al 2000). Mechanicall y, CNTs have been described as being 100 times as strong as steel (Foust 2003; Collins a nd Avouris 2000) and one sixth the weight of steel (Bernholc et al. 2000). The mechani cal strength of carbon nanotubes arises from the tube configuration of the perfect a rrangement of the covalently bonded C-C bonding in graphitic carbon. Theoreti cal results show the Young’s modulus of SWNTs ranging from 1-5 TPa, as compared to theoretical valu es of in-plane graphite as 1 TPa (Nalwa et al. 2000; Overney, Zhong, and Tomanek 1993). Experimental studies of both SWNTs and MWNTs using Atomic Force Microscopy and Transmission Electron Microscopy to

PAGE 28

5 obtain Young’s Modulus confirmed values abov e 1 TPa (Krishnan et al. 1998; Muster et al. 1998; Salvetat et al. 1999; Treacy et al. 1996; Wong et al. 1997). Carbon nanotube structure directly affects their electronic tran sport capabilities (Bernholc et al.2000). CNTs can be metallic, se mi conducting or semi metallic (Dai et al. 2002). SWNTs can be in three structural forms: (1) zigzag, (2) armchair, and (3) chiral. A nanotube with a zigzag conformation is form ed when the graphene sheet is bisected normal to a threefold axis forming a vector. Upon rolling up the sheet, the ends of the chiral vector meet and n and m are integers in the equation Ch= n a1 +m a2 and determine the helicity of the tube. The resultin g chiral angle for a zigzag tube is 0o and n or m are zero. An armchair nanotube has a chiral angle of 30o and n is equal to m Chiral nanotubes have chiral angles that are 0< <30o. A SWNT is metallic if n-m is divisible by 3, in all other cases the nanotube is tw o-thirds conducting and one third metallic (Dresselhaus, Dresselhaus, a nd Eklund 1996; Dresselhaus, Dr esselhaus, and Avouris et al. 2001). For example, a (9,0) nanotube is a zigzag nanotube and metallic. Nanotubes having an armchair configuration are always metallic and tubes that have a zigzag and chiral configuration can be either metallic or semi c onducting (Nalwa et al. 2000). Theoretical results on the elec tronic behavior of carbon nano tubes are usually conducted on the simplest form of CNTs, SWNTs (D resselhaus, Dresselhaus, and Eklund 1996); however, many of the experimental results are based on multi-walled carbon nanotubes (Charlier and Michenaud 1993; Charlier 1994; Saito, Dresselhaus, and Dresselhaus 1993; Lambin and Philippe et al. 1994; Lamb in and Charlier et al. 1994). There presently exist three major routes for the growth of carbon nanotubes: arcdischarge, laser ablation, and chemical va por deposition (CVD). Multi-walled carbon nanotubes (MWNTs) can be grown using an arc-discharge between two graphitic electrodes (Nalwa et al 2000; Dresselhaus, Dresselhaus, and Avouris 2001) or CVD (Dai et al. 2002) which involves flowing a hydrocar bon gas such as acetylene and ethylene through a tube reactor fo r a specified amount of time while heating a catalyst material to temperatures ranging from 500oC to 1000oC (Dai et al. 2002). Arc-discharge produces MWNTs that are extremely stra ight and contain very limite d defects. MWNTs produced via CVD are known to have a high density of defects on the sidewalls; however, this

PAGE 29

6 method has been shown to produce quantities ra nging from kilograms to ton levels (Dai et al. 2002), which is in contrast to the gram quan tities obtained from the arc-discharge method ( Dai et al. 2002). Single walled carbon nanotubes (SWNTs) ma y be grown by the arc-discharge, laser ablative, CVD, and high pressure catalytic d ecomposition of carbon monoxide (HiPCO) methods. The arc discharge involve s co-evaporating a meta l catalyst such as iron or cobalt in methane gas and is known to produce high quality tubes (Nalwa et al 2000; Dresselhaus, Dresselhaus, and Ekl und 1996; Dresselhaus, Dresselhaus, and Avouris 2001). Laser ablation has also been used to grow high quality SWNTs with limited defects. Chemical Vapor Deposition (CVD) is another method that has been used successfully to grow carbon nanotubes (B aker 1978; Tibbetts 1984; Tibbetts 1990). CVD has produced SWNTs with limited to no stru ctural defects and in large quantities. SWNTs have also been grown by using the HiPCO method in which high pressure CO serves as the carbon source and Fe(CO)5 as the catalyst at high temperatures to produce high quality SWNTs with limited defects ( Ti bbetts 1984; Cheng et al 1998; Nikolaev et al. 1999). As a result of nanotube growth, byproducts such as amorphous carbon, metal catalyst, and graphitic layers exist in the ra w nanotube material. Currently, there exist a number of methods used to purify carbon na notubes: acid oxidation, gas phase oxidation, filtration, and chromatography (Dresselhaus, Dr esselhaus, and Avouris 2001; Rao et al 2001). In order to obtain tube s with the highest purity, a combination of chemical and physical methods should be employed (Shelimov et al. 1998). Several characterization techniques have been used to better understand the structure and behavior of carbon nanotubes. Atomic Fo rce Microscopy (AFM) which provides resolution at the atomic level (Nalwa et al. 2000) has been used to understand the surface topography of CNTS as well as stud y nanotube elastic strength (Simonis et al 2000; Forro et al. 2000; Krishnan et al. 1998; Muster et al. 1998; Salvetat et al. 1999; Treacy et al. 1996; Wong et al. 1997). Tr ansmission Electron Microscopy (TEM) has been used to observe the structure of th e SWNTs and MWNTs, chemical composition of

PAGE 30

7 purified and unpurified carbon nanotubes, and diameter. High Resolution Transmission Electron Microscopy (HRTEM), the technique used when Sumio Iijima discovered MWNTs (Iijima 1991), can be used to understand the order and disord er present in the nanotube walls (Forro et al.2000). Scanning Electron Microscopy (SEM) can be used to determine nanotube diameter size and length (Simonis and Volodin et al. 2000; Avouris et al.1999) as well as the purity of CNTs (Dai 2001; Dresselhaus, Dresselhaus and Avouris 2001). When analyzed via Rama n Spectroscopy, metallic and semi-conducting SWNTs can be differentiated from peak assi gnments (Saito and Ka tura 2001; Cooper and Young 2000). Raman also provides information pertaining to tube or ientation, diameter, and metallic character (Saito and Katura 2001; Dresselhaus, Dresselhaus and Avouris 2001). Polymer Carbon Nanotube Composites The goal of creating polymer carbon nanot ube composites is to develop a product that exploits the unique elec trical and mechanical prope rties of the carbon nanotube, while maintaining the ideal properties, low cost, and processing ability of the polymer matrix. However, the incorporation of CNTS in the polymer matrix has been hindered due to the difficulties in achieving uniform dispersion of the nanotubes in the polymer matrix. Due to the strong van der Waal s forces, carbon nanotube s usually exist in bundles. Much research has been focused on modifying the surface of carbon nanotubes to reduce bundling, thus increasing applicab le uses (Bahr and Tour 2002; Pompeo and Resasco 2002; Dyke and Tour 2004; Hirsch 2002; Rao and Satishkumar et. al 2001). These include coating the surface with a c onducting polymer via plasma polymerization (Coleman and Dalton 2000), chemical dopi ng (Rao and Satishkumar et. al 2001), ultrasonic dispersion in a solvent (Chen and Hamon et al. 1998; Hamon and Chen 1999 et al.; Chiang and Brinson et al. 2001; Coleman 2000; Huang and Lin et al.), and grafting (Pompeo and Resasco 2002). These forms of functionalization of the carbon nanotube begin with defects in the tube. Defects se rve as primary starting points for covalent functionalization on the SWNT surface (Hirsch 2 002). Defects can be located at the tips of the tube after removing the end caps, on the sidewalls, or points of curvature on the tube. In addition to the above mentioned modification techniques, several processing

PAGE 31

8 methods are employed to disperse nanotubes in a polymer matrix. Methods such as sonication (Koshio and Yudasaka et al. 2001), in situ polymerization (Park and Ouanies 2002), solution mixing (O’Rourke Muisener et al. 2001, Harmon et. al 2002, Tatro et. al 2002, Clayton et. al 2004), and melt mixing are common techniques (Potschke, Fornes and Paul 2002; Zou and Feng et. al 2003; O’ Rourke Muisener et al. 2001, Harmon et. al 2002, Tatro et. al 2002, Clayton et. al 2004). Compression mo lding (Haggenmueller and Gommans et. al 2000) and electrospinning spinni ng (Sen and Zhao et. al 2004) have also been used. In most cases a combination of th ese processes are used. In situ polymerization/sonication (Chapter 3), solu tion mixing/sonication (Chapters 4 and 7), and melt mixing (Chapter 4), were used in this work. Grimes et al. (2000) utilized in-situ polymerization in the presence of nanotubes to produce a polymer composite. Park et. al (2002) also employed in-situ polymerization to fabricate polymer nanocomposites. The resulting polyimide/SWNT composites had un iform dispersion, optical transparency and increased conductivity. It has been stated that certain free radi cal initiators open bonds in CNTs. When present dur ing the addition poly merization of methyl methacrylate monomer carbon nanotubes were shown to partic ipate in the polymerization process (Jia et al. 1999). Thus, in-situ polymerization of a monomer in the presence of carbon nanotubes increases the possibi lity of chemical adhesion at the polymer-carbon nanotube interface. Dispersion in a polymer soluti on is often preceded by dispersing the carbon nanotubes in a suitable solvent. A solvent of choice must dissolve the polymer or be miscible with the solvent that dissolves the po lymer, as well as disperse the nanotubes. Carbon nanotubes are insoluble in most organic solvents. However, solvents such as N,N dimethylformamide (DMF) (Park and Ounaie s 2002), toluene, chlo roform (Kymakis, Alexandou and Amaratunga 2002), and ethanol (Zou and Feng et.al 2004) have been found to nicely disperse carbon na notubes in the polymer matrix It is safe to say that due to the metallic character of CNTS, the more polar the solvent, the better it will perform in dispersing the carbon nanotubes. So dium dodedcyl sulfate has also been used to disperse carbon nanotubes (Assael and Chen et.al 2004). Melt mixing is a process that can be eith er extensive (distributive) or intensive

PAGE 32

9 (dispersive) (Jones 1989). Intensive mixi ng (melt compounding) was the form of melt mixing used in the experiment detailed in Chapter 4. It differs from extensive mixing (melt blending) because it usually involves a phy sical change of the material, a high shear force to bring about this cha nge, and the polymer must exist in its rubbery or molten state in order for successful dispersion of the two components ( Jone s 1989). Single-walled, multiwalled, and soot have been dispersed in poly (methyl methacrylate) using a branbury compounding mixer (O’Rourke Muisen er et al. 2001, Harm on et. al 2002, Tatro et. al 2002, Clayton et. al 2004). Melt mixing can also be accomplished by using extrusion. Researchers have ut ilized twin screw extruders to disperse nanotubes into the polymer matrix (Potschke, Fo rnes and Paul 2002; Zou and Fe ng et. al 2003). Sonication is a vital component in the processing step. It has served as a beneficial tool in dispersing nanotubes in the monomer or polymer soluti on. However, sonication has been found to shorten nanotube length (Koshio and Yudasaka et al. 2001). Despite this occurrence, the resulting polymer nanotube composites display properties that are an improvement on the polymer properties. Mechanically, carbon nanotubes serve as great reinforcement agents to the polymer matrix. If uniformed dispersion is achieved throughout th e polymer matrix, the carbon nanotubes will absorb much of the energy caused by the applied stress (Sen and Zhao et. al 2004; Qian and Dickey et al. 2000). The incorporation of 1% (by weight) multi walled nanotubes (MWNTs) in polystyrene resulted in a 36%-42% increase in elastic modulus and a 25% increase in the br eak stress (Qian and Dickey et al. 2000). MWNTs were shown to increase the viscosit y of polycarbonate with increasing nanotube loading; thus, resulting in an increase in elastic melt properties (P otschke, Fornes, and Paul 2002). Aligned single walled carbon nanotubes incorporated into poly (methyl methacrylate) resulted in fibers that exhibite d an increase in elastic modulus as nanotube loading increased (Haggenmueller and Gomman s et al. 2000). Sen et. al (2004) found that ester functionalized SW NTs improved the mechanical properties of polyurethane more than nonfunctionalized tubes. This s upports the theory that uniform dispersion is vital in order to achieve load transfer, as well as strong a dhesion at the polymer-nanotube interface.

PAGE 33

10 Carbon nanotubes have also improved th e electrical properties of polymer matrices. Poly (3-octythiophene)/SWNT composites exhibited an increase in conductivity by five orders of magnitude with an increase in loading from 0 to 20% (by weight) (Kymakis, Alexandou, and Amaratunga 2002). Poly(ethyl methacrylate)/SWNT composites with nanotube concentrations rangi ng from 0-23 wt% were measured in the 500 MHz to 5.50 GHz complex permittivity spectra and found to increase the magnitude of the relative permittivity (Grimes and Mungl e et. al 2000). Kymakis and Amaratunga (2004) optimized characterization via ab sorption spectroscopy by experimentally studying the optical properties of nanocompos ite films as a function of volume fraction and theoretically by using the Maxwell-Garnet t effective medium theory. The absorption spectra of polymer-CNT composites were also used to better unders tand the dielectric behavior of the material (Kymakis, Alexandou, and Amaratunga 2002). Design, Fabrication, and Characterization The work presented in chapters 3-7 of this dissertation collectively focus on the design, fabrication, and characterization of polymer carbon nanotube composites. According to McCrum, Buckley, and Bucknell (1997) designing a successful plan for the production of polymeric material encompasse s (1) understanding the known and desired physical and chemical properties of the material; (2) the a ppropriate processing techniques for the material and how processi ng will effect the material’s properties; and (3) economic factors. Fabrication includes the polymerizati on and polymer processing steps and will greatly influence structure-prop erty relationships. For example, different polymerization processes lead to differences in stereoregularit y, molecular weight, branching, or crosslinking be havior (Stevens 1999). Li kewise, polymer processing (molding, extrusion, and blending ) can influence the crystal linity, strength, and modulus of a polymer. Characterization of polymers most comm only utilizes thermal and spectroscopic methods (Stevens 1999). Thermal analysis can be used to investigate the effect of the nanoparticles on polymer ch ain mobility and conformation. Adhesion strength between the polymer-nanoparticle inte rface can also be suggested as a result of

PAGE 34

11 relaxation behavior. Literature has stated that ca rbon nanotubes can be used to identify or detect polymer relaxations (Zhao and Wood 2001). Spectroscopic characterization (Ultraviolet Spectroscopy) assists in determ ining optical transparency and dispersion quality (extent of scattering). Techniques su ch as microhardness, index of refraction, gel permeation chromatography and scanning elec tron microscopy serve as supporting tools to further understand the behavior of polymer nanocomposites and optimize these characterization techniques for future studies.

PAGE 35

12 CHAPTER 2 Polymer Relaxations and Instrumentation Theory All polymers display a glass transition temperature which is indicative of the primary relaxation. The Tg region is representative of an increase in free volume (empty spaces that allow for chain mobility ) and corresponds to synchronized motion of about 10 to 100 carbon atoms in the main chain (Seymour and Carraher 1988, Gedde 1999). This transition is de fined by three theories: kine tic, thermodynamic, and free volume. The kinetic theory defines the tran sition as being a kinetic process without a defined thermodynamic glass transition. The th ermodynamic theory stat es that the glass transition temperature is a kine tic process; however, equilibrium properties do exist. The free volume theory defines the transition as the temperature where there exists enough empty space to allow for the atoms in the main chain to slip past each other. This is known as the iso-free volume state (Gedde 1999) According to Doolitle, the main chain cooperative motion is described as the viscosity of the polymer and can be related to the fractional free volume using equation (2.1). f BA (2.1) Where is the viscosity, a a nd b are constants and f is the fractional free volume. The fractional free volume is defined as f = f/f is the free volume per gram and is the specific volume at a particular temperature (McCrum 1967, Gedde 1999).

PAGE 36

13 The Dolittle equation can be used to derive the William-Landel-Ferry (WLF) equation. As stated above, molecular motion depends on the amount of free space available at a certain temper ature; the WLF equation (2.2) describes this relationship. Log aT = r rT T C T T C 2 1) ( (2.2) The temperature shift function, aT, corresponds to frequency (Gedde 1999; Emran 2000; McCrum 1967) and is expressed as: aT = T/ s = T/Ts ( is the average relaxation times at T and Tr ), Tr is the reference temperature or the Tg value, and T is the given temperature. The WLF universal constants are expressed as C1 with a value of 17.44 and C2 with a value of 51.6. The universal cons tants of the WLF equation are used to determine the fractional free volume (fg) and the coefficient of thermal expansion (f) of the polymer in the primary relaxa tion region (equations 2.3 and 2.4) (Gedde 1999, McCrum 1967, Emran 2000). fg = ) ( 303 21C B ( 2.3) f = 2C fg (2.4) Doolittle states that B = 1 (Doolittle 1952,Gedde 1999). A ccording to equation (2.3), fg is 0.025 for all polymers, a theory first proposed by Fox and Flory (1950); however, studies have shown that these values can vary w ith different polymer systems (McCrum 1967, Ferry 1961). The temperature dependency of the relaxation in amorphous polymers follows WLF. The temperature dependency of sec ondary relaxations of amorphous polymers and relaxations in semi crystalline polymers typi cally conform to a linear behavior and are obtained by using the Arrhenius equation (2.5) (McCrum, Read and Williams 1967).

PAGE 37

14 ln f = ln fo – RT Ea (2.5) Where f is the frequency, fo is the reference frequency, R is the gas constant (1.987 cal/mol K), and T is the temperature in Kelvin (Gedde 1999; Meier 1978; Aklonis, McKnight, and Shen 1972). Figure 2.1 depict s WLF and Arrhenius behavior when the log of the frequency is plotted against temperature. Figure 2.1. Plot of typica l WLF and Arrhenius behavior. The and relaxations are described as the part ial reorientation (rotation) of side groups attached to the main chain. These side groups can be either rigid or flexible in their attachment to the main chain and must contain dipoles (Runt and Fitzgerald 1997; McCrum, Read and Williams 1967). The extent of rotation of these side groups is characterized by the loss factor peaks plotte d against temperature. Energy required to

PAGE 38

15 rotate these groups (apparent activation energies) is determ ined from the slope of the Arrhenius plot which is derived by plotting the inverse of the temperature at peak maximum against the natural log of the frequency. Polymer Relaxations in poly (methyl meth acrylate) and poly(4-methyl-1-pentene) Poly (methyl methacrylate) (PMM A), an amorphous polymer, and poly (4-methyl-1-pentene) (P4M1P), a semicrystalline polymer, were used in this work. Amorphous polymers are highly transparent and resemble a glass. These structures contain chains that exist in a disordered state consisting of a tangled arrangement. On the other hand, semi-crystalline polymers exist in an ordered structure. Semi-crystalline polymers are not 100% crystalline; they contain amorphous regions that link the crystalline regions (Stevens 1999). Crys talline polymers usually have stronger interactions between chains as a result of being highly st ereo regular, having no chain branching or containing highly polar groups. They are also more resistant to solvents, tougher, stiffer, and more opaque than an amorphous polymer. However, P4M1P not only has non-polar groups but is transparent in its isotatic arrangement (Lopez et. al 1992). The structure and group arrangement of a polymer has a great effect on the polymer’s property. For example, the gla ss transition temperature will increase with increased cross-linking density, chain symmetry, number and size of bulky side groups, or the number of polar groups. A decrease in Tg will occur if the side groups are flexible, if there is no chain symmetry, or upon th e addition of a plasti cizer (Seymour and Carraher 1988). Certain properties can be obtained from mechanical and dielectric analysis as a result of the molecular motions or torsiona l vibrations and large scale rotations of molecular groups (Bershtein and Egorov 1994) induced by external forces as a function of time, frequency, and temperature. When an applied force is introduced to the sample and then removed, the polymer molecules return to their original undisturbed state. This process is known as a molecular relaxation (or transition) (Bershtein and Egorov 1994; Stevens 1999; McCrum, Read and Williams 1967). Molecular relaxations are denoted by and symbols (Bershtein and Egorov

PAGE 39

16 1994; Stevens 1999; McCrum 1967). The region occurs at high temperatures and high frequencies and is referred to as the primary process. The and regions occur at decreasing temperatures and are known as s econdary processes or sub-glass transitions (McCrum, Read and Williams 1967; Gedde 1999). The region in an amorphous polymer is indicative of main chain segmen tal motion, and is related to the glass transition temperature (Tg). This is the region in which the polymer undergoes softening and moves from a glassy state to a rubbery or molten state. The secondary regions, and, are defined as the hindered rotations of side groups in an amorphous polymer. However, in a crystalline polymer, the region is considered th e crystalline portion of the polymer and is located near or around the melting temperature (Tm). The melting temperature occurs when the polymer chains, which are arranged in ordered crystals, fall out of their aligned state and move into a disordered state in which the polymer chains begin to move without restraint. The region in a crystalline polymer is often referred to as a and is considered the amorphous portion of the polymer or the glass transition region (Gedde 1999; McCrum, Read and Williams 1967). Poly (methyl methacrylate) (PMMA) exhibits and relaxations under mechanical force and only and relaxations with a subsequent merging of the and relaxations to produce the relaxation under dielectric a nd mechanical forces. The structure of PMMA is depicted in Figure 2.2. The transition is the segmental motion of the polymer main chain, and the transition is the hindered rotation of the ester side group of the C-C bond. The relaxation is defined as the rotation of either the methyl attached to the ester side chain or the -methyl attached to the main chain. The relaxation is not typically seen in dielectric spectroscopy due to the lack of polarity of the methyl side groups. Indicative of ma ny lower polyalkyl methacrylates, the relaxation is the dominate loss peak both dielectrically a nd mechanically (McCrum 1967, Ishida 1969, Sasabe and Saito et al. 1968). Poly (4-methyl-1-pentene) (P4M1P) is dielec trically inactive due to the lack of polar groups in the structure. However, a study conducted by Lee and Hiltz (1984) characterizes P4M1P under dielec tric analysis (Lee and Hiltz 1984). It has been reported

PAGE 40

17 that the polymer exhibits two mechanical relaxations in the and regions (Woodard, Sauer and Wall 1961). The relaxation is the gla ss transition region (52oC-67oC) and the is the low temperature (-123oC) transition. Several literature sources refer to the mechanical and dielectric gl ass transition region as the a region and the low temperature relaxation as sc)In this study, the glass transition re gion of PMP will be referred to as (a) or a and the low temperature peak as Literature has also observed a relaxation at high temperatures above 125oC; this transition will be referred to as the c transition (Choy et al. 1981). Figure 2.3 is an illustration of the stru cture of P4M1P. A more in depth discussion on the mechanical relaxation behavior of P4M1P will be discussed in Chapter 7.

PAGE 41

18 C CH 2 C O O CH3CH3n Figure 2.2 Structure and relaxations in poly(methyl methacrylate). (a) alpha () relaxation, (b) beta () relaxation, and (c) gamma () relaxation. C CH 2 C O O CH 3 CH 3 n C CH 2 C O O CH 3 CH 3 n

PAGE 42

19 Figure 2.3. Structure and relaxatio ns of poly (4methyl-1-pentene). CH2CH CH2 CH n CH3CH3 (a)

PAGE 43

20 Instrumentation Thermal Analysis A polymer’s transitions induced by dielectr ic or mechanical forces are expressed as loss peaks when characterized under ther mal analysis (TA Instruments 1998, Bershtein and Egorov 1994). A detailed explanation of the instrumentation used to characterize these polymer transitions via thermal analysis is discussed in the following section. This section is followed by a discussion on othe r techniques used to characterize polymers that, when compounded with thermal analys is, provides additional information on the structure-property relationship of the material. Differential Scanning Calorimetry (DSC) Differential Scanning Calorimetry (DSC) is designed to measure temperature and heat flow changes related to material tran sitions as a function of time and temperature (TA Instruments 1998). DSC is the most common technique employed to perform thermal analyst on materials and is used to measure the glass and melting transition temperatures, heat capacity, and heat of tran sition, degree of crystallinity, cross linking, degree of cure, oxidative stability of polymer s, kinetics, and purity (TA Instruments 1998; Hohne, Hemminger, and Flammersheim 1996). Samples of varying compositions such as films, fibers, powders, solutions and composites can be analyzed via DSC (TA Instruments 1998). The basic operational theory of the DSC is based on a difference in heat flow between a sample and a reference. A sample pan consisting of a small quantity of sample (between 2 – 10 mg) and an empty reference pan of exact or similar weight to the sample pan prior to the addition of sample sit on elevated su rfaces of the thermoelectric (constantan) disk. The thermoelectric disk is composed of metallic material and allows for heat transfer. It is located above th e heating block as shown in Figure 2.4 (TA Instruments 1998).

PAGE 44

21 Figure 2.4. Schematic of DSC cell. Taken from TA Instruments 1998. The differential heat flow to the sample and reference is monitored by thermocouples located underneath the constantan disk. The thermocouples m easure the differential heat flow using equation 2.6. DR T dt dQ (2.6) Where dQ/dt is the heat flow, T is the difference in temperature between the sample and reference, and RD is the thermal resistance of the c onstantan disk (TA Instruments 1998). Quantitative data are plotted as heat flow (dQ/dt) against temperature. When the sample pan and reference pan are placed under an inert environment and heat is conducted at the same rate, th e amount of heat increase will be the same. However, if the sample experiences a heat related change (i.e. morphological changes), heat will either be absorbed or evolved, thus exhibiting an endothermic or exothermic process. An endothermic process occurs in polymers at their glass transition temperature

PAGE 45

22 (Tg). Polymers at Tg exhibit absorption of heat and an increase in heat capacity. This is seen as a downward shift in the baseline of the thermograms (plots) (TA Instruments 1998; Collins, Bares and Billmeyer 1973; Stevens 1999) as shown in Figure 2.5. Another endothermic process is the temperature at which semi crystalline polymers melt (Tm). An exothermic process occurs in semi crystalline polymers at their crystallization temperature (Tc). Figure 2.5. Polymer transitions characterized via Differential Scanning Calorimetry. Illustration taken fr om TA Instruments DSC Brochure 2004. Polymer morphology can be affected by pro cessing. Thus, to ensure that all samples have the same thermal history, heating above the transition temperatures are performed twice. Samples are initially heat ed above their glass tr ansition and/or melting temperatures, cooled below their Tg and heated again. The gl ass transition temperatures and melting temperatures are taking from this second heating. A TA Instruments DSC 2920 was used to obtain thermal properties. Collected data were analyzed on TA Instruments software Universal Analysis 2000. The glass transition temperature is taken as the infl ection point of the curve. The melting temperature is taken by performing a linear integration and norma lizing the area above the endothermic melting peak. The percent cr ystallinity can also be determined by the software. The equation (2.7) uses the heat of fusion value in joules /gram (standard heat) of a 100% crystalline poly mer (TA Instruments 1998).

PAGE 46

23 % crystallinity = (area/standard heat) x 100 (2.7) Prior to the collection of data, the base line is calibrated using TA Instruments DSC Calibration Program 2000 Baseline calibration involve s measuring the heat flow in an empty cell with set parameters on the te mperature range and heating rate needed for data collection. All samples in a data set s hould be collected usi ng the same baseline. Dielectric Analysis (DEA) Dielectric Analysis is used to measur e the electric response of materials when placed in an electric field. As with all thermal analysis of pol ymers, dielectric behavior is a function of frequency, time, and te mperature (TA Instru ments 1998). Although extremely useful in analyzing properties such as frequency dependence of thermal transitions, degree of cure, activation energi es, resin flow and cure to name a few, dielectric analysis is limited to polymers that contain dipoles. Figur es 2.2 and 2.3 are the molecular structures of poly (methyl me thacrylate) (PMMA) and poly (4-methyl -1 pentene) (P4M1P). PMMA has dipoles on the ester group and P4M1P does not have any dipoles, thus PMMA is an ideal polymer to analyze via diel ectric analysis and P4M1P is considered dielectrically inact ive. Further, the conductive nature of carbon nanotubes is believed to further enhance the dielectric response of non conductive polymers (Clayton 2005; Tatro 2004; Zhao, Wood and Wagner 2001). The direct properties measured under di electric analysis are capacitance and conductance. Capacitance is a material’s ability to store elec trical charge and conductance is the ability to tr ansfer electrical charge. The measurement of capacitance and conductance allows the effect of an applie d electric field on polym er relaxations to be quantified through the relative permitt ivity or dielectric constant (e ) and the loss factor (e ). Manipulation of capacitance and conducta nce measurements was calculated via TA Instruments software program Universal Analysis 2000 and the analyzer used in this study was a TA Instruments DEA 2970.

PAGE 47

24 Debye first introduced concepts regarding th e dielectric behavior of material. The measurements quantified via dielectric an alysis are based on Debye’s equation for permittivity (2.8) and loss factor (2.10). Permittivity or dielectric constant ( e ’) is the extent of dipole alignment and is propor tional to capacitance. For parallel plate electrodes the permittivity is the ratio of the polymer’s capacitance in a dielectric cell to the capacitance of the same dielectric cell unde r vacuum and is expressed in equation 2.8 (TA Instruments 2000). 2) 2 ( 1 ) ( f e e e eu r u (2.8) Where eu is the unrelaxed permittivity due to induced dipoles, er is the relaxed permittivity, is the molecular relaxation time, f is the frequency, and 2f is considered the angular frequency. The permittivity due to dipole alignment is defined as 2) 2 ( 1 ) ( f e eu r (TA Instruments 1998). A e cd eo (2.9) Where c is capacitance, d is plate spacing, A is the electrode plate area, and oe is the constant representing the absolute permittivity of free space (8.85 x 10-12 F/m). The dielectric loss factor e represents the energy required to align the dipoles or move ions and is proportional to conductance (a materials abili ty to transfer electrical charge). Equation 2.11 repr esents the expression for parallel plate electrodes. o u rfe f f e e e 2 ) 2 ( 1 2 ) (2 (2.10)

PAGE 48

25 ofe RA d e2 (2.11) Where R is the resistance in ohms and is the ionic conductivity term. Above the glass transition region and melting temp eratures the polymer begins to flow and the behavior of the unrestricted ions can be analyzed (TA Instruments 1998), thus ionic conductivity can be plotted against temperature at high te mperatures. In add ition to the dielectric constant, the loss factor a nd the ionic conductivity, Tan or loss tangent can also be obtained. e e tan (2.12) Figures 2.6 to 2.9 are plots generated via TA Instruments Universal Analysis 2000 that represent plots of permittivity, loss factor loss tangent, and ionic conductivity against temperature of PMMA.

PAGE 49

26 Neat PMMA 1 Hz 100000 Hz 2 4 6 8Permittivity -150-5050150Temperature (C) Universal V3.4C TA Instruments Figure 2.6. DEA dielectric permittivity ( e’) plotted against temperature.

PAGE 50

27 Neat PMMA 1 Hz 100000 Hz 0 1Loss Factor -150-5050150Temperature (C) Universal V3.4C TA Instruments Figure 2.7. DEA dielectric loss factor ( e”) plotted against temperature.

PAGE 51

28 Neat PMMA 1 Hz 100000 Hz 0.0 0.2Tan Delta -150-5050150Temperature (C) Universal V3.4C TA Instruments Figure 2.8. DEA dielectric loss tangent ( ) plotted against temperature.

PAGE 52

29 Neat PMMA 1 Hz 100000 Hz 0 10000 20000 30000 40000Ionic Conductivity (pmho/cm) -150-5050150Temperature (C) Universal V3.4C TA Instruments Figure 2.9. DEA ionic conductivity ( ) plotted against temperature.

PAGE 53

30 The basic operational theory of the dielec tric analysis involves placing a sample between two gold plated sensor s or electrodes. A voltage in the form of a sinusoidal signal is applied which results in an alternating electric fi eld (TA Instruments 1998). The sample is then polarized or oriented at th e same frequency of the electric field; however, there exist a phase angle shift ( ) which is possibly a result of the local restriction of the dipole alignment which in turn delays the dielectric response of the material (Havriliak and Havriliak 1997; TA Instruments 2000). Th e applied voltage and the current can be compared to determine the phase angle shift (as shown in Figure 2.10). The subsequent current is then expressed as capacitance and conductance as shown in Figure 2.10. Figure 2.10. Representation of phase angl e shift between the applied voltage and current in dielectric analysis. Illustration taken from TA Instruments DEA Advantage Manual 1998.

PAGE 54

31 Figure 2.11. Dielectric cap acitance plotted against conductance. Illustration taken from TA Instruments DEA Advantage Manual 1998. In this study, the parallel plate sensors were used for all samples. Each sensor was cleaned and calibrated prior to use. The parallel plate sensor is primarily used to measure molecular relaxations and bulk prope rties of a polymer. Whereas single surface ceramic sensors are (according to TA Instrume nts) used to measure cure experiments and surface properties of polymers. The lower se nsor of the parallel electrodes (Figure 2.12) is placed on the furnace and is responsible for applying the voltage, resulting in an electric field that polarizes the sample. The top sensor is attached to the ram and measures the generated current. In order to control and correct for noise as a result of stray capacitance and fringing, a guard ring is located on the outer edges of the top sensor. The temperature is controlled and measured by the resistance temperature detector (RTD) which is the outer ring of the bottom sensor (TA Instruments 1998).

PAGE 55

32 Figure 2.12. DEA parallel plate sensor. Il lustration taken from TA Instruments DEA Advantage Manual 1998. Dynamic Mechanical Analysis (DMA) Dynamic mechanical analysis is used to measure viscoelastic behavior. Polymers exhibit viscoelastic properties because th eir behavior under applied stresses is a combination of a true elastic solid and a true liquid (T A Instruments 1998; Aklonis, MacKnight and Shen 1972; Sepe 1998). When a fo rce or stress is applied to a true elastic solid the material will deform, but will recover completely when the force is removed. A true liquid will deform under stress and will not recover when the stress is removed (TA Instruments 1998; Aklonis, MacKnight and Sh en 1972; Sepe 1998). However, when an external force or stress is applied and remove d, a viscoelastic entity will exhibit both elasticity and flow. The deformation is de pendent on time and temperature and creates a strain (TA Instruments 1998; Aklonis, M acKnight and Shen 1972; Sepe 1998). The portion of the strain that recovers when the stress is rem oved is considered the elastic (E’) portion and represents the ener gy (applied stress) stored by th e material. The portion that does not recover is the viscous (E”) portion and represents the dissipation or dispersion of applied energy (TA Instruments 1998). The dynamic mechanical analyzer can be used to measure in transient mode (Creep and Stress Relaxation) or oscillatory (dynamic mechan ical). Creep measurements involve applying a constant stress and measurin g the resulting strain as a function of time (TA Instruments 1998). Stress relaxation applies a strain and the stress required to

PAGE 56

33 maintain that strain is measured as a functi on of time. In this work, dynamic mechanical analysis was the testing method used. An os cillatory force (frequency) was applied and the resulting deformation was measured. Base d on the behavior of the oscillating phase angle shift stress and strain valu es can be calculated. In an elastic solid the phase angle shift would be 0o; for an ideal liquid the shift would be 90o; and for a viscoelastic polymer, the phase angle shift would fall between 0o and 90o. Figures 2.13 and 2.14 represent the oscilla tory behavior of all three scenarios (TA Instruments 1998). Figure 2.13. Mechanical phase angle shif ts for an ideal liquid an elastic solid. Illustration taken from TA Inst ruments DMA Advantage Manual 1998.

PAGE 57

34 Figure 2.14. Mechanical phase angle shifts for a viscoelastic polymer. Illustration taken from TA Inst ruments DMA Advantage Manual 1998. The modulus is the ratio of stre ss to strain. The stress generate d from an applied strain is the complex stress ( *) and can be used to calculat e the complex modulus (E*). The complex modulus defines a materi als resistance to a deformati on and is characteristic of both elastic and viscous propert ies. The phase angle shift is then used to obtain the elastic stress ( ’) and viscous stress ( ”). Subsequently, the el astic storage modulus (E’) and the loss modulus ( E”) can calculated from ’ and ” ( TA Instruments 1998). A Dynamic Mechanical Analyzer 2980 was us ed in the studies presented in this body of work. It can be used to analyze the vi scoelastic properties of samples of various shapes and sizes using different clamps. More over, since the modulus of a material is not dependent on the dimensions, the DMA 2980 inst rument control software calculates the modulus depending on the clamp type. Because the modulus is a measure of how stiff a material is, these equations take into account the sample stiffness (K ) which is dependent on geometry. K is a materials ability to resi st deformation and is defined as the force applied divided by the amplitude of the deform ation. The tension film/fiber clamp was used in this study. The tension film can be used for samples that are 2 mm thick or less. The modulus is then obtained by

PAGE 58

35 A L K Es (2.13) Where Ks is the measured stiffness. The stress and strain calculations for the tension clamp are as follows: o oA P (2.14) o oL L (2.15) Where o is stress, o is strain, P is the applied force, L is the cumulative change in sample length, Lo is the initial sample length, and Ao. Figure 2.15 is a graphic repr esentation of the tension cl amp (TA Instruments 1998). Viscoelastic behavior is time (frequency) and temperature dependent; thus, for all studies presented, samples with estimated dimensi ons of 19 x 6 x 2 mm we re measured from temperatures ranging from -150oC to 300oC and frequencies from 1 Hz to 100 Hz. For all samples the amplitude was set at 5 microns.

PAGE 59

36 Figure 2.15. Dynamic mechanical tension f ilm clamp. Illustration taken from TA Instruments DMA Advantage Manual 1998. Typically, the storage modulus and loss modulus are plo tted against temperature (Figure 2.16). The storage modulus provide s useful information regarding structure property relationships and failure analysis of the sample (TA Instruments 1998; Sepe 1998). The loss modulus identifies regions of structural mobility as expressed through molecular relaxations and their behavior dependence on time, frequency, and temperature.

PAGE 60

37 0 100 200 300 400 500[ ] Loss Modulus (MPa) 0 2000 4000 6000 8000 10000Storage Modulus (MPa) -150 -100 -50 0 50 100 150Temperature (C) Universal V3.4C TA Instruments Figure 2.16. Storage modulus (E) a nd loss modulus (E) plotted against temperature. Illustration taken from TA Instruments DMA Advantage Manual 1998.

PAGE 61

38 Spectroscopy Optical spectroscopy is the study of the in teraction between molecular species and radiant energy in the ultraviolet, visible, and infrared regions of the electromagnetic spectrum (ES). The six primary occurrences in optical spectroscopy are absorption, emission, scattering, fluorescence, phos phorescence, and chemiluminescence (Skoog and Leary 1992). Further, in order to obtain spectra, a typical spectroscope must include an energy source, a transparent container for the sample, a device that isolates the specified region of the ES in question and lastly, a detector (Skoog and L eary 1992). Absorption spectroscopy in the ultr aviolet and visible regions was used in this body of work (Chapter 3) and thus will be the focus of discussion in this section. Absorption spectroscopy is the analysis of the amount of light transmitted by an absorbing species when placed between a li ght source and a detector (Rao 1967). The absorption of this light energy results in a ch ange in energy in the electronic, vibrational or rotational behavior present in the absorbi ng species. Therefore, the increase in energy is thus the same as the energy of the photon and can be mathematically expressed by the following equation: E=h= hc 2.16 Where h is Planck’s constant (6.63x10-34 m2 kg/s), is the frequency of the radiation, is the wavelength, and c is the speed of light (3.00x108 m/s) ( Rao 1967). Electronic transitions involve the absorption of the greatest amount of energy and rotational transitions involve the least amount of energy (Rao 1967). Species that absorb in the far infrared region will experience a change in rota tional energy; those that absorb in the near infrared region will experience rotational and vi brational changes; changes in vibrational, rotational and electronic energies will occur in species that absorb in the ultraviolet region (Rao 1967, Skoog and Leary 1992).

PAGE 62

39 Ultraviolet Visible Spe ctroscopy (UV-Vis) Ultraviolet visible (UV-Vi s) spectroscopy has been a helpful aid in polymer characterization. UV-Vis spectroscopy can a ssist in identifying unr eacted monomer, as well as the presence of initiator and inhib itors. It can also be used to determine copolymer composition, and end group com position (Stevens 1999). The ultraviolet visible region of the electromagnetic spect rum ranges from 200 nm to 800 nm (Carey 1996; Skoog and Leary 1992; Rao 1967). The near ultraviolet region ranges from 200 nm to 400 nm and the vacuum of far ultraviole t region is below 200 nm (Rao 1967). The visible region ranges from 400 nm to 800 nm. The theoretical basis of UV-Vis spectroscopy is to identify electronic transi tions between energy levels. In order to accomplish this a molecule must contain a chromophore (unsaturated bonds) such as C=C, C=O, and N=N (Rao 1967) or unshared ou ter electrons localized about the atom ( i.e., O, N) (Skoog and Leary 1992). These ch romophores must interact with and absorb energy at a particular wavelength. These ch romophores contain elec trons that are easily excited from a low energy bonding orbital to a high energy non bonding orbital (Skoog and Leary 1992, Rao 1967). Once ex cited, electrons can go through *, n*, n* or transitions. An electron that undergoes a absorbs in the vacuum or far ultraviolet region; thus, absorption at these short wavelengths require a large amount of energy to excite an electron in the bonding orbital to the antibonding orbital. Saturated molecules bonded to atoms wh ich contain nonbonding elec tron pairs exhibit n* transitions. These transitions require less energy than the transitions and usually occur in the far to near UVVIS re gion (150 nm to 250 nm) (Rao 1967, Skoog and Leary 1992). Finally, n* and transitions involve the absorption of radiation by saturated molecules occurring at longer wavelengths ( > 200nm) and requiring less energy (Rao 1967, Skoog and Leary 1992). The instrumentation used in the study de tailed in chapter 2 was a Hewlet Packard 8452A Diode Array Spectrophotometer and th e operating software was HP 89531A. A deuterium lamp was used to allow for abso rptions in the range of 190 nm to 820 nm.

PAGE 63

40 Light from the deuterium lamp passes in a single beam through a source lens, then proceeds through the sample in which energy is absorbed at certain wavelengths by absorbing species. Once the beam has left th e sample, it is scattere d onto the diode array detector by a grating, resulting in a quantifie d value for absorbance or transmittance as a function of wavelength (Hewle t Packard 1990, Emran 2000). For polymer characterization via UV-Vis th in films and polymer solutions can be used to obtain transmission or absorbance spec tra. For thin films, all films to be compared must be of the same thickness and ai r serves as the blank or reference. When analyzing polymer solutions, polymers must be di ssolved in a solvent that is then used as the reference blank. Solutions and solvents are placed in a quartz cuvette with a path length if 1.000 cm to obtain spectr a (Hewlet Packard 1990, Emaran 2000). Other Characterization Techniques Microhardness Hardness is a measure of a material’s resi stance to local deformation (Calleja and Fakirov 2000, Tabor 1951). Mi crohardness testing involves measurements with force loads that are less than 1 N (Calleja a nd Fakirov 2000). Micr ohardness testing of polymers is dependent on the viscoelastic behavior and thus provides supporting information regarding the mechanical propertie s such as strength, elasticity, and modulus of the material (Calleja and Fakirov 2 000). There are three aspects of hardness measurements: scratch hardness, indentati on hardness and dynamic hardness. Dynamic hardness (the type used in th is study) involves applying a spec ified force to the material’s surface for a set amount of time resulting in an indentation of a particular size. With dynamic testing there are several methods used to obtain micohardne ss values: Brinell, Vickers, Knoop, Rockwell, and Scleroscopy. These methods differ by the shape of the indenter (Calleja and Fakirov 2000). A Lei ca VMHT MOT with a Vickers indenter was used. The Vickers indenter is a squa re pyramid of diamond and the apex angles between non-adjacent sides measure 136o (Leica 1999).

PAGE 64

41 The Vickers microhardness (HV) valu es are determined by the following equation: 2 24 1854 2 sin 2d F d F A F HV (2.17) Where F is the applied force, A is the surf ace area of the imprint, d is the average diagonal length of the imprint, and is the angle (Leica 1999; Calleja and Fakirov 2000). Supporting Instrumentation Several characterization techniques were performed by outside agencies. These characterization techniques will be discussed briefly in the following section. Index of Refraction The interaction between a materi al and light is quantified as its refractive index and is defined as ni = iv c (2.18) where n is the refractive index, i is the frequency, v is the velocity of light in the medium and c is the velocity of light in a vacuum (Skoog and Leary 1992). When light passes at a certain angle between two transparent medi ums with different densities the light undergoes a change in velocity and refrac tion (direction) (Skoog and Leary 1992). The degree of refraction is ex pressed by Snell’s Law: 2 1 1 2 2 1sin sin v v n n (2.19)

PAGE 65

42 where 1 and 2 are the angles of incidence and refraction, n1 and n2 are the refractive indices of the incident an d refracting mediums, and v1 and v2 are the velocities of light moving through the incident and refr acting mediums (Skoog and Leary 1992). The refractive index can be measured by us ing a vacuum or air as the reference, however, air is most often used, and thus equation 2.19 is simplified to 2 1sin ) (sin air (2.20) The refractive index squared is equal to the dielectric constant of a polymer at low frequencies (Van Krevlan 1990). This relations hip is discussed in detail in Chapter 3 of this book. Samples used in the study discu ssed in Chapter 3 were sent to an outside agency (Optical Polymer Research, Gainesvi lle FL) for refractometry measurements. Measurements were obtained on an Inde x Instruments CLR 12-70 Contact Lens Refractometer. Gel Permeation Chromatography (GPC) Gel Permeation Chromatography (GPC) is also referred to as size exclusion chromatography (SEC). It is used in polymer technology to determine polymer molecular weights and molecular weight di stribution (polydispers ity) (Stevens 1999; Skoog, West and Holler 1992). A polymer dissolv ed in a column safe solvent such as tetrahydrofuran (THF) is injected into a co lumn packed with a hi ghly porous material. This material is a cross-linked polymer (usually polystyrene cross-linked with divinylbenzene) that will not chemically reac t with the polymer to be tested (Stevens 1999; Rosen 1993; Skoog and Leary 1992). Molecu lar weight values are obtained by the separation of polymer molecules according to size; however, it is also believed that separation occurs as a re sult of a polymer’s hydrodynamic volume (Stevens 1999,

PAGE 66

43 Burfield and Doi 1983). The smaller the polymer molecule, the longer it will take to pass through the column due to the ability of the small molecule to fit into the pores of the packing material (Stevens 1999; Skoog and Leary 1992). Once the molecules pass through the column a detector is used to de termine polymer fractions. The most common detectors are refractive index, infrared, or ultraviolet detectors (Stevens 1999). GPC was done by TSE Industrie s in St. Petersburg Flor ida. Molecular weights were determined only for neat polymer samples due to the lack of information on the interaction between the column and carbon nanot ubes. GPC is discussed in Chapter 3. Scanning Electron Microscopy (SEM) Scanning Electron Microscopy (SEM) is typically used to study surface topology such as the dispersion of pigments in paint, cracking of coatings, adhesive failures, and phase boundaries in polymer blends (Stevens 1999). Research in polymer nanotube composites has utilized SEM to investigat e the presence of nanotubes on the polymer surface (Tatro et al. 2002) and nanotubes embedd ed into the polymer matrix when a cross section portion is used (Wu and Fitzgerald et al 2000, Clayton 2005). In SEM an electron beam and a beam in a cathode ray tube is simultaneously scanned across the surface of the sample. A signal is then produced by scatte red electrons resulting in an image with a three dimensional appearance (Stevens 1999). SEM images were captured on a Hitach i S-800 scanning electron microscope. Images were taken of the polymer’s cro ss section and coated with 10-15 nm of gold/palladium alloy. The SEM is located in the Nanomaterials and Nanomanufacturing Research Center in the Department of Engin eering at the University of South Florida. Gamma Irradiator The JL Shepard Mark I cesium-137 gamma i rradiator is located at All Children’s Hospital in St. Petersburg, FL and is owned and maintained by the University of South Florida’s Radiation and Safety Office. According to doc umentation provided by the

PAGE 67

44 University of South Florida’s Radiat ion Safety Office, cesium-137 decays to barium-137, releasing a gamma ray. Scheme 2.1 is the process of decay and Figure 2.17 is a representation of the irradiator. The unstable cesium emits a beta particle as the neutron changes to a proton, forming the metastable form of 137Ba. The half life of metastable 137Ba is 2.6 minutes at which time a gamma ray is released and measured. The half life of 137Cesium is thirty years. 55Cs -156Ba (metastable) 56Ba (stable) (Scheme 2.1)

PAGE 68

45 JL Sheppard Cs-137 gamma irradiator (806-3) St. Pete ACH/ARC Position 7 6 5 4 3 2 1 Figure 2.17. Illustrati on of the JL Sheppard 137Cs Gamma Irradiator Samples were placed near position 1 (closest to the source). For the studies presented in this book (Chapters 3 and 4) rates were determined by placing dosimeters in a semicircular arrangement 6 cm from the s ource. The dosimeters were Harshaw TLD-400 (CaF2/Mn) thermoluminescent ribbon dosimeters (TLDs) with dimensions of 0.32 cm x 0.32 cm x 0.09 cm. Dose rates for each dosimeter were read on a Harshaw 5500 TLD reader (Tatro et al. 2002).

PAGE 69

46 CHAPTER 3 Transparent Poly (methyl methacrylate)/Si ngle walled carbon nanotube Composites with Increased Dielectric Consta nts: Polymerized via In-situ Free Radical Polymerization using Heat, UV Light, and Ionizing (Gamma) Radiation Initiation Sources This study explored a novel combination of incorporating in-situ polymerization (Chapter 1-Polymer Carbon Nanotube Compos ites), sonication, and three initiation sources (thermal energy (heat), UV light, a nd gamma radiation) to create polymer carbon nanotube composites. The monomer was polymerized in the presence of carbon nanotubes and initiated via the sources mentioned above. Single walled carbon nanotubes were used and the surfaces of the tubes were not modified prior to use. Further, the in-situ polymeriza tion process is believed to increase chemical adhesion at the polymer-carbon nanotube interface (Chapter 1) (Jia et al 1991). The combination of techniques used in this study resulted in th e production of films w ith increased dielectric properties while limiting the loss in optical transparency. Methyl methacrylate (MMA) monomer was us ed as the polymer matrix. MMA is a vinyl monomer that is polym erized via radical mechanisms (Reetz, Yagci, and Mishra 1998). Free radical vinyl polym erization proceeds via th ree major processes: (1) initiation, (2) propagation, a nd (3) termination (Yagci and Mishra 1998). As noted by Reetz, Yagci, and Mishra these three initiation sources can initiate polymerization by producing free radicals. Thermal and UV sources supply enough energy to cause bond breakage of atomic bonds. High energy ( rays, x rays, and energy rich particle rays) initiation results in the transfer of electrons from ions or atoms to an acceptor molecule that then undergoes bond di ssociation (Reetz, Yagci, a nd Mishra 1998). This is accomplished either by (a) the Compton Effect (b) the direct interaction between an extracted electron and the atom, or (c) the di rect homolytic bond breakage (Reetz, Yagci, and Mishra 1998; Skoog and Leary 1992). The major difference in these three methods is in the form of energy supplied for the initiation step; propagation, transfer and

PAGE 70

47 termination are in effect identical proce sses (Reetz, Yagci, an d Mishra 1998). Polymerizing PMMA via free radical mechanisms has been shown to result in syndiotactic character of about 60% (McC rum 1967; Bovey and Tiers 1960; Fox and Schnecko 1962). Further, as the polymerizati on temperature is lowered, there is an increase in syndiotacticity; pol ymerizations occurring below 0oC can exist in both the syndiotactic or isotactic form and can undergo crystallization (McCrum 1967). Syndiotacticity occurs when si de groups are attached to the polymer backbone in an alternating arrangement (Stevens 1999; Se ymour and Carraher 1988) (Appendix A). The location of side groups on the chain, th eir lengths, bulkiness a nd degree of branching may hinder rotation, lead to plasticizing eff ects, influence chain entanglement and may alter the free volume of the polymer; thus aff ecting the chemical and physical properties of the polymer (Stevens 1999, McCrum 1967). The isotactic form of PMMA has a lower glass transition temperature (~ 45oC) than the syndiotactic form (~ 105-115oC). Dielectric data show that the process (glass transition region) in PMMA occurs at a lower temperature in the isot actic form than in the syndiotactic form. Further, the process is the stronger relaxa tion in syndiotactic PMMA and the weaker relaxation in the isotactic form (McCrum 1967). The experimental work presented in this chapter focuses on applying conventional polymerization, processing, and characteriza tion techniques to a new phase of polymer composites: composites that fuse material that exists within the nanoand macrolength scales. Differential Scanning Calorimetry was used to determine the glass transition temperatures of all samples. Dielectric Analysis (DEA) was em ployed to understand the effect of electrical conduc tive nanotubes on an insulating polymer. DEA dielectric constant values were also used to correlate with the refractive index values to provide information on nanotube effect on the polariz ation of the polar groups in PMMA as related to optical dispersion. Dynamic Mechanical Analys is (DMA) characterized the mechanical properties of the neat and compos ite samples as a function of temperature and frequency; microhardness was used as s upporting data to DMA and determined mechanical properties as a function of load a pplied to the surface at room temperature. Lastly, Ultraviolet visible spectroscopy was util ized to determine the optical transparency

PAGE 71

48 and quality of dispersion of the composite s. This work provides useful design, fabrication, and characterizati on methods that contribute to the future use of polymer nanotube composites in i ndustrial applications. The production of polymer nanotube composites with increased optical transparency and dielectric cons tants provides practical material s for specific applications such as electromagnetic shielding and elec tronic charge dissipati on (RW Evans et al. 1997 NASA report; Joseph Smith et al. 2003 NAS A report). For instance, spacecraft design must include electrical conductive ma terial that will limit any electromagnetic interference generated by the d eep space environment. Further, Smith et al.(2003) stated that low color/solar absorptivity () while maintaining electr ical conductivity (for electrostatic charge dissipation) are desired pr operties for the design of materials used for spacecraft equipment. Polymers serve as ideal replacements for electrically conductive metals for spacecrafts because they are lightwe ight, inexpensive, easy to process, and can be doped with conductive fillers that enhance their conductivity (Evan et al. 1997).

PAGE 72

49 Experimental Materials The methyl methacrylate monomer and the column packing material used to remove the monomethyl ether hydroquinone (M EHQ) inhibitor were all purchased from Sigma Aldrich (Milwaukee, WI). The UV photoinitiator, 1-phenyl -2-hydroxy-2-methyl1 propanone (Benacure 1173), was kindly provi ded by the Mayzo Corporation (Norcross, GA). Purified laser ablated single-walled carbon nanotubes (SWNT) were provided by the Center for Nanotechnology/NASA Ames Co rporation (Moffet Field, CA). The Certified A.C.S. grade methylene chloride (d ichloromethane) solvent was purchased from Fisher Scientific (Pittsburgh, PA). Single-Walled Carbon Nanotube Preparation Raw laser ablation material provi ded by NASA Johnson Space Center was purified according to a previously published procedure (Liu et al. 1998). The raw carbon nanotube material was refluxed in 2.6 M nitr ic acid for approximately 160 hours and then diluted with double distilled water. This so lution was then centrif uged at 4000 rpm. The solvent mixture was decanted and the sample was again suspended in double distilled water. This step was repeated two more ti mes in order to remove the acid from the nanotubes. The solution was th en filtered through a cellulose nitrate filter and dried at 60C in a vacuum oven to form a buckypaper. Polymer-Nanotube Composite Synthesis Poly (methyl methacrylate)/singlewalled carbon nanotube (PMMA/SWNT) composites were prepared via in-situ polymer ization. Pure SWNTs, 0.26% by weight, were placed in 9.5 g of deinhibited met hyl methacrylate monomer. The mixture was sonicated with a Branson Sonifier 450 sonicator (see Chapter 4 for illustration) until the nanotubes were finely dispersed in th e methyl methacrylate monomer. The

PAGE 73

50 monomer/SWNT mixtures were then placed in glass sample vials. A 0.5% (w/w) concentration amount of th e photoinitiator, 1-phenyl-2hydroxy-2-methyl-1 propanone (Benacure 1173), was added to the monomer/SWNT mixture after sonication. In order to remove any presence of oxygen gas, nitrogen gas was bubbled through the mixture for 1 minute. Methods of Polymerization Free radicals were created as a result of the cleavage in which the photoinitiator, 1-phenyl-2-hydroxy-2-methyl-1 propanone, undergoes when exposed to UV light (Appendix A) (Bradley, eds. 1998). The initiator was designed to initiate photopolymerization; however, this material was effectively us ed for all three initiation methods. UV Polymerization : a spectroline UV light (=220nm-280nm; E = 9.03 x 1019J7.10 x 10-19J) was used to expose the methyl methacrylate monomer in the presence of single-walled nanotubes (MMA/SWNT) to short wave UV radiation for five hours. The sample was placed 19 mm from the UV source (Figure 3.1). Thermal Polymerization: the MMA/SWNT mixture was heated in an oven at 70oC for 16 hours. Polymerization via Gamma radiation : a cesium-137 gamma source (Chapter 2) was used to expose the MMA/SWNT mixture to gamma radiation for 41 hours at a dose rate of 985 rads/min and a total dose of 2.42 Mr ads. All samples were post-cured in the oven for 4 hours at a temperature of 70oC. Neat PMMA sample s were prepared in a similar manner. After the polymerizations, the neat and co mposite samples were then dissolved in methylene chloride to make a 5% solution (w/w ). The samples were placed in a vacuum oven with a liquid nitrogen tr ap for a total of eight da ys at a temperature of 80oC to remove any residual solvent.

PAGE 74

51 Figure 3.1. Illustration of UV Polymerization process.

PAGE 75

52 Molding Samples characterized via dynamic mechani cal analysis and dielectric analysis were compression molded in a Carver Press at 135oC and 3000 pounds of pressure. Differential Scanning Calorimetry A TA Instruments 2920 Differential Sca nning Calorimeter (DSC) was used to obtain glass transiti on temperatures (Tg) of the polymer and composite samples. A sample amount between 2 –10 mg was obtained from the solvent evaporated film. The samples were heated to 145oC at a heating rate of 10oC per minute to ensure that all samples had the same thermal history. The samples were then cooled with liquid nitrogen to room temperature and reheated to 145oC. The Tg values were taken from the second heat. Gel Permeation Chromatogrphy (GPC) Molecular weight values were collect ed via gel permeation chromatography (GPC). Measurements were taken on a Perk in Elmer series 200 Liquid Chromatograph with two PLgel 5m mixed columns in series. The GPC was equipped with a Perkin Elmer series 200 Refractive Index Detector a nd a Perkin Elmer 785A UV-VIS Detector. Solvent evaporated films of the neat samples were analyzed and the GPC was calibrated with polymer standards. Each sample wa s dissolved in tetrahydrofuran (THF) and measured with a flow rate of 0.9 mL/min. Ultraviolet Visible Spectrphotometer A HP 8452A spectrophotometer was used to record ultraviole t-visible spectra. Films were prepared from PMMA/SWNT co mposite/methylene chloride solutions via solvent evaporation. Solvent evaporated films with a thickness of 0.127 mm were

PAGE 76

53 scanned with air as the b ackground. The PMMA/SWNT/CH2Cl2 solution was placed in a quartz cuvette with a 1 cm path length to obtain UVVIS spectra. Methylene chloride served as the reference blank. Dielectric Analysis Dielectric data were collected on a TA Instruments 2970 DEA using parallel plate sensors. Compression molded samples had an inner diameter of 27 mm and thicknesses that ranged from 0.4 to 1.7 mm. Samples we re scanned in the temperature range between 200 and -150oC with increments of –5oC under a nitrogen purge. Scanning frequencies ranged from 1 Hz to 1.0 x 105 Hz. A maximum force of 250 N was applied to all samples for the entire scan. Refractive Index Films with a thickness of 0.127 mm were measured on an Index Instruments, CLR 12-70 contact lens refract ometer to obtain refractive index values at 589 nm at 25oC. The instrument sensitivity was + 0.005. Dynamic Mechanical Analysis Mechanical data were collected on a TA Instruments 2980 Dynamic Mechanical Analyzer in the tension film mode. Measurem ents were taken from 1 Hz to 100 Hz and 150oC to 190oC in 5oC increments. The average sample dimensions were 19 x 6 x 2 mm.

PAGE 77

54 Microhardness The Vickers hardness number (HV) for each sample was determined with a Leica VMHT MOT with a Vickers indenter. The valu es were taken from the average of four indents. A horizontal and a vertical read ing was taken on each indent. A load of 500g and a dwell time of 20s was used. Results and Discussion Differential Scanning Calorimetry Glass transition temperatures (Tg) were obtained for all samples (Table 3.1). Figures 3.2-3.7 represent DSC thermograms of the solvent evaporated films. The Tg values for the UV and gamma polymerized n eat PMMA samples were lower than their composite counterparts. Literature states that the glass transiti on temperature of the polymer matrix increases with the addition of single-walle d carbon nanotubes (Harmon et al. 2001; Muisener et al. 2002; Wei, Srivas tava, and Cho 2002); however, the increase in this study was not significant due to the low concentrations of carbon nanotubes used. Table 3.1. DSC data. Gla ss Transition Temperatures (Tg) for PMMA and PMMA/SWNT samples polymerized via thermal, UV and initiation sources. Sample Tg(oC) Sample Tg(oC) Thermal Neat PMMA 118 Thermal PMMA/SWNT Composite 118 UV Neat PMMA 119 UV PMMA/SWNT Composite 122 Neat PMMA 120 PMMA/SWNT Composite 121

PAGE 78

55 118.60C(I)112.90C 123.70CUV polymerized neat PMMA -0.4 -03 -02 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V3.4C TA In s Figure 3.2. DSC data. Tg value for UV polymerized neat PMMA.

PAGE 79

56 121.88C(I)115.13C 124.02CUV polymerized PMMA/SWNT Composite -0.1 0.0 0.1 0.2 0.3Heat Flow (W/g) 20406080100120140160Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 3.3. DSC data. Tg value for UV polymerized PMMA/SWNT composite.

PAGE 80

57 119.72C(I)114.78C 123.64CGamma polymerized neat PMMA -0.3 -0.2 -0.1 0.0 0.1Heat Flow (W/g) 20406080100120140160Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 3.4. DSC data. Tg value for Gamma () polymerized neat PMMA.

PAGE 81

58 121.30C(I)11501C 12526CGamma polymerized PMMA/SWNT Composites -0.3 -0.2 -0.1 0.0 0.1Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 3.5. DSC data. Tg value for Gamma () polymerized PMMA/SWNT composite.

PAGE 82

59 11767C(I) 111.88C 121.47CThermally polymerized PMMA -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 3.6. DSC data. Tg value for the thermally polymerized neat PMMA.

PAGE 83

60 118.18C(I) 112.65C 122.64CThermally polymerized PMMA/SWNT Composite -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20406080100120140160Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 3.7. DSC data. Tg value for thermally polymerized PMMA/SWNT Composite.

PAGE 84

61 Gel Permeation Chromatography Weight average (Mw) and Number average (Mn) molecular weights as well as the polydispersity (Mw/Mn) of the neat PMMA samples pol ymerized via all three methods were determined and are listed in Table 3.2. The sample polymerized via thermal energy exhibited the highest weight av erage molecular weight of 3.0x106 g mol-1 and a polydispersity of 4.4. The molecular we ights of UV and gamma polymerized samples are 2.9 x 105 and 5.4 x 105 g mol-1 with polydispersities of 1.8 and 1.9, respectively. The high molecular weight of the sample thermally initiated is a result of the gel effect or Tromsdorff effect which is a typical occu rrence for the thermal polymerization of bulk methyl methacrylate (Ebewele 2000; Stevens 1990). The molecular weight study proves that the initiator, 1phenyl-2-hydroxy-2-methyl-1 propanone can be successfully used for all three polymerization techniques. Due to the possible interactions between the carbon nanotubes and the column packing materials, composite materials were not tested. Table 3.2. GPC Results for UV, and Heat Polymerized PMMA and PMMA/SWNT Composites. Sample Mn Mw ( g mol-1) Mw/Mn UV neat 1.6 x 105 2.9 x 105 1.8 Gamma neat 2.8 x 105 5.4 x 105 1.9 Thermal neat 7.1 x 105 3.0 x 106 4.4

PAGE 85

62 Ultraviolet-Visible Spectroscopy Transparent films were produced from the PMMA/SWNT composite/methylene chloride solution as depicted in Figure 3.8. These PMMA/SWNT films are compared to the neat PMMA prepared by the same ma nner and a PMMA/SWNT composite prepared via melt blending (Harmon et al. 2001; Muisener et al. 2002). Figure 3.9 shows that the UV-Vis spectra of all three composites exhibited a transmittance of 50% and higher at or above 300 nm. Transparent PMMA/SWNT composite films were achieved only if films were prepared immediately after dissolution. Figure 3.10 are UV-Vis spectra of the PMMA/SWNT/CH2Cl2 solution over time. After dissolution, the solution had an observable light grey tint w ith no evident carbon nanotubes other than the color. However, over time the carbon nanotubes agglomerated, resulting in a decrease in transmittance. If the solvent is evaporated prior to reagglomeration, films with limited loss in transparency results. If the agglomer ated solutions are agitated, the particulates break apart and solutions can be cast into films with limited loss in transparency. Figure 3.8. Films (1.5 mm) of (a) neat PMMA (b) heat polymerized composite, (c) polymerized composite, (d) UV polymerized composite.

PAGE 86

63 0 20 40 60 80 100 200300400500600700800wavenumber (nm)% Transmittance Figure 3.9. UV-Vis Spectra of PM MA/SWNT Composites from 200-800nm. Heat Polymerized UV Polymerized Gamma polymerized Neat

PAGE 87

64 300400500600700800 0 10 20 30 40 50 60 70 80 90 100 Elasped Time (1) 0 h (2) 1.5 h (3) 4.0 h (4) 22.0 h (5) 25.0 h % TransmittanceWavelength (nm) Figure 3.10. UV-Vis sp ectra of PMMA/SWNT/CH2Cl2 solution.

PAGE 88

65 Scanning Electron Microscopy The characterization studies of the po lymer/SWNT composites were further supported with images captured via scanni ng electron microscopy (SEM). SEM images of the solvent evaporated films (Figure 3.11-3.13) validate the presence of carbon nanotubes in the PMMA matrix regardle ss of the three methods employed for polymerization. It can be safely stated that dissolution of composites in methylene chloride does not remove carbon nanot ubes from the polymer matrix. Figure 3.11. SEM image of UV polymerized PMMA/SWNT.

PAGE 89

66 Figure 3.12. SEM image of gamma polymerized PMMA/SWNT.

PAGE 90

67 Figure 3.13. SEM image of heat polymerized PMMA/SWNT.

PAGE 91

68 Dielectric Analysis Dielectric analysis measures loss factor (’), permittivity (”), and Tan Two clear relaxations in PMMA, the and are evident. The relaxation is defined as main chain segmental motion; whereas, the relaxation is considered the hindered rotation of the ester side group attached to the polymer ma in chain. Figures 3.14-3.19 represent loss factor spectra of PMMA and PMMA/SWNT composites from 6 Hz to 1 x 105 Hz and 150oC to 220oC. The loss factor plots were used to determine the activation energies for the relaxation via Arrhenius plot s (Figures 3.20-3.25) as de scribed in Chapter 2. These values are listed in Table 3.3. The activa tion energies are comparable to the activation energies previously cited in literatu re for PMMA and PM MA/carbon nanotube composites (McCrum, Read and Williams 1967; Ta tro et al. 2004; Musisener et al. 2002). Based on the data, the carbon nanotubes (at low concentrations) did not hinder or ease the rotation of the ester side gr oup about the C-C bond that atta ches it to the polymer main chain. Table 3.3 DEA Data. Activation energies of the transition (1-300 Hz). Sample Neat ( kcal/mol) Composite ( kcal/mol) UV Polymerized 18 18 Polymerized 17 19 Heat Polymerized 16 17

PAGE 92

69 UV Polymerized neat PMMA 6Hz100000Hz 0 1Loss Factor -150-5050150Temperature (C) Universal V3.4C TA Instruments Figure 3.14. DEA loss factor data for UV polymerized neat PMMA.

PAGE 93

70 UV Polymerized PMMA/SWNT Composite 6Hz100000Hz 0.0 0.5 1.0Loss Factor -150-5050150Temperature (C) Universal V3.4C TA Instruments Figure 3.15. DEA loss factor data fo r UV polymerized PMMA/SWNT composite.

PAGE 94

71 Gamma Polymerized neat PMMA 6Hz100000Hz 0.0 0.5 1.0Loss Factor -150-100-50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.16. DEA loss factor data for polymerized neat PMMA.

PAGE 95

72 Gamma Polymerized PMMA/SWNT 6Hz100000Hz 0.0 0.5 1.0Loss Factor -150-100-50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.17. DEA loss factor data for the polymerized PMMA/SWNT composite.

PAGE 96

73 Heat polymerized neat PMMA 6Hz 100000 Hz 0.0 0.5 1.0Loss Factor -150-100-50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.18. DEA loss factor data fo r the heat polymerized neat PMMA.

PAGE 97

74 Heat polymerized PMMA/SWNT Composite 6Hz 100000 Hz 0.0 0.5 1.0Loss Factor -150-100-50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.19. DEA loss factor data for heat polymerized PMMA/SWNT composite.

PAGE 98

75 Arrenhius Plot of transition for UV polymerized neat PMMA from 1 to 300 Hz y = -8888.3x + 32.008 R2 = 0.99650 1 2 3 4 5 6 7 0.0020.00220.00240.00260.00280.003000320.00340.00360.00381/T (K)ln frequency Ea= 18 kcal/mol Figure 3.20. Arrenhius Plot of transition for the UV Polymerized neat PMMA sample. Arrenhius Plot of transition for UV polymerized PMMA/SWNT Composite from 1 to 300 Hz y = -9108.4x + 32.766 R2 = 0.99790 1 2 3 4 5 6 0.0020.002200024000260.002800030.00320.00340.00360.00381/T (K)ln frequency Ea = 18 kcal/mol Figure 3.21. Arrenhius Plot of transition for the UV Polymerized PMMA/SWNT Composite.

PAGE 99

76 Arrenhius Plot of transition for polymerized neat PMMA from 1 to 300 Hz y = -8400.5x + 30.146 R2 = 0.9970 1 2 3 4 5 6 0002000220.00240.0026000280.003000320.0034000360.00381/T (K)ln frequency Ea= 17 kcal/mol Figure 3.22. Arrenhius Plot of transition for the gamma () polymerized neat PMMA sample. Arrenhius Plot of transition for polymerized PMMA/SWNT Composite from 1 to 300 Hz y = -9383.6x + 33.81 R2 = 0.98810 1 2 3 4 5 6 7 0.0020002200024000260.00280.0030.00320.00340.00360.00381/T (K)ln frequency Ea= 19 kcal/mol Figure 3.23. Arrenhius Plot of transition for the gamma() polymerized PMM/SWNT Composite.

PAGE 100

77 y = -8303.6x + 29.995 R2 = 0.99520 1 2 3 4 5 6 7 0.0020.00220.00240.00260.002800030.0032000340.0036000381/T (K)ln frequency Ea= 16 kcal/molArrenhius Plot of transition for oven polymerized neat PMMA from 1 to 300 Hz Figure 3.24. Arrenhius Plot of transition for the heat polymerized neat PMMA sample. Arrenhius Plot of transition for oven polymerized PMMA/SWNT Composite from 1 to 300 Hz y = -8349.8x + 30.651 R2 = 0.99480 1 2 3 4 5 6 0.0020.00220.0024000260.00280.003000320.00340.0036000381/T (K)ln frequency Ea= 17 kcal/mol Figure 3.25. Arrenhius Plot of transition for the heat polymerized PMMA/SWNT.

PAGE 101

78 Figures 3.26-3.31 are loss factor plots at 30 Hz, 60 Hz, and 100 Hz of the neat PMMA and PMMA/SWNT samples prepared vi a UV, heat, or gamma polymerization. The high temperature region corresponds to the relaxation. At low frequencies, the and relaxations in PMMA are separated; however at higher frequencies the relaxation is overcome by the relaxation and less apparent (R ibelles and Calleja 1985). In syndiotactic PMMA, the dominate relaxation is less dependent on temperature than the weaker relaxation and moves at a faster rate th an the primary relaxation, resulting in merging and the presence of the relaxation (Bergman et al. 1998, McCrum, Read and Williams 1967). Much discussion has centered on understanding the process; which is considered to be a result of the cooperative motion of the main chain ( process) and local hindered rotation of the es ter side group about the C-C bond ( process) attached to the main chain ( Bergman et al. 1998; Ga rwe et al. 1996, McCrum 1967). The intensity and presence of the peak for PMMA is dependent on the extent of and merging and conductivity effects (Chapter 6) that may impede the complete occurrence of this relaxation. The relaxation conforms to WLF (Chapter 2), but due to contributions from merging in all samples, WLF parameters c ould not be clearly defined for the three frequencies in which the relaxation appeared. It is also significant to note the diffe rence in peak intensity between the UV and gamma polymerized neat samples and their composite counterparts in the region, as indicated in Figure 3.32 (a) and (b). This confirms that the carbon nanotubes provide additional dipoles to th e polymer matrix.

PAGE 102

79 UV polymerized neat PMMA 30Hz, 60 Hz, 100 Hz 00 02 0.4 06 08Loss Factor -50 0 50 100 150 200Temperature (C) y;p Universal V3.4C TA Instruments Figure 3.26. and relaxation of UV polymerized neat PMMA at 30 Hz, 60 Hz, and 100 Hz. UV polymerized PMMA/SWNT 30Hz, 60 Hz, 100 Hz 00 02 0.4 06 08Loss Factor -50 0 50 100 150 200Temperature (C) Universal V3.4C TA Instruments Figure 3.27. and relaxation of UV polymerized PMMA/SWNT at 30 Hz, 60 Hz, and 100 Hz. region region

PAGE 103

80 Gamma polymerized neat PMMA 30Hz, 60 Hz, 100 Hz 00 02 0.4 06 08Loss Factor -50 0 50 100 150 200Temperature (C) Universal V3.4C TA Instruments Figure 3.28. and relaxation of gamma polymeri zed neat PMMA at 30 Hz, 60 Hz, and 100 Hz. Gamma polymerized PMMA/SWNT 30Hz, 60 Hz, 100 Hz 00 02 0.4 06 08Loss Factor -50 0 50 100 150 200Temperature (C) Universal V3.4C TA Instruments Figure 3.29. and relaxation of gamma polymer ized PMMA/SWNT at 30 Hz, 60 Hz, and 100 Hz. region region

PAGE 104

81 Heat polymerized neat PMMA 30Hz, 60 Hz, 100 Hz 0.0 0.2 0.4 0.6 0.8Loss Factor -50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.30. and relaxation of heat polymerized neat PMMA at 30 Hz, 60 Hz, and 100 Hz. Heat polymerized PMMA/SWNT 30Hz, 60 Hz, 100 Hz 0.0 0.2 0.4 0.6 0 8 Loss Factor -50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.31. and relaxation of heat polymeri zed PMMA/SWNT at 30 Hz, 60 Hz, and 100 Hz. region region

PAGE 105

82 60 Hz 0 1Loss Factor -150 -50 50 150Temperature (C) UV Polymerized Neat PMMA Oven Polymerized Neat PMMA Gamma Polymerized Neat PMMAUniversal V3.4C TA Instruments Figure 3.32.a. DEA loss factor at 60 Hz of UV, gamma, and heat polymerized neat PMMA. 60 Hz 0 1Loss Factor -150 -50 50 150Temperature (C) UV Polymerized PMMA/SWNT Oven Polymerized PMMA/SWNT Gamma Polymerized PMMA/SWNTUniversal V3.4C TA Instruments Figure 3.32.b. DEA loss factor at 60 Hz of UV, gamma, and heat polymerized PMMA/SWNT.

PAGE 106

83 PMMA contains two methyl groups: (1) the methyl attached to the main chain and (2) the methyl attached to the ester side group. The region in PMMA is the rotation of these methyl groups and occurs at low temp eratures. However, these groups are not usually detected under dielectric analysis due to the non-polar nature of the methyl group. Such relaxations are typically characterized under mechanical analysis (McCrum, Read, and Williams 1967). However, a weak relaxation is present in the gamma and heat initiated composites from -130 to -30oC as displayed in Figures 3.33 and 3.34. This peak is difficult to determine in the UV polymerized composites. The relaxation has also been seen in PMMA/MWNT composites befo re irradiation (Tatro et al. 2004). It has been stated in literature that carbon nanotubes can be used to identify or detect polymer transitions (Zhao, Wood, and Wagner 2001). This behavior may be a result of changes in the chemical structure induced by the met hod of initiation, as well as the increase interaction at the polymer-nanotube interface. The permittivity or dielectric constant is defined as the amount of alignment of the dipoles in an electric field (TA Instru ments 1998). The experimental data published for polymer-SWNT composites (Harmon et al 2001; Muisener et al. 2002; Brosseau, Beroual, and Boudida 2000; Kusy, Whitley, and Kalachandra 2001) exhibit an increase in permittivity for the composite samples as compar ed to the neat samples. The permittivity data obtained in this study agreed with resu lts previously published. Table 3.4 shows the permittivity at 100 Hz and 25oC. The UV polymerized samples displayed the largest increase in the dielectric constant, followed by the thermally polymerized samples and the polymerized samples, respectively. The dielectric constant can be correlated to the refractive index in order to better understand the electronic nature of the polymer and the effect of carbon nanotubes in the the polymer matrix. Van Krevelen (Van Krev elan 1990) states that if the sample in question is a non-polar insulator, the dielec tric constant for low frequencies can be expressed by = n2 where n is the refractive index. If the difference between the dielectric constant and square d refractive index is large, the disparity is the result of permanent dipoles and semi-conductive character in the samples (Van Krevelan 1990). Table 3.5 lists values of the dielectric constant ( ) and the squared refractive index (n2).

PAGE 107

84 The difference between and n2 for the PMMA/SWNT composites is larger than that observed in the neat PMMA samples. Th e largest difference is seen in the UV polymerized composite sample. The publishe d refractive index of PMMA is 1.49 at 589 nm (Keyes 1988). The permanent dipoles in the ester side group of PMMA explain the difference between and n2 in neat PMMA samples (McCrum, Read, and Williams 1967). The carbon nanotubes, which are known to have one-third metallic character, and two-thirds semi-conductive character, contri bute to the conductiv e nature of the composites. The refractive indices of the composite samples are slightly lower than their neat counterparts. This proves that while diel ectric properties are notably affected by the presence of carbon nanotubes even at small concentration, optical properties are less sensitive to the presence of nanotubes. Table 3.4. DEA data. Dielectric consta nt values of PMMA and PMMA/SWNT Composites. Sample RI (n) ’@ 25oC/100Hz n2 UV Neat 1.4916 4.9857 2.2248 UV Composite 1.4913 7.2933 2.2239 Gamma Neat 1.4919 3.7113 2.2257 Gamma Composite 1.4913 4.5639 2.2239 Thermal Neat 1.4919 3.7368 2.2257 Thermal Composite 1.4910 4.7517 2.2230 Table 3.5. Refractive Inde x and Dielectric constant values of PMMA and PMMA/SWNT Composites. ’ @ 25oC 100 Hz Sample Neat Composite Gamma 3.71 4.56 0.85 UV 4.99 7.29 2.3 Thermal 3.74 4.75 1.01

PAGE 108

85 0.0 0.2Loss Factor -130 -30Temperature (C) Gamma Polymerized PMMA/SWNT Gamma Polymerized neat PMMAUniversal V3.4C TA Instruments relaxation Figure 3.33. DEA data. Pl ot depicting enhanced relaxation of the Gamma polymerized PMMA/SWNT Composite compared to the neat PMMA.

PAGE 109

86 000 005 0.10Loss Factor -130 -80 -30Temperature (C) Oven Polymerized neat PMMA Oven Polymerized PMMA/SWNTUniversal V3.4C TA Instruments relaxation Figure 3.34. DEA data. Plot depicting enhanced relaxation of the heat polymerized PMMA/SWNT Composite compared to the neat PMMA.

PAGE 110

87 Dynamic Mechanical Analysis Dynamic mechanical analysis (DMA) is used to characterize the viscous (E”) and elastic properties (E’) of a polymer. When characterized via DMA, PMMA clearly exhibits all three relaxations: and Figures 3.35-3.40 are loss modulus (E’) plots of all six composites. Activ ation energies of the samples were obtained by plotting the temperature at the maximum peak height of th e beta transition agains t the natural log of the frequency (Figures 3.41-3.46). All sa mples (neat and composites) had activation energies (Table 3.6) comparable to those cited in literature for PMMA and PMMA/SWNT composites (McCrum, Read and Williams 1967; Harmon et al. 2001, Muisener et al. 2002; Tatro et al. 2002). St orage modulus (E”) va lues were tabulated (Table 3.7) at 10 Hz and -85oC, 25oC and 100oC. These temperatures were chosen because they coincide with the and relaxation regions of the polymer as seen in the mechanical loss plots (E’). In all samples the storage moduli decr eases with increasing temperature. This is consistent with th e viscoelastic behavior of polymers. As temperature is increased, the mobility of th e molecules in the polymer chain increases leading to a softening of the polymer ma in chain and structural failure at high temperatures. Table 3.6. DMA Data: Activation Energies of transition Neat ( kcal/mol) Composite (kcal/mol) UV 18 18 Thermal 18 18 Gamma 18 18

PAGE 111

88 UV Polymerized neat PMMA ( 1-100 Hz) 0 100 200 300 400 500Loss Modulus (MPa) -150-100-50050100150200Temperature (C) Instrument : 2980 DMA V 1 7 B Universal V3.4C TA Instruments Figure 3.35. DMA data. Loss Modulus Plot of UV polymerized neat PMMA.

PAGE 112

89 UV Polymerized PMMA/SWNT ( 1-100 Hz) 0 100 200 300 400 500Loss Modulus (MPa) -150-100-50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.36. DMA data. Loss Modulus Plot of UV polymerized PMMA/SWNT Composite.

PAGE 113

90 Oven Polymerized neat PMMA ( 1-100 Hz) 0 100 200 300 400 500Loss Modulus (MPa) -150-100-50050100150200Temperature (C) sue 980 Universal V3.4C TA Instruments Figure 3.37. DMA data. Loss Modulus Pl ot of heat polymerized PMMA/SWNT Composite.

PAGE 114

91 Oven Polymerized PMMA/SWNT ( 1-100 Hz) 0 100 200 300 400 500Loss Modulus (MPa) -150-100-50050100150200Temperature (C) sue 980 Universal V3.4C TA Instruments Figure 3.38. DMA data. Loss Modulus Plot of heat polymerized PMMA/SWNT Composite.

PAGE 115

92 Gamma Polymerized neat PMMA ( 1-100 Hz) 0 100 200 300 400 500Loss Modulus (MPa) -150 -100 -50 0 50 100 150 200Temperature (C) Universal V3.4C TA Instruments Figure 3.39. DMA data. Loss Modulus Plot of Gamma ( ) Polymerized neat PMMA.

PAGE 116

93 Gamma Polymerized PMMA/SWNT ( 1-100 Hz) 0 100 200 300 400 500Loss Modulus (MPa) -150-100-50050100150200Temperature (C) Universal V3.4C TA Instruments Figure 3.40. DMA data. Loss Modulus Plot of Gamma ( ) Polymerized PMMA/SWNT Composite.

PAGE 117

94 Arrenhius Plot of UV Polymerized neat PMMA y = -9932x + 35.504 R2 = 0.99820 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0.0030.0031000320.00330.003400035 0.00361/T (K)ln frequency Figure 3.41. Arrenhius Plot for the transition of the UV Polymerized neat PMMA. Arrenhius Plot of UV Polymerized PMMA/SWNT y = -9808.8x + 35.104 R2 = 0.99880 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0.0030.00310003200033 0.0034 0.0035 00036 0.00371/T (K)ln frequency Figure 3.42. Arrenhius Plot for the transition of the UV Polymerized PMMA/SWNT.

PAGE 118

95 Arrenhius Plot of neat PMMA polymerized via Heat y = -8901.1x + 32.141 R2 = 0.9980 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0.0030.00310003200033 0.0034 0.0035 00036 0.00371/T (K)ln frequency Figure 3.43. Arrenhius Plot for the transition of the heat polymerized neat PMMA. Arrenhius Plot of neat PMMA/SWNT polymerized via Heat y = -10171x + 36.333 R2 = 0.99890 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0.003050.00310.003150.00320.003250.00330003350003400034500035000355000361/T (K)ln frequency Figure 3.44. Arrenhius Plot for the transition of the heat polymerized PMMA/SWNT.

PAGE 119

96 Arrenhius Plot of neat PMMA polymerized via Gamma Radiationy = -10504x + 37.467 R2 = 0.99830 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0.003050.00310.003150.00320.003250.00330003350003400034500035000355000361/T (K)ln frequency Figure 3.45. Arrenhius Plot for the transition of the gamma polymerized neat PMMA. Arrenhius Plot of neat PMMA/SWNT polymerized via Gamma Radiationy = -9325.8x + 33.488 R2 = 0.99820 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0.0030.00310003200033 0.0034 0.0035 00036 0.00371/T (K)ln frequency Figure 3.46. Arrenhius Plot for the transition of the gamma polymerized PMMA/SWNT.

PAGE 120

97 Table 3.7. DMA data. Storage Modulus (E”) values at 10 Hz and -85oC, 25oC and 100oC. Microhardness The Vickers hardness number was tabulated (Table 2.8) for each sample. There was not any noticeable difference in the hardness between the neat and composite samples. Because hardness is dependent on the viscoelastic behavior of polymers (Calleja and Fakirov 2000), these values further support the mechanical data. Table 3.8. Microhardness Data for UV, heat, gamma polymerized PMMA and PMMA/SWNT Nanocomposites. Sample Storage Modulus 10 Hz -85oC 25oC (MPa) 100oC UV Neat 7355 3957 1440 UV Composite 7414 3943 1340 Gamma Neat 7491 3832 863 Gamma Composite 7242 3945 950 Thermal Neat 6916 3820 1159 Thermal Composite 6028 3187 1014 Sample Neat Composites UV 21.5 0.1 21.0 0.4 Thermal 20.6 0.2 17.6 0.2 Gamma 19.7 0.5 20.2 0.8

PAGE 121

98 Conclusions This study proves that PMMA/SWNT films with increased optical transparency, as compared to a melt blended PMMA/SWNT composite, can be successfully fabricated via free radical polymerization using the same photoinitiator and thre e different initiation sources. The combination of sonication, in situ polymerization, dissolution and solvent evaporation are essential components to the fabr ication of such films. It is also evident that the source of initiation has an effect on certain properties as seen in the high molecular weight of the heat initiated polymer sample as compared to the UV and gamma polymerized samples. Characterization of polymer carbon nanotube composites via dielectric analysis supports previously stated liter ature that carbon nanotubes can be used to detect polymer relaxations. The dielectric properties of an insulating polymer increased when carbon nanotubes are added, even at low concentratio ns. There also exists interaction between the carbon nanotubes and polymer matrix as shown by the enhanced relaxation in the region of the dielectric loss plots. The dielectric consta nt and refractive index were successfully correlated using Maxwell’s Relationship and provided information concerning nanotube effect on pol arization and optical dispersi on of a dielectric material. The mechanical properties we re not enhanced with the addition of carbon nanotubes in this study. Fabricating composites with al igned nanotubes would allow for the nanotubes to absorb the energy from the load more eff ectively; thus, increasing the strength of the composite. The end use implications of the compos ite formed and the method used suggest the ability to design and fabri cate material to be used for electromagnetic shielding and electrostatic charge dissipati on. Although higher loadings would be needed in some instances to ensure efficient performance, this study provides concrete data suggesting that even at a low nanotube concentrati on electrical conductivity is achieved.

PAGE 122

99 CHAPTER 4 Ionizing (Gamma) Radiation Effects on PMMA/Soot composites When a particle contains enough energy to re move an electron from a molecule or atom, it is considered an ionizing species. Further, the intera ction between ionizing species with molecules and atoms leads to the rearrangement or breakage of molecular bonds. Ionizing radiation exists in two form s: (1) direct ionization and (2) indirect ionization. Ionization occurs directly when an ionizing speci es (i.e., electrons, positrons or alpha particles) interact with the electrons of a molecule or atom. Indirect ionization requires an intermediate step that produces x-rays, -rays and high energy electrons that serve as ionizing species (Clegg and Collyer 1991; Tatro et al. 2002). For example, neutrons do not posses enough energy to interact with electrons; how ever, these neutrons can interact with an atom’s nucleus. This interaction leads to ra dioactive decay of the nucleus and the release of the above menti oned high energy particle rays and electrons (Clegg and Collyer 1991). Gamma ( ) radiation was the energy source used in this study. Cobalt 60 (half life 5.3 years) and 137 Cs (half life 30 years) are the most common forms of gamma sources used to study radiation effects on polymers. Cobalt emits 1.25 MeV rays and 0.66 MeV rays are emitted from the cesium source (Reetz, Yagci and Mishra 2000; Clegg and Collyer 1991). For more speci fics regarding the gamma source used in this study refer to Chapter 2. The specific reactions that occur within a polymer as a result of radiation exposure depend greatly on the st ructure of the polymer and the atmosphere in which the exposure takes place. A polymer can experien ce degradation as well as exhibit resistance to radiation. The formation of free radical s, scission of the main chain (resulting in a decrease in molecular weight), cross-linki ng (resulting in an increase in molecular weight), and the formation of peroxides a nd volatile gaseous products (Clegg and Collyer 1991; Reich and Stivala 1971) ar e all possible products of radiation exposure. The

PAGE 123

100 resistance of polymers to radiation depends on the extent of molecular changes due to the irradiation. The polymer used in this study, poly (m ethyl methacrylate) (PMMA), is known to undergo main chain scission when exposed to io nizing radiation. The effect of radiation exposure on PMMA has been extensivel y studied (Kudoh and Sasuga et. al1996; Shrempel and Witthulm 1997; Sayyah and Sabbah et. al 1997; Harmon, Gaynor and Taylor 1993; Bertolucci and Harmon 1998). It has been found that as a result to exposure, a decrease in the polymer’s gla ss transition temperature, as well as the evolution of products such as monomer, carbon dioxide, carbon monoxide, methane, propane or hydrogen occurs (Reich and S tilva 1971; Goyanes and Benites et. al. 1996; Goyanes and Benites et. al. 1997) Being that there exist se veral applications in which polymers are exposed to radia tion, there is much intere st in developing a polymer composite material that exhibits increased radiation resistance. Single walled (SWNT) and multiwalle d (MWNT) carbon nanotubes have been found to increase the radiation resistance of polymeric material. PMMA composites are commonly used to determine the effects of various additives on polymer radiation stability. Previous studies indicate that the addition of ar omatic groups to the polymer, either within the structure or as a part of the composite increases radiation resistance (Clegg and Collyer 1991; Clough and Gillen 1991). The use of carbon nanotubes as additives in polymer matrices, including PMMA has been widely studied (O’Rourke Muisener et. al 2002; Harmon et. al 2001; Haggenmueller and Gommans et. al. 2000; Jia and Wang et al 1999; Tatro et.al 2002). Studi es have shown that the addition of SWNTs or MWNTs carbon nanotubes can increase radiation re sistance of corresponding polymer/nanotube composite materials (O’Rour ke Muisener et. al 2002; Harmon et. al 2001; Tatro et.al 2002). Both types of carbon nanotubes have unpurified components known as soot. Soot is composed of carbon na notubes, metal catalyst, fullerenes and other amorphous carbons. This chapter explores the use of the less pu re, but considerably less expensive soot as a radiation hardening component in polymer/nanotube composites.

PAGE 124

101 Experimental Materials The methyl methacrylate monomer (MMA) a nd the column packing material that was used to remove the monomethyl ethe r hydroquinone (MEHQ) inhibitor were all purchased from Sigma Aldrich (Milwaukee, W I). The initiator, 2,2’azobis ( 2,4dimthylpentane nitrile) (Vazo 52) was purchased from DuPont (Wilmington, DE). The solvents used (all A.C.S. grade): methylene chloride (dicloromethane), N, N, dimethylformamide and methanol were purchased from Fisher Scientific (Pittsburg, PA). Soot was obtained from the Center of Na notechnology at NASA Ames Research Center (Moffett Field, CA). Composite preparation Methyl methacrylate monomer was deinhib ited using a packed column to remove the monomethyl ether hydroquinone (MEHQ) inhibitor. 0.2% of the initiator 2,2’azobis (2,4-dimethylpentane nitrile) (Vazo 52) was added to the monomer. Dry nitrogen was then bubbled through the mixture for 1 minute to remove oxygen. The monomer/initiator mixture was placed in sample vials and heat ed in the oven for 26 hours at a temperature of 60C. After polymerization, polymer samp les were dissolved in dichloromethane to make a 10% (by wt.) solution. PMMA was th en precipitated in meth anol and dried in a vacuum oven at 125C for 4 days. Drie d PMMA was dissolved in N,N dimethyl formamide (DMF). Soot (1% by wt) was soni cated using a Branson Sonifier 450 ( Figure 4.1) in DMF for two hours. The sonicated soot was then added to the sonicated polymer/DMF solution. The PMMA/soot/DMF mixture was sonicated for an additional two hours. After sonication, th e mixture was precipitated out in methanol. The resulting material was placed in a vacuum oven for 5 da ys at 145C. In order to make composites with lower concentrations (0.25%, 0.5%) of soot, the dried 1% PMMA/soot composite was mixed with neat PMMA polymer in a C. W. Brabender Plasticorder with a

PAGE 125

102 bandbury mixer attachment (Figure 4.2) for 5 minutes at 210C. Samples were then molded in a carver press for 5 minutes at a pressure of 5000 pounds and a temperature of 135C. Compression molded samples (excluding the controls) were -irradiated in air at room temperature via a 137Cs source. The dose rate was constant at 985 rads/min for a total dose of 6 Mrad. Differential Scanning Calorimetry The glass transition temperatures (Tg) of the polymer samples were obtained on a TA Instruments 2920 Differential Scanning Calorimeter (DSC). A sample amount between 2–10 mg was obtained from the co mpression molded disc. The samples were heated to 145oC at a rate of 10oC per minute to insure that all samples had the same thermal history. Then the sample was cool ed with liquid nitrogen to room temperature and reheated to 145oC. The Tg values were taken from the second heat as the inflection point of the curve (Hatakeyama and Quinn 1999). Dynamic Mechanical Analysis Mechanical data were collected on a TA Instruments 2980 Dynamic Mechanical Analyzer (DMA). The instrument mode wa s set to measure a tension film using frequencies ranging from 1 to 100 Hz with an amplitude of 5 microns at a temperature range from -150oC to 190oC. The average sample size was 19 x 6 x 2 mm. Microhardness The Vickers hardness number (HV) for each sample was determined with a Leica VMHT MOT with a Vickers indenter. The values were taken from the average of four indents. A horizontal and a vertical read ing was taken on each indent. A load of 500g and a dwell time of 20s was used.

PAGE 126

103 Scanning Electron Microscopy The morphologies of the fractured surfaces of the composites with soot in PMMA matrix were observed using a Hitachi S800 scanning electron microscope. The fracture surfaces were coated with 15 nm thin films of evaporated gold/palladium alloy. The applied voltage depended on magnification. Figure 4.1 Illustra tion of a sonicator.

PAGE 127

104 Figure 4.2. Illustration of bandbury mixer.

PAGE 128

105 Results and Discussion Differential Scanning Calorimetry Glass transition temperatures of PMMA-b ased composites were measured before, immediately after and four months after expos ure to gamma radiation. Four month aging was necessary to compensate for the unstabl e radiolysis products and free radicals reacting within sample over time. The Tg for the neat PMMA was 124oC. As the concentration of the soot incr eased the glass transition temperatures decreased slightly (as compared to the neat) before radiation exposure as shown in Table 4.1. This indicates that impurities in the soot have a plasticizing effect This trend does not agree with studies on the glass transition temperature of SWNTs (M uisener et al. 2002; Harmon et al. 2001) and MWNTs (Tatro et al. 2004) in PMMA. In these studies, the glass transition temperatures increased as the concentra tion of the carbon nanotubes increased. Table 4.1. DSC data. Glass transition temperatures (Tg) of pure PMMA and PMMA/soot composites before irradiation, immediately after i rradiation and four months after irradiation. After irradiation the glass transition temperatures decreased for all samples except the 0.5% PMMA/soot composite. The extent of decrease was greatest for the neat PMMA. The DSC plots for all samples befo re, after, and 4 months after radiation exposure are represented in Figures 4.3-4.14. It is significant to note that the glass transition temperature of the soot composites appeared to recover after 4 mont hs of post irradiation aging at room temperature. This phenomenon was not observed in the neat PMMA samples. While it is not possible at Sample Tg before irradiation (oC) Tg immediately after exposure to 6Mrads (oC) Tg 4 months after exposure to 6Mrads (oC) Neat PMMA 124 114 115 0.25% PMMA/soot 122 116 120 0.5% PMMA/soot 119 119 120 1% PMMA/soot 122 114 118

PAGE 129

106 this time to assign a direct molecular mechanism for this recovery, it is tempting to speculate on possible reasons for the increase in the glass transition temperatures of the composites. Since free radicals persist in ir radiated samples for periods of months (Clegg and Collyer 1991; Clough and Gillen 1991), the soot may undergo reactions with the free radicals forming tighter matrix structures. Additionally, lo w molecular weight radiolysis products responsible for the decrease in Tg may migrate to the surface of the soot material and return the matrix to its original, un-plastici zed state. It is believed that soot particles can agglomerate at higher concentrations re sulting in a smaller soot surface area, less efficient radiation absorption and less effici ent incorporation of low molecular weight radiolysis products. This explains higher ra diation resistance of 0.5% composite as compared to the 1% composite.

PAGE 130

107 123.52C(I) 11800C 126.73C -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.3. DSC data. Glass Transition Temperature of Neat PMMA before radiation exposure. Neat PMMA Non irradiated

PAGE 131

108 114.41C(I) 11162C 12236C -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.4. DSC data. Glass Transition Temperature of Neat PMMA tested immediately after radiation exposure. Neat PMMA -Irradiated

PAGE 132

109 11523C(I) 108.07C 11864C -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.5. DSC data. Glass Transition Temperature of Neat PMMA tested 4 months after radiation exposure. Neat PMMA Irradiated 4mths

PAGE 133

110 122.00C(I) 116.30C 127.01C -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.6. DSC data. Glass Transition Te mperature of 0.25% PMMA/soot before radiation exposure. 0.25%PMMA/soot Non irradiated

PAGE 134

111 116.10C(I) 110.34C 122.46C -0.3 -0.2 -0.1 0.0 0.1Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Comment: irradiated 6.1Mrads Instrument: 2920 MDSC V2.6A Exo Up Universal V36C TA Instruments Figure 4.7. DSC data. Glass Transition Te mperature of 0.25% PMMA/soot tested immediately after radiation exposure. 0.25% PMMA/soot Irradiated

PAGE 135

112 120.01C(I) 113.32C 123.61C -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Coeeu sue90SC6 Exo Up Universal V36C TA Instruments Figure 4.8. DSC data. Glass Transition Te mperature of 0.25% PMMA/soot tested 4 months after radiation exposure. 0.25% PMMA/soot Irradiated 4mths

PAGE 136

113 11892C(I) 113.09C 122.22C -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.9. DSC data. Glass Transition Te mperature of 0.5% PMMA/soot before radiation exposure. 0.5% PMMA/soot Non irradiated

PAGE 137

114 11890C(I) 11505C 122.44C -0.3 -0.2 -0.1 0.0 0.1Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Comment: irradiated 6.1Mrads Instrument: 2920 MDSC V2.6A Exo Up Universal V36C TA Instruments Figure 4.10. DSC data. Glass Transition Te mperature of 0.5% PMMA/soot tested immediately after radiation exposure. 0.5% PMMA/soot Irradiated

PAGE 138

115 118.98C(I) 113.08C 121.99C -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.11. DSC data. Glass Transition Te mperature of 0.5% PMMA/soot tested 4 months after radiation exposure. 0.5% PMMA/soot Irradiated 4mths

PAGE 139

116 122.38C(I) 115.39C 126.53C -0.3 -0.2 -0.1 0.0 0.1Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Exo Up Universal V36C TA Instruments Figure 4.12. DSC data. Transition Temper ature of 1% PMMA/soot tested before radiation exposure. 1% PMMA/soot Non irradiated

PAGE 140

117 114.13C(I) 106.48C 11884C -0.5 -0.4 -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Comment: irradiated 6.1 Mrads Instrument: 2920 MDSC V2.6A Exo Up Universal V36C TA Instruments Figure 4.13. DSC data. Glass Transition Temperature of 1% PMMA/soot tested immediately after radiation exposure. 1% PMMA/soot Irradiated

PAGE 141

118 118.48C(I) 11292C 12263C -0.3 -0.2 -0.1 0.0Heat Flow (W/g) 20 40 60 80 100 120 140 160Temperature (C) Coepessed5,5deg,5000ps,aeos sue90SC6 Exo Up Universal V36C TA Instruments Figure 4.14. DSC data. Glass Transition Te mperature of 1% PMMA/soot tested 4 months after radiation exposure. 1% PMMA/soot Irradiated 4mths

PAGE 142

119 Scanning Electron Microscopy Figure 4.15 is a SEM image of unpurif ied carbon nanotubes and Figures 4.164.18 depict SEM images of the fractured surfaces of neat PMMA, 0.5% and 1% composite samples before radiation exposure. All images were captured at the same resolution. There is a noticeable differen ce in the surface morphol ogy between the neat and composite samples. As the soot concentr ation increases from 0% to 1% the size of the morphological features decreases indica ting possible adhesion failure between the soot particles and the polymer matrix. Figure 4.15. SEM of unpur ified carbon nanotubes (soot).

PAGE 143

120 Figure 4.16. SEM images before radiation exposure of neat PMMA.

PAGE 144

121 Figure 4.17. SEM images before ra diation exposure of 0.5% PMMA/soot.

PAGE 145

122 Figure 4.18. SEM images before ra diation exposure of 1% PMMA/soot.

PAGE 146

123 Microhardness Microhardness measurements were conducted for all samples before and after irradiation, shown in Table 4.2. The Vick ers hardness numbers support the trends that have been previously stated in literature. After exposure to radi ation, the neat PMMA behaves as expected with a decrease in hardness. The composite samples show an increase in hardness after radiation expos ure, with the 0.5% PMMA/soot composite having the greatest potential for resistance to radiation. Table 4.2. Vickers hardness numbers of neat PMMA and PMMA/soot samples before and after the irradiation. Dynamic Mechanical Analysis PMMA exhibits three cl ear transitions when characterized under dynamic mechanical analysis (DMA): The corresponds to main chain molecular motion, corresponds to the rotation of the ester side group, and corresponds to the rotation of the methyl side group (McCrum, Read and Williams 1967). Figures 4.194.22 are loss modulus (E”) plots of PMMA and PMMA/soot composites vs. temperature, noting the three typical transitions. The loss modulus (E”) is an expres sion of the viscous properties associated with the polymer’s ability to di ssipate mechanical energy. The loss modulus values were recorded before, immediately af ter and 4 months after radiation exposure. The activation energies for the transitions, shown in Table 4.3, were determined from the E” spectra. These values were obtain ed by taking the inverse of the temperature atmaximum peak height plotted against the na tural log of the frequency (Figures 4.234.34). Sample Control Immediately after Irradiation Neat 0.25% 0.50% 21.2 0.43 20.3 0.22 18.9 0.19 20.2 0.02 21.1 0.08 24.1 0.19 1% 21.1 0.47 21.8 0.49

PAGE 147

124 A linear relationship showing Arrhen ius behavior was obtained for the ransition. The values of activation energies are simila r for all the samples except for the 0.25% composite tested four mont hs after exposure. The r eason for the 0.25% composite deviation is not apparent at th is time. These activation energy values reported in this paper are also consistent with the previously published data on PMMA transitions (Harmon et al. 2001; McCrum, Read and Williams 1967; Muisener et al. 2002). Sample Before radiation exposure kcal/mol Immediately after exposure to 6 Mrads kcal/mol 4 months after exposure to 6 Mrads kcal/mol Neat PMMA 0.25% PMMA/soot 0.5% PMMA/soot 16 18 15 17 16 16 14 24 15 1% PMMA/soot 17 20 14 Table 4.3. DMA data. Activation energies of transitions for neat PMMA and PMMA/soot composites.

PAGE 148

125 30 Hz 0 100 200 300 400 500 600 700 800Loss Modulus (MPa) -150 -100 -50 0 50 100 150 200Temperature (C) PMMA Irradiated PMMA Non Irradiated PMMA Irradiated 4 mthsUniversal V3.4C TA Instruments Figure 4.19. DMA data. Loss Modulus (E) of Neat PMMA at 30 Hz tested before, immediately after and 4 mont hs after radiation exposure.

PAGE 149

126 30 Hz 0 200 400 600 800Loss Modulus (MPa) -150 -100 -50 0 50 100 150 200Temperature (C) 025%PMMA/soot Irradiated 0.25%PMMA/soot Non Irradiated 0.25%PMMA/soot Irradiated 4mthsUniversal V3.4C TA Instruments Figure 4.20. DMA data. Loss Modulus (E) of 0.25% PMMA/soot composite at 30 Hz tested before, immediately after and 4 months after radiation exposure.

PAGE 150

127 30 Hz 0 200 400 600 800Loss Modulus (MPa) -150 -100 -50 0 50 100 150 200Temperature (C) 05%PMMA/soot Irradiated 0.5%PMMA/soot Non Irradiated 0.5%PMMA/soot Irradiated 4 mthsUniversal V3.4C TA Instruments Figure 4.21. DMA data. Loss Modulus (E) of 0.5% PMMA/soot composite at 30 Hz tested before, immediately after a nd 4 months after radiation exposure.

PAGE 151

128 30 Hz 0 100 200 300 400 500 600 700 800Loss Modulus (MPa) -150 -100 -50 0 50 100 150 200Temperature (C) 1% PMMA/soot Irradiated 1% PMMA/soot Non Irradiated 1% PMMA/soot Irradiated 4 mthsUniversal V3.4C TA Instruments Figure 4.22. DMA data. Loss Modulus (E) of 1% PMMA/soot composite at 30 Hz tested before, immediately after and 4 months after exposure.

PAGE 152

129 Arrenhius Plot of Transition for Nonirradiated PMMAy = -8161.5x + 29.289 R2 = 0.99210 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea= 16 kcal/mol Figure 4.23. Arrenhius Plot of Transition for nonirradiated PMMA Arrenhius Plot of Transition for Irradiated Neat PMMAy = -8364.2x + 30.017 R2 = 0.99560 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea = 17 kcal/mol Figure 4.24. Arrenhius Plot of Transition for irradiated PMMA.

PAGE 153

130 Arrenhius Plot of Transition for Neat PMMA tested 4 months after radiation exposurey = -7243.3x + 26.46 R2 = 0.99460 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Figure 4.25. Arrhenius Plot of Transition for irradiated PMMA tested 4 months after radiation exposure. Arrenhius Plot of Transition for Non irradiated 0.25% PMMA/Soot y = -9248.8x + 32.703 R2 = 0.99470 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea = 18 kcal/mol Figure 4.26. Arrhenius Plot of Transition for non irra diated 0.25%PMMA/soot. Ea = 14 kcal/mol

PAGE 154

131 Arrenhius Plot of Transition for Irradiated 0.25% PMMA/Soot y = -8171.4x + 29.509 R2 = 0.99330 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea = 16 kcal/mol Figure 4.27. Arrhenius Plot of Transition for irradiat ed 0.25% PMMA/soot. Arrenhius Plot of Transition for 0.25% PMMA/Soot tested 4 months after radiation exposure y = -12274x + 45.838 R2 = 0.99220 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Figure 4.28. Arrhenius Plot of Transition for 0.25%PMMA /soot tested 4 months after radiation exposure. Ea = 24 kcal/mol

PAGE 155

132 Arrenhius Plot of Transition for Non irradiated 0.5% PMMA/Soot y = -7550.6x + 28.629 R2 = 0.99490 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Figure 4.29. Arrhenius Plot of Transition for non irra diated 0.5%PMMA/soot. Arrenhius Plot of Transition for Irradiated 0.5% PMMA/Soot y = -8148.1x + 29.436 R2 = 0.99390 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea = 16 kcal/mol Figure 4.30. Arrhenius Plot of Transition for irradiated 0.5% PMMA/soot. Ea = 15 kcal/mol

PAGE 156

133 Arrenhius Plot of Transition for 0.5% PMMA/soot tested 4 months after radiation exposurey = -7328.1x + 27.195 R2 = 0.99430 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea = 15 kcal/mol Figure 4.31. Arrhenius Plot of Transition 0.5% PMMA/soot tested 4 months after radiation exposure. Arrenhius Plot of Transition for Non irradiated 1% PMMA/Soot y = -8382.9x + 30.076 R2 = 0.99570 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Figure 4.32. Arrhenius Plot of Transition for non irra diated 1% PMMA/soot. Ea = 17 kcal/mol

PAGE 157

134 Arrenhius Plot of Transition for Irradiated 1% PMMA/Soot y = -10275x + 36.838 R2 = 0.99750 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Figure 4.33. Arrhenius Plot of Transition for irradi ated 1% PMMA/soot. Arrenhius Plot of Transition for 1% PMMA/Soot tested 4 months after radiation exposure y = -7082.9x + 25.661 R2 = 0.99110 0.5 1 15 2 25 3 35 4 45 5 0.003000310.003200033000340.0035000360.00371/T (K)ln frequency Ea = 14 kcal/mol Figure 4.34. Arrhenius Plot of Transition for 1% PMMA /soot tested 4 months after radiation exposure.Ea = 20 kcal/mol

PAGE 158

135 Conclusions The PMMA/Soot samples were characteri zed before, immediately after, and 4 months after exposure to radiation. The 0.5% composite sample did not exhibit a decrease in Tg immediately after exposure. The neat PMMA had the greatest decrease immediately after radiation exposure and did not recover after 4 months; however, the 0.25% and 1% composites did recover afte r 4 months. The data obtained via DSC analysis shows increased radiation resistan ce of 0.25% and 0.5% PMMA/soot composites as compared to pure PMMA. The 0.5% soot composite exhibited the greatest extent of radiation hardness. 1% PMMA/soot composite did not exhibit increased radiation hardness. The behavior of the 1% composite may be a result of st rong agglomeration of soot particles. A comparison of radia tion resistance study of PMMA/soot composites with similar studies conducted on PMMA/SWNT and PMMA/MWNT composites (Muisener et al. 2002; Harmon et al. 2001; Tatr o et al. 2004) shows th at the single-walled and multi-walled carbon nanotubes are more suitab le fillers than the unpurified soot for mechanical radiation resistance.

PAGE 159

136 CHAPTER 5 Characterization of PMMA/soot comp osites via Dielectric Analysis This chapter explores the effect of s oot (unpurified carbon nanot ubes) (Chapter 4) on the sub-glass relaxations of poly (methyl methacrylate) when an electric field is applied. The PMMA/soot composites discussed in this chapter are the same nonirradiated samples discussed in Chapter 4. Chapter 7 discusses dielectric behavior present at and above the glass transition re gion of these composites. PMMA has four possible relaxations: and The occurrence of the relaxation is dependent on the method of testing. Because the methyl groups do not contain dipoles, they are not detected under dielectric analysis; however, under dynami c mechanical analysis the relaxation can be clearly seen (McCrum, Read, and Williams 1967). Dielectrically, the presence of the transition is dependent on th e extent of merging of the and transitions and/or the effects of conductivity at and above the glass transition region. Mechanically, the and transitions are always present. Dielectric analysis, as discussed in Chapter 2, provides information on a polymer’s electrical properties by measuring th e extent of molecular motion as related to dipole alignment. Further, information such as activation energies and dielectric relaxation strengths ( ) can be determined via DEA. Dielectric relaxation strengths can be determined via the Cole-Cole, Davids on-Cole, and the Havriliak Negami methods which are based on the Debye equations for a single relaxation (eqs. 5.1-5.3) (Gedde 1999; Havriliak and Havriliak 1997; McCrum Read and Williams 1967). Equations (5.2) and (5.3) represent the re al and imaginary parts of the complex dielectric constant (Chapter 2).

PAGE 160

137 E UR Ui 1 )( (5.1) 2 21 )(E UR U (5.2) 2 21 )(E E UR (5.3) The dielectric relaxation time is described as and is indicative of the time needed for molecules to reorient themselves is the angular frequency, and U is defined as the unrelaxed state and corresponds to high fr equencies where dipole relaxation does not occur; and R is the relaxed state and corresponds to low frequencies where dipoles align with the electric field (McCrum, Read and Williams 1967, Gedde 1999, Havriliak and Havriliak 1997). The Cole-Cole equation for plotting a si ngle relaxation (5.4 ) displays semicircular behavior for small rigid molecules and dilute solutions of polar liquids as indicated in Figure 5.1; howev er, deviation from semi-circu lar (Figure 5.2) behavior occurs when there is a distribution of re laxation times which is typical for polymer systems (McCrum, Read, and Williams 1967; Emran 2000; Cole and Cole 1941). Based on the equation (5.4), an Argand diagram plotting against is created. The radius of the semi-circle is defined as 2 /)(UR (McCrum, Williams and Read 1967). The values of R and U represent the two points that intersect the x-axis. 2 2'' 22 )( 2 )(UR UR (5.4) The dielectric relaxation strength which co rresponds to the amount of dipoles per unit volume and the extent of their alignment due to molecular motion (Emran 2000; Runt and

PAGE 161

138 Fitzgerald 1997), is determin ed by taking the difference of the relaxed and unrelaxed states (Equation 2). U R (5.5) Cole and Cole modified the single relaxation equations (5.2 and 5.3) to compensate for deviations in the curve (Cole and Cole 1941; McCrum, Read, and Willams 1967). Cole and Cole replaced (1+ i ) with 1+ (i o). is the curve broadening term where 0 < 1. A single relaxation behavior occurs when = 1, as approaches 0 the loss maximum becomes broader and the dielectric constant flattens; thus indicat ing a deviation from Debye behavior. ) ( 1 ) ( ) ( *o U R Ui (5.6) The Cole-Cole method applies to relaxa tions that are symmetrical; DavidsonCole method (Equation 5.7) introduced an asymmetric (skewed) parameter and the HavrilaikNegami method (Equation 5.8) co mbines equations 5.6 and 5.7 (McCrum, Read and Williams 1967, Gedde 1999, Havriliak and Havriliak 1997). ) 1 ( ) ( ) ( *o U R Ui (5.7) ) )) ( 1 ( ) ( ) ( *o U R Ui (5.8) The aim of this chapter is to provide dielectric characte rization methods for polymersoot systems. Activation energies were calculated for the process in neat

PAGE 162

139 PMMA, 0.25% PMMA/soot and 1% PMMA/soot com posites. Dielectric strengths were obtained for the and processes by fitting the data to Havriliak-Negami parameters.

PAGE 163

140 Figure 5.1. Argand ( Cole-Cole) plot of loss factor plotted against the dielectric permittivity. Figure 5.2. Complex Cole-Cole plot of loss factor plotted against the dielectric permittivity.

PAGE 164

141 Experimental The non irradiated soot composites were prepared using the method discussed in Chapter 4. Dielectric analysis was c onducted on a TA Instruments 2970 DEA using parallel plate sensors. Compression molded samples had an inner diameter of 27 mm and thickness that ranged from 1.2 to 1.7 mm. Samples were scanned from 200 to -150oC at -5oC increments under a nitrogen purge. Scanning frequencies ranged from 1 Hz to 1.0 x 105 Hz. A maximum force of 250 N wa s applied to all samples for the entire scan.. Results and Discussion Figures 5.3-5.5 are loss factor ( ’) plots for neat PMMA, 0.25 % PMMA/soot and 1% PMMA/soot composites. Figure 5.6 is a plot comparing the loss plots for all three samples. The 1% PMMA/soot composite has a much higher loss factor value as compared to the neat and 0.25% composite. Studies conducted by Harmon’s research group on polymer nanotube composites have revealed a relaxation in the region of PMMA (Tatro et al. 2004; Clayton et al. 2005). It has been reported that carbon nanotubes can detect or identify polymer relaxations (Zhao, Wood, and Wagner 2001). A pronounced low temperature (-150oC to -50oC) relaxation is present in the 1% PMMA/s oot composite (Figure 5.5). It can be safely stated, based on reproducible data, th at carbon nanotube based material enhances the relaxation region of PMMA under dielectric analysis. Activation energies were obtained for the process from Arrhenius plots (Figures 5.7-5.9) and are listed in Table 1. Activa tion energies for the composites are similar to that of the neat polymer, indicating that the presence of the nanotubes did not affect side chain rotation. Figure 5.10 is a plot of ’ vs. temperature, the permittivity increased with an increase in soot concentration. This tr end is comparable to studies conducted on

PAGE 165

142 PMMA/MWNT and PMMA/SWNT composites (Tatro et al. 2004; Muisener et al. 2002) and confirm that the soot part icles enhanced the dielectric properties of the polymer matrix. Table 5.1. DEA data. Activation Energies of transition for neat PMMA and PMMA/soot composites. Sample Ea (kcal/mol) Neat 19 0.25% PMMA/soot 17 1% PMMA/soot 18

PAGE 166

143 neat PMMA 1Hz100000Hz 0.0 0.5 1.0 1.5Loss Factor -150 -50 50 150 Temperature (C) Universal V3.4C TA Instruments Figure 5.3. DEA loss factor ( ) plot for neat PMMA

PAGE 167

144 0.25% PMMA/soot 1Hz100000Hz 0.0 0.5 1.0 1.5Loss Factor -150 -50 50 150 Temperature (C) g Universal V3.4C TA Instruments Figure 5.4. DEA loss factor ( ) plot for 0.25% PMMA/soot.

PAGE 168

145 1% PMMA/soot 1Hz100000Hz 0.0 0.5 1.0 1.5Loss Factor -150 -50 50 150 Temperature (C) Universal V3.4C TA Instruments Figure 5.5. DEA loss factor ( ) plot for 1% PMMA/soot. enhanced region

PAGE 169

146 60Hz 0.0 1.5Loss Factor -150 -50 50 150 Temperature (C) neat PMMA 0.25% PMMA/soot 1% PMMA/soot Universal V3.4C TA Instruments Figure 5.6. DEA loss factor plot ( ) at 60 Hz of neat PMMA, 0.25% PMMA/soot, 1% PMMA/soot.

PAGE 170

147 Arrhenius Plot for the relaxation in neat PMMAy = -9498.9x + 33.59 R2 = 0.99710 1 2 3 4 5 0.0030.00310.00320.00330.00340.00350.0036 1/T(K)ln frequency Ea= 19 kcal/mol Figure 5.7. Arrhenius Plot for neat PMMA. Arrhenius Plot for the relaxation in 0.25% PMMA/sooty = -8800.4x + 31.347 R2 = 0.9992 0 05 1 15 2 25 3 35 4 45 5 000302000312000322000332000342000352 1/T (K )ln frequency Ea= 17 kcal/mol Figure 5.8. Arrhenius Plot for 0.25% PMMA/soot. Arrhenius Plot for the relaxation in 1% PMMA/sooty = -9281.5x + 32.733 R2 = 0.9957 0 1 2 3 4 5 0003000310003200033000340003500036 1/T(K)ln frequency Ea= 18 kcal/mol Figure 5.9. Arrhenius Plot for 1% PMMA/soot.

PAGE 171

148 2 4 6 8 10 12 14 16Permittivity -100 0 100 200 Temperature (C) neat PMMA 0.25 % PMMA/soot 1% PMMA/soot Universal V3.4C TA Instruments Figure 5.10. DEA permittivity ( ) at 60 Hz of neat PMMA and PMMA/soot composites.

PAGE 172

149 Figures 5.11 to 5.16 are Cole-Cole plots of the neat and composite samples at 40oC, representative of the transition region, and 100oC, corresponding to the merging region. The process for PMMA is considered a noncooperative simple relaxation associated with the rotation of the ester side group and the process is considered a complex cooperative relaxation between the and processes (Starkweather 1981, McCrum, Read and Williams 1967). At both temperatures reported, the dielectric relaxation strength increased as the soot concentration incr eased. This behavior was expected due to the incorporation of more dipoles in the system from the soot. The symmetric broadening term ( was slightly lower for the 1 % composites for both the and process; but all samples deviated from symmetrical semi-c ircular behavior ( =1). The term was similar for all samples; the value ( = ~0.50) indicating that the curve broadens at high frequencies (McCrum, Read and Williams 1967). Sample U R 40oC () Neat 3.82 6.77 0.493 0.399 2.95 0.25% PMMA/soot 3.92 7.04 0.497 0.401 3.12 1% PMMA/soot 4.67 11.05 0.490 0.353 6.38 100oC () Neat 3.65 7.06 0.500 0.500 3.41 0.25% PMMA/soot 3.88 7.41 0.507 0.569 3.53 1% PMMA/soot 4.33 10.87 0.515 0.401 6.54 Table 5.2. DEA data. Havriliak Negami parameters and relaxation strengths for neat PMMA, 0.25% PMMA/soot, and 1% PMMA/soot composites.

PAGE 173

150 4.04.55.05.56.06.57.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 RU"' Figure 5.11. Cole-Cole plot of the relaxation for neat PMMA at 40oC. 4.045505.560657.0 0.1 02 03 0.4 05 06 0.7 08 09 10 "' Figure 5.12. Cole-Cole plot of the relaxation for 0.25% PMMA/soot at 40oC. 56789101112 03 0.4 05 06 0.7 08 09 10 Figure 5.13. Cole-Cole plot of the relaxation for 1 % PMMA/soot at 40oC.

PAGE 174

151 4045505560657075 01 02 03 04 05 06 07 08 09 10 Ru"' Figure 5.14. Cole-Cole plot of the relaxation for neat PMMA at 100oC. 4045505560657075 01 02 03 04 05 06 07 08 09 10 Ru"' Figure 5.15. Cole-Cole plot of the relaxation for 0.25% PMMA/soot at 100oC. 56789101112 04 05 06 07 08 09 10 Ru"' Figure 5.16. Cole-Cole plot for the relaxation of 1 % PMMA/soot at 100oC.

PAGE 175

152 Conclusion The dielectric behavior of polymer/soo t composites below the alpha process is comparable to that of PMMA/SWNT and PMMA/MWNT composites. The relaxation region in PMMA was enhanced in the 1% PMMA/soot composite. This occurrence is consistent with previously published data. The distribution of relaxation times obtained from the dielectric strength increased with th e increase in soot concentration for both the and processes. This increase was due to an increase in dipoles introduced into the system. The Havriliak-Negami and parameters were similar for all samples. Thus, all samples deviated from semi-circular beha vior and exhibited skewed (asymmetric) behavior in the and processes. Dielectric characterization techniques were found to apply to polymer/soot composites, thus providing new avenues in wh ich to characterize and further understand the effect of nanoparticle s in the polymer matrix.

PAGE 176

153 CHAPTER 6 Examination of DC Conductivity and Interfac ial Polarization of polymer nanotube composites Dielectric spectroscopy measures the electr ical properties of polymeric material, specifically the materials ability to store el ectric charge and its ability to transfer electrical charge (Chapter 2) (TA Instruments 1999; Havril iak et al. 1997). Changes in electric properties correlate to the molecula r mobility of the polymer; increased mobility results in the alignment of dipoles when an electric field is app lied (TA Instruments 1998; Emran 2000). Alignment or polarization in an el ectric field occurs wh en a dielectric (i.e., polymeric material) is placed between two para llel electrodes, resulting in a separation of charges within the polymers. Polarization can occur by four routes depending on time, frequency and strength: elec tronic, atomic, dipolar orie ntation, and interfacial polarization (Simon 1994). Electr onic polarization occurs around 1015 Hz and is the result electron distortion in an electric fi eld. Atomic polarization occurs around 1013 Hz an is defined as the distortion of atomic bonds. Dipolar polarization (10-1 to 109 Hz) occurs as a result of molecular motion of pe rmanent dipoles in a polymer ( Simon 1994; TA Instruments 1999). The last type of polar ization, interfacial, occurs around 100 Hz and is also known as the Maxwell-Wagner-S illars (MWS) effect. It arises in heterogeneous material that differ in perm ittivity and conductivity (Simon 1994, Maxwell 1892; Wagner 1914; Sillars 1937; Steeman et al. 1991;MacKinnon et al. 1992; Korkakas 1993). Polarization is induced as a result of space charge buil d up at the interfaces of the materials involved. At temperatures above the glass transiti on region, the polymer exhibits viscous flow allowing for the movement of ions and other impurities and an increase in DC resistance. The increase in DC conduction influences the loss maximum and can mask high temperature relaxations such as the pr imary (glass transition) relaxation. This

PAGE 177

154 influence is expressed in the fo llowing equation 6.1 (Simon 1994). f x RCo o 1010 8 1 1 (6.1) where R is the DC resistance (ohms), is the angular frequency, is the specific conductivity, and 1.8 x 1010 is 1 over the absolute permittivity of free space multiplied by 2 In order to expose these relaxations, conductivity effects must be removed by treating the data with electric modulus (Starkweather and Avakian 1992; Pissis and Kyritsis 1997). The electric modulus, M*, is the inverse of the complex permittivity, *, where = ’-i ” and M i M M *which can be separated into the real and imaginary parts which are expressed in the following equations: 1) ( i M i M (6.2) 2 2) ( ) ( M (6.3) 2 2) ( ) ( M (6.4) A study conducted by Tsangaris et al. (1998) us ed the electric modulus along with ColeCole plots to understand the effect of pol ymers with metallic fillers on interfacial polarization. In this chapter, the influence of car bon nanotubes on (1) the high temperature conductivity relaxation and (2) the interfacial pol arization or Maxwell-Wagner-Sillars effect and on PMMA will be discussed. The DC conductivity of the neat and composite samples was determined as a function of temperature, and activation energies were determined via Arrhenius plots. Interfacial polarization was determined by treating the

PAGE 178

155 data with electric modulus and plotting the data using the cole-cole method. HavriliakNegami and parameters were used to understand the extent of deviation from Debye behavior at the high temperatur e and to understand the behavior of CNTs and soot in the conductivity relaxation region. The aim of this chapter is to provide better understanding on the role of electrical conductive fillers on polymeric systems. Composite samples were prepared by the methods presented in Chapter 3 for the P MMA/SWNT composites and Chapter 4 for the PMMA/soot.

PAGE 179

156 Results and Discussion PMMA/SWNT composites Figures 6.1-6.6 represent the AC conduc tivity plotted against frequency for the neat PMMA and PMMA/SWNT composites polymerized via thermal energy, UV light, and radiation (Chapter 3). The values for DC conductivity were obtained by extrapolation to zero frequenc y and used to obtain activatio n energies via Arrhenius plots (Starkweather and Avakian 1992; Pissis and Kyritsis 1997; Po lizos and Kyritsis et al. 2000). The dc values for the neat and composite samp les are tabulated in Table 6.1. As expected, the conductivity in creases with incr easing temperature for all samples, correlating with an increase in viscous flow observed at te mperatures above the glass transition region, thus inducing the movement of ions and other conductive impurities (Simon 1994). Further, there is an increase in conductivity values in the composite samples as compared to the neat sample s. This observation proves that the carbon nanotubes do in fact contribu te to the conductive nature of the polymer at these high temperatures. Temperatures from 125oC to 175oC in 10oC increments were used to obtain the Arrhenius plots (Figures 6.7 and 6.12) and did conform to Arrhen ius behavior; however, there was a change in this behavior at 185oC (Figures 6.13 and 6.18). At this temperature a change in the conductivity mechanism occurs. This trend is apparent in all samples. Table 6.2 lists the activation energies fo r the neat and composite samples. The differences in the activation energies between the neat and composite samples are small; however, the composite samples all exhi bit a decrease in activation energies. Starkweather and Avakian (1992) concluded that the mobility of ions depend on the ability of polymer segments to move out of the way. Thus, a lower activation energy indicates that the polymer chains do not hinde r ion mobility, nor does the presence of the carbon nanotubes. It is interesting to note the relationship between activation energies and molecular motion within the polymer (Starkweather 1981). The differences in activation energies in

PAGE 180

157 UV NeatUV CompThermal NeatThermal Comp Neat CompoC dc dc oC dc dc oC dc dc1251.17E-101.55E-091252.24E-105.42E-101258.14E-116.55E-09 1352.52E-104.05E-091355.81E-101.25E-091351.56E-101.49E-08 1455.97E-109.47E-091451.33E-092.85E-091453.34E-103.07E-08 1551.51E-091.96E-081552.96E-095.69E-091557.54E-105.77E-08 1653.02E-093.69E-081656.03E-099.93E-091651.66E-099.79E-08 1755.00E-096.25E-081751.09E-081.43E-081753.21E-091.50E-07 1856.79E-098.71E-081851.73E-081.18E-081855.34E-092.00E-07 1955.26E-098.28E-081952.06E-084.72E-091956.52E-092.19E-07 Table 6.1. DC Conductivity values for PMMA and PMMA/SWNT composites.

PAGE 181

158 PMMA between that of the process (18-19 kcal/mol) (C hapter 3) indicative of side group rotation below Tg, those reported here which ar e indicative of a conductivity relaxation above Tg, and the energies reported fo r the glass transition region (110kcal/mol) (McCrum, Read and Williams 1967) are due to simple or complex relaxations. Activation energies for the glass transition process are usually high (Starkweather and Avakian 1992) due to coope rative motion between the main chain and neighboring groups, as there is an increase in temperature, th ere is an increase in flow and a deviation from viscoelastic behavi or. Activation entropies approaching zero signifies noncooperative motion. It has been reported that the act ivation entropy for the process of PMMA is equal to zero. At temper atures above the glass transition region, the activation entropy is much lo wer than that of the Tg region, also signifying a simple noncooperative relaxatio n mechanism (Starkweather 1991; Simon 1994). Table 6.2. DEA Data. Activation Ener gies for neat PMMA and PMMA/SWNT samples polymerized via UV light. Ea ( kcal/mol) UV Heat Neat PMMA 11.96 11.97 11.55 PMMA/SWNT 11.41 10.29 9.69

PAGE 182

159 0.1110100100010000100000 1E-10 1E-9 1E-8 1E-7 1E-6 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oCIonic Conductivity ac (S/m)f (Hz) Figure 6.1. AC conductivity plotte d against frequency for neat PMMA polymerized via UV light.

PAGE 183

160 0.1110100100010000100000 1E-10 1E-9 1E-8 1E-7 1E-6 195 185 175 165 155 145 135 125Ionic Conductivity ac(S/m)f (Hz) Figure 6.2. AC conductivity plotte d against frequency for PMMA/SWNT polymerized via UV light.

PAGE 184

161 0.1110100100010000100000 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6 Ionic Conductivity ac (S/m)f (Hz) 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oC Figure 6.3. AC conductivity plotte d against frequency for neat PMMA polymerized via radiation.

PAGE 185

162 0.1110100100010000100000 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oC Ionic Conductivity ac (S/m)f (Hz) Figure 6.4. AC conductivity plotte d against frequency for PMMA/SWNT polymerized via radiation.

PAGE 186

163 0.1110100100010000100000 1E-10 1E-9 1E-8 1E-7 1E 6 Ionic Conductivity ac (S/m)f(Hz) 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oC Figure 6.5. AC conductivity plotte d against frequency for neat PMMA polymerized via thermal energy.

PAGE 187

164 0.1110100100010000100000 1E-10 1E-9 1E-8 1E-7 1E-6 pygy Ionic Conductivity ac (S/m)f (Hz) 195oC 155oC 185oC 145oC 175oC 135oC 165oC 125oC Figure 6.6. AC conductivity plotte d against frequency for PMMA/SWNT polymerized via thermal energy.

PAGE 188

165 220225230235240245250255 -100 -98 -96 -94 -92 -90 -88 -86 -84 -82 -80 Ea=11.96 kcal/mol y=-6.0204x+5.18473 R2= 0.99795Log dc (S/m)1000/T (K-1) Figure 6.7. Arrhenius Plot of Neat PMMA polymerized via UV light from 125oC-175oC. 220225230235240245250255 -90 -88 -86 -84 -82 -80 -78 -76 -74 -72 -70 Ea=11.41 kcal/mol y=-5.7411x+5.66097 R2= 0.99739Log dc(S/m)1000/T ( K-1) Figure 6.8. Arrhenius Plot of Neat PMMA polymerized via UV light from 125oC-175oC.

PAGE 189

166 2.202.252.302.352.402.452.50255 -9.8 -9.6 -9.4 -9.2 -9.0 -8.8 -8.6 -8.4 -8.2 -8.0 -7.8 py gy Ea= 11.55 kcal/mol y=-6.0473x+5.56945 R2= 0.99913Log dc (S/m)1000/T Figure 6.9. Arrhenius Pl ot of PMMA/SWNT polymeriz ed via thermal energy from 125oC-175oC. 2.202.252302.352.402.452.502.55 -9.4 -9.2 -9.0 -8.8 -8.6 -8.4 -8.2 -8.0 -7.8 Ea= 9.69 kcal/mol y= -5.17748x + 3.78792 R2= 0.9948Log dc ( S/m)1000/T (K) Figure 6.10. Arrhenius Plot of PMMA /SWNT polymerized via thermal energy from 125oC-175oC.

PAGE 190

167 2.202.252.302.352.402.452.502.55 -10.2 -10.0 -9.8 -9.6 -9.4 -9.2 -9.0 -8.8 -8.6 -8.4 Ea = 11.97 kcal/mol y= -5.81339x + 4.46611 R2= 0.99859Log dc (S/m)1000/T (K) Figure 6.11. Arrhenius Plot of neat P MMA polymerized via g radiation from 125oC-175oC. 2.202.252.302.352.402.452.502.55 -8.2 -8.0 -7.8 -7.6 -7.4 -7.2 -7.0 -6.8 Ea = 10.29 kcal/mol y = -4.87617x + 4.10989 R2= 0.9969Log dc (S/m)1000/T (K) Figure 6.12. Arrhenius Plot of neat PMMA polymerized via g radiation from 125oC-175oC.

PAGE 191

168 210215220225230235240245250255 -100 -98 -96 -94 -92 -90 -88 -86 -84 -82 -80 185oC 175oC Log dc(S/m)1000/T (K-1)2.102.152.202.252.302.352.402.452.502.55 -9.0 -8.8 -8.6 -8.4 -8.2 -8.0 -7.8 -7.6 -7.4 -7.2 -7.0 185oC 175oC py g Log dc(S/m)1000/T ( K-1) Figure 6.13. Log dc vs. temperature of Neat PMMA polymerized via UV ight from 125oC-195oC. Figure 6.14. Log dc vs. temperature of PMMA /SWNT polymerized via UV light from 125oC195oC.

PAGE 192

169 210215220225230235240245250255 -102 -100 -98 -96 -94 -92 -90 -88 -86 -84 -82 -80 185oC 175oC Log dc (S/m)1000/T (K-1) Figure 6.15. Log dc vs. temperature of neat PMMA polymerized via radiation from 125oC-195oC. 210215220225230235240245250255 -84 -82 -80 -78 -76 -74 -72 -70 -68 -66 185oC 175oC Log dc (S/m)1000/T (K-1) Figure 6.16. Log dc vs. temperature of P MMA/SWNT polymerized via radiation from 125oC-195oC.

PAGE 193

170 210215220225230235240245250255 -100 -98 -96 -94 -92 -90 -88 -86 -84 -82 -80 -78 -76 185oC 175oC Log dc (S/m)1000/T (K-1) Figure 6.17. Log dc vs. temperature of neat PMMA polymerized via thermal energy from 125oC-195oC. 210215220225230235240245250255 -94 -92 -90 -88 -86 -84 -82 -80 -78 185oC 175oC Log dc (S/m)1000/T (K-1) Figure 6.18. Log dc vs. temperature of PMMA/SWNT polymerized via thermal energy from 125oC-195oC.

PAGE 194

171 PMMA/soot composites Figures 6.19-6.21 are AC conductivity plots for the neat PMMA, 0.25% soot composite, and the 1% soot composite. Figures 6.226. 24 are the arrhenius plots used to obtain the activation energies for the soot composites. Further, the soot composites exhibited a similar change in conduction mechanism at 175oC as seen in Figure 6.25 a-c. Table 6.3 lists conductivity activation energies for n eat PMMA and PMMA/soot composites. The Ea values are higher for the composites as compar ed to the neat sample. As mentioned in the previous section, the mobility of ions depends on the ability of polymer segments to move out of the way (Starw eather and Avakian 1992). Highe r activation energies are indicative of increased cooperation of the pol ymer chains, thus hindering the mobility of the ions. However, in the case of the PMMA/soot composites, it is possible that the components that constitute soot and their si ze may be responsible for the obstruction or, these components could also associate with th e polymer and form bridges, thus hindering movement. DC conductivity values are pl otted in Table 6.4. As seen in the PMMA/SWNT composite, the soot composite s exhibited the same increase in conductivity at the temperatures reported. Table 6.3. Activation energies for PMMA/soot composites above the glass transition region. Sample Ea ( kcal/mol) Neat PMMA 12.98 0.25% PMMA/soot 15.15 1% PMMA/soot 15.45

PAGE 195

172 Table 6.4. DC Conductivity values for PMMA and PMMA/SWNT composites. Neat PMMA 0.25% PMMA/soot 1% PMMA/soot oC dcdcdc 125 5.10E-11 4.85E-11 2.91E-10 135 1.42E-10 1.72E-10 9.94E-10 145 3.65E-10 4.76E-10 2.92E-09 155 8.52E-10 1.19E-09 8.25E-09 165 1.82E-09 3.10E-09 2.02E-08 175 3.35E-09 6.92E-09 4.29E-08 185 4.53E-09 1.27E-08 7.36E-08 195 3.92E-09 1.51E-08 8.02E-08

PAGE 196

173 0.1110100100010000100000 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oC Ionic Conductivity ac (S/m)f (Hz) Figure 6.19. Ionic Conductivity vs. Log f of neat PMMA.

PAGE 197

174 0.1110100100010000100000 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oC Ionic Conductivity ac (S/m)f (Hz) Figure 6.20. Ionic Conductivity vs. Log f of 0.25% PMMA/SWNT.

PAGE 198

175 0.1110100100010000100000 1E-11 1E-10 1E-9 1E-8 1E-7 1E-6 pyyqy 195oC 185oC 175oC 165oC 155oC 145oC 135oC 125oC Ionic Conductivity ac (S/m)f (Hz) Figure 6.21. Ionic Conductivity vs. Log f of 1% PMMA/SWNT.

PAGE 199

176 2.202.252.302.352.402.452.502.55 -10.5 -10.0 -9.5 -9.0 -8.5 Ea= 12.98 kcal/mol y= -6.53004 + 6.14674 R2= 0.99858Log dc(S/m)1000/T (K-1) Figure 6.22. Arrhenius Plot of neat PMMA from 125oC – 175oC.

PAGE 200

177 2.202.252.302.352.402.452.502.55 -10.5 -10.0 -9.5 -9.0 -8.5 -8.0 Ea= 15.15 kcal/mol y= -7.62684x + 8.888769 R2= 0.99919Log dc(S/m)1000/T (K-1) Figure 6.23. Arrhenius Plot of 0.25% PMMA/soot from 125oC – 175oC.

PAGE 201

178 2.202.252.302.352.402.452.502.55 -9.5 -9.0 -8.5 -8.0 -7.5 -7.0 y = -7.77489x + 10.03396 R2= 0.9989 Ea= 15.45 kcal/mol Log dc(S/m)1000/T (K-1) Figure 6.24. Arrhenius Plot of 1% PMMA/soot from 125oC to 175oC.

PAGE 202

179 2.102.152.202.252.302.352.402.452.502.55 -10.5 -10.0 -9.5 -9.0 -8.5 -8.0 185oC 175oC Log dc (S/m)1000/T (K-1)2.102.152.202.252.30 2.352.402.452.502.55 -10.5 -10.0 -9.5 -9.0 -8.5 -8.0 -7.5 185oC 175oC Log dc (S/m)1000/T (K-1) (a) (b) 2.102.152.202.252.302.352.402.452.502.55 -9.5 -9.0 -8.5 -8.0 -7.5 -7.0 185oC 175oC Log dc(S/m)1000/T (K-1) (c) Figure 6.25. DC Conductivity vs. te mperature of (a) neat PMMA, (b) 0.25% PMMA/soot, (c) 1% PMMA/soot.

PAGE 203

180 Interfacial Polarization Figures 6.26 and 6.27 are plots of neat PMMA and PMMA/SWNT samples polymerized via UV light of the loss factor (”) plotted against log frequency. Figures 6.28 and 6.29 are the above mentioned plots tr eated with electric modulus. The M” vs. log frequency plots reveal peaks that are not apparent in the ” vs. log frequency plots. This behavior is consistent for all PMMA/ carbon nanotube composites presented in this dissertation. This occurre nce is indicative of a Maxwell-Wagner-Sillars relaxation process. Tsangaris et al. (1998) used Cole -Cole plots to explain the MWS effect for heterogeneous polymeric systems. When treat ed with electric modulus and plotted via the ColeCole method (Chapter 5), the ex tent of the MWS effect can be better understood. Cole-Cole plots (Figures 6.30 and6.31) were constructed for neat and composite samples polymerized via UV light (Chapter 3) and for neat PMMA and PMMA/soot composites (Chapters 4 and 5). Da ta presented in this study was consisted with the findings of Tsangaris et al. (1998). The increase in nanotube concentration (an increase in heterogeneity) shifted the proce ss toward lower frequencies and resulted in a more completely formed semi-circle, t hus indicative of MWS behavior. Some researchers have observed a positive intercept on the M’ axis at high filler loads. This behavior is indicative of macrscopic in-hom ogeneity (agglomeration) and charge buildup (Tsangaris et al. 1998; Starweather and Av akian 1992). However, results obtained in this work show that the composites exhibit, on a macroscopic level, excellent dispersion of the nanotubes within the matrix. This can be seen from the zer o intercept of the M” vs. M’ argand plots (Figures 6.30 and 6.31). This behavior is s upported by the findings presented in the paper pub lished by Tsangaris (1998). At 180oC the semi-circle is completely formed for both the neat and composite samples and resembles Debye behavior. When compounded with the high temperature behavior seen in Figures 6.13-6.18 and 6.25 it can be stated that the high temperature behavior is a result of a change in conducti on mechanism. It has been stated that PMMA approaches a liquid-li quid state around 212oC (Hedvig 1977). It has been further stated that conductivity relaxations exhibit a singl e relaxation time (Johari and Pathmanathan

PAGE 204

181 1988). Both would exhibit Debye behavi or. For the plots representing 130oC and 150oC, another relaxation process is present and is associated with dipoloar polarization. There also exists asymmetric behavi or or deviation from Debye behavior. The presence of a dipolar relaxation was also seen in th e work of Tsangaris et al. (1998). To further understand the asymmetry (associated with a distribution of relaxation times) the Havriliak Negami fitting parameters (Chapter 5) were applied to the samples treated with electric modulus at 130oC, 150oC, and 180oC. The skewing and broadening terms for the neat PMMA and PMMA/SWNT samples approached Debye behavior for all temperatures reported as listed in ta bles 6.5 and 6.6. The Debye like behavior observed with the Havriliak Negami parameters for all samples is possibly due to the approaching liquidliquid pro cess of PMMA as well as th e nature of the conductivity relaxation, and confirms the behavior seen in the above Cole-Cole plots. The neat PMMA and PMMA/soot samples also approach ed Debye behavior. Va lues could not be determined for the neat PMMA and 0.25% PMMA/soot samples at 130oC due to the dipolar relaxation occurring in the glass transition region of the polymer at these temperatures.

PAGE 205

182 01110100100010000100000 0 20 40 60 80 100 120 "Log f 130oC 140oC 150oC 160oC 170oC 180oC Figure 6.26. DEA data. Loss factor plotted against log frequency for neat PMMA above the glass transition temperature. 01110100100010000100000 -200 0 200 400 600 800 1000 1200 1400 "Log f 130oC 140oC 150oC 160oC 170oC 180oC Figure 6.27. DEA data. Loss factor plotted against log frequency for PMMA/SWNT above the glass transition temperature.

PAGE 206

183 01110100100010000100000 000 001 002 003 004 005 006 007 008 009 010 130oC 140oC 150oC 160oC 170oC 180oC Log f Figure 6.28. DEA data. Electric modulus treated Loss factor plotted against log frequency for neat PMMA above th e glass transition temperature. 01110100100010000100000 000 001 002 003 004 005 006 007 008 009 010 "Log f 130oC 140oC 150oC 160oC 170oC 180oC Figure 6.29. DEA data. Electric Modulus treated Loss factor plotted against log frequency for PMMA/SWNT above the glass transition temperature.

PAGE 207

184 Figure 6.30. Cole-Cole plot with electric modulus treatment of neat PMMA and PMMA/SWNT at 130oC, 150oC, 180oC.

PAGE 208

185 Figure 6.31. Cole-Cole plot with electric modulus treatment of neat PMMA and PMMA/soot composites at 130oC,150oC, 180oC.

PAGE 209

186 Table 6.5. Havriliak Negami values for neat PMMA and PMMA/SWNT samples polymerized via UV light at 130oC, 150oC and 180oC. Table 6.6. Havriliak Negami values for neat PMMA and PMMA/soot samples 130oC, 150oC and 180oC. o 130oC PMMA 0.000 0.1220 0.7040 0.9100 PMMA/SWNT 0.000 0.1000 0.8927 0.9870 150oC PMMA 0.000 0.1300 0.7120 0.9590 PMMA/SWNT 0.000 0.1300 0.7100 0.9780 180oC PMMA 0.000 0.1400 0.6910 0.9990 PMMA/SWNT 0.000 0.1100 0.9320 1.0000 o 130oC PMMA ------------------------0.25% PMMA/soot -------------------------1% PMMA/soot 0.0000 0.0860 0.8109 0.9853 150oC PMMA 0.0000 0.1340 0.8797 0.9974 0.25% PMMA/soot 0.0000 0.1300 0.8394 0.9214 1% PMMA/soot 0.0000 0.0930 0.7963 0.9883 180oC PMMA 0.0000 0.1440 0.8633 0.9659 0.25% PMMA/soot 0.0000 0.1400 0.8034 0.9777 1% PMMA/soot 0.0000 0.1050 0.7648 0.9693

PAGE 210

187 Conclusions Plots of log DC conductivity vs. temper ature followed Arrhenius behavior from 125oC-175oC. At 185oC a change in condction mechanis m was observed in all neat and composite (PMMA/SWNT and PMMA/soot) samp les. These plots were used to determine activation energies for neat and composite samples. The difference between activation energies for the neat PMMA samples and PMMA/SWNT samples was not large enough to draw conclusions However, the activation en ergies reveal that the high temperature process is considered a nonc ooperative simple relaxation as stated by Starkweather and Avakian (1992). The Ac tivation energies for the PMMA/soot composites increased as compared to the neat PMMA. It is probable that the composition of the soot contributed to th e higher activation energies. So ot particles could obstruct the polymer chain in a way that prevents the chains from moving out of the way, thus hindering the movement of ions and other conductive impurities. Soot particles and polymer chains could also form br idges, also preventing movement. Cole-Cole plots were effectively used to determine the extent of interfacial polarization in the composite samples. Plots sh ifted toward the origin with an increase in carbon nanotube or soot con centration indicating a Maxwel l-Wagner-Sillars process. The data revealed that at the concentr ations used in both the polymer-SWNT and polymer-soot composites good dispersion was ach ieved within the polymer matrix as indicated by the zero in tercept observed on the M’ axis. Cole-Cole plots were fitted to Havriliak Negami to determine the and parameters. Parameters revealed that at temperatures above the glass transition te mperature relaxation be havior approached Debye behavior for a single relaxation. Th is behavior compounded with the change in conduction mechanism seen in the DC conducti vity plots (Figures 6.13-6.18 and 6.25) and the shape and movement of the Cole-Cole plots confirm that the approach to Debye behavior from 130oC to 180oC corresponds to the nature of the conductivity relaxation to exhibit a single relaxation time and the dimished influence of viscoelasticity at these high temperatures.

PAGE 211

188 CHAPTER 7 Preparation of Poly (4-methyl-1-pentene)/single-walled carbon nanotubes The incorporation of carbon nanotubes (CNTs) into polymer matrices has resulted in composites that exhibit increased thermal stability, modulus, strength, electrical and optical properties (Shaffer et al. 1999; Jin et al. 2001; Hagge nmueller et al. 2000; Jia et.al 1999; Ounaies et al. 2003, Park et al. 2005, Ta tro et al. 2004; Siochi et al. 2003; Clayton et al. 2005). Several investigations have c oncluded that carbon nanot ubes can also act as a nucleating agents for polymer crystallizati on (Ryan et al. 2004; Cadek et al. 2004, Ruan et al. 2003). Various processing techniques have been employed to uniformly disperse the nanotubes in an attempt to increase interac tion at the polymer/nanotube interface (Shaffer et al. 1999; Jin et al. 2001; Haggenmueller et al. 2000; Ounaies et al 2003, Park et al. 2005, Tatro et al. 2004; Siochi et al. 2003; Clayton et al. 200 5). Our laboratory has extensively studied technique s that have been shown to effectively disperse carbon nanotubes into a polymer matrix. These st udies have produced polymer nanocomposites with enhanced mechanical, dielectric, optical and radiation resistant properties (Clayton et al. 2005; Tatro et.al 2004; Mu isener et al. 2002). In th is study we have employed an interesting procedure of pretreating the carbon nanotubes with a polar solvent, N, NDimethylformimide (DMF), then dispersi ng them via sonication in a halogenated hydrocarbon, cyclohexyl chloride, which also dissolves the non polar polymer, poly (4methyl-1-pentene) (P4M1P). Studies have show n that certain polar solvents such as N,N, dimethylformamide (DMF) (Park et al. 2002 ; Chen et. al 1998). and chlorobenzene (Robinson and Lee et al. 2003) have been effective in dispersing carbon nanotubes. However, these solvents do not dissolve th e non polar polymer. Other studies have reported the use of a binary solvent system consisting of a halogenated hydrocarbon and a non ionic surfactant as a dispersing agen t to create a polyolefin/carbon nanotube

PAGE 212

189 composite (Marissen and van Es, 2003). Poly (4-methyl-1-pentene) (P4M1P) is a line ar hydrocarbon. The isotatic form of this polymer is highly crystalline, yet is optically transparent as a result of having a crystalline phase with a lower density (0.828 g/cm3) than the amorphous phase (0.838 g/cm3) (Lopez et.al 1992). P4M1P is known to have high thermal stability; and the distinctive density behavior of the crystal line and amorphous phases has generated a great deal of research and speculation regarding the influence of this behavior on the properties of this material (Lopez et al. 2004; Choy 1981; Penn 1966; Lee and Hiltz 1984; Woodward et al. 1961; Mi yoshi et al. 2004). P4M1P could potentially serve as an ideal matrix material for high performance polymer-nanotube composites as well as a suita ble alternative to polyethylene. Although polyethylene (a hydrocarbon) has low density, high strength, high modulus, and good chemical resistance it is hard to proc ess and has a low melting temperature (130oC). P4M1P possess mechanical properties comparable to those of PE, but is easier to process and has a higher melting temperature (240oC) (Lopez et al. 1992). Composites of PE and CNTs have been produced and found to have enhanced properties; however, literature does conclude that these composites are limited in areas where high temperature applications are needed (Ruan et al. 2003). A composite of P4M1P/SWNT with 0.5% CNT loading was produced. Samples were characterized via dynamic mechanical analysis (DMA), microhardness (MH), and differential scanning calorimetr y (DSC). Nanotube dispersion was captured via optical microscopy. The processing technique employe d to fabricate the composite resulted in uniform dispersion. The dispersion quality and the effect of the nanotubes on the viscoelastic properties are pr esented and discussed.

PAGE 213

190 Experimental Materials The poly (4 methyl-1-pentene) and cyclohe xyl chloride solvent were purchased from Sigma Aldrich (Milwaukee, WI). The N,N, dimethylformamide solvent was obtained from Fisher Scientific (Pittsburgh, PA). Purified laser ablated single-walled carbon nanotubes (SWNT) were provided by the Center for Nanotechnology/NASA Ames Corporation (Moffett Field, CA). Single-Walled Carbon Nanotube Preparation Raw laser ablation material provided by NASA Johnson Space Center was purified as described elsewhere (Liu et al. 1998). The raw nanotubes were refluxed in 2.6 M nitric acid for approximately 160 hours and then di luted with double dist illed water. This solution was then centrifuged (4000 rpm), the solvent mixture decanted, and the sample was again suspended in double distilled water. This step was repeated two more times in order to remove the acid from the nanotube s. Finally, the solution was filtered through a cellulose nitrate filter and dried at 60C in a vacuum oven to form a buckypaper (Clayton et al. 2005). Polymer-Nanotube Composite Synthesis Commercial low molecular weight poly (4-methyl-1-pentene) with a measured Tm of 235oC was dissolved in cycl ohexyl chloride at 110oC to make a 3.5% solution. Laser ablated SWNTs were sonicated in N,N-dimethylformamide (DMF) using a Branson Sonifer 450 for 1 hour. The nanotube/DMF di spersion was placed in a vacuum oven at 80oC to remove the solvent. The DMF treate d nanotube paper was then dispersed in cyclohexyl chloride via sonication for 6 hours. The nanotube/solvent mixture was added to the polymer solution and sonicated together for 1 hour. The polymer/nanotube/cyclohexyl

PAGE 214

191 chloride mixture was placed in a warm beaker lined with teflon film and the solvent was allowed to evaporate at room temperature fo r 12 hours, and then placed in a vacuum oven at 80oC to remove any residual solvent. Th e dried composite with 0.5% (by wt) of SWNTs was compression molded for analysis. Pieces were placed between Kapton film and stainless steel pl ates and pressed for 5 minutes at 5000 pounds of pressure at a temperature of 246oC. Neat PMP was prepared in the same manner. After processing, the measured Tm for the neat and composite sample was 235oC. Dynamic Mechanical Analysis The viscoelastic properties were coll ected on a TA Instruments 2980 Dynamic Mechanical Analyzer (DMA). The mode was set to measure a tension film from frequencies ranging from 1 to 100 Hz with an amplitude of 5 microns. The average sample size was 14.4 x 5.8 x 1.3 mm. Because measurements are time, temperature and frequency dependent a temperat ure range was taken from -150oC to 300oC. Microhardness The Vickers hardness number (HV) for each sample was determined with a Leica VMHT MOT with a Vickers indenter. The values were taken from the average of four indents. A horizontal and a vertical reading were taken on each indent. A load of 500g and a dwell time of 20s were used. HV values we re expressed in MPa by multiplying by 9.807. Differential Scanning Calorimetry Melt temperatures (Tm) and percent crystallinity were obtained on a TA Instruments 2920 DSC. A sample amount between 2–10 mg was obtained from the compression molded sample. The samples were heated to 300oC at a rate of 5oC per minute to insure that all samples had the same thermal history. The sample was cooled with liquid nitrogen to room temperature and reheated to 300oC. The Tm and percent crystallinity values were taken from the second heat. Percent crystall inity values were calculated based on a 100%

PAGE 215

192 crystalline polymer with a heat of fusion of 61.7 J/g (Zoller et al 1986; Miyoshi et al. 2004; Reddy et al. 1997). Optical Microscopy A film of the polymer solution/nanotube mixt ure was cast onto a glass slide. Solvent was allowed to evaporate at room temperature fo r 12 hours and then placed in a vacuum oven. The film was mounted between two glass slides and images were captured on a Leica Microsystems Optical Microscope.

PAGE 216

193 Results and Discussion Literature states that halogenated hydr ocarbons, cyclohexane and cyclohexene are typical solvents used to dissolve poly ethylene and poly (4-methyl-1-pentene). Cyclohexane dissolved P4M1P, but was not effective in dispersi ng the nanotubes. 1-chlorohexane did not dissolve the polymer nor was it efficient at dispersing the nanotubes (Figure 7.1a). Cyclohexyl chlo ride was found to create a uniformed solvent/nanotube mixture (Figure 7.1b) as well as a uniformed solvent/polymer/nanotube mixture (Figure 7.1c). Figure 7.2 is an optic al micrograph of the P4M1P thin film. The picture in the inset is that of the neat. Figure 7.1. (a) carbon nanotubes sonicated in 1-chlorohexane pretreated with DMF, (b) carbon nanotubes sonicated in cy clohexyl chloride pretreated with DMF, (c) carbon nanotubes/cyc lohexyl chloride/polymer.

PAGE 217

194 Figure 7.2. Optical Micrograph of (a) neat P4M1P and (b) 0.5% P4M1P/SWNT composite. 10 x 0.3 magnification.

PAGE 218

195 Dynamic Mechanical Analysis P4M1P has three reported mechanical relaxations: the a also referred to as a) ranging from 20oC – 67oC (Woodward et al. 1961; Miyos hi et al. 2004; Reddy et al. 1997), a broad high temperature relaxation (c) ranging from 105oC – 135oC (Lopez et al. 1992; Reddy et al. 1997; Choy et al. 1981; Miyoshi et al. 2004) and a low temperature peak (orsc) was also observed at -123oC (Woodard et al. 1961) and -140oC (Choy et al. 1981). The low temperature relaxation () was not seen in the frequency range used for this study. It is defined as the rotation of the side groups and depends on the amount of amorphous characte r present (Lopez et al. 1986). The a transition is the glass transition region a ssociated with the segmental motion of the polymer main chain (Penn 1966; Choy et al. 1981). The nature of the c transition is associated with motions within the crystalline phase and is believed to be an expansion of the unit cell parameter a (Lopez et al. 1992, Penn 1986, Ranby et al. 1962). Figure 7.3 is a plot of the loss modulus (E”) plotted against temperature for the neat and composite samples from -150oC to 250oC and 1 Hz to 60 Hz. The loss modulus of the composite sample increases with the addition of the carbon nanotubes. The high temperature relaxation (c) is more pronounced in the composite sample as compared to the neat. The percent crystallin ity, as determined from DSC plots, (Figures 7.4 and 7.5) for the neat and composite sa mples was 68% and 74%, respectively. The elastic modulus (E’) represents the material’s stiffness. The stiffness of the composite at 60 Hz and -50oC, 25oC, and 50oC is higher than that of the ne at as indicated in Table 7.1, with the highest modulus existi ng at temperatures below the Tg region (Figure 7.6). Further, an increase in stiffness should correlate to an increase in th e percent crystallinity of the polymer (Gedde 1999). To further s upport the increase in vi scoelastic properties, the composite had a Vickers hardness number of 97 MPa as compared to 82 MPa for the neat. The enhanced relaxation intensity of the crystalline region (c) is indicative of increased interaction between the carbon nanotubes and pol ymer matrix. Studies have shown that carbon nanotubes can act as nucleatin g agents (Ryan et al. 2004; Cadek et al.

PAGE 219

196 2004, Ruan et al. 2003; Bhattacharyya 2003). It was shown that uniform dispersion and good interfacial bonding between CNTs and pol yethylene resulted in secondary crystal growth, thus enhancing the duc tility of the composite (Rua n et al. 2003). Further, a crystalline layer formed on MWNTs, contributed to the enhanced mechanical properties of polyvinylalcohol/MWNT composite s (Cadek et al. 2004). 0 50 100 150 200 250Loss Modulus (MPa) -150-5050150250Temperature (C) – – – – neat PMP ––––––– 0.5% PMPCNTUniversal V3.4C TA Instruments Figure 7.3. DMA data. Loss Modulus (E”) plotted against temperature for neat P4M1P and P4M1P/SWNT. a c

PAGE 220

197 23533C 229.00C 42.04J/g 6813 % crystallized -0.6 -0.4 -0.2 0.0 0.2Heat Flow (W/g) -200-100 0 100200300400 Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 7.4. DSC data. DSC Plot of neat P4M1P.

PAGE 221

198 235.29C 227.76C 4536J/g 7351 % crystallized -0.6 -0.4 -0.2 0.0 0.2Heat Flow (W/g) -200-100 0 100200300400 Temperature (C) Exo Up Universal V3.4C TA Instruments Figure 7.5. DSC data. DSC Pl ot of neat P4M1P/SWNT.

PAGE 222

199 Tg = 37C Tg = 43C60 Hz E' = 2409MPa -50C E' = 3716MPa -50C 0 1000 2000 3000 4000 5000Storage Modulus (MPa) 0 50 100 150 200 250Loss Modulus (MPa) -150 -50 50 150 250 Temperature (C) neat PMP 0.5% CNTPMP Universal V3.4C TA Instruments Figure 7.6. DMA data at 60Hz of E and E. E (MPa) @ 60 Hz -50oC 25oC 50oC Neat PMP 2409 1710 918 0.5 % PMP/CNT 3716 2713 1494 Table 7.1. Storage Modulus (E) values at 60 Hz and -50oC, 25oC, and 50oC.

PAGE 223

200 In semi-crystalline polymers, the glass tran sition region is restri cted by crystals and exhibit broader relaxa tions than in the Tg region of fully amorphous polymers (Gedde 1999). Thus, glass transition temperatures are difficult to decipher in differential scanning calorimetry plots. However, DMA is a useful tool in dete rmining these values. Moreover, being that relaxa tions are time, temperature and frequency dependent, Tg values reported from DMA must specify th e frequency in which the temperature is reported. The glass transition temperatures for the neat and composite samples taken at 60 Hz were found to be 37oC and 43oC. The maximum loss peak height obtained from DMA will shift to higher temperatures. In a narrow temperature rang e, the shift or frequency is linear (Gedde 1999). Temperature dependency of semi-crystalline polymers conforms to Arrhenius behavior (McCrum 1967). Figures 7.7 and 7. 8 are Arrhenius plots of neat P4M1P and the composite. Activation energies were obtained by multiplying the slope of the line by the gas constant (1.987cal/mol K). The neat had an activation energy of 59 kcal/mol with that of the composite being 76 kcal/mol. The energy needed to induce flow in the composite was higher. The reason for this in crease is two-fold: (1) the presence of the nanotubes hindering chain movement and (2) the presence of a crystal layer on the CNTs, thus increasing the crystallinity in this region which in turn restricts the mobility of the amorphous region. Activation energies associat ed with viscous flow are large due to the cooperative behavior present in this region (Starkweather 1981). Lee and Hiltz (1984) obtained an activation energy of 106 kcal/m ol and Choy et al. (1981) reported 60 kcal/mol. Activation energies vary dependi ng on the method used for testing, thus they are only approximations.

PAGE 224

201 Arrhenius Plot for the a relaxation in Neat P4M1Py = -30031x + 98.428 R2 = 0.99460 1 2 3 4 5 6 0.00310.003120.003140.003160.00318 0.00320.003220.003240.003260.003281/T (K )ln frequency Ea= 59 kcal/mol Figure 7.7. Arrhenius Plot for neat P4M1P from 1Hz to 100 Hz. Arrhenius Plot for the a relaxation in P4M1P/SWNTy = -38485x + 123.96 R2 = 0.98090 1 2 3 4 5 6 0.003090.003110.003130.00315 0.003170.003190.003211/T (K )ln frequency Ea= 76 kcal/mol Figure 7.8. Arrhenius Plot for 0.5% P4M1P/SWNT from 1 Hz to 100 Hz.

PAGE 225

202 The Williams, Landel and Ferry equation (7.1) accounts for curvature present in the Arrhenius plot (Gedde 1999; Starkweather 1981). In this study, the values for C1, C2, and the reference temperature To (Tg) were obtained from a curv e fitting program (Gao 1997); aT represents the shift factor or frequency and T is the given temperatur e. Table 2 lists the values reported by Penn (1966) and L ee and Hiltz (1984). Deviations from the universal constants are typical due to variations in the glas s transition temperatures and the methods used to obtain these values (McCrum 1967). ) ( ) ( log2 1 o o TT T C T T C a (7.1) Table 7.2. WLF shift constants for poly (4-methyl-1-pentene) and P4M1P/SWNT. Sample To C1 C2 Universal constants ----17.4 51.6 Neat PMP 32.6 9.90 56.3 0.5%PMP/CNT37.7 10.2 48.1 Lee and Hiltz* ----20.7 37.0 Penn* 25.0 17.3 40.4

PAGE 226

203 The WLF shift constants, C1 and C2, can be used to predict mechanical behavior of a polymer over a wide range of frequencies. In this study, 1 Hz, 3Hz, 6Hz, 10Hz, 30 Hz, 60 Hz, and 100 Hz were used to obtain mechanical data. To further understand the behavior of P4M1P as a function of time and temperature over a wide range of frequencies a master curve was generated uti lizing the WLF shift constants. Figure 7.9 is a plot of master curves for the neat and co mposite samples. It is clear that over a wide range of frequencies and temperatures, P4M1P c onforms to WLF. Figure 7.10 is a plot of the glass transition region of P4M1P using the WLF shift constants. These results are comparable to WLF treatment of P4M1P prev iously published (Penn 1966; Lee and Hiltz 1984). 2030405060708090100 -6 -4 -2 0 2 4 Log f (Hz)T (oC) Neat P4M1P P4M1P/SWNT Composite Figure 7.9. Master Curve of neat P4M1 P and P4M1P/SWNT composite from 3 x 10-6 Hz to 1000 Hz and a temperature range of 20oC to 100oC.

PAGE 227

204 203040506070 -4 -3 -2 -1 0 1 2 3 4 Log f (Hz)T(oC) Neat P4M1P P4M1P/SWNT Composite Figure 7.10. Master curve of reported Tg region for P4M1P using WLF shift constants. The WLF constants can also be used to calculate the fractional free volume (fg) and the thermal expansion coefficient (f) (Table 7.3) of a polymer (Aklonis et al. 1972; Emran 2000). Equations 7.2 and 7.3 were used to calculate fg and f where B is equal to 1. 1) 303 2 (C B fg (7.2) 2C fg f (7.3) fg defines the amount of unoccupied space betw een chain segments as a result of chain segment packing (Aklonis et al. 1972). Conclusions can not be made based on the calculated fractional free volume and coeffici ent of thermal expansion for the neat and composite sample due to the small loading of carbon nanotubes; however, it can be stated that the composite can be used in applicati ons in which the pure polymer is desired.

PAGE 228

205 Table 7.3 WLF constants and calculated fractional free volume and expansion of thermal coefficient values. Sample To C1 C2 fg af Neat P4M1P 32.6 9.90 56.3 0.0439 0.779 0.5%P4M1P/CNT 37.7 10.2 48.1 0.0430 0.884

PAGE 229

206 Conclusions Carbon nanotubes were successfully incorpor ated into poly(4-methyl-1-pentene). The processing technique employed was found to be effective in dispersing the nanotubes in the polymer. Further, analysis of the composite confirmed that the nanotubes did in fact serve as a good reinforcement agent fo r the polymer. The composite sample exhibited an increase in modulus and gl ass transition temperature as observed via dynamic mechanical analysis. The crystalline region as noted in the loss modulus data was found to enhance with the addition of carbon nanotubes, indicat ing good interaction between the polymer-nanot ube interface. Experimental data for the composite sa mple was fitted to WLF parameters and found to be consistent with values obtained for neat poly (4-methyl -1-pentene) in this study and previously published results (Penn 1966; Lee and Hiltz 1984); thus characterization techniques can be ex tended to polymer-nanotube composites.

PAGE 230

207 CHAPTER 8 Conclusions The research presented in this disserta tion provides useful processing techniques and characterization methods for polymer carbon nanotube composites. It is evident that the design and fabrication of polymer nanotube composites are vital steps in achieving a nanocomposite with desired properties for speci fic applications. During the course of this work, techniques such as sonica tion, melt mixing, so lution mixing, and in situ polymerization were utilized to disperse th e carbon nanotubes in the polymer matrix. The combination of sonication, in situ polymerization, dissolution and solvent evaporation proved to be a su ccessful fabrication technique in producing thin films with enhanced optical properties as compared to a composite fabricated via melt compounding (Clayton et al. 2005; Muisener et al. 2002). Carbon nanotubes were found to increase the dielectric properties of the polymer, while limiting the loss in transparency. CNTs were found to enhance the low temperature dielectric relaxation in PMMA, indicating an interaction at the polymer-nanotube interf ace at low temperatures. The dielectric constant and refractive index were successfu lly correlated using Maxwell’s Relationship and provided new information concerning nano tube effect on polarization and optical dispersion of a dielectric material. The incorporation of unpurified carbon na notubes proved to enhance radiation resistance at loadings smaller than 1%. The results obtained by DSC compounded with the results obtained from microhardness measur ements revealed that the 0.5% composite exhibited the highest degree of radiation resistance as compared to the neat, 0.25% and 1% composite samples. Characterization via DSC, MH, and SEM confirmed that the composite with 1% soot loading experience d strong agglomeration, thus decreasing its resistance to radiation. A comparison of radiation re sistance study of PMMA/soot composites with similar studies c onducted on PMMA/SWNT and PMMA/MWNT

PAGE 231

208 composites (Muisener et al. 2002; Harmon et al. 2001; Tatro et al. 2004) shows that the single-walled and multi-walled carbon nanotubes are more suitable fillers than the unpurified soot for mechanical radiation resist ance. However, dielectric data revealed that soot particles enhanced polymer relaxa tions and enhanced the conductive nature of the polymer. Employing a binary solvent system allowe d for the successful incorporation of carbon nanotubes into a non polar semi-crystalline polymer, resu lting in a composite with increased mechanical properties. Carbon nanotubes served as nucleating agents, thus increasing the crystallinity of the compos ite which contributed to the increase in mechanical properties. The ability of the CNTs to nucleate provided an ideal environment for good interaction at the pol ymer-nanotube interface. The neat and composite samples were found to conform to WLF behavior. Conventional polymer characte rization techniques were ut ilized in this work and were found to provide great insight on the influence of carbon nanotubes on the polymer matrix. Dielectric Analysis was found to be the most sensitive technique. Further, mathematical modeling of dielectric data provided information regarding dipolar relaxation processes and conducti on behavior of the composites. Dielectric analysis via Cole-Cole plots and Havriliak Negami para meters determined that good dispersion was achieved in the composites. The end use implications of the compos ites discussed and the methods used to fabricate and characterize them have provided useful data for the future of nanotechnology and polymer nanotechnology in ar eas such as space exploration, offshore oil drilling, automotives, and electronics. The need for light weight and/or high performance polymeric material and the high strength and good electr ical properties of carbon nanotubes has led to millions of dollars in research funding.

PAGE 232

209 REFFERENCES Aklonis, J.J., W.J. MacKnight, and M. Shen. 1972. An Introduction to Polymer Viscoelasticity. New York: Wiley-Interscience. Assael, M.J., C.F. Chen,I. Metaxa, W.A. Wakeham. 2004. Thermal Conductivity of Carbon Nanotubes in Water. NIST Symposium. Assouline, E., A. Lustiger, A.H. Barber, C.A. Cooper, E. Klein, E. Wachtel, H.D. Wagner. 2003. Journal of Polymer Science: Part B: Polymer Physics 41:520. Bahr J. and J.M. Tour. 2002.Journal of Materials Chemistry.12. Bandyopadhyay, S., A.J. Hsieh, and E.P. Giannells. 2002. PMMA Nanocomposites Synthesized by Emul sion Polymerization. In Polymer Nanocomposites – Synthesis, Characterization and Modeling, Ed. Krishnamoorti R., and R. Vaia. ACS Symposium Series, Vol. 804. Washingt on, DC: American Chemical Society. Bershtein, V. A., V. M. Egorov. 1994. Differential Scanning Calorimetry of Polymers. Physics,Chemistry,Analysis,Technology. New York: Ellis Horwood Limited. Bergman, R., F. Alvarez, A. Alegria, and J. Colmenero. 1998. Journal of Chemical Physics 109: 7546. Bernholc, J., M.B. Nardelli, J-L Fattebert, D. Orlikowski, C. Roland, and Q. Zhao. 2000 Mechanical Properties and Electro nic Transport in Carbon Nanotubes. In Science and Application of Nanotubes, Ed. D. Tomanek and R.J. Enbody. New York: Kluwer Academic. Bertolucci, R. P. H. and Harmon, J. P. 1998. Polymer Engineering and Science 38:699. Bhattacharyya, A.R., T.V. Sreekumar, T. Liu, S. Kumar, L.M. Ericson, R.H. Hauge, R.E. Smalley. 2003. Polymer 44:2373. Bovey, F.A. and G.V.D. Tiers. 1960. J. Polymer Sci. 44:173. Bradley, G., Ed. 1998. Photinitiators for Free Ra dical Cationic and Anionic Photopolymerization, 2nd ed. New York: John Wiley and Sons Publishers. Brosseau, C., A. Beroual, A. Boudida. 2000. Journal of Applied Physics, 88:12.

PAGE 233

210 Burfield, B.R. and Y. Doi. 1983. Macromolecules, 16:702. Cadek, M.,J.N. Coleman, K.P. Ryan, V. Nicolo si, G. Bister, A. Fonesca, J.B. Nagy, K. Szostak, F. Beguin, W.J. Blau. 2004. Nano Lett. 2: 353. Calleja, F.J.B. and S. Fakirov. 2000. Microhardness of Polymers. Cambridge: Cambridge University Press. Charlier, J.C. and J.P. Michenaud. 1993. Phys. Rev. Lett. 70:1858. Chen J., M.A. Hamon, H. Hu, Y. Chen, A. Rao, P.C. Eklund, R.C. Haddon.1998. Science, 282. Cheng H., F. Li, G. Su, H. Pan, M. Dresselhaus. 1998. Appli. Phys. Lett. 72:3282. Chiang, I.W., B.E. Brinson, R.E. Smalle y, J.L. Margrave, and R.H. Hauge. 2001. J. Phys. Chem. B, 105. Choy, C.L., W.K. Luk, F.C. Chen. 1981. Polymer 22:543. Clayton, L.M., A. Sikder, A. Kumar, M. Ci nke, M. Meyyappan, T. Gerasimov, and J.P. Harmon. 2005. Advanced Functional Materials 15. Clayton, L.M., T. G. Gerasimov, M. Cinke, M. Meyyappan, J.P. Harmon. 2004. Polymer Bulletin, 52. Clegg, D.W. and A.A. Collyer. 1991. Irradi ation Effects on Polymers. New: York: Elsevier Science Publishers. Clough, R. L. and Gillen, K. T. 1991. Radiation Resistance of Polymers and Composites. Irradiation Effects on Polymers. New York: Elsevier Science Publishers. Cole, K.S., R.H. Cole. 1941. Journal of Chemical Physics 9: 341. Coleman, J. N, A.B. Dalton, S. Curran, A. R ubio, A.P. Davey, A. Drury, B. McCarthy, B. Lahr, P.M. Ajayan, S.R. Roth, R.C. Barkle, W.J. Blau. 2000. Advanced Materials,12. Dyke, CA. and J.M. Tour. 2004. ChemistryA European Journal. 10: 812. Collins, A., J. Barnes, F.W. Billmeyer, Jr. 1973. Experiments in Polymer Science. New York: John Wiley and Sons.

PAGE 234

211 Collins, P.G., and Ph. Avouris. 2000. Nanotubes for eletronics. Scientific American. Cooper, C.A. and R.J. Young. 2000. Investigation of the defo rmation of carbon nanotube composites through the use of raman spectroscopy. In Science and Application of Nanotubes, Ed. D. Tomanek and R.J. Enbody. New York: Kluwer Academic. Dai H. 2002. Surface Science 500:218. Davey, M.E. 2004. Manipulating Molecules: Fe deral Support for Nanotechnology Research. Congressional Research Service. The Library of Congress. Dresselhaus, M.S. G. Dresselhaus, and Ph. Avouris. Eds. 2001. Carbon Nanotubes: Synthesis, Structure, Properties, and Applications. Germany: Springer. Dresselhaus, M.S., G. Dresselhaus, P.C. Eklund. 1996. Science of Fullerenes and Carbon Nanotubes. New York: Academic Press. Ebewele, R.O. 2000. Polymer Science and Technology. Boca Raton, FL: CRC Press. Emran, S.K. 2000. Viscoelastic Properties of Dendrimers, Dendrimer Blends, and Dendrimer Gels. University of Sout h Florida: Doctor al Dissertation. Evans, R.W. 1997. Design Guidelines for Shielding Effectiveness, Current Carrying Capability, and the Enhancement of C onductivity of Composite Materials. NASA Report, NASA CR-4784. Ferry, J.D. 1961. Viscoelastic Properties of Polymers. New York: Wiley. Feynman R. 1960. “There’s Plenty of Ro om at the Bottom,” Speech given at the Californina Institute of Technology (Decem ber 29, 1959). Published in February 1960 in Engineering and Science. Forro, L. J.P Salvetat, J-M. Bonard, R. Bacsa, N.H. Thomson, S. Garaj, L. Thien-Nga, R. Gaal, A. Kulik, B. Ruzicka, L. Degior gi, A. Bachtold, C. Schonenberger, S. Pekker, K. Hernadi. 2000. Elelctron and Mechanical Properties of Carbon Nanotubes. In Science and Application of Nanotubes, Ed. D. Tomanek and R.J. Enbody. New York: Kluwer Academic. Foust, J. 2003. The Space elevator: going up? (Part 2). Article appeared in The Space Review: essays and commentary about the final frontier September 22, 2003. Web Article http://www.thespacereview.com/article/48/1 Fox, T.G. and P.J. Flory. 1950. J.Appl. Phys. 21:581. Fox, T.G. and H.W. Schnecko. 1962. Polymer 3: 575.

PAGE 235

212 Gao, H. 1996. Viscoeleasticity of P-Alkylated and Halogenated Polystyrenes. University of South Florida: Doctoral Dissertation. Garwe, F., A. Schonhals, H. Lockwenz, M. Beiner, K. Schroter, and E. Donth. 1996. Macromolecules 29:247. Gedde, U. 1999. Polymer Physics. London: Chapman and Hall. Goyanes, S. N., Benites, G. M., Rubi olo, G. H. and Marzocca, A. J. 1996. Journal de Physique III 6:587. Goyanes, S. N., Benites, G. M., Gonzalez, J. J., Rubiolo, G. H. and Marzocca, A. J. 1997. Polymer Testing. 16:7. Grimes, C.A. C. Mungle, D. Kouzoudis, S. Fang, P.C. Eklund. 2000. Chemical Physics Letters. 319:460. Hamon, M.A., J. Chen, H. Hu, Y. Chen, M.E. Itkis, A. Rao, P.C. Eklund, R.C. Haddon. 1999. Advanced Materials, 11. Haggenmueller, R., H.H. Gommans, A.G. Rinz ler, J.E. Fischer, and K.I, Winey. 2000. Chemical Physics Letters 330:219. Harmon, J.P., Muisener, P., L. Clayton, J. D’ Angelo, A. K. Sikder, A. Kumar, M. Meyyappan, A. M. Cassell. 2001. Material Research Society Symposium Proceedings, 697. Harmon, J. P., Gaynor, J. F. and Taylor, A. G. 1993. Radiat. Phys. Chem. 41:153. Hatakeyama, T., Quinn, F.X. 1999. Thermal Analysis, Fundament als and Applications to Polymer Science. New York: John Wiley and Sons. Havriliak Jr. S. and S.J. Havriliak. 1997. Dielectric and Mechani cal Relaxation in Materials. New York: Hanser Publishers. Havriliak, S. 1968. Polymer 9: 289. Hedvig, P. 1977. Dielectric Spectroscopy of Polymers. New York: John Wiley and Sons. Hewlet Packard. 1990. HP 8452A Diode A rray Spectrophotometer Handbook. HP Part No. 08452-90002. Hirsch, A. 2002.Angew. Chem. Int. Ed. 41: 1853. Hohne, G. W. H., W. Hemminge r, H. J. Flammersheim. 1996. Differential ScanningCalorimetry, An Introduction for Practitioners. Springer.

PAGE 236

213 Huang, W., Y.Lin, S. Taylor, J. Ga illard, A.M. Rao, Y. Sun. 2002. Nano Letters 2:231. Ijima,S. 1991. Nature 354:56. Jia, Z., Wang, C. Xu, J Liang, B. Wei, D. Wu, S. Zhu. 1999. Materials Science and Engineering A A271: 395. Jin, Z., K.P. Pramoda, G. Xu, S.H. Goh. 2001. Chem. Phys. Lett. 337: 43. Johari, G.P. and P. Pathmanathan. 1988. Physics and Chemistry of Glasses 29: 219. Keyes D.L.1988. Optical Testing Characterization. In Engineering Plastics: Engineered Materials Handbook. Ed. C.A. Dostal. Vol. 2 Engineering Plastics. Ohio: ASM International. Klabunde, K.J. 2001. Nanoscale Materials in Chemistry. New York: Wiley-Interscience. Korkakas, G., C.M. Gomez and C.B. Bucknell. 1993. Plast., Rubber and Comp. Proc. And Applications 19:285. Koshio A., M. Yudasaka, M. Zhang, S. Iijima. 2001. Nano Letters 1:361. Krishnamoorti R. and R. Vaia. 2002. Polymer Nanocomposites: Introduction. In Polymer Nanocomposites – Synthesis, Characterization and Modeling, Ed. Krishnamoorti R., and R. Vaia. ACS Sym posium Series, Vol. 804. Washington, DC: American Chemical Society. Krishnan, A., E. Dujardin, T.W. Ebbesen, P.N. Yanilos, and M.M. J. Treacy. 1998. Phys. Rev. B. 58: 14013. Kudoh, H., T. Sasuga, T. Seguchi and Y. Katsumura. 1996. Polymer 37:4663. Kusy, R.P., J.Q. Whitley, S. Kalachandra. 2001. Polymer, 42:2585. Kymakis, E. G.A.J. Amaratunga. 2004. Synthetic Metals 142:161. Kymakis, E., I. Alexandou, G.A.J. Amaratunga. 2002. Synthetic Metals 127:59. Lambin, Ph., J.C. Charlier, and J.P. Michen aud. 1994. In Progress in Fullerene Research, H. Kuzmany, J. Fink, M. Mehring, and S. Roth ( Ed). World Scientific, Singapore, 130. Lambin, Ph., L. Phillipe, Charlier, J.C. and J.P. Michenaud. 1994. Computational Materials Science 2: 350.

PAGE 237

214 Lee, G.F. and T.H. Hiltz. 1984. Journal of Applied Polymer Science 29:3057. Leica. 1999. Leica VMHT MOT Operating Instructions. Austria: Leica. Liu, J., A. G.Rinzler, H. Dai, J. H. Hafner, R. K. Bradley, P. J. Boul, A. Lu, T. Inverson, K. Shelimov, C. B. Huffman, F. Rodrigue z-Macias, Y.-S. Shon, T. R. Lee, D. T. Colbert, R. E.Smalley. 1998. Science 280: 1253. Lopez, L.C., G.L. Wilkes, P.M. Stricklen, S.A.White. 1992. Journal of Macromol. Sci. Chem. 32:301. MacKinnon, A.J., S.D. Jenkins, P.T. McGrail and R.A. Pethrick. 1992. Macromolecules 34:3252. Marissen, R. and M.A. van Es. 2003. Method of Producing hi gh Strength elongated products containing carbon nanotubes. European Patent. EP 1336672A1. Martel, R. H.R. Shea, and Ph. Avouris. 1999. Journal of Physical Chemistry B. 103. Martel, R., R. Shea and P. Avouris.1999. Nature 398:299. Maxwell, J.C. 1989. Electricity and Magnetism, Vol.1. Oxford: Clarendon Press. Meier, D.J. 1978. Molecular Basis of Transistions and Relaxations. London: Gordon and Breach Science Publishers. Mishra, M.K., Y. Yagci,. Eds. 1998. In Handbook of Radical Vinyl Polymerization. New York: Marcel Dekkar. Miyoshi, T., O. Pascui, and D. Reichert. 2004. Macromolecules 37: 6460. McCrum, N.G., C.P. Buckley, and C.B. Bucknall.1997. Principles of Polymer Engineering, 2nd Ed. New York: Oxford University Press. McCrum, N.G., B.E. Read, and G. Williams.1967. Anelastic and Dielectric Effects in Polymeric Solids. New York: Dover Publications, Inc.. Morton-Jones, D.H. 1989. Polymer Processing. New York: Chapman and Hall. Muisener, P., L. Clayton, J.D'Angelo, J. Harmon, A.K. Sikder, A. Kumar, M. Meyyappan, B. Chen, and A. Cassell. 2002. J. Mater. Research 17:2507. Muster, J., M Burghard, S. Roth, G.S. Dusberg, E. Hernandez, A. Rubio. 1998. J. Vac. Sci. Technol. 16:2796.

PAGE 238

215 Nalwa, H.S.Ed. 2000. Handbook of Nanostructure d Materials and Nanotechnology. Vol. 5. New York: Academic Press. National Research Council. 1994. Polymer Science and Engi neering: The Shifting Research Frontiers.Washington, D.C.: National Academy Press. Nikolaev P., M.J. Bronikowski, R.K. Bradley, F. Rohmund, D.T. Colbert, K.A. Smith, R.E.Smalley. 1999. Chem. Phys. Lett. 313:91. Ounaies, Z., C. Park, K.E. Wise, E. Siochi, and J.S. Harrison. 2003. Composites Science and Technology 63:1637. Overney,G., W. Zhong, and D. Tomanek. 1993. Zeitschrift fr Physik D Atoms Molecules and Clusters (Z. Phys. D) 27: 93. Paine, R.T., E.A. Pruss, G.L. Wood, C. Schw ierkowski, R.F. Hill, C. Chapelle, and W.J. Kroenke. 2002. Boron Nitride Fillers for Organic Polymer Composites. In Polymer Nanocomposites – Synthesis, Characterization and Modeling, Ed. Krishnamoorti R., and R. Vaia. ACS Sym posium Series, Vol. 804. Washington, DC: American Chemical Society. Park C., Z. Ounaies, K.A. Watson, R.E. Crooks, J. Smith, S.E. Lowther, J.W. Connell, E.J. Siochi, J.S. Harrison, T.L. St. Clair. 2002. Chemical Physics Letters 364. Penn, R.W. 1966. Journal of Polymer Science: Part A-2 4:559. Petrovic, Z.S., I. Javnis, A. Waddon, and G.J. Banhegyi. 2000. Journal of Applied Polymer Science 76: 133. Pissis, P., A. Kyritsis. 1997. Solid State Ionics 97:105. Polizos, G., A. Kyritsis, P. Pissis, V.V. Shilov, V.V. Shevchenko. 2000. 136-137:1139. Pompeo F. and D.E. Resasco.2002. Nano Letters. Potschle, P., T.D. Fornes, D.R. Paul. 2002. Polymer Quian, D., E.C. Dickey, R. Andrews, and T. Rantell. 2000. Applied Physics Letters 76: 2868. Rao, C.N.R., B.C. Satishkumar, A. Govindaraj, M. Nath. 2001. Chem Phys Rev.2:78. Rao, C.N. R., 1967. Ultra-Violet and Visible Spect roscopy Chemical Applications. New York: Plenum Press.

PAGE 239

216 Ratner, M. and D. Ratner. 2003. Nanotechnology: A Gentle In troduction to the next Big Idea. New Jersey: Prentice Hall. Reddy.S., P. Desai, A.S. Abhiraman, H.W. Beckham, A.S. Kulik, H.W. Spiess. 1997. Macromolecules 30:3293. Reich, L. and S. Stivala. 1971. Elements of Polymer Degradation. New York: McGrawHill. Ruan, S.L., P. Gao, X.G. Yang, T.X. Yu. 2003. Polymer 44:5643. Ribelles, J.L., R. Diaz-Calleja. 1985. Journal of Polymer Science: Polymer Physics 23:1297. Robinson, L.A., S-B Lee, K.B. Teo, M. C hhowalla, G.A.J Amaratunga, W.I. Milne, D.A. Williams, D.G. Hasko, H. Ahmed. 2003 Nanotechnology 14. Rosen, S.L. 1993. Fundamental Principles of Polymeric Material. New York: John Wiley and Sons, Inc. Runt, J.P. and J.J. Fitzgerald. 1997. Dielec tric Spectroscopy pf Polymeric Materials Fundamentals and Applications. Washingt on, DC: American Chemical Society. Ryan, K.P., S.M. Lipson, A. Drury, M. Cadek, M. Ruether, S.M. O’ Flaherty, V. Barron, B.McCarthy, H. J. Byrne, W.J. Blau, J. N. Coleman. 2004. Chem. Phys. Letters 391:329. Sandler, J., M.S.P. Shaffer, T. Prassee, W. Bauhofer, K. Schulte, and A.H. Windle. 1999. Polymer 40. Saito, R. and H. Kataura. 2001. Optical Properties and Raman Spectroscopy of Carbon Nanotubes. In Carbon Nanotubes: Synthesis, Structure, Properties, and Applications. Eds. Dresselhaus, M.S. G. Dresselhaus, and Ph. Avouris.Germany: Springer. Saito, R. G. Dresselhaus, M.S. Dresselhaus. 1993. Journal of Applied Physics, 73:494. Salvetat, J.P., G.A.D. Briggs, J.M. Bonard, R.R. Basca, A.J. Kulik, T. Stockli, N.A. Burnham, L. Forro. 1999. Phys. Rev. Lett. 82: 944. Sayyah, S. M., Sabbah, I. A., Ayahub, M. M. H., Barsoum, B. N. and Elwy, E. 1997. Polymer Degradation and Stability. 58:1. Starweather, H.W. Jr. 1981. Macromolecules 14:1277.

PAGE 240

217 Sepe, M. 1998. Dynamic Mechanical Analysis for Plastics Engineering. New York:Plastics Design Library. Sen, R., B. Zhao, D. Perea, M. Itkis, H. Hu, J. Love, E. Bekyarova, and R. Haddon. 2004. Nano Letters 4: 459. Seymour, R.B. and C.E. Carraher, Jr., 1992. Polymer Chemistry An Introduction, 3rd Ed. New York: Marcel Dekker. Shaffer, M.S.P., A.H. Windle.1999. Adv. Mater., 11: 937. Shelimov K.B., R.O. Esenaliev, A.G. Rinzler, C.B. Huffman, R.E. Smalley. 1998. Chemical Physics Letters, 282. Shrempel, F. and Witthuhn, W. 1997. Nuclear Instruments a nd Methods in Physics Research B. 132:430. Sillars, R.W. 1937. IEEE Journal 80:378. Simon, G.P. 1986. Dielectric Relaxation Spectroscopy of Thermoplastic Polymers and Blends. Sydney: Pergamon Press. Simonis, P. A. Volodin, E. Seynaeve, Ph. Lambin and Ch. Van Haesendonck. 2000. Comparative Study of Colied Carbon Nanotube by Atomic Force Micrscopy and Scanning Electron Microscopy. In Science and Application of Nanotubes, Ed. D. Tomanek and R.J. Enbody. Ne w York: Kluwer Academic. Siochi, E.J., P.T. Lillehei, K.E. Wise, C. Park, J.H. Rouse. 2003. Design and characterization of ca rbon nanotube composites. NASA Langley Research Center. Paper presented at the 4th Intl. Symp. In MEMS and Nanotechnology, Charlotte, NC. Skoog, D.A. and J.J. Leary. 1992. Principles of Inst rumental Analysis. New York: Saunders College Publishing. Skoog, D.A., D.M. West, and F.J. Holler. 1992. Fundamentals of Analytical Chemistry. Texas: Saunders College Publishing. Smalley, R. Summer 1999. Dr. Smalley addr essed the U.S. House of Representatives: Committee on Science, Subcommitt ee on Basic Research. Hearing: “Nanotechnology: The State of Nano-Scie nce and Its Prosp ects for the Next Decade. Washington, D.C.: June 22, 1999.

PAGE 241

218 Smith, J.G. K.A. Watson, C.M. Thompson, and J.W. Connell. 2003. Carbon Nanotube/Space Durable Polymer Nanocompos ite Films for Eletrostatic Charge Dissipation. NASA report. Starweather, H.W. Jr., P. Avakian. 1992. Journal of Polymer Scie nce: Part B: Polymer Physics 30:637. Steeman, P.A.M., F.H.J. Maurer and M.A. van Es. 1991. Polymer 32:351. Stephen, C., T.P. Nguyen, M. Lamy de la Chap elle, S. Lefrant, C. Journet, P. Bernier. 2000. Synthetic Metals 108:139 Stevens, M.P. 1999. Polymer Chemistry: an Introduction, 3rd Ed. New York: Oxford. TA Instruments.1998. Differential Scanning Ca lorimetry Operator’s Manual. Delaware: TA Instruments. TA Instruments.1998. Dielectric Analyz er Operator’s Manual. Delaware: TA Instruments. TA Instruments.2000. Dielectric Analyz er Operator’s Manual. Delaware: TA Instruments. TA Instruments.1998. Dynamic Mechanical An alyzer Operator’s Manual. Delaware: TA Instruments. TA Instruments. 2004. DSC Brochu re. Delaware: TA Instruments. Tabor, D. 1951. The Hardness of Metals. New York: Oxford University Press. Tatro, S., L.M. Clayton, P.A. Muis ener, A.M. Rao, J.P. Harmon. 2004. Polymer 45:1971. Tatro, S.R. 2002. Ionizing Radiation Effects on Polymers and Polymer Nanocomposites. University of South Florida: Doctoral Dissertation. Tibbetts, G.G. J. Cryst. Growth. 1984. 66:632. Tibbetts, G.G. Filaments and Composites. 1990. In Carbon Fibers. Kluwer Academic, Amsterdam, pp.73-94. Treacy, M.M. J., T.W. Ebbesen, J.M. Gibson. 1996. Nature 381: 678. Tsangaris, G.M., G.C. Psarras, and N. Kouloumbi. 1998. Journal of Materials Science 33:2027.

PAGE 242

219 University of South Florida. n.d. US F Gamma Irradiator User Training Handout. University of South Florida O ffice of Radiation and Safety. Van Krevelan, D.W.. 1990. Properties of Polymers: Their Correlation with chemical Structure; their Numerical Estimat ion and Predication from Additive Group Contribution. Amsterdam: Elsevier Scientific Publishing. Wagner, K.W. 1914. Arch. Electrotech 2:371. Wei, C., D. Srivastava, K.J. Cho. 2004. Nano Letters 4:1949. Williams G. And D.C. Watts. 1971. Multiple Dielectric Relaxation Processes in Amorphous Polymers as a function of frequency, temperature and applied pressure. In Dielectric Properties of Polymers. Ed. Frank E. Karasz. New York: Plenum Press. Wong E.W., P.E. Sheehan, C.M. Lieber. 1997 Science. 277:1971. Woodard, A.E. J.A. Sauer, R.A. Wall. 1961. Journal of Polymer Science L:117. Wu, P.K., J. Fitz-Gerald, A. Pique, D.B. Chrisey, and R.A. McGill. 2000. Materials Research Society Symposium 617:J2.3.1. Zhao, Q., J. Wood, and H.D. Wagner. 2001. J. Polymer Sci: Part B: Polym Phys 39:1492. Zhou, Q., Y. Nakamura, S. Inaoka, M. Park, Y. Wang, X. Fan, J. Mays, R. Advincula. 2002. Surface-Initiated Anionic Polymerization: Tethered Polymer Brushes on Silicate Flat Surfaces. In Polymer Nanocomposites – Synthesis, Characterization and Modeling, Ed. Krishnamoorti R., and R. Vaia. ACS Symposium Series, Vol. 804. Washington, DC: American Chemical Society. Zoller, P., H.W. Starweather Jr., and G.A. Jones. 1986. Journal of Polymer Science: Part B: Polymer Physics 24: 1451. Zou, Y., Y. Feng, L. Wang, X. Liu. 2004. Carbon 42:271.

PAGE 243

220 APPENDICES

PAGE 244

221 Appendix A. Chapter 3 Figure A.1. Syndiotactic PMMA.

PAGE 245

222 Appendix A. Con’t Scheme A. 1. Photochemical decom position of 1-phenyl-2-hydroxy-2-methyl-1 propanone Thermal Methods for DSC, DEA, DMA DSC Segment Description Data storage: off Equilibrate at 25.00 C Isothermal for 2.00 min Data storage: on Ramp 10.00 C/min to 145.00 C DEA Segment Description Data storage: off Equilibrate at 200.00 C Isothermal for 3.00 min Data storage: on Isothermal for 2.00 min Frequency sweep Increment -5.00 C Repeat segment 5 until -155.00 C DMA Segment Description Equilibrate at -150.00 C Isothermal for 5.00 min Isothermal for 2.00 min Frequency sweep Increment 5.00 C Repeat segment 3 until 190.00 C O C C OH CH3CH3 O C C OH CH3CH3 + .

PAGE 246

223 APPENDIX B. Chapter 4 Thermal Methods for DSC and DMA DSC Segment Description Data storage: off Equilibrate at 25.00 C Isothermal for 2.00 min Data storage: on Ramp 10.00 C/min to 145.00 C DMA Segment Description Equilibrate at -150.00 C Isothermal for 5.00 min Isothermal for 2.00 min Frequency sweep Increment 5.00 C Repeat segment 3 until 190.00 C

PAGE 247

224 APPENDIX C. Chapter 5. DEA Segment Description Data storage: off Equilibrate at 200.00 C Isothermal for 3.00 min Data storage: on Isothermal for 2.00 min Frequency sweep Increment -5.00 C Repeat segment 5 until -155.00 C

PAGE 248

225 APPENDIX D. Chapter 7. DSC Segment Description Data storage: off Equilibrate at -100.00 C Isothermal for 2.00 min Data storage: on Ramp 5.00 C/min to 300.00 C Mark end of Cycle 1 Equilibrate at -100.00oC Isothermal for 2.00 min Mark end of Cycle 2 Ramp 5.00 C/min to 300.00 C Mark end of Cycle 3 DMA Segment Description Equilibrate at -150.00 C Isothermal for 5.00 min Data Storage: on Isothermal for 2.00 min Frequency sweep Increment 5.00 C Repeat segment 4 until 300.00 C

PAGE 249

About the Author LaNetra Michelle Clayton received her B.S. degree from Florida A & M University in 1998. She officially entered the Ph.D. program in chemistry at the University of South Florida in 2001. LaNetra’s research, under the direction of Dr. Julie P. Harmon, produced three publications at the time her graduation. He r research also generated two invention disclosures. She has co-authored an addi tional seven articles, one other invention disclosure, and participated in seven presentations. Ms Clayton was also awarded a fellowship funded by the National Science Foundation.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681154
003 fts
005 20060215104639.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051012s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001031
035
SFE0001031
040
FHM
c FHM
090
2005
1 100
Clayton, LaNetra.
4 245
The design, fabrication, and characterization of polymer-carbon nanotube composites
h [electronic resource] /
by LaNetra Clayton.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (Ph.D.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 249 pages.
520
ABSTRACT: The design, fabrication, and characterization of polymer-carbon nanotube (CNT) composites have generated a significant amount of attention in the fields of materials science and polymer chemistry. The challenge in fabricating composites that exploit the unique properties of the CNT and the ideal processing ability and low cost of the polymer is in achieving a uniform dispersion of the filler in the polymer matrix. This body of work focuses on (1) techniques employed to disperse CNTs into a polymer matrix and (2) the effects of CNTs on the mechanical and electrical properties of the polymer. Poly (methyl methacrylate) (PMMA), an amorphous polymer, and poly (4-methyl-1-pentene) (P4M1P), a semi crystalline polymer, were chosen as the matrices. Non-functionalized single-walled carbon nanotubes and soot (unpurified carbon nanotubes) were chosen as the filler material.In the first study, single-walled carbon nanotubes (SWNTs) were sonicated in methyl methacrylate monomer and initiated via thermal energy, UV light, and gamma radiation. Composite films with increased dielectric constants and unique optical transparency were produced. Samples were characterized using differential scanning calorimetry, dielectric analysis, and dynamic mechanical analysis. Refractive Indices were obtained and correlated to the dielectric constant using Maxwells relationship. PMMA/soot composites were fabricated in the second study. Dispersion was accomplished by way of sonication and melt compounding. The PMMA/soot composites were exposed to gamma radiation, with a 137Cs gamma source, in order to investigate how the filler affects the polymers ability to resist radiation. Samples were characterized by differential scanning calorimetry, dielectric analysis, and dynamic mechanical.
590
Adviser: Julie P. Harmon.
Co-adviser:
653
Nanotechnology.
Poly(methyl methacrylate).
Poly(4-methyl-1-pentene).
Interfacial polarization.
Dielectric analysis.
0 690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
856
u http://digital.lib.usf.edu/?e14.1031