Citation
Dynamic Antifouling Structures and Actuators Using EAP Composites

Material Information

Title:
Dynamic Antifouling Structures and Actuators Using EAP Composites
Creator:
BOHN, CLAYTON CLAVERIE ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Conducting polymers ( jstor )
Electrodes ( jstor )
Oxidizers ( jstor )
Polymerization ( jstor )
Polymers ( jstor )
Polypyrroles ( jstor )
Pyrroles ( jstor )
Solvents ( jstor )
Strain gauges ( jstor )
Surface roughness ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Clayton Claverie Bohn. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/7/2004
Resource Identifier:
56814116 ( OCLC )

Downloads

This item has the following downloads:

bohn_c ( .pdf )

bohn_c_Page_012.txt

bohn_c_Page_058.txt

bohn_c_Page_018.txt

bohn_c_Page_187.txt

bohn_c_Page_135.txt

bohn_c_Page_078.txt

bohn_c_Page_119.txt

bohn_c_Page_005.txt

bohn_c_Page_155.txt

bohn_c_Page_084.txt

bohn_c_Page_185.txt

bohn_c_Page_039.txt

bohn_c_Page_134.txt

bohn_c_Page_132.txt

bohn_c_Page_014.txt

bohn_c_Page_214.txt

bohn_c_Page_173.txt

bohn_c_Page_147.txt

bohn_c_Page_174.txt

bohn_c_Page_040.txt

bohn_c_Page_050.txt

bohn_c_Page_019.txt

bohn_c_Page_146.txt

bohn_c_Page_094.txt

bohn_c_Page_089.txt

bohn_c_Page_209.txt

bohn_c_Page_153.txt

bohn_c_Page_203.txt

bohn_c_Page_154.txt

bohn_c_Page_002.txt

bohn_c_Page_021.txt

bohn_c_Page_128.txt

bohn_c_Page_034.txt

bohn_c_Page_255.txt

bohn_c_Page_180.txt

bohn_c_Page_198.txt

bohn_c_Page_072.txt

bohn_c_Page_268.txt

bohn_c_Page_243.txt

bohn_c_Page_103.txt

bohn_c_Page_247.txt

bohn_c_Page_022.txt

bohn_c_Page_242.txt

bohn_c_Page_169.txt

bohn_c_Page_108.txt

bohn_c_Page_161.txt

bohn_c_Page_259.txt

bohn_c_Page_149.txt

bohn_c_Page_215.txt

bohn_c_Page_258.txt

bohn_c_Page_055.txt

bohn_c_Page_122.txt

bohn_c_Page_178.txt

bohn_c_Page_171.txt

bohn_c_Page_248.txt

bohn_c_Page_263.txt

bohn_c_Page_160.txt

bohn_c_Page_143.txt

bohn_c_Page_073.txt

bohn_c_Page_190.txt

bohn_c_Page_111.txt

bohn_c_Page_197.txt

bohn_c_Page_227.txt

bohn_c_Page_141.txt

bohn_c_Page_254.txt

bohn_c_Page_229.txt

bohn_c_Page_138.txt

bohn_c_Page_006.txt

bohn_c_Page_074.txt

bohn_c_Page_096.txt

bohn_c_Page_186.txt

bohn_c_Page_152.txt

bohn_c_Page_127.txt

bohn_c_Page_223.txt

bohn_c_Page_199.txt

bohn_c_Page_165.txt

bohn_c_Page_109.txt

bohn_c_Page_008.txt

bohn_c_Page_240.txt

bohn_c_Page_063.txt

bohn_c_Page_170.txt

bohn_c_Page_083.txt

bohn_c_Page_097.txt

bohn_c_Page_256.txt

bohn_c_Page_211.txt

bohn_c_Page_106.txt

bohn_c_Page_217.txt

bohn_c_Page_204.txt

bohn_c_Page_090.txt

bohn_c_Page_020.txt

bohn_c_Page_079.txt

bohn_c_Page_265.txt

bohn_c_Page_184.txt

bohn_c_Page_100.txt

bohn_c_Page_066.txt

bohn_c_Page_044.txt

bohn_c_Page_003.txt

bohn_c_Page_116.txt

bohn_c_Page_201.txt

bohn_c_Page_151.txt

bohn_c_Page_085.txt

bohn_c_Page_142.txt

bohn_c_Page_236.txt

bohn_c_Page_125.txt

bohn_c_Page_071.txt

bohn_c_Page_172.txt

bohn_c_Page_267.txt

bohn_c_Page_060.txt

bohn_c_Page_036.txt

bohn_c_Page_139.txt

bohn_c_Page_028.txt

bohn_c_Page_241.txt

bohn_c_Page_232.txt

bohn_c_Page_156.txt

bohn_c_Page_010.txt

bohn_c_Page_177.txt

bohn_c_Page_069.txt

bohn_c_Page_208.txt

bohn_c_Page_166.txt

bohn_c_Page_196.txt

bohn_c_Page_017.txt

bohn_c_Page_148.txt

bohn_c_Page_087.txt

bohn_c_Page_047.txt

bohn_c_Page_107.txt

bohn_c_Page_048.txt

bohn_c_Page_261.txt

bohn_c_Page_086.txt

bohn_c_pdf.txt

bohn_c_Page_054.txt

bohn_c_Page_009.txt

bohn_c_Page_114.txt

bohn_c_Page_264.txt

bohn_c_Page_075.txt

bohn_c_Page_140.txt

bohn_c_Page_253.txt

bohn_c_Page_011.txt

bohn_c_Page_046.txt

bohn_c_Page_249.txt

bohn_c_Page_238.txt

bohn_c_Page_027.txt

bohn_c_Page_158.txt

bohn_c_Page_067.txt

bohn_c_Page_081.txt

bohn_c_Page_092.txt

bohn_c_Page_252.txt

bohn_c_Page_064.txt

bohn_c_Page_228.txt

bohn_c_Page_093.txt

bohn_c_Page_004.txt

bohn_c_Page_150.txt

bohn_c_Page_023.txt

bohn_c_Page_193.txt

bohn_c_Page_205.txt

bohn_c_Page_131.txt

bohn_c_Page_235.txt

bohn_c_Page_068.txt

bohn_c_Page_033.txt

bohn_c_Page_251.txt

bohn_c_Page_099.txt

bohn_c_Page_206.txt

bohn_c_Page_195.txt

bohn_c_Page_029.txt

bohn_c_Page_117.txt

bohn_c_Page_088.txt

bohn_c_Page_167.txt

bohn_c_Page_231.txt

bohn_c_Page_220.txt

bohn_c_Page_176.txt

bohn_c_Page_237.txt

bohn_c_Page_082.txt

bohn_c_Page_104.txt

bohn_c_Page_221.txt

bohn_c_Page_051.txt

bohn_c_Page_112.txt

bohn_c_Page_200.txt

bohn_c_Page_102.txt

bohn_c_Page_179.txt

bohn_c_Page_130.txt

bohn_c_Page_210.txt

bohn_c_Page_212.txt

bohn_c_Page_057.txt

bohn_c_Page_257.txt

bohn_c_Page_222.txt

bohn_c_Page_043.txt

bohn_c_Page_091.txt

bohn_c_Page_245.txt

bohn_c_Page_110.txt

bohn_c_Page_025.txt

bohn_c_Page_213.txt

bohn_c_Page_262.txt

bohn_c_Page_076.txt

bohn_c_Page_194.txt

bohn_c_Page_230.txt

bohn_c_Page_105.txt

bohn_c_Page_182.txt

bohn_c_Page_016.txt

bohn_c_Page_032.txt

bohn_c_Page_126.txt

bohn_c_Page_145.txt

bohn_c_Page_159.txt

bohn_c_Page_183.txt

bohn_c_Page_041.txt

bohn_c_Page_080.txt

bohn_c_Page_168.txt

bohn_c_Page_052.txt

bohn_c_Page_031.txt

bohn_c_Page_049.txt

bohn_c_Page_124.txt

bohn_c_Page_007.txt

bohn_c_Page_113.txt

bohn_c_Page_042.txt

bohn_c_Page_098.txt

bohn_c_Page_030.txt

bohn_c_Page_233.txt

bohn_c_Page_077.txt

bohn_c_Page_024.txt

bohn_c_Page_234.txt

bohn_c_Page_163.txt

bohn_c_Page_065.txt

bohn_c_Page_164.txt

bohn_c_Page_189.txt

bohn_c_Page_123.txt

bohn_c_Page_045.txt

bohn_c_Page_118.txt

bohn_c_Page_059.txt

bohn_c_Page_246.txt

bohn_c_Page_070.txt

bohn_c_Page_181.txt

bohn_c_Page_026.txt

bohn_c_Page_157.txt

bohn_c_Page_115.txt

bohn_c_Page_192.txt

bohn_c_Page_188.txt

bohn_c_Page_226.txt

bohn_c_Page_175.txt

bohn_c_Page_038.txt

bohn_c_Page_260.txt

bohn_c_Page_001.txt

bohn_c_Page_216.txt

bohn_c_Page_035.txt

bohn_c_Page_202.txt

bohn_c_Page_062.txt

bohn_c_Page_120.txt

bohn_c_Page_191.txt

bohn_c_Page_266.txt

bohn_c_Page_219.txt

bohn_c_Page_250.txt

bohn_c_Page_244.txt

bohn_c_Page_053.txt

bohn_c_Page_056.txt

bohn_c_Page_162.txt

bohn_c_Page_137.txt

bohn_c_Page_207.txt

bohn_c_Page_121.txt

bohn_c_Page_129.txt

bohn_c_Page_015.txt

bohn_c_Page_101.txt

bohn_c_Page_144.txt

bohn_c_Page_133.txt

bohn_c_Page_225.txt

bohn_c_Page_037.txt

bohn_c_Page_224.txt

bohn_c_Page_095.txt

bohn_c_Page_013.txt

bohn_c_Page_218.txt

bohn_c_Page_239.txt

bohn_c_Page_061.txt

bohn_c_Page_136.txt


Full Text












DYNAMIC ANTIFOULING STRUCTURES AND ACTUATORS USING EAP
COMPOSITES













By

CLAYTON CLAVERIE BOHN JR.


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Clayton Claverie Bohn Jr.
































This dissertation is dedicated to my loving parents Jack and Kathy Higgins and Clay and
Susan Bohn for all there guidance and support throughout my life.















ACKNOWLEDGMENTS

I would like to thank my advisor and committee chair, Dr. Anthony Brennan, for

providing me with the opportunity to further develop my understanding of polymers and

engineering as a science and an art. He has provided infinite patience, guidance, and

support throughout this project. He has continually challenged me as a researcher and a

man to grow and find solutions that are not always obvious. I would like to thank the rest

of my committee, Dr. Ronald Baney, Dr. Chris Batich, Dr. Amelia Dempere, and Dr.

John Reynolds, for their time and valued critique of this dissertation. In addition I would

like to further thank Dr. Reynolds for his support, guidance, and extensive knowledge of

conducting polymers and electrochemistry during this project. I would also like to thank

him for providing me the use of his electrochemical equipment and the knowledge of his

graduate students. I would especially like to thank my coworkers on this project, Dr.

Said Sadki and Dr. Myoungho Pyo, for being there as friends, coworkers, and teachers.

Dr. Sadki provided the critical knowledge of electrochemistry necessary to get this

project started. Dr. Pyo provided the exceptional skill and knowledge of the subject

necessary to develop the EcAu systems and to get the project to where it is now. To them

I owe all my knowledge of conducting polymers and electrochemistry, and without their

help and guidance throughout this project none of this would have been possible. I would

like to thank Dr. Elisabeth Smela for her continued guidance and support on the

development of the EcAu systems. I would also like to thank Dr. Peter Ifju for help on

developing the strain sensitive actuator technology. Dr. Ifju provided extensive









knowledge and support on the proper use and application of strain gages and initial strain

gage equipment usage.

I would also like to especially thank my past and present group members. They

have provided me with invaluable knowledge, assistance, and friendship during my years

in the Brennan research group. Jeanne McDonald and Jeremy Mehlem provided

exceptional help and guidance during my initial year in the research group. Both of them

have been invaluable as coworkers and friends. I would like to thank them and their

spouses for providing me with a very enlightening and enjoyable start in this group. I

would like to give special thanks to Michelle Carmen, Thomas Estes, Adam Feinberg,

Amy Gibson, Brian Hatcher, Nikhil Kothurkar, Chuck Seegert, Jim Schumacher, Leslie

Wilson, and Wade Wilkerson who have proven to be exceptional friends and coworkers.

I would also like to thank the rest of my research group for their support: Kenneth

Williams and Kiran Karve. I would also like to thank the rest of my friends and

colleagues in the department that have provided me with support: Brett Almond, Brian

Cuevas, Iris Enriquez, Brent Gila, Josh Stopek, Dan Urbaniak, and Amanda York. Out of

this group I would especially like to thank Jamie Rhodes and Paul Martin for also

providing me with spare parts and expertise needed to help build and keep equipment

running. I would also like to especially thank Jennifer Wrighton; with out her this whole

place would probable fall apart and no work would ever get done. She had been a good

friend and a great secretary to me and the rest of the group for the past 4 years.

I would also like to thank my family and friends for their love and support

throughout my life. I would especially like to thank my brother David for always being

there and providing a comical and uplifting side to life. I wish him luck in fatherhood









and all that he does. And I would again like to thank him for always being a true brother

and friend.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ......... .................................................................................... iv

LIST OF TABLES ...................................... ...................... ............xii

LIST OF FIGURES ............. ........................ ................. .... .... .........xiii

ABSTRACT ........ .............. ............ .. ...... ........... .......... xxiv

1 INTRODUCTION ............... ..................................................... 1

2 B A C K G R O U N D ..................... .... ................................ ........ ........ .......... .. ....

2.1 Electroactive Polym er Actuators ................................ ......................... ........ 6
2.1.1 Introduction to EAPs ............................................................................. 6
2.1.1 C conducting Polym ers ........................................ ........................... 10
2.1.2 Conducting Polymer Synthesis ....................................... ...............11
2.1.3 Conducting Polym er Actuators ...................................... ............... 13
2.2 Electrical R resistance Strain G ages ............................................ ............... 16
2 .2 .1 Strain G age T heory ................. .............................. ............... ... 18
2.2.2 Strain Gage Materials and Construction................... ...............20
2.2.3 Strain G age A accuracy ..................................................................... ..... 23
2.3 A nti-Fouling/Foul-Release Coatings................................................................. 25
2.4 Electrow etting........ .. ................................... .. .. ........ ................. 30
2.5 Dynam ic Surfaces ........ ..... ........................ ....... .... .......... .............. .. 33
2 .5 .1 P o ly p y rro le ................................................. ................ 3 4
2.5.2 Poly(3-m ethylthiophene) .................................... .......................... ........ 35
2 .5 .3 P oly (p -ph eny len e) ........................................................... .....................36

3 INITIAL IN SITUEVALUATION OF CP'S VIA STRAIN GAGE TECHNIQUE .38

3 .1 In tro du ctio n ...................................... ............................ ................ 3 8
3 .2 M materials and M ethods .............................................................. .....................38
3.2.1 M materials .............................................................................. 38
3.2.2 Strain Gages............................................ ........ 39
3.2.3 Conducting Polym er Synthesis........................................ ............... 39
3.3 Polypyrrole (PPy/TO S) R esults..........................................................................40
3.3.1 Determination of Electropolymerization Conditions for PPy/TOS.............40
3.3.2 PPy/TOS Cyclic Voltammetry and Strain Response..............................41









3.3.3 PPy/TOS Multi-Cycle Strain Response..........................................43
3.3.4 PPy/TOS Square-Wave Potential Experiments............... ...................44
3.4 PED OP Results ............................................ .. .... .................. 46
3.4.1 Electrochemical Analysis of PEDOP ............... ............................... 46
3.4.2 PEDOP M ulti-Cycle Strain Response................................ ... ..................46
3.4.3 Effects of Cyclic Scan Rate on the Strain Response of PEDOP ..............49
3.4.4 PEDOP Square-Wave Potential Experiments ........................................51
3.5 PB ED O T-C z R results .................................................. .............................. 52
3.5.1 PBEDOT-Cz Introduction ....................................................................52
3.5.2 PBEDOT-Cz Electrochemical Conditions ...........................................53
3.5.3 PBEDOT-Cz Strain Response (-0.8 V to 0.6 V) .......................... 54
3.5.4 PBEDOT-Cz Strain Response (-0.8 V to 1.0 V) .......................... 55
3.5.5 PBEDOT-Cz Strain Response (-0.8 V to 1.2 V) .......................... 55
3.5.6 Overall Results for PBEDOT-Cz ................................ ............... 56
3.6 Overall Comparison of PPy, PEDOP, and PBEDOT-Cz ...............................58
3.7 Effects of Interlayer Adhesion ................... ........................... ..................59
3 .8 C o n clu sio n s.................................................. ................ 6 1

4 INSITU STRAIN MEASUREMENTS OF CP'S ON ENHANCED AU SURFACES66

4 .1 Intro du action ...................................... ............................ ................ 6 6
4 .2 M materials and M ethods .............................................................. .....................66
4.2.1 M materials ................... ..................... ......................... 66
4.2.2 Electrochemical Gold Deposition Solution ................................................67
4.2.3 Evaporated and Electrochemically Deposited Gold................................67
4.2.4 Conducting Polymer Synthesis...................... ........... ..............68
4.3 Improved Interlayer Adhesion Utilizing Electrochemically Deposited Au
Surfaces (EcA u) ...................................... ..... ............... .... .. ........ .... 68
4.4 Effects of Surface Roughness on EcAu Morphology...............................70
4.5 Electrochemical Deposition of PPy ............................................... ...............77
4.6 Improved PPy Adhesion to EcAu Treated Surfaces...........................................82
4.7 IN SITU PPy/EvAu/PI Actuator Results............... ....... ............ ........83
4.7.1 Counter Ion Effects on PPy Strain Response ............................................83
4.7.2 Effects of Potential Limiting on PPy Strain Response ..............................85
4.7.3 Effects of PPy Film Thickness on Strain Response ...................................86
4.7.4 Effects of Polymerization Potential on PPy Strain Response ....................90
4.8 IN SITU PPy/EcAu/EvAu/PI Actuator Results .................................. ...............91
4.8.1 Effects of PPy electropolymerization charge and surface roughness factor92
4.8.1.1 Effects of surface roughness factor .............. .... ......... ............... 92
4.8.1.2 Effects of PPy electropolymerization charge on strain (PPy film
thickness) .......................... ............................ ... ... ............. 95
4.8.1.3 Combined effects of PPy electropolymerization charge and surface
roughness factor ............................... ...... ........ ..... ................ .. 97
4.8.2 Counter Ion Effects on PPy Strain Response ..........................................98
4.8.3 Frequency response ............................................................................. 99
4 .8 .4 L ifetim e .................................................................. 10 0
4 .9 C on clu sion s................................................................................. ............. 103









5 RAPID ELECTRODE PATTERNING FOR USE IN ADVANCED CONDUCTING
POLYER ACTUATORS AND DYNAMIC SURFACES ....................................104

5 .1 Intro du action ............. .. ... ... ............. ................................................ 104
5.2 Rapid Electrode Patterning Techniques.....................................105
5.2.1 Rapid Electrode Patterning................ ........................ ............... 105
5.2.2 Line Patterning ................... ................ ........................... 106
5.3 L near A ctuators ....................... .. ........................ .. ...... .... ...........108
5.3.1 B background ....................................................... ..... .. ............. 108
5.3.2 Cantilever based linear actuators......................... ................ 109
5.4 Actuator Based Dynam ic Surfaces ............. ............. ................ ............... 119
5 .5 C o n clu sio n s................................................. ................ 12 5

6 PDMSe BASED DYNAMIC NON-TOXIC ANTI-FOULING SURFACE
COATINGS ..................................... ................................ ........... 127

6.1 Introduction.................................................... ........................... ....... 127
6.2 M materials and M ethods .............................................. ............................. 129
6.2.1 M materials .................................... ....................... ..... ......... 129
6.2.2 G elatin Preparation .............................................................. ... ............ 129
6.2.3 Polydimethylsiloxane Elastomer Preparation....................................130
6.2.4 Soluble Polypyrrole Preparation ............... .......... ... .... ........... 130
6.2.5 Chemical Formation of Conducting Polymer IPN Systems.....................130
6.2.6 Supercritical CO2 Solution Formation of Conducting Polymer IPN
System s .................... .............................................. ........ .................131
6.2.7 Electrochemical Formation of Conducting Polymer IPN Systems ..........132
6.2.8 Mechanical Property Testing of PPy/PDMSe IPN systems..................... 133
6.2.9 A TR -FTIR A analysis ....................................................... .... ........... 133
6.2.10 O ptical M icroscopy ........................................ .......................... 134
6.2.11 Electrochemical Analysis ............. ................................ ............... 134
6.2.12 ED S M apping .......... .. ................ ..... ..... .... .... ........ .... 134
6.3 B background Study ................... .... ... .................. .. .......... .............. .. 135
6.3.1 Electrochemical formation of Conducting Polymer IPN Systems...........136
6.3.2 Chemical formation of Conducting Polymer IPN Systems......................138
6.3.3 Conducting Polymer Blend Formation by Solution Blending ...............140
6.4 Polypyrrole/PDMSe IPN Formation.......................................................141
6 .4 .1 Introdu action ............................... ..................................14 1
6.4.2 Supercritical Carbon Dioxide Solution Doping .......................................142
6.4.2.1 Introduction ..... ... .. ............... .. ...... ................ ... 142
6.4.2.2 EtOH/scCO2 prepared IPNs .......................... .... .................143
6.4.2.3 IPA /scCO 2 prepared IPN s .............................................................147
6.4.3 Tetrahydrofuran Solution Doping .................................... ............... 154
6.4.3.1 Effects of hydration of sample conductivity ............................... 156
6.4.3.2 Effects of PDMSe surface segregation....................................158
6.4.3.3 Effects of FeC13/THF doping on PDMSe mechanical properties ..162
6 .5 C on clu sion s................................................ .................. 164









7 TP AND TPE BASED DYNAMIC, NON-TOXIC, ANTI-FOULING SURFACE
C O A T IN G S ...................................... .............................................. 16 7

7.1 IN TR O D U C TIO N ...................................................... ................................. 167
7.2 M ATERIALS AND M ETHODS ............................................ ............... 169
7.2.1 M materials .................. ............... .. ....................... ..... ......... 169
7.2.1.1 G general C hem icals ....................... .. ....... ..... ............... ... 169
7.2.1.2 PPy/Santoprene Sample Preparation and Mounting ..................169
7.2.1.3 Polydimethylsiloxane Elastomer Preparation .............................170
7.2.1.4 ATS Coupling Agent Preparation .............. ................................. 170
7.2.2 Sam ple Preparation M ethods................................................................ 170
7.2.2.1 Sam ple M counting ................................................. .. ............. 171
7.2.2.6 M icropatterning ............................................ 173
7.2.2.7 TP Film Formation by Spin Casting and Spraying ......................173
7.2.2.8 Conductivity D eterm ination................................... ... ..................174
7.3 POLYPYRROLE/8281-65 IPN SYSTEMS..................... .................1.75
7.3.1 PPy/Santoprene 8281-65 Contact Angle Measurements........................175
7.3.2 Determination of PPy Content in PPy/8281-65 Samples......................176
7.4 POLYPYRROLE/271-55 IPN SYSTEMS.................... .. ................. 178
7.4.1 Determination of PPy Content in PPy/271-55 Samples.........................178
7.4.2 Effects of Counter Ion Exchange on PPy/271-55 Conductivity.............180
7.4.3 Formation of poly(3-methylthiophene)/271-55 IPN systems.................83
7.4.4 M icropatterning of 271-55 ... ......... ................... ..................................185
7.5 EDS and SEM ANALYSIS of SANTOPRENE' TPE's ...................................187
7.6 PPy/POLYSULFONE SYSTEMS ............................ ..................... 194
7.8 C O N C L U SIO N S ....................................................................... ...................196

8 DYNAMIC MODULUS MAPPING OF PPY/SANTOPRENE BLENDS ...........198

8.1 IN T R O D U C T IO N ................................................................... ..................... 198
8.2 M ATERIALS AND M ETHODS ............................................ ............... 199
8.2 .1 M materials ...................................... ..... .... ......................... 199
8.2.2 Sample Preparation M ethods .............. ........................................ 199
8.2.2.1 PPy/TPE (8211-65) Sample Preparation............... ..............199
8.2.2.2 Sample M counting ............ ..... ....... ... ...... .... ............. 199
8.2.2.3 Electrochemical Polymerization of Pyrrole ................................200
8.2.3 Hysitron Nano-DM A........ ......... .. ......... .................. .... 200
8.2.3.1 Hysitron Introduction .............. .. ...... ................. 200
8.2.3.2 Experimental conditions.......................................... 202
8.3 HYSITRON NANO-DMA and AFM ANALYSIS................. ... .................204
8.3.1 Nano-DMA mapping of 271-55 and PPy/271-55 samples...................204
8.3.2 Dynamic fluid cell nano-DMA mapping ofPPy/8211-65 ..................207
8 .4 C O N C L U SIO N S ..................................................................... .....................2 15

9 CONCLUSIONS AND FUTURE WORK.................... ............. ........... 216

A B B R E V IA T IO N S .............................................................................. .....................22 7









L IST O F R EFE R EN C E S ........................................................................... ........ .......... 229

B IO G R A PH IC A L SK E T C H ........................................... ...........................................243















LIST OF TABLES

Table p

7.1 Contact angle values obtain for PPy/8281-65 dual element sample in Instant Ocean
artificial sea water versus a Ag/AgCl reference electrode ...................................175

7.2 Weight and volume change data for the formation of PPy/8281-65 IPN systems using
the FeC13/THF solvent soak process................................. ........................ 177















LIST OF FIGURES


Figure p

2.1 Bipolar unit charge migration as depicted for polythiophene..............................10

2.2 Example of one of the possible mechanisms for the oxidative polymerization of
polypyrrole. .......................................... ............................ 12

2.3 Examples of basic bending (cantilever) conducting polymer actuators; A) bilayer
actuator, B) backbone type actuator, C) shell type actuator............................... 15

2.4 Diagram of a polyimide based electrical resistance strain gage and EAP actuator
setu p ...............................................................................18

2.5 Thermally induced apparent strain for Constantan (Advanced), Isoelastic, and
K arm a alloy s. ...................................................... ................. 22

2.6 Diagram showing the different species breakdown and layering of biofoul film
form action .............................................................................26

2.7 Design of electrowetting device: (a) no applied electrical potential (hydrophobic
surface); (b) with applied electrical potential hydrophilicc surface). Fluid is
pumped by continuously cycling the applied electrical potential ..........................30

2.8 Voltage required to induce a AO of 400 (from 120-80) versus dielectric layer
thickness for Teflon AF based EWOD device, with P = 2.0 and Ebreakdown = 2x1016
V /cm ..................................................................33

2.9 Relative surface charge of different conducting polymers. ....................................34

2.10 Oxidation (-) and reduction (--) potentials of poly pyrrole (PPy), polyaniline (PA),
poly(3-methylthiophene (PMeT), and poly(p-phenylene) (PPP). ............................36

2.11 Monomer and polymer structures for A) polypyrrole, B) poly(3-methylthiophene),
and C) poly(p-phenylene)......................................................... 37

3.1 SEM micrograph of surface morphology of PPy/TOS film prepared in 1.0 M
LiC104 at a potential of 0.65 V. SEM image taken of an uncoated sample at 1000X
and 15 K eV ....................................................................... ... ....... ....... 41









3.2 (a) PPy/TOS cyclic voltammetry (v = 10 mV/s) and (b) in situ strain response of a
9.6 [tm film prepared in aqueous 1.0 M LiC104...............................43

3.3 In situ multi-cycle cyclic voltammetry strain response of a 9.6 ,s film prepared in
aqueous 1.0 M LiC10 4 ............................................. .... .. ... .. ........ .... 44

3.4 In situ square-wave strain response of a 9.6 [m film prepared in aqueous 1.0 M
L iC 10 4. ..................................................................................45

3.5 Cyclic Voltammetry (v = 100 mV/s) of a PEDOP film produced from aqueous 1.0
M L iC 10 4 at E = 0.5 V and t = 200 s............................................. .....................47

3.6 SEM micrograph of a PEDOP film prepared at 0.6 V in 1.0 M LiC104. SEM
image taken of an uncoated sample at 1000X and 15 KeV. ................................48

3.7 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of a 10.6 [tm
PEDOP film in aqueous 1.0 M LiC104. ........................................ ...............48

3.8 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of 10.6 [tm
PEDOP film in aqueous 1.0 M LiC104. Data from figure 3.7 has been replotted vs.
tim e ........................................................................................ .4 9

3.9 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of a 10.6 [tm
PEDOP film in aqueous 1.0 M LiC104. ........................................ ...............50

3.10 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of 10.6 [tm
PEDOP film in aqueous 1.0 M LiC104 replotted vs. time. ............................50

3.11 In situ square-wave strain response of a 10.6 Es film prepared in aqueous 1.0 M
L iC 10 4. ..................................................................................5 1

3.12 Cyclic voltammetry (100 mV/s) of PBEDOT-Cz in 0.1 M TBAP/CAN................53

3.13 Normalized in situ cyclic strain response of 9.9 |jm PBEDOT-Cz film (-0.8 V to
0.6 V) for the 1st, 2n, 5th, and 10th scans in 1.0 M LiC104......................................54

3.14 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film (-0.8 V to
1.0 V) for the 1t, 2nd, 5th, and 10th scans in 1.0 M LiC104 ......................................55

3.15 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film (-0.8 V to
1.2 V) for the 1st, 2nd, 5th, and 10th scans in 1.0 M LiC104 ......................................56

3.16 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film scanned
from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (5th scan) in 1.0 M LiC104 ........................ 57

3.17 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film scanned
from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (10th scan) in 1.0 M LiC104 ......................58









3.18 Comparison of in situ cyclic strain response of PPy, PEDOT, and PBEDOT-Cz in
aqueous 1.0 M LiC104 ........ .. ..... ............ ...... ............. .. .. .... ... ...... .. 59

3.19 SEM micrograph of delamination of PPy resulting from long-term cycling of the
actuator; 150X .................................... ............................... ..........62

3.20 SEM micrograph of delamination of PPy resulting from exposure to high vacuum;
2 5 X ........................................................................................6 2

3.21 Enlarged SEM micrograph of region "A" in Figure 4.20; 1350X ...........................63

3.22 SEM micrograph of porous PPy at PPy-Au interface; 5000X.............................63

3.23 Enlarged SEM micrograph of region "B" in Figure 4.20; 100X ............................64

3.24 SEM micrograph of PPy nodules remaining of Au substrate after PPy delamination;
1000X ...................................................................................64

3.25 SEM micrograph of PPy surface after delamination from Au substrate during long-
term repetitive cycling of the actuator; 250X ......................................................65

3.26 SEM micrograph of exposed PPy-Au interface exhibiting PPy nodule growth;
1000X ...................................................................................6 5

4.1 Correlation between surface roughness factor, nominal EcAu thickness and EcAu
deposition charge. EcAu thickness was determined by cross-section SEM .........70

4.2 SEM micrograph of EvAu deposited on smooth polyimide (PI); 4000X ..............72

4.3 SEM micrograph of 3 minute EcAu deposition on smooth EvAu/PI; 4000X .........72

4.4 SEM micrograph of 10 minute EcAu deposition on smooth EvAu/PI; 4000X .......73

4.5 SEM micrograph of 30 minute EcAu deposition on smooth EvAu/PI; 4000X .......73

4.6 SEM micrograph of 60 minute EcAu deposition on smooth EvAu/PI; 4000X .......74

4.7 SEM micrograph of EvAu (r = 2.89) deposited on rough PI strain gage; 4000X....74

4.8 SEM micrograph of EcAu (r = 6.17, 2.5 min.) deposited on rough EvAu/PI strain
g ag e ; 4 0 0 0 X ....................................................... ................ 7 5

4.9 SEM micrograph of EcAu (r = 10.04, 10 min.) deposited on rough EvAu/PI strain
g ag e ; 4 0 0 0 X ....................................................... ................ 7 5

4.10 SEM micrograph of EcAu (r = 18.90, 30 min.) deposited on rough EvAu/PI strain
g ag e ; 4 0 0 0 X ....................................................... ................ 7 6









4.11 SEM micrograph ofEcAu (r = 24.50, 60 min.) deposited on rough EvAu/PI strain
g ag e ; 4 0 0 0 X ....................................................... ................ 7 6

4.12 SEM micrograph of PPy (18.9 C/cm2) deposited on smooth EvAu/PI; 4000X.......78

4.13 SEM micrograph of PPy (18.9 C/cm2) deposited on smooth EvAu/PI treated with
EcAu for 10 m in.; 4000X ..................................................................... 78

4.14 SEM micrograph of PPy (18.9 C/cm2) deposited on smooth EvAu/PI treated with
EcAu for 30 m in.; 4000X ..................................................................... 79

4.15 SEM micrograph of PPy (18.9 C/cm2) deposited on smooth EvAu/PI treated with
EcAu for 60 m in.; 4000X ..................................................................... 79

4.16 Cross-section SEM micrograph of EcAu (60 min.) deposited on smooth EvAu/PI;
4 000X ................................................................................... 8 1

4.17 Cross-section SEM micrograph of PPy (18.9 C/cm2) deposited on smooth EvAu/PI
treated with EcAu for 60 min.; 4000X............................................. ...............81

4.18 Surface plot of calculated PPy film thickness as a function of surface roughness
factor (r) and electropolymerization charge (C/cm2) ............................................ 82

4.19 Strain response of a 2.83 C/cm2 PPy/EvAu/PI actuator during potential cycling at 5
mV/s in aqueous NaC104, LiC104, CsC104, NaNO3, and NaCl solutions...............84

4.20 Strain response of a 2.83 C/cm2 PPy/EvAu/PI actuator during potential stepping
(100 s/step) in NaC104 between -0.6 V and (a) 0.1 V, (b) 0.2 V, (c) 0.3 V, (d) 0.4
V (e) 0.5 V and (f) 0.6 V ............................................... ...................... ......... 86

4.21 In situ strain response of PPy/EvAu/PI actuators of varying PPy film thickness
during potential stepping between -0.6 V and 0.4 V in aqueous NaC104.
Electropolymerization charge densities were 0.79 C/cm2 (2.8 |tm, calculated), 1.6
C/cm2 (5.7 |tm, cal.), 2.4 C/cm2 (8.6 |tm, cal.), 3.2 C/cm2 (11.4 |tm, cal.), 4.0
C/cm2 (14.3 |tm, cal.), and 4.8 C/cm2 (17.1 [tm, cal.). ........................................ 88

4.22 In situ charge response of PPy/EvAu/PI actuators of varying PPy film thickness
during potential stepping between -0.6 V and 0.4 V in aqueous NaC104.
Electropolymerization charge densities were 0.79 C/cm2 (2.8 |tm, calculated), 1.6
C/cm2 (5.7 |tm, cal.), 2.4 C/cm2 (8.6 [tm, cal.), 3.2 C/cm2 (11.4 [tm, cal.), 4.0 C/cm2
(14.3 |tm cal.), and 4.8 C/cm 2 (17.1 |tm cal.)....................................... ............ 89

4.23 In situ strain response of a 2.83 C/cm2 PPy/EvAu/PI actuator during potential
stepping between -0.6 V and 0.4 V in aqueous NaC104. PPy was prepared
potentiostatically at (a) 0.7 V, (b) 0.8 V, (c) 0.9 V, and (d) 1.0 V.........................91









4.24 In situ strain response of a 1.18 C/cm2 PPy/EcAu/EvAu/PI actuator during potential
stepping between -0.6 V and 0.4 V in aqueous NaC104. With surface roughnesses
factors of(a) 2.89, (b) 6.17, (c) 7.13, (d) 10.04, (e) 18.90, and (f) 24.50. ...............93

4.25 Effects of surface roughness factor (r) on the overall change in strain (As) of a 1.18
C/cm2 PPy/EcAu/EvAu/PI actuator during potential stepping between -0.6 V and
0.4 V in aqueous N aC10 4. ............................................ .............................. 94

4.26 Effects of surface roughness factor (r) on the strain rate (.il/sec) of a 1.18 C/cm2
PPy/EcAu/EvAu/PI actuator during potential stepping between -0.6 V and 0.4 V in
aqueous NaC104. ........................... ........ .... .. .... .......... .... 95

4.27 In situ strain response of PPy/EcAu/EvAu/PI actuator during potential stepping
between -0.6 V and 0.4 V in aqueous NaC104. With surface roughnesses factors of
18.90 and PPy electropolymerization charge of (a) 0.79 C/cm2, (b) 1.6 C/cm2, (c)
2.4 C/cm2, (d) 3.2 C/cm2, (e) 4.0 C/cm2, and (f) 4.8 C/cm2. ................................96

4.28 Maximum change in strain response of PPy/EcAu/EvAu/PI actuators during
potential stepping between -0.6 V and 0.4 V in aqueous NaC104 with increasing
electropolymerization charge. Measured on surface roughnesses of (a) r = 2.89
(EvAu), and EcAu samples ofr = (b) 6.17, (c) 10.00, (d) 18.90, and (e) 24.50 .....97

4.29 Surface plot of the normalized overall strain response of PPy as a function of
electropolymerization charge and surface roughness factor (r). ...........................98

4.30 In situ Strain response of a 8.0 C/cm2 PPy/EcAu/EvAu/PI actuator during potential
cycling at 5 mV/s in aqueous (a) LiC104, (b) NaNO3, and (c) NaCl solutions........99

4.31 In situ normalized (relative to 0.01 Hz) strain and charge response of 1.18 C/cm2
PPy as a function of frequency on (a) r = 2.89 (EvAu) and (b) r = 10.0 (EcAu)
surfaces. Inset shows the charge efficiency (tE/C) as a function of frequency. ...100

4.32 In situ normalized strain and charge response of 1.18 C/cm2 PPy (inadequate Argon
purge) on r = 5.1 (EvAu) and r = 22.1 (EcAu) treated actuators with repeated
potential stepping between -0.6 V (10 s) and 0.4 V (20 s) ..............................102

4.33 In situ normalized strain and charge response of 1.18 C/cm2 PPy on r = 3.43
(EvAu), r = 8.26 (EcAu), and r = 18.90 (EcAu) treated actuators with repeated
potential stepping between -0.6 V (10 s) and 0.4 V (20 s) ..............................102

5.1 Negative patterns of linear actuator produced using Microsoft Powerpoint (A) and
printed on Kapton used to fabricate linear actuators (B) and a negative pattern used
for dynam ic surfaces (C ) ............................................................. ................. ...105

5.2 Rapid electrode patterning process for the development of conducting polymer
devices................................ ... ................. ............... .......... 106









5.3 Diagram of basic linear actuator design and AutoCAD model used to predict the
overall developed linear strain for the device ................................ .. ...............110

5.4 Initial PPy linear actuator based on 10mm segments deposited on copper foil.
Picture of oxidized state enhanced to show placement of PPy. ...........................111

5.5 PPy linear actuator based on 10mm segments deposited on EvAu coated polyimide
strip ...................................... ................................................... . 1 12

5.6 PPy linear actuator based on 20mm segments deposited on EvAu coated polyimide
strip ...................................... ................................................... . 1 12

5.7 Single sided rapid electrode pattern design and illustration of how it works. The
green and blue areas represent the separate electrode patterns (working and counter
e le c tro d e ) ................................................................................................ .... 1 1 3

5.8 Diagram of linear actuator fabrication process. Initial negative CAD electrode
design (A) negative electrode design printed on substrate (B) produced patterned
gold electrode from the mask (C) electrochemically deposited conducting polymer
on the patterned gold electrode (D ) ..................................... .......... ............... 114

5.9 Pictures of linear actuator fabrication process. CAD design of negative electrode
pattern (A) negative pattern printed on substrate (B) gold coated patterned substrate
before acetone wash (C) patterned gold electrode after pattern removal (D)
electrochemically deposited conducting polymer on the patterned gold electrode
(E ) ................... ........................................................................ 1 1 4

5.10 Side view diagram of bilayer (A) and backbone (B) type linear actuator designs. 115

5.11 WLOP image of the polyimide substrate used in the construction of linear actuators
before patterning (M agnification = 25X). ............ ...................... ....................117

5.12 WLOP image of the polyimide substrate coated with evaporated gold
(M agnification = 25X ). ................................................................................... ..... 117

5.13 WLOP image of electrochemically deposited gold on top of evaporated gold coated
polyimide (M agnification = 25X). ............. ..................... ............. 118

5.14 WLOP image of polypyrrole (+0.8V to 2.24 C/cm2) electrochemically deposited in
EcAu/EvAu coated polyimide (Magnification = 25X). ........... ...............118

5.15 WLOP image of patterned PPy/EcAu/EvAu on polyimide (Magnification = 25X). 118

5.16 WLOP image of exposed base polyimide substrate between two patterned
PPy/EcAu/EvAu wires (M agnification = 25X) ....................................................119

5.17 Schematic of conducting polymer actuator based dynamic surface. .....................1119


xviii









5.18 Images of 0.75, 0.50, and 0.25 pt line patterns produced with Xerox Phaser 6200
laser jet, HP Deskjet 6122, and Lexmark i3 inkjet printers (Magnification = 7X).121

5.19 Images of solid printed pattern areas produced with Xerox Paser 6200 laser jet, HP
Deskjet 6122, and Lexmark i3 inkjet printers (Magnification = 14X).................121

5.20 Images of 0.75 pt blue photomask (A) and enhanced image of resulting patterned
photoresist coated glass slide (B) (M agnification = 7X)..................................... 124

5.21 Images of printed field and 0.75 pt brown photomask with black toner cartridge (A)
and without black toner cartridge (B) (Magnification = 7X)..............................124

5.22 Examples of computer designed patterns used to make blue (A) and brown (B)
p h o to m a sk s ...................................... .............................................. 12 4

6.1 Images of captive air bubble contact angle measurements on PPy/PDMSe oxidizer
insertion film s ................................................. ................. 140

6.2 Captive air bubble contact angle data for PPy/PDMSe oxidizer insertion films
conducted at potentials of-0.8, 0.0, and +0.8 V versus Ag/AgCl in distilled H2O.140

6.3 Phase diagram for carbon dioxide...................................... ........................ 143

6.5 Optical images of degraded PDMSe samples after supercritical CO2 doping with
0.1M FeC13 dopant and lvol% ethanol cosolvent for 24hrs........................ 144

6.6 Optical images of untreated PDMSe (A), and PPy/PDMSe IPNs prepared by EtOH
(B) and EtOH/scCO2 (C) oxidizer insertion methods. ........................................146

6.7 Changes in PDMSe sample weight during EtOH/scCO2 prepared PPy/PDMSe IPN
sy stem s. ......................................................................... 14 7

6.8 Weight change data for IPA/scCO2 prepared PPy/PDMSe prepared samples. ..... 148

6.9 Weight change data for EtOH/scCO2 prepared PPy/PDMSe prepared samples ...148

6.10 Optical images (Leica G27) of PPy/PDMSe samples prepared by EtOH/scCO2 (A)
and IPA/scCO2 (B) oxidizer insertion methods (mag. = 2X). Sample placed on top
of printed text to show transparency. ..............................149

6.11 Optical images (Leica G27) of PPy/PDMSe batches prepared by EtOH/scCO2 (A)
and IPA/scCO2 (B) oxidizer insertion methods showing the differences in
homogeneity of PPy formation between samples. ................................................ 149

6.12 Optical images (Axioplan II) of PPy/PDMSe samples prepared by EtOH/scCO2 (A)
and IPA/scCO2 (B) oxidizer insertion methods (magnification = 100X; scale bar =
2 0 0 p m ). ......................................................................... 15 0









6.13 Optical images (Axioplan II) of PPy/PDMSe samples prepared by EtOH/scCO2 (A)
and IPA/scCO2 (B) oxidizer insertion methods (magnification = 200X; scale bar =
100l m ). ...........................................................150

6.14 Weight change of PDMSe samples exposed to different supercritical CO2
treatm ents. ....................................................................... 153

6.15 Average weight change in PDMSe samples exposed to different supercritical CO2
treatments and after an 18 hr degassing period. .................. ............................. 153

6.16 Current response of PPy/PDMSe samples prepared by THF solvent soaking and
IPA/scCO2 techniques along with untreated PDMSe. Cyclic voltammetry
conducted at +0.8 V at 10 mV/sec versus Ag/AgCl in Instant Ocean artificial sea
w after. ........................................................................... 15 6

6.17 Effects of sample hydration on cyclic voltammetry for wet and dry PPy/PDMSe
sam ples. .............................................................................157

6.18 ATR-FTIR analysis of PPy/PDMSe IPN systems prepared by various solution
soaking procedures. Inset box shows location of characteristic pyrrole ring
vibrations (1500-1600 cm ) ......... ................. ............................. ............... 159

6.19 EDS spectra of cross sectioned FeC13/THF doped PDMSe..................................160

6.20 EDS mapping of cross sectioned FeC13/THF doped PDMSe (magnification = 70X).161

6.21 EDS chlorine mapping of cross sectioned FeC13/THF doped PDMSe showing a
-60-100 [im surface layer of predominantly pure PDMSe (mag. =70X). .............161

6.22 Strain at break and peak stress data for the degradation of Silastic T2 PDMSe with
prolonged exposed to a 5 wt% FeC13/THF solution ........................ ...........163

6.23 High and low strain modulus data for the degradation of Silastic T2 PDMSe with
prolonged exposed to a 5 wt% FeC13/THF solution ........................ ...........164

7.1 Preparation scheme for the formation of PDMSe encapsulated PPy composite
samples for contact angle measurement and marine testing ...............................172

7.2 PDMSe encapsulated PPy composite samples for testing in aqueous/marine
environments using a single (A) and dual (B) element designs...........................173

7.3 Change in contact angle for a PPy/8281-65 dual element sample in Instant Ocean
artificial sea water versus a Ag/AgCl reference electrode ...................................176

7.4 Weight loss data for the formation of PPy/271-55 IPN systems. ...........................180

7.5 Volume loss data for the formation ofPPy/271-55 IPN systems...........................180









7.6 Calculated PPy/271-55 film conductivity based on conductive layer thickness.
Value for 0.07 cm is actual conductivity of PPy/271-55 film assuming bulk
conductive. .......................................... ........................... 183

7.7 Example of 5rm biomimetic sharklet pattern on PDMSe (A) and patterned silicon
wafer used to create the patterned PDMSe. SEM micrographs taken at 1000X
(scale bar = 50 rim; WD = 15 mm and EV = 15 KeV, and AuPd coated). ...........185

7.8 Optical micrograph of 5[m biomimetic sharklet pattern on 271-55 produced at
1850C (A) and patterned silicon wafer used to create the patterned 271-55. Images
taken at 1000X .................................................................... 186

7.9 SEM micrographs of sharklet patterned Santoprene 271-55 patterned at 1850C;
3300X (A) and 2500X 400 tilt (B). Images taken at 15 KeV with a working
distance of 15 mm and a AuPd surface coating. .............................................186

7.10 SEM micrographs of sharklet patterned Santoprene 271-55 patterned at 2000C;
500X (A) and 1000X (B). Images taken at 15 KeV with a working distance of 15
m m and a AuPd surface coating .............. ..................... ........... .... ........... 187

7.11 Cross sectional EDS spectrum of FeC13 doped Santoprene 271-55 (A) and 8211-65
(B) showing the presence of Fe and Cl. ...................................... ............... 188

7.12 Cross sectional EDS mapping of FeC13 doped Santoprene 271-55 taken at 180X;
scale bar = 200 m ............................................ .......................... ...... ... 189

7.13 Cross sectional EDS mapping of FeC13 doped Santoprene 8211-65 taken at 85X;
scale bar = 500 m ............................................ .......................... ...... ... 190

7.14 Cross sectional EDS mapping for Cl in FeC13 doped PDMSe (A), 271-55 (B), and
8211-65 (C). Magnification and scale bars equal to 70X, 180X, and 85X and 500
rlm, 200 rm, and 500 rm respectively. Images taken at 15 KeV and 15 mm on
carbon coated cross section samples. ........................................ ............... 191

7.15 Cross sectional EDS mapping of FeC13 doped PPy/271-55 taken at 3000X; scale bar
= 10 rm. Images taken at 15 KeV and 15 mm on carbon coated cross section
sam ples. .............................................................................192

7.16 Cross sectional EDS mapping of FeC13 doped PPy/8211-65 taken at 3000X; scale
bar = 10 rm. Images taken at 15 KeV and 15 mm on carbon coated cross section
sam ples. .............................................................................193

7.17 Cross sectional SEM secondary (A) and backscatter (B) electron micrographs of
PPy/271-55. Images taken at 3000X and scale bar equal to 10 im. Images taken at
15 KeV and 15 mm on carbon coated cross section samples..............................194









7.18 Cross sectional SEM secondary (A) and backscatter (B) electron micrographs of
PPy/8211-65. Images taken at 3000X and scale bar equal to 10 im. Images taken
at 15 KeV and 15 mm on carbon coated cross section samples...........................194

7.19 Optical micrographs of PSU films sprayed from 5 wt% (A) and 10 wt% (B)
PSU/THF solutions. M agnification = 7X ............................................................196

8.1 Diagram of three plate capacitor transducer used in the Hystitron Triboindenter.201

8.2 Images of cono-spherical (A) and Berkovich (B) Hysitron tips. Images obtained
from www.hysitron.com (magnification unknown)........ ...... ....... ............203

8.3 Diagram (A) and optical image (B) of fluid cell setup inside the Hysitron
Triboindenter ..................................... .. .. .. ...... ........ ..... 204

8.4 Optical micographs of Santoprene 271-55 (A) and PPy/271-55 (B) samples obtain
by the onboard optical camera of the Hysitron Triboindenter (magnification =
5 X ) ............................................................................ .2 0 5

8.5 3D surface plots of Santoprene 271-55 (A) and PPy/271-55 (B) surfaces. Scan
area is equal to 50 [im x 50 [im square........................................ ............... 205

8.6 Nano-DMA mapping images of Santoprene 271-55 showing uniform surface
phase and modulus morphology. Scan area is 50 itm x 50 tm square ...............206

8.7 Nano-DMA mapping images of PPy/271-55 showing uniform surface phase and
modulus morphology. Scan area is 50 [im x 50 [im square ...............................207

8.8 3D surface topography images of PPy/8211-65 at applied potentials of -0.7 V (A),
0.0 V (B), and +0.7 V (C). Scan area is 25 [im x 25 [im square.........................208

8.9 AFM optical micrograph (A) of PPy/8211-65 with corresponding surface
topography (B) and phase contrast (C) images taken at a 50.im square scan area.
Scale bars equal to 15 m ........................................................... .....................209

8.10 AFM surface topography, phase imaging, and 3D surface plot of PPy/8211-65 at
50 tm (A), 10 tm (B), 5 .im (C), and 1 .im (D) square scan areas. Scale bars equal
to 15 rm, 3 rm, 1.5 rim, and 0.3 [m respectively. ............. ............. ............... 210

r 8.11 Complex modulus maps for PPy/8211-65 at an applied potential of -0.8 V (A), 0.0
V (B ), and + 0.8 V (C ). ................................. ............ .................... ............. 2 12

8.12 Complex modulus line scan data for PPy/8211-65 obtained from the location
depicted in figure 7.28 .......... ................................ .................. .. .... ....... 2 12

8.13 3D surface plots of PPy/8211-65 under the applied potentials of-0.8 V (A), 0.0 V
(B ), and + 0.8 V (C ). .......................... .................... ... .. .. .... ...............2 13









8.14 Average surface modulus for PPy and PPy/8211-65 systems as a function of the
applied potential. .......................... .... ........................ ............ .......... ..... 2 14


xxiii















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DYNAMIC ANTIFOULING STRUCTURES AND ACTUATORS USING EAP
COMPOSITES
By

Clayton Claverie Bohn Jr.

August 2004

Chair: Anthony Brennan
Major Department: Materials Science and Engineering

By utilizing strain gage technology it is possible to directly and continuously

measure the electrochemically induced strain response of EAP actuators. Strain sensitive

actuators were constructed by directly vapor depositing gold (EvAu) on polyimide strain

gages which are capable of measuring strain with an accuracy of +/- 1 pE. Strain

sensitive actuators were used to evaluate the strain response of polypyrrole (PPy),

poly(3,4-ethylenedioxypyrrole) (PEDOP) and poly(3,6-bis(2-(3,4-ethylenedioxy)thienyl)-

N-carbazole) (PBEDOT-Cz). PPy was shown to produce significantly higher strain when

compared to PEDOP and PBEDOT-Cz. The resulting overall strain for the materials was

236, 33, and 35 pE respectively. From the initial investigation, adhesion of the EAP to

the EvAu layer was identified as a major factor in the resulting lifetime and strain

response of these actuators. Therefore an electrochemically deposited Au layer (EcAu)

was deposited on top of the EvAu layer to improve the adhesion of the EAP to the

working electrode. By changing the surface roughness from r = 3.43 (EvAu) to r = 8.26


xxiv









and 18.00 (EcAu) the normalized strain response after 2000 cycles increases from 45% to

60% and 68% respectively. Also by changing the surface roughness from r = 5 to r = 23,

the resulting strain response increases from -100 ,s to 600-800 ,s for PPy.

By incorporating conducting polymers such as polypyrrole into elastomeric base

materials such as PDMSe and Santopene TPEs a tough durable dynamic non-toxic

antifouling surface coating was formed. These coatings utilize the dynamic changes in

polymer charge, modulus, and swelling that naturally occur during the redox cycling of

conducting polymers to dynamically change the surface properties of the resulting films.

Dynamic changes in contact angle (surface energy) of 110 and 210 have been measured

for PPy/TPE (+0.5V driving potential) and PPy/PDMSe (1.0V driving potential) IPN

systems.

Using a Hysitron Triboindenter equipped with a electrochemical fluid cell setup;

dynamic fluid cell nano-DMA mapping measurements were conducted on PPy/TPE

systems. Dynamic surface modulus changes of -20-40% were measured for the

PPy/TPE composite systems when switched between their reduced to oxidized states. All

samples showed a decrease in the sample modulus when switched from the reduced to the

oxidized state. This is mainly associated with the influx of water and counter ions into

the conducting polymer phase during the oxidation process. This in turn increases the

free volume of the conducting polymer phase and also acts to plasticize the phase.

Changes in the surface topography associated with the redox cycling of the composite

structure were also observed during these experiments. This in turn results in a dynamic

surface coating that is capable of changing its surface energy, modulus, and topography

by changing applied electrical potential.














CHAPTER 1
INTRODUCTION

Smart materials such as electroactive polymers (EAPs) or more specifically

conducting polymers (CPs), such as polyaniline and polypyrrole (PPy), have gained a lot

of interest as candidates for various actuator applications such as active hinges, anti-

vibration systems, micro-catheter steerers, micro-valves and pumps, mechanical shutters,

and robotics due to their ability to undergo controllable volumetric changes under a given

stimulus.1-1 More specifically, conducting polymers have been shown to undergo

volumetric changes during electrochemical oxidation and reduction due to the

electrochemical transport of ions and solvent molecules into and out of the polymeric

system.11-23 Actuators are constructed by directly depositing conducting polymers,

electrochemically, onto a metalized flexible substrate, such as polyimide and polyester.

By doing this it is possible to directly convert induced volumetric changes into a

mechanical (bending) type of actuation.6' 24-36 The development of actuation in these

systems is very similar in nature to the motion developed in bi-metallic strip (i.e.,

thermostats). One layer changes length with respect to the other inducing a curling

motion to develop in the direction away from the expanding layer. During

electrochemical switching (oxidation and reduction) the CP layer expands and contracts

respectively while the flexible substrate maintains its original dimensions. This induces a

bending motion to develop in the direction opposite the expanding CP layer.

The amount of strain (A = AL/Lo, where AL is the overall change in length and Lo

is the original length) that is developed by these CP systems during electrochemical









switching is the main physical properties that controls the amount of actuation produced

by these devices. Therefore precise measurement of the strain developed by these

systems is an essential part in understanding the physical properties and behavior of these

materials and devices. A major factor in understanding these materials is that the

reported physical properties for CP systems vary from one study to another depending on

the characterization techniques utilized and how the sample was prepared (counter ions,

potentials, concentrations, etc.). The smallest change in any of these factors can produce

dramatic changes in the overall performance (physically and electrochemically) of these

materials. Some common characterization techniques used to study the in-plane strain

(sample elongation) of these systems are high-speed video capture, laser displacement,

and load/stress sensors (Instron/MTS type). With digital video, the motion of the

conducting polymer actuator is digitally recorded and imported to a computer for detailed

analysis.11, 37-39 Accurate measurements of deflection and elongation can be obtained

however this involves post processing of the recorded data and results can vary

depending on what method was used to calculate the strain. Real time laser displacement

meters or extensometers have also been utilized to evaluate the degree of motion induced

during electrochemical switching.26'40 41 These systems allow for accurate, real time

monitoring of tip displacement of the actuator system. Tip displacement can then be

converted to strain at a latter time. Use of force/displacement meters or load cells is

commonly used in the study of conducting polymers actuators under linear actuation

conditions.13' 42-48 Out-of-plane strain (sample thickening) for PPy has also been

measured using atomic force microscopy and has been shown to be significantly higher

than the in-plane strain.49









A new method has been developed for the in situ measurement of the in-plane

strain response developed by cantilever (bi-layer) style CP actuators. With this technique

it is possible to measure very precisely the average strain developed in the system during

electrochemical switching. This method employs the utilization of Au coated electrical

resistance strain gages as the working electrode (flexible metalized substrate) during

electrochemical switching. Strain gages are widely used in many industries (automotive,

aeronautical, naval, construction, etc.) for precision in situ spot measurements of induced

strains in many materials. These devices are capable of measuring strain with a precision

of 1 t (tE = 1.10-6 ) and have a strain limit of 5% (50,000 tEs). These devices are

also utilized as the sensing components of load cells and therefore are capable of

measuring stresses when applied in the right configuration.

Strain gages are typically constructed of a constantan foil grid that is encapsulated

in a flexible polyimide substrate, but other gages are available utilizing different grid and

backing materials. These devices are very flexible in use as well as installation and are

capable of measuring strain under various load conditions and environments, including

cyclic strains. This ability to conduct in situ cyclic strain measurements with high

precision is ideal for applications in EAP actuators and sensors due to the relatively low

strain produced and cyclic nature of their motion. This technique allows for detailed real-

time analysis of the effects of various actuation, electrochemical polymerization, and

actuator construction factors such as CP film thickness, working electrode morphology,

counter ion type, driving potential, polymerization potential, and scan rate or switch

speed.









Another issue associated with the development and use of conducting polymer

based actuators is interlayer adhesion. The interface between the flexible Au coated

substrate and stiffer conducting polymer is subjected to large amounts of stress during the

redox cycling process. The repeated cyclic swelling and deswelling of the conducting

polymer layer eventually initiates crack formation at this interface. The cracks then

propagate with continued cycling and eventually caused the complete delamination of the

conducting polymer layer and the failure of the device.

A new electrochemically deposited Au (EcAu) layer was developed to improve the

interlayer adhesion between the conducting polymer and Au electrode layers. The EcAu

treatment greatly increased the base electrode surface area by the growth of Au crystals

on the preexisting evaporated Au layer. These crystals range in size from sub micron to

10-20im which can be controlled by the electrodeposition time or current density. These

surface treatments have been shown to greatly improve overall actuator performance and

lifetime when applied. Performance variable that are improved include properties such as

produced strain, strain rate, frequency response, lifetime, and potentially the stress out put

from these systems.

By applying the lessons learned from the previous mentioned work a novel

dynamic surface coating was developed by the incorporation of conducting polymers

such as polypyrrole into elastomeric material such as PDMSe and thermoplastic

elastomers (TPEs). These coatings have potential applications as dynamic non-toxic

antifouling coatings for marine and industrial applications. Biofouling is estimated to

cost the US Navy alone over $1 billion per year by increasing the hydrodynamic drag of









naval vessels 50, 51 thus resulting in decreased range, speed, and maneuverability of naval

vessels and increases the fuel consumption by up to 30-40%.52, 53

It has been shown that the settlement and adhesion of fouling marine organisms

are effected by surface properties such as surface energy, modulus, and topography. Due

to genetic diversity of the various fouling marine species the effects of surface properties

on settlement and adhesion changes from species to species. Therefore the use of single

surface coating with fixed surface properties can not deter the settlement and growth of

all marine organisms. In this approach the dynamic changes in polymer charge, modulus,

and topography associated with the redox cycling of conducting polymers is used to

develop a dynamic surface coating with variable properties. This should in turn provide a

wider range of deterrence to settlement and growth than can be obtained from non-

dynamic surfaces.














CHAPTER 2
BACKGROUND

2.1 Electroactive Polymer Actuators

2.1.1 Introduction to EAPs

Electroactive polymers (EAP) are a class of polymers that respond to electrical

stimuli by changing their properties and shape. As described by Bar-Cohen 54 these

materials can be separated into two classes: electronic and ionic electroactive polymers.

The electronic EAPs class of materials is driven by an applied electric field or Coulomb

forces. The second class of ionic EAPs is driven by the movement of ions and solvent

into and out of the polymeric structure.

The electronic EAPs include ferroelectric polymers, dielectric EAPs,

electrostrictive graft elastomers, electrostrictive paper, electroviscoelastic elastomers,

electrorheological fluids (ERF), and liquid crystal elastomer (LCE) materials. Ionic

EAPs consist of ionic polymer gels (IPG), ionomeric polymer-metal composites (IPMC),

conducting polymers (CP), and carbon nanotubes (CNT).

Ferroelectric polymers are typically characterized by the piezoelectric materials.

The piezoelectric effect was discovered in 1880 by Pierre and Paul-Jacques Curie. They

found that by compressing certain crystals such as quartz along the appropriate axes a

voltage is produced. The following year they discovered that the opposite was true as

well. If a voltage was applied to these crystals they in turn elongated. Ferroelectricity is

the phenomenon of piezoelectricity that is observed in nonconducting crystalline or

dielectric materials such as poly(vinylidene fluoride) and its copolymers that exhibit









spontaneous polarization under a given electrical stimuli. Piezoelectric materials are

characterized as being semi-crystalline with an inactive amorphous phase. These

materials are relatively stiff with a Young's modulus between 1-10 GPA. A relatively

high driving potential of about 200 MV/m, which is very close to the materials dielectric

breakdown point, is required to activate the piezoelectric materials. Piezoelectric

materials are capable of producing strains on the order of 2% (5% strain possible at 150

V/im with specially modified materials) with very rapid speed and are able to be

actuated in air, vacuum, and aqueous environment.

Dielectric EAPs are a class of materials that are characterized as having a low

modulus with a high dielectric constant. These polymers are capable of inducing large

actuation strains with the application of a given electric field. Actuators are typically

constructed by placing/laminating electrodes on opposite sides of the elastomeric

dielectric material. Once an electric field is applied the electrodes are electrically

attracted to each other applying a compressive force on the elastomeric dielectric material

causing it to deform and expand in the transverse direction. The dielectric EAP's also

require a large driving potential on the order of 100 V/itm, but are capable of producing

strains on the order of 10-300%. As with piezoelectric materials this driving voltage is

very close to the breakdown point of the material.

Electrostrictive graft elastomers have a flexible amorphous backbone with grafted

polar polymer segments that are capable of forming crystalline domains. The crystalline

domains react to the applied electrical field. These materials are capable of producing

strains on the order of 4% and are readily processable.









Electroviscoelastic elastomers and electrorheological fluids are both characterized

as materials that have dielectric particles dispersed throughout the bulk of the material.

The most basic example of an ERF is that of dielectric particles (0.1-100 .im) dispersed

in insulating base material such as silicone oil. Under an applied electrical field a dipole

moment is induced on the particles which then align and form chains increasing the

viscosity of the fluid. These fluids take on the consistency of gels under the applied

electrical field. Electrovisoelastic elastomers are similar in structure but utilize a

crosslinked insulating phase instead of a fluid phase. Under an applied electrical field

(<6 V/rim) the particles interact and increase the shear modulus of the material by as

much as 50%. Both of these materials have applications as active dampeners.

Ionic EAPs such as ionic polymer gels (IPG) are driven or actuated by the

movement counter ions in the material. IPGs are capable of producing strains on the

order of 400% under chemical and electrical stimuli. However they suffer from

extremely long actuation times on the order of 20 minutes for a 400% increase in size. If

these gels are placed in an electrolyte solution and an electrical stimulus is applied the

material will expand and in all directions. However if the gel is used in a dry

environment and electrodes are placed on opposite sides of the gel it will form a bending

actuator. As one side becomes more alkaline (cathode) and the other more acidic (anode)

the migration of ions through the material results in an expansion on the cathodic side

and a contraction on the anodic side. Besides having relatively long actuation times the

large induced strains tend to damage the applied electrodes after only a couple of cycles.

Ionomeric polymer-metal composite actuators operate on the mobility of cations in

the composite structure. Upon the application of a low driving voltage (1-10 V) cations









are transported from one side of the actuator to the other through negatively charged

channels in the polymer film. Acid functional fluorocarbon polymers such as Nafion

(sulfonate based) and Flemion carboxylatee based) are two examples of polymers used

in these composites. Electrodes are applied chemically from solution by implanting

metal ions (gold, platinum, etc.) throughout the hydrophobic regions of the actuators

surface. These materials are characterized as producing relatively large strain at

frequencies below 1Hz. As with most ionic EAPs, as the switching frequency increases

the produced strain decreases.

Carbon nanotubes (CNT) have recently been identified as a class of ionic EAPs.

They are very similar to conducting polymers in that they are nearly completely

composed of a conjugated carbon-carbon bond network (excluding defects in the

structure). However unlike conducting polymers the actuation produced by carbon

nanotubes is due to the change in the carbon-carbon bond length as the material is

oxidized and reduced. Upon removal (oxidation) of electrons a net positive charge is

formed across the conjugated network. This net positive charge on the carbon nuclei

causes the carbon nuclei to repel one another resulting in an increase in the overall length

of the CNT. Upon injection of electrons (reduction) into the conjugated carbon network

an increase in carbon-carbon bond length is also induced. These devices have been

shown to produce strains on the order of 1% (along length of nanotube) and are capable

of withstanding temperatures in excess of 10000C. This allows for the development of

new high temperature actuators that far exceed the temperature capabilities of current

systems.









The main EAP of interest for this research is the conducting polymer. These

materials are comprised of polymers that have a completely conjugated backbone such as

polypyrrole, polyanaline, and poly(p-phenylene).

2.1.1 Conducting Polymers

Conducting polymers have recently received a lot of attention due to their ability to

generate various responses under electrical or chemical stimuli along with the ability to

be electrical conductors. Most notable for this application is volumetric changes that

occur during chemical or electrochemical oxidation and reduction. The changes in

volume are due to development of charge along the polymer backbone.








Bipolar unit








Figure 2.1 Bipolar unit charge migration as depicted for polythiophene.

Conducting polymers are characterized as having a fully conjugated polymer

backbone with an extended 7t-system. This allows for electron and charge delocalization

along the polymer backbone as shown in Figure 2.1.55 The volume change associated

with oxidation of conducting polymers in electrolyte solutions (salt solutions, body

fluids, etc.) is complex but is mainly induced by the influx of solvent and anions into the

conducting polymer matrix to balance the developed increasing charge resulting in a









volume expansion of the material.11-23 During reduction the material starts to return to its

neutral (uncharged) state and anions are expelled from the polymer matrix resulting in a

volume contraction. There have been numerous studies examining the developed strain

and force associated with the redox swelling and de-swelling of conducting polymers.23'

29, 33, 56-62

2.1.2 Conducting Polymer Synthesis

Electrochemical polymerization is generally carried out by the application of an

oxidizing potential equal to or greater than the oxidation potential of the monomer.55 63

The general mechanism for the polymerization of conducting polymers involves the

formation of a radical-cation on the monomer structure. The propagation step involves

the subsequent coupling of two oxidized monomers to form a dicationic dimer upon

which two hydrogen atoms are expelled and a stable dimer is formed.

Oxidation of the dimer followed by the subsequent coupling with an oxidized

monomer, dimer, or oligomer results in the continued propagation of the conjugated

polymer chain. The exact mechanism for the oxidative polymerization of conducting

polymers is unknown but a series of possible mechanisms is shown by Sadki et al..64

Figure 2.2 illustrates one of the more accepted polymerization mechanisms for

polypyrrole.

The properties of a conducting polymer are dependent upon many variables. The

main factors are polymerization/oxidation potential, electrochemical polymerization

conditions (potentiostatic, galvanostatic, and variations thereof), dopants (C1-, C104-,

etc.), as well as solvents and electrodes that were used during the electropolymerization

process.








H H
\CC/
S\-c

H C C H


H H
\-c/
C-C

H h~-


-e


H H
\_/
H C-C/ H



H


H H
C__c/
/ \\


H
H H
\ /
SCC

HC/H
'?7-


H H H H
\ / \ /
C-C C C

H C 7* eHHC C H

H H


H H
H H


H H
\ / H




H \
H H


/C/ /H H
I/ I\ / H
H,/ / N C /


H H
H H
Figure 2.2 Example of one of the possible mechanisms for the oxidative polymerization
of polypyrrole. Adapted from Sadki et al..64

Conducting polymers can also be formed via chemical polymerization methods

such as Lewis acid, ring opening metathesis, and transition metal catalyzed coupling









polymerizations.55, 63 The most common chemical method utilized to make conducting

polymers (due to its simplicity) is the use of Lewis acids such as FeC13, FeC104, and

ammonium persulfate.

The polymerization mechanism is very similar to that of the electrochemical

polymerization; however electron transfer is too oxidizing species instead of an electrode.

During polymerization an electron is transferred from the monomer (oxidation) to

Fe(III)C13 which in turn is reduced to Fe(II) leaving C1- (counter ion) to balance the

positive charge developed on the monomer during the oxidation process. Chemical

polymerization is a bulk polymerization method that is a distinct advantage compared to

electrochemical methods. However, chemical synthesis with Lewis acids is not as

controllable as the electrochemical methods and is known to produce material with

greater defects (disrupted conjugation) densities, thereby decreasing the overall

conductivity of the material.

2.1.3 Conducting Polymer Actuators

Conducting polymer actuators are candidates for various actuator applications

including robotics, artificial muscles, microvalves, catherter steerers, antivibration

systems, and multiple other systems.1-7, 9, 10, 65, 66

There are numerous forms of conducting polymer actuators; however all CP

actuators are typically based on the idea of a conducting polymer laminated to a flexible

conductive substrate. The most basic design involves the deposition of a CP film on one

side of a conducting flexible substrate (sputtered Au on polyimide or PET) and utilizes an

external counter electrode for operation.

During redox cycling the CP expands (-1-2% in-plane 29, 44, 67and -30% out-of-

plane49) and contracts due to the movement of ions into and out of the CP structure









respectively. This volume change can then be converted into a bending type actuation by

deposition of the CP film onto an inactive but flexible substrate such as polyimide or

PET. As the CP is oxidized it expands while the flexible substrate maintains its original

length. This causes a bending moment in the direction away from the expanding CP

layer. This type of actuator is typically referred to as a bilayer actuator and has been

studied extensively.6' 24-36, 45, 65, 66, 68-70

A modification of the bilayer design is the backbone type design. In the backbone

type design a CP layer is deposited on either side of the flexible substrate. One side acts

as the anode (working electrode) while the other side acts as the cathode (counter

electrode). During redox switching one side expands (anode) while the other side

contracts (cathode). When the driving potential is flipped there roles reverse and the

bending occurs in the opposite direction. This push-pull technique results in an enhanced

bending moment or increased strain performance.

In order for these types of actuators to work they must be run in an electrolyte fluid

for ion transport. Another configuration is the shell type design. The shell type actuator

is basically a modification of the backbone type design. However in this case, the

flexible substrate is replaced with a flexible adhesive polyelectrolyte (source for ion

transport) and the whole device is encapsulated to prevent dehydration of the system.

These type devices can be actuated in air and do not require an external ion source such

as an electrolyte solution.

Figure 2.3 shows some examples of typical CP bending actuators. Linear actuators

have also been constructed by wrapping CP fibers around a flexible electrode or vice









versa. In this case the CP fibers expand and contract in length (axial direction) and a

linear type actuation is produced.44' 46' 6 71




Ad hesive
h r Polyelectrolyte
















SConducting Protective
Polymer Layer
(a) (b) (c)


Figure 2.3 Examples of basic bending (cantilever) conducting polymer actuators; A)
bilayer actuator, B) backbone type actuator, C) shell type actuator.

These actuating devices are relatively easy to make and due to new lithographic

micropatteming techniques they can be fashioned into almost any configuration

imaginable making them very versatile. These devices are usually driven with an applied

voltage in the range of 1-5 V depending on the polymer used. However there actuation

time (strain rate) and performance are very dependent on polymerization and actuation

conditions.









One of the main drawbacks to CP actuators is that the whole actuation/swelling

process is diffusion limited. Thinner layers have an improved strain rate; however they

also produce lower overall stresses and strains than thicker films. Therefore as one

increases the CP film thickness to increase the generated stress and strain the overall

speed of the system decreases rapidly. These devices are also subject to large amounts of

shear at the CP-substrate interface and are prone to interfacial cracking and delamination

when cycled for prolonged periods of time.54'72'73

2.2 Electrical Resistance Strain Gages

The amount of strain developed by the EAP systems during electrochemical

switching is one of the main physical properties that controls the amount of actuation

produced by the EAP devices. Measurement of the developed strain is an essential part

in understanding the physical properties and behavior of these materials and devices.

Reported physical properties for EAP's vary from one study to another depending on the

characterization technique and sample preparation.

Some common characterization techniques used to study the in-plane strain of these

systems are high-speed video capture, laser displacement, and load/stress sensors

(Instron/MTS type). With digital video, the motion of the conducting polymer actuator

can be recorded and imported to a computer for detailed analysis.11' 37-39 Accurate

measurements of deflection and elongation can be obtained with this technique, but it

involves post processing of the data recorded. Another drawback of the video system is

as actuation speed increases the speed and resolution of the video capture system also has

to increase to record accurate data; therefore the price of the system has to increase as

well.









Real time laser displacement meters or laser extensometers have also been used to

evaluate the degree of motion induced during redox switching.26'40'41 These systems

enable accurate, real time monitoring of tip displacement of the actuator system. Tip

displacement can then be converted to strain at a latter time. Force/displacement meters

or load cells are also commonly used in the study of conducting polymers actuators under

linear actuation conditions.13, 42-48 Out-of-plane strain has also been measured using

atomic force microscopy.49

We have developed a new technique for the in-situ measurement of the strain

response developed by cantilever (bi-layer) style EAP actuators. This method employs

the utilization of gold coated electrical resistance strain gages (Figure 2.4) as the working

electrode during electrochemical switching. Strain gages are widely used in many

industries (automotive, aeronautical, naval, construction, etc.) for precision in-situ spot

measurements of induced strains in many materials. These devices are capable of

measuring strain with a precision of 1-10-6 E (= ~.) and have a strain limit of 5%

(50,000 tE). These devices are also capable of measuring stresses when applied in the

right configuration, such as in load cells.

Electrical resistance strain gages are typically constructed of a constantan foil grid

that is encapsulated in a flexible polyimide shell, but other gages are available utilizing

different grid and backing materials. As the constantan foil grid is deformed (during

redox switching) the electrical resistance of the grid changes, this change in electrical

resistance is directly related to the strain induced on the grid/system. The strain (E) is

reported as the change in the grid length (AL) relative to initial grid length (Lo), E=AL/Lo.









constantan grid





1. -EAP







polyimide solder tabs


Figure 2.4 Diagram of a polyimide based electrical resistance strain gage and EAP
actuator setup. Adapted from www.Vishay.com.

These devices are very flexible in use as well as installation and are capable of

measuring strain under various load conditions and environments, including cyclic

strains. This ability to measure cyclic strains with high precision is ideal for applications

in EAP actuators and sensors due to the relatively low strain produced and cyclic nature

of their motion.

2.2.1 Strain Gage Theory

The basic principles allowing for the development of modern electrical resistance

strain gages were discovered in 1856 by Lord Kelvin. He observed that under tensile

loading of copper and iron wires the resistance of the wires increased for a given amount

of strain. He also discovered that the iron wire exhibited a larger change in resistance

than copper for a given amount of strain. He finally applied a Wheatstone bridge to

accurately measure the resistance changes developed in these systems.









From this it can be said that the resistance of a wire is a function of the strain

applied to that wire, different materials produce different resistance changes for a given

strain (sensitivity, SA). Therefore the resistance (R) of a uniform wire can be written as:


R=p (1.1)
A

where p is the specific resistance of the material, L is the wire length, and A is the cross-

sectional area of the wire. The sensitivity of the wire can then be describes as the

resistance change of the wire per unit of initial resistance divided by the applied strain



dRR0
SA (1.2)


By combining equations 1.1 and 1.2 and rearranging the sensitivity term can be derived

to determine the sensitivity as a function of the Poisson's ratio (v, E/E ) of the material

and its change in specific resistance due to the applied strain, Eq. 1.3.


S = 1+ 2 +dpp (1.3)


Equation 1.3 can be broken down into two parts, the effects of dimensional changes

during applied stain (1+2v) and the effects of specific resistivity (iLQ-m)with applied

strain ([dp/p]/s). Most metallic alloys have an SA value between 2 to 4 with the value of

(1+2v) varying from 1.4 to 1.7. This gives a range of 0.3 to 2.6 for the change in specific

resistance, which can be quite significant compared to the effects of dimensional changes.

The change in specific resistance is due to the number of free electrons and the variation

of their mobility with the applied strain.









The actual strain is determined from the gage factor (Sg). This factor is

experimentally determined for each batch of foil gages by applying a known strain to a

series of gages mounted on a specially designed cantilever beam and measuring there

resistance response to the applied strain. This is expressed as

AR
R = SgE, (1.4)


where Ea is the applied axial strain.

2.2.2 Strain Gage Materials and Construction

A major factor in choosing an appropriate gage alloy is that the strain sensitivity is

linear over a wide range of applied strains. This allows for the use of a single calibration

factor and insures that this calibration factor will not change with the various degrees of

strain that the gage might see.

Some other factors to consider is that the alloy's SA should not change significantly

as the material enters the plastic regime, the alloy should also have a high specific

resistance, thermal stability, and not be significantly affected by temperature change.

Temperature compensation can typically be controlled for a particular alloy by the

addition of trace impurities and heat treatment. Temperature compensating strain gages

reduce the effects of the AR/R induced by temperature changes to less than 10-6/oC.

The most common alloy used is Advanced or Constantan. The Constantan alloy is

comprised of 45% nickel and 55% copper and has a specific resistance of 0.49 .iQ-m

which allows for the construction of smaller gage patterns with high resistance. Due to

electrical circuit requirements the minimum gage resistance needs to be above 100 Q to

prevent overloading of the power supply and to minimize gage self heating (resistive

heating). Thus the minimum gage element strand length is on the order of 4" when made









of the finest standard wire. To keep the overall gage length down the sensor elements are

typically folded back and forth to form a grid pattern (Figure 2.4). Typical gage

resistances are 120 and 350 Q, but higher resistance gages are available.

Some other common gage alloys are Nichrome V (80 Ni, 20 Cr; SA = 2.2),

Isoelastic (36 Ni, 8 Cr, 0.5 Mo, 55.5 Fe; SA= 3.6), Karma (74 Ni, 20 Cr, 3 Al, 3 Fe; SA

2.0), Armour D (70 Fe, 20 Cr, 10 Al; SA= 2.0), and Alloy 479 (92 Pt, 8 W; SA= 4.1).

There are advantages and disadvantages to the use of these alloys when compared to

Constantan. For example Isoelastic has a higher sensitivity (3.6) and higher fatigue

strength compared to Constantan, allowing for more precise measurements under high

cyclic strains exceeding 1500 [ig. However Isoelastic is extremely sensitive to

temperature effects. A change in gage temperature of 1 C will result in an apparent

strain change of 300 400 [i.

Karma has similar properties compared to Constantan but has a higher fatigue limit

and excellent time stability, allowing for extended measurement periods (weeks to

months). Karma also has a larger use range (up to 260 C) than Constantan (up to 204

C). Karma is difficult to solder, which makes it difficult to attach lead wires. Figure 2.5

shows the thermally induced apparent strain for the Constantan (Advanced), Isoelstic,

and Karma alloys.










Temperature, OC
-50 0 +50 +100 +150 +200 +250
+500r r r r r Constantan ( Is ,
*4o.00. .- ---------r- -,- --
+300 -- --i --I--,---,--T --

Karma alloys. Taken from Daly et al., Figure
S200 -- -----------------------------

















The versatility of this process allows for the production of a variety of gage sizes and

shapes. However the patterned metal grid is very thin and fragile. This makes it prone to

damage by distortion, wrinkling, and tearing of the grid element.

A solution to this problem is to back the metal film elements with flexible sheets

like polyimide (0.025 mm thick). This improves the mechanical stability of the gage and

handling. It also improves bonding of the gage to various surfaces including conductive

ones (polyimide backing acts as an electrical insulator between the grid element and the

substrate). The gage can also be encapsulated with a top layer of polyimide film for use

in adverse conditions like under water and in areas prone to high levels of dust/debris.

Other substrates like very thin high modulus epoxy (transducer applications) and glass

fiber or phenolic reinforced epoxies (high level cyclic strain and high temperature

applications) are also used under special conditions. These gages can be affixed to a
-Jon -4- --I .
-200 -.-- -,--,--,--.-- ----
-300 ----,- --.--.--.--.--.-


-00 0 +100 +200 +300 +400 +500
Temperture, OF

Figure 2.5 Thermally induced apparent strain for Constantan (Advanced), ]soelastic, and
Karma alloys. Taken from Dally et al., Figure 6.2 4

Current strain gages use metal foil grids produced from a photoetching process.

The versatility of this process allows for the production of a variety of gage sizes and

shapes. However the patterned metal grid is very thin and fragile. This makes it prone to

damage by distortion, wrinkling, and tearing of the grid element.

A solution to this problem is to back the metal film elements with flexible sheets

like polyimide (0.025 mm thick). This improves the mechanical stability of the gage and

handling. It also improves bonding of the gage to various surfaces including conductive

ones (polyimide backing acts as an electrical insulator between the grid element and the

substrate). The gage can also be encapsulated with a top layer of polyimide film for use

in adverse conditions like under water and in areas prone to high levels of dust/debris.

Other substrates like very thin high modulus epoxy (transducer applications) and glass

fiber or phenolic reinforced epoxies (high level cyclic strain and high temperature

applications) are also used under special conditions. These gages can be affixed to a









variety of surfaces utilizing epoxy cement, cyanoacrylate cement, polyester adhesives,

and ceramic cements depending on the testing/application conditions.

2.2.3 Strain Gage Accuracy

Electrical resistance strain gages are considered one of the most accurate methods

for determining strain available. These gages are capable of measuring strain with a

precision on the order of 1 p i. This is due to the ability to produce gages with resistance

accuracy of 0.3% and with gage factors (calibration constant) certified to 0.5%.

However there is some error associated with the transverse sensitivity of the gages.

The end loops place a small portion of the gage in the transverse direction. So any

transverse strain in the system will increase the resistivety of the gage resulting in a false

increase in the measured linear strain. Modern foil gages have enlarged end loops which

decrease the sensitivity of the gage to transverse strain in the system. By enlarging the

end loops the overall resistance of these segments is very low and therefore any change in

the resistance due to the transverse strain will be negligible. The only time error due to

transverse strain is not a factor is where the transverse sensitivity factor (Kt) of the gage

is zero or when the applied stress field is uniaxial.

The overall change in resistance for a strain gage is a result of the applied strains

(axial, transverse, and shear) and the sensitivity of the gage to each of these strains. This

is expressed as:

AR
--= SE' + S,E, + Sa), (1.5)
R

where Sa, St and Ss are the sensitivities of the gage to the applied axial, transverse and

shearing strains and ea, at and eat are the applied axial, transverse, and shearing strains.









The sensitivity of the gage to the shearing stress is negligible and is usually ignored.

Thus reducing equation 1.5 down to


R= S ( +KE,) (1.6)


where Kt = St/Sa (transverse sensitivity factor; %).

By assuming the effects of Kt and at are negligible; the gage calibration factor or

gage factor can be determined. The gage factor as a function of the applied axial strain is

expressed as

AR
S= R (1.7)

where Sg is the gage factor. This factor is used to convert the resistance changes

observed in the gage to the applied strain seen by the gage.

The amount of error that occurs if only the gage factor is considered in

determining the strain is expressed as

= K,(,/, +v) .100 (1.8)
1- voK,

where Vo is the Poisson's ratio of the material being tested. If you assume applied axial

and transverse strain are the same (st/Ea = 1, usually Sa > Et .'. Kt/Ea < 1), a Poisson's ratio

of 0.285 (Poisson's ratio for calibration beam) and a Kt of -0.006 (-0.6%, Kt for strain

gages utilized during dissertation work) the error associated with transverse strain is

-0.77%.

The strain gages utilized for this work were directly bonded to the conducting

polymer being studied (w/ evaporated gold interface). So the Poisson's ratio and the

applied strain field are both factors of the conducting polymer being studied. During









electrochemical growth of conducting polymers there is no axial or transverse orientation

(to the gage element) of the polymer so during redox cycling the induced strain field

would be uniform in both the axial and transverse direction (to grid element). By

assuming this both the Et/Sa and Vo terms become one (applied strain field and expansion

of CP is biaxial). Assuming this the calculated error associated with the transverse strain

would be -1.19%.

2.3 Anti-Fouling/Foul-Release Coatings

Biofouling is the result of marine organisms settling, attaching, and growing on

submerged marine surfaces. The biofouling75 process is initiated within minutes of a

surface being submerged in a marine environment by the absorption of dissolved organic

materials (conditioning film). Once this conditioning film is deposited, bacteria

(unicellular algae) will colonize the surface within hours of submersion. The resulting

biofilm produced from the colonization of the bacteria is referred to as microfouling or

slime and can reach thicknesses on the order of 500 im. Macrofouling species may

eventually colonize on top of the microfouling or slime layer.

Soft and hard fouling are two different classifications of macrofouling. Soft

fouling consists of algae and invertebrates such as anemones, hydroids, soft corals,

sponges, and tunicates. Hard fouling consists of invertebrate species like barnacles,

muscles, and tubeworms.50 This results in a multilayer structure with each layer having

its own unique properties and adhesion mechanisms.

To further complicate things, the species makeup for each layer is dependent on the

geographical location and the makeup of the underlying structure. This issue is further

compounded due to the fact that there are 12 well-defined geographical zones in the









world's oceans with varying salinity, clarity, temperature, amount and type of

micronutrients, and number and type of fouling organisms.52

















Substrate








IIlU. ll lt.lk3al tl d
-wvnmsfdxs Mla 5M ItiL I

Figure 2.6 Diagram showing the different species breakdown and layering of biofoul film
formation.

Biofouling is estimated to cost the US Navy alone over $1 billion per year by

increasing the hydrodynamic drag of naval vessels.50' 51 This in turn decreases range,

speed, and maneuverability of naval vessels and increases the fuel consumption by up to

30-40%.52, 53

Anti-fouling and foul-release coatings are two main approaches used for combating

biofilm formation. Anti-fouling coatings prevent or deter the settling of biofouling

organisms on a surface by the use of leached biocides. Foul-release coatings control

biofilm formation by modifying surface properties in such a way as to prevent the









formation of strong adherent bond between the biofoulant and the surface. This reduces

the work required to remove them.

The first method is typically accomplished through the use of anti-fouling coatings

containing heavy metal biocides, such as cuprous oxide or tributyltin (chloride and

oxide), which are either tethered to the coated surface or are released from the surface

into the surrounding environment. Use of these coatings has caused problems in the

marine ecosystem, especially in shallow bays and harbors where the biocides can

accumulate. As such the use of tributyltin has been banned in many parts of the world

and its use will be completely banned worldwide on all naval vessels by the International

Maritime Organization in January 2008.50, 51

The second concept for controlling biofouling is the use of foul-release coatings.

Foul-release coatings control biofilm formation through the use of engineered surfaces

with controlled surface properties such as surface energy, modulus, and roughness to

minimize biofoulant adhesion.50, 52, 76-79 More recently the use of nano and micro-scale

topographies has come into interest.51'80-82

The interest in the use of surfaces with controlled/tailored surface energies for foul-

release coating is due to the fact that wetting of the surface with biological glue (or any

other fluid) is controlled by the surface tension/energy. By controlling the surface energy

of a material the wettability of it's surface and therefore bond formation can be

controlled.52 Work done by Pasmore and Bowman 76 has shown that the percent removal

of biofilms, Pseudomonas aeruginosa under given conditions, increases with a decreased

in contact angle (increase in surface energy) and decreases with an increase in surface

roughness. Similar results were produced on smooth glass, electropolished 316 stainless









steel, and PTFE samples which indicated a biofilm accumulation -35% lower than for

rougher surfaces.77

Another factor of interest is the modulus of the material being used. Gray and

Loeb 78 have shown that the degree of settlement of various organisms decreased with a

decrease in modulus of crosslinked PDMS samples. From this it can be said that the

force required to remove an adhered biofilm is a factor of the settled surfaces surface

energy and surface modulus, this is in agreement with the Kendal theory 52

Pc= 47rEw la3 /(1_ 2) (1.9)


crc = 4Ew /ra(- v2) (1.10)

which gives the critical pull-off force (Pc) and critical crack propagation stress (oc) as a

function of elastic modulus (E) and the work of adhesion (wa; Wa = 2-y). The work of

adhesion is related to the surface energy by the interfacial tension (y), where "a" is the

contact radius and v is the Poisson's ratio.

Another current area of interest is the use of micro-topographies in the control of

biofoulant spore settlement and adhesion.51 80-82 By manipulating surface topography,

through the use of micropatteming, the overall surface energy and therefore

hydrophilicity or hydrophobicity of a coating material can be manipulated. Many studies

have shown the importance of surface roughness on the settling of spores. However the

importance of surface feature size and shape has begun to be studied only recently. Work

done by Verran and Boyde (2001) has shown that macro-scale surface features (>10 im)

are relatively unimportant in cell settlement since cell dimensions are much smaller than

the surface features. It has been suggested that the shape, scale, and periodicity of

surface features may influence the settlement of barnacle larvae (Hills & Thomason,









1998; Lapointe & Bourget, 1999; Berntsson et al.,2000), as reviewed by Callow et al..80

Work by Brennan and coworkers 80,81 has shown that the settlement of spores can be

influenced by the use of ridge and pillar micro-topographies on PDMSe surfaces with

feature sizes and spacings on the order of 1.5-20 im. In this study the minimum spacing

between features was 5[m which was similar in size to the diameter of the cells being

studied, therefore the cells settled into the valleys between the features. From this work it

was determined that the pattern spacing should be reduced to 2-3 .im to increase the work

required to settle on the patterned surface. Recently a new biomimiticly engineer

surface topography (B.E.S.T. or sharklet) utilizing 2-3 rm feature sizes has been

developed. This pattern has been shown to reduce the settlement of Ulva spore (green

algae) by -85%.

Similar results were found for flocked surfaces by Kim et al..51 They found that by

flocking a smooth PVC surface with a heterogeneous mixture of nylon fibers (90% 1.8

denier and 1.27 mm long; 10% 3.0 denier and 2.54 mm long) that they could influence

the settlement and growth of different marine species. However they found that the

flocked surface inhibited the growth of some species but had no effect or even enhanced

the growth of other species.

From this it can be said that surface features play an important role in the

settlement and adhesion of biofouling organisms to a particular surface. However, it has

been shown that different factors influence different species in different ways, i.e., what

inhibits the growth of one species might enhance the growth of another. This makes it

unlikely that a single coating with a fixed set of surface parameters will be effective in

preventing biofilm formation for all species. Therefore a surface coating with









dynamically variable surface properties (surface energy, modulus, and roughness) would

be more effective in the prevention or retardation of biofilm formation.

2.4 Electrowetting

Electrowetting is the process of changing the surface wetability (surface tension) of

a metal electrode by rearrangement and or formation of an electronic double layer (EDL)

at the electrode's surface due to an applied electrical potential. The electrowetting (EW)

process has been extensively studied for pure metal electrodes with electrolyte

solutions.83-85 EW devices have been limited to uses in polar media due to the nature of

the formed EDL.

The EDL is formed from the transfer of electrons from the electrode to redox-active

species in the fluid medium. The electrical stability of the EDL limits the use of these

devices to low voltages, as low as -1 V.86, 87 However the induced change in contact

angle (AO) is proportional to the amount of charge developed at the electrode surface

thereby limiting the overall AO that can be produced. Two major applications for this

technology are in micro-fluidic devices and MEMS type applications. A schematic of an

EW capillary pump is depicted in Figure 2.7.8

Ca) Teflon Mb
S.. ... ..... ........ ........................ .. .. ........
..... ... ... ................ .. ....... .. ......

d Electrolyte E.D1

S Substate -_____ / -_ __.1
[Ecadrode (Au) -
V
Figure 2.7 Design of electrowetting device: (a) no applied electrical potential
(hydrophobic surface); (b) with applied electrical potential hydrophilicc
surface). Fluid is pumped by continuously cycling the applied electrical
potential.87









Recently, it has been shown that the application of a thin dielectric layer (PTFE,

SiO2, etc.) between the electrode and fluid can enhance this EW behavior. This allows

the pumping/wetting of virtually any fluid medium. This behavior is referred to as

electrowetting-on-dielectric (EWOD). However a higher electrical potential is required to

drive these systems.86-89 Typical operating voltages for EWOD devices typically exceed

100-200 V. Yet, more recently, Moon and coworkers produced EWOD devices

operating as low as 15V.86

EWOD devices also are resistant to corrosion due to the protection offered to the

electrode by the dielectric layer and can produce larger AO when a hydrophobic dielectric

(higher initial contact angle, lower initial surface energy) is used (i.e., PTFE) when

compared to EW based devices. The dielectric layer blocks electron transfer from the

electrode to the fluid medium; however it sustains the high electric field at the interface

due to charge redistribution when a potential is applied.

The relationship between the applied electrical potential (V) and the resultant

surface tension (y) are expressed in Lippman's equation:

1
y =yo cV2 (1.11)
2

which when combined with the Young's equation:

YSL = YSG YLG CO (1.12)

can yield the resultant contact angle (0) according to Lippman-Young's equation:

1 1
cos= cos0o+ I- IcV2 (1.13)
YLG 2

Where yo and 90 are the surface tension and contact angle of the solid-liquid

interface when there is no electrical field across the interface layer (surface tension and









contact angle at point of zero charge), YSL, YSG, and YLG are the solid-liquid, solid-gas,

and liquid-gas surface tensions, V is the applied electrical potential, and c is the specific

capacitance of the dielectric layer (c = / t; F/cm2) where so is the permittivity of a

vacuum, P is the dielectric constant of the dielectric layer, and t is the dielectric layer

thickness.

The Lippman-Young's equation predicts that the overall AO can be increased by

increasing the applied electrical potential and the dielectric constant of the material being

used and by decreasing the dielectric layer thickness. Therefore the thinner the dielectric

layer used the lower the electrical potential required to induce a given AO.

However the dielectric breakdown voltage is proportional to the thickness of the

dielectric layer. Therefore at a certain dielectric thickness the required electrical potential

to induce a given AO will exceed the dielectric breakdown potential. Thus, there is a

thickness limit on the dielectric layer which is dependent on the dielectric material used.

An example of this was shown by Moon et al..86

Moon and coworkers calculated that the voltage required to induce a 400 AO for a

Teflon AF based EWOD device (s = 2.0 and Ebreakdown = 2x1016 V/cm ) as a function of

dielectric layer thickness. For dielectric layer thicknesses of less than 0.2 gm the voltage

required to induce a 400 change in 0 was higher than the dielectric breakdown potential

of the Teflon AF (see Figure 2.8).

In order to prevent the dielectric breakdown of thin film dielectrics Moon and

coworkers deposited a 700A layer of barium strontium titanate between the platinum

electrode and the 200A Teflon AF outer layer. This resulted in a EWOD device









capable of operating at electrical potentials as low as 15V and generating a AO on the

order of 400 (from 1200800).86


1 2 0 / . .-.. . . . . .
OiCekcric
S00 btbakdown
80 / ett,,ity ,~o;itb i
SIofor 120V O")
t6O
> 60



20 before EWOD

0 0.2 0.4 0.6 0.8 1.0
Thkicknss of Teflon* AF [hm]


Figure 2.8 Voltage required to induce a AO of 400 (from 1200-800) versus dielectric
layer thickness for Teflon AF based EWOD device, with F = 2.0 and
Ebreakdown = 2x1016 V/cm.86

2.5 Dynamic Surfaces

The volume expansion and contraction experienced by conducing polymers during

electrochemical switching between there oxidized and reduced states is also associated

with other physical property changes in the system. During the oxidation process the

developed positive charge induced along the CP backbone will also result in a change in

the surface charge or energy of the system. The influx of counter ions and associated

solvent that drives the volume expansion during oxidation also results in a drop in the

modulus of these materials. By incorporating CPs into other more flexible and durable

base materials (such as elastomers, thermoplastic elastomers, thermoplastics, etc.) the

dynamic properties of the CP can be utilized to produce coatings with dynamic surface

properties.









Changes in surface energy have been studied under both chemical 90,91 and

electrochemically 92, 93 induced redox systems for PPy, polyaniline (PANI), poly(3-

(pyrrolyl)-alkanoic acid), poly(3-octylthiophene) (P30T), and poly(3-hexylthiophene)

(P3HT). The results of these studies demonstrate that oxidizing the conducting polymer

decreases the Sessile drop contact angle (increase surface energy) of water.

An example of this behavior was reported by Gregory et al..91 They found that by

chemically oxidizing and reducing PANI, PPy, and P3HT contact angle changes of 36.5,

16.80, and 31.50 could be induced. However Bartlett et al. 94 stated that by using

functionalized thiophenes, i.e., alkanoic acids such as carboxylic, butanoic, and

pentanoic, a higher contact angle (AO= 80, 200, 300 respectively) is measured for the

oxidized state. This is attributed to the protonation of the alkanoic acids (pKa = 6.5, 5.8,

6.1 respectively) at a low pH (oxidizing environment).

By utilizing various conducting polymer systems it was hypothesized that it should

be possible to tailor particular systems to produce desired surface properties changes,

compatibilities, and redox switching speeds. The relative differences in oxidation

potential of various conducting polymers are shown in Figure 2.9.


PPy P3MeT PPP


I I I
(+) oxidized (neutral) reduced (.)

Figure 2.9 Relative surface charge of different conducting polymers.

2.5.1 Polypyrrole

To date a majority of the literature work conducted on conducting polymers has

been on polypyrrole (PPy) (Figure 2.11 A). PPy is characterized by high stability in its









oxidized form due to its oxidation potential -0.2 V which is close to the 02 reduction

potential at -0.2-0.3 V.95 Therefore neutral PPy will be oxidized by 02 to form its

oxidized conducting form when exposed to air. PPy has gained a lot of exposure recently

in the field of conducting polymer based artificial muscles due to its ability to produce a

large volume expansion (-1-3% longitudinally and -35% in thickness on a bound

surface49) when redox cycled between the oxidized and reduced state. This large volume

change induced during redox cycling will be beneficial in the modification of the PPy-

PDMS surface modulus. PPy can be synthesized chemically or electrochemically in

various media. The chemical polymerization can be facilitated in the presence of Lewis

Acids such as FeC13 or ammonium persulfate along with codopants such as NaC104.55' 96-
101

2.5.2 Poly(3-methylthiophene)

Polythiophene and its derivatives have received a lot of attention lately due to their

stability in the oxidized as well as reduced states.5' 39, 55, 90, 91, 95, 96, 102-108 They also

possess many highly desirable electrical, optical, and redox properties. The thiophene

monomer can easily be derivatized using a number of chemistries. It has been shown that

by changing the substituents on the thiophene ring the oxidation potential of the resulting

monomer and polymer can be varied between 1.20-2.00 V and 0.70-1.45 V

respectively.96 Poly(3-methylthiophene) (PMeT) (Figure 2.11 B) in particular has an

oxidation potential of -0.8 V and reaches a fully reduced state at -0.2 V vs Ag/AgC1.

These values lie well above the 02 reduction potential and above the H20 oxidation

potentials allowing for good stability in both forms. This can be seen vs. SCE (not

Ag/AgC1) along with the oxidation potentials for PPy and PPP in Figure 2.10.95 PMeT

can also be polymerized in a similar fashion as PPy.









oxidation reduction


H 20 0

I ^I
ppp PMeT PA PMeT
I dP IPP

1.2 0.8 0.4 0 -04

E- Va ll/ C*


Figure 2.10 Oxidation (-) and reduction (--) potentials of poly pyrrole (PPy), polyaniline
(PA), poly(3-methylthiophene (PMeT), and poly(p-phenylene) (PPP).95

2.5.3 Poly(p-phenylene)

One of the disadvantages of PPy is its high stability in its oxidized (charged) form.

This hinders the return of the PPy-PDMSe surface back to its original (unchargred)

surface energy. Poly(p-phenylene) (PPP) (Figure 2.11 C) on the other hand exhibits

exceptional stability in its neutral form. The oxidation potential of PPP is around +1.2 V

which is very close to the oxidation potential of water, therefore water will reduce the

oxidized form of PPP to its more stable neutral form.95 PPP is also characterized as being

highly crystalline, difficult to process, insoluble, and exhibiting high resistance to

oxidation, radiation, and thermal degradation.55 96 PPP can be synthesized from benzene

in the presence of Lewis acid such as FeC13 (-70 C) and AlC13 (-37 C) along with an
additional oxidizing agent (codapoant).55 96, 109-114
additional oxidizing agent (codapoant).












N


C


Figure 2.11 Monomer and polymer structures for A) polypyrrole, B) poly(3-
methylthiophene), and C) poly(p-phenylene).














CHAPTER 3
INITIAL IN SITUEVALUATION OF CP'S VIA STRAIN GAGE TECHNIQUE

3.1 Introduction

Strain sensitive actuators were developed utilizing commercially available strain

gage technology. This is the first time sensors of this type have been used in the in situ

evaluation of conducting polymer based actuators.56 Strain sensitive actuators were used

to evaluate the in situ strain response of PPy/TOS based actuator films during redox

cycling using cyclic voltammetry and square-wave potential stepping. The strain

responses of PEDOP and PBEDOT-Cz were also evaluated and compared to PPy/TOS.

It is believed that PPy's high strain response is due to its ability to crosslink through the

3, 4 positions on the monomer structure.64'104 By placing substituents in these positions

the degree of crosslinking can be controlled. PEDOP was chosen as a candidate for study

due to its blocked 3, 4 positions, which should produce a linear polymer structure with a

minimal degree of crosslinking. PBEDOT-Cz has been shown to undergo a non-

reversible oxidative reaction at about 1.15 V, which has been attributed to a possible

crosslinking reaction.107 This ability to potentially crosslink was the reason that

PBEDOT-Cz was chosen for comparison.

3.2 Materials and Methods

3.2.1 Materials

Pyrrole was purchased from Sigma-Aldrich and was filtered through neutral

alumina (Brockman activity 1; Fisher Scientific) until colorless before use to remove any

impurities. 3,6-bis(2-(3,4-ethylenedioxy)thienyl)-N-carbazole (BEDOT-Cz) and 3,4-









ethylenedioxypyrrole (EDOP) used in this study were synthesized by the Reynolds

Research group (University of Florida Chemistry Department).107 115,116 Acetonitrile

(ACN), lithium perchlorate (LiC104), sodium perchlorate (NaC104), p-toluenesulfonic

acid Na salt (NaTOS), and tetrabutylammonium were purchased from Sigma-Aldrich and

used as received. Sodium nitrate (NaNO3) was purchased from Mallinckrodt Chemicals

and sodium sulfite (Na2SO3) was purchased from Fisher Scientific, both were used as

received. Deionized water (18 MQ, Millipore system) was used in all experiments and

the solutions were deoxygenated by bubbling argon prior to electropolymerization and

redox switching of the CP.

3.2.2 Strain Gages

Electrical resistance strain gages (CEA-06-500UW-120 and EA-06-20BW-120;

Figure 2.4) were purchased from Vishay Measurements Group Inc. All strain gages were

cleaned prior to Au deposition and use with ethanol to remove any surface oils and

debris. These gages were treated with various Au treatments and then coated with the

appropriate CP to form strain sensitive actuators. A Vishay Measurements Group Inc.

P3500 strain indicator was used to measure the resistance changes in the strain gage and

thereby produce the strain measurement. EvAu coated strain gages (no CP) that were

exposed to electrolyte solutions and potentiostatically cycled gave a zero strain reading.

This was done to verify that the strain gages were unaffected by the potential cycling

during redox switching and that all strain measurements acquired were the result of the

CP's and not the strain gage.

3.2.3 Conducting Polymer Synthesis

In the initial study EvAu was vapor deposited thermally onto strain gages to a

thickness of ca. 700 A (measured by quartz microbalance). These gages were used as the









working electrode during electropolymerization and redox switching of the CP's. All

electrochemical work was carried out utilizing an EG&G Princeton Applied Research

273A potentiostat/galvanostat utilizing the CorrWare software package, a platinum foil

counter electrode and an Ag/AgCl (BAS MF2052) reference electrode at room

temperature. All potentials given are relative to this reference electrode. Conducting Cu

tape (1181, 3M) was utilized to make electrical connections between the EvAu layer and

the working electrode lead from the potentiostat. Polypyrrole (PPy) was synthesized

potentiostatically (E = 0.65 V, t = 500 s) from an aqueous solution of 0.2 M pyrrole, 0.1

M p-toluenesulfonic acid Na salt (NaTOS), and 1.0 M lithium perchlorate. This resulted

in a film thickness of ca. 9.6 [m measured by a Sloan Dektak 3030 profilometer.

PEDOP was synthesized potentiostatically (E = 0.6 V, t = 10,000 s) from an aqueous

solution of 0.01 M 3,4-ethylenedioxypyrrole, 0.1 M p-toluenesulfonic acid Na salt, and

1.0 M lithium perchlorate. The resulting film thickness was 10.6 im. PBEDOT-Cz was

synthesized potentiostatically (E = 0.8 V, t = 3000 s) from an aqueous solution of 0.01 M

Bis-EDOT-Cz, 0.1 M acetonitrile, and 1.0 M tetrabutylammonium. The corresponding

film thickness of PBEDOT-Cz was 9.9 im.

3.3 Polypyrrole (PPy/TOS) Results

3.3.1 Determination of Electropolymerization Conditions for PPy/TOS

In order to obtain the optimal strain response from the strain sensitive actuators the

PPy/TOS electropolymerization conditions were first evaluated on a gold-button working

electrode (MF-2014 Au electrode (AUE), Bioanalytical Systems, Inc.). PPy/TOS films

were prepared by cyclic voltammetry (-0.8 V to 1.0 V) in an aqueous solution of 0.2 M

pyrrole, 0.1 M NaTOS and 1.0 M LiC104. The pyrrole oxidation peak was observed at

0.65 V, which corresponds to the optimal electropolymerization voltage for these









electrochemical conditions. The adhesion of the PPy/TOS film to the gold-button

working electrode was tested using a tape (810, 3M) peel off method. PPy/TOS films

electropolymerized at a constant potential of 0.65 V produced the most adherent films,

while films polymerized at voltages exceeding 0.8 V exhibited very poor adhesion to the

electrode.

3.3.2 PPy/TOS Cyclic Voltammetry and Strain Response

Utilizing these electropolymerization conditions (E = 0.65 V) a PPy/TOS film was

deposited on a Au coated strain gage (CEA-06-500UW-120) for 500 seconds and resulted

in a film thickness of 9.6 [tm (Sloan Dektak 3030). PPy/TOS films have been shown to

produce films of high conductivity (150 S/cm) and tensile strength (73.4 MPa).61' 117 A

representative SEM micrograph of PPy/TOS is shown in Figure 3.1.





















Figure 3.1 SEM micrograph of surface morphology of PPy/TOS film prepared in 1.0 M
LiC104 at a potential of 0.65 V. SEM image taken of an uncoated sample at
1000X and 15 KeV.

PPy/TOS samples were scanned from -0.8 V to 0.4 V at 10 mV/s in 1.0 M aqueous

LiC104. The resulting strain response measured by the strain sensitive actuator followed









the cyclic voltammetry data (Figure 3.2). These results were obtained during the 5th

potential cycle to attain reproducible cyclic voltammetry and strain data. The strain

response was collected manually at 0.2 V intervals while the cyclic voltammogram was

acquired directly from the potentiostat. The PPy/TOS cyclic voltammetry (Figure 3.2a),

on the strain gage, is relatively broad due to the high surface area of the working

electrode (ca. 100 mm2) and large film thickness (9.6 [tm). The PPy reduction peak is

centered at -0.2 V. The overall change in strain (As) for the PPy/TOS systems was on the

order of 236 gE.

Upon further examination (Figure 3.2b) it is evident that a small initial decrease in

the strain response (contraction) is present, starting at -0.8 V, and is followed by a much

larger increases (expansion) in strain as the oxidation potential is reached. This is similar

to results previously published in literature 37 and is evident in all samples of PPy/TOS,

PEDOP, and PBEDOT-Cz under cyclic voltammetry. It is also evident that there is

hysteresis in the strain response. It is believed to be due to ion and solvent

concentrations, in the CP, never reaching equilibrium under these scan conditions (10

mV/s).44 This also causes the strain response to initially continue to rise on the reverse

scan, resulting in the maximum strain response to be observed at 0.2 V instead of 0.4 V.

Similar results have been obtained during electrochemical quartz crystal microbalance

(EQCM) experiments conducted with PPy. These results are attributed to the diffusion of

the dopant ions and the corresponding solvent molecules into and out of the polymer

structure during redox cycling.







43



A)

6 -




2-




-2

-4


B)


300


250-7

200

_0




150


so --------------
-1.0 -0 8 -0.6 -0.4 -02 0.0 0.2 0,4 0,6

E (V) versus AgAg C I
Figure 3.2 (a) PPy/TOS cyclic voltammetry (v = 10 mV/s) and (b) in situ strain response
of a 9.6 [pm film prepared in aqueous 1.0 M LiC104.

3.3.3 PPy/TOS Multi-Cycle Strain Response

Strain sensitive actuators were also used to acquire multi-cycle strain data as shown

in Figure 3.3. It is evident from the multi-cycle data that the strain response does not

return to its initial starting value, but has a positive drift. The difference in the initial

starting points for the 2nd and 5th cycle is on the order of 100 p, and decreases with each










cycle. However the overall change in strain for each cycle is reproducible on the order of

200 tgE. The drift in cycle measurements is attributed to the film possibly not returning to

the same state of solvation from cycle to cycle. It has also been suggested that the drift

could be part of the break-in process for these materials (i.e., the film remembers it is

bent and there is some molecular rearrangement). This break-in process is evident during

the 1st cycle (Figure 3.3) by the very erratic strain response, which starts to normalize

after a couple of scans.



400
-*-- \a9 cycle
-T- 2na cycle
300 ~- 'w cr '------
4 l1h cy-,e ". --h- -3_
hi cycle




J100 ----


0


-100



-1.0 -0.8 -0.6 -04 -0.2 0.0 02 0.4 0.6
E (V versusAg'AgCI
Figure 3.3 In situ multi-cycle cyclic voltammetry strain response of a 9.6 tE film
prepared in aqueous 1.0 M LiC104.

3.3.4 PPy/TOS Square-Wave Potential Experiments

Square-wave potential experiments were also utilized to evaluate the strain

response of PPy/TOS. PPy/TOS films were stepped from -0.8 V to 0.4 V and then back

to -0.8 V five times with a 50 s hold time at each potential (Figure 3.4). These stepping










experiments resulted in a reproducible As of 70 ps. The minimum and maximum strain

values were collected manually at -0.8 V and 0.4 V respectfully.




80
-*- Micro Strain (Ls) 0.6
Voltage (V)
0.4
60
S0.2

0.0
40

20 / I -0.42



-0.6

0 -0.8

... -1.0
0 100 200 300 400 500
Time (s)
Figure 3.4 In situ square-wave strain response of a 9.6 pm film prepared in aqueous 1.0
M LiC104.

The overall change in the strain response obtained by square-wave voltammetry is

less than that obtained by cyclic voltammetry; however the time required to complete one

cycle from each experiment is different. The time required to complete one cycle is 240 s

for cyclic voltammetry while the time required to complete one cycle by square-wave

stepping is 100 s. The strain produced by oxidative doping of these devices is diffusion

limited. Therefore longer cycle times allows more ion and solvent molecules to diffuse

into and out of these films (i.e., higher degree of oxidative doping), which produces a

higher overall degree of strain.









3.4 PEDOP Results

3.4.1 Electrochemical Analysis of PEDOP

Poly(3,4-ethylenedioxypyrrole) (PEDOP) was recently introduced as a conducting

polymer with a very low redox switching potential and a high stability to repeated redox

switching in aqueous electrolytes. 115' 11 Initial electrochemical evaluations were

conducted on a gold-button electrode as with PPy/TOS. PEDOP electropolymerized

potentiostatically from aqueous solutions of 0.01 M EDOP, 0.1 M NaTOS, and 1.0 M

LiC104 at 0.5 V produced nicely adherent and electroactive films. As with PPy/TOS,

PEDOP electropolymerized at potentials greater than or equal to 0.8 V produced film

with poor adhesion, these films spontaneously delaminated from the gold-button

electrode. PEDOP films produced at 0.5 V were electrochemically characterized by

cyclic voltammetry, scanned at 100 mV/s from 0.0 V to -1.2 V, in aqueous 1.0 M

LiC104. The resulting cyclic voltammetry is shown in Figure 3.5.

From the cyclic voltammetry it is evident that PEDOP's redox switching potential

is much lower than that of PPy/TOS. PEDOP has a measured half-wave potential (El/2)

of-0.6 V. This low redox switching potential is attributed to the electron donating

alkoxy substituents producing a highly electron-rich polyhetrocycle backbone. This

electron-rich nature produces high stability to air and aqueous electrolytes in the doped

and conducting form. This allows PEDOP to be held in the oxidized (conducting) form

for extended periods of time without degradation and also leads to the high cyclic

stability of this material.

3.4.2 PEDOP Multi-Cycle Strain Response

EDOP was electropolymerized potentiostatically at E = 0.60 V for 10,000 s from a

aqueous solution of 0.01 M EDOP, 0.1 M NaTOS, and 1.0 M LiC104 to produce a










PEDOP film of 10.6 tm on a Au coated strain gage. A representative SEM micrograph

of the surface morphology of PEDOP is shown in Figure 3.6. As can be seen the surface

morphology of PEDOP films is comparable to that of PPy. The strain response of

PEDOP during cyclic voltammetry was evaluated at a scan rate of 10 mV/s from -0.8 V

to 0.0 V in an aqueous solution of 1.0 M LiC104. The resulting strain response (Figure

3.7) is considerably less than that attained from PPy/TOS with a As that ranged from 36

FE to 47 sE with an average As of 42.6 Es for the 2nd through 10th cycles. The same data

has been replotted against a time axis, in Figure 3.8, to improve resolution of the strain

response from cycle to cycle.

13


0-2-


0-1 I


0-I
E e,





-0.2




-1.2 -I. -OL. -0.o -0.4 -0.2 OL.
E (V versw s AgAgCI
Figure 3.5 Cyclic Voltammetry (v = 100 mV/s) of a PEDOP film produced from aqueous
1.0 M LiC104 at E = 0.5 V and t = 200 s.







48



























Figure 3.6 SEM micrograph of a PEDOP film prepared at 0.6 V in 1.0 M LiC104. SEM
image taken of an uncoated sample at 1000X and 15 KeV.


60
S -0- 1 ,t .: -1

40 1- rd : I:
4 .th .: : I
h ,: -1


20 ------4---- --- -----h
-0-0-- *,h ,: 4,1









-40 ------



-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2

E (V) versus Ag/AgCI
20

















Figure 3.7 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of a 10.6 m:
PEDOP film in aqueous 1.0 M LiC104.
-20


-40


-60 ,
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2
E (V) versus Ag/AgCl
Figure 3.7 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of a 10.6 [tm
PEDOP film in aqueous 1.0 M LiC104.











60 0.2
-*-- Micro Strain (LEs)
Voltage (V)
40 0.0


20 -0.2 o


0 -0.4


-20 -0.6


-40 -0.8


-60 -, -1.0
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 3.8 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of 10.6 pm
PEDOP film in aqueous 1.0 M LiC104. Data from Figure 3.7 has been
replotted vs. time.

3.4.3 Effects of Cyclic Scan Rate on the Strain Response of PEDOP

The effects of scan rate were also evaluated. PEDOP films were

electropolymerized as previously stated and then scanned at 100 mV/s by cyclic

voltammetry from 0.0 V to -0.8 V in aqueous LiC104. The resulting strain response was

significantly less than that obtained at 10 mV/s. Due to the speed of the redox switching,

data was only collected at each end of the potential scan (i.e., 0.0 V and -0.8 V). The

average As was 6.6 Es for the first 10 cycles (Figures 3.9 and 3.10). However iot should

be noted that the 1st and 6th cycle both produced a As of -1 Es while the rest of the cycles

all had As values between 7 Es and 9 aE. This dramatic decrease in the strain response is

expected due to the diffusion limited swelling nature of this and other conducting

polymers. The scan rate of this experiment was conducted 10 times faster than the

previous experiment, thus not allowing for full oxidative doping of the polymer network.







50





i st ccde
6 3l 2ncycle
-v- 3rd cyde
-w- 4th cycle
4 5th cycle
Sth cycle
S-- 72th cyc











-1 -0.8 o0.6 -.C4 -02 0.0 0.2

E (V) versus ,iAg Cl
-1.0h cyc--
LaV0 A 10th cyclet












Figure 3.9 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of a 10.6
km PEDOP film in aqueous 1.0 M LiC104.



8 02
-- Micro Strin (OZ)
6 --Volnge N,


4


D



.:4 | I II




I -
-0.

















-, -r r -- -1.0

0 20 40 60 80 100 120 140 160
Time (s)
Figure 3.10 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of 10.6
km PEDOP film in aqueous 1.0 M LiC104 replotted vs. time.









3.4.4 PEDOP Square-Wave Potential Experiments

Square-wave potential experiments were also conducted on PEDOP. Using the

electropolymerization conditions stated above a 10.6 [m film was stepped from -0.8 V to

0.0 V with a 50 s equilibrium hold at each voltage. The resulting square-wave strain

response exhibited a As that varied from 11 E, to 33 E, with an average value of 21 jEs.

The strain response with the corresponding potential sweep is plotted versus time in

Figure 3.11.

As stated earlier, it is believed that a cross-linked structure, like PPy, will develop a

higher degree of strain/swelling than a non cross-linked structure. The cross-links act to

tie the whole polymer network together; therefore a small change (strain due to swelling)

in one section of the network will exert an affect (strain) on the rest of the networked

structure.

35 .2
-- Micro Strain (IE)

0,2




S15 -- 0-4
0
i- / \/ \/\ l\1I





-1-08


0 100 200 300 -00 500
Time (s)
Figure 3.11 In situ square-wave strain response of a 10.6 tp film prepared in
aqueous 1.0 M LiC104.









During anion insertion (oxidation) the cross-linked polymer network would be

highly strained due to inability of the polymer chains to separate/relax through simple

chain motion. By blocking the 3- and 4- positions on the pyrrole ring PEDOP cannot

form cross-links and forms a linear structure through the 2- and 5- positions. This lack of

cross-linking would allow the polymer network to relax through simple chain motions

during ion transport into and out of the structure (during redox switching). Therefore

anion insertion into the polymer network would result in a much lower degree of overall

strain than in a cross-linked network. However with most things, this can be overdone.

As the cross-link density of the network increases the stiffness of the network also

increases. Thus requiring an increased amount of force (potential driving force) to

develop the same amount of strain in the system. As the cross-link density of the

polymer network increases the diffusion rate of counter ions in the system will decrease.

Also as the cross-link density increases the modulus increases and the film can become

too rigid to deflect/strain resulting in zero strain development and therefore no actuation.

3.5 PBEDOT-Cz Results

3.5.1 PBEDOT-Cz Introduction

Poly [3,6-2(2-(3,4-ethylenedioxythienyl)-carbazole] (PBEDOT-Cz) and other bis-

EDOT derivitized carbazoles were found to be of interest, by Reynolds et al., for there

ability to form three distinct redox states (neutral, cation-radical, and di-cation forms) at

low potentials and to be stable to thousands of redox switches.107 These low switching

potentials are attributed to the electron-rich biEDOT structure in the PBEDOT-Cz

polymer repeat structure. Its was found during electrochemical analysis that a

irreversible higher potential redox process at 1.15 V (Figure 3.12) was present during the

first cyclic voltammetry scan but was absent for all subsequent scans. This process could









not be attributed to over oxidation (breakdown) of the polymer backbone. This was

theorized by Reynolds et al. to be a possible cross-linking reaction, however no

spectroscopic evidence of cross-linking has been reported to date.

PBEDOT-Cz was chosen as a candidate for evaluation due to this possible cross-

linking reaction. It was theorized that the cross-linking of PBEDOT-Cz would lead to an

improved strain response in the material when compared to non-cross-linked materials.

This was based on the observation that PPy's strain response is related to its crosslinked

64, 104
structure.64104



BO-







-2nd scan





Potential cross-linking rxn (1st scan)



1.2 1, 02 U, 04 0.2 ,0 U G-a *06 1,1

Figure 3.12 Cyclic voltammetry (100 mV/s) of PBEDOT-Cz in 0.1 M TBAP/CAN.
Adapted from Figure 8 of Reynolds et al..107

3.5.2 PBEDOT-Cz Electrochemical Conditions

The effects of possible cross-linking at 1.15 V 107 in PBEDOT-Cz was evaluated in

samples that were electropolymerized from a solution of 0.01 M BEDOT-Cz and 0.1 M

tetrabutlyammonium perchlorate in acetonitrile. Samples were cycled from -0.8 V to 0.6,









1.0, and 1.2 V at 10 mV/s in 1.0 M aqueous LiC104 successively. Data was collected for

the 1st, 2nd, 5th, and 10th scans at 0.2 V intervals.

3.5.3 PBEDOT-Cz Strain Response (-0.8 V to 0.6 V)

The cyclic strain response acquired from -0.8 V to 0.6 V exhibited a very irregular

shape for the first four scans but then normalized after the 5th scan. This is due to the

normal break in period for these materials, as discussed earlier for PPy and PEDOP. The

normalized cyclic strain data for the 1st, 2nd, 5th, and 10th cycles (Figure 3.13) exhibits a

As range from 15 wts to 21 wts. The average change in strain was 17[ts.


-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8
E (V) versus Ag/AgCI
Figure 3.13 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film (-
0.8 V to 0.6 V) for the 1st, 2nd, 5t, and 10th scans in 1.0 M LiC104.













30

20 -I -- 5.h ,: I
-20I- C0th ,: 6 ,: ----'"__-_
u.-.- --------.. ,

10

0 ..., .


-o .
10


-20


-3 0 .
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

E (V) versus Ag/AgCI
Figure 3.14 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film (-
0.8 V to 1.0 V) for the 1st, 2nd, 5th, and 10th scans in 1.0 M LiC104.

3.5.4 PBEDOT-Cz Strain Response (-0.8 V to 1.0 V)

The cyclic strain data acquired from -0.8 V to 1.0 V was very regular in shape and

exhibited its max strain at -0.2 V on the reverse scan as shown in Figure 3.14. The As

measured for this set of scans ranged from 28 sE to 37 tse with an average value of 33 Es.

This is a change of 16 ps and 0.6 V between the average As measured for the -0.8 V to

0.4 V and -0.8 V to 1.0 V scans.

3.5.5 PBEDOT-Cz Strain Response (-0.8 V to 1.2 V)

The strain data acquired from -0.8 V to 1.2 V is also very regular in shape when

compared to the scans obtained from -0.8 V to 0.4 V. This set of data was obtained

above the potential cross-linking potential of 1.15 V for PBEDOT-Cz, however there are

no significant visual differences in the shape or measured values of the strain data from

above and below this potential cross-linking voltage. The As ranged from 34 ts to 35 ts










with an average value of 34.75 gE for the 1st, 2nd, 5th, and 10th scans. These scans are

shown in Figure 3.15. This is a change of 1.75 pE and 0.2 V between the average As

measured for the -0.8 V to 1.0 V and -0.8 V to 1.2 V scans. This data does not support

the argument of potential crosslinking of PBEDOT-Cz at potentials above 1.15 V.




30
--- ,nd ,: .: "F
5- t[:h .: .. -
20 -- hlOth,: -
m.3

10 : : ~

o '

u ,.


-10 -



-20 ......
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
E (V) versus Ag/AgCI
Figure 3.15 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film (-
0.8 V to 1.2 V) for the 1st, 2nd, 5t, and 10th scans in 1.0 M LiC104.

3.5.6 Overall Results for PBEDOT-Cz

The overall change in strain for the 0.6, 1.0, and 1.2 V scans ranged from 15-21,

28-37, and 34-35 tE, with average values of 17, 33, and 34.75 tE respectively. These

values are significantly lower than those obtained from PPy. However there is no

significant difference in the data obtained from the -0.8 V to 1.0 V and -0.8 V to 1.2 V

scans (below and above potential cross-linking voltage). The normalized strain response

of the 5th and 10th cycles for all three potential ranges is shown in Figures 3.16 and 3.17

respectfully.











The slight increase in the strain response between the scans from -0.8 V to 1.0 V

and 1.2 V is most likely due to the increased driving potential of 0.2 V. This combined

with the fact that no spectroscopic evidence of cross-linking for PBEDOT-Cz could be

found by Reynolds et al.,107 leads to the conclusion that the PBEDOT-Cz cross-linking

reaction is non-existent or so minimal that no enhancement in the strain production is

evident. However the overall strain response of PBEDOT-Cz is comparable to that of

PEDOP. This is due to the inability of PBEDOT-Cz to crosslink as with PEDOP. This

low strain response combined with the cost to produce the BEDOT-Cz monomer makes

PBEDOT-Cz a poor candidate for actuator construction.





30
--- -0 -...v tc: .


20



'-10 -
V
CO *
0 1 Y '
S.0
0 S
0 --



-10



-2 0 .
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

E (V) versus Ag/AgCI
Figure 3.16 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film
scanned from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (5th scan) in 1.0 M LiC104.













30

-0 4. ..v 1 0,.
-U- -0 .'. .. t 1 : Ov
20--i---------.---.----t,--,----* ''...'9 -.



",, ,-
20




S10

0
U 0- S-





*
-20 ---------------------------------

-20
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
E (V) versus Ag/AgCI
Figure 3.17 Normalized in situ cyclic strain response of 9.9 |tm PBEDOT-Cz film
scanned from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (10th scan) in 1.0 M LiC104.

3.6 Overall Comparison of PPy, PEDOP, and PBEDOT-Cz

If the strain data for PPy/TOS, PEDOP, and PBEDOT-Cz is normalized and plotted

on the same graph it is easy to see that PPy/TOS produced significantly more strain than

PEDOP and PBEDOT-Cz (Figure 3.18). It is believed that the higher strain response of

PPy/TOS over PEDOP and PBEDOT-Cz is due to its ability to crosslink during

electropolymerization.64 104

The possible ability of PBEDOT-Cz to cross-link at a potential of 1.15 V has been

shown to be non-existent or at least so minimal that it is not significant in the strain

production of this system.

From this study it was shown that it is possible to obtain a detailed strain response

measurement of various CP's directly and precisely utilizing standard strain gages

technology to construct strain sensitive actuators. These gages determine strain by










directly measuring the strain induced on the embedded constantan grid (inside the

actuator) not the overall deflection of the actuator tip like most measurement techniques.

This provides a detailed measurement of the actual strain produced internally in the

system.




250


200 -


S 150

CU
" 100

50


0


-50
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

E (V) versus Ag/AgCI
Figure 3.18 Comparison of in situ cyclic strain response of PPy, PEDOT, and
PBEDOT-Cz in aqueous 1.0 M LiC104.

3.7 Effects of Interlayer Adhesion

Adhesion of the CP to the EvAu coated substrate was found to be a potential

problem during long term repeated redox switching. In our study PPy, PEDOP, and

PBEDOT-Cz were all found to have poor adhesion to a gold-button working electrode

when electropolymerized at potentials > 0.8 V. However large-scale (>70%)

delamination of PPy was seen after long term cycling (>500 cycles) of actuators (Figure

3.19) and one film completely delaminated from a Au coated polyimide sample when

exposed to a high vacuum (- 1 105 Torr) during SEM examined (Figure 3.20).


-* FF PiP 04 I
". -y-- PED-,.P 1 0..
-*- PFBEDCT-C. i 1 0 '...
SF'BEDC ',T-C. ii 2 '...
* S 1I 1__'_









Examination of the image of the sample that delaminated in the SEM, it is obvious

that the PPy surface in contact with the Au layer is porous and irregular in texture

(Figures 3.21 and 3.22), which leads to poor interlayer adhesion between the PPy and the

Au substrate. Also regions of small nodules or nucleation sites of PPy were left behind

on the Au substrate during the delamination process (Figures 3.23 and 3.24). The same

features are also evident in SEM's of the underside of a PPy film that cracked and

delaminated during repeated redox switching.

The partially delaminated PPy film was rolled back to expose the surface that was

in contact with the Au substrate for further examination by SEM (Figure 3.25). The

adhesion of the PPy nodules to the Au substrate was non-homogenous upon

delamination. From closer examination it can be seen that these nodules grow at the

surface of the PPy film, and either lie directly on the surface or in some of the many

pores present at PPy-Au interface (Figure 3.26).

Delamination of the CP film most likely initiates in isolated regions with high

stress concentrations that eventually propagate into total delamination. As delamination

of the CP film from the substrate propagates a loss of electrical and physical contact

between the CP and the working electrode develops. This decreases the ability to redox

switch the CP film, thus retarding swelling/oxidation of the CP and therefore the

development of strain in these systems. This combined with a decreasing amount of

physical contact between the CP and the substrate reduces the overall movement and the

total lifetime of actuators made from these systems. Once total delamination of the CP

from the working electrode occurs no strain can be developed by the system and the

actuator will totally stop functioning.









3.8 Conclusions

From this study it was shown that strain gage technology can readily be utilized as

an inexpensive and highly accurate method of evaluating the resultant physical properties

of different electrochemical polymerization and actuation conditions on a given

conducting polymer structure. And due to the vast array of different strain gage designs it

is possible to incorporate this technology into almost any actuator design.

It has also been shown that PPy produced significantly higher strains than PEDOP

and PBEDOT-Cz under the given electrochemical conditions. PBEDOT-Cz also

produced higher strain responses around and above the potential crosslinking potential of

-1.15 V than at lower potentials. However it is believe that the increased strain response

is due to higher applied driving potentials than the possible crosslinking reaction due to

the lack of any electrochemical, spectroscopic, or strain/performance evidence.

It has also been determined that the interfacial adhesion of the conducting polymer

to the EvAu coated surface is a potential problem for long term actuator lifetimes.

During repeated electrochemical cycling the induced cyclic strains initiate micro-crack

formation between the CP and EvAu layers which then grow and eventually result in

device failure. All utilized conducting polymers (PPy, PEDOP, and PBEDOT-Cz) were

found to exhibit poor adhesion to EvAu when electropolymerized at potentials of 0.8 V

and greater. The most obvious case of this was for PPy which was shown to undergo

large-scale delamination from the EvAu substrate when subjected to long term redox

switching (Figure 3.19) and upon exposure to high vacuum during SEM examination

(Figure 3.20). It is important to improve interlayer adhesion to increase CP actuator

performance and overall lifetime.































Figure 3.19 SEM micrograph of delamination of PPy resulting from long-term cycling
of the actuator: 150X


Figure 3.20 SEM micrograph of delamination of PPy resulting from exposure to high
vacuum; 25X






























Enlarged SEM micrograph of region "A" in Figure 4.20; 1350X


SEM micrograph of porous PPy at PPy-Au interface; 5000X


Figure 3.21


Figure 3.22




























Figure 3.23 Enlarged SEM micrograph of region "B" in Figure 4.20; 100X


Figure 3.24 SEM micrograph of PPy nodules remaining of Au substrate after PPy
delamination; 1000X




























Figure 3.25 SEM micrograph of PPy surface after delamination from Au substrate
during long-term repetitive cycling of the actuator; 250X


Figure 3.26 SEM micrograph of exposed PPy-Au interface exhibiting PPy nodule
growth; 1000X














CHAPTER 4
IN SITU STRAIN MEASUREMENTS OF CP'S ON ENHANCED AU SURFACES

4.1 Introduction

Adhesion of CP's to EvAu coated substrates was identified as a potential problem

during the previous study. Poor adhesion of the CP to the substrate will lead to low or no

strain development and will greatly reduce the working lifetime for these actuators due to

eventual delamination of the CP from the substrate. PPy, PEDOP, and PBEDOT-Cz

were found to exhibit poor adhesion to Au when electropolymerized on Au-button

electrodes at potentials of 0.8 V and greater. PPy was also shown to undergo large-scale

delamination when subjected to long term redox switching (Figure 3.19) and upon

exposure to high vacuum during SEM examination (Figure 3.20). The purpose of this

study was to improve the interfacial adhesion between the conducting polymers and the

substrate. Further characterization of the effects of various polymerization and actuation

conditions on these systems was conducted on both standard and enhanced interfacial

surfaces using the strain sensitive actuator technology.

4.2 Materials and Methods

4.2.1 Materials

Pyrrole was (Sigma-Aldrich) was filtered through neutral alumina (Brockman

activity 1; Fisher Scientific) until colorless before use to remove any impurities.

Acetonitrile (ACN), lithium perchlorate (LiC104), sodium perchlorate (NaC104), p-

toluenesulfonic acid Na salt (NaTOS), and tetrabutylammonium were all used as received

from Sigma-Aldrich. Sodium nitrate (NaNO3) was purchased from Mallinckrodt









Chemicals and sodium sulfite (Na2SO3) was purchased from Fisher Scientific, both were

used as received. Deionized water (18 MQ, Millipore system) was used in all

experiments and the solutions were deoxygenated for 15 minutes by bubbling argon prior

to electropolymerization and redox switching of the CP.

4.2.2 Electrochemical Gold Deposition Solution

Electrochemical deposition of Au (EcAu) on thermally evaporated Au (EvAu)

coated polyimide substrates utilized a solution of Oromerse SO Part B replenisherr"

(Na3Au(SO3)2), a commercially available gold plating solution purchased from Technic

Inc. EcAu deposition was carried out in a solution of 25% (10 mL) Na3Au(SO3)2 and

75% (30 mL) of aqueous 1.7 M Na2SO3 as reported for electroless Au deposition.119

4.2.3 Evaporated and Electrochemically Deposited Gold

Thermally evaporated Au (EvAu) was vacuum deposited on the bottom side of the

strain gages to the desired thickness. All electrochemical work was carried out utilizing

an EG&G Princeton Applied Research 273A potentiostat/galvanostat utilizing the

CorrWare software package, a platinum foil counter electrode and an Ag/AgCl (BAS

MF2052) reference electrode at room temperature. All potentials given are relative to

this reference electrode.

Conducting Cu tape (1181, 3M) was utilized to make electrical connections

between the EvAu layer and the working electrode lead from the potentiostat.

Electrochemically deposited Au (EcAu) samples were then prepared on EvAu samples

potentiostatically from a diluted solution of Na3Au(SO3)2 (mentioned above) at -0.9 V.

EcAu layer thickness (and therefore surface roughness) was controlled by varying the

cathodic charge during electrochemical deposition.









The surface roughness factor (r) of the Au layer was determined by taking the ratio

of the electrochemical area, determined from the charge required to reduce the surface

oxide layer (potential cycling between 0.0 V and 1.5 V in aqueous 50 mM H2S04),120 to

the geometric area. The surface roughness factor varied from r = 2.89 for EvAu to r =

6.17 to r = 24.5 for the EcAu samples. Surface morphology was examined by utilizing a

scanning electron microscope (JEOL 6400) and white light optical profilometry

(Wyko/Veeko NT1000).

4.2.4 Conducting Polymer Synthesis

In this study EvAu was thermally vapor deposited to a thickness of ca. 1.0 [tm

(measured by cross-section SEM), some EvAu samples when subsequently treated with

EcAu of varying thicknesses (-1-10 [m) as described above. PPy films of varying

thicknesses were prepared potentiostatically on EvAu and EcAu samples from a solution

of 0.1 M pyrrole in 0.1 M aqueous sodium perchlorate at E = 0.9 V unless otherwise

mentioned.

The high surface roughness of EcAu samples made it difficult to determine PPy

film thickness directly; therefore polymerization charge density (C/cm2) was utilized to

compare PPy film thicknesses in this study. The polymerization charges ranged from

0.35 to 5.42 C/cm2. By assuming a 100% efficiency for the PPy deposition reaction and

a charge thickness ratio of 0.28 C/(_m cm2) 121 this correlates to a PPy film thickness of

-1.25-19.40 pm on a smooth surfaces.

4.3 Improved Interlayer Adhesion Utilizing Electrochemically Deposited Au
Surfaces (EcAu)

Interlayer adhesion between PPy and the EvAu coated substrate has been enhanced

by the electrochemical deposition of Au (EcAu) onto the EvAu surface. The EcAu









treatment increases the surface area of the working electrode via the growth of Au

crystals on the EvAu surface. EcAu films were deposited on EvAu coated PI substrates

(EcAu/EvAu/PI samples) potentiostatically at -0.9 V from a solution consisting of 25%

Na3Au(SO3)2 (Oromerse SO Part B replenisherr") and 75% aqueous 1.7 M Na2SO3.

The EcAu layer thickness (Au crystal height), and therefore surface roughness (r),

was initially controlled by limiting the deposition time on smooth PI samples and then

was later controlled by monitoring the total charge passed during the electrochemical

deposition process for all strain gage samples. The surface roughness factor (r), of the

Au surfaces, was determined by taking the ratio of the surface area measured

electrochemically, obtained from the charge required to reduce the surface oxide layer,120

to the geometric surface area of the working electrode. The electrochemical surface area

was measured by potential cycling the Au surface between 0.0 V and 1.5 V in an aqueous

50 mM H2SO4 solution. The charge passed during re-oxidation of the Au oxide

monolayer was converted to the surface area using a factor of 0.43 mC/cm2.












20 10 -

I10 8


oS =
2 Z



0 2 4 5
EcAu Deposition C charge (Crk m)
Figure 4.1 Correlation between surface roughness factor, nominal EcAu thickness and
EcAu deposition charge. EcAu thickness was determined by cross-section
SEM.

Figure 4.1 illustrates the relationship between the surface roughness factor (r) and

EcAu deposition charge. The EcAu layer thicknesses were measured by cross-section

SEM.

4.4 Effects of Surface Roughness on EcAu Morphology

Many factors such as electrolyte type, pH, deposition potentials, additives, and

substrate structure (surface texture) have all been shown to affect the resulting surface

morphology of electrochemically deposited Au surfaces.122-126 The effects of substrate

structure on EcAu morphology can easily be seen from SEM micrographs of the EcAu

deposited on smooth EvAu/PI substrates versus the rough EvAu/PI surface of the strain

gages. EcAu growth on smooth EvAu/PI surfaces results in the formation of large 5-

point start structures. The Au crystals grow about normal to the EvAu/PI surface on the

smooth samples. Crystal growth starts at many small nucleation sites. As the crystals

continue to grow they coalesce to form fewer but larger crystal structures.









However the application side (backside) of the PI strain gages is roughened

(sanded) to help promote adhesion of the strain gage to the sample it is being applied to.

This roughened surface (surface roughness factor (r) = 2.89) produces very irregularly

shaped Au crystal structures when grown to high charge densities (3.67 C/cm2, r = 18.90

and 4.72 C/cm2, r = 24.50). At lower charge densities (0.24 C/cm2, r = 6.17 and 0.71

C/cm2, r = 10.04) the Au crystal structures are very regular and resemble these obtained

on the smooth EvAu/PI samples.

However, the irregular surface of the strain gages causes the Au crystals to grow

off the normal axis causing the crystals to collide at random angles and coalesce to form

spiky tree-like structures when grown to high deposition charges. SEM micrographs

(4000X) of the different Au substrate surface roughness generated by varying the EcAu

deposition time between 0 and 60 minutes on smooth EvAu/PI samples are shown in

Figures 4.2 4.6. The corresponding SEM micrographs of Au surfaces deposited on

roughened EvAu/PI strain gages are shown in Figures 4.7 4.11. The topographies are

very different from those obtained by chemical deposition of Au from a similar solutions

on glass and polycarbonate substrates.127 128































/Au deposited on smooth polyimide (PI); 4000X


Figure 4.3 SEM micrograph of 3 minute EcAu deposition on smooth EvAu/PI; 4000X


Figure 4.2































IVI micrograpn or lu minute LCAU aepositon on smooin LVAU,


minute EcAu deposition on smooth


Figure 4.4


; 4000X


Figure 4.5


/Au/PI; 4000X





























Figure 4.6 SEM micrograph of 60 minute EcAu deposition on smooth EvAu/PI; 4000X





















Figure 4.7 SEM micrograph of EvAu (r = 2.89) deposited on rough PI strain gage;
4000X





























Figure 4.8 SEM micrograph ofEcAu (r = 6.17, 2.5 min.) deposited on rough EvAu/PI
strain gage; 4000X


Figure 4.9 SEM micrograph ofEcAu (r
strain gage; 4000X


10.04, 10 min.) deposited on rough EvAu/PI




Full Text

PAGE 1

DYNAMIC ANTIFOULING STRUCTURES AND ACTUATORS USING EAP COMPOSITES By CLAYTON CLAVERIE BOHN JR. A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Clayton Claverie Bohn Jr.

PAGE 3

This dissertation is dedicated to my loving parents Jack and Kathy Higgins and Clay and Susan Bohn for all there guidan ce and support thr oughout my life.

PAGE 4

ACKNOWLEDGMENTS I would like to thank my advisor and committee chair, Dr. Anthony Brennan, for providing me with the opportunity to further develop my understanding of polymers and engineering as a science and an art. He has provided infinite patience, guidance, and support throughout this project. He has continually challenged me as a researcher and a man to grow and find solutions that are not always obvious. I would like to thank the rest of my committee, Dr. Ronald Baney, Dr. Chris Batich, Dr. Amelia Dempere, and Dr. John Reynolds, for their time and valued critique of this dissertation. In addition I would like to further thank Dr. Reynolds for his support, guidance, and extensive knowledge of conducting polymers and electrochemistry during this project. I would also like to thank him for providing me the use of his electrochemical equipment and the knowledge of his graduate students. I would especially like to thank my coworkers on this project, Dr. Said Sadki and Dr. Myoungho Pyo, for being there as friends, coworkers, and teachers. Dr. Sadki provided the critical knowledge of electrochemistry necessary to get this project started. Dr. Pyo provided the exceptional skill and knowledge of the subject necessary to develop the EcAu systems and to get the project to where it is now. To them I owe all my knowledge of conducting polymers and electrochemistry, and without their help and guidance throughout this project none of this would have been possible. I would like to thank Dr. Elisabeth Smela for her continued guidance and support on the development of the EcAu systems. I would also like to thank Dr. Peter Ifju for help on developing the strain sensitive actuator technology. Dr. Ifju provided extensive iv

PAGE 5

knowledge and support on the proper use and application of strain gages and initial strain gage equipment usage. I would also like to especially thank my past and present group members. They have provided me with invaluable knowledge, assistance, and friendship during my years in the Brennan research group. Jeanne McDonald and Jeremy Mehlem provided exceptional help and guidance during my initial year in the research group. Both of them have been invaluable as coworkers and friends. I would like to thank them and their spouses for providing me with a very enlightening and enjoyable start in this group. I would like to give special thanks to Michelle Carmen, Thomas Estes, Adam Feinberg, Amy Gibson, Brian Hatcher, Nikhil Kothurkar, Chuck Seegert, Jim Schumacher, Leslie Wilson, and Wade Wilkerson who have proven to be exceptional friends and coworkers. I would also like to thank the rest of my research group for their support: Kenneth Williams and Kiran Karve. I would also like to thank the rest of my friends and colleagues in the department that have provided me with support: Brett Almond, Brian Cuevas, Iris Enriquez, Brent Gila, Josh Stopek, Dan Urbaniak, and Amanda York. Out of this group I would especially like to thank Jamie Rhodes and Paul Martin for also providing me with spare parts and expertise needed to help build and keep equipment running. I would also like to especially thank Jennifer Wrighton; with out her this whole place would probable fall apart and no work would ever get done. She had been a good friend and a great secretary to me and the rest of the group for the past 4 years. I would also like to thank my family and friends for their love and support throughout my life. I would especially like to thank my brother David for always being there and providing a comical and uplifting side to life. I wish him luck in fatherhood v

PAGE 6

and all that he does. And I would again like to thank him for always being a true brother and friend. vi

PAGE 7

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ............................................................................................................xii LIST OF FIGURES .........................................................................................................xiii ABSTRACT ...................................................................................................................xxiv 1 INTRODUCTION........................................................................................................1 2 BACKGROUND..........................................................................................................6 2.1 Electroactive Polymer Actuators ............................................................................6 2.1.1 Introduction to EAPs ....................................................................................6 2.1.1 Conducting Polymers .................................................................................10 2.1.2 Conducting Polymer Synthesis ...................................................................11 2.1.3 Conducting Polymer Actuators ..................................................................13 2.2 Electrical Resistance Strain Gages .......................................................................16 2.2.1 Strain Gage Theory .....................................................................................18 2.2.2 Strain Gage Materials and Construction .....................................................20 2.2.3 Strain Gage Accuracy .................................................................................23 2.3 Anti-Fouling/Foul-Release Coatings ....................................................................25 2.4 Electrowetting .......................................................................................................30 2.5 Dynamic Surfaces .................................................................................................33 2.5.1 Polypyrrole .................................................................................................34 2.5.2 Poly(3-methylthiophene) ............................................................................35 2.5.3 Poly(p-phenylene) ......................................................................................36 3 INITIAL IN SITU EVALUATION OF CPS VIA STRAIN GAGE TECHNIQUE.38 3.1 Introduction...........................................................................................................38 3.2 Materials and Methods.........................................................................................38 3.2.1 Materials.....................................................................................................38 3.2.2 Strain Gages................................................................................................39 3.2.3 Conducting Polymer Synthesis...................................................................39 3.3 Polypyrrole (PPy/TOS) Results............................................................................40 3.3.1 Determination of Electropolymerization Conditions for PPy/TOS............40 3.3.2 PPy/TOS Cyclic Voltammetry and Strain Response..................................41 vii

PAGE 8

3.3.3 PPy/TOS Multi-Cycle Strain Response......................................................43 3.3.4 PPy/TOS Square-Wave Potential Experiments..........................................44 3.4 PEDOP Results.....................................................................................................46 3.4.1 Electrochemical Analysis of PEDOP.........................................................46 3.4.2 PEDOP Multi-Cycle Strain Response........................................................46 3.4.3 Effects of Cyclic Scan Rate on the Strain Response of PEDOP................49 3.4.4 PEDOP Square-Wave Potential Experiments............................................51 3.5 PBEDOT-Cz Results............................................................................................52 3.5.1 PBEDOT-Cz Introduction..........................................................................52 3.5.2 PBEDOT-Cz Electrochemical Conditions.................................................53 3.5.3 PBEDOT-Cz Strain Response (-0.8 V to 0.6 V)........................................54 3.5.4 PBEDOT-Cz Strain Response (-0.8 V to 1.0 V)........................................55 3.5.5 PBEDOT-Cz Strain Response (-0.8 V to 1.2 V)........................................55 3.5.6 Overall Results for PBEDOT-Cz...............................................................56 3.6 Overall Comparison of PPy, PEDOP, and PBEDOT-Cz.....................................58 3.7 Effects of Interlayer Adhesion..............................................................................59 3.8 Conclusions...........................................................................................................61 4 IN SITU STRAIN MEASURMENTS OF CPS ON ENHANCED AU SURFACES66 4.1 Introduction...........................................................................................................66 4.2 Materials and Methods.........................................................................................66 4.2.1 Materials.....................................................................................................66 4.2.2 Electrochemical Gold Deposition Solution................................................67 4.2.3 Evaporated and Electrochemically Deposited Gold...................................67 4.2.4 Conducting Polymer Synthesis...................................................................68 4.3 Improved Interlayer Adhesion Utilizing Electrochemically Deposited Au Surfaces (EcAu).....................................................................................................68 4.4 Effects of Surface Roughness on EcAu Morphology...........................................70 4.5 Electrochemical Deposition of PPy......................................................................77 4.6 Improved PPy Adhesion to EcAu Treated Surfaces.............................................82 4.7 IN SITU PPy/EvAu/PI Actuator Results...............................................................83 4.7.1 Counter Ion Effects on PPy Strain Response.............................................83 4.7.2 Effects of Potential Limiting on PPy Strain Response...............................85 4.7.3 Effects of PPy Film Thickness on Strain Response...................................86 4.7.4 Effects of Polymerization Potential on PPy Strain Response....................90 4.8 IN SITU PPy/EcAu/EvAu/PI Actuator Results....................................................91 4.8.1 Effects of PPy electropolymerization charge and surface roughness factor92 4.8.1.1 Effects of surface roughness factor..................................................92 4.8.1.2 Effects of PPy electropolymerization charge on strain (PPy film thickness)................................................................................................95 4.8.1.3 Combined effects of PPy electropolymerization charge and surface roughness factor......................................................................................97 4.8.2 Counter Ion Effects on PPy Strain Response.............................................98 4.8.3 Frequency response....................................................................................99 4.8.4 Lifetime....................................................................................................100 4.9 Conclusions.........................................................................................................103 viii

PAGE 9

5 RAPID ELECTRODE PATTERNING FOR USE IN ADVANCED CONDUCTING POLYER ACTUATORS AND DYNAMIC SURFACES.......................................104 5.1 Introduction.........................................................................................................104 5.2 Rapid Electrode Patterning Techniques..............................................................105 5.2.1 Rapid Electrode Patterning.......................................................................105 5.2.2 Line Patterning.........................................................................................106 5.3 Linear Actuators.................................................................................................108 5.3.1 Background...............................................................................................108 5.3.2 Cantilever based linear actuators..............................................................109 5.4 Actuator Based Dynamic Surfaces.....................................................................119 5.5 Conclusions.........................................................................................................125 6 PDMSe BASED DYNAMIC NON-TOXIC ANTI-FOULING SURFACE COATINGS..............................................................................................................127 6.1 Introduction.........................................................................................................127 6.2 Materials and Methods.......................................................................................129 6.2.1 Materials...................................................................................................129 6.2.2 Gelatin Preparation...................................................................................129 6.2.3 Polydimethylsiloxane Elastomer Preparation...........................................130 6.2.4 Soluble Polypyrrole Preparation...............................................................130 6.2.5 Chemical Formation of Conducting Polymer IPN Systems.....................130 6.2.6 Supercritical CO 2 Solution Formation of Conducting Polymer IPN Systems..........................................................................................................131 6.2.7 Electrochemical Formation of Conducting Polymer IPN Systems..........132 6.2.8 Mechanical Property Testing of PPy/PDMSe IPN systems.....................133 6.2.9 ATR-FTIR Analysis.................................................................................133 6.2.10 Optical Microscopy................................................................................134 6.2.11 Electrochemical Analysis.......................................................................134 6.2.12 EDS Mapping.........................................................................................134 6.3 Background Study..............................................................................................135 6.3.1 Electrochemical formation of Conducting Polymer IPN Systems...........136 6.3.2 Chemical formation of Conducting Polymer IPN Systems......................138 6.3.3 Conducting Polymer Blend Formation by Solution Blending.................140 6.4 Polypyrrole/PDMSe IPN Formation...................................................................141 6.4.1 Introduction..............................................................................................141 6.4.2 Supercritical Carbon Dioxide Solution Doping.......................................142 6.4.2.1 Introduction....................................................................................142 6.4.2.2 EtOH/scCO 2 prepared IPNs...........................................................143 6.4.2.3 IPA/scCO 2 prepared IPNs..............................................................147 6.4.3 Tetrahydrofuran Solution Doping............................................................154 6.4.3.1 Effects of hydration of sample conductivity..................................156 6.4.3.2 Effects of PDMSe surface segregation...........................................158 6.4.3.3 Effects of FeCl 3 /THF doping on PDMSe mechanical properties..162 6.5 Conclusions.........................................................................................................164 ix

PAGE 10

7 TP AND TPE BASED DYNAMIC, NON-TOXIC, ANTI-FOULING SURFACE COATINGS..............................................................................................................167 7.1 INTRODUCTION..............................................................................................167 7.2 MATERIALS AND METHODS.......................................................................169 7.2.1 Materials...................................................................................................169 7.2.1.1 General Chemicals.........................................................................169 7.2.1.2 PPy/Santoprene Sample Preparation and Mounting....................169 7.2.1.3 Polydimethylsiloxane Elastomer Preparation................................170 7.2.1.4 ATS Coupling Agent Preparation..................................................170 7.2.2 Sample Preparation Methods....................................................................170 7.2.2.1 Sample Mounting...........................................................................171 7.2.2.6 Micropatterning..............................................................................173 7.2.2.7 TP Film Formation by Spin Casting and Spraying........................173 7.2.2.8 Conductivity Determination...........................................................174 7.3 POLYPYRROLE/8281-65 IPN SYSTEMS.......................................................175 7.3.1 PPy/Santoprene 8281-65 Contact Angle Measurements........................175 7.3.2 Determination of PPy Content in PPy/8281-65 Samples.........................176 7.4 POLYPYRROLE/271-55 IPN SYSTEMS.........................................................178 7.4.1 Determination of PPy Content in PPy/271-55 Samples...........................178 7.4.2 Effects of Counter Ion Exchange on PPy/271-55 Conductivity...............180 7.4.3 Formation of poly(3-methylthiophene)/271-55 IPN systems...................183 7.4.4 Micropatterning of 271-55.......................................................................185 7.5 EDS and SEM ANALYSIS of SANTOPRENE TPEs....................................187 7.6 PPy/POLYSULFONE SYSTEMS.....................................................................194 7.8 CONCLUSIONS................................................................................................196 8 DYNAMIC MODULUS MAPPING OF PPY/SANTOPRENE BLENDS...........198 8.1 INTRODUCTION..............................................................................................198 8.2 MATERIALS AND METHODS.......................................................................199 8.2.1 Materials............................................................................................199 8.2.2 Sample Preparation Methods....................................................................199 8.2.2.1 PPy/TPE (8211-65) Sample Preparation........................................199 8.2.2.2 Sample Mounting...........................................................................199 8.2.2.3 Electrochemical Polymerization of Pyrrole...................................200 8.2.3 Hysitron Nano-DMA................................................................................200 8.2.3.1 Hysitron Introduction.....................................................................200 8.2.3.2 Experimental conditions.................................................................202 8.3 HYSITRON NANO-DMA and AFM ANALYSIS............................................204 8.3.1 Nano-DMA mapping of 271-55 and PPy/271-55 samples.......................204 8.3.2 Dynamic fluid cell nano-DMA mapping of PPy/8211-65.......................207 8.4 CONCLUSIONS................................................................................................215 9 CONCLUSIONS AND FUTURE WORK...............................................................216 ABBREVIATIONS.........................................................................................................227 x

PAGE 11

LIST OF REFERENCES.................................................................................................229 BIOGRAPHICAL SKETCH...........................................................................................243 xi

PAGE 12

LIST OF TABLES Table page 7.1 Contact angle values obtain for PPy/8281-65 dual element sample in Instant Ocean artificial sea water versus a Ag/AgCl reference electrode.....................................175 7.2 Weight and volume change data for the formation of PPy/8281-65 IPN systems using the FeCl3/THF solvent soak process......................................................................177 xii

PAGE 13

LIST OF FIGURES Figure page 2.1 Bipolar unit charge migration as depicted for polythiophene...................................10 2.2 Example of one of the possible mechanisms for the oxidative polymerization of polypyrrole...............................................................................................................12 2.3 Examples of basic bending (cantilever) conducting polymer actuators; A) bilayer actuator, B) backbone type actuator, C) shell type actuator.....................................15 2.4 Diagram of a polyimide based electrical resistance strain gage and EAP actuator setup.........................................................................................................................18 2.5 Thermally induced apparent strain for Constantan (Advanced), Isoelastic, and Karma alloys............................................................................................................22 2.6 Diagram showing the different species breakdown and layering of biofoul film formation..................................................................................................................26 2.7 Design of electrowetting device: (a) no applied electrical potential (hydrophobic surface); (b) with applied electrical potential (hydrophilic surface). Fluid is pumped by continuously cycling the applied electrical potential............................30 2.8 Voltage required to induce a of 40 (from 12080) versus dielectric layer thickness for Teflon AF based EWOD device, with = 2.0 and E breakdown = 2x10 16 V/cm.........................................................................................................................33 2.9 Relative surface charge of different conducting polymers......................................34 2.10 Oxidation (-) and reduction (--) potentials of poly pyrrole (PPy), polyaniline (PA), poly(3-methylthiophene (PMeT), and poly(p-phenylene) (PPP).............................36 2.11 Monomer and polymer structures for A) polypyrrole, B) poly(3-methylthiophene), and C) poly(p-phenylene).........................................................................................37 3.1 SEM micrograph of surface morphology of PPy/TOS film prepared in 1.0 M LiClO 4 at a potential of 0.65 V. SEM image taken of an uncoated sample at 1000X and 15 KeV...............................................................................................................41 xiii

PAGE 14

3.2 (a) PPy/TOS cyclic voltammetry ( = 10 mV/s) and (b) in situ strain response of a 9.6 m film prepared in aqueous 1.0 M LiClO 4 .......................................................43 3.3 In situ multi-cycle cyclic voltammetry strain response of a 9.6 film prepared in aqueous 1.0 M LiClO 4 ..............................................................................................44 3.4 In situ square-wave strain response of a 9.6 m film prepared in aqueous 1.0 M LiClO 4 .................................................................................................................45 3.5 Cyclic Voltammetry ( = 100 mV/s) of a PEDOP film produced from aqueous 1.0 M LiClO 4 at E = 0.5 V and t = 200 s........................................................................47 3.6 SEM micrograph of a PEDOP film prepared at 0.6 V in 1.0 M LiClO4. SEM image taken of an uncoated sample at 1000X and 15 KeV.....................................48 3.7 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of a 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 ....................................................................48 3.8 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 Data from figure 3.7 has been replotted vs. time. .................................................................................................................49 3.9 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of a 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 ....................................................................50 3.10 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 replotted vs. time.......................................50 3.11 In situ square-wave strain response of a 10.6 film prepared in aqueous 1.0 M LiClO4. .................................................................................................................51 3.12 Cyclic voltammetry (100 mV/s) of PBEDOT-Cz in 0.1 M TBAP/CAN.................53 3.13 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film (-0.8 V to 0.6 V) for the 1 st 2 nd 5 th and 10 th scans in 1.0 M LiClO 4 .......................................54 3.14 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film (-0.8 V to 1.0 V) for the 1 st 2 nd 5 th and 10 th scans in 1.0 M LiClO 4 .......................................55 3.15 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film (-0.8 V to 1.2 V) for the 1 st 2 nd 5 th and 10 th scans in 1.0 M LiClO 4 .......................................56 3.16 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film scanned from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (5 th scan) in 1.0 M LiClO 4 .........................57 3.17 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film scanned from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (10 th scan) in 1.0 M LiClO 4 .......................58 xiv

PAGE 15

3.18 Comparison of in situ cyclic strain response of PPy, PEDOT, and PBEDOT-Cz in aqueous 1.0 M LiClO 4 ..............................................................................................59 3.19 SEM micrograph of delamination of PPy resulting from long-term cycling of the actuator; 150X..........................................................................................................62 3.20 SEM micrograph of delamination of PPy resulting from exposure to high vacuum; 25X. .................................................................................................................62 3.21 Enlarged SEM micrograph of region A in Figure 4.20; 1350X...........................63 3.22 SEM micrograph of porous PPy at PPy-Au interface; 5000X.................................63 3.23 Enlarged SEM micrograph of region B in Figure 4.20; 100X.............................64 3.24 SEM micrograph of PPy nodules remaining of Au substrate after PPy delamination; 1000X. .................................................................................................................64 3.25 SEM micrograph of PPy surface after delamination from Au substrate during long-term repetitive cycling of the actuator; 250X...........................................................65 3.26 SEM micrograph of exposed PPy-Au interface exhibiting PPy nodule growth; 1000X. .................................................................................................................65 4.1 Correlation between surface roughness factor, nominal EcAu thickness and EcAu deposition charge. EcAu thickness was determined by cross-section SEM...........70 4.2 SEM micrograph of EvAu deposited on smooth polyimide (PI); 4000X................72 4.3 SEM micrograph of 3 minute EcAu deposition on smooth EvAu/PI; 4000X.........72 4.4 SEM micrograph of 10 minute EcAu deposition on smooth EvAu/PI; 4000X.......73 4.5 SEM micrograph of 30 minute EcAu deposition on smooth EvAu/PI; 4000X.......73 4.6 SEM micrograph of 60 minute EcAu deposition on smooth EvAu/PI; 4000X.......74 4.7 SEM micrograph of EvAu (r = 2.89) deposited on rough PI strain gage; 4000X....74 4.8 SEM micrograph of EcAu (r = 6.17, 2.5 min.) deposited on rough EvAu/PI strain gage; 4000X.............................................................................................................75 4.9 SEM micrograph of EcAu (r = 10.04, 10 min.) deposited on rough EvAu/PI strain gage; 4000X.............................................................................................................75 4.10 SEM micrograph of EcAu (r = 18.90, 30 min.) deposited on rough EvAu/PI strain gage; 4000X.............................................................................................................76 xv

PAGE 16

4.11 SEM micrograph of EcAu (r = 24.50, 60 min.) deposited on rough EvAu/PI strain gage; 4000X.............................................................................................................76 4.12 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI; 4000X.......78 4.13 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI treated with EcAu for 10 min.; 4000X.........................................................................................78 4.14 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI treated with EcAu for 30 min.; 4000X.........................................................................................79 4.15 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI treated with EcAu for 60 min.; 4000X.........................................................................................79 4.16 Cross-section SEM micrograph of EcAu (60 min.) deposited on smooth EvAu/PI; 4000X. .................................................................................................................81 4.17 Cross-section SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI treated with EcAu for 60 min.; 4000X.....................................................................81 4.18 Surface plot of calculated PPy film thickness as a function of surface roughness factor (r) and electropolymerization charge (C/cm 2 )...............................................82 4.19 Strain response of a 2.83 C/cm 2 PPy/EvAu/PI actuator during potential cycling at 5 mV/s in aqueous NaClO 4 LiClO 4 CsClO 4 NaNO 3 and NaCl solutions...............84 4.20 Strain response of a 2.83 C/cm 2 PPy/EvAu/PI actuator during potential stepping (100 s/step) in NaClO 4 between .6 V and (a) 0.1 V, (b) 0.2 V, (c) 0.3 V, (d) 0.4 V, (e) 0.5 V, and (f) 0.6 V........................................................................................86 4.21 In situ strain response of PPy/EvAu/PI actuators of varying PPy film thickness during potential stepping between .6 V and 0.4 V in aqueous NaClO4. Electropolymerization charge densities were 0.79 C/cm2 (2.8 m, calculated), 1.6 C/cm2 (5.7 m, cal.), 2.4 C/cm2 (8.6 m, cal.), 3.2 C/cm2 (11.4 m, cal.), 4.0 C/cm2 (14.3 m, cal.), and 4.8 C/cm2 (17.1 m, cal.)............................................88 4.22 In situ charge response of PPy/EvAu/PI actuators of varying PPy film thickness during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 Electropolymerization charge densities were 0.79 C/cm 2 (2.8 m, calculated), 1.6 C/cm 2 (5.7 m, cal.), 2.4 C/cm 2 (8.6 m, cal.), 3.2 C/cm 2 (11.4 m, cal.), 4.0 C/cm 2 (14.3 m, cal.), and 4.8 C/cm 2 (17.1 m, cal.).........................................................89 4.23 In situ strain response of a 2.83 C/cm 2 PPy/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 PPy was prepared potentiostatically at (a) 0.7 V, (b) 0.8 V, (c) 0.9 V, and (d) 1.0 V...........................91 xvi

PAGE 17

4.24 In situ strain response of a 1.18 C/cm 2 PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 With surface roughnesses factors of (a) 2.89, (b) 6.17, (c) 7.13, (d) 10.04, (e) 18.90, and (f) 24.50................93 4.25 Effects of surface roughness factor (r) on the overall change in strain () of a 1.18 C/cm 2 PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 .........................................................................................94 4.26 Effects of surface roughness factor (r) on the strain rate (/sec) of a 1.18 C/cm2 PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO4......................................................................................................95 4.27 In situ strain response of PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 With surface roughnesses factors of 18.90 and PPy electropolymerization charge of (a) 0.79 C/cm 2 (b) 1.6 C/cm 2 (c) 2.4 C/cm 2 (d) 3.2 C/cm 2 (e) 4.0 C/cm 2 and (f) 4.8 C/cm 2 .....................................96 4.28 Maximum change in strain response of PPy/EcAu/EvAu/PI actuators during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 with increasing electropolymerization charge. Measured on surface roughnesses of (a) r = 2.89 (EvAu), and EcAu samples of r = (b) 6.17, (c) 10.00, (d) 18.90, and (e) 24.50......97 4.29 Surface plot of the normalized overall strain response of PPy as a function of electropolymerization charge and surface roughness factor (r)...............................98 4.30 In situ Strain response of a 8.0 C/cm 2 PPy/EcAu/EvAu/PI actuator during potential cycling at 5 mV/s in aqueous (a) LiClO 4 (b) NaNO 3 and (c) NaCl solutions........99 4.31 In situ normalized (relative to 0.01 Hz) strain and charge response of 1.18 C/cm 2 PPy as a function of frequency on (a) r = 2.89 (EvAu) and (b) r = 10.0 (EcAu) surfaces. Inset shows the charge efficiency (/C) as a function of frequency....100 4.32 In situ normalized strain and charge response of 1.18 C/cm 2 PPy (inadequate Argon purge) on r = 5.1 (EvAu) and r = 22.1 (EcAu) treated actuators with repeated potential stepping between .6 V (10 s) and 0.4 V (20 s)....................................102 4.33 In situ normalized strain and charge response of 1.18 C/cm 2 PPy on r = 3.43 (EvAu), r = 8.26 (EcAu), and r = 18.90 (EcAu) treated actuators with repeated potential stepping between .6 V (10 s) and 0.4 V (20 s)....................................102 5.1 Negative patterns of linear actuator produced using Microsoft Powerpoint (A) and printed on Kapton used to fabricate linear actuators (B) and a negative pattern used for dynamic surfaces (C)........................................................................................105 5.2 Rapid electrode patterning process for the development of conducting polymer devices....................................................................................................................106 xvii

PAGE 18

5.3 Diagram of basic linear actuator design and AutoCAD model used to predict the overall developed linear strain for the device........................................................110 5.4 Initial PPy linear actuator based on 10mm segments deposited on copper foil. Picture of oxidized state enhanced to show placement of PPy..............................111 5.5 PPy linear actuator based on 10mm segments deposited on EvAu coated polyimide strip.........................................................................................................................112 5.6 PPy linear actuator based on 20mm segments deposited on EvAu coated polyimide strip.........................................................................................................................112 5.7 Single sided rapid electrode pattern design and illustration of how it works. The green and blue areas represent the separate electrode patterns (working and counter electrode)................................................................................................................113 5.8 Diagram of linear actuator fabrication process. Initial negative CAD electrode design (A) negative electrode design printed on substrate (B) produced patterned gold electrode from the mask (C) electrochemically deposited conducting polymer on the patterned gold electrode (D)........................................................................114 5.9 Pictures of linear actuator fabrication process. CAD design of negative electrode pattern (A) negative pattern printed on substrate (B) gold coated patterned substrate before acetone wash (C) patterned gold electrode after pattern removal (D) electrochemically deposited conducting polymer on the patterned gold electrode (E)...........................................................................................................................114 5.10 Side view diagram of bilayer (A) and backbone (B) type linear actuator designs.115 5.11 WLOP image of the polyimide substrate used in the construction of linear actuators before patterning (Magnification = 25X)...............................................................117 5.12 WLOP image of the polyimide substrate coated with evaporated gold (Magnification = 25X)...........................................................................................117 5.13 WLOP image of electrochemically deposited gold on top of evaporated gold coated polyimide (Magnification = 25X)..........................................................................118 5.14 WLOP image of polypyrrole (+0.8V to 2.24 C/cm 2 ) electrochemically deposited in EcAu/EvAu coated polyimide (Magnification = 25X)..........................................118 5.15 WLOP image of patterned PPy/EcAu/EvAu on polyimide (Magnification = 25X).118 5.16 WLOP image of exposed base polyimide substrate between two patterned PPy/EcAu/EvAu wires (Magnification = 25X)......................................................119 5.17 Schematic of conducting polymer actuator based dynamic surface.......................119 xviii

PAGE 19

5.18 Images of 0.75, 0.50, and 0.25 pt line patterns produced with Xerox Phaser 6200 laser jet, HP Deskjet 6122, and Lexmark i3 inkjet printers (Magnification = 7X).121 5.19 Images of solid printed pattern areas produced with Xerox Paser 6200 laser jet, HP Deskjet 6122, and Lexmark i3 inkjet printers (Magnification = 14X)...................121 5.20 Images of 0.75 pt blue photomask (A) and enhanced image of resulting patterned photoresist coated glass slide (B) (Magnification = 7X)........................................124 5.21 Images of printed field and 0.75 pt brown photomask with black toner cartridge (A) and without black toner cartridge (B) (Magnification = 7X).................................124 5.22 Examples of computer designed patterns used to make blue (A) and brown (B) photomasks.............................................................................................................124 6.1 Images of captive air bubble contact angle measurements on PPy/PDMSe oxidizer insertion films.........................................................................................................140 6.2 Captive air bubble contact angle data for PPy/PDMSe oxidizer insertion films conducted at potentials of -0.8, 0.0, and +0.8 V versus Ag/AgCl in distilled H 2 O.140 6.3 Phase diagram for carbon dioxide..........................................................................143 6.5 Optical images of degraded PDMSe samples after supercritical CO2 doping with 0.1M FeCl3 dopant and 1vol% ethanol cosolvent for 24hrs..................................144 6.6 Optical images of untreated PDMSe (A), and PPy/PDMSe IPNs prepared by EtOH (B) and EtOH/scCO 2 (C) oxidizer insertion methods............................................146 6.7 Changes in PDMSe sample weight during EtOH/scCO 2 prepared PPy/PDMSe IPN systems...................................................................................................................147 6.8 Weight change data for IPA/scCO 2 prepared PPy/PDMSe prepared samples.......148 6.9 Weight change data for EtOH/scCO 2 prepared PPy/PDMSe prepared samples....148 6.10 Optical images (Leica G27) of PPy/PDMSe samples prepared by EtOH/scCO 2 (A) and IPA/scCO 2 (B) oxidizer insertion methods (mag. = 2X). Sample placed on top of printed text to show transparency......................................................................149 6.11 Optical images (Leica G27) of PPy/PDMSe batches prepared by EtOH/scCO 2 (A) and IPA/scCO 2 (B) oxidizer insertion methods showing the differences in homogeneity of PPy formation between samples..................................................149 6.12 Optical images (Axioplan II) of PPy/PDMSe samples prepared by EtOH/scCO 2 (A) and IPA/scCO 2 (B) oxidizer insertion methods (magnification = 100X; scale bar = 200m)...................................................................................................................150 xix

PAGE 20

6.13 Optical images (Axioplan II) of PPy/PDMSe samples prepared by EtOH/scCO 2 (A) and IPA/scCO 2 (B) oxidizer insertion methods (magnification = 200X; scale bar = 100m)...................................................................................................................150 6.14 Weight change of PDMSe samples exposed to different supercritical CO 2 treatments...............................................................................................................153 6.15 Average weight change in PDMSe samples exposed to different supercritical CO 2 treatments and after an 18 hr degassing period......................................................153 6.16 Current response of PPy/PDMSe samples prepared by THF solvent soaking and IPA/scCO2 techniques along with untreated PDMSe. Cyclic voltammetry conducted at .8 V at 10 mV/sec versus Ag/AgCl in Instant Ocean artificial sea water.......................................................................................................................156 6.17 Effects of sample hydration on cyclic voltammetry for wet and dry PPy/PDMSe samples...................................................................................................................157 6.18 ATR-FTIR analysis of PPy/PDMSe IPN systems prepared by various solution soaking procedures. Inset box shows location of characteristic pyrrole ring vibrations (1500-1600 cm -1 )...................................................................................159 6.19 EDS spectra of cross sectioned FeCl 3 /THF doped PDMSe....................................160 6.20 EDS mapping of cross sectioned FeCl 3 /THF doped PDMSe (magnification = 70X).161 6.21 EDS chlorine mapping of cross sectioned FeCl 3 /THF doped PDMSe showing a ~60-100 m surface layer of predominantly pure PDMSe (mag. =70X)..............161 6.22 Strain at break and peak stress data for the degradation of Silastic T2 PDMSe with prolonged exposed to a 5 wt% FeCl 3 /THF solution...............................................163 6.23 High and low strain modulus data for the degradation of Silastic T2 PDMSe with prolonged exposed to a 5 wt% FeCl 3 /THF solution...............................................164 7.1 Preparation scheme for the formation of PDMSe encapsulated PPy composite samples for contact angle measurement and marine testing..................................172 7.2 PDMSe encapsulated PPy composite samples for testing in aqueous/marine environments using a single (A) and dual (B) element designs.............................173 7.3 Change in contact angle for a PPy/8281-65 dual element sample in Instant Ocean artificial sea water versus a Ag/AgCl reference electrode.....................................176 7.4 Weight loss data for the formation of PPy/271-55 IPN systems............................180 7.5 Volume loss data for the formation of PPy/271-55 IPN systems...........................180 xx

PAGE 21

7.6 Calculated PPy/271-55 film conductivity based on conductive layer thickness. Value for 0.07 cm is actual conductivity of PPy/271-55 film assuming bulk conductive..............................................................................................................183 7.7 Example of 5m biomimetic sharklet pattern on PDMSe (A) and patterned silicon wafer used to create the patterned PDMSe. SEM micrographs taken at 1000X (scale bar = 50 m; WD = 15 mm and EV = 15 KeV, and AuPd coated).............185 7.8 Optical micrograph of 5m biomimetic sharklet pattern on 271-55 produced at 185C (A) and patterned silicon wafer used to create the patterned 271-55. Images taken at 1000X.......................................................................................................186 7.9 SEM micrographs of sharklet patterned Santoprene 271-55 patterned at 185C; 3300X (A) and 2500X 40 tilt (B). Images taken at 15 KeV with a working distance of 15 mm and a AuPd surface coating.....................................................186 7.10 SEM micrographs of sharklet patterned Santoprene 271-55 patterned at 200C; 500X (A) and 1000X (B). Images taken at 15 KeV with a working distance of 15 mm and a AuPd surface coating.............................................................................187 7.11 Cross sectional EDS spectrum of FeCl 3 doped Santoprene 271-55 (A) and 8211-65 (B) showing the presence of Fe and Cl..................................................................188 7.12 Cross sectional EDS mapping of FeCl 3 doped Santoprene 271-55 taken at 180X; scale bar = 200 m.................................................................................................189 7.13 Cross sectional EDS mapping of FeCl 3 doped Santoprene 8211-65 taken at 85X; scale bar = 500 m.................................................................................................190 7.14 Cross sectional EDS mapping for Cl in FeCl 3 doped PDMSe (A), 271-55 (B), and 8211-65 (C). Magnification and scale bars equal to 70X, 180X, and 85X and 500 m, 200 m, and 500 m respectively. Images taken at 15 KeV and 15 mm on carbon coated cross section samples......................................................................191 7.15 Cross sectional EDS mapping of FeCl 3 doped PPy/271-55 taken at 3000X; scale bar = 10 m. Images taken at 15 KeV and 15 mm on carbon coated cross section samples...................................................................................................................192 7.16 Cross sectional EDS mapping of FeCl 3 doped PPy/8211-65 taken at 3000X; scale bar = 10 m. Images taken at 15 KeV and 15 mm on carbon coated cross section samples...................................................................................................................193 7.17 Cross sectional SEM secondary (A) and backscatter (B) electron micrographs of PPy/271-55. Images taken at 3000X and scale bar equal to 10 m. Images taken at 15 KeV and 15 mm on carbon coated cross section samples.................................194 xxi

PAGE 22

7.18 Cross sectional SEM secondary (A) and backscatter (B) electron micrographs of PPy/8211-65. Images taken at 3000X and scale bar equal to 10 m. Images taken at 15 KeV and 15 mm on carbon coated cross section samples.............................194 7.19 Optical micrographs of PSU films sprayed from 5 wt% (A) and 10 wt% (B) PSU/THF solutions. Magnification = 7X..............................................................196 8.1 Diagram of three plate capacitor transducer used in the Hystitron Triboindenter .201 8.2 Images of cono-spherical (A) and Berkovich (B) Hysitron tips. Images obtained from www.hysitron.com (magnification unknown)...............................................203 8.3 Diagram (A) and optical image (B) of fluid cell setup inside the Hysitron Triboindenter ........................................................................................................204 8.4 Optical micographs of Santoprene 271-55 (A) and PPy/271-55 (B) samples obtain by the onboard optical camera of the Hysitron Triboindenter (magnification = 5X)..........................................................................................................................205 8.5 3D surface plots of Santoprene 271-55 (A) and PPy/271-55 (B) surfaces. Scan area is equal to 50 m x 50 m square...................................................................205 8.6 Nano-DMA mapping images of Santoprene 271-55 showing uniform surface phase and modulus morphology. Scan area is 50 m x 50 m square.................206 8.7 Nano-DMA mapping images of PPy/271-55 showing uniform surface phase and modulus morphology. Scan area is 50 m x 50 m square..................................207 8.8 3D surface topography images of PPy/8211-65 at applied potentials of -0.7 V (A), 0.0 V (B), and +0.7 V (C). Scan area is 25 m x 25 m square...........................208 8.9 AFM optical micrograph (A) of PPy/8211-65 with corresponding surface topography (B) and phase contrast (C) images taken at a 50m square scan area. Scale bars equal to 15m.......................................................................................209 8.10 AFM surface topography, phase imaging, and 3D surface plot of PPy/8211-65 at 50 m (A), 10 m (B), 5 m (C), and 1 m (D) square scan areas. Scale bars equal to 15 m, 3 m, 1.5 m, and 0.3 m respectively.................................................210 r 8.11 Complex modulus maps for PPy/8211-65 at an applied potential of -0.8 V (A), 0.0 V (B), and +0.8 V (C)............................................................................................212 8.12 Complex modulus line scan data for PPy/8211-65 obtained from the location depicted in figure 7.28............................................................................................212 8.13 3D surface plots of PPy/8211-65 under the applied potentials of -0.8 V (A), 0.0 V (B), and +0.8 V (C)................................................................................................213 xxii

PAGE 23

8.14 Average surface modulus for PPy and PPy/8211-65 systems as a function of the applied potential.....................................................................................................214 xxiii

PAGE 24

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DYNAMIC ANTIFOULING STRUCTURES AND ACTUATORS USING EAP COMPOSITES By Clayton Claverie Bohn Jr. August 2004 Chair: Anthony Brennan Major Department: Materials Science and Engineering By utilizing strain gage technology it is possible to directly and continuously measure the electrochemically induced strain response of EAP actuators. Strain sensitive actuators were constructed by directly vapor depositing gold (EvAu) on polyimide strain gages which are capable of measuring strain with an accuracy of +/1 Strain sensitive actuators were used to evaluate the strain response of polypyrrole (PPy), poly(3,4-ethylenedioxypyrrole) (PEDOP) and poly(3,6-bis(2-(3,4-ethylenedioxy)thienyl)-N-carbazole) (PBEDOT-Cz). PPy was shown to produce significantly higher strain when compared to PEDOP and PBEDOT-Cz. The resulting overall strain for the materials was 236, 33, and 35 respectively. From the initial investigation, adhesion of the EAP to the EvAu layer was identified as a major factor in the resulting lifetime and strain response of these actuators. Therefore an electrochemically deposited Au layer (EcAu) was deposited on top of the EvAu layer to improve the adhesion of the EAP to the working electrode. By changing the surface roughness from r = 3.43 (EvAu) to r = 8.26 xxiv

PAGE 25

and 18.00 (EcAu) the normalized strain response after 2000 cycles increases from 45% to 60% and 68% respectively. Also by changing the surface roughness from r = 5 to r = 23, the resulting strain response increases from ~100 to 600-800 for PPy. By incorporating conducting polymers such as polypyrrole into elastomeric base materials such as PDMSe and Santopene TPEs a tough durable dynamic non-toxic antifouling surface coating was formed. These coatings utilize the dynamic changes in polymer charge, modulus, and swelling that naturally occur during the redox cycling of conducting polymers to dynamically change the surface properties of the resulting films. Dynamic changes in contact angle (surface energy) of 11 and 21 have been measured for PPy/TPE (.5V driving potential) and PPy/PDMSe (.0V driving potential) IPN systems. Using a Hysitron Triboindenter equipped with a electrochemical fluid cell setup; dynamic fluid cell nano-DMA mapping measurements were conducted on PPy/TPE systems. Dynamic surface modulus changes of ~20-40% were measured for the PPy/TPE composite systems when switched between their reduced to oxidized states. All samples showed a decrease in the sample modulus when switched from the reduced to the oxidized state. This is mainly associated with the influx of water and counter ions into the conducting polymer phase during the oxidation process. This in turn increases the free volume of the conducting polymer phase and also acts to plasticize the phase. Changes in the surface topography associated with the redox cycling of the composite structure were also observed during these experiments. This in turn results in a dynamic surface coating that is capable of changing its surface energy, modulus, and topography by changing applied electrical potential. xxv

PAGE 26

CHAPTER 1 INTRODUCTION Smart materials such as electroactive polymers (EAPs) or more specifically conducting polymers (CPs), such as polyaniline and polypyrrole (PPy), have gained a lot of interest as candidates for various actuator applications such as active hinges, anti-vibration systems, micro-catheter steerers, micro-valves and pumps, mechanical shutters, and robotics due to their ability to undergo controllable volumetric changes under a given stimulus. 1-10 More specifically, conducting polymers have been shown to undergo volumetric changes during electrochemical oxidation and reduction due to the electrochemical transport of ions and solvent molecules into and out of the polymeric system. 11-23 Actuators are constructed by directly depositing conducting polymers, electrochemically, onto a metalized flexible substrate, such as polyimide and polyester. By doing this it is possible to directly convert induced volumetric changes into a mechanical (bending) type of actuation. 6, 24-36 The development of actuation in these systems is very similar in nature to the motion developed in bi-metallic strip (i.e., thermostats). One layer changes length with respect to the other inducing a curling motion to develop in the direction away from the expanding layer. During electrochemical switching (oxidation and reduction) the CP layer expands and contracts respectively while the flexible substrate maintains its original dimensions. This induces a bending motion to develop in the direction opposite the expanding CP layer. The amount of strain ( = L/Lo, where L is the overall change in length and Lo is the original length) that is developed by these CP systems during electrochemical 1

PAGE 27

2 switching is the main physical properties that controls the amount of actuation produced by these devices. Therefore precise measurement of the strain developed by these systems is an essential part in understanding the physical properties and behavior of these materials and devices. A major factor in understanding these materials is that the reported physical properties for CP systems vary from one study to another depending on the characterization techniques utilized and how the sample was prepared (counter ions, potentials, concentrations, etc.). The smallest change in any of these factors can produce dramatic changes in the overall performance (physically and electrochemically) of these materials. Some common characterization techniques used to study the in-plane strain (sample elongation) of these systems are high-speed video capture, laser displacement, and load/stress sensors (Instron/MTS type). With digital video, the motion of the conducting polymer actuator is digitally recorded and imported to a computer for detailed analysis. 11, 37-39 Accurate measurements of deflection and elongation can be obtained however this involves post processing of the recorded data and results can vary depending on what method was used to calculate the strain. Real time laser displacement meters or extensometers have also been utilized to evaluate the degree of motion induced during electrochemical switching. 26, 40, 41 These systems allow for accurate, real time monitoring of tip displacement of the actuator system. Tip displacement can then be converted to strain at a latter time. Use of force/displacement meters or load cells is commonly used in the study of conducting polymers actuators under linear actuation conditions. 13, 42-48 Out-of-plane strain (sample thickening) for PPy has also been measured using atomic force microscopy and has been shown to be significantly higher than the in-plane strain. 49

PAGE 28

3 A new method has been developed for the in situ measurement of the in-plane strain response developed by cantilever (bi-layer) style CP actuators. With this tecnique it is possible to measure very precisely the average strain developed in the system during electrochemical switching. This method employs the utilization of Au coated electrical resistance strain gages as the working electrode (flexible metalized substrate) during electrochemical switching. Strain gages are widely used in many industries (automotive, aeronautical, naval, construction, etc.) for precision in situ spot measurements of induced strains in many materials. These devices are capable of measuring strain with a precision of 1 ( = 110 -6 ) and have a strain limit of 5% (50,000 ). These devices are also utilized as the sensing components of load cells and therefore are capable of measuring stresses when applied in the right configuration. Strain gages are typically constructed of a constantan foil grid that is encapsulated in a flexible polyimide substrate, but other gages are available utilizing different grid and backing materials. These devices are very flexible in use as well as installation and are capable of measuring strain under various load conditions and environments, including cyclic strains. This ability to conduct in situ cyclic strain measurements with high precision is ideal for applications in EAP actuators and sensors due to the relatively low strain produced and cyclic nature of their motion. This technique allows for detailed real-time analysis of the effects of various actuation, electrochemical polymerization, and actuator construction factors such as CP film thickness, working electrode morphology, counter ion type, driving potential, polymerization potential, and scan rate or switch speed.

PAGE 29

4 Another issue associated with the development and use of conducting polymer based actuators is interlayer adhesion. The interface between the flexible Au coated substrate and stiffer conducting polymer is subjected to large amounts of stress during the redox cycling process. The repeated cyclic swelling and deswelling of the conducting polymer layer eventually initiates crack formation at this interface. The cracks then propagate with continued cycling and eventually caused the complete delamination of the conducting polymer layer and the failure of the device. A new electrochemically deposited Au (EcAu) layer was developed to improve the interlayer adhesion between the conducting polymer and Au electrode layers. The EcAu treatment greatly increased the base electrode surface area by the growth of Au crystals on the preexisting evaporated Au layer. These crystals range in size from sub micron to 10-20m which can be controlled by the electrodeposition time or current density. These surface treatments have been shown to greatly improve overall actuator performance and lifetime when applied. Performance variable that are improved include properties such as produced strain, strain rate, frequency response, lifetime, and potentially the stress out put from these systems. By applying the lessons learned from the previous mentioned work a novel dynamic surface coating was developed by the incorporation of conducting polymers such as polypyrrole into elastomeric material such as PDMSe and thermoplastic elastomers (TPEs). These coatings have potential applications as dynamic non-toxic antifouling coatings for marine and industrial applications. Biofouling is estimated to cost the US Navy alone over $1 billion per year by increasing the hydrodynamic drag of

PAGE 30

5 naval vessels 50, 51 thus resulting in decreased range, speed, and maneuverability of naval vessels and increases the fuel consumption by up to 30-40%. 52, 53 It has been shown that the settlement and adhesion of fouling marine organisms are effected by surface properties such as surface energy, modulus, and topography. Due to genetic diversity of the various fouling marine species the effects of surface properties on settlement and adhesion changes from species to species. Therefore the use of single surface coating with fixed surface properties can not deter the settlement and growth of all marine organisms. In this approach the dynamic changes in polymer charge, modulus, and topography associated with the redox cycling of conducting polymers is used to develop a dynamic surface coating with variable properties. This should in turn provide a wider range of deterrence to settlement and growth than can be obtained from non-dynamic surfaces.

PAGE 31

CHAPTER 2 BACKGROUND 2.1 Electroactive Polymer Actuators 2.1.1 Introduction to EAPs Electroactive polymers (EAP) are a class of polymers that respond to electrical stimuli by changing their properties and shape. As described by Bar-Cohen 54 these materials can be separated into two classes: electronic and ionic electroactive polymers. The electronic EAPs class of materials is driven by an applied electric field or Coulomb forces. The second class of ionic EAPs is driven by the movement of ions and solvent into and out of the polymeric structure. The electronic EAPs include ferroelectric polymers, dielectric EAPs, electrostrictive graft elastomers, electrostrictive paper, electroviscoelastic elastomers, electrorheological fluids (ERF), and liquid crystal elastomer (LCE) materials. Ionic EAPs consist of ionic polymer gels (IPG), ionomeric polymer-metal composites (IPMC), conducting polymers (CP), and carbon nanotubes (CNT). Ferroelectric polymers are typically characterized by the piezoelectric materials. The piezoelectric effect was discovered in 1880 by Pierre and Paul-Jacques Curie. They found that by compressing certain crystals such as quartz along the appropriate axes a voltage is produced. The following year they discovered that the opposite was true as well. If a voltage was applied to these crystals they in turn elongated. Ferroelectricity is the phenomenon of piezoelectricity that is observed in nonconducting crystalline or dielectric materials such as poly(vinylidene fluoride) and its copolymers that exhibit 6

PAGE 32

7 spontaneous polarization under a given electrical stimuli. Piezoelectric materials are characterized as being semi-crystalline with an inactive amorphous phase. These materials are relatively stiff with a Youngs modulus between 1-10 GPA. A relatively high driving potential of about 200 MV/m, which is very close to the materials dielectric breakdown point, is required to activate the piezoelectric materials. Piezoelectric materials are capable of producing strains on the order of 2% (5% strain possible at 150 V/m with specially modified materials) with very rapid speed and are able to be actuated in air, vacuum, and aqueous environment. Dielectric EAPs are a class of materials that are characterized as having a low modulus with a high dielectric constant. These polymers are capable of inducing large actuation strains with the application of a given electric field. Actuators are typically constructed by placing/laminating electrodes on opposite sides of the elastomeric dielectric material. Once an electric field is applied the electrodes are electrically attracted to each other applying a compressive force on the elastomeric dielectric material causing it to deform and expand in the transverse direction. The dielectric EAPs also require a large driving potential on the order of 100 V/m, but are capable of producing strains on the order of 10-300%. As with piezoelectric materials this driving voltage is very close to the breakdown point of the material. Electrostrictive graft elastomers have a flexible amorphous backbone with grafted polar polymer segments that are capable of forming crystalline domains. The crystalline domains react to the applied electrical field. These materials are capable of producing strains on the order of 4% and are readily processable.

PAGE 33

8 Electroviscoelastic elastomers and electrorheological fluids are both characterized as materials that have dielectric particles dispersed throughout the bulk of the material. The most basic example of an ERF is that of dielectric particles (0.1-100 m) dispersed in insulating base material such as silicone oil. Under an applied electrical field a dipole moment is induced on the particles which then align and form chains increasing the viscosity of the fluid. These fluids take on the consistency of gels under the applied electrical field. Electrovisoelastic elastomers are similar in structure but utilize a crosslinked insulating phase instead of a fluid phase. Under an applied electrical field (<6 V/m) the particles interact and increase the shear modulus of the material by as much as 50%. Both of these materials have applications as active dampeners. Ionic EAPs such as ionic polymer gels (IPG) are driven or actuated by the movement counter ions in the material. IPGs are capable of producing strains on the order of 400% under chemical and electrical stimuli. However they suffer from extremely long actuation times on the order of 20 minutes for a 400% increase in size. If these gels are placed in an electrolyte solution and an electrical stimulus is applied the material will expand and in all directions. However if the gel is used in a dry environment and electrodes are placed on opposite sides of the gel it will form a bending actuator. As one side becomes more alkaline (cathode) and the other more acidic (anode) the migration of ions through the material results in an expansion on the cathodic side and a contraction on the anodic side. Besides having relatively long actuation times the large induced strains tend to damage the applied electrodes after only a couple of cycles. Ionomeric polymer-metal composite actuators operate on the mobility of cations in the composite structure. Upon the application of a low driving voltage (1-10 V) cations

PAGE 34

9 are transported from one side of the actuator to the other through negatively charged channels in the polymer film. Acid functional fluorocarbon polymers such as Nafion (sulfonate based) and Flemion (carboxylate based) are two examples of polymers used in these composites. Electrodes are applied chemically from solution by implanting metal ions (gold, platinum, etc.) throughout the hydrophobic regions of the actuators surface. These materials are characterized as producing relatively large strain at frequencies below 1Hz. As with most ionic EAPs, as the switching frequency increases the produced strain decreases. Carbon nanotubes (CNT) have recently been identified as a class of ionic EAPs. They are very similar to conducting polymers in that they are nearly completely composed of a conjugated carbon-carbon bond network (excluding defects in the structure). However unlike conducting polymers the actuation produced by carbon nanotubes is due to the change in the carbon-carbon bond length as the material is oxidized and reduced. Upon removal (oxidation) of electrons a net positive charge is formed across the conjugated network. This net positive charge on the carbon nuclei causes the carbon nuclei to repel one another resulting in an increase in the overall length of the CNT. Upon injection of electrons (reduction) into the conjugated carbon network an increase in carbon-carbon bond length is also induced. These devices have been shown to produce strains on the order of 1% (along length of nanotube) and are capable of withstanding temperatures in excess of 1000C. This allows for the development of new high temperature actuators that far exceed the temperature capabilities of current systems.

PAGE 35

10 The main EAP of interest for this research is the conducting polymer. These materials are comprised of polymers that have a completely conjugated backbone such as polypyrrole, polyanaline, and poly(p-phenylene). 2.1.1 Conducting Polymers Conducting polymers have recently received a lot of attention due to their ability to generate various responses under electrical or chemical stimuli along with the ability to be electrical conductors. Most notable for this application is volumetric changes that occur during chemical or electrochemical oxidation and reduction. The changes in volume are due to development of charge along the polymer backbone. SSSSSSSSSSSSSSSSB i p o l a r u n i t Figure 2.1 Bipolar unit charge migration as depicted for polythiophene. Conducting polymers are characterized as having a fully conjugated polymer backbone with an extended -system. This allows for electron and charge delocalization along the polymer backbone as shown in Figure 2.1. 55 The volume change associated with oxidation of conducting polymers in electrolyte solutions (salt solutions, body fluids, etc.) is complex but is mainly induced by the influx of solvent and anions into the conducting polymer matrix to balance the developed increasing charge resulting in a

PAGE 36

11 volume expansion of the material. 11-23 During reduction the material starts to return to its neutral (uncharged) state and anions are expelled from the polymer matrix resulting in a volume contraction. There have been numerous studies examining the developed strain and force associated with the redox swelling and de-swelling of conducting polymers. 23, 29, 33, 56-62 2.1.2 Conducting Polymer Synthesis Electrochemical polymerization is generally carried out by the application of an oxidizing potential equal to or greater than the oxidation potential of the monomer. 55, 63 The general mechanism for the polymerization of conducting polymers involves the formation of a radical-cation on the monomer structure. The propagation step involves the subsequent coupling of two oxidized monomers to form a dicationic dimer upon which two hydrogen atoms are expelled and a stable dimer is formed. Oxidation of the dimer followed by the subsequent coupling with an oxidized monomer, dimer, or oligomer results in the continued propagation of the conjugated polymer chain. The exact mechanism for the oxidative polymerization of conducting polymers is unknown but a series of possible mechanisms is shown by Sadki et al.. 64 Figure 2.2 illustrates one of the more accepted polymerization mechanisms for polypyrrole. The properties of a conducting polymer are dependent upon many variables. The main factors are polymerization/oxidation potential, electrochemical polymerization conditions (potentiostatic, galvanostatic, and variations thereof), dopants (Cl, ClO 4 etc.), as well as solvents and electrodes that were used during the electropolymerization process.

PAGE 37

12 Figure 2.2 Example of one of the possible mechanisms for the oxidative polymerization of polypyrrole. Adapted from Sadki et al.. 64 CCCNCHHHHHCCCNCHHHHHCCCNCHHHHHCCCNCHHHHHCCCNCHHHHHCCCNCHHHHHCCCNCHHHHCCCNCHHHHCCCNCHHHHCCCNCHHHHCCCNCHHHHCCCNCHHHHCCCNCHHHHHCCCNCHHHHHCCCNCHHHHHCCCNCHHHHH2H+e-CCCNCHHHHH Conducting polymers can also be formed via chemical polymerization methods such as Lewis acid, ring opening metathesis, and transition metal catalyzed coupling

PAGE 38

13 polymerizations. 55, 63 The most common chemical method utilized to make conducting polymers (due to its simplicity) is the use of Lewis acids such as FeCl 3 FeClO 4 and ammonium persulfate. The polymerization mechanism is very similar to that of the electrochemical polymerization; however electron transfer is too oxidizing species instead of an electrode. During polymerization an electron is transferred from the monomer (oxidation) to Fe(III)Cl 3 which in turn is reduced to Fe(II) leaving Cl (counter ion) to balance the positive charge developed on the monomer during the oxidation process. Chemical polymerization is a bulk polymerization method that is a distinct advantage compared to electrochemical methods. However, chemical synthesis with Lewis acids is not as controllable as the electrochemical methods and is known to produce material with greater defects (disrupted conjugation) densities, thereby decreasing the overall conductivity of the material. 2.1.3 Conducting Polymer Actuators Conducting polymer actuators are candidates for various actuator applications including robotics, artificial muscles, microvalves, catherter steerers, antivibration systems, and multiple other systems. 1-7, 9, 10, 65, 66 There are numerous forms of conducting polymer actuators; however all CP actuators are typically based on the idea of a conducting polymer laminated to a flexible conductive substrate. The most basic design involves the deposition of a CP film on one side of a conducting flexible substrate (sputtered Au on polyimide or PET) and utilizes an external counter electrode for operation. During redox cycling the CP expands (~1-2% in-plane 29, 44, 67 and ~30% out-of-plane 49 ) and contracts due to the movement of ions into and out of the CP structure

PAGE 39

14 respectively. This volume change can then be converted into a bending type actuation by deposition of the CP film onto an inactive but flexible substrate such as polyimide or PET. As the CP is oxidized it expands while the flexible substrate maintains its original length. This causes a bending moment in the direction away from the expanding CP layer. This type of actuator is typically referred to as a bilayer actuator and has been studied extensively. 6, 24-36, 45, 65, 66, 68-70 A modification of the bilayer design is the backbone type design. In the backbone type design a CP layer is deposited on either side of the flexible substrate. One side acts as the anode (working electrode) while the other side acts as the cathode (counter electrode). During redox switching one side expands (anode) while the other side contracts (cathode). When the driving potential is flipped there roles reverse and the bending occurs in the opposite direction. This push-pull technique results in an enhanced bending moment or increased strain performance. In order for these types of actuators to work they must be run in an electrolyte fluid for ion transport. Another configuration is the shell type design. The shell type actuator is basically a modification of the backbone type design. However in this case, the flexible substrate is replaced with a flexible adhesive polyelectrolyte (source for ion transport) and the whole device is encapsulated to prevent dehydration of the system. These type devices can be actuated in air and do not require an external ion source such as an electrolyte solution. Figure 2.3 shows some examples of typical CP bending actuators. Linear actuators have also been constructed by wrapping CP fibers around a flexible electrode or vice

PAGE 40

15 versa. In this case the CP fibers expand and contract in length (axial direction) and a linear type actuation is produced. 44, 46, 60, 71 Figure 2.3 Examples of basic bending (cantilever) conducting polymer actuators; A) bilayer actuator, B) backbone type actuator, C) shell type actuator. These actuating devices are relatively easy to make and due to new lithographic micropatterning techniques they can be fashioned into almost any configuration imaginable making them very versatile. These devices are usually driven with an applied voltage in the range of 1-5 V depending on the polymer used. However there actuation time (strain rate) and performance are very dependent on polymerization and actuation conditions.

PAGE 41

16 One of the main drawbacks to CP actuators is that the whole actuation/swelling process is diffusion limited. Thinner layers have an improved strain rate; however they also produce lower overall stresses and strains than thicker films. Therefore as one increases the CP film thickness to increase the generated stress and strain the overall speed of the system decreases rapidly. These devices are also subject to large amounts of shear at the CP-substrate interface and are prone to interfacial cracking and delamination when cycled for prolonged periods of time. 54, 72, 73 2.2 Electrical Resistance Strain Gages The amount of strain developed by the EAP systems during electrochemical switching is one of the main physical properties that controls the amount of actuation produced by the EAP devices. Measurement of the developed strain is an essential part in understanding the physical properties and behavior of these materials and devices. Reported physical properties for EAPs vary from one study to another depending on the characterization technique and sample preparation. Some common characterization techniques used to study the in-plane strain of these systems are high-speed video capture, laser displacement, and load/stress sensors (Instron/MTS type). With digital video, the motion of the conducting polymer actuator can be recorded and imported to a computer for detailed analysis. 11, 37-39 Accurate measurements of deflection and elongation can be obtained with this technique, but it involves post processing of the data recorded. Another drawback of the video system is as actuation speed increases the speed and resolution of the video capture system also has to increase to record accurate data; therefore the price of the system has to increase as well.

PAGE 42

17 Real time laser displacement meters or laser extensometers have also been used to evaluate the degree of motion induced during redox switching. 26, 40, 41 These systems enable accurate, real time monitoring of tip displacement of the actuator system. Tip displacement can then be converted to strain at a latter time. Force/displacement meters or load cells are also commonly used in the study of conducting polymers actuators under linear actuation conditions. 13, 42-48 Out-of-plane strain has also been measured using atomic force microscopy. 49 We have developed a new technique for the in-situ measurement of the strain response developed by cantilever (bi-layer) style EAP actuators. This method employs the utilization of gold coated electrical resistance strain gages (Figure 2.4) as the working electrode during electrochemical switching. Strain gages are widely used in many industries (automotive, aeronautical, naval, construction, etc.) for precision in-situ spot measurements of induced strains in many materials. These devices are capable of measuring strain with a precision of 1-6 (=) and have a strain limit of 5% (50,000 ). These devices are also capable of measuring stresses when applied in the right configuration, such as in load cells. Electrical resistance strain gages are typically constructed of a constantan foil grid that is encapsulated in a flexible polyimide shell, but other gages are available utilizing different grid and backing materials. As the constantan foil grid is deformed (during redox switching) the electrical resistance of the grid changes, this change in electrical resistance is directly related to the strain induced on the grid/system. The strain () is reported as the change in the grid length (L) relative to initial grid length (L o ), =L/L o

PAGE 43

18 EAPconstantan gridpolyimidesolder tabs EAPconstantan gridpolyimidesolder tabs Figure 2.4 Diagram of a polyimide based electrical resistance strain gage and EAP actuator setup. Adapted from www.Vishay.com These devices are very flexible in use as well as installation and are capable of measuring strain under various load conditions and environments, including cyclic strains. This ability to measure cyclic strains with high precision is ideal for applications in EAP actuators and sensors due to the relatively low strain produced and cyclic nature of their motion. 2.2.1 Strain Gage Theory The basic principles allowing for the development of modern electrical resistance strain gages were discovered in 1856 by Lord Kelvin. He observed that under tensile loading of copper and iron wires the resistance of the wires increased for a given amount of strain. He also discovered that the iron wire exhibited a larger change in resistance than copper for a given amount of strain. He finally applied a Wheatstone bridge to accurately measure the resistance changes developed in these systems.

PAGE 44

19 From this it can be said that the resistance of a wire is a function of the strain applied to that wire, different materials produce different resistance changes for a given strain (sensitivity, S A ). Therefore the resistance (R) of a uniform wire can be written as: LRA (1.1) where is the specific resistance of the material, L is the wire length, and A is the cross-sectional area of the wire. The sensitivity of the wire can then be describes as the resistance change of the wire per unit of initial resistance divided by the applied strain (): 0AdRRS (1.2) By combining equations 1.1 and 1.2 and rearranging the sensitivity term can be derived to determine the sensitivity as a function of the Poissons ratio (, xz ) of the material and its change in specific resistance due to the applied strain, Eq. 1.3. 12AdS (1.3) Equation 1.3 can be broken down into two parts, the effects of dimensional changes during applied stain (1+2) and the effects of specific resistivity (m)with applied strain ([d/]/). Most metallic alloys have an S A value between 2 to 4 with the value of (1+2) varying from 1.4 to 1.7. This gives a range of 0.3 to 2.6 for the change in specific resistance, which can be quite significant compared to the effects of dimensional changes. The change in specific resistance is due to the number of free electrons and the variation of their mobility with the applied strain.

PAGE 45

20 The actual strain is determined from the gage factor (S g ). This factor is experimentally determined for each batch of foil gages by applying a known strain to a series of gages mounted on a specially designed cantilever beam and measuring there resistance response to the applied strain. This is expressed as g aRSR (1.4) where a is the applied axial strain. 2.2.2 Strain Gage Materials and Construction A major factor in choosing an appropriate gage alloy is that the strain sensitivity is linear over a wide range of applied strains. This allows for the use of a single calibration factor and insures that this calibration factor will not change with the various degrees of strain that the gage might see. Some other factors to consider is that the alloys S A should not change significantly as the material enters the plastic regime, the alloy should also have a high specific resistance, thermal stability, and not be significantly affected by temperature change. Temperature compensation can typically be controlled for a particular alloy by the addition of trace impurities and heat treatment. Temperature compensating strain gages reduce the effects of the R/R induced by temperature changes to less than 10 -6 /C. The most common alloy used is Advanced or Constantan. The Constantan alloy is comprised of 45% nickel and 55% copper and has a specific resistance of 0.49 m which allows for the construction of smaller gage patterns with high resistance. Due to electrical circuit requirements the minimum gage resistance needs to be above 100 to prevent overloading of the power supply and to minimize gage self heating (resistive heating). Thus the minimum gage element strand length is on the order of 4 when made

PAGE 46

21 of the finest standard wire. To keep the overall gage length down the sensor elements are typically folded back and forth to form a grid pattern (Figure 2.4). Typical gage resistances are 120 and 350 but higher resistance gages are available. Some other common gage alloys are Nichrome V (80 Ni, 20 Cr; S A = 2.2), Isoelastic (36 Ni, 8 Cr, 0.5 Mo, 55.5 Fe; S A = 3.6), Karma (74 Ni, 20 Cr, 3 Al, 3 Fe; S A = 2.0), Armour D (70 Fe, 20 Cr, 10 Al; S A = 2.0), and Alloy 479 (92 Pt, 8 W; S A = 4.1). There are advantages and disadvantages to the use of these alloys when compared to Constantan. For example Isoelastic has a higher sensitivity (3.6) and higher fatigue strength compared to Constantan, allowing for more precise measurements under high cyclic strains exceeding 1500 However Isoelastic is extremely sensitive to temperature effects. A change in gage temperature of 1 C will result in an apparent strain change of 300 400 Karma has similar properties compared to Constantan but has a higher fatigue limit and excellent time stability, allowing for extended measurement periods (weeks to months). Karma also has a larger use range (up to 260 C) than Constantan (up to 204 C). Karma is difficult to solder, which makes it difficult to attach lead wires. Figure 2.5 shows the thermally induced apparent strain for the Constantan (Advanced), Isoelstic, and Karma alloys.

PAGE 47

22 Figure 2.5 Thermally induced apparent strain for Constantan (Advanced), Isoelastic, and Karma alloys. Taken from Dally et al., Figure 6.2. 74 Current strain gages use metal foil grids produced from a photoetching process. The versatility of this process allows for the production of a variety of gage sizes and shapes. However the patterned metal grid is very thin and fragile. This makes it prone to damage by distortion, wrinkling, and tearing of the grid element. A solution to this problem is to back the metal film elements with flexible sheets like polyimide (0.025 mm thick). This improves the mechanical stability of the gage and handling. It also improves bonding of the gage to various surfaces including conductive ones (polyimide backing acts as an electrical insulator between the grid element and the substrate). The gage can also be encapsulated with a top layer of polyimide film for use in adverse conditions like under water and in areas prone to high levels of dust/debris. Other substrates like very thin high modulus epoxy (transducer applications) and glass fiber or phenolic reinforced epoxies (high level cyclic strain and high temperature applications) are also used under special conditions. These gages can be affixed to a

PAGE 48

23 variety of surfaces utilizing epoxy cement, cyanoacrylate cement, polyester adhesives, and ceramic cements depending on the testing/application conditions. 2.2.3 Strain Gage Accuracy Electrical resistance strain gages are considered one of the most accurate methods for determining strain available. These gages are capable of measuring strain with a precision on the order of 1 This is due to the ability to produce gages with resistance accuracy of 0.3% and with gage factors (calibration constant) certified to 0.5%. However there is some error associated with the transverse sensitivity of the gages. The end loops place a small portion of the gage in the transverse direction. So any transverse strain in the system will increase the resistivety of the gage resulting in a false increase in the measured linear strain. Modern foil gages have enlarged end loops which decrease the sensitivity of the gage to transverse strain in the system. By enlarging the end loops the overall resistance of these segments is very low and therefore any change in the resistance due to the transverse strain will be negligible. The only time error due to transverse strain is not a factor is where the transverse sensitivity factor (K t ) of the gage is zero or when the applied stress field is uniaxial. The overall change in resistance for a strain gage is a result of the applied strains (axial, transverse, and shear) and the sensitivity of the gage to each of these strains. This is expressed as: aattsatRSSSR (1.5) where S a S t and S s are the sensitivities of the gage to the applied axial, transverse and shearing strains and a t and at are the applied axial, transverse, and shearing strains.

PAGE 49

24 The sensitivity of the gage to the shearing stress is negligible and is usually ignored. Thus reducing equation 1.5 down to (aattRSKR ) (1.6) where K t = S t /S a (transverse sensitivity factor; %). By assuming the effects of K t and t are negligible; the gage calibration factor or gage factor can be determined. The gage factor as a function of the applied axial strain is expressed as g aRSR (1.7) where S g is the gage factor. This factor is used to convert the resistance changes observed in the gage to the applied strain seen by the gage. The amount of error that occurs if only the gage factor is considered in determining the strain is expressed as ()1001ttaootKK (1.8) where o is the Poissons ratio of the material being tested. If you assume applied axial and transverse strain are the same ( t / a = 1, usually a > t t / a < 1), a Poissons ratio of 0.285 (Poissons ratio for calibration beam) and a K t of -0.006 (-0.6%, K t for strain gages utilized during dissertation work) the error associated with transverse strain is 0.77%. The strain gages utilized for this work were directly bonded to the conducting polymer being studied (w/ evaporated gold interface). So the Poissons ratio and the applied strain field are both factors of the conducting polymer being studied. During

PAGE 50

25 electrochemical growth of conducting polymers there is no axial or transverse orientation (to the gage element) of the polymer so during redox cycling the induced strain field would be uniform in both the axial and transverse direction (to grid element). By assuming this both the t / a and o terms become one (applied strain field and expansion of CP is biaxial). Assuming this the calculated error associated with the transverse strain would be -1.19%. 2.3 Anti-Fouling/Foul-Release Coatings Biofouling is the result of marine organisms settling, attaching, and growing on submerged marine surfaces. The biofouling 75 process is initiated within minutes of a surface being submerged in a marine environment by the absorption of dissolved organic materials (conditioning film). Once this conditioning film is deposited, bacteria (unicellular algae) will colonize the surface within hours of submersion. The resulting biofilm produced from the colonization of the bacteria is referred to as microfouling or slime and can reach thicknesses on the order of 500 m. Macrofouling species may eventually colonize on top of the microfouling or slime layer. Soft and hard fouling are two different classifications of macrofouling. Soft fouling consists of algae and invertebrates such as anemones, hydroids, soft corals, sponges, and tunicates. Hard fouling consists of invertebrate species like barnacles, muscles, and tubeworms. 50 This results in a multilayer structure with each layer having its own unique properties and adhesion mechanisms. To further complicate things, the species makeup for each layer is dependent on the geographical location and the makeup of the underlying structure. This issue is further compounded due to the fact that there are 12 well-defined geographical zones in the

PAGE 51

26 worlds oceans with varying salinity, clarity, temperature, amount and type of micronutrients, and number and type of fouling organisms. 52 Figure 2.6 Diagram showing the different species breakdown and layering of biofoul film formation. Biofouling is estimated to cost the US Navy alone over $1 billion per year by increasing the hydrodynamic drag of naval vessels. 50, 51 This in turn decreases range, speed, and maneuverability of naval vessels and increases the fuel consumption by up to 30-40%. 52, 53 Anti-fouling and foul-release coatings are two main approaches used for combating biofilm formation. Anti-fouling coatings prevent or deter the settling of biofouling organisms on a surface by the use of leached biocides. Foul-release coatings control biofilm formation by modifying surface properties in such a way as to prevent the

PAGE 52

27 formation of strong adherent bond between the biofoulant and the surface. This reduces the work required to remove them. The first method is typically accomplished through the use of anti-fouling coatings containing heavy metal biocides, such as cuprous oxide or tributyltin (chloride and oxide), which are either tethered to the coated surface or are released from the surface into the surrounding environment. Use of these coatings has caused problems in the marine ecosystem, especially in shallow bays and harbors where the biocides can accumulate. As such the use of tributyltin has been banned in many parts of the world and its use will be completely banned worldwide on all naval vessels by the International Maritime Organization in January 2008. 50, 51 The second concept for controlling biofouling is the use of foul-release coatings. Foul-release coatings control biofilm formation through the use of engineered surfaces with controlled surface properties such as surface energy, modulus, and roughness to minimize biofoulant adhesion. 50, 52, 76-79 More recently the use of nano and micro-scale topographies has come into interest. 51, 80-82 The interest in the use of surfaces with controlled/tailored surface energies for foul-release coating is due to the fact that wetting of the surface with biological glue (or any other fluid) is controlled by the surface tension/energy. By controlling the surface energy of a material the wettability of its surface and therefore bond formation can be controlled. 52 Work done by Pasmore and Bowman 76 has shown that the percent removal of biofilms, Pseudomonas aeruginosa under given conditions, increases with a decreased in contact angle (increase in surface energy) and decreases with an increase in surface roughness. Similar results were produced on smooth glass, electropolished 316 stainless

PAGE 53

28 steel, and PTFE samples which indicated a biofilm accumulation ~35% lower than for rougher surfaces. 77 Another factor of interest is the modulus of the material being used. Gray and Loeb 78 have shown that the degree of settlement of various organisms decreased with a decrease in modulus of crosslinked PDMS samples. From this it can be said that the force required to remove an adhered biofilm is a factor of the settled surfaces surface energy and surface modulus, this is in agreement with the Kendal theory 52 : 34/(1CaPEwa 2)v (1.9) 24/(1Ca ) E wav (1.10) which gives the critical pull-off force (P C ) and critical crack propagation stress ( C ) as a function of elastic modulus (E) and the work of adhesion (w a ; w a = 2). The work of adhesion is related to the surface energy by the interfacial tension (), where a is the contact radius and is the Poissons ratio. Another current area of interest is the use of micro-topographies in the control of biofoulant spore settlement and adhesion. 51, 80-82 By manipulating surface topography, through the use of micropatterning, the overall surface energy and therefore hydrophilicity or hydrophobicity of a coating material can be manipulated. Many studies have shown the importance of surface roughness on the settling of spores. However the importance of surface feature size and shape has begun to be studied only recently. Work done by Verran and Boyde (2001) has shown that macro-scale surface features (>10 m) are relatively unimportant in cell settlement since cell dimensions are much smaller than the surface features. It has been suggested that the shape, scale, and periodicity of surface features may influence the settlement of barnacle larvae (Hills & Thomason,

PAGE 54

29 1998; Lapointe & Bourget, 1999; Berntsson et al.,2000), as reviewed by Callow et al.. 80 Work by Brennan and coworkers 80, 81 has shown that the settlement of spores can be influenced by the use of ridge and pillar micro-topographies on PDMSe surfaces with feature sizes and spacings on the order of 1.5-20 m. In this study the minimum spacing between features was 5m which was similar in size to the diameter of the cells being studied, therefore the cells settled into the valleys between the features. From this work it was determined that the pattern spacing should be reduced to 2-3 m to increase the work required to settle on the patterned surface. Recently a new biomimiticly engineerd surface topography (B.E.S.T. or sharklet) utilizing 2-3 m feature sizes has been developed. This pattern has been shown to reduce the settlement of Ulva spore (green algae) by ~85%. Similar results were found for flocked surfaces by Kim et al.. 51 They found that by flocking a smooth PVC surface with a heterogeneous mixture of nylon fibers (90% 1.8 denier and 1.27 mm long; 10% 3.0 denier and 2.54 mm long) that they could influence the settlement and growth of different marine species. However they found that the flocked surface inhibited the growth of some species but had no effect or even enhanced the growth of other species. From this it can be said that surface features play an important role in the settlement and adhesion of biofouling organisms to a particular surface. However, it has been shown that different factors influence different species in different ways, i.e., what inhibits the growth of one species might enhance the growth of another. This makes it unlikely that a single coating with a fixed set of surface parameters will be effective in preventing biofilm formation for all species. Therefore a surface coating with

PAGE 55

30 dynamically variable surface properties (surface energy, modulus, and roughness) would be more effective in the prevention or retardation of biofilm formation. 2.4 Electrowetting Electrowetting is the process of changing the surface wetability (surface tension) of a metal electrode by rearrangement and or formation of an electronic double layer (EDL) at the electrodes surface due to an applied electrical potential. The electrowetting (EW) process has been extensively studied for pure metal electrodes with electrolyte solutions. 83-85 EW devices have been limited to uses in polar media due to the nature of the formed EDL. The EDL is formed from the transfer of electrons from the electrode to redox-active species in the fluid medium. The electrical stability of the EDL limits the use of these devices to low voltages, as low as ~1 V. 86, 87 However the induced change in contact angle () is proportional to the amount of charge developed at the electrode surface thereby limiting the overall that can be produced. Two major applications for this technology are in micro-fluidic devices and MEMS type applications. A schematic of an EW capillary pump is depicted in Figure 2.7. 87 Figure 2.7 Design of electrowetting device: (a) no applied electrical potential (hydrophobic surface); (b) with applied electrical potential (hydrophilic surface). Fluid is pumped by continuously cycling the applied electrical potential. 87

PAGE 56

31 Recently, it has been shown that the application of a thin dielectric layer (PTFE, SiO 2 etc.) between the electrode and fluid can enhance this EW behavior. This allows the pumping/wetting of virtually any fluid medium. This behavior is referred to as electrowetting-on-dielectric (EWOD). However a higher electrical potential is required to drive these systems. 86-89 Typical operating voltages for EWOD devices typically exceed 100-200 V. Yet, more recently, Moon and coworkers produced EWOD devices operating as low as 15V. 86 EWOD devices also are resistant to corrosion due to the protection offered to the electrode by the dielectric layer and can produce larger when a hydrophobic dielectric (higher initial contact angle, lower initial surface energy) is used (i.e., PTFE) when compared to EW based devices. The dielectric layer blocks electron transfer from the electrode to the fluid medium; however it sustains the high electric field at the interface due to charge redistribution when a potential is applied. The relationship between the applied electrical potential (V) and the resultant surface tension () are expressed in Lippmans equation: 2012cV (1.11) which when combined with the Youngs equation: cosSLSGLG (1.12) can yield the resultant contact angle () according to Lippman-Youngs equation: 2011coscos2LGcV (1.13) Where 0 and 0 are the surface tension and contact angle of the solid-liquid interface when there is no electrical field across the interface layer (surface tension and

PAGE 57

32 contact angle at point of zero charge), SL SG and LG are the solid-liquid, solid-gas, and liquid-gas surface tensions, V is the applied electrical potential, and c is the specific capacitance of the dielectric layer ( 0/c t ; F/cm 2 ) where 0 is the permittivity of a vacuum, is the dielectric constant of the dielectric layer, and t is the dielectric layer thickness. The Lippman-Youngs equation predicts that the overall can be increased by increasing the applied electrical potential and the dielectric constant of the material being used and by decreasing the dielectric layer thickness. Therefore the thinner the dielectric layer used the lower the electrical potential required to induce a given However the dielectric breakdown voltage is proportional to the thickness of the dielectric layer. Therefore at a certain dielectric thickness the required electrical potential to induce a given will exceed the dielectric breakdown potential. Thus, there is a thickness limit on the dielectric layer which is dependent on the dielectric material used. An example of this was shown by Moon et al.. 86 Moon and coworkers calculated that the voltage required to induce a 40 for a Teflon AF based EWOD device ( = 2.0 and E breakdown = 2x10 16 V/cm ) as a function of dielectric layer thickness. For dielectric layer thicknesses of less than 0.2 m the voltage required to induce a 40 change in was higher than the dielectric breakdown potential of the Teflon AF (see Figure 2.8). In order to prevent the dielectric breakdown of thin film dielectrics Moon and coworkers deposited a 700 layer of barium strontium titanate between the platinum electrode and the 200 Teflon AF outer layer. This resulted in a EWOD device

PAGE 58

33 capable of operating at electrical potentials as low as 15V and generating a on the order of 40 (from 12080). 86 Figure 2.8 Voltage required to induce a of 40 (from 12080) versus dielectric layer thickness for Teflon AF based EWOD device, with = 2.0 and E breakdown = 2x10 16 V/cm. 86 2.5 Dynamic Surfaces The volume expansion and contraction experienced by conducing polymers during electrochemical switching between there oxidized and reduced states is also associated with other physical property changes in the system. During the oxidation process the developed positive charge induced along the CP backbone will also result in a change in the surface charge or energy of the system. The influx of counter ions and associated solvent that drives the volume expansion during oxidation also results in a drop in the modulus of these materials. By incorporating CPs into other more flexible and durable base materials (such as elastomers, thermoplastic elastomers, thermoplastics, etc.) the dynamic properties of the CP can be utilized to produce coatings with dynamic surface properties.

PAGE 59

34 Changes in surface energy have been studied under both chemical 90, 91 and electrochemically 92, 93 induced redox systems for PPy, polyaniline (PANI), poly(3-(pyrrolyl)-alkanoic acid), poly(3-octylthiophene) (P3OT), and poly(3-hexylthiophene) (P3HT). The results of these studies demonstrate that oxidizing the conducting polymer decreases the Sessile drop contact angle (increase surface energy) of water. An example of this behavior was reported by Gregory et al.. 91 They found that by chemically oxidizing and reducing PANI, PPy, and P3HT contact angle changes of 36.5, 16.8, and 31.5 could be induced. However Bartlett et al. 94 stated that by using functionalized thiophenes, i.e., alkanoic acids such as carboxylic, butanoic, and pentanoic, a higher contact angle (= ~ 8, 20, 30 respectively) is measured for the oxidized state. This is attributed to the protonation of the alkanoic acids (pKa = 6.5, 5.8, 6.1 respectively) at a low pH (oxidizing environment). By utilizing various conducting polymer systems it was hypothesized that it should be possible to tailor particular systems to produce desired surface properties changes, compatibilities, and redox switching speeds. The relative differences in oxidation potential of various conducting polymers are shown in Figure 2.9. PPy P3MeT PPP oxidized reduced (neutral) (-) (+) Figure 2.9 Relative surface charge of different conducting polymers. 2.5.1 Polypyrrole To date a majority of the literature work conducted on conducting polymers has been on polypyrrole (PPy) (Figure 2.11 A). PPy is characterized by high stability in its

PAGE 60

35 oxidized form due t o its oxidation potential ~ -0.2 V which is close to the O2 reduction poteny n 6in the oxidized as well as reduced states.5, 39, 55, 90, 91, 95, 96, 102-108 They also ctrical, optical, and redox properties. The thiophene monoulting idation tial at ~ -0.2-0.3 V. 95 Therefore neutral PPy will be oxidized by O 2 to form its oxidized conducting form when exposed to air. PPy has gained a lot of exposure recentlin the field of conducting polymer based artificial muscles due to its ability to produce alarge volume expansion (~1-3% longitudinally and ~35% in thickness on a bound surface 49 ) when redox cycled between the oxidized and reduced state. This large volume change induced during redox cycling will be beneficial in the modification of the PPy-PDMS surface modulus. PPy can be synthesized chemically or electrochemically ivarious media. The chemical polymerization can be facilitated in the presence of Lewis Acids such as FeCl 3 or ammonium persulfate along with codopants such as NaClO 4. 55, 9101 2.5.2 Poly(3-methylthiophene) Polythiophene and its derivatives have received a lot of attention lately due to their stability possess many highly desirable ele mer can easily be derivatized using a number of chemistries. It has been shown that by changing the substituents on the thiophene ring the oxidation potential of the resmonomer and polymer can be varied between 1.20-2.00 V and 0.70-1.45 V respectively. 96 Poly(3-methylthiophene) (PMeT) (Figure 2.11 B) in particular has an oxidation potential of ~0.8 V and reaches a fully reduced state at ~0.2 V vs Ag/AgCl. These values lie well above the O 2 reduction potential and above the H 2 O oxpotentials allowing for good stability in both forms. This can be seen vs. SCE (not Ag/AgCl) along with the oxidation potentials for PPy and PPP in Figure 2.10. 95 PMeT can also be polymerized in a similar fashion as PPy.

PAGE 61

36 Figure 2.10 Oxidation (-) and reduction (--) potentials of poly pyrrole (PPy), polyaniline (PA), poly(3-methylthiophene (PMeT), and poly(p-phenylene) (PPP).95 2.5.3 Poly(p-phenylene) One of the disadvantages of PPy is its high stability in its oxidized (charged) form. enylene) (PPP) (Figure 2.11 C) on the other hand exhibits excepe eing e This hinders the return of the PPy-PDMSe surface back to its original (unchargred) surface energy. Poly(p-ph tional stability in its neutral form. The oxidation potential of PPP is around +1.2 V which is very close to the oxidation potential of water, therefore water will reduce thoxidized form of PPP to its more stable neutral form. 95 PPP is also characterized as bhighly crystalline, difficult to process, insoluble, and exhibiting high resistance to oxidation, radiation, and thermal degradation. 55, 96 PPP can be synthesized from benzenin the presence of Lewis acid such as FeCl 3 (~70 C) and AlCl 3 (~37 C) along with an additional oxidizing agent (codapoant). 55, 96, 109-114

PAGE 62

37 NHSNHSABC Figure 2.11 Monomer and polymer structures for A) polypyrrole, B) poly(3-methylthiophene), and C) poly(p-phenylene).

PAGE 63

CHAPTER 3 INITIAL IN SITU EVALUATION OF CPS VIA STRAIN GAGE TECHNIQUE 3.1 Introduction Strain sensitive actuators were developed utilizing commercially available strain gage technology. This is the first time sensors of this type have been used in the in situ evaluation of conducting polymer based actuators. 56 Strain sensitive actuators were used to evaluate the in situ strain response of PPy/TOS based actuator films during redox cycling using cyclic voltammetry and square-wave potential stepping. The strain responses of PEDOP and PBEDOT-Cz were also evaluated and compared to PPy/TOS. It is believed that PPys high strain response is due to its ability to crosslink through the 3, 4 positions on the monomer structure. 64, 104 By placing substituents in these positions the degree of crosslinking can be controlled. PEDOP was chosen as a candidate for study due to its blocked 3, 4 positions, which should produce a linear polymer structure with a minimal degree of crosslinking. PBEDOT-Cz has been shown to undergo a non-reversible oxidative reaction at about 1.15 V, which has been attributed to a possible crosslinking reaction. 107 This ability to potentially crosslink was the reason that PBEDOT-Cz was chosen for comparison. 3.2 Materials and Methods 3.2.1 Materials Pyrrole was purchased from Sigma-Aldrich and was filtered through neutral alumina (Brockman activity 1; Fisher Scientific) until colorless before use to remove any impurities. 3,6-bis(2-(3,4-ethylenedioxy)thienyl)-N-carbazole (BEDOT-Cz) and 3,438

PAGE 64

39 ethylenedioxypyrrole (EDOP) used in this study were synthesized by the Reynolds Research group (University of Florida Chemistry Department). 107, 115, 116 Acetonitrile (ACN), lithium perchlorate (LiClO 4 ), sodium perchlorate (NaClO 4 ), p-toluenesulfonic acid Na salt (NaTOS), and tetrabutylammonium were purchased from Sigma-Aldrich and used as received. Sodium nitrate (NaNO 3 ) was purchased from Mallinckrodt Chemicals and sodium sulfite (Na 2 SO 3 ) was purchased from Fisher Scientific, both were used as received. Deionized water (18 M, Millipore system) was used in all experiments and the solutions were deoxygenated by bubbling argon prior to electropolymerization and redox switching of the CP. 3.2.2 Strain Gages Electrical resistance strain gages (CEA-06-500UW-120 and EA-06-20BW-120; Figure 2.4) were purchased from Vishay Measurements Group Inc. All strain gages were cleaned prior to Au deposition and use with ethanol to remove any surface oils and debris. These gages were treated with various Au treatments and then coated with the appropriate CP to form strain sensitive actuators. A Vishay Measurements Group Inc. P3500 strain indicator was used to measure the resistance changes in the strain gage and thereby produce the strain measurement. EvAu coated strain gages (no CP) that were exposed to electrolyte solutions and potentiostatically cycled gave a zero strain reading. This was done to verify that the strain gages were unaffected by the potential cycling during redox switching and that all strain measurements acquired were the result of the CPs and not the strain gage. 3.2.3 Conducting Polymer Synthesis In the initial study EvAu was vapor deposited thermally onto strain gages to a thickness of ca. 700 (measured by quartz microbalance). These gages were used as the

PAGE 65

40 working electrode during electropolymerization and redox switching of the CPs. All electrochemical work was carried out utilizing an EG&G Princeton Applied Research 273A potentiostat/galvanostat utilizing the CorrWare software package, a platinum foil counter electrode and an Ag/AgCl (BAS MF2052) reference electrode at room temperature. All potentials given are relative to this reference electrode. Conducting Cu tape (1181, 3M) was utilized to make electrical connections between the EvAu layer and the working electrode lead from the potentiostat. Polypyrrole (PPy) was synthesized potentiostatically (E = 0.65 V, t = 500 s) from an aqueous solution of 0.2 M pyrrole, 0.1 M p-toluenesulfonic acid Na salt (NaTOS), and 1.0 M lithium perchlorate. This resulted in a film thickness of ca. 9.6 m measured by a Sloan Dektak 3030 profilometer. PEDOP was synthesized potentiostatically (E = 0.6 V, t = 10,000 s) from an aqueous solution of 0.01 M 3,4-ethylenedioxypyrrole, 0.1 M p-toluenesulfonic acid Na salt, and 1.0 M lithium perchlorate. The resulting film thickness was 10.6 m. PBEDOT-Cz was synthesized potentiostatically (E = 0.8 V, t = 3000 s) from an aqueous solution of 0.01 M Bis-EDOT-Cz, 0.1 M acetonitrile, and 1.0 M tetrabutylammonium. The corresponding film thickness of PBEDOT-Cz was 9.9 m. 3.3 Polypyrrole (PPy/TOS) Results 3.3.1 Determination of Electropolymerization Conditions for PPy/TOS In order to obtain the optimal strain response from the strain sensitive actuators the PPy/TOS electropolymerization conditions were first evaluated on a gold-button working electrode (MF-2014 Au electrode (AUE), Bioanalytical Systems, Inc.). PPy/TOS films were prepared by cyclic voltammetry (.8 V to 1.0 V) in an aqueous solution of 0.2 M pyrrole, 0.1 M NaTOS and 1.0 M LiClO 4 The pyrrole oxidation peak was observed at 0.65 V, which corresponds to the optimal electropolymerization voltage for these

PAGE 66

41 electrochemical conditions. The adhesion of the PPy/TOS film to the gold-button working electrode was tested using a tape (810, 3M) peel off method. PPy/TOS films electropolymerized at a constant potential of 0.65 V produced the most adherent films, while films polymerized at voltages exceeding 0.8 V exhibited very poor adhesion to the electrode. 3.3.2 PPy/TOS Cyclic Voltammetry and Strain Response Utilizing these electropolymerization conditions (E = 0.65 V) a PPy/TOS film was deposited on a Au coated strain gage (CEA-06-500UW-120) for 500 seconds and resulted in a film thickness of 9.6 m (Sloan Dektak 3030). PPy/TOS films have been shown to produce films of high conductivity (150 S/cm) and tensile strength (73.4 MPa). 61, 117 A representative SEM micrograph of PPy/TOS is shown in Figure 3.1. Figure 3.1 SEM micrograph of surface morphology of PPy/TOS film prepared in 1.0 M LiClO 4 at a potential of 0.65 V. SEM image taken of an uncoated sample at 1000X and 15 KeV. PPy/TOS samples were scanned from -0.8 V to 0.4 V at 10 mV/s in 1.0 M aqueous LiClO 4 The resulting strain response measured by the strain sensitive actuator followed

PAGE 67

42 the cyclic voltammetry data (Figure 3.2). These results were obtained during the 5th potential cycle to attain reproducible cyclic voltammetry and strain data. The strain response was collected manually at 0.2 V intervals while the cyclic voltammogram was acquired directly from the potentiostat. The PPy/TOS cyclic voltammetry (Figure 3.2a), on the strain gage, is relatively broad due to the high surface area of the working electrode (ca. 100 mm 2 ) and large film thickness (9.6 m). The PPy reduction peak is centered at -0.2 V. The overall change in strain () for the PPy/TOS systems was on the order of 236 Upon further examination (Figure 3.2b) it is evident that a small initial decrease in the strain response (contraction) is present, starting at .8 V, and is followed by a much larger increases (expansion) in strain as the oxidation potential is reached. This is similar to results previously published in literature 37 and is evident in all samples of PPy/TOS, PEDOP, and PBEDOT-Cz under cyclic voltammetry. It is also evident that there is hysteresis in the strain response. It is believed to be due to ion and solvent concentrations, in the CP, never reaching equilibrium under these scan conditions (10 mV/s). 44 This also causes the strain response to initially continue to rise on the reverse scan, resulting in the maximum strain response to be observed at 0.2 V instead of 0.4 V. Similar results have been obtained during electrochemical quartz crystal microbalance (EQCM) experiments conducted with PPy. These results are attributed to the diffusion of the dopant ions and the corresponding solvent molecules into and out of the polymer structure during redox cycling.

PAGE 68

43 Figure 3.2 (a) PPy/TOS cyclic voltammetry ( = 10 mV/s) and (b) in situ strain response of a 9.6 m film prepared in aqueous 1.0 M LiClO 4 3.3.3 PPy/TOS Multi-Cycle Strain Response Strain sensitive actuators were also used to acquire multi-cycle strain data as shown in Figure 3.3. It is evident from the multi-cycle data that the strain response does not return to its initial starting value, but has a positive drift. The difference in the initial starting points for the 2 nd and 5 th cycle is on the order of 100 and decreases with each

PAGE 69

44 cycle. However the overall change in strain for each cycle is reproducible on the order of 200 The drift in cycle measurements is attributed to the film possibly not returning to the same state of solvation from cycle to cycle. It has also been suggested that the drift could be part of the break-in process for these materials (i.e., the film remembers it is bent and there is some molecular rearrangement). This break-in process is evident during the 1 st cycle (Figure 3.3) by the very erratic strain response, which starts to normalize after a couple of scans. Figure 3.3 In situ multi-cycle cyclic voltammetry strain response of a 9.6 film prepared in aqueous 1.0 M LiClO 4 3.3.4 PPy/TOS Square-Wave Potential Experiments Square-wave potential experiments were also utilized to evaluate the strain response of PPy/TOS. PPy/TOS films were stepped from .8 V to 0.4 V and then back to .8 V five times with a 50 s hold time at each potential (Figure 3.4). These stepping

PAGE 70

45 experiments resulted in a reproducible of 70 The minimum and maximum strain values were collected manually at .8 V and 0.4 V respectfully. Time (s) 0100200300400500 Micro Strain () 020406080 E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.6 Micro Strain ( Voltage (V) Figure 3.4 In situ square-wave strain response of a 9.6 m film prepared in aqueous 1.0 M LiClO 4 The overall change in the strain response obtained by square-wave voltammetry is less than that obtained by cyclic voltammetry; however the time required to complete one cycle from each experiment is different. The time required to complete one cycle is 240 s for cyclic voltammetry while the time required to complete one cycle by square-wave stepping is 100 s. The strain produced by oxidative doping of these devices is diffusion limited. Therefore longer cycle times allows more ion and solvent molecules to diffuse into and out of these films (i.e., higher degree of oxidative doping), which produces a higher overall degree of strain.

PAGE 71

46 3.4 PEDOP Results 3.4.1 Electrochemical Analysis of PEDOP Poly(3,4-ethylenedioxypyrrole) (PEDOP) was recently introduced as a conducting polymer with a very low redox switching potential and a high stability to repeated redox switching in aqueous electrolytes. 115, 118 Initial electrochemical evaluations were conducted on a gold-button electrode as with PPy/TOS. PEDOP electropolymerized potentiostatically from aqueous solutions of 0.01 M EDOP, 0.1 M NaTOS, and 1.0 M LiClO 4 at 0.5 V produced nicely adherent and electroactive films. As with PPy/TOS, PEDOP electropolymerized at potentials greater than or equal to 0.8 V produced film with poor adhesion, these films spontaneously delaminated from the gold-button electrode. PEDOP films produced at 0.5 V were electrochemically characterized by cyclic voltammetry, scanned at 100 mV/s from 0.0 V to .2 V, in aqueous 1.0 M LiClO 4 The resulting cyclic voltammetry is shown in Figure 3.5. From the cyclic voltammetry it is evident that PEDOPs redox switching potential is much lower than that of PPy/TOS. PEDOP has a measured half-wave potential (E 1/2 ) of .6 V. This low redox switching potential is attributed to the electron donating alkoxy substituents producing a highly electron-rich polyhetrocycle backbone. This electron-rich nature produces high stability to air and aqueous electrolytes in the doped and conducting form. This allows PEDOP to be held in the oxidized (conducting) form for extended periods of time without degradation and also leads to the high cyclic stability of this material. 3.4.2 PEDOP Multi-Cycle Strain Response EDOP was electropolymerized potentiostatically at E = 0.60 V for 10,000 s from a aqueous solution of 0.01 M EDOP, 0.1 M NaTOS, and 1.0 M LiClO 4 to produce a

PAGE 72

47 PEDOP film of 10.6 m on a Au coated strain gage. A representative SEM micrograph of the surface morphology of PEDOP is shown in Figure 3.6. As can be seen the surface morphology of PEDOP films is comparable to that of PPy. The strain response of PEDOP during cyclic voltammetry was evaluated at a scan rate of 10 mV/s from .8 V to 0.0 V in an aqueous solution of 1.0 M LiClO 4 The resulting strain response (Figure 3.7) is considerably less than that attained from PPy/TOS with a that ranged from 36 to 47 with an average of 42.6 for the 2 nd through 10 th cycles. The same data has been replotted against a time axis, in Figure 3.8, to improve resolution of the strain response from cycle to cycle. Figure 3.5 Cyclic Voltammetry ( = 100 mV/s) of a PEDOP film produced from aqueous 1.0 M LiClO 4 at E = 0.5 V and t = 200 s.

PAGE 73

48 Figure 3.6 SEM micrograph of a PEDOP film prepared at 0.6 V in 1.0 M LiClO4. SEM image taken of an uncoated sample at 1000X and 15 KeV. E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.2 Micro Strain -60-40-200204060 1st cycle 2nd cycle 3rd cycle 4th cycle 5th cycle 6th cycle 7th cycle 8th cycle 9th cycle 10th cycle Figure 3.7 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of a 10.6 m PEDOP film in aqueous 1.0 M LiClO 4

PAGE 74

49 Time (s) 02004006008001000120014001600 Micro Strain () -60-40-200204060 E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.2 Micro Strain () Voltage (V) Figure 3.8 In situ multi-cycle cyclic voltammetry (10 mV/s) strain response of 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 Data from Figure 3.7 has been replotted vs. time. 3.4.3 Effects of Cyclic Scan Rate on the Strain Response of PEDOP The effects of scan rate were also evaluated. PEDOP films were electropolymerized as previously stated and then scanned at 100 mV/s by cyclic voltammetry from 0.0 V to .8 V in aqueous LiClO 4 The resulting strain response was significantly less than that obtained at 10 mV/s. Due to the speed of the redox switching, data was only collected at each end of the potential scan (i.e., 0.0 V and .8 V). The average was 6.6 for the first 10 cycles (Figures 3.9 and 3.10). However iot should be noted that the 1st and 6 th cycle both produced a of while the rest of the cycles all had values between 7 and 9 This dramatic decrease in the strain response is expected due to the diffusion limited swelling nature of this and other conducting polymers. The scan rate of this experiment was conducted 10 times faster than the previous experiment, thus not allowing for full oxidative doping of the polymer network.

PAGE 75

50 Figure 3.9 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of a 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 Figure 3.10 In situ multi-cycle cyclic voltammetry (100 mV/s) strain response of 10.6 m PEDOP film in aqueous 1.0 M LiClO 4 replotted vs. time.

PAGE 76

51 3 .4.4 PEDOP Square-Wave Potential Experiments ted on PEDOP. Using the electr to d earlier, it is believed that a cross-linked structure, like PPy, will develop a highe Square-wave potential experiments were also conduc opolymerization conditions stated above a 10.6 m film was stepped from 0.8 V0.0 V with a 50 s equilibrium hold at each voltage. The resulting square-wave strain response exhibited a that varied from 11 to 33 with an average value of 21 The strain response with the corresponding potential sweep is plotted versus time in Figure 3.11. As state r degree of strain/swelling than a non cross-linked structure. The cross-links act to tie the whole polymer network together; therefore a small change (strain due to swelling) in one section of the network will exert an affect (strain) on the rest of the networked structure. Figure 3.11 In situ square-wave strain response of a 10.6 film prepared in aqueous 1.0 M LiClO4.

PAGE 77

52 During anion insertion (oxidation) the cross-linked polymer network would be highly strained due to inability of the polymer chains to separate/relax through simple chain motion. By blocking the 3and 4positions on the pyrrole ring PEDOP cannot form cross-links and forms a linear structure through the 2and 5positions. This lack of cross-linking would allow the polymer network to relax through simple chain motions during ion transport into and out of the structure (during redox switching). Therefore anion insertion into the polymer network would result in a much lower degree of overall strain than in a cross-linked network. However with most things, this can be overdone. As the cross-link density of the network increases the stiffness of the network also increases. Thus requiring an increased amount of force (potential driving force) to develop the same amount of strain in the system. As the cross-link density of the polymer network increases the diffusion rate of counter ions in the system will decrease. Also as the cross-link density increases the modulus increases and the film can become too rigid to deflect/strain resulting in zero strain development and therefore no actuation. 3.5 PBEDOT-Cz Results 3.5.1 PBEDOT-Cz Introduction Poly [3,6-2(2-(3,4-ethylenedioxythienyl)-carbazole] (PBEDOT-Cz) and other bis-EDOT derivitized carbazoles were found to be of interest, by Reynolds et al., for there ability to form three distinct redox states (neutral, cation-radical, and di-cation forms) at low potentials and to be stable to thousands of redox switches. 107 These low switching potentials are attributed to the electron-rich biEDOT structure in the PBEDOT-Cz polymer repeat structure. Its was found during electrochemical analysis that a irreversible higher potential redox process at 1.15 V (Figure 3.12) was present during the first cyclic voltammetry scan but was absent for all subsequent scans. This process could

PAGE 78

53 not be attributed to over oxidation (breakdown) of the polymer backbone. This was theorized by Reynolds et al. to be a possible cross-linking reaction, however no spectroscopic evidence of cross-linking has been reported to date. PBEDOT-Cz was chosen as a candidate for evaluation due to this possible cross-linking reaction. It was theorized that the cross-linking of PBEDOT-Cz would lead to an improved strain response in the material when compared to non-cross-linked materials. This was based on the observation that PPys strain response is related to its crosslinked structure. 64, 104 Figure 3.12 Cyclic voltammetry (100 mV/s) of PBEDOT-Cz in 0.1 M TBAP/CAN. Adapted from Figure 8 of Reynolds et al.. 107 2n d scan Potential cross-linking rxn (1 s t scan) 3.5.2 PBEDOT-Cz Electrochemical Conditions The effects of possible cross-linking at 1.15 V 107 in PBEDOT-Cz was evaluated in samples that were electropolymerized from a solution of 0.01 M BEDOT-Cz and 0.1 M tetrabutlyammonium perchlorate in acetonitrile. Samples were cycled from -0.8 V to 0.6,

PAGE 79

54 1.0, and 1.2 V at 10 mV/s in 1.0 M aqueous LiClO 4 successively. Data was collected for the 1 st 2 nd 5 th and 10 th scans at 0.2 V intervals. 3.5.3 PBEDOT-Cz Strain Response (-0.8 V to 0.6 V) The cyclic strain response acquired from .8 V to 0.6 V exhibited a very irregular shape for the first four scans but then normalized after the 5 th scan. This is due to the normal break in period for these materials, as discussed earlier for PPy and PEDOP. The normalized cyclic strain data for the 1 st 2 nd 5 th and 10 th cycles (Figure 3.13) exhibits a range from 15 to 21 The average change in strain was 17. E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.60.8 Micro Strain () -20-15-10-505101520 1st cycle 2nd cycle 5th cycle 10th cycle Figure 3.13 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film (-0.8 V to 0.6 V) for the 1 st 2 nd 5 th and 10 th scans in 1.0 M LiClO 4

PAGE 80

55 E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.01.2 Micro Strain () -30-20-100102030 1st cycle 2nd cycle 5th cycle 10th cycle Figure 3.14 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film (-0.8 V to 1.0 V) for the 1 st 2 nd 5 th and 10 th scans in 1.0 M LiClO 4 3.5.4 PBEDOT-Cz Strain Response (-0.8 V to 1.0 V) The cyclic strain data acquired from .8 V to 1.0 V was very regular in shape and exhibited its max strain at .2 V on the reverse scan as shown in Figure 3.14. The measured for this set of scans ranged from 28 to 37 with an average value of 33 This is a change of 16 and 0.6 V between the average measured for the .8 V to 0.4 V and .8 V to 1.0 V scans. 3.5.5 PBEDOT-Cz Strain Response (-0.8 V to 1.2 V) The strain data acquired from .8 V to 1.2 V is also very regular in shape when compared to the scans obtained from .8 V to 0.4 V. This set of data was obtained above the potential cross-linking potential of 1.15 V for PBEDOT-Cz, however there are no significant visual differences in the shape or measured values of the strain data from above and below this potential cross-linking voltage. The ranged from 34 to 35

PAGE 81

56 with an average value of 34.75 for the 1 st 2 nd 5 th and 10 th scans. These scans are shown in Figure 3.15. This is a change of 1.75 and 0.2 V between the average measured for the .8 V to 1.0 V and .8 V to 1.2 V scans. This data does not support the argument of potential crosslinking of PBEDOT-Cz at potentials above 1.15 V. E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.01.21.4 Micro Strain () -20-100102030 1st cycle 2nd cycle 5th cycle 10th cycle Figure 3.15 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film (-0.8 V to 1.2 V) for the 1 st 2 nd 5 th and 10 th scans in 1.0 M LiClO 4 3.5.6 Overall Results for PBEDOT-Cz The overall change in strain for the 0.6, 1.0, and 1.2 V scans ranged from 15-21, 28-37, and 34-35 with average values of 17, 33, and 34.75 respectively. These values are significantly lower than those obtained from PPy. However there is no significant difference in the data obtained from the .8 V to 1.0 V and .8 V to 1.2 V scans (below and above potential cross-linking voltage). The normalized strain response of the 5 th and 10 th cycles for all three potential ranges is shown in Figures 3.16 and 3.17 respectfully.

PAGE 82

57 The slight increase in the strain response between the scans from .8 V to 1.0 V and 1.2 V is most likely due to the increased driving potential of 0.2 V. This combined with the fact that no spectroscopic evidence of cross-linking for PBEDOT-Cz could be found by Reynolds et al., 107 leads to the conclusion that the PBEDOT-Cz cross-linking reaction is non-existent or so minimal that no enhancement in the strain production is evident. However the overall strain response of PBEDOT-Cz is comparable to that of PEDOP. This is due to the inability of PBEDOT-Cz to crosslink as with PEDOP. This low strain response combined with the cost to produce the BEDOT-Cz monomer makes PBEDOT-Cz a poor candidate for actuator construction. E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.01.21.4 Micro Strain () -20-100102030 -0.8V to +0.6V -0.8V to +1.0V -0.8V to +1.2V Figure 3.16 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film scanned from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (5 th scan) in 1.0 M LiClO 4

PAGE 83

58 E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.01.21.4 Micro Strain () -20-100102030 -0.8V to +0.6V -0.8V to +1.0V -0.8V to +1.2V Figure 3.17 Normalized in situ cyclic strain response of 9.9 m PBEDOT-Cz film scanned from -0.8 V to 0.4 V, 1.0 V, and 1.2 V (10 th scan) in 1.0 M LiClO 4 3.6 Overall Comparison of PPy, PEDOP, and PBEDOT-Cz If the strain data for PPy/TOS, PEDOP, and PBEDOT-Cz is normalized and plotted on the same graph it is easy to see that PPy/TOS produced significantly more strain than PEDOP and PBEDOT-Cz (Figure 3.18). It is believed that the higher strain response of PPy/TOS over PEDOP and PBEDOT-Cz is due to its ability to crosslink during electropolymerization. 64, 104 The possible ability of PBEDOT-Cz to cross-link at a potential of 1.15 V has been shown to be non-existent or at least so minimal that it is not significant in the strain production of this system. From this study it was shown that it is possible to obtain a detailed strain response measurement of various CPs directly and precisely utilizing standard strain gages technology to construct strain sensitive actuators. These gages determine strain by

PAGE 84

59 directly measuring the strain induced on the embedded constantan grid (inside the actuator) not the overall deflection of the actuator tip like most measurement techniques. This provides a detailed measurement of the actual strain produced internally in the Figure 3.1 system. 8 Comparison of in situ cyclic strain response of PPy, PEDOT, and PBEDOT-Cz in aqueous 1.0 M LiClO4. Adhesion Adhesion of the CP d to be a potential probl e rs (Figure exposed to a high vacuum (~ 1 10-5 Torr) during SEM examined (Figure 3.20). E (V) versus Ag/AgCl -1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.01.21.4 Micro Strain ( -50050100150200250 PPy ( 0.4V ) PEDOP ( 0.0 V ) PBEDOT-Cz ( 0.6 V ) PBEDOT-Cz ( 1.0 V ) PBEDOT-Cz (1.2 V ) 3.7 Effects of Interlayer to the EvAu coated substrate was foun em during long term repeated redox switching. In our study PPy, PEDOP, andPBEDOT-Cz were all found to have poor adhesion to a gold-button working electrodwhen electropolymerized at potentials 0.8 V. However large-scale (>70%) delamination of PPy was seen after long term cycling (>500 cycles) of actuato3.19) and one film completely delaminated from a Au coated polyimide sample when

PAGE 85

60 Examination of the image of the sample that delaminated in the SEM, it is obviothat the PPy surface in contact with the Au layer is porous and irregular in texture us (Figuhind was mination by SEM (Figure 3.25). The adhes many total delamination. As delamination of thend the res 3.21 and 3.22), which leads to poor interlayer adhesion between the PPy and the Au substrate. Also regions of small nodules or nucleation sites of PPy were left beon the Au substrate during the delamination process (Figures 3.23 and 3.24). The same features are also evident in SEMs of the underside of a PPy film that cracked and delaminated during repeated redox switching. The partially delaminated PPy film was rolled back to expose the surface thatin contact with the Au substrate for further exa ion of the PPy nodules to the Au substrate was non-homogenous upon delamination. From closer examination it can be seen that these nodules grow at the surface of the PPy film, and either lie directly on the surface or in some of thepores present at PPy-Au interface (Figure 3.26). Delamination of the CP film most likely initiates in isolated regions with high stress concentrations that eventually propagate into CP film from the substrate propagates a loss of electrical and physical contact between the CP and the working electrode develops. This decreases the ability to redox switch the CP film, thus retarding swelling/oxidation of the CP and therefore the development of strain in these systems. This combined with a decreasing amount of physical contact between the CP and the substrate reduces the overall movement atotal lifetime of actuators made from these systems. Once total delamination of the CPfrom the working electrode occurs no strain can be developed by the system and the actuator will totally stop functioning.

PAGE 86

61 3.8 Conclusions From this study it was shown tha t strain gage technology can readily be utilized as an inexpensive and highly accurateg the resultant physical properties of difz also produ sponse s. Durinn method of evaluatin ferent electrochemical polymerization and actuation conditions on a given conducting polymer structure. And due to the vast array of different strain gage designs it is possible to incorporate this technology into almost any actuator design. It has also been shown that PPy produced significantly higher strains than PEDOP and PBEDOT-Cz under the given electrochemical conditions. PBEDOT-C ced higher strain responses around and above the potential crosslinking potential of~1.15 V than at lower potentials. However it is believe that the increased strain reis due to higher applied driving potentials than the possible crosslinking reaction due to the lack of any electrochemical, spectroscopic, or strain/performance evidence. It has also been determined that the interfacial adhesion of the conducting polymerto the EvAu coated surface is a potential problem for long term actuator lifetime g repeated electrochemical cycling the induced cyclic strains initiate micro-crack formation between the CP and EvAu layers which then grow and eventually result idevice failure. All utilized conducting polymers (PPy, PEDOP, and PBEDOT-Cz) werefound to exhibit poor adhesion to EvAu when electropolymerized at potentials of 0.8 Vand greater. The most obvious case of this was for PPy which was shown to undergo large-scale delamination from the EvAu substrate when subjected to long term redox switching (Figure 3.19) and upon exposure to high vacuum during SEM examination (Figure 3.20). It is important to improve interlayer adhesion to increase CP actuator performance and overall lifetime.

PAGE 87

62 Figure 3.19 cycling Figure 3.20 SEM micrograph of delamination of PPy resulting from exposure to high vacuum; 25X SEM micrograph of delamination of PPy resulting from long-termof the actuator; 150X

PAGE 88

63 Figure 3.21 Enlarged SEM micrograph of region A in Figure 4.20; 1350X Figure 3.22 SEM micrograph of porous PPy at PPy-Au interface; 5000X

PAGE 89

64 Figure 3.23 Figure 3.24 SEM micrograph of PPy nodules remaining of Au substrate after PPy delamination; 1000X Enlarged SEM micrograph of region B in Figure 4.20; 100X

PAGE 90

65 Figure 3.25 SEM micrograph of PPy surface after delamination from Au sub strate during long-term repetitive cycling of the actuator; 250X Figure 3.26f exposed PPy-Au interface exhibiting PPy nodule growth; 1000X SEM micrograph o

PAGE 91

CHAPTER 4 STRAIN MEASURMENTS OF CPS ON ENHANCED AU SURF4.1 Introduction Adhesion of CPs to EvAu coated substrates was identified as a potenduring the previous study. Poor adhesion of the CP to the substrate will lead to low olopment and will greatly reduce the working lifetime for these actuaination of the CP from the substrate. PPy, PEDOP, and PBEDOT-Cz were found to exhibit poor adhesion to Au when electropolymerized on Au-button IN SITUACES tial problem r no strain devetors due to eventual delamelectrodes ndergo large-scale delamistudy was to imrs and the substrate. F4.2.1 Materials an Acetonitrile (ACN), lithium perchlorate (LiClO4), sodium perchlorate (NaClO4), p-toluenesulfonic acid Na salt (NaTOS), and tetrabutylammonium were all used as received from Sigma-Aldrich. Sodium nitrate (NaNO3) was purchased from Mallinckrodt at potentials of 0.8 V and greater. PPy was also shown to u nation when subjected to long term redox switching (Figure 3.19) and upon exposure to high vacuum during SEM examination (Figure 3.20). The purpose of this prove the interfacial adhesion between the conducting polymeurther characterization of the effects of various polymerization and actuation conditions on these systems was conducted on both standard and enhanced interfacial surfaces using the strain sensitive actuator technology. 4.2 Materials and Methods Pyrrole was (Sigma-Aldrich) was filtered through neutral alumina (Brockm activity 1; Fisher Scientific) until colorless before use to remove any impurities. 66

PAGE 92

67 Chemicals and sodium sulfite (Na2SO3) was purchased from Fisher Scientific, both were used as received. Deionized water (18 M, Millipore system) was used in all experiments and the solutions were deoxygeinutes by bubbling argon prior to 4.2.2 Electrochemical Gold Depos 9 f the g Conducting Cu tape (11ectrical connections u layer and the working electrode lead from the potentiostat. Electrs 332.9 V. e nated for 15 m electropolymerization and redox switching of the CP. ition Solution Electrochemical deposition of Au (EcAu) on thermally evaporated Au (EvAu) coated polyimide substrates utilized a solution of Oromerse SO Part B replenisher (Na 3 Au(SO 3 ) 2 ), a commercially available gold plating solution purchased from Technic Inc. EcAu deposition was carried out in a solution of 25% (10 mL) Na 3 Au(SO 3 ) 2 and75% (30 mL) of aqueous 1.7 M Na 2 SO 3 as reported for electroless Au deposition. 114.2.3 Evaporated and Electrochemically Deposited Gold Thermally evaporated Au (EvAu) was vacuum deposited on the bottom side ostrain gages to the desired thickness. All electrochemical work was carried out utilizinan EG&G Princeton Applied Research 273A potentiostat/galvanostat utilizing the CorrWare software package, a platinum foil counter electrode and an Ag/AgCl (BAS MF2052) reference electrode at room temperature. All potentials given are relative to this reference electrode. 81, 3M) was utilized to make el between the EvA ochemically deposited Au (EcAu) samples were then prepared on EvAu samplepotentiostatically from a diluted solution of NaAu(SO) (mentioned above) at 0EcAu layer thickness (and therefore surface roughness) was controlled by varying thcathodic charge during electrochemical deposition.

PAGE 93

68 The surface roughness factor (r) of the Au layer was determined by taking the ratio of the electrochemical area, determined from the charge required to reduce the oxide layer (potential cycling between 0.0 V and 1.5 V in aqueous 50 mM H surface om r = 2.89 for EvAu to r = phology was examined by utilizing a scannth PPy films of varying thickn ses in this study. The polymerization charges ranged from 0.35 tn and ickness of Surfaces (EcAu) Interlayer adhesion between PPy and the EvAu coated substrate has been enhanced by the electrochemical deposition of Au (EcAu) onto the EvAu surface. The EcAu 2 SO 4 ), 120 to the geometric area. The surface roughness factor varied fr 6.17 to r = 24.5 for the EcAu samples. Surface mor ing electron microscope (JEOL 6400) and white light optical profilometry (Wyko/Veeko NT1000). 4.2.4 Conducting Polymer Synthesis In this study EvAu was thermally vapor deposited to a thickness of ca. 1.0 m (measured by cross-section SEM), some EvAu samples when subsequently treated wi EcAu of varying thicknesses (~1-10 m) as described above. esses were prepared potentiostatically on EvAu and EcAu samples from a solution of 0.1 M pyrrole in 0.1 M aqueous sodium perchlorate at E = 0.9 V unless otherwise mentioned. The high surface roughness of EcAu samples made it difficult to determine PPy film thickness directly; therefore polymerization charge density (C/cm2) was utilized tocompare PPy film thicknes o 5.42 C/cm2. By assuming a 100% efficiency for the PPy deposition reactioa charge thickness ratio of 0.28 C/(m cm 2 ) 121 this correlates to a PPy film th~1.25-19.40 m on a smooth surfaces. 4.3 Improved Interlayer Adhesion Utilizing Electrochemically Deposited Au

PAGE 94

69 treatm then l charge passed during the electrochemical mples. The surface roughness factor (r), of the Au su area s converted to the surface area using a factor of 0.43 mC/cm2. ent increases the surface area of the working electrode via the growth of Au crystals on the EvAu surface. EcAu films were deposited on EvAu coated PI substrates(EcAu/EvAu/PI samples) potentiostatically at .9 V from a solution consisting of 25% Na 3 Au(SO 3 ) 2 (Oromerse SO Part B replenisher) and 75% aqueous 1.7 M Na 2 SO 3 The EcAu layer thickness (Au crystal height), and therefore surface roughness (r), was initially controlled by limiting the deposition time on smooth PI samples andwas later controlled by monitoring the tota deposition process for all strain gage sa rfaces, was determined by taking the ratio of the surface area measured electrochemically, obtained from the charge required to reduce the surface oxide layer, 120to the geometric surface area of the working electrode. The electrochemical surfacewas measured by potential cycling the Au surface between 0.0 V and 1.5 V in an aqueous 50 mM HSO solution. The charge passed during re-oxidation of the Au oxide monolayer wa 2 4

PAGE 95

70 Figure 4.1 Correlation between surface roughness factor, nominal EcAu thickness and -section SEM. es, and substrate structure (surface texture) have all been shown to affect the resulting surface morphology of electrochemically deposited Au surfaces.122-126 The effects of substrate structure on EcAu morphology can easily be seen from SEM micrographs of the EcAu deposited on smooth EvAu/PI substrates versus the rough EvAu/PI surface of the strain gages. EcAu growth on smooth EvAu/PI surfaces results in the formation of large 5-point start structures. The Au crystals grow about normal to the EvAu/PI surface on the smooth samples. Crystal growth starts at many small nucleation sites. As the crystals continue to grow they coalesce to form fewer but larger crystal structures. EcAu deposition charge. EcAu thickness was determined by crossFigure 4.1 illustrates the relationship between the surface roughness factor (r) and EcAu deposition charge. The EcAu layer thicknesses were measured by cross-section SEM. 4.4 Effects of Surface Roughness on EcAu Morphology Many factors such as electrolyte type, pH, deposition potentials, additiv

PAGE 96

71 However the application side (backside) of the PI strain gages is roughened (sanded) to help promote adhesion of the strain gage to the sample it is being applied to. This roughened surface (surface roughness factor (r) = 2.89) produces very irregularly shaped Au crystal structures when grown to high charge densities (3.67 C/cm2and 4.72 C/cm2, r = 24.50). At lower charge densities (0.24 C/cm2, r = 6.17 and 0.71 2, r = 10.04) the Au crystal structures are very regular and resemble these obtained ooth EvAu/PI samples. However, the irregular surface of the strain gages causes the Au crystals to grow r = 18.90 C/cmon the smoff the norm spiky tree-like structures when grown to high deposition charges. SEM micrographs (4000X) of the different Au substrate surface roughness generated by varying the EcAu deposition time between 0 and 60 minutes on smooth EvAu/PI samples are shown in Figures 4.2 4.6. The corresponding SEM micrographs of Au surfaces deposited on roughened E127, 128 al axis causing the crystals to collide at random angles and coalesce to form vAu/PI strain gages are shown in Figures 4.7 4.11. The topographies are very different from those obtained by chemical deposition of Au from a similar solutions on glass and polycarbonate substrates.

PAGE 97

72 Figure 4.2 SEM micrograph of EvAu deposited on smooth polyimide (PI); 4000X Figure 4.3 SEM micrograph of 3 minute EcAu deposition on smooth EvAu/PI; 4000X

PAGE 98

73 Figure 4.4 Figure 4.5 SEM micrograph of 30 minute EcAu deposition on smooth EvAu/PI; 4000X SEM micrograph of 10 minute EcAu deposition on smooth EvAu/PI; 4000X

PAGE 99

74 Figure 4.6 SEM micrograph of 60 minute EcAu deposition on smooth EvAu/PI; 4000X Figure 4.7 SEM micrograph of EvAu (r = 2.89) deposited on rough PI strain gage; 4000X

PAGE 100

75 Figure 4.8 SEM micrograph of EcAu (r = 6.17, 2.5 min.) deposited on rough EvAu/PI strain gage; 4000X Figure 4.9 crograph of EcAu (r = 10.04, 10 min.) deposited on rough EvAu/PI strain gage; 4000X SEM mi

PAGE 101

76 Figure 4.10 SEM micrograph of EcAu (r = 18.90, 30 min.) deposited on rough EvAu/PI strain gage ; 4000X ; 4000X Figure 4.11 SEM micrograph of EcAu (r = 24.50, 60 min.) deposited on rough EvAu/PI strain gage

PAGE 102

77 4.5 Electrochemical Deposition of PPy s were electropolymerized potentiostatically on EvAu/PI and EcAu/EvAu/PI m a solution of 0.1 M pyrrole in aqueous 0.1 M NaClO4 at 0.9 V, unless entioned. Figures 4.12 (EvAu), 4.13 (10 min. EcAu), 4.14 (30 min. EcAu), and 4.15 (60 mEM micrographs (4000X) which show the changes in PPy (18.9 C/cmorphology with increased EcAu deposition time (increased surface roughnfactor) on smooth PI samples. The typical nodular growth (for thick PPy filmEvAu/PI) of PPy becomes relatively flat and conforms nicely to the crystal structures as PPy filmsurfaces frootherwise min. EcAu) are S2) surface mess s on ). This reduces the overall PPy film thickness and increases the overall surface area for a given electropolym and results in enactuators as m the surface roughness increases for EcAu/EvAu/PI samples (Figures 4.12 and 4.15 erization charge. By reducing the cross-sectional area of the PPy filmincreasing the surface area the diffusion properties of the film can be improved, i.e., ions and corresponding solvent molecules can diffuse into and out of the film faster. This hanced strain properties (developed strain and strain rate) of the PPy easured during redox switching of the strain sensitive actuators.

PAGE 103

78 Figure 4.12 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI; 4000X Figure 4.13 SEM micrograph of PPy (18.9 C/cm2) deposited on smooth EvAu/PI treated with EcAu for 10 min.; 4000X

PAGE 104

79 Figure 4.14 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI treated with EcA u for 30 min.; 4000X Direct determination of the PPy film thickness was difficult due to the high surface roughness of some EcAu/EvAu/PI samples (Figures 4.16 and 4.17, EcAu/EvAu/PI and PPy/EcAu/EvAu/PI samples respectively, 60 min EcAu treatment time, 4000X). Figure 4.15 SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth EvAu/PI treated with EcAu for 60 min.; 4000X

PAGE 105

80 Therefore thcomponomer polymsame ame techniqu. Assuming ration of 0.28 C/mcm2mooth p as a function of electropolymerization charge and surface roughness. Assuming the same numbsurface rougme oferefore as the working eleintain the same e total charge (C/cm2) passed during electropolymerization was utilized to are the effects of the amount of PPy on the strain response. The charge passed during electropolymerization can be directly related to the number of moles of merized, therefore by monitoring the charge passed we can accurately deposit the ount of PPy on each sample independent of surface roughness. This is the same used to control the EcAu surface roughness. The PPy polymerization charge ranged from 0.35 C/cm2 to 5.42 C/cm2100% efficiency for the anodic deposition reaction and a charge thickness,121 this correlates to PPy film thicknesses of 1.25 m to 19.4 m on s EvAu/PI samles. Using this information it is possible to calculate the PPy thickness er of moles of pyrrole monomer are polymerized on all surfaces independent of hness for a given polymerization charge, we can assume that the total volu PPy is also the same for all film for a given polymerization charge. Thctrode surface area increases the PPy thickness has to decrease to ma total volume of PPy. The equations used for this calculation are given below: 12211-2-121AAEPCGiven:T=Tr=T=AA0.28Ccmm (4.1) 212 -2-1rr0.28Ccmme surface roughness factor, EPC is the electropo 11EPCTherefore:T=TT= (4.2) where r is thlymerization charge (C/cm2), and A1,2 and T1,2 are the surface areas and PPy film thicknesses on surfaces 1 and 2

PAGE 106

81 respectfully. A surface plot of the calculated PPy film thickness as a function of electropolymerization charge and surface roughness is shown in Figure 4.18. Figure 4.16 Cross-section SEM micrograph of EcAu (60 min.) deposited on smooth EvAu/PI; 4000X EvAu/PI treated with EcAu for 60 min.; 4000X Figure 4.17 Cross-section SEM micrograph of PPy (18.9 C/cm 2 ) deposited on smooth

PAGE 107

82 Figure 4.18PPy Adhesion to EcAu Treated Surfaces s were washed conducted wsurfaces (11 samsampbonding pro Electropolymerization Charge (C/cm2) 123456 Surface Roughness (r) 510152025 1.0 m 2.0 m 3.0 m 4.0 m 5.0 m 6.0 m 7.0 m 8.0 m 9.0 m 10.0 m 11.0 m 12.0 m 13.0 m 14.0 m 15.0 m 16.0 m 17.0 m 18.0 m 19.0 m 20.0 m 21.0 m 22.0 m 1.0 m2.0 m Surface plot of calculated PPy film thickness as a function of surface roughness factor (r) and electropolymerization charge (C/cm 2 ) 4.6 Improved Interlayer adhesion of PPy to EvAu/PI and EcAu/EvAu/PI was tested by a tape (810, 3M) peel-off test. PPy (0.30 C/cm2) was electrochemically deposited on EvAu (r = 2.89) and EcAu/EvAu (r 10) coated strain gage samples. The PPy filmin double distilled water and dried in air for at least 24 hrs before testing. Tests were ith a 90 pull-of angle. PPy films were readily removed from all EvAu/PI surfaces while no delamination was noticed for films deposited on the EcAu/EvAu/PI ples). This is attributed to the larger surface area of the EcAu/EvAu/PI les providing more area for PPy bonding to Au and also physical/mechanical vided by the nooks and crannies present on the EcAu surface.

PAGE 108

83 4.7 IN SITU PPy/EvAu/PI Actuator Results Initial studies were conducted utilized a 2.28 C/cm2 PPy film (~ 10 m) to investigate the effects of g redox switching, Pdependent mduring redox switching.these filmNaClO4, duced at 0.6 V for 60ilms were cycle) counter ion type (anion), potential limiting durinPy film thickness, and electropolymerization potential on the overall strain response of PPy/EvAu/PI-strain gage actuators. 4.7.1 Counter Ion Effects on PPy Strain Response It has been well documented that the actuation response of CP actuators is any factors including the specific counter ion and solvent species utilized 9, 12, 36, 129, 130 It is also believed that the swelling behavior of s is anion dominant. The strain response of ClO4 doped (polymerized in NaClO4) PPy/EvAu/PI actuators was evaluated during redox switching by potential cycling in 0.1 M aqueous +Na + NO 3 and Na + Cl electrolyte solutions. The PPy films were fully re s and then cycled between -0.6 V to 0.4 V at 5 mV/s. The f d five times in each electrolyte to obtain a reproducible strain response. After cycling in the ClO 4 solution the films were thoroughly washed with double distilled water and then cycled in NO 3 and Cl solutions respectively. These experiments were conducted on three ClO 4 doped samples and one NO 3 doped (polymerized in NaNO 3sample to verify the results. There was no significant difference in the ClO 4 doped andNO 3 doped samples. PPy films were also cycled in aqueous Li ClO + 4 and Cs ClO + 4 solutions to verify the performance of the ClO 4 anion. Figure 4.19 shows the resultant normalized strain responses for these materials.

PAGE 109

84 444339 rse t own by De Rossi et al..44 PPy oxidizes t about 0.2 V therefore the film is reduced during the initial part of the scan. PPy films cycled in NaNO3 again showed an initial decrease in strain followed by a large increase Figure 4.19 Strain response of a 2.83 C/cm 2 PPy/EvAu/PI actuator during potential cycling at 5 mV/s in aqueous NaClO, LiClO, CsClO, NaNO, and NaCl solutions. PPy films cycled in aqueous ClO 4 solutions exhibited values of 147 134 and 136 for NaClO 4 LiClO 4 and CsClO 4 respectfully with an average of 1 between the three samples. The maximum strain was obtained at 0.3 V on the revescan for all three tests. This is due to the slow diffusion process of the ion migration intoand out off the PPy film. Both the LiClO 4 and CsClO 4 samples exhibited an initial slighdecrease in strain between .4 V and .1 V on the forward scan which was then followed by a large increase in strain, however the NaClO 4 samples did not exhibit thisinitial decrease in strain. This is indicative of cation influx during the initial reduction process. 11-13, 19, 20, 22 This process has also been sh a E(V) versus Ag/AgCl -0.6-0.4 -0.20.00.20.40.6 Mic -4040120 rra) o St in ( -20020 60 80100140160 NaClO4 CsClO4 LiClO4 NaNO3 NaCl

PAGE 110

85 in strathe over, with a re This is a d and 1.8 timNaCl. Simdoped PPy.e film was linearly proportional to the solvated anion volume. The solvated ionic larger solvated ionic volume than Cl, which resulted in a 1.8 times increase in strain betwefy that ial d between in with the maximum strain value obtained at 0.3 V on the reverse scan. However all change in strain was smaller for NaNO3 samples compared to NaClO4 of 114 This is a difference of 33 between the two. Similar results weobtained for samples cycled in NaCl. But NaCl produced an even lower of 81 ifference of 66 between NaCl and NaClO4 and a difference of 33 between NaCl and NaNO3. In review NaClO4 exhibited a strain response 1.3 times higher than NaNO3es higher than NaCl. While NaNO3 exhibited a 1.4 times higher strain response that ilar results have been observed by Kaneto et al. for poly(styrene sulfonate) 131 They found that swelling during oxidation due to anion incorporation into th volumes of Cl and ClO 4 are 25 3 and 56 3 respectively. ClO 4 has about a 2.2 times en the two counter ions for these systems. This suggests that anion exchange during redox switching is the dominant factor controlling swelling and therefore strain production. 4.7.2 Effects of Potential Limiting on PPy Strain Response The effects of potential limiting on the PPy strain response during redox switching was also investigated utilizing strain sensitive actuators. This was also done to verithe consumed charge followed the strain response. A square-wave potential was used tostep the films between their oxidized and reduced (-0.6 V) forms. The oxidation potentwas varie 0.1 V and 0.6 V at 0.1 V increments. The films were held at each potential for 100 s starting in the reduced state. Figure 4.20 shows the square-wave strain response as a function of oxidation potential. The strain responses were normalized for

PAGE 111

86 comparative purposes. From the inset of Figure 4.20 it can be seen that the normalized strain response and consumed charge have an almost linear dependency on the oxidation potential and follow each other nicely. This also supports that the number of anions inthe film, determined from the consumed charge and controlled by the doping level/oxidation potential, is a dominant factor in the production of strain in these systems. that the reduction/contraction process was faster than the oxidaFigur 69 It can also be seen tion process, especially at the lower oxidation potentials. e 4.20 Strain response of a 2.83 C/cm 2 PPy/EvAu/PI actuator during potential stepping (100 s/step) in NaClO 4 between .6 V and (a) 0.1 V, (b) 0.2 V, (c) 0.3 V, (d) 0.4 V, (e) 0.5 V, and (f) 0.6 V 4.7.3 Effects of PPy Film Thickness on Strain Response The effects of PPy film thickness on the overall strain performance of PPy/EvAu/PI actuators were evaluated utilizing the strain sensitive actuator technology. PPy films were electropolymerized on a EvAu/PI strain gage to six different charge densities: 0.79 Time (sec) 050100150 200250 Mic Strin ( roa) -160-120-400 -200 -80 E(V) versus Ag/AgCl 0.00.40.8 Nra et Stin () 0180240 306090120150210 Charge Density (Cm-2) *c 0.150.210.36 0.180.240.270.300.33 abd cef

PAGE 112

87 C/cm2, 1.6 C/cm2, 2.4 C/cm2, 3.2 C/cm2, 4.0 C/cm2, and 4.8 C/cm2. The film were fully oxidized at 0.4 V and then potentiostatically stepped to .6 V and held for 100 s to allow the strain response to equilibrate and then stepped back to 0.4 V and held for an additional 100 s. This was done five times to obtain reproducible results. As from Figure 4.21 as the electropolymerization charge density was increased the ovstrain response of the actuator also increased. However the given amount of change in strain for a given increase in film thickness decreased with incr can be seen erall easing charge densities. The samelectropolymhere a 4.8 C/cmse in strain fllows: 1.6 C/cm, 3.2 C/cm. 2.8 tim 0.79 C/cm thickness .8 C/cm2 PPyrease in strain for a 4.1, 5.1, and 6.1 ared to 0.79 C/cm2 PPy films. e trend can be seen in the charge data (Figure 4.22). PPy films erized to 0.79 C/cm2 (2.8 m, calculated) produced a of 59 W2 PPy films (17.1 m, cal.) produced a of 236 this is a 4 times increaor a 6.1 times increase in PPy film thickness. The rest of the data is as fo2 (5.7 m, cal.) PPy = 113 2.4 C/cm2 (8.6 m, cal.) PPy = 162 2 (11.4 m, cal.) PPy = 186 and 4.0 C/cm2 (14.3 m, cal.) PPy = 212 An interesting trend is that the 1.6 C/cm2 and 2.4 C/cm2 PPy films produced a 1.9 and es increase in strain for a calculated 2.0 and 3.1 times increase in the PPy filmthickness respectively (same for electropolymerization charge) when compared to the 2 PPy films. There is almost a direct relationship between PPy film and the produced strain response of these films. However, 3.2 C/cm2, 4.0 C/cm2, and 4 films produced a 3.2, 3.6, and 4 times inc times increase in PPy film thickness respectfully when comp A simpler way of looking at this is to normalize the overall change in strain () to the PPy film thickness by dividing by the calculated PPy thickness (T PPy ). The resulting /T PPy ratio gives the strain () produced per unit thickness (m) of PPy. The

PAGE 113

88 resulting /T PPy (/m) ratios for the above PPy/EvAu/PI strain gage samples are 21.07, 19.82, 18.84, 16.32, and 13.80 for 2.8 m, 5.7 m, 8.6 m, 11.4 m, and 17.1 m PPy films respectfully. As can be seen, the amount of strain produced for a given thickness of PPy decreases as PPy film thickness increases. Figure 4.21 In situ strain response of PPy/EvAu/PI actuators of varying PPy film NaClO4. Electropolymerization charge densities were 0.79 C/cm2 (2.8 m(11.4 m, cal.), 4.0 C/cm2 (14.3 m, cal.), and 4.8 C/cm2 (17.1 m, cal.). thickness during potential stepping between .6 V and 0.4 V in aqueous calculated), 1.6 C/cm2 (5.7 m, cal.), 2.4 C/cm2 (8.6 m, cal.), 3.2 C/cm2 Time (s) Micro Srain ( -250-50 -200-150-100050 0.79 C/cm2 t) 1.6 C/cm2 2.4 C/cm2 3.2 C2 /cm 4.0 C/cm 2 4.8 C/cm2 050100150200

PAGE 114

89 Figure 4.22 m m, 2 more strainnature of thproperties. As the film thickness increases the diffusion of ions into the film is slowed and therefore the deeper regions of the film (closer to the electrode) do not swell as much as the surface regions. Therefore a strain gradient or swelling gradient is developed in the film and the given increase in strain for a given increase in film thickness decreases as film thickness increases. Also as the film thickness increases the effects of Ohmic drop 48 (film resistance) becomes more significant. As the film thickness increases the resistance Time (s) 050100150200 Charge (C) -1.2-1.0-0.8-0.6-0.4 -0.20.00.2 0.79 C/cm2 1.6 C/cm2 2.4 C/cm 3.2 C/cm2 4.0 C/cm2 4.8 C/cm2 In situ charge response of PPy/EvAu/PI actuators of varying PPy filthickness during potential stepping between .6 V and 0.4 V in aqueous NaClO4. Electropolymerization charge densities were 0.79 C/cm2 (2.8 calculated), 1.6 C/cm2 (5.7 m, cal.), 2.4 C/cm2 (8.6 m, cal.), 3.2 C/cm(11.4 m, cal.), 4.0 C/cm2 (14.3 m, cal.), and 4.8 C/cm2 (17.1 m, cal.). PPy film thickness is an important factor in the overall strain performance of PPy based actuators. The thicker the film the more material there is to swell and therefore is produced. However as the film thickness increases the diffusion-limited ese materials becomes a greater influence on the overall mechanical

PAGE 115

90 of the filmalso invine the optims. response on for PPy(dogreated -wave 0.7 V. The switching charge density, during PPy oxidation, (inset Figure 4.23) for all causes a potential gradient through the cross section of the film. In result different regions of the film are exposed to different oxidation potentials and as shown previously lower oxidation potentials produce lower strains. 4.7.4 Effects of Polymerization Potential on PPy Strain Response The effects of electropolymerization potential on the strain response of PPy were estigated utilizing strain sensitive actuators. This was also done to determal electropolymerization potential to produce the maximum strain for these systemIt is believed that by controlling the electropolymerization potential the cross-link density of the resultant PPy film can also be controlled. Higher cross-linking densities should produce systems yielding higher degree of strain, however if the cross-link density becomes to high the film will become too rigid and this will decrease the overall strainf the actuator. Recently, Maw et al. 132 demonstrated that bending deformatiodecylbenzenesulfonate)/Au/PI laminates prepared galvanostatically was r at low deposition current densities (0.4 mA/cm 2 ) when compared to higher deposition current densities (40 mA/cm 2 ). PPy films (2.83 C/cm 2 ) were prepared potentiostatically at 0.7 V, 0.8 V, 0.9 V, an1.0 V and then subsequently were switched between 0.6 V and 0.4 V by squarepotentials, with 100 s steps. The resultant strain response (shown in Figure 4.23) was found to be inversely proportional to the polymerization potential at potentials greater than 0.7 V. Films prepared at potentials of 0.9 V and 1.0 V had an overall strain response on the order of ~20 % less than films prepared at 0.7 V. This trend was also later confirmed on EcAu samples; however it was also shown on EcAu samples that the strain response increased linearly with the electropolymerization potential at potentials

PAGE 116

91 films were found to be similar indicating that the decrease in the strain response is notdue to a decrease in the electroactivity of the PPy films. It can be inferred that at electropolymerization potentials above 0.7 V the increasing cros s-link density of the PPy .83 C/cm2 PPy/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO. PPy was the Au film start to hamper strain development in these systems. Figure 4.23 In situ strain response of a 2 Time (sec) 050100150200250 Mico Strin ( ra) -200-80 -160-120-400 Polymerization Potential (V) 0.70.80.91.0 Nerain () 4 prepared potentiostatically at (a) 0.7 V, (b) 0.8 V, (c) 0.9 V, and (d) 1.0 V. 4.8 IN SITU PPy/EcAu/EvAu/PI Actuator Results As stated earlier, by electrochemically depositing Au (EcAu) on EvAu coated substrates adhesion of CPs can be dramatically improved due to the increased surface roughness of the EcAu layer. By increasing the adhesion of the CP to the substrateoverall strain response of the actuator should be increased. Also be utilizing the EcAu surface modification the overall CP film thickness is reduced improving diffusion and reducing the effects of Ohmic drop in the film. Studies were conducted on EcAu treated EvAu/PI strain gages to investigate these effects. The effects of surface roughness (Ec t St 140 150160170180 Cha De (C) 0.180.200.22 0.240.260.28 acd b rgensity*cm-2

PAGE 117

92 film thickness), PPy electropolymerization charge (PPy film thickness), counter-ion typfrequency response, and overall actuator lifetime were evaluated on EcAu/EvAu/Pgages actuators. These results were also compared to the results obtained on EvAu/Pi strain gages. e, I stain thickness ancontroterials. electropolyamelectrode suount of ss. The efor EvAu and r = 6.17, 7.13, 10.04, 18.90, and 24.50 for EcAu. Three 1.18 C/cm2 PPy films o 0.4 d 4.8.1 Effects of PPy electropolymerization charge and surface roughness factor The two main factors of interest in these experiments are the effects of PPy filmd EcAu film thickness or surface roughness factor. Both of these factors are lled by the electrochemical charge passed during the deposition of these maHowever the PPy film thickness is also dependent on the surface roughness factor. As stated earlier, as the surface roughness factor increases, for a given PPy merization charge, the resulting PPy film thickness decreases. The total ount of PPy present on the working electrode remains the same; however the working rface area is increased. Resulting in the distribution of the same amPPy over a larger surface area and thereby decreasing the overall PPy film thickne 4.8.1.1 Effects of surface roughness factor ffects of Au surface roughness were tested for surface roughnesses of r = 2.89 were tested on each surface roughness by potentially stepping between -0.6 V tV in 0.1 M aqueous NaClO4. It was shown that for a given PPy film thickness (specificpolymerization charge) the strain response increases up to an optimal surface roughnessfactor and then decreased when this optimal point was passed. The square-wave strain response is shown in Figure 4.24. The resulting for each surface roughness is plottein Figure 4.25, from this it can be seen that the overall change in strain increases from 80 for r = 2.89 (EvAu, PPy = 1.46 m cal.) up to 166 for r = 10.04 (0.42 m cal.) and

PAGE 118

93 then falls back off to 150 for r = 18.90 (0.22 m cal.) and 107 for r = 24.50 (0.17 m cal.). Figure 4.24 In situ strain response of a 1.18 C/cm PPy/EcAu/EvAu/PI actuator during surface roughnesses factors of (a) 2.89, (b) 6.17, (c) 7.13, (d) 10.04, (e) 18.90, rate (/sec) response (Figure 4.26). The strain8 2potential stepping between .6 V and 0.4 V in aqueous NaClO 4 With and (f) 24.50. Time (sec) 050100150200250 Micro Stra () in -200-120-80-40 40 r = 2.89 -1600 abdef c r = 6.17 r = 7.13 r = 10.04 r = 18.90 r = 24.50 The same trend can be seen in the strain rate increases from 5.69 /sec during expansion (1.75 /sec during contraction) for r = 2.89 up to 30.68 /sec (8.68 /sec) for r = 10.04 and then falls back off to 18.3/sec (7.00 /sec) for r = 18.90 and 16.19 /sec (5.25 /sec) for r = 24.50. This is due to the overall decrease in PPy film thickness. For a given PPy polymerization charge, as the surface roughness factor increases and the PPy film thickness decreases. There is an optimal point whereby the PPy film thickness becomes insufficient to deflect/strain the strain gage based actuator. This point is dependent on the PPy polymerization charge and will be shown later. However as the film thickness decreases

PAGE 119

94 the diffusion rate of ions in the film increases producing a higher strain rate. This is trueuntil the ov erall strain response of the actuator is hindered by the EcAu film thickness. Surface Roughness Factor (r) 051015202530 Net Strain () 6080100120140160180200 Figure 4.25 Effects of surface roughness factor (r) on the overall change in strain () of a 1.18 C/cm 2 PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO 4

PAGE 120

95 Figure 4.26The ber of surface roughnesses. PPy samples were electropolymerized at 0.79 C/cm2, 1.6 C/cm2, 2.4 C/cm2, 3.2 C/cm2, 4.0 C/cm2, and 4.8 C/cm2 and then potentially stepped between 0.6 V and 0.4 V (100 s steps) in 0.1 M NaClO4. An example is shown in Figure 4.27 for a surface roughness factor of 18.9. Unlike the surface roughness effects, the strain response was shown to increase with increasing PPy electropolymerization charge for all samples. The resulting for PPy of 0.79 C/cm2 (0.15 m cal.), 1.6 C/cm2 (0.30 m cal.), 2.4 C/cm2 (0.45 m cal.), 3.2 C/cm2 (0.60 m cal.), 4.0 C/cm2 (0.76 m cal.), and 4.8 C/cm2 (0.91 m cal.) is 162 270 339 389 450 and 498 respectively on a surface roughness of 18.90 (~30 min EcAu treatment). The same effects of PPy film thickness can be seen here as was seen for the PPy/EvAu/PI samples. As the PPy thickness increases the increase in strain for a given increase in PPy film Surface Roughness Factor (r) 0510152025 Strain Rate (/sec) 0510152025 3035 Expansion Rate Contraction Rate Effects of surface roughness factor (r) on the strain rate (/sec) of a 1.18 C/cm2 PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO4. 4.8.1.2 Effects of PPy electropolymerization charge on strain (PPy film thickness) effects of PPy electropolymerization charge were also evaluated on a num

PAGE 121

96 thickness decreases. The resulting strain to thickness ratios (/TPPy) are 1080, 900, 753, 648, 592, and 547 for the PPy samples mentioned above respectfully. The effects of PPy electropolymerization charge for different surface roughness values of 2.89, 6.17, 10.00, 18.90, and 24.50 are shown in Figure 4.28. Figure 4.27 In situ strain response of PPy/EcAu/EvAu/PI actuator during potential stepping between .6 V and 0.4 V in aqueous NaClO 4 With surface roughnesses factors of 18.90 and PPy electropolymerization charge of (a) 0.79C/cm, (b) 1.6 C/cm, (c) 2.4 C/cm, (d) 3.2 C/cm, (e) 4.0 C/cm, and (fC/cm 22222) 4.8 2 Time (s) 050100150200 Micro Srain () -500-400-200-1000100 0.79 C/cm2 1.6 C/cm2 t -600 -300 2.4 C/cm2 3.2 C/cm 2 4.0 C/cm2 4.8 C/cm2 ad bc e f

PAGE 122

97 Figure 4.28 with factor The overall combined effects of PPy electropolymerization charge and surface roughness hown in Figure 4.electropolymerization charge and surface roughness decrease as the value of these factors increase. If you look at the effects of electropolymerization charge and surface roughness separately it is evident that the slope of the strain versus factor plot is greatest initially and then decreases with increased factor level. It is evident that the increased strain response from both of these factors will eventually reach a plateau at an optimal value. Electropolymerization Charge (C*cm-2) 012345 Micro Strain () 0100200300400500 600 r = 2.89 r = 6.17 r = 10.00 r = 18.90 r = 24.50 Maximum change in strain response of PPy/EcAu/EvAu/PI actuators during potential stepping between .6 V and 0.4 V in aqueous NaClO4 increasing electropolymerization charge. Measured on surface roughnesses of (a) r = 2.89 (EvAu), and EcAu samples of r = (b) 6.17, (c) 10.00, (d) 18.90, and (e) 24.50. 4.8.1.3 Combined effects of PPy electropolymerization charge and surface roughness factor (r) on strain response are tied together nicely in the surface plot s29. From the surface plot it is evident that the effects of

PAGE 123

98 Electropolymerization Charge (C/cm2) 12345 Surface Roughness (r) 5101520 100 200 300 400 500 Figure 4.29 Surface plot of the normalized overall strain response of PPy as a funcof electropolymerization charge and surface roughness factor (r). ter Ion Effects on PPy Strain Response tion 4.8.2 CounCounter ion effects were also evaluated on EcAu/EvAu/PI strain gage actuators and lts were found to similar to those found on EvAu/PI strain gage actuators. HoweAu NaCl. Samples 4443 the resu ver the overall was much higher for the EcAu samples when compared to Evsamples. PPy/EcAu/EvAu/PI strain gages (PPy = 8 C/cm 2 r = 18.9) were cycled between .6 V and 0.4 V at 5 mM/s in aqueous LiClO 4 NaNO 3 andcycled in LiClO 4 produced a of 457. While samples cycled in NaNO 3 and NaCl produced of 278 and 219 respectfully. The results for the EvAu samples were 147 133 136 114 and 81 for NaClO, LiClO, CsClO, NaNO, and NaCl respectfully. It should be noted that the PPy electropolymerization charge for the EvAu samples was 2.28 C/cm2 (~10 m) while the electropolymerization charge for

PAGE 124

99 the EcAu samples was 8 C/cm2 (1.5 m). However, the LiClO4 produced a 2.1 timgrater strain response than NaCl and as stated before the solvated ion size for ClO-. This is a much closer correlation between the strain response and solvated ionic size than in EvAu samples. The cyclic voltammetry strain response is es 4~2.2 times larger than Clshown in Figure 4.30. aCl EvAu systems at all frequencies (Figure 4.31). The charge response exhibited a similar Figure 4.30 In situ Strain response of a 8.0 C/cm 2 PPy/EcAu/EvAu/PI actuator during potential cycling at 5 mV/s in aqueous (a) LiClO 4 (b) NaNO 3 and (c) Nsolutions. 4.8.3 Frequency response The frequency response of these systems was also evaluated using the strain sensitive actuators. PPy (1.18 C/cm2) on EvAu (r = 2.89) and EcAu (r = 10.00) was stepped between -0.6 V and 0.4 V in aqueous 0.1 M NaClO 4 at increasing frequencies from 0.01 to 0.5 Hz. Strain decreased as a function of frequency for both systems; however the strain response for the EcAu systems remained consistently higher than for E (V) versus Ag/AgCl -0.6-0.4-0.20.00.20.40.6 Micr o Strain () 100 0 -100 200300400500 LiClO4 NaNO3 NaCl

PAGE 125

100 response to frequency as the strain response. The charge efficiency (/C) was also evaluated and is shown in the inset of Figure 4.31. The charge efficiency of PPy on EcAu was significantly higher than on EvAu surfaces. The greater efficiency of the EcAsystems can be a result of improved adhesion of the PPy to the electrode and smaller Ohmic drop in the syste u m due to the thinner PPy layer. The effects of EcAu treatments on actuator lifetime were also initially investigated back to -0.6 V (10 sec). EvAu samples only exhibited 11.5% and 35% of their initial Frequency (Hz) 0.00.10.20.30.40 Normalized Change 0.00.20.40.60.81.0 Frequency (Hz) 0.00.20.40.6 Charge Efficiency (C) 0100200300400 b (strain)b (charge)a (strain)a (charge) ba .5 Figure 4.31 In situ normalized (relative to 0.01 Hz) strain and charge response of 1.18 C/cm 2 PPy as a function of frequency on (a) r = 2.89 (EvAu) and (b) r = 10.0 (EcAu) surfaces. Inset shows the charge efficiency (/C) as a function of frequency. 4.8.4 Lifetime by potentially stepping of 1.13 C/cm2 PPy films on EvAu (r = 5.1) and EcAu (r = 22.1) surfaces. One cycle consisted of stepping from -0.6 V (10 sec) to 0.4 V (20 sec) and then strain and charge responses respectively after 900 cycles. Consequently EcAu samples

PAGE 126

101 only exhibited 25% and 46% (after 900 cycles) and 12% and 41% (after 1200 cycles) of their initials strain and charge responses respectively. Both of these systems exhibitepoor overall lifetimes for both their strain and charge responses, Figure 4.32. It was later determined that the electropolymerization solutions were not purged with argon adequately prior to electropolymerization. The presence of oxygen can interfere with the electropolymerization process yielding d poor PPy films. A second set of experims systemEvAu systemsmed imse ents was conducted with 1.18 C/cm2 PPy films on EvAu (r = 3.43) and EcAu (r = 8.26 and 18.90) surfaces for 2000 cycles. The electropolymerization solutions were purged with argon for 15 min prior to electropolymerization. Both systems (EvAu and EcAu) exhibited a decrease in the normalized strain response; however the EcAu systemstabilized after 1000 cycles while the EvAu systems continued to decrease. The EcAu s retained about 60% of their initial strain response after 2000 cycles while the s retained only 40%. The change in the normalized consumed charge was aller than the change in strain. All three samples exhibited a normalized consumcharge of ~75%. This improved strain response after 2000 cycles is attributed to proved interlayer adhesion between PPy and EcAu. Figure 4.33 shows the normalized strain respon of the EvAu and EcAu systems tested.

PAGE 127

102 Figure 4.32 In situ normalized strain and charge response of 1.18 C/cm 2 PPy (inadequate Argon purge) on r = 5.1 (EvAu) and r = 22.1 (EcAu) treated actuators with repeated potential stepping between .6 V (10 s) and 0.4 V (20 s). Figure 4.33 In situ normalized strain and charge response of 1.18 C/cm2 PPy on r = 3.43 (EvAu), r = 8.26 (EcAu), and r = 18.90 (EcAu) treated actuators with repeated potential stepping between .6 V (10 s) and 0.4 V (20 s). Number of Redox Switches 02004006008001000120014001600180020002200 Normrainsp alized St Reonce(%) 02040100 6080 Strain: EvAu (r = 3.43) Charge: EvAu (r = 3.43) Strain: EcAu (r = 8.26) Charge: EcAu (r = 8.26) Strain: EcAu (r = 18.0) Charge: EcAu (r = 18.0) Number of Redox Switches 0200400600800100012001400 Normad 060 Strain: EvAu (r = 5.1) 80100 Charge: EvAu (r = 5.1) Strain: EcAu (r = 22.1) Charge: EcAu (r = 22.1) lize Responce (%) 2040

PAGE 128

103 4.9 Conclusions The strain of PPy/Au/PI bending actuators during redox switching of PPy was directly and quantitatively monitored using an in-situ strain gage, a sensitive and reliable thod. The results support previously reported trends for the effect of deposition potentials and redox switching of PPy. The roughness of the substrate significantly affects the maximum strain attainable. When the surface roughness and the amount of PPy are judiciously chosen,the actuator exhibits 3 times higher strain than PPy actuators prepared with a thermevaporated Au surface. It is also important that long-term stability is significantly proved by good adhesion between PPy and the EcAu. These effects are probably a result of better adhesion and electrical contact due to mechanical interlocking of the PPy me ally im and the metal electrode.

PAGE 129

CHAPTER 5 ING FOR USE IN RAPID ELECTRODE PATTERN ADVANCED CONDUCTING A critical step in the advancement of conducting polymer actuators and devices is s rapidly and accurately at a low production cost. Conducting polymers altimagination and their ability to fabricate the desired electrode pattern. 133-136 With these techniques it is possible to rapidly design and fabricate virtually any electrode pattern desired within a matter of hours and at a very low cost compared to standard electrode fabrication. er actuators based on the basic cantilever/bimorph design but capable of producing linear strains of up to 24% depending on the CP segment length (in-plane strain (linear stain) on standard CP bimorph actuators is typically ~1% max). The design of these linear actuators is based on altered bimorph and shell type (CP on both sides of flexible substrate) configurations with staggered CP segments that have independent oxidation/reduction control. POLYER ACTUATORS AND DYNAMIC SURFACES 5.1 Introduction the ability to develop new electrode pattern hough inherently susceptible to polymerization conditions are easily applied to any electrode surface in any configuration. Therefore the design/form and capabilities of these devices are only limited to the designers Here we describe the development and use of a new rapid patterning technique very similar to but developed separately from the line patterning technique developed by MacDiarmid and coworkers. The initial use of this technique was to produce linear conducting polym 104

PAGE 130

105 The aforementioned rapid patterning technique was also used to develop the base electrode structure for a new actuator design which results in a thin film surfaces with dynamic surface properties. 5.2.1 Rapid Electrode Patterning the printer. For simint (A) and printed on Kapton used to fabricate linear actuators (B) and a negative 5.2 Rapid Electrode Patterning Techniques The rapid patterning technique developed in our lab utilizes standard laser jet or ink jet printers and a desktop computer to produce the patterns. A negative image of the desired pattern is first electronically designed on the computer using any number of drawing or drafting software programs, such as AutoCAD, PowerPoint, and Adobe Illustrator which can be used to produce detailed and accurate patterns. However the fidelity of the final printed pattern is directly dependent on the resolution of ple pattern designs and for ease of use Microsoft PowerPoints drawing features are more than adequate. Once the desired negative pattern is created it is then printed onto the desired substrate utilizing standard office laser jets or ink jet printers. Figure 5.1 Negative patterns of linear actuator produced using Microsoft Powerpopattern used for dynamic surfaces (C)

PAGE 131

106 Kapton polyimide and PET (standard laser jet transparencies) were used as thesubstrate materials in this research. Figure 5.1 shows two examples of different patterndesigned using Powerpoint. O s nce the negative pattern is printed on the substrate the whole device is coate standard sputter or ess 127, 128gold (or other metal) depositing techniques can al on ited er 5.2.2 Line Patterning The m and coworkers is applicationl layer. The previous method utilizes standard metallization techniques followed by electropolymerization of the conducting d with gold (or any other desired material) using thermal coating techniques. Electrol so be used to form the electrode layer. After the gold coating process the device is rinsed/stripped with acetone and ethanol to remove the printed ink pattern and the goldtop of the pattern. This stripping process leaves only the gold that was directly deposon the substrate. This is the final step in forming the electrode layer. After this stepstandard electropolymerization techniques are used to deposit the desired conducting polymer yielding the final CP device. Figure 5.2 depicts the rapid electrode patterning technique. Figure 5.2 Rapid electrode patterning process for the development of conducting polymdevices. ethod of line patterning developed by MacDiarmid 133-135 very similar to the method described above. The difference in the two techniques is the of the conducting polymer and/or meta

PAGE 132

107 polymed. rencies to a commercial dispersion of poly(3,4-ethylenedioxythiophene) liquid crystal displays.134-136 Work currently being conducted in the Reynolds research group (Univ. Florida Department of Chemistry) utilizes a modified line patterning techniquelectrode formation technique uses electroless gold deposition to deposit gold onto the standard electopolymerization techniques are used to form the final CP device. r. Line patterning uses selective chemical absorption to form the conducting polymer and/or metal layers. The substrate, after it is patterned, is exposed to a solution or vapor containing the material (conducting polymer or metal solution) to be depositeDue to the chemical and/or physical differences between the substrate and the pattern lines the material will deposit selectively either on the substrate, the printed lines, or both. MacDiarmid 133 showed that by exposing a patterned substrate (PET transparency) to an aqueous polypyrrole solution for 20 minutes and by changing the polypyrrole dopant ion from toluenesulfonate to chloride he could get the PPy to preferentially deposit on the printed lines or both the printed lines and the bare substrate respectively. By ultrasonicating the film in toluene the printed lines are removed and only the PPy deposited directly on the substrate remains. It has also shown that by exposing standardPET transpa (PEDOT, Baytron P Bayer Corp.) that the PEDOT preferentially wets the transparencies and not the printed, patterned lines. By repeatedly exposing the patterned transparency to the PEDOT solution and then drying a patterned PEDOT film was be produced on the transparency. These CP patterned substrates were then used to produce simple push button type switches, field effect transistors, and as the electrode layers for e for use in photochromic devices and polymer light emitting diodes. Their patterned substrate instead of a conducting polymer. The pattern is then removed and

PAGE 133

108 5.3 Linear Actuators 5.3.1 Background As stated earlier there has been a lot of work put into the development of polymer based actuators. One of the most promising polymers of choice is the class of conductpolymers. However most of the research that has been conducted in this area has been on the development of bending/cantilever type actuators. Currently there is a drive to develop polymeric actuators capable of producing linear strain. The two main forms of linear actuators being developed are free standing conducting polymer films ing inear coworkers ve l. llinear strains of 15.1%. By using other common organic solvents such as propylene carbonate 44, 71, 137, 138 and fibers. 46, 139, 140 Conducting polymers typically are capable of producing an in-plane strain (lstrain) of only ~1%. This was also the case of CP linear actuators developed by Della Santa 44 and Madden 71 et al. Della Santa showed that polypyrrole linear actuators made from free standing films are capable of producing strains of 1%. Madden andhave produced encapsulated PPy linear actuators capable of producing linear strains of 2%. However these designs have been improved on by Bay and Hara. Bay et al. 138 haproduced linear actuators capable of producing strains on the order of 12% from free standing films of PPy doped with alkylbenzene sulfonate electropolymerized on gold foiHara et al. 137 formed highly flexible PPy (free standing film) linear actuators by electropolymerizing PPy from a methyl and butyl benzoate solutions of tetra-n-butyammonium tetrafluoroborate (TBABF 4 ) and pyrrole on different electrode materials. Films produced from methyl benzoate/TBABF 4 solutions on titanium electrodes and redox cycled in NaPF 6 aqueous solutions produced linear stains of 12.4%. While filmsproduced from butyl benzoate/TBABF 4 solutions on nickel electrodes produced

PAGE 134

109 and acetonitrile during electropolpable of producing strains of 8When aqueous solutions of NaBF4 were employed during electrter rain ectively. However as the anion size is increaL tilever ting s t ymerization actuators ca 10% can be formed. opolymerization brittle films capable of producing strains of only 0.3% were formed. Qi et al. 139 have produced polyaniline (PANi) fibers capable of producing linear strains of 0.51-1.18% in aqueous solutions. They showed that by changing the counion used during redox cycling. As the counter ion size increased from fluorine (F, 10.3 ) to chlorine (Cl, 31.0 ) and bromine (Br, 31.0 ) the produced linear stincreased from 0.76% to 1.05% and 1.18% resp sed beyond bromine to BF 4 (31.0 ), ClO 4 (52.9 ), and CF 2 SO 3 (70.4 ), the produced linear strain decreases from 0.57, to 0.51, and to 0.40%, respectively. u et al. have produced dry PANi solid-in-hollow fiber actuators (solid fiber insidea hollow fiber filled with a solid polymer electrolyte (SPE) gel) capable of producing linear strain on the order of 0.90%. These values are comparable to those obtained by Mazzoldi and coworkers of 0.55% for dry perchlorate (ClO 4 ) doped PANi/SPE fiber actuators. 5.3.2 Cantilever based linear actuators The linear actuator design developed in our lab is based on the standard candesign used in typical bimorph actuators. The basic design consists of placing alternaconducting polymer segments on adjacent sides of a flexible conductive substrate. Thihas the same effect as placing multiple cantilever actuators end to end so the final actuated device forms an S shape. The pattern is formed by masking off sections of the substrate with tape to preventhe electrochemical deposition of the conducting polymer. After the CP deposition

PAGE 135

110 process the tape is removed and the device is ready to be used. An example of this design and an AutoCAD model of this device is depicted in Figure 5.3. Figure 5.3 Diagram of basic linear actuator design and AutoCAD model used to predict the overall developed linear strain for the device. The initial design consisted of alternating 10mm strips of PPy placed on adja cent sides ith an overall dimension of 2mm X 50mm. the device by controlling the length of the PP, -s is higher than the actual experimental values. of a thin copper foil. This device produces a linear strain of -9.75% during oxidation (contraction) as depicted in Figure 5.4. Subsequent revisions of this design employed a gold coated (EvAu) flexible polyimide substrate instead of a solid copper foil. Utilizing this design the PPy segment lengths was produced in 5, 10, and 20 mm segments w We can adjust the overall strain produced by y segment length. The overall strain produced by the device also increases with longer segments. The resulting strains produced by actuators with 5, 10, 20 mm PPy segment lengths are: -2.44, -12.41, and -23.84% respectively. The deflection of the actuators was also modeled in AutoCAD 2000 and the calculated strains were -3.6714.18, and -49.62% respectively. Thi

PAGE 136

111 The model is based on a uniform 1% (max reported strain for PPy switching) produced in the PPy layer resulting in a perfect S or sine wave shape i strain n the device (as shown in the moduced in this layer is probably mental strain. By back calcularequired to produce the experime curvature in each of the PPy segm straight in the reduced state which is not the casages of the y altern f the Figure 5.4 Initial PPy linear actuator based on 10mm segments deposited on copper foil. odel shown in Figure 5.3). However the actual strain pruch less the 1%, which would account for the lower experimting it was determined that the strain ntal strain values was 0.66, 0.93, and 0.81% respectively. However as stated above this model assumes a completely uniforments and that the actuator is completelye in the actual devices. Figures 5.3, 5.4, and 5.6 show im actual devices in action. From this it can be said that linear actuators can be fabricated b ating the placement of conducting polymer segments on opposite sides of a flexibleconductive substrate. However better control over the placement and alignment oconducting polymer segments is needed to improve the linearity of the actuation. Picture of oxidized state enhanced to show placement of PPy.

PAGE 137

112 Figure 5.5 PFigure 5.6 PPpolyimucing linear strain. A new mor the fabrication ofover conducting polyming the created utilizing two separate ique places both the working and counter electrode on the actuator itself (Figure 5.7) Py linear actuator based on 10mm segments deposited on EvAu coated polyimide strip. y linear actuator based on 20mm segments deposited on EvAu coated ide strip. This first generation device was crude and simplistic but effective in prodethod of rapid electrode patterning was developed f these actuators to improve reliability and linearity. More accurate control er placement is accomplished by designing and pattern electrode pattern electronically. Utilizing this technique, a new electrode design was electrodes patterns on a single actuator. This techn

PAGE 138

113 Figure 5.7 Single sided rapid electrode pattern design and illustration of how it works. The green and blue areas represent th e separate electrode patterns (working and counter electrode). As stated abative of the pattern utilizing computers. After the electrodestandard laseconductive mation techniques. The coated n. This leavest is deposited directly on the substrate and not the printed patterand ove, the electrode pattern is created by first designing the neg aided design or other computer drawing program design is created it is then printed onto the desired substrate using either a r or ink jet printer. After the substrate is patterned it is then coated with terial (typically gold) using standard metalliza substrate is subsequently washed in acetone and ethanol to strip off the printed patter only the gold tha n. After this step, the desired conducting polymer can be electrochemically deposited on the gold electrode creating the finished linear actuator (Figures 5.8 5.9).

PAGE 139

114 Figure 5.8 Diagram of linear actuator fabrication process. Initial negative CAD electroddesign (A) negative electrode design printed on substrate (B) produced patterned gold electrode f e rom the mask (C) electrochemically deposited Figure 5.9 Pictures of linear actuator fabrication process. CAD design of negative patterned substrate before acetone wash (C) patterned gold electrode after pattern removal (D) electrochemically deposited conducting polymer on the patterned gold electrode (E). Even though this design shows promise, a more advanced design was fabricated utilizing the backbone type actuator concept. As described earlier, the backbone design employs two conducting polymer layers placed adjacently to each other on opposite sides of the flexible substrate. In this design one side acts as the working electrode while the other side serves as the counter electrode. Therefore as one side is oxidized and expands conducting polymer on the patterned gold electrode (D). electrode pattern (A) negative pattern printed on substrate (B) gold coated

PAGE 140

115 the other side is reduced and contracts. This simultaneous push-pull strain and force produced by the device. In this advanced linear actuatoprinted on both sides of the fle designs are depicted in Figure 5.10. process increases the r design the negative electrode pattern is exible substrate and then processed. Side views of both of thesOne nment or double side the substrae pattern is printed again. However if the pattern reg Figure5.10 Side view diagram of bilayer (A) and backbone (B) type linear actuator designs. problem with the backbone type linear actuators is the alig registration of the electrode patterns on opposite sides of the substrate. To print the d pattern the electrode pattern is printed on one side of the substrate and thente is simply turned over and th istration is off it will cause misalignment of the printed electrodes pattern. This will cause twisting and other erratic movement in the linear actuator. Due to this problem care must be taken to balance the designed grid of patterned electrodes in the

PAGE 141

116 computer software (see Figure 5.10 B) and that the substrate is aligned and centered properly in the printer. Another issue is the interlayer adhesion of these devices. Failure at either the conducting polymer/Au interface or the Au/substrate interface is detrimental to the performance and lifetime of the device. The effects of the conducting polym er/Au the overall peically deposited go Due to the printing process somer to Au treatment process can also be usede surface as well imtrate is not properly cleaned priot adhere well to the subs the device inoperable. If residual debris is not removed from the substrate prior to Au deposition or if the ced in the and is interlayer adhesion were discussed in the previous chapter. Previously, it was shown that rformance of the device was enhanced by applying an electrochemld surface onto the preexisting evaporated gold surface. However a new problem arises with this patterning technique. residual debris and oils can contaminate the substrate surface prior to Au deposition. The substrate should be cleaned with acetone and ethanol priodeposition process. The application of a short (~1 minute) plasma to thoroughly clean the substrate surface and will roughen thproving the adhesion of the deposited Au layer. If the subsr to Au deposition then it is likely that the deposited Au will notrate and will eventually delaminate from the surface rendering toner particles overspray the unmasked areas wholes/defects can be produevaporated gold electrode. White light optical profilometry (Wyko NT1000) was used to evaluate the surface roughness and printed line fidelity for polyimide, EvAu polyimide, EcAu polyimidePPy/EcAu polyimide substrates (Figures 5.11-16, magnification = 25X). The NT1000capable of surface height measurements from 0.1 nm to 1 mm and is non-contact

PAGE 142

117 therefore the surface is not altered during measurements. The EcAu treatment was conducted at -0.9 V for 10 minutes on EvAu coated polyimide and PPy was electres ochemically deposited at +0.8 V to 2.24 C/cm 2 The root mean squared surface roughness (RMS or R q ) of these surfaces changfrom ~1.05 m for Pi to ~315 nm for EvAu/Pi, ~631 nm EcAu/EvAu/Pi, and ~7 mPPy/EcAu/EvAu/Pi. Some surface defects are also present in these deposited films. Figure 5.13 shows a defect in the EcAu/EvAu/Pi electrode structure caused by a stray toner particle (over spray) from the pattern printing process. Figure 5.11 WLOP image of the polyimide substrate used in the construction of linear actuators before patterning (Magnification = 25X). Figure 5.12 WLOP image of the polyimide substrate coated with evaporated gold (Magnification = 25X).

PAGE 143

118 Figure 5.13 WLOP image of electrochemically deposited gold on top of evaporated gcoated polyimide (Magnification = 25X). old Figure 5.14 WLOP image of polypyrrole (+0.8V to 2.24 C/cm2) electrochemically deposited in EcAu/EvAu coated polyimide (Magnification = 25X). Figure 5.15 WLOP image of patterned PPy/EcAu/EvAu on polyimide (Magnification = 25X).

PAGE 144

119 Figure 5.16 WLOP image of exposed base polyimide substrate between two patterned PPy/EcAu/EvAu wires (Magnification = 2 5X). correspond to line widths of 264 m, 176 m, and 88 m respectively. The patterns were tested on thrree different printers, i.e., Xerox Phaser 6200 laser jet printer, HP Deskjet 6122, and Lexmark i3 Inkjet. Figure 5.17 Schematic of conducting polymer actuator based dynamic surface. Initially, the desired pattern with three different line thicknesses, i.e., 0.75, 0.50, and 0.25 pt, was printed on standard PET transparencies. The patterned line thicknesses 5.4 Actuator Based Dynamic Surfaces Another application for this technology is in the development of conducting polymer actuator based dynamic surfaces. In this project a patterned conducting polymer electrode/actuator is deposited beneath a flexible surface coating. Therefore as the conducting polymer expands and contracts during redox cycling it will replicate its pattern in the flexible top coating (Figure 5.17).

PAGE 145

120 The Xerox Phaser 6200 is capable of printing at 2400 X 600 dpi in color and 1200 X 1200 dpi in black and white (BW). The HP Deskjet 6122 is capable of printing at 4800 X 1200 dpi (optimized) in color and 1200 X 1200 dpi BW. And the Lexmark i3 has a print resolution of 2400 X 1200 dpi in color. The patterns produced on the Xerox Phaser 6200 in black and white at a resolution of 1200 X 1200 dpi exhibited too much overspray in the unpatterned areas and small dispersed holes in the solid patterned areas. This ects in the final Au electrode. The Au electrodes caused def produced on the Xerox Phaser were not completely conductive due e electrode structure. Similar effects were produced from other laser jet printers at this resolution. The less overspray compared to the Phaser. However, the solid patterned area had fewer holes that were larger compared to those printed on the Phaser printer. ilar in reso to multiple large defects in th HP Deskjet 6122, which has a higher printing resolution in black and white, produced The next set of tests utilized color printing due to higher resolution. The Deskjet is capable of printing at 4800 X 1200 dpi in color using computer software enhanced mode, which does not translate to a real line image resolution. The Lexmark i3 produces a true color line printing resolution of 2400 X 1200 dpi. The i3 produced patterns were sim lution to the Deskjet generated lines, but have less overspray and more uniform density in the solid patterned areas, i.e., no holes in the printed pattern(Figures 5.18 and 5.19).

PAGE 146

121 Figure 5.18 Images of 0.75, 0.50, and 0.25 pt line patterns produced with Xerox Phas(Magnification = 7X). er 6200 laser jet, HP Deskjet 6122, and Lexmark i3 inkjet printers The Lexmark i3 produced the best overall patterns in blue ink compared to the Xerox Phaser 6200 and HP Deskjet 6122 printers, which are black and white. However direct gold coating of the patterned inkjet transparencies is ineffective. Standard inkjet Figure 5.19 Images of solid printed pattern areas produced with Xerox Paser 6200 laser jet, HP Deskjet 6122, and Lexmark i3 inkjet printers (Magnification = 14X).

PAGE 147

122 transparencies are coated with a hydrophilic surface coating to improve the wetting and drying characteristics of the ink on the transparency. This in turn prevents smearing of the printed ink on the transparency surface. If this coating is left intact the deposited gold will be deposited on this coating and not the base PET substrate. Once the substrate is washed to remove the printed pattern the Au that was deposited on the nonpatterned areas will also be removed. Removal of this this surface coating is removed then the ink will not wet or dry correctly on the PET substrate. This results in the ink beading up smearing and thus rendering the pattern useless. Another alternative to printing and forming the electrode surface directly on the PET transparency is to use the printed PET transparency as an inexpensive photomask for use in photolithographic processing. Utilizing this technique PET photomasks can be produced rapidly and inexpensively with various pattern features. However the of the mask is still limited by the printing resolution of the printer. Also th resolution e PET photomask can be used in the patterning of almost any substrate. This is ask to develop it. If a positive photoresist is used the areas that are exposed to the UV light degrades and are subsequently rem advantageous in the case of the actuator based dynamic surfaces. A stiff substrate is required in this application to direct all the produced strain in the system toward the surface of the flexible top coating. Photolithography is carried out by spin coating a photoresist onto the surface that is being patterned. The photoresist is then exposed to UV light through the patterned photom oved during a rinsing process. This leaves only the photoresist that was masked (unexposed) from the UV light by the photomask. The sample is then gold coated in a similar process to that described previously for the

PAGE 148

123 rapid patterning process. The remaining patterned photoresist is then removed leaving only an Au pattern. Initially, the blue pattern produced on the Lexmark i3 printer was evaluated as a photomask. However during the photolithographic processing some of the UV light penetrated through the blue photomask. The partial exposure of the underlying photoresist can result in a random speckled pattern to develop in the photoresist. This can caused random holes and other defects to be form in the patterned photoresist resulting in random Au spots to be produ ced on the patterned electrode surface were the holese ts changed ter also m 1). ch darker and less transparent than a similar mask produosure were. Figure 5.20 shows 7X optical images of the 0.75 pt blue photomask and thresultant patterned photoresist (computer enhanced/colored) on a glass slide developed from this photomask. The random holes are shown in image B and appear as white spoin the patterned brown photoresist. To overcome this problem the ink color was to brown. However the black toner cartridge had to be removed from the printer to maintain good pattern resolution. When the black cartridge is installed the prinapplies black ink which has a much lower dpi than the color ink. This produced randolarge black ink spots which degraded the overall resolution of the pattern (Figure 5.2The final printed photomask was mu ced from blue ink. This in turn prevented the UV light from penetrating the photomask and exposing the underlying photoresist. This eliminated the partial expof the photoresist without any loss in pattern resolution. Examples of the blue and brown photomask used to pattern glass slides are presented in Figure 5.22.

PAGE 149

124 Figure 5.20 Images of 0.75 pt blue photomask (A) and enhanced image of resultFigure 5.21 Images of printed field and 0.75 pt brown photomask with black toner cartridge (A) and without black toner cartridge (B) (Magnification = 7X). Figure 5.22 Examples of computer designed patterns used to make blue (A) ing patterned photoresist coated glass slide (B) (Magnification = 7X). and brown (B) photomasks.

PAGE 150

125 5.5 Conclusions A new patterning technique for the rapid development of various has been described in this chapter. The patterning technique was used to develop er based linear actuators. Standard office equipm electrode patterns conducting polyment such as desktop to produce various electrode patterns. The patterns facilitated the rapid generation of linear actuators at a mInitial linecapable of producing linear stlengths respectively. The developed uelectrodes onrnal counter electrode sed in the back bone type devices. In these devices conducting polymer layers are placed on the working electrode while the opposite layer s used is causes one layer to expand the device. It was show directly resolution and lower degree of over spray. This in turn produces patterns and final devices with higher fidelity and accuracy. However the ink jet printers suffer from one major set back. The transparencies used during ink jet printing are coated with a computers and printers were utilized in conjunction with photolithographic methods inimal cost. ar actuator designs were basic in construction and were shown to be rain of 2.44, 12.41, 23.84% for 5, 10, 20 mm PPy segment new rapid patterning technique was used to develop new advanced linear actuator designs. Both a single sided and double sided design was tilizing this technique. These designs place both the working and counter the same surface. This eliminates the need for an exte in the system. The double sided design utilizes the same push-pull type technique u opposite sides of the flexible substrate and as one layer is used as as the counter electrode. Thwhile the other contracts increasing the overall developed stress and strain produced by n that the final device fidelity or pattern fidelity isdependent on the resolution of the printer used. Ink jet printers were shown to produce patterns of higher quality compared to laser jet printers due to there higher printing

PAGE 151

126 hydrophilic coating used to improveying of the ink. This coating prevensive photomasks. Utilizing these photomasks glass slides coate jet d the wetting and dr nts the deposition of the gold electrode layer directly on the transparency. This prevents the gold from sticking to the transparency substrate rendering the pattern useless. If this coating is removed the ink from the printer will not stick and dry to the transparency also rendering the pattern useless. This makes the use of ink jet printers inadequate for the production of electrode designs. However these patterns have been shown to be suitable as inexpe d with standard photoresist and developed with standard photolithography techniques were patterned and used to produce the base electrode for conducting polymer actuator based dynamic surfaces. The only problem experienced with the use of inkprinted photomasks was with the partial exposure of the photoresist under the maskeareas. This was found to be due to the color used to produce the photomask. Blue ink was found to produce patterns of high fidelity however the blue ink allows some of the UV light used to develop the photoresist to penetrate through the mask creating small holes in the final electrode pattern. It was found that by changing the mask color from blue to brown the partial exposure of the photoresist could be eliminated.

PAGE 152

CHAPTER 6 PDMSE BASED DYNAMIC NON-TOXIC ANTI-FOULING SURFACE COATINGS6.1 Introduction As mentioned earlier biofouling is the result of marine organisms settling, attaching, and growing on submerged marine surfaces and is a very expensive and complex problem facing military, commercial, and private naval vessels and struThis process is initiated within minutes of expose of the surface to the marine environment and is initiated by the absorption of dissolved organic materials onto the submerged surface forming a conditioning film. Once this film is formed various mariorganisms settle and then grow on this surface and on top of other organisms already growing on this surface. This is further complicated by to the fact that there are 12 welldefined geographical zones in the worlds oceans with varying salinity, clarity, temperature, amount and type of micronutrients, and number and type of fouling organisms. Due to the vast number and diversity of marine fouling organisms each ctures. ne -52having its own unique properties and adhesion mechanisms it is difficult to produce a single coating that prevents the settlement of all fouling marine species. It has been shown that the three major surface properties that control the settlement and growth of fouling marine species are surface energy, surface modulus, and surface topography.50-52, 76-82 It has also been shown that different marine species react differently to each of these properties. Therefore while one set of surface properties might prevent one marine species from settling and growing on a given surface the same set of properties might increase the settlement and growth of other marine species.141, 142 127

PAGE 153

128 This makes it very difficult to produce a single marine coating that can prevent the settlement and growth of all fouling marine species. For example, Rittschof and coworker141ined that barnacles prefer to settle owith lower initial wettability. This wconducting settlement assays on bryozization p a s of the electrre s determ n surfaces with higher initial wettability while bryozoans prefer to settle on surfaces as determined by oans and barnacles on surfaces with varying surface energies (through silanof glass surfaces). The approach of this research is to utilize the electrochemically induced changes inthe bulk properties of conducting polymers (CP) during redox cycling to develosurface coating with dynamic surface properties. During electrochemical switching (redox cycling) of conducting polymers in an electrolyte medium (salt solution) counter ions and solvent are drawn into and expelled from the CP matrix due to changes in theinduced charge on CP backbone charge. This influx and expulsion of counter ions and solvent is due to a charge neutralization process which results in a swelling and contraction of the CP matrix. The swelling process also induces a decrease in the overall modulus of the CP material. 94, 143-147 By incorporating conducting polymers into traditional polymeric surface coating (paint or appliqu/tile coatings) thus combining the effect ochemically induced CP charge and swelling with these materials it is possible to produce a single durable surface coating with variable surface energy, modulus, and topography. Dynamic composite films based on PPy and PDMSe have been produced that acapable of generating a 21 change in contact angle under an electrical stimuli. It is felt

PAGE 154

129 that this dynamic change in contact angle might be sufficient to perturb the settlemeand growth of marine organisms. nt ties. Lithium perchlorate (LiClO4), sodium perchlorate (NaClO4), and gelatiA), m) nd the crosslinker solution are then heated to ~38-40C in an oven for 1hr. The g 6.2 Materials and Methods 6.2.1 Materials Pyrrole was purchased from Sigma-Aldrich and was filtered through a neutral alumina (Brockman activity 1; Fisher Scientific) column until colorless before use to remove any impuri n Type B ~255 bloom (Mn = 100,000) were purchased from Sigma-Aldrich and used as received. Ammonium persulfate (APS), dodecylbenzene sulfonic acid (DBSor alkylbenzene sulfonic acid), iron (III) chloride (FeCl 3 ), ethanol, methanol, and tetrahydrofuran were purchased from Fisher Scientific and used as received. All deionized (ultrapure) water was produced using an 18 M Millipore system. Artificial sea water was made using Instant Ocean (obtained from Fisher Scientific) synthetic seasalt and deionized water. 6.2.2 Gelatin Preparation Gelatin films (yellow/orange in color) were prepared by mixing 10wt% gelatin (thermally denatured bovine collagen) in ultrapure water. NaN 3 was added (~200 ppas an anti-microbial. Gelatin films were allowed to gel and dry for 3hr at 25C. Crosslinked gelatin films were prepared by adding 50 vol% crosslinker solution (4.5 wt%glutaraldehyde solution in ultrapure water) after the gelatin films is dried. The dried gelatin film a elatin films un-gel at 40C allowing the crosslinker to mix and react with the gelatin.

PAGE 155

130 6.2.3 Polydimethylsiloxane Elastomer Preparation PDMSe films were prepared us ing Silastic T2 resin (Dow Corning). The resin and curing agent are mixed in is then thoroughly mixed er vacuum until most of the bubbles are removed. The resin is then poure or 0 ml r 1.5 ed to run for ~36 hrs and then terminated by adding excess metha to 1-2 s of rial were soaked for 24 hrs in a solvent oxidizer (FeCl3) solution. The oxidizer solution consisted of either 0.1 M or 1 M FeCl3 in EtOH, IPA, H2O, scCO2, or a 10:1 weight ratio. The resin and degassed und d into sheets and cured between glass or PET cover glass plates using spacers toachieve the desired thickness. The cast films can then be cured at room temp (25C) f24 hrs or at 50, 60, 65, or 80C for 5, 4, 2-3, or 1.5 hrs respectively. 6.2.4 Soluble Polypyrrole Preparation Soluble polypyrrole was prepared from a technique reported by Song et al.. 98 Pyrrole (0.3 mol) and dodecylbenzene sulfonic acid (0.15 mol) are dissolved in 40ultrapure water. The solution is cooled to 0C in an ice bath and allowed to mix fohrs while being degassed by bubbling N 2 After mixing a aqueous solution of ammoniumpersulfate (0.045 mol in 200 ml ultrapure H 2 O) is added drop wise under N 2 using an addition funnel. The reaction was allow nol. The solution is then filtered and washed with ultrapure water and methanolremove unreacted pyrrole, APS, and DBSA. After filtering the resulting PPy powder is dried at 25C for 12 hrs under vacuum. The dried PPy powder plus an additionalwt% DBSA are then dissolved (dispersed) in chloroform using ultrasonication. 6.2.5 Chemical Formation of Conducting Polymer IPN Systems Conducting polymer/base material IPN systems were prepared using an oxidizer insertion method. The general process scheme is described as follows. Samples filmthe base mate

PAGE 156

131 THF. Initially a perchlorate codopant (0.1 M or 1 M L iClO4 or NaClO4) was added to insertd saturated ark brown-orange/black in color. The color of the soCl3 as black in color. The dark brown-orange color of the solution indica a i-automatic scCO2 critica, the ClO4 anion instead of Cl. After soaking the doped films were dried under vacuum for 1 hr to remove any residual solvent. The films were then placed in a sealed container containing pyrrole monomer vapor for 24 hrs. After exposure to the pyrrole monomer the films were drieunder vacuum for 1 hrs to remove any residual pyrrole monomer. Initial THF solution soaking experiments were conducted using a FeCl3/THF solution. This solution was d lution indicates the degree of hydration the FeCl3 has undergone. Hydrated Fe(FeCl3 hexahydrate) is yellowish/orange in color while dry FeCl3 is dark black in color;the same is true for FeCl3 in solution. The same effect was observed in FeCl3/IPA solutions. It was found that if regular (~95% pure) IPA was used to make FeCl3 solutions the solutions turned orange in color while if ultra pure IPA (~+99%) was used the resulting solution w tes that the THF was slightly wet and that the some of the FeCl3 had gone to its hydrated form which is less reactive and potentially less soluble in the PDMSe than the dry form. 6.2.6 Supercritical CO 2 Solution Formation of Conducting Polymer IPN Systems Supercritical CO 2 solution doping of base materials was conducted using Tousimis Samdri 780A critical point dryer. The Samdri 780A is sem l point dryer equipped with a 1.25 diameter by 1.25 height reaction chamber (volume 26 ml). The operating pressure and temperature are ~82.7 bar 5-7% (~1200 psi 5-7%) and 33-39C. The basic construction of this unit does not allow pressuretemperature control or for the recovery of the CO 2 after the reaction process.

PAGE 157

132 Base material samples (on a rack), cosolvent, and oxidants are placed in the reaction chamber and then the chamber i s sealed and cooled to ~ 0C. The chamber is then f is ve set to a pressure a little nd er the re anges in sample weight were monitored throughout the process to determine 6.2.7 ing illed with liquid CO 2 while maintaining the chamber temperature at ~ 0C. Thisincreases the CO 2 charge in the system. Once the chamber is filled it is then heated to~33-39C thus increasing the chamber pressure to above the critical pressure of ~82.7 bar. The pressure is maintained by an automatic pressure relief val above 82.7 bar. If the pressure exceeds this set point the CO 2 is automaticallyvented to the room. After the samples are soaked in scCO 2 for the desired amount of time the chamber is slowly vented at a rate of ~100 psi/min until the chamber is returned to room pressure. Care must be taken not too vented or decompressed the chamber too quickly the scCO 2 that is still dissolved inside the sample will return to the gas state apotentially destroy/explode the samples. In order to determine the actual solvent (scCO 2 ) volume of the scCO 2 chambchamber volume was measured and the volumes of the samples and sample rack wesubtracted from the chamber volume. This was used to calculate the volume percent cosolvent and molar concentrations of the oxidizers. The ch the extent of the FeCl 3 and PPy incorporation into the PDMSe base material. Electrochemical Formation of Conducting Polymer IPN Systems The electropolymerization of PPy inside various base materials is conducted using the following technique. Base material films are solution cast on one side of the workelectrode (either EvAu coated polyimide or stainless steel electrode (other side coated with polyurethane)). The coated working electrode was allowed to soak in the monomer/salt solution for 20 minutes to allow for sample hydration. Pyrrole

PAGE 158

133 electrinutes 5 s the average of five samples. 6.2.9 P he crystal geometry was 50 mm X 10 mme opolymerization was conducted in side the base material at +0.7 V for 130 min an aqueous 0.1 M pyrrole, 0.1M NaClO 4 solution. 6.2.8 Mechanical Property Testing of PPy/PDMSe IPN systems An Instron 1122 load frame updated with an MTS ReNew/E upgrade package andTestWorks 4 software was used to evaluate the tensile strength, stain at break, and modulus of PDMSe dog bones samples. An Interface SMT1-22 (22lb capacity) load cellwas used for these measurements. PDMSe samples were prepared by casting uncuredPDMSe between PET cover glass plates with 1.2 mm spacers and cured at 80C for 1.hrs. Samples were then punched from the PDMSe sheets with an ASTM D 1822-68 Type L bog bone die with 3/8 tabs. Tests were conducted at a cross head speed of 2 inches per minute and displacement was measured with an MTS LX1500 laser extensometer. Data values are reported a ATR-FTIR Analysis ATR-FTIR analysis was conducted on a Nicolet 20SXB FTIR equipped with anATR stage and Omnic ESP 4.1 software. A KRs-5 (Thallium Bromide-Iodide) 45 SP(surface plasmon polariton) ATR crystal was used. T X 2 mm. This technique has a average penetration/sampling depth of 0.6 to 6m depending on the refractive index of the sample and ATR crystal used along with th sampling wavelength. Scans were conducted from 400-4000 cm -1 with a resolution of 4 cm -1 (data spacing = 1.928 cm -1 ) in absorbance mode; scan number = 32. Measurements were acquired using a DTGS KBr detector and KBr beam splitter and mirror velocity of 1.5825 cm/sec; gain = 2.

PAGE 159

134 6.2.10 Optical Microscopy Optical microscopy was used to monitor the phas e size and distribution of PPy in pe (2X objective) equipwere re ucted in electrode and an The CV current-voltage response of the material was monit and s ee minutes prior to EDS mapping. the PPy/PDMSe materials. For this a Leica G27 dissecting microsco ped with a Sony CCD-IRIS color CCD camera and Canopus DV Raptor image captures software and a Zeiss Axioplan II microscope (high magnification images) used to capture optical images of the PPy/PDMSe systems. A standard green filter was used with the Axioplan II microscope analysis to improve contrast between the PPy andbulk PDMSe. 6.2.11 Electrochemical Analysis In order to investigate the electrochemical properties of the materials sample wesubjected to cyclic voltammetry measurements. These measurements were condI.O. at .8 V with a sweep rate of 10 mV/sec vs. an Ag/AgCl reference Au wire working electrode. ored to determine electroactivity and conductivity of the samples. A Omega HHM57 digital multimeter was used to conduct quick conductivity measurements on these materials. Conductivity measurements were taken on each side from corner to corner (diagonal). This results in four measurement locations for each sample. These measurements were repeated three times in random order. 6.2.12 EDS Mapping EDS mapping measurements were conducted on a JEOL SEM 6400 at 15 KeVa 15 mm working distance. PDMSe samples were doped with a 5 wt% FeCl 3 /THF solution for 24 hrs and then dried overnight under vacuum. After drying the samplewere carbon coated for thr

PAGE 160

135 6.3 Background Study d aterial swells in the monomer and electrolyte plied inducing the electrochemical polymerization of the cogh n n appropriate ling the base material (EtOH, IPA, scCO2, THF, etc.). The base matergree e Three possible routes for the direct incorporation of conducting polymers into basematerials have been proposed. The three methods include electrochemical, chemical, ansolution or bulk blending. Both electrochemical and chemical routes to the formationPPy/PDMSe were used in this study. In the electrochemical method the working electrode is coated with an aqueous or organic compatible material depending on the type of electrochemical electrolyte medium/system being used. As the base m solution an oxidizing potential is ap nducting polymer inside the base material. This results in the formation of interpenetrating polymer network (IPN) of conducting polymer inside the base material which initiates at the electrode surface and grows out towards the films surface. The degree of interpenetration of the conducting polymer can be controlled by the electropolymerization potential and deposition time. Another method for the formation of conducting polymer IPN systems is throuthe utilization of chemical oxidants such as FeCl 3 Two different methods can be used inthe formation of these networks. The first method is referred to as oxidizer insertion. Ithis method a given concentration of the chemical oxidizer is dissolved in a solvent capable of swel ial is then soaked in this solution for a predetermined time resulting in the deposition of the oxidizer inside the base material. The soak time determines the deof oxidant insertion into the base material and also the dispersion of the oxidant into thmaterial.

PAGE 161

136 Shorter exposure times ten in the surface layer than in the bulk re es result in more uniform bulk deposition of the oxidaymer in a ethod sed. This method is referred to as monol All electrochemistry was done utilizing an Ag/AgCl reference electrode unless otherwise noted. Crosslinked gelatin films were cast on EcAu (T=1 hr, I=100 ma) d to deposit more oxidant esulting in materials with higher surface conductivities due to the higher concentration of the resulting conducting polymer on the surface of the material. However the resultant material is not bulk conductive and is subject to electrode damagdue to surface contact. Longer soak tim nt leading to more bulk like conductivity but with lower surface conductivities. After the soaking process the doped base material is exposed to the conducting polmonomer or monomer vapor resulting in the formation of a conducting polymer IPN inside the base material. The doped base material is soaked in either pure monomer (if liquid) or a solvent monomer solution. Vapor phase polymerization is conducted sealed chamber containing a small amount of volatile monomer; the doped base materialis suspended above the monomer. The second chemical polymerization method is similar to the previous mexcept the oxidizer and monomer soak steps are rever mer insertion due to the monomer soak step being done first instead of oxidizer soaking. In this process the penetration of the CP IPN can be altered by the oxidizer solvent used. If an aqueous oxidizer solution is used and the base material is not water swellable then the CP formation will be limited the surface layer of the base material. If the base material is swellable in the oxidizer solvent then a deeper CP IPN formation wiloccur. 6.3.1 Electrochemical formation of Conducting Polymer IPN Systems In the initial tests electrochemically prepared CP/gelatin IPN systems were investigated

PAGE 162

137 treatete the the still looked like gelatin (shiny black). The PPy/gter ith to form in spots randomly dispersed across the electrode surface inside the gelatin e gelatin film. The rking d EvAu coated polyimide films (in a mold) using the previous described method and allowed to dry over night. The crosslinked gelatin films tended to delaminacorner of the EcAu/EvAu coating from the polyimide substrate when dried due to contraction of the gelatin. Polypyrrole was electrochemically polymerized inside the crosslinked gelatin film (t = 50 minutes). The gelatin film was allowed to presoak in the monomer/salt solution for ~ 20 minutes prior to the electropolymerization process. This was done to allow the gelatin film to completely hydrate. After 50 minutes the gelatin film visually turned auniform light grey/black color and had a non-conductive surface. Due to the nonconductive surface PPy formation was continued for an additional 50 minutes. The film turned completely black after 15 minutes. However after 50 minutes the PPy/gelatin film surface elatin EcAu/EvAu film completely delaminated from the polyimide substrate afthe electropolymerization process. The surface was completely (uniform) black in color and nonconductive. This experiment was repeated on a stainless steel electrode (other side coated wpolyurethane). Pyrrole electropolymerization was conducted at +0.7 V for 130 minutes in an aqueous 0.1 M pyrrole, 0.1M NaClO 4 solution. The gelatin films were not allowed to presoak in the monomer/salt solution prior to electropolymerization. Polypyrrole started film after 1hr. The PPy spots continued to grow but did not totally cover th esulting PPy/gelatin film was easily delaminated from the stainless steel wor

PAGE 163

138 electrode after it was dried. The side of the film that was in contact with the working electrode was conductive while the outer surface was nonconductive. In both cases the electropolymerization process was too short to allow for the complete formation of polyp yrrole throughout the bulk of the material resulting in a nonco surface effects of the base m ng the rm ion method was also used to form PPy/gelatin films. Cross y described however 1 M and 3 M of FeCl3 were added in with the gelatin powder. Both mixtures did not gel and the experiments were terminated at this point. The gelation of nductive sample surface. The polymerization process begins at the working electrode and propagates out toward the film surface. Longer electropolymerization times (>130 min.) should result in the formation of a more completely bulk conductive PPy/gelatin film. However once the PPy reaches the surface it would most likely form acomplete PPy layer on top of the PPy/gelatin film thus negating the aterial. In this research a film surface comprising of a composite or IPN structureis desired not a completely PPy surface. Also from the first experiment by presoakigelatin film in the monomer solution prior to electropolymerization a more unifopyrrole polymerization occurs instead of the spotty growth described above. 6.3.2 Chemical formation of Conducting Polymer IPN Systems The oxidizer insert linked gelatin films were soaked for 4hrs in the oxidizer solution (1 M FeCl3/LiClO4; aqueous). These samples were prepared using pyrrole vapor and a pyrrole solution (1M; aqueous). The samples were stored in an aqueous LiClO4 solution (0.1 M; aqueous) until tested. All samples had a shiny black nonconductive surface. This suggests that the surface was mainly gelatin and not PPy. Another set of experiments was carried out using gelatin where the gelatin films were predoped with oxidizer prior to gelation. Gelatin films were prepared as previousl

PAGE 164

139 gelatin is an ionic process; it is believed that the gelation process was interrupted due tcharge screening from the FeCl3 salt. o e). lar) is anol nomer vapor mples in hen the iment The oxidizer insertion method was also tried using Silastic T2 resin (PDMSThe choice of solvent is very important in this process. A solvent that is capable of dissolving the oxidizer (polar) used and swelling the base material (PDMSe, nonporequired. The degree that the solvent is capable of swelling the base material dictateshow deep the oxidizer will penetrate into and deposit inside the base material. In this experiment ethanol was used because it is a good solvent for the oxidizer FeCl3, but a poor solvent for PDMSe. This caused surface swelling of the PDMSe. The deposition of the oxidizer was limited to the surface layer of the film. Thus PPy modification of the PDMSe films was also localized to the surface. PDMSe films were prepared as previously described (0.1 M FeCl3/NaClO4 ethsolution; 70 Hrs). The doped PDMSe films were then exposed to pyrrole mo for 48. Captive air bubble contact angle measurements were conducted on these saultrapure water. A reproducible contact angle change of 21 was measured wpotential was switched between V (Figure 6.1). The contact angles for this experwere 23 31 40 and 44 degrees for potentials of +1.0, 0.0, -0.6, and -1.0 V respectively (Figures 6.2).

PAGE 165

140 Figure 6.1 Images of captive air bubble contact angle measurements on PPy/PDMSe oxidizer insertion films. Figure 6.2 Captive air bubble contact angle da 2role/DBSA was prepared as previously described. Two DMSe films were placed in a 1wt% PPy, 2wt% DBSA chloroform solution and allowed to soak for ~20hrs. This resulted in the complete degradation of the PDMSe samples o determine what caused the degradation of the PDMSe films two new PDMSe films were placed in chloroform and allowed to soak for 24hrs. After 24hrs both films ta for PPy/PDMSe oxidizer insertion films conducted at potentials of -0.8, 0.0, and +0.8 V versus Ag/AgCl in distilledHO. 6.3.3 Conducting Polymer Blend Formation by Solution Blending A solution blending process was also tried to form PPy/PDMSe blends. In this experiment soluble polypyr P T 1525-1.5 2040-1.0-0.50.00.51.01.5E(V) versus Ag/AgClContd 30354550act Angle (eg)

PAGE 166

141 were come chloroformmajor com(http://www.dynaloy.com/silicone_products.html) produces a commercial grade series of 6.4.1 Introduction As stated above solvent soaking is a viable method for the preparation of polypyrrole/PDMSe IPN systems. However solvents like ethanol are not the best choice for this base material. Ethanol is a good solvent in the sense that oxidizers such as FeCl are readily soluble in it; but ethanol is not a good solvent for the swelling of PDMSe. Ethanol is a poor solvent for PDMSe and is only capable of solvating/swelling a very limited surface layer of the PDMSe sample. This results in PPy/PDMSe IPN formation only in a very thin surface layer, thus limiting the conductivity of the resultant film and pletely intact with no signs of degradation. The films were left sealed in th for over 4 weeks with no signs of degradation. It is believed that the sulfonic acid group of the excess DBSA caused the degradation of the PDMSe network. And it has now been determined that DBSA is the ponent in some commercial silicone solvents and removers for both cured and uncured silicone. Dynaloy Engineered Chemistries silicone solvents/removers under the tradename Dynasolve. Dynasolve consists of either methylene chloride or naptha (60-100%) and 10-30% DBSA which are capable of removing silicone coating in under 24hrs. The methylene chloride containing product is the fastest acting of the series but can also attack other materials present such as metals, epoxies, and polyurethanes. By replacing the methylene chloride with different grades of naptha the side reactions are limited unless the system is contaminated with water. From this it was determined that the solution blending process previously described is not suitable for use with PDMSe. 6.4 Polypyrrole/PDMSe IPN Formation 3

PAGE 167

142 the overall durability of this conductive layer. Due to the very thin layer of PPy/PDMSany surface damage can potentially destroy the electrical pro e perties of the film and theref g the base material to greater degree sit the ling, and low cost. Supercritical CO2 is formed when CO2 is compressed andtemperature and bout 73.8bar (1070psi) and 31.1C so most scCO2 processes run at tempesi). A to ore the dynamic properties of the film. In order to prevent loss of electrical properties a more bulk conductive material is needed. In order to form a more bulk conductive system deeper IPN formation is needed. The use of better solvents that are capable of swellin(solubility parameter closer to that of the base material) should result in the formation of IPN systems that penetrates deeper into the bulk of the sample. However the solvent must also be able to adequately solvate the desired oxidizer/dopant in order to depooxidant into the base material. 6.4.2 Supercritical Carbon Dioxide Solution Doping 6.4.2.1 Introduction Supercritical carbon dioxide (scCO 2 ) has gained a lot of attention lately as being an environmentally friendly replacement for organic solvents due to its non-toxic, non-flammable nature, ease of recyc heated above its critical point. The critical pressure for CO 2 is a ratures between 32-49C and pressures between 73.8-241.3bar (1070-3500pphase diagram for CO 2 is depicted in Figure 6.3. Once in the supercritical phase CO 2 exhibits the properties of both a liquid and a gas. The lower surface tension of scCO 2 compared to liquid CO 2 enhances the ability wet and penetrate various materials. However by maintaining some of its liquid properties, scCO 2 is also able to solvate mixtures. The organic nature of supercritical CO 2 allows it to swell and penetrate organic materials very easily. Inorganic compounds

PAGE 168

143 such as inorganic oxidizers are relatively insoluble in scCO 2 however the solubility of scCO 2 can be tuned by varying the process temperature and pressure. Also small amounts of co-solvents such as ethanol and me thanol can also be added to the system to increany http://www.pprc.org/pprc/p2tech/co2/co2intro.html 6.4.2.2 EtOH/scCO2 prepared IPNs 2 versus the ethanol soak treatment. For this study ethanol (EtOH) was added to the scCO2 system as a cosolvent to improve the solubility of FeCl3 in scCO2. Iron chloride is not 31.1 se the solubility of inorganic species. One nice feature of scCO 2 is the ability to ease of removal of any contaminates/additives such as co-solvents, oxidizers, and oils from the system by lowering the pressure and/or temperature below the supercritical point. When this is done the CO 2 returns to the gas phase and can easily be bled off and recycled leaving acontaminated in the bottom of the reaction chamber. Figure 6.3 Phase diagram for carbon dioxide. Adapted from Figure 1 on 73.8 solidliquidsupercritical gasfluidPressure (bar)Temperature (C) -56.4 5.2 73.8 31.1 solidliquidsupercritical gasfluidPressure (bar)Temperature (C) -56.4 5.2 An initial study was conducted to compare the effectiveness of the scCOtreatment

PAGE 169

144 very soluble in scCO 2 but by adding co-solvents such as ethanol the solubility can be greatly improved. For this first study a 0.1M FeCl 3 /NaClO 4 solution was made in EtOH/scCO 2 (1vol% EtOH) and pure EtOH. Two PDMSe samples (cylinders; ~24.75mm X 9.5mm v 1.75ml) were placed in the scCO2 chamber with FeCl3/NaClO4 and E2 treatment was e one into solution) and the PDMSe sampe chamber temperature had reachthe samples remdegraded yellow resin remoxidizend relatively sample. tOH cosolvent. The sample chamber was then sealed and the scCOconducted for 24hrs as described previously. The experiment was duplicated in purEtOH and was conducted in a 50ml polypropylene centrifuge tube with 14ml of a 0.1MFeCl 3 /NaClO 4 solution in EtOH. After 6 hrs the scCO 2 system was visually checked and the oxidizer was not visible through the sight glass (g les appeared swollen and puffy white in appearance. At this point thed 42C. After 24hrs the scCO2 chamber was slowly vented and oved. The PDMSe samples were completely destroyed and only a ained in the chamber (Figure 6.5). However after 24hrs in the r EtOH solution the EtOH soak samples were completely intact aunchanged except for a slight color change due to oxidizer insertion into the PDMSe Figure 6.5 Optical images of degraded PDMSe samples after supercritical CO2 doping with 0.1M FeCl3 dopant and 1vol% ethanol cosolvent for 24hrs.

PAGE 170

145 In order to determine the cause of the degradation of the EtOH/scCO 2 PDMSe samples a scCO 2 experiment was conducted with 2vol% EtOH and oxidizer. Afterthe two PDMSe samples were intact but swollen and white in appearance. The swhitish appearance slowly disappeared over about a 1 hr period and the samples returned to there original condition. This is believed to be due to the EtOH dispersed inside the PDMSe sample. From this it is believed that the degradation of EtOH/scCO 24hrs wollen ng). e ples e samples exSe. Optical 2 PDMSe samples was primarily due to the oxidizer concentration. In order to prevent sample degradation the scCO 2 oxidizer insertion method was repeated but at different oxidizer and co-solvent concentrations. The oxidizer concentration as dropped from 0.1M to 0.01M and the EtOH concentration were increased to 2vol% from 1vol%. After a 24hr treatment, the two samples were completely intact prior to the decompression of the sample chamber (chamber ventiAfter decompression the samples split and fractured slightly due to the inability of the CO 2 to diffuse out of the center of PDMSe samples fast enough. From this it was determined that a slower decompression rate is needed to prevent the destruction of thsamples. After the doping treatments both the EtOH/scCO2 and EtOH prepared samples were placed in a 50ml PP centrifuge tube with 2ml of pyrrole monomer and allowed to soak for 24hrs. After 24hrs the samples were removed and allowed to dry (evaporate residual pyrrole monomer) at room temperature and pressure. The EtOH treated samexhibited a slight color change turning light transparent brown in appearance while thEtOH/scCO2 prepared samples turned completely opaque black in color. Both sets of hibited a shiny surface texture similar in appearance to pure PDM

PAGE 171

146 imageFigur 2e method ation of omer in vacuum for 1 hr. These weight changes (Figure 6.7) are attributed to the scCO2n, e s of untreated PDMSe and EtOH doped and EtOH/scCO 2 doped PPy/PDMSe samples are shown in Figure 6.6. e 6.6 Optical images of untreated PDMSe (A), and PPy/PDMSe IPNs prepared byEtOH (B) and EtOH/scCO (C) oxidizer insertion methods. This result shows that the EtOH/scCO 2 treatment is a much more effectivfor the doping (oxidizer insertion) used for the subsequent vapor phase polymerizPPy than EtOH solvent soaking techniques. The sample weight varied during processing of PPy/PDMSe IPNs on PDMSe films(18 mm X 18 mm X 2 mm) using 0.01M FeCl 3 /NaClO 4 and 2vol% EtOH in scCO 2 The weight increases with EtOH/scCO 2 oxidant insertion (24 hrs) and pyrrole monexposure (48 hrs). But then the weight decreases below the initial sample weight after being dried process extracting uncured resin and oils from the PDMSe samples. In additioresidual solvent and monomer present in the films were removed during the vacuum drying process. The PPy/PDMSe films were inhomogeneous visually. This is most likely an indication of the PPy distribution inside the PDMSe. ATR-FTIR analysis (Figure 6.18) of these films, which were also nonconductive, reveals pure PDMSe on thsurface identical to an untreated Silastic T2 PDMSe.

PAGE 172

147 50.052.0 Figure 6.7 Py/PDMSe Isopropyl alcohol (IPA) is a better solvent for PDMSe than EtOH and is still capable of solvating FeCl. The use of IPA was evaluated in an attempt to improve the dispersion and formation of PPy inside PDMSe films. Additional oxidizer was also tested. The IPA was used in conjunction with scCO and the results are compared to the EtOH/scCO system. Seven PDMSe samples (18 mm X 18 mm X 2 mm) were treated in both 2 vol% IPA/scCO and EtOH/scCO solutions containing 0.02 M FeCl and 0.02 M NaClO. Samples were allowed to soak in the cosolvent/scCO solutions for 24 hrs and then were soaked in pyrrole monomer vapor for 24 hrs. Sample weights were measured before oxidizer insertion (solvent soaking), between oxidizer insertion and pyrrole monomer exposure, and after pyrrole monomer exposure. Changes in PDMSe sample weight during EtOH/scCO 2 prepared P IPN systems. 6.4.2.3 IPA/scCO 2 prepared IPNs 3 2 2 2 2 3 4 2 40.042.044.046.048.0 Sample Weight (mg) Series1 Series2 Series3 Series4 Series5 Series6 Series7 Series8 Series9 Initial wt.After scCO2After pyrroleAfter vacuum

PAGE 173

148 4304404504604704804905005105201234567Sample #Weight (mg) Initial wt. After scCO2 After pyrrole Figure 6.8 Weight change data for IPA/scCO 2 prepared PPy/PDMSe prepared samples. 420e 540 440460480500520Wight (mg) Initial wt. After scCO2 After pyrrole 1234567Sample # Figure 6.9 Weight change data for EtOH/scCO 2 prepared PPy/PDMSe prepared samples. The samples prepared using the IPA/scCO 2 oxidizer insertion method resulted in a slight overall increase in sample weight while the IPA/scCO 2 preparation method resulted in a slight overall decrease in sample weight (Figures 6.8 and 6.9). The weight change is 2 preparation method resulted in a slight overall decrease in sample weight (Figures 6.8 and 6.9). The weight change is probably due to a com probably due to a com bination between the effects of oxidizer insertion and subsequent PPy formation and the extraction of soluble components from the PDMSe films during the scCO2 treatment. The effects of this are clearly visible in the optical images. Optical bination between the effects of oxidizer insertion and subsequent PPy formation and the extraction of soluble components from the PDMSe films during the scCO2 treatment. The effects of this are clearly visible in the optical images. Optical

PAGE 174

149 EtOH/scCO (A) and IPA/scCO (B) oxidizer insertion methods showing the images of IPA/scCO2 and EtOH/scCO2 prepared PPy/PDMSe samples are shown in Figures 6.10 6.13. Figure 6.10 Optical images (Leica G27) of PPy/PDMSe samples prepared by EtOH/scCO2 (A) and IPA/scCO2 (B) oxidizer insertion methods (mag. = 2X). Sample placed on top of printed text to show transparency. Figure 6.11 Optical images (Leica G27) of PPy/PDMSe batches prepared by 2 2 differences in homogeneity of PPy formation between samples.

PAGE 175

150 Figure 6.12 Optical images (Axioplan II) of PPy/PDMSe samples prepared by EtOH/scCO2 (A) and IPA/scCO2 (B) oxidizer insertion methods (magnification = 100X; scale bar = 200m). Figure6.13 Optical images (Axioplan II) of PPy/PDMSe samples prepared by EtOH/scCO2 (A) and IPA/scCO2 (B) oxidizer insertion methods (magnification = 200X; scale bar = 100m). From the optical images it is evident that the dispersion and PPy phase size rom the EtOH and IPA scCO produced fes prepared by EtOH/scCO2 have a relatively uniform particle size on the order of ~9-22 m. However the samples prepared by the IPA/scCO2 method have relatively large particles on the order of ~20-100 m with a much finer dispersed particle phase with particles on the order of 1-2 m. 2 treatments is different. The sampl

PAGE 176

151 This is believed to be mainly due to the improved solubility of IPA and the resulting FeCl3/IPA solution in the PDMSe material. Both sets of samples are visually dark black in color but have different degrees of opacity. Samples prepared by 2 have a splotchy non-uniform appearance while the IPA/scCO2 sampre uniform appearance when viewed with the naked eye. From the images above 0) the letters PPy are clearly visible in the EtOH/scCO2 samples while they isible in the IPA/scCO2 sample. The reason for this is apparent from Figures EtOH/scCOles have a mo(Figures 6.1are not vorm PPy phases (bla 6.13). While the samples uniformmicomsystemhe initial sample weitreatms to g ted an additional wtavg of -0.02 mg, for an overall wtavg of -0.86 6.12 and 6.13. The EtOH/scCO 2 samples have finely dispersed and locally unifck speck in the optical image, Figure 6.12 and prepared by IPA/scCO2 exhibit much larger PPy phases with a finely dispersed and secondary PPy phase which appears as the dark smoky regions in the optical crograph. In order to investigate the weight change effects of extraction of soluble ponents from the PDMSe films a series of tests were conducted on four sample s (n=7). The sample sets consisted of controls (no treatment) and samples exposed to scCO 2 scCO2 with 1vol% EtOH, and scCO 2 with 1 vol% IPA for 24 hrs. Tght was recorded and then the samples weights were rerecorded after the 24 hr ent and then again at 1hr and 18 hrs (air dry) after the 24 hr scCO 2 treatmentallow for complete offgassing of CO 2 and solvent evaporate from inside the samples. The weight changes between each step and overall weight changes were thencalculated. The control samples exhibited an average weight change (wt avg ) of -0.76 mafter the scCO 2 treatment. After 1hr they exhibited an additional wt avg of -0.09 mg and after 18 hrs they exhibi

PAGE 177

152 mg coavg nd 1 mg, l d ss (scCO2 treatment and 18 hrs degassing). 9% for wt hat the IPh mpared to the initial sample weights. For samples treated with scCO 2 the wt scCO 2 treatment, wt avg 1 hr, and wt avg 18 hrs were: -3.02 mg, -0.49 mg, and -0.34 mgThis resulted in an overall wt avg of -3.85 mg. For samples treated with scCO 2 avol% EtOH the wt avg scCO 2 wt avg 1 hr, and wt avg 18 hrs were: +12.59 mg, -17.57 mg, and -6.47 mg resulting in an overall wt avg of -11.45 mg. The samples treated with scCO 2 and 1 vol% IPA showed the largest initial swelling/weight change when compared to scCO 2 and EtOH/scCO 2 samples. Howeverthe overall changes in weight were not much higher than that for the EtOH/scCO 2 samples. The wt avg scCO 2 wt avg 1 hr, and wt avg 18 hrs were: +23.69 mg, -27.13 and -9.23 mg resulting in an overall wt avg of -12.67 mg. Due to differences in initiasample weight the % weight changes were calculated for the initial scCO 2 treatment anthe overall proce The %wt for just the scCO 2 process is -0.17%, -0.66%, +2.56%, and 5.1the controls, scCO 2 scCO 2 -EtOH, and scCO 2 -IPA respectively. While the overall %for the samples were -0.19%, -0.84%, -2.36%, -2.78% respectively. This data is displayed in Figures 6.14 and 6.15. Due to the fact that the samples prepared by the IPA/scCO 2 method had an average weight change of about twice that obtained for the EtOH/scCO 2 samples after the scCO 2 treatment, but only exhibited an extra -0.42% decrease in final sample weight compared to the EtOH/scCO 2 samples, it can be said t A/scCO 2 method is more efficient at swelling the PDMSe material but that botsystems extract about the same amount of soluble material from the PDMSe films or that both systems extracted out all the soluble material from the system.

PAGE 178

153 440490Aveim 450460470480500510g. wght (g) controls scCO2 410420430Initial 24hr scCO21hr 18hr Figure 6.14 Weight change of PDMSe samples exposed to different supercritical CO 2 treatments. EtOH/scCO2 IPA/scCO2 Figure 6.15 Average weight change in PDMSe samples exposed to different supercritical d into the PDMSe atrix during scCO2 processing. When this is accounted for in the previous IPA/scCO2 CO 2 treatments and after an 18 hr degassing period. From this it can be said that on average ~2-3% of the samples weight is extractedfrom the samples during scCO2 treatments with EtOH and IPA. This complicated the determination of how much oxidizer and polypyrrole are incorporate m -0.170.19-0.6684-2.5.192.56-4-2-113Avg. weight c --0.36-2.78-302% scCO2% totalhange ( 456%) controls scCO2 EtOH/scCO2 IPA/sc CO2

PAGE 179

154 and EtOH/scCO2 PPy/PDMSe samples it equates to ~4.1-5.1 wt% and ~1.7-2.7 wt% PPy for the IPA and EtOH samples. 6.4.3 Tetrahydrofuran Solution Doping As previously shown scCO2 treatment is an effective way to incorporate oxidant into PDMSe for the polymerization of pyrrole and other conducting polymers. However due to the limited sample compartment size only a limited number of samples (currently 7) can be prepared per cycle (24 hrs per cycle). A solvent soak system is more advantageous due to the ability to prepare large number of samples per cycle. However it is difficult to find a solvent that is a good solvent for both PDMSe (for swelling) and inorganic oxidizers (FeCl3, NaClO4, etc.). brigh In a paper by Zoppi and De Paoli,148 THF was found to be a good solvent for FeCl3. It was also shown to swell EPDM rubber (ethylene-propylene-5-ethylidene-2-norborene) used to form PPy/EPDM conductive blends. From their work, it is known that ~15-30% PPy is needed to produce conductive blends. The solubility parameter () for THF is 18.5 (MPA)1/2 while the for PDMSe is 14.9-15.59 (MPA)1/2. So from this information, THF should be a good choice as a solvent for the FeCl3 and the PDMSe system that we are currently using. An initial test was conducted with a saturated solution of FeCl3 in THF. PDMSe samples were soaked in a saturated solution for 41hrs. When removed, the samples were t orange in color and very swollen. The samples were then degassed in a hood for 7hrs at which time they were placed in a centrifuge tube filled with pyrrole vapor. The samples immediately (15 sec) started to turn black around the edges and slightly on the sample face.

PAGE 180

155 The resulting PPy/PDMSe samples were jet black in color and completely opaque when viewed with the naked eye or under the Zeiss microscope as previously done uld be taken for these samples. osed to for er and THF. These PPy/PDMSe samples appea if a urface. lips making sure the clips did not come near the waters surface. Both the therefore no transmission optical images co Another set of experiments was conducted with a 10wt% FeCl 3 solution in THF. The PDMSe samples were soaked in the FeCl 3 /THF solution for 24hrs and then dried under vacuum for 1hr to remove the residual THF. The samples turned cloudy opaque yellow in color from a clear transparent color. Once dried the samples were exppyrrole monomer vapor for 96 hrs to ensure full PPy formation for the given FeCl 3 doping. After pyrrole monomer exposure the samples were again dried under vacuum2 hrs to remove any residual pyrrole monom red to have no bulk conductivity when tested with an Omega HHM57 digital multimeter. Two samples each of IPA/scCO 2 and THF/scCO 2 were placed in a 50 ml PP centrifuge tube containing 15 ml Instant Ocean artificial sea water (I.O.) to determinesample hydration influenced the conductivity of the samples two THF prepared. The THF prepared samples had a rough flat black surface while the IPA/scCO 2 samples had smooth shiny black s The samples were allowed to soak in I.O. for ~64 hrs to fully hydrate prior to electrochemical analysis. The THF prepared samples turned the I.O. light brown in color. This is probably due to residual Fe(III) and Fe(II) leaching from the samples. However the IPA/scCO 2 samples produced no color change in the I.O. salt solution. The samples were patted dry with Kimwipes and then suspended in the solution from alligator c

PAGE 181

156 untreaing ammetry exper and IPA/scCO2 techniques along with untreated PDMSe. Cyclic voltammetry artificial sea water. Visual inspection of the samples after the cyclic voltammetry experiment showed that the THF PPy/PDMSe sample had increased in size by ~10% during the 64 hr I.O. soak when compared to the other samples. This increase in volume suggests that the PPy/PDMSe samples are uptaking water and swelling during the I.O. soak. 6.4.3.1 Effects of hydration of sample conductivity The effects of sample hydration were investigated by conducting cyclic voltammetry experiments in I.O. on dry PPy/PDMSe samples prepared by the THF method. Cyclic voltammetry was conducted at .0 V with a rate of 10 mV/sec vs. ted PDMSe and the IPA/scCO 2 prepared PPy/PDMSe samples exhibited no electrochemical response. However the PPy/PDMSe sample prepared by THF soakshowed a low electrochemical response (0.27 mA/cm 2 ) during the cyclic volt iment. The electrochemical response of the PDMSe and two PPy/PDMSe samplesis shown in Figure 6.16. -0.050.000.30-0.8-0.6-0.4-0.200.20.40.60.81 0.050.100.150.200.25-1I (mA/cm2) THF PPy/PDMSe IPA PPy/PDMSe PDMSe E(V) versus Ag/AgCl Figure 6.16 Current response of PPy/PDMSe samples prepared by THF solvent soaking conducted at .8 V at 10 mV/sec versus Ag/AgCl in Instant Ocean

PAGE 182

157 Ag/AgCl. A dry sample was suspended in the I.O. solution and cycled five times; experiment lasted approximately 34 minutes. After the first five cycles the samples wasallowed to soak for ~90 minutes in the I.O. solution and then the experiment was repeated for another 20 cycles resulting in a total of 25 cycles being run on the sample and a total I.O. immersio the n time of approximately 4.5 hrs. The current draw at the potential of +0.8 V was 3.6 nA during the first cycle which increased to 78.1 nA for the fifth cycle. During the second half of this experiment the peak current for the 6th cycle increased to 0.145 mA (145 A). This equates to the samples being in the I.O. salt solution for a total of 120 minutes. The final cycle (25th cycle) had a peak current of 0.258 mA; this equates to 253 minute soak time in the salt solution. This is a significant increase of ~71667% in current draw over the entire experiment, which equates to a decrease in resistance from 278 M to 4 K at a 1.0 V potential (Figure 6.17). Figure 6.17 Effects of sample hydration on cyclic voltammetry for wet and dry PPy/PDMSe samples. E(V) versus Ag/AgCl -0.8-0.5-0.30.30.50.8-1.00.0 1.0 I(mAcm2 /) -0.100.100.250.30 Dry sample 1st 5 cycles Dry samples n ext 20 cycles Wet sample (3 day soak in I.O.) 25th cycle = 253 m in. 0.000.050.150.20 6th cycle = 120 min. in Instant Ocean 1st five cycles in Instant Ocean -0.05

PAGE 183

158 6.4.3.2 Effects of PDMSe surface segregation An interesting trait noticed for all the THF prepared PPy/PDMSe composites is they exhibited bulk conductivity and swelled when soaked in I.O.. Samples prepaother methods (EtOH, IPA, scCO red by ell flat o be due to sample handling in between the oxidizer insertion and the pyrrole mono, placed flat, MSe with Kimw surface layer is removed and a pure PDMSe surface is detected by ATR-FTIR. 2 soaks) exhibited no bulk conductivity and do not swin the I.O. solution. The samples that were conductive and swelled all have a roughblack surface while nonconductive samples all have a flat shiny black surface. This was determined t mer exposure steps. Samples were non-conductive if the surface was rinsedon Kimwipe to dry, or manipulated in any way. The nonconductive surfaces were shiny and black similar to the surfaces produced from the previous scCO 2 treatments. Due to the fact the PDMSe rearranges its surface depending on the environment. Itis believed that the formation of true PPy/PDMSe IPN systems is hindered on the PDsurface. This trend was discussed earlier in the studies on the scCO 2 systems by ATR/FTIR analysis. It is believed that the same thing happens in the THF systems. ATR-FTIR analysis of PPy/PDMSe samples prepared by EtOH, IPA, scCO2, and THF treatments was conducted using a Nicolet 20SXB FTIR. There is no detectable difference in the ATR spectra of samples prepared by EtOH, IPA and scCO2 methods (Figure 6.18). However THF prepared samples have a characteristic peak for PPy at ~1540 cm-1.149-152 This peak is characteristic of the pyrrole ring vibrations (C-C and C=C ring vibrations; 1500-1600 cm-1). The rest of the characteristic peaks for PPy are confounded with the peaks associated with PDMSe. When the THF prepared PPy/PDMSe (same sample of previous) sample surface is cleaned with light whipping ipe and EtOH the PPy

PAGE 184

159 Figursolution soaking procedures. Inset box shows location of characteristic t is ace e 6.18 ATR-FTIR analysis of PPy/PDMSe IPN systems prepared by various pyrrole ring vibrations (1500-1600 cm -1 ). This could explain batch inconsistency with respect to conductivity. Handling of the samples between the oxidizer insertion and pyrrole monomer exposure potentiallyremoves or disrupts the deposited FeCl 3 layer which would heighten this effect. It is also believed that the conductive samples have a PPy surface layer thasitting directly on top of a predominately pure PDMSe layer that and it is not truly incorporated into the PDMSe sample. This is supported by the observation that the PPy/PDMSe samples mark easily, i.e., PPy flacks off when the sample surface is rubbed. The conductivity of the samples is easily destroyed by minor surface abrasion. Removing or disrupting the FeCl 3 deposited on or near the PDMSe sample surfprevents formation of a uniform, continuous PPy surface layer. This ultimately makes the surface nonconductive.

PAGE 185

160 The effects of PDMSe surface segregation was evaluated using cross sectional EDS mapping on FeCl3/THF doped PDMSe samples. The EDS spectrum for the FeCl3 doped PDMSe is shown in Figure 6.19. EDS mapping was conducted for silicon, chlorine, and iron (Figure 6.20). From the EDS mapping for iron and chlorine (FeCl3), it can be shown that the doped PDMSe samples have a ~60-100 m surface layer relatively devoid of FeCl3 (Figure 6.21). Chlorine was selected to illustrate the surface segregation of PDMSe due to its higher EDS signal intensity compared to iron. Figure 6.19 EDS spectra of cross sectioned FeCl3/THF doped PDMSe. 3be mumple FeCl is still present in this surface layer however its concentration is appears ro ch lower than in the bulk of the sample. Subsequent exposure of this sample to pyrrole monomer vapor would result in the formation of a PPy/PDMSe IPN. However; the majority of the PPy would be buried in the bulk of the samples and not on the sasurface thus resulting in a non-conductive sample due to a predominantly pure PDMSe surface.

PAGE 186

161 Figure 6.20 EDS magnification = 70X). Figure 6.21 EDS chlorine mapping of cross sectioned FeCl3/THF doped PDMSe showing a ~60-100 m surface layer of predominantly pure PDMSe (mag. =70X). apping of cross sectioned FeCl3/THF doped PDMSe (m Silastic T2 PDMSe doped w/ FeCl3/THF 97 m 62 m Silastic T2 PDMSe doped w/ FeCl3/THF Silastic T2 PDMSe doped w/ FeCl3/THF 97 m 62 m

PAGE 187

162 FeCleak the in%/hr). After this point the strain at break slowly decreases form the 156% for the 6 hr soak to about 90% for the 20 hr soak (rate = ~4 %/hr). The modulus data, shown in Figure 6.23, shows an initial decrease in modulus from the unmodified PDMSe (1.22 MPa low strain and 3.35 MPa high strain) to the 1 and 2 hr treatment times (~0.91 MPa low, ~2.57 MPa high). After this the modulus increased and stabilizes for the 3-6 hr treatments (~1.47 MPa low, ~3.65 MPa high) after this the modulus again decreases and stabilizes for the 12, 16, and 20 hr treatments (~0.67 MPa low, ~2.09 MPa high). This trend is evident in both the low and high strain modulus values. The initial increase then stabilization in modulus up to the 6 hr mark is most likely due to the extraction to uncured oligomers, oils, and other soluble components. After this point the decrease in modulus is most likely due to the possible degradation of 6.4.3.3 Effects of FeCl3/THF doping on PDMSe mechanical properties The effects of the FeCl3/THF doping process on the mechanical properties of PDMSe samples were investigated using tensile testing methods. The tensile strength, stain at break, and modulus of PDMSe dog bones samples was evaluated as a function of 3 solution exposure time. As can be seen from Figure 6.22 there is an overall decrease in the strain at break and peak stress (tensile strength) as the FeCl3 soak time is increased. The strain at brfor untreated PDMSe is about 230% with a peak stress of 5.96 MPa. The decrease in peak stress occurs in almost a linear fashion. The values for peak stress decrease from itial point (untreated) of 5.96 MPa to 1.44 MPa for the 20 hr soak. In contrast the decrease in the strain at break seems to occur in two different stages due to a change in the degradation rate between the 5 and 6 hr point. The strain at break rapidly decreases from 230% for the untreated sample to the about 130% for the 5 hr soak time (rate = ~20

PAGE 188

163 the PDMSe. The extraction of unreacted oligomers, oils, and other soluble components is also evident from the weight loss data produced for the scCO2, EtOH/scCO2, and IPA/scCO2 treatments on PDMSe. The samples run at the 24 hr treatment times were too brittle to mount and run tensile tests on. The PDMSe samples change form a clear elastic material to an opaque yellow/orange material that tends to fracture if not handled correctly after the 24 hr exposure to the oxidizer solution. However; after exposure to the pyrrole monomer vapor for 24 hrs the samples regain their elastic nature and durability. The decreases in mechanical properties suggest that the Silastic T2 PDMSe used in these exo nd other solubFigure 6.22 Strain at break and peak stress data for the degradation of Silastic T2 periments is not a suitable choice for prolonged exposure to oxidizers due t oxidative degradation and extraction of unreacted silicone oligomers, oils, a le components. PDMSe with prolonged exposed to a 5 wt% FeCl 3 /THF solution. 0.00100.00150.00200.000510152025Strain @ Brea0.002.003.004.00Petress (MPa 50.00250.00FeCl3/THF Soak (Hrs)k (1.005.006.00ak S) 300.00350.00%)7.00 Strain @ Break (%) Peak Stress (MPa)

PAGE 189

164 4.50 0.001.002.002.503.500510152025FeCl/THF Soak Time (Hrs)Modul (MP Low Strain High Strain 0.501.503.004.003usa) Figure 6.23 High and low strain modulus data for the degradation of Silastic T2 PDMwith prolonged exposed to a 5 wt% FeCl Se It was also shown that by changing the solvent sycCO2. ples have really low conductivities ( resistivety > 300 M) which could then be increased with exposure to an electrolyte solution. 3 /THF solution. 6.5 Conclusions From this set of experiments it was shown that it is possible to form dynamic PPy/PDMSe composite structures which are capable of producing a 21 change in contact angle when switched between 1.0 V. stem that the overall loading/formation of PPy inside PDMSe can be increased thus improving the overall properties of the system. Supercritical CO 2 cosolvent solutions were shown to be effective in doping PDMSe samples when compared to plain ethanol systems. In particular isopropyl alcohol/scCO 2 mixtures were shown to increase the PPy loading in PDMSe to 4-5 wt% compared to 2-3 wt% for ethanol/s It was later determined that THF was a better solvent medium for the FeCl 3 doping of PDMSe than either of the scCO 2 methods. THF was used to form conductive PPy/PDMSe composite systems with conductivities in the range of 278 M to 4 K depending on sample hydration. Dry sam

PAGE 190

165 However it was determined that the PPy/PDMSe surface layers are unstable due to The dical roperties of PDMSe were also investigated. It was found that the FeCl3/THF soaking process seriously decreased the tensile strength (peak stress) and strain at break of the that was initially investigated and needs further work is the formation and use of vinyl and/or silanol functional pyrrole monomers. By incorporating pyrrole functional groups PDMSe surface segregation. From EDS mapping and FTIR/ATR measurement techniques it was shown that the Silastic T2 PDMSe tends to rearrange its surface prior to PPy formation forming a predominantly pure PDMSe surface layer. From the EDS mapping for the chlorine and iron atoms from the FeCl3 it was shown that the PDMSe sample formed a 60-100 m surface layer that was much lower in concentration of FeCl3 than the bulk of samples. Exposure of this sample to pyrrole vapor would result in the formation of a PPy/PDMSe composite system with the majority of the PPy being buried in the bulk of the material resulting in a non-conductive sample. egradative effects of the FeCl 3 /THF doping process on the mechan p PDMSe samples. As the oxidizer doping process was carried out the tensile strength and strain at break decreased from 5.96 MPa and 230% to 1.44 MPa and 90%, respectively. This was also evident in both the low and high strain moduli. The untreated modulus values were 1.22 and 3.35 MPa at the low and high strain values where after the 20 hr treatment the modulus values were 0.66 and 2.04 MPa respectively. After 24 hrs the samples were too brittle to handle and could not be run. From this; it is obvious that improved techniques for the formation of PPy/PDMSe IPN systems are needed. It is believed that by chemically binding the PPy into the PDMSe network that the surface segregation of the PDMSe can be overcome. One route

PAGE 191

166 in the PDMSe network or by incorporating silanol or vinyl functional groups into the PPy struct ure further PPy/PDMSe formation would be chemically bound together thus restricting there movement and segregation.

PAGE 192

CHAPTER 7 TP AND TPE BASED DYNAMIC, NON-TOXIC, ANTI-FOULING SURFACE COATINGS 7.1 INTRODUCTION Polydimethylsiloxane elastomer (PDMSe) was previously studied as a base material for the formation of PPy/PDMSe IPN systems due to its low surface energy anmodulus. However due to the low modulus of PDMSe the film durability becomes an issue. Due to this concern, efforts have been put forth to d increase the durability of PDMlastomer Systems. The systems currently being studied are Santoprene 8211-65, 8281-65, and 271-55. Santoprene is a PP/EPDM (ethylene propylene diene monomer) based TPE material. EPDM is mechanically melt phase mixed into PP and is then lightly crosslinked to form dispersed elastomer phases in the PP matrix. Santoprene 271-55 is a FDA approved food grade TPE with physical properties comparable to that of Silastic Se coatings as well as investigating the use of other low modulus rubber type materials for foul release coatings. Another alternative to PDMSe is the thermoplastic elastomer (TPE) class of materials. These materials are generally characterized as having elastomer phases dispersed in thermoplastic materials such as polypropylene. They impart a relatively low modulus with increased durability (tear strength, hardness, and modulus) due to the thermoplastic phase of the material. PPy IPNs systems have been formed in three different TPEs manufactured by Advanced E 167

PAGE 193

168 T2 PDMSe, but with a higher elongation to break (330% compared to 250% for Silastic T2). Santoprene 8211-65 and 8281-65 are both from a newer series of Santoprene TPEhardness, and tear and tensile strallowing for a broader range of uses. E) sily njection and blow molding equipment. The only differe eries tics tic (TP) material with excelgth, s that have better oxidative stability and physical properties (elongation at break, engths) than 271-55 thus Santoprene 8211-65 is a general purpose grade thermoplastic elastomer (TPthat is formulated to replace thermoset elastomers such as chloroprene and EPDM. 8281-65 is a FDA approved biocompatible TPE. This material is typically used in short termblood contact medical conditions such as in catheters and septum. Both 8211-65 and 8281-65 are non-hygroscopic, shear sensitive and can be eaprocessed using standard extrusion, i ence between 8211-65 and 8281-65 is that more stringent control is used in thmanufacturing processing of 8281-65 to comply with FDA regulations. The 8000 smaterials have a higher durometer hardness (65 vs. 47), tear strength (150 psi vs. 130 psi), and elongation at break (530% vs. 250%) than Silastic T2 PDMSe. Other prospective choices for a substrate material are engineering thermoplassuch as Udel polysulfone. Polysulfone is a very hard thermoplas lent oxidative stability (thermal and hydrolytic (hot water and steam) stability, inorganic acid and base resistance) and physical properties (modulus, tensile strenand elongation at break) comparable to polycarbonate. Udel polysulfone is approved for use in food, water, and medical applications. Typical applications include membrane (water purification, gas separation), medical

PAGE 194

169 (surgical trays, nebulizers, and humidifiers), food services (microwave cookware, etc.), and plu mbing (hot water fixtures and fittings). ldrich), iron (III) cer r g/cm3, salinity = ~28-34 ppt). 7.2.1.McMaster-Carr in the form of a 0.03 thick sheet. Samples were cut to size (18 mm x 18 mm) and then used as received. PPy IPNs have been formed with polysulfone resulting in an extremely hard anddurable surface coating that is resistant to surface abrasion and damage. An added advantage is that it can be applied from solution, thus allowing for more conventional application techniques. Another benefit of polysulfone is that it can be functionalized with various substituents (carboxylic acid, sulfonic acid, etc.) allowing for easy manipulation of the surface and bulk properties of the material. This will in turn produceIPN systems with easily modifiable properties. 7.2 MATERIALS AND METHODS 7.2.1 Materials 7.2.1.1 General Chemicals Pyrrole (Sigma-Aldrich) was filtered, before use, through a neutral alumina (Brockman activity 1; Fisher Scientific) column until colorless to remove impurities. Lithium perchlorate (LiClO 4 ) and sodium perchlorate (NaClO 4 ) (Sigma-A hloride (FeCl 3 ) (Fisher Scientific), tetrahydrofuran (Fisher Scientific), and allyltriethoxy silane (Gelest) were used as received unless otherwise stated. Deionized (ultrapure) water was produced using an 18 M Millipore system. Artificial sea watwas made using Instant Ocean artificial sea salt (Fisher Scientific) and deionized wate(1 ml I.O. in 30.7 ml H 2 O; specific gravity = 1.020-1.023 2 PPy/Santoprene Sample Preparation and Mounting Advanced Elastomer Systems (AES) Santoprene 271-55 was purchased from

PAGE 195

170 Santoprene 8211-65 was obtained directly from Advanced Elastomer Systems (AES) in the form of 4.38 x 4.38 x 0.06 plaques. Plaques were prepared by injection moldie 04C nd s are removed (~10-15 min.). The resin is then applied and cured pling Agent Preparation for the preparation of the allyltriethoxy silane (ATS) coupl1.58 2-3 281-65 samples. treatment the samples were placed under vacuum for 1 ng on a 170 Ton Vandoren injection molding machine to a thickness of 0.08. Thplaques were then compression molded down to a thickness of 0.06 by heating at 2for 5 minutes followed by a 10 minute cool down period before removal from the mold7.2.1.3 Polydimethylsiloxane Elastomer Preparation PDMSe was prepared using Silastic T2 resin (Dow Corning). The resin acuring agent are thoroughly mixed (~10 min.) in a 10:1 weight ratio. The resin is then degassed under vacuum until most of the bubble at room 50C for 5 hrs. 7.2.1.4 ATS Cou The general procedure ing is as follows. In a 50 ml polypropylene cup, add 30 ml EtOH (95%) and ml H 2 O and stir with magnetic stir bar. Adjust pH of the solution to 4.5-5.5 by addingdrop-wise glacial acetic acid (~2 drops). After the pH is adjusted add 0.17 ml ATS andlet mix for 5 minutes. Apply ATS solution to cleaned glass slide and let react forminutes. Rinse ATS treated glass slides with EtOH and place in 120C oven for 10 minutes. After this you can apply the uncured PDMSe as usual. 7.2.2 Sample Preparation Methods The oxidizer insertion method was used to form PPy/TPE IPN systems. The general procedure for this process is described as follows using 8 Santoprene 8281-65 film was cut into 18 mm x 18 mm squares to produce samples. Nine of these samples were placed in a 5 wt% FeCl 2 /THF solution and allowed to soak for 24 hrs. After the 24 hr

PAGE 196

171 hr to he r 7.2.2. Se. nd flame treatment. After cleaning lead wiresnners a o wire) is then cleaned sing a radio frequency discharge argon plasma treatmrom so protec dry (remove residual THF). The dried 8281-65 samples appeared to shrink after tdrying process. They were initially swollen when removed from the THF solution; prioto drying. The samples were then placed directly in a sealed PP centrifuge tube containing 2 ml of pyrrole monomer for 24 hrs to facilitate the formation of PPy. After the pyrrole monomer treatment the samples were again dried under vacuum to remove any residual pyrrole monomer. 1 Sample Mounting In order to conduct contact angle measurements samples were produced using thestandard FeCl 3 /THF method mentioned above. Samples were then mounted to glass slides using Ag epoxy and encapsulated with PDM Glass slides were cleaned with ethanol a were super glued to the glass slides. Sample mounting pads and lead wire ruwere painted onto the glass slide using Ag-filled epoxy. The Ag-filled epoxy was then cured for 30 minutes at 175C. The bottom sides of the samples were then coated with thin layer of Ag-filled epoxy and placed on top of the mounting pads. The samples werethen sandwiched between two glass plates and placed in a 175C oven for 30 minutes tcure. The substrate surface (glass slide, Ag-filled epoxy, and lead and oxidized prior to ATS treatment u ent (Ar plasma) for 1 minute at ~50 mTorr. The sample surface is protected fthe argon plasma using a layer of double sided sticky tape. This protective layer al ts the sample surface during the preceding ATS and PDMSe backfilling process.The substrate is then treated with an ATS coupling agent to improve the adhesion of the PDMSe encapsulant to the glass substrate. A top cover plate (PET covered glass

PAGE 197

172 slide) was applied to the sample and the edges of the glass slides were sealed with scotch tape (create a small mold). Samples were then encapsulated by pumping uncured PDMSe resin in bettwo glass plates with a syringe pump. A 30 cc syringe and an 18 gauge needle were usedfor this process. The flow rate of the syringe pump was varied throughout the backfillinprocess due to the increasing in v ween the g iscosity of the PDMSe as it started to cure at room as then cured in an oven at 50C for 5hrs. samples for contact angle measurement and marine testing. Samples were mounted this way in single and double element configurations. Single element samples were used for contact angle measurements and require the use of an external stainless steel counter electrode. However for marine testing a dual element design was developed. In the dual element design both of PPy/TPE elements alternate roles as the working and counter electrodes. Therefore the need for a separate counter electrode is eliminated. The layouts for the single and dual element designs are shown in Figure 7.2. temperature. The PDMSe w After this the top cover plate and protective tape layer are removed and the sample is ready to use. A general schematic of this process is depicted in Figure 7.1. Figure 7.1 Preparation scheme for the formation of PDMSe encapsulated PPy composite

PAGE 198

173 Figure 7.2 PDMSe encapsulated PPy composite samples for testing in aqueous/marine environments using a single (A) and dual (B) element designs. 7.2.2.6 Micropatterning Hot embossing methods were employed to micropattern Santoprene TPEs. Patterned silicon wafers, patterned with the negative of the sharklet pattern, were glued to standard glass slides (pattern side up). TPE films were then placed in between the glass elastomer. ad to be removed carefu re prepared containing 1, 5, 10, and 20 wt% FeCl3 and 5 wt% PSU. Both the FeCl3 and PSU were pre-dissolved in separate THF solutions and then mixed to form solutions of the above concentrations. PSU/FeCl3 solutions in THF slides and the patterned silicon wafers. The pieces were held together using standard binder clips to form molds. The molds were placed in an oven at 185C and 200C for 35 minutes. The molds were removed from the oven and allowed to cool before extracting the The TPE tended to stick to the patterned silicon wafer and h lly in order to prevent damage to the pattern. This was accomplished by slowlypulling the elastomer off of the patterned silicon wafer. The patterned surface was then imaged using optical and scanning electron microscopy. 7.2.2.7 TP Film Formation by Spin Casting and Spraying FeCl 3 /PSU films were formed using spin casting and spraying techniques. These films where then used to form PPy/PSU composite structures. FeCl 3 /PSU solutions we

PAGE 199

174 were then spin cast onto cleaned glass microscope slides for ~ 30 seconds at 1000, 1400 and 2000 rpm. The PSU/FeCl3 solution was placed drop wise onto the slide surfacethe surface appeared to be completely coved. During the spin casting process the THF is rapidly evaporated from the sample until coated slider for 24 uum for 1hr to remove any residual pyrrole monomer. The films turned opaque black rapidly upon exposure to the pyrrole monomer vapor. 7.2.2.8 Conductivity Determination Sample conductivity was checked either by a hand held digital multimeter (Omega eter were taken diagonally from corner to corner on both the front and back sides ople is then calculated from the following equation: therefore no vacuum drying was required after the spin casting process. The FeCl 3 /PSUes were placed in a sealed PP tube containing ~2 ml pyrrole monom hrs. The films were then dried under vac HHM57) or on a four point (4-pt) conductivity probe. Conductivety measurements taken by digital multim f the sample. This results in a total of 4 different measurements taken for each sample. Four point conductivity probe measurements were taken using a Keithley 181 nanovolt meter and a Keithley 224 programmable current source. The 4-pt conductivity probe applies a current between the outer two probes and the resultant voltage is measured from the inner two probes. The conductivity () of the sam ln2I V (7.where I is the current applied (amps), V is the measured voltage (volts), and is the sample thickness (cm). 1)

PAGE 200

175 The voltage was measured at 5 different currents; 10, 20, 30, 40, and 50 A. These measurements were taken at three different spots on the sample surface and then repeated on the opposite side of the samples (N =6). From this the conductivity of the samps ment within which the samples are expected to operate. or PPy/8281-65 dual element sample in Instant s a Ag/AgCl reference electrode. .0, and Ag/AgCl counter electrode was used for all electrochemical experiments. Three captive air bubbles (repeated twice) were collected at each potential. The potentials were chosen in a random order to produce 6 e (-e 7.1). This value then dropped to 38 and 30 for the Average383041StDev223 les was determined using equation 7.1. 7.3 POLYPYRROLE/8281-65 IPN SYSTEMS 7.3.1 PPy/Santoprene 8281-65 Contact Angle Measurements The dynamic character of the PPy/8281-65 samples contact angle measurementwere conducted in Instant Ocean artificial sea water. This approximates the environ Table 7.1 Contact angle values obtain fOcean artificial sea water versu Dual element PPy/8281-65 samples were preconditioned by cyclic voltammetry prior to testing. The samples were cycled from .5 V at 10 mV/sec for 10 cyclesContact angle values (captive air bubble technique) were then collected at +0.5, 00.5 V from one of the element surfaces. A 0.0 V+0.5 V-0.5 V38314435334538273936N/A43422937372838 measurements at each potential. The average contact angle value for the reduced stat0.5 V) was 41 3 (Tabl

PAGE 201

176 oxidizef re 7.3. Figure 7.3 Change in contact angle for a mple in Instant Ocean artiftrode. ation process on 8281-65 a more detailed weight and volume change experiment was conducted. This was done to determine the amount of FeCl and PPy deposited in the Santoprene base material. Two samples sets (N=7) were used; one being pure THF soak (control) and the other a 5 wt% FeCl/THF soak. The average initial sample weights and volumes were 399.6 mg, 461.94 mm3 and 401.0 mg, 463.53 mm3 for the THF and FeCl/THF samples respectively. After the 24 hr solution treatment the average sample weights and volumes were 224.3 mg, 262.16 mm3 and 235.1 mg, 268.89 mm3for the THF and FeCl/THF samples respectively. This d states of 0.0 V and +0.5 V, respectively. This resulted in an 11 change in contact angle for a 1 V difference in potential. This is comparable to the 21 change produced from a 2 V difference in potential for the PPy/PDMSe samples. The results othis experiment are shown if table 7.1 and Figu 253045-0.6-0.4-0.200.20.40.6E(V) versus Ag/AgClont A (deg) 3540actngle C PPy/8281-65 dual element saicial sea water versus a Ag/AgCl reference elec7.3.2 Determination of PPy Content in PPy/8281-65 Samples In order to investigate the effects of the PPy form 3 3 3 3

PAGE 202

177 equates to a 44%, 43% and 41%, 42% change in the sample weights and volumes aftthe treatment process. The THF samples lost an extra 3% of weight and 1% of volumduring the solvent soak process compared to the FeCl er e sample weights and volumes were ms e whole process of 40-43% and 38-40% respectively. This weight chang 3 /THF samples. After the pyrrole monomer exposure the average 227.7 mg, 277.19 mm3 and 240.1 mg, 284.70 mm3 for the THF and FeCl3/THF samples respectively. This equates to a 1.53%, 5.73% and 2.13%, 5.88% increase in sample weight and volume for the THF and FeCl3 samples during the polymerization step. From this set of experiments it was not possible to determine the amount of FeCl3 and PPy deposited in the system. It can be said that there is not much of a difference between the control and experimental samples during the PPy formation process. This is most likely due to the samples not being thoroughly dried after the solution and monomer soak steps. Table 7.2 Weight and volume change data for the formation of PPy/8281-65 IPN systeusing the FeCl3/THF solvent soak process. V (THF) V (FeCl 3 ) V Wt (THF) Wt (Fe Cl 3 ) Wt PPy=15.0315.810.78(mm 3 )0.00340.0 However there is a 41-43% decrease in both the sample weight and volume from the 24 hr exposure to the THF based solutions and an overall change in sample weight and volume over th 0500.0016(gm)PP% Total =-39.99-38.581.41(%)-43.0053-40.11692.8884(%) y (%)=5.735.880.15(%)1.53492.13280.5979(%)Total =-184.74-178.835.91(mm3)-0.1718-0.16090.0110(gm) e is due to soluble components being extracted from the system. This is likely dueto the extraction of extender oils and thus decreases the elastic properties of the system. The overall differences in the PPy formation process between the THF and FeCl 3 /THF soaked samples is shown in table 7.2.

PAGE 203

178 7.4 POLYPYRROLE/271-55 IPN SYSTEMS PPy/271-55 samples were prepared in the same manner as previou sly described for PPy/8es 71-55 ba AFM analysis section. hrs and not the FeCl3/THF oxidizer solution. The other two samples sets were FeCl/THF systems and were exposed to a 5 wt% FeCl/THF solution for 24 hrs. The initial sample weights and volumes were recorded and then the samples were placed in their appropriate solutions and allowed to soak for 24 hrs on a rotator. The average initial sample weights and volumes for all the sample sets were 248.9 mg .2 and 247.8 mm 9.2 respectively. 281-65 and PPy/PDMSe formation. Contact angle measurements were conducted on PPy/271-55 samples in Instant Ocean artificial sea water. These measurements turned out to be inconclusive resulting in very erratic valuand bubble shapes. This is believed to be due random surface defects present in the 2 se material as obtained from the manufacturer. The previous PDMSe and 8281-65 samples were cured against smooth glass andmelt pressed against polished aluminum resulting a initially smooth sample surface. Surface defects in Santoprene TPE materials used as received are shown later in the Hysitron and 7.4.1 Determination of PPy Content in PPy/271-55 Samples In order to determine the amount of PPy formed during the doping and polymerization process weight and volume change experiments were conducted on 3 sets of 9 samples. The first sample set was a control set that was just exposed to THF for 24 333

PAGE 204

179 A fter the 24 hr solvent soak the control set lost 38% of the sample weight and 36% of sample volume; wthe sample weight and 3respend PPy deposited in the samples. However for ed samples maintained abouther the tem or to ta rom off white to dark black, and all the samples were conductive. This data is shown in Figures 7.4 and 7.5. hile the FeCl 3 treated samples lost about 31% of 3% of the sample volume. After the pyrrole monomer vapor exposure the control set lost an additional 3% of its weight and 10% of its volume for a total weight and volume loss of 41% and 46% ctively. The FeCl 3 treated samples also lost an additional 3% of their weight but only lost 5% of their volume for a total weight and volume loss of 34% and 38% respectively. Due to the continued weight and volume loss after the pyrrole monomer vapor exposure it can be said that there was still residual THF in the samples that was not completely removed during the vacuum drying process. This complicated the determination of the amount of FeCl 3 a both the solvent soak and monomer soak process the FeCl 3 treat 7% more of their initial weight compared to the control set. Even though all the sample set started out with roughly the same sample weights and volumes the FeCl 3 treated samples average weight and volume were 7.8% and 8.7% higher than the control set after the PPy formation. This can be attributed to eitdeposition of FeCl 3 inside the 271-55 sample or to the formation of the PPy IPN sys both of these processes. It is difficult to separate the two due to the FeCl 3 and PPy weight and volume dabeing confounded with the THF solvent evaporation. The presence of PPy is evident inthe systems due to the color change f

PAGE 205

180 100200260300UnmodifiedSolvent SoakMonomer Soakght (mg) 240280 THF Soak FeCl3/THF Soak FeCl3/THF Soak 120140160180220Wei Figurxchange on PPy/271-55 Conductivity nce Cl-. e 7.4 Weight loss data for the formation of PPy/271-55 IPN systems. 100160220280Voe3) 180200240260lum (mm THF Soak FeCl3/THF Soak FeCl3/THF Soak 120140UnmodifiedSolvent SoakMonomer Soak Figure 7.5 Volume loss data for the formation of PPy/271-55 IPN systems. 7.4.2 Effects of Counter Ion E All PPy IPN samples have been made up to this point using FeCl 3 as the oxidizer. This results in Cl being inserted into the PPy matrix as the counter ion. From the previous studies with the conducting polymer actuators we saw an improved performaof the actuators when using the ClO 4 counter ion compared to

PAGE 206

181 An attempt was made to make PPy using Fe(ClO4)3 instead of the typical FeCl3. This would result in the perchlorate ion being used as the counter ion. However when a solution of Fe(ClO4)3 in THF was made the Fe(ClO4)3 (crystalline) sank to the bottom of the flask and started to dissolve. However after about a minute the whole solution started to turn cloudy white. After letting the mixture stir for ~30 minutes the mixture was still cloudy white in appearance and small white crystals were visible. After stopping the stirring and letting the mixture settle a large amount of white crystals precipitated out of solution and the solution turned a transparent brown color. The amount of crystals formed during this reaction was in great exces s of the mples were ductivities were testedgain 64hrs later and amount of Fe(ClO4)3 initially added to the THF. It is believed that Fe(ClO4)3 is able to ring-open polymerize the THF and form polybutylene glycol (poly(butylene oxide)) or various oligomers.153, 154 In order to incorporate the ClO4into the PPy structure a solvent exchange process was performed. PPy/271-55 samples (N=9) were prepared using the FeCl3/THF preparation method. After the final drying process the samples were placed in a 3wt% NaClO4 solution in THF for 24hrs. The 3 wt% NaClO4/THF solution was clear with no visible signs of crystal formation or other side reactions between the NaClO 4 and THF. After the sa soaked for 24hrs they were dried under vacuum for 1hr and then the con on three random samples and compared to standard PPy/271-55 samples that were prepared at the same time. The resistivity of the samples was measured at 16 k 5 and 18 k 17 for the ClO 4 and Cl samples. Three more random samples were measured a

PAGE 207

182 were stems were run under real enviroent ill n (sea water = ~2.72% NaCl, ~19,500 ppm Cl~10,770 ppm a+, 67asured cm); which results in a bulk sample conductivity of 9.57er gher due to its thinner nature. Calculations were done assumi measured at 38 k 12and 17 k 13 respectively. There is no significant difference between these values so it appears that there is no real difference between these two sample sets. It is unclear if there is a difference in conductivity between the counter ions used or if the counter ion exchange even took place. One way to remedy this would be to run the experiment in an aqueous NaClO 4 solution and apply a cyclic potential to the samples. This would force the ion exchange process. However the same thing would happen when these sy nmental conditions. When these samples are submerged in a marine environmand cycled the counter ions associated with the film, form the PPy formation process, wbe exchanged with the anion present in the highest concentration. In sea water this happens to be the Cl anio N ppm Br ). After this experiment the conductivity of the PPy/272-55 samples was meon a 4-point conductivity probe at currents of 10, 20, 30, 40, and 50 A. The PPy/27155 sample thickness was 0.7mm (0.07 E -7 9.09E -9 S/cm 2 This is comparable to values obtain for other CP/elastomblend systems which typically range 10 -6 10 -9 S/cm 2 155-158 This measurement is for the bulk conductivity of the whole film. This assumes that the sample is bulk conductive; however if the sample is not completely bulk conductive the conductivity of the conductive layer would be hi ng a conductive layer thickness (film thickness) of 700m (0.07 cm, bulk conductive), 100 m, 50 m, 25 m, and 5m. This resulted in film conductivities of

PAGE 208

183 9.57E -7 6.70E -6 1.34E -5 2.68E -5 6.70E -5 and 1.34E -4 S/cm 2 respectively. This datgraphically shown in Figure 7.6. a is f poly(3-methylthiophene)/271-55 IPN systems and P tubes Figure 7.6 Calculated PPy/271-55 film conductivity based on conductive layer thickness. Value for 0.07 cm is actual conductivity of PPy/271-55 film assuming bulk conductive. 7.4.3 Formation o Poly(3-methylthiophene) (PMeT) was also incorporated into the 271-55 material in the same manner as the PPy/271-55. By incorporating different conducting polymers,like PMeT, into base materials different surface properties can be produced. PMeT has different oxidation and reduction potentials than PPy and is capable of producing a net negative charge during reduction unlike PPy. This would result in a dynamic surface that switched between two hydrophilic states (positively and negatively charges) with a hydrophobic state in between the two. 271-55 sample films (N=18) were doped with FeCl 3 as previously described then dried under vacuum for 4 hrs. The samples were split in half and placed in Pcontaining ~3 ml of 3-methylthiophene and pyrrole monomer respectively. Samples 0.0E+004.0E-05PPy thickness 2.0E-058.0E-051.4E-040.07 cm100 micron50 micron25 micron10 micron5 micronConducvi/ 6.0E-051.0E-041.2E-041.6E-04tity (Scm2)

PAGE 209

184 placed in the MeT monomer exhibited no initial color change when exposed to the monomer vapor. However the sam ples placed in pyrrole monomer started to turn black completelyof the spectively. the om this it was determined that the use of PMeT was u instantaneously upon exposure to the monomer vapor. After 24 hrs both sets of samples were removed from the monomer vapor and dried under vacuum for 1 hr. The pyrrole exposed samples were completely black in color while the MeT exposed samples were generally greenish/brown in color (FeCl3 insertion) with dispersed black regions on the sample surface. The PMeT samples were non-conductive except on the black areas. The PMeT samples appeared to have turned darker during the vacuum drying process. The PPy/271-55 and PMeT/271-55 samples were placed in separate PP tubes and sealed for storage. After being sealed in the PP tube for 25 days the PMeT/271-55 had turned black in color and had a completely conductive surface. The resistively PMeT/271-55 and PPy/271-55 samples was 20-30 M and ~15 k re Methylthiophene can be chemically polymerized by FeCl 3 in solution at room temperature much like pyrrole. The vapor pressure of 3-methylthiophene is 42 mmHg compared to 65 mmHg for pyrrole with boiling points of 115-117C and 129-131C respectively so both monomers vaporize relatively easily. It is currently unclear weather the kinetics of the vapor phase polymerization of 3-methylthiophene are lower than that of pyrrole or if the presence of oxygen inhibits polymerization of 3-methylthiophene. Fr ndesired due to the long polymerization times utilizing the current polymerization process.

PAGE 210

185 Other oxidizer systems can be used in the polymerization process but the moscommon one is an AlCl t is oxidizer system would be undesed 5 TPE. les patterhe Figure 7.7 Example of 5m biomimetic sharklet pattern on PDMSe (A) and patterned silicon wafer used to create the patterned PDMSe. SEM micrographs taken at 1000X (scale bar = 50 m; WD = 15 mm and EV = 15 KeV, and AuPd coated). Images taken by James Schumacher. 3 /CuCl 3 mixture. Currently the use of copper in marine coating isunder the process of being banned therefore the use of th irable. 7.4.4 Micropatterning of 271-55 Santoprene 271-55 was micropatterned with a biomimetic sharklet pattern using a hot embossing method. A biomometic sharklet pattern developed by the Brennan research group at the University of Florida has been shown to greatly reduce the settlement of Ulva spore (green algae) when patterned on Silastic T2 PDMSe comparto smooth unpatterned PDMSe (Figure 7.7). Hot embossing methods were employed to micro pattern Santoprene 271-5Samples patterned at 185C produced the best results (Figures 7.8 and 7.9). However there are obvious defects in the pattern features produced at this temperature. Samp ned at 200C replicated very poorly resulting in virtually no pattern transfer to t271-55 (Figure 7.10).

PAGE 211

186 Figure 7.8 Optical micrograph of 5m biomimetic sharklet pattern on 271-55 produced at -55. Images taken at 1000X. Figure 7.9 SEM micrographs of sharklet patterned Santoprene 271-55 patterned at 185C (A) and patterned silicon wafer used to create the patterned 271 185C; 3300X (A) and 2500X 40 tilt (B). Images taken at 15 KeV with a working distance of 15 mm and a AuPd surface coating. Images taken by James Schumacher.

PAGE 212

187 Figure 7.10 SEM micrographs of sharklet patterned Santoprene 271-55 patterned at 200C; 500X (A) and 1000X (B). Images taken at 15 KeV with a working distance of 15 mm and a AuPd surface coating. Images taken by James Schumacher. initial patterning test show This s that it is possible to micropattern on Santoprene maremTHFpreferentially swell and deposit FeCl3 in the EPDM phases of the Santoprene TPE material. By using the EDS mapping technique it is possible to determine the location and dispersion of the Fe and Cl in the Santoprene material. It is also possible to determined that the location and dispersion of the clay filler in the Santoprene material due to the presence of silicon and aluminum. terials. However improvements in the patterning technique need to be made. The current patterning process utilizes standard office binder clips to apply the pressure for patterning. By using a more standard heated compression mold higher pressures can be applied potentially improving the replicated pattern. Also by reproducing the pattening technique under vacuum it might be possible to improve the replicated pattern by oving the air in the cavities of the negative pattern (silicon wafer). 7.5 EDS and SEM ANALYSIS of SANTOPRENE TPEs is a good solvent for EPDM but not for PP therefore the FeCl 3 /THF should

PAGE 213

188 Santoprene 271-55 and 8211-65 samples were doped with FeCl3 using the previously described FeCl3/THF solution method. Cross section samples were prepared by cutting the samples in half with a razor blade. After this the samples were mounted onto a SEM stub and coated with carbon prior to SEM analysis. Carbon was used to coat the sample instead of Au to prevent masking/blocking of the EDS and backscatter signal the SEM. Due to differences in sample thickness (0.03 vs. 0.06 for 271-55 and 8211-65) fromthe magnifmapped at 180X and 85X respectively (Figures 7.11-7.13). 1and ication of the two mapping images are different. 271-55 and 8211-65 were From the EDS spectra (Figure 7.11) the content of alumina filler is much higher in the 8211-65 materials than the 271-55 where the silica filler content is higher in the 2755. From the EDS map of 271-55 the FeCl 3 tends to have a higher concentration closerto the samples surface with the center of the material being lower in concentration. The dispersion of FeCl 3 is more uniform throughout the bulk of the sample for 8211-65. Figure 7.11 Cross sectional EDS spectrum of FeCl 3 doped Santoprene 271-55 (A) 8211-65 (B) showing the presence of Fe and Cl.

PAGE 214

189 Figure 7.12 Cross sectional EDS mapping of FeCl3 doped Santoprene 271-55 taken at 180X; scale bar = 200 m.

PAGE 215

190 igure 7.14 shows the difference between the dispersion of the FeCl3 in PDMSe, 271-55, and 8211-65 elastomers. The FeCl3 tends to concentrate in the center of the Figure 7.13 Cross sectional EDS mapping of FeCl 3 doped Santoprene 8211-65 taken at85X; scale bar = 500 m. F

PAGE 216

191 PDMSe samples (also confirmed with ATR-FTIR; Figure 6.18) while in both the Santoprene materials the FeCl3 is more uniformly dispersed. This results in a more niform polymerization of pyrrole in the Santoprene base materials resulting in a more ples u bulk conductive material compared to PDMSe. Figure 7.14 Cross sectional EDS mapping for Cl in FeCl3 doped PDMSe (A), 271-55 (B), and 8211-65 (C). Magnification and scale bars equal to 70X, 180X, and 85X and 500 m, 200 m, and 500 m respectively. Images taken at 15 KeV and 15 mm on carbon coated cross section samples. This type of analysis was also conducted on PPy/271-55 and PPy/8211-65 sam(Figures 7.15 and 7.16). From this the dispersion of the clay filler (Si, Al, and O) and the PPy (Fe and Cl) can be detected. The formation of PPy should only occur where the FeCl3 was deposited, therefore by mapping for Fe and Cl the location of the PPy can be determined.

PAGE 217

192 Figure 7.15 Cross sectional EDS mapping of FeCl3 doped PPy/271-55 taken at 3000X; scale bar = 10 m. Images taken at 15 KeV and 15 mm on carbon coated cross section samples. The 271-55 and 8211-65 samples have large clusters on the order of 1-8 m of clay filler. The dispersion of the FeCl3 deposited in the 271-55 base material is in the form of large cluster as well with dimensions on the order of 2-5 m. A finer uniform dispersion FeCl3 is also present. However the dispersion of the FeCl3 deposited in the 8211-65 is

PAGE 218

193 almFigure 7.16 Cross sectional EDS mapping of FeCl doped PPy/8211-65 taken at 3000X; cross section samples. ost completely uniform. There is no evidence of the large clusters of FeCl3 that are evident in the 271-55 material. 3 scale bar = 10 m. Images taken at 15 KeV and 15 mm on carbon coated

PAGE 219

194 The distribution of higher density phases (clay filler and FeCl 3 ) in PPy/271-55 andPPy/8211-65 are also shown in th e cross sectional SEM secondary and backscatter ages PPy ed into the thermoplastic material polysulfone (PSU). Polysulfone has excellent oxidative resistance and mechanical properties. The overall mechanical properties of this material are comparable to that of polycarbonate. By electron images, Figures 7.17 and 7.18. These images were taken at 3000X. Figure 7.17 Cross sectional SEM secondary (A) and backscatter (B) electron micrographs of PPy/271-55. Images taken at 3000X and scale bar equal to 10 m. Imtaken at 15 KeV and 15 mm on carbon coated cross section samples. Figure 7.18 Cross sectional SEM secondary (A) and backscatter (B) electron micrographs of PPy/8211-65. Images taken at 3000X and scale bar equal to 10 m. Images taken at 15 KeV and 15 mm on carbon coated cross section samples. 7.6 PPy/POLYSULFONE SYSTEMS has also been incorporat

PAGE 220

195 utilizi filmsamonto the PE the 10 psi. A ccontrast of the film. The filmoothes and most uniform surface texture. While the films sprayed from the 10 wt% solution ng hard thermoplastic materials such as polysulfone very hard and durable dynamic surface coatings should be able to be formed. Also these systems can be applied byconventional paint application processes. To investigate the effects of substrate material and FeCl3 concentration FeCl3/PSU s were cast onto glass and PET substrates at various concentrations of FeCl3. Concentrations of 1, 5, 10, and 20 wt% FeCl3 were used. The 1 wt% FeCl3 solution produced films of a light grey color that were non-conductive while the 5, 10, and 20 wt% films turned dark opaque black in color with an increasing level of opacity. Films cast onto the cleaned glass slides produced a non-uniform surface morphology. The ples had a blochy appearance with shiny and dull sections. However the films cast T covered slides had a uniform shiny black surface. There was no real difference in the conductivity of the PPy/PSU films cast on glass and PET covered glass. But as the FeCl3 concentration was increased the overall conductivity of the samples increased. Samples produced from the 5 wt% solution had a resistivity range of ~450-1000 k, while the samples produced from the 10 wt% solution had a resistivity of 120-600 k. The samples prepared from the 20 wt% solution hadhighest conductivity with a resistivity range of 14-200 k. Spray application techniques were also investigated. Solution concentrations were evaluated at 5 and 10 wt% PSU. Argon was used to the spray the films at a pressure of ommercial blue thermoplastic dye was added to the solution to improve the s sprayed from the 5 wt% solution produced the sm

PAGE 221

196 tended to fibrous in nature. An example of this is shown in Figure 7.19 at a magnification of 7X. The films formed a very hard and sc ratch resistant layer on the glass substrate. Howet more is s er. proved mechanical prope From this it has been shown that it is possible to form conducting polymer composite structures with dynamic surface properties such as surface energy, modulus, ver the films easily delaminate from the glass when exposed to water. This was initially thought to be due to the FeCl 3 potentially degrading the PSU and making ihydrophilic or due to the hydration and swelling of the incorporated PPy. However thdoes not happen for the films cast on the PET. Also pure PSU that was applied to glasslides in the same manner also delaminate from the glass substrate. It turns out that PSUdoes not bind well to glass and simply de-wets the glass surface when exposed to watPSU has also been solution cast onto the biomimetic sharklet design. The PSU replicated the sharklet pattern nicely and results in patterns with im rties compared to PDMSe. From this it should be possible to form hard dynamic micropatterned surface coatings. Figure 7.19 Optical micrographs of PSU films sprayed from 5 wt% (A) and 10 wt% (B)PSU/THF solutions. Magnification = 7X. 7.8 CONCLUSIONS

PAGE 222

197 and topography. PPy was incorporated into Santoprene 8281-65 TPE and wproduce a dynamic con as shown to tact angle change of 11 under a .5 V stimulus. The contact angle proceg. From this it was shown that the dispersion of was shown to change from 30 in its oxidized form at +0.5 V to 41 in its reduced form at -0.5 V. PPy/8211-65 samples were then shown to produce a dynamic surface modulus change when switched between its two redox states. Micro contact patterning of a biomimetic sharklet pattern developed here at the University of Florida was accomplished on Santoprene 271-55 and Udel polysulfone. This pattern has been shown to reduce the settlement of Ulva spore on patterned PDMSe by ~85% when compared to smooth PDMSe. By incorporating the micro patterning ss with the dynamic surface materials the overall performance of these systems should be greatly enhanced. Also the distribution and morphology FeCl 3 deposited inside Santoprene TPEs was investigated using EDS mappin FeCl3 inside Santoprene 271-55 and 8211-65 was relatively uniform when compared to PDMSe. PDMSe was previously shown to surface segregate and force the FeCl3 into the center of the sample resulting in an overall non-conductive material. Where as the Santoprene materials have a uniform FeCl3 dispersion throughout the bulk of the material producing conductive samples with a very stable surface PPy layer.

PAGE 223

CHAPTER 8 DYNAMIC MODULUS MAPPING OF PPY/SANTOPRENE BLENDS 8.1 INTRODUCTION In order to investigate the dynamic modulus properties of the PPy/TPE systems nano-DMA measurements were conducted using a Hysitron Triboindenter equipped with the nano-DMA package. Hysitron nano-DMA is performed in a similar fashion to standard DMA testingtechniques. A load is applied to a sample at a given frequency and the resulting displacement and phase shift between the frequency of the displacement and applied load is meus l t and c frequency test. In this test a constant quasistatic load and dynamic load amplitude are used while the frequency is ramped. The Triboindenter is capable of applying and measuring frequencies from 1-300 Hz; however below 10 Hz the testing time becomes too long to be feasible and above 200 Hz the displacement signal amplitude becomes very small making it difficult to measure the phase and amplitude response of the material. Therefore the typical measurement ranges is around 10-200 Hz. asured. By monitoring the displacement and phase shift of the sample under varioloads and frequencies or temperatures the viscous and elastic responses of the materiacan be determined. The Hysitron nano-DMA accomplishes this by applying a small force/load to the sample surface by oscillating the tip on the sample surface. As this is done the applied force and oscillation frequency are monitored and the displacemenphase lag produced in the sample is measured. The most common technique used for nano-DMA is the ramping dynami 198

PAGE 224

199 8.2 MATERIALS AND METHODS 8.2.1 Materials P(Brockman activity 1; Fisher Scirless to remove impurities. Iron (re m. Artificial sea water was made using Instant Ocean artificres er for 24 hrs to facilitate the formation of PP d lides atterned yrrole (Sigma-Aldrich) was filtered, before use, through a neutral alumina entific) column until colo III) chloride (FeCl 3 ) (Fisher Scientific) and tetrahydrofuran (Fisher Scientific) weused as received unless otherwise stated. Deionized (ultrapure) water was produced using an 18 M Millipore syste ial sea salt (Fisher Scientific) and deionized water (1 ml I.O. in 30.7 ml H 2 O; specific gravity = 1.020-1.023 g/cm 3 salinity = ~28-34 ppt). 8.2.2 Sample Preparation Methods 8.2.2.1 PPy/TPE (8211-65) Sample Preparation The oxidizer insertion method was used to form PPy/TPE IPN systems as previously descrbed. Santoprene 8211-65 films were cut into 18 mm x 18 mm squato produce samples. Samples were then exposed to a 5 wt% FeCl 2 /THF solution for 24 hrs and then dried for 1 hr. The samples were then placed directly in a sealed PP centrifuge tube containing 2 ml of pyrrole monom y. After the pyrrole monomer treatment the samples were again dried undervacuum for 1 hr to remove any residual pyrrole monomer. 8.2.2.2 Sample Mounting Samples were then mounted to glass slides using Ag-filled epoxy and encapsulatewith PDMSe using the procedure previously described in chapter 7. The base electrode structure (Ag-filled epoxy) was painted onto clean glass sand then cured at 175C for 30 minutes. Samples were then mounted onto the p

PAGE 225

200 glass slide using Ag-filled epoxy and then sandwiched between two glass plates and cured as above. te surface (glass slide, Ag-filled epoxy, and lead wire) was then cleaned by Arted ss Se late the sample. The 0C for 5hrs. After this the top cover plate and ready to use. 8.2.2.r tificial 8.2.3.ducer (Figure 8.1) to pro The substra plasma treatment for 1 minute at ~50 mTorr. The sample surface was protecfrom the Ar plasma using a layer of double sided sticky tape. After the cleaning procethe substrate was treated with an ATS coupling agent to improve the adhesion of the PDMSe encapsulant to the glass substrate. A top cover plate (PET covered glass slide) was applied to the sample and the edges of the glass slides were sealed with scotch tape (creating a small mold). PDMwas then backfilled in-between the two glass plates to encapsu PDMSe was then cured in an oven at 5 protective tape layer are removed and the sample is 3 Electrochemical Polymerization of Pyrrole Polypyrrole was electrochemically polymerized on an EvAu coated polyimide films from a 0.1 M aqueous solution of pyrrole and NaClO 4 using a stainless steel counteelectrode and an Ag/AgCl reference electrode. The films were grown to a current density of 4 C/cm 2 The PPy films were then broken in by cycling the samples between .8 V at 10 mV/sec ten times in I.O. arsea water prior to nano-DMA measurements. 8.2.3 Hysitron Nano-DMA 1 Hysitron Introduction The Hysitron Triboindenter uses a three plate capacitance trans duce small tip displacements (<5m) and to measure and apply loads to the sample

PAGE 226

201 surface. The whole transducer is mounted on a stepper motor assemble to produce lascale displacemen rge ts (>5m). ) en the center and bottom plates. te toring how much DC The Triboindenter is capable of performing nano-DMA measurements between 110-200 Hz. At measurement speeds of less then 10 Hz the experimental timscan speed). The sampling tip is mounted on the center plate and its position is controlled by a DC current passed between it and the top and bottom plates. By applying a DC currentbetween the center and top plates the center plate is electrostatically attracted to the top plate produces a negative displacement in the tip (retraction). The opposite (extensionoccurs when the DC current is applied betwe The position of the center plate and therefore the tip is monitored by two separaAC currents. AC currents are passed between the center plate and the top and bottom plates and are 180 out of phase in the zero position. By monitoring the phase shift between the two AC signals the position of the center plate is determined. The force/load exerted on the sample is determined by moni current is required to maintain the center plate position Figure 8.1 Diagram of three plate capacitor transducer used in the Hystitron Triboindenter. 300 Hz but is typically run from e becomes exorbitant and to speed greater than 200 Hz the response signal becomes difficult to detect (response signal amplitude decreases with increased

PAGE 227

202 8.2.3.2 Experimental conditions Nano-DMA mapping w as conducted on dry 271-55 and PPy/271-55 samples using erzian conta-65 and PPy samp 0 mm) with a stainless steel diagram and photograph of the fluid cell setup are shown in Figure 8.3. Due to the viscosity of the artificial sea water in the fluid cell the load on the transducer N during the sample approach. This caused the Triboindenter the ramping dynamic frequency test from 10 Hz-200 Hz. A quasistatic load (pre-load) of 2 N and dynamic load of 1 N was used. The tip used during these experiments was a diamond cono-spherical tip (Figure 8.2A) with a tip radius of 0.94 mand a cone angle of 60. Under these conditions this equates to a calculated contact areaof about 100-200 nm if a sample surface modulus of 1-4 GPa is used (circular H ct calculator with sample radius approaching infinity; http://grove.ufl.edu/~wgsawyer/Laboratory/Software/Software.html ). The tip has a modulus of 1140 GPa and a Poissons ratio of 0.07. Dynamic fluid cell nano-DMA mapping was conducted on PPy/8211 les driven by a Princeton Applied Research 263A-2 potentiostat. The fluid cell wasconstructed using a standard glass Petri dish (150 mm X 2 counter electrode and Ag/AgCl reference electrode. The stainless steel counter electrode was placed on the bottom of the fluid cell adjacent to the sample for the initial experiments. In subsequent experiments (experiments discussed here) the counter electrode was wrapped around the inside edge of the Petri dish wall. The sample was mounted in the center of the fluid cell and covered with artificial sea water. A The detection of the sample surface during tip approach is controlled by the load measured on the tip. Once a load equal to the quasistatic load is detected the Triboindenter assumes it has found the sample surface and starts scanning that surface. repeatedly exceeded 2

PAGE 228

203 to falsely engage the surface about 5-10 mm above the actual sample surface (visual obserhe and is typically used for ed very flat tip with a large contact area eFigurobtained from www.hysitron.com (magnification unknown). vation). To overcome this problem a higher quasistatic load of 10 N was used during the fluid cell experiments. The dynamic load was 1 N. Ramping dynamic frequency test were conducted from 10 Hz-200 Hz. Tmachine scan rate was 0.25 Hz with scan areas of 25 m x 25 m, 10 m x 10 m, or 1 m x 1 m. The tip velocity was 5 m/sec. A Berkovich fluid tip (Figure 8.2B) was used for these experiments due to the fact that no other fluid tip was available. The Berkovich tip is not ideal for polymeric materialsindention testing of harder materials and metals. The Berkovich tip has a total includangle of 142.3 with a half angle of 65.35 making it a specially in polymers. The average radius of curvature for these tips is typically between 100 and 200 nm. A more suitable choice for a fluid tip is the cono-sherical tips. e 8.2 Images of cono-spherical (A) and Berkovich (B) Hysitron tips. Images

PAGE 229

204 Figure 8.3 Diagram (A) and optical image (B) of fluid cell setup inside the Hysitron Triboindenter. 8.3 HYSITRON NANO-DMA and AFM ANALYSIS 8.3.1 Nano-DMA mapping of 271-55 and PPy/271-55 samples A 50 m x 50 m square area was mapped on 271-55 and PPy/271-55 samples. One ofg the e Fromdepths up to 900 nm while the change in surface height for the untreated 271-55 is on the order of 200-400 nm. the first things noticed is the appearance of dark black phases in the sample surface after the formation of the PPy IPN system. These phases were detected usinoptical camera mounted in the Triboindenter and are shown in Figure 8.4 (magnification = 5X). These phases are either due to clustering of PPy on/in the sample surface or duto surface defects present in the material. These features are also present in the 3D surface topography images (Figure 8.5). this it can be seen that the PPy/271-55 sample surface is much more irregular than the untreated 271-55 sample. The cracks present in the PPy/271-55 material reach

PAGE 230

205 Figure 8.4 Optical micographs of Santoprene 271-55 (A) and PPy/271-55 (B) samples obtain by the onboard optical camera of the Hysitron Triboindenter (magnification = 5X). Figure 8.5 3D surface plots of Santoprene 271-55 (A) and PPy/271-55 (B) surfaces. e surface defects have also been detected in untreamodulus mapping data of the samples surfaces reveals that there is a distinct difference between the 271-55 and PPy/271-55 samples. The phase and modulus Scan area is equal to 50 m x 50 m square. However it should be noted that larg ted 271-55 samples using modulus mapping. In one area a hole with approximate dimensions of 50 m long and about 40 m wide was detected in untreated 271-55. The depth of the defect was in excess of 2 m. The Hysitron tip lost contact with the surface in the defect so its exact depth could not be determined. Phase and

PAGE 231

206 im-55 samrface compFigure 8.6 Nano-DMA mapping images of Santoprene 271-55 showing uniform surface ages for the untreated 271-55 show a relatively uniform surface morphology with a surface complex modulus in the range of 0.5-0.7 GPa (Figure 8.6). While the PPy/271ples have a very erratic non-uniform surface morphology and an average sulex modulus in the range of 0.4 GPa to 2.5-3.5 GPa (8.7). phase and modulus morphology. Scan area is 50 m x 50 m square.

PAGE 232

207 Figure 8.7 Nano-DMA mapping images of PPy/271-55 showing uniform surface phase and modulus morphology. Scan area is 50 m x 50 m square. 8.3.2 Dynamic fluid cell nano-DMA mapping of PPy/8211-65 Using modulus mapping and nano-DMA testing, changes in complex, storage, and loss modulus along with changes in tan delta and surface topography were measured in situ while switching the material between its oxidized and reduced states.

PAGE 233

208 Dynamic fluid cell nano-DMA analysis of PPy/8211-65 revealed a phase separated in the samimodulus of the scannonly thing thFigure 8.8 als of -0.7 V (A), 0.0 V (B), and +0.7 V (C). Scan area is 25 m x 25 m square. An AFM analysis of the sample surface was conducted to confirm the morphology observed with the Hysitron (Figure 8.9). These measurements were conducted on a dry sample in tapping mode on a Veeco Dimension 3100 AFM. Surface imaging of 50 m, morphology comparable to that seen for PPy/271-55 (Figure 8.7B). Initially it was believed that these darker phases were PPy rich phases. A 25 m x 25 m scan of the dark phase (Figure 8.8) produced images with large amounts of tip effects. A difference ple morphology was detected due to changes in redox states. Tip effects are usually seen when the surface being imaged is very rough and features sizes are much smaller than the tip contact area. In this case you actually end up aging the tip with the sample surface instead of imaging the surface with the tip. Yet, even with the tip effects present; changes in the average surface med area were detectable. The average surface modulus (complex) of the sample at the applied potentials of +0.7 V, 0.0 V, and -0.7 V were 6.47 2.76, 6.83 2.83, and 10.85 0.68 GPa respectively. Due to the unknown tip area for the Birkovich tip the at can be said is that there is about a 40% decrease in surface modulus when switched from the reduced state to the oxidized state. 3D surface topography images of PPy/8211-65 at applied potenti

PAGE 234

209 10 mnot so revealed somelar res couldFigure 8.9 AFM optical micrograph (A) of PPy/8211-65 with corresponding surface area. Scale bars equal to 15m. 5 m, and 1 m square scan areas was conducted. These AFM images reveal that the phase contrast observed in the optical images is due to large surface defects and PPy segregation. This supports the observations in the initial fluid cell nano-DMA imaging of PPy/8211-65. The AFM topography and phase imaging of the sample surface al interesting surface morphology (Figure 8.10). Some surface features with reguangles are evident in the 10 m and 5 m scan area images. It is not clear at this time what these features are due to and further investigation needs to be done. These featu be the result of FeCl 3 salt deposition in the sample surface, the presence of clay filler particles, or the formation of PPy. topography (B) and phase contrast (C) images taken at a 50m square scan

PAGE 235

210 Figure 8.10 AFM surface topography, phase imaging, and 3D surface plot of PPy/8211-65 at 50 m (A), 10 m (B), 5 m (C), and 1 m (D) square scan areas. Sbars equal to 15 m, 3 m, 1.5 m, and 0.3 m respectively. The surface morphology of the PPy in the PPy/8211-65 material is still not evidenFuture studies will include SPM (scanning probe microscopy) imaging of the PPy/TPE terials using the AFM setup. By using SPM techniques the spreading resistivity of the ple surface can be mapped. This should reveal the location and morphology of the conductive regions on the sample surface thus giving the location and morphology of the PPy phase. cale t. masam to surface defphase regions. The scan area was reduced to 10 m square for this set of measurements. The scan area was also translated 30 m in the positive X direction from the initial tip After the determining the phase contrast observed in the optical images was dueects; dynamic fluid cell nano-DMA experiments were conducted on the lighter

PAGE 236

211 engagement point. This was done to insure that the sample surface being imaged wasaged due to the initial tip engagement process. Due to the fluid cell setup false engagements were a problem during the initial tip approach to the sample surface (tip stops about 5-10 mm above sample surface). This was overcome by increasing the quick approach set point (quasistatic load). The set point value is the static load the tip places on the sample during the nano-DMA measurements and is also used to automatically detect the surface. Once the tip experiences a force oequal or greater value than the set point the Triboindenter algorithm converts to surfpping. The tip engagement force (set point) is typically 2 N but due to the false engagement problems the set point had to be increased to 10 N. This high force could potentially damage the sample surface during the initial approach process. The dynamload (oscillating load) used for these experiments was 1 N. not damf ace maic scanned. F0 x 10m scan area changed fromthe average complex modulus for the whole sample surface. This data also shows th Under these conditions there were no tip effects evident for the sampled area or this location the average surface complex modulus for the 1 7.71 0.70 GPa to 14.23 2.00 GPa and 13.23 1.75 GPa under the applied potentials of +0.8 V, 0.0 V, and -0.8 V respectively. This equates to a 42% decrease in surface modulus when switched from the reduced state to the oxidized state. Line scan modulus data was also extracted from the complex modulus map. The line scan data for the positions shown in Figure 8.11 are exhibited in Figure 8.12. This data agrees with the values obtained for e presence of a surface feature of about 1m in diameter that does not change its modulus with the applied potential as much as the rest of the sample surface.

PAGE 237

212 Figurer 8.11 Complex modulus maps for PPy/8211-65 at an applied potential of -0.8 V (A), 0.0 V (B), and +0.8 V (C). Figure 8.12 Complex modulus line scan data for PPy/8211-65 obtained from the location depicted in Figure 7.28 By looking at the 3D surface plots obtained from these measurements it can be seen that the naturally occurring surface features present on the sample surface swell and de-swell with the changes in the applied potential. The 3D surface plots show a ridge (top center of the plot) that waves up and down from the sample surface when the potential is switched between -0.8 V and +0.8 V (Figure 8.13). This demonstrates the dynamic surface topography capabilities of these types of materials. With improved processing 05100246810Position (m)omplx M 152025usGPa) Ceodul ( -0.8 V 0.0 V +0.8 V

PAGE 238

213 and patterning techniques precise engineered dynamic surface topographies should be e produced, able to bFigure 8.13 3D surface plots of PPy/8211-65 under the applied potentials of -0.8 V (A), 0.0 V (B), and +0.8 V (C). mmsampcondu The average surface complex modulus changed from 11.95 1.64 GPa to 11.90 2.02 GPa and 14.92 5.76 GPa under an applied potential of +0.8 V, 0.0 V, and -0.8 V respectively over a 1 m square area. On this scale the decrease in surface modulus was only ~20% when switched from the reduced to the oxidized state. The modulus changes easured for this area are not as significant as those produced in the previous easurements. However they are on the same order as the previous measurements and follow the same trend as before. This could be due to the smaller scan size and the le morphology of the area being mapped. As a comparative study dynamic fluid cell nano-DMA measurements were cted on pure PPy. The PPy sample were electrochemically switched between -0.8 V, 0.0 V, and +0.8 V and nano-DMA measurements were conducted. The average surface complex modulus changed from 2.42 1.03 GPa to 2.34 0.15 GPa and 25.83 7.53 GPa for potentials of +0.8 V, 0.0 V and -0.8 V respectively. This equates to a 90% decrease in the surface modulus when switched from the reduced to oxidized state. Conducting polymers have previously been shown to change their modulus during

PAGE 239

214 electrochemical redox switching. 94, 143-147 The average surface complex modulus for PPand PPy/8211-65 are p y lotted against each other in Figure 8.14. Figure 8.14 Average surface modulus for PPy and PPy/8211-65 systems as a function of active t ties e amount of conducting polymer incorpther the applied potential. The modulus change produced by the PPy sample is significantly larger than recorded for the PPy/821-65 materials. This is expected since the PPy is the rematerial in this system. By incorporating it into the 8211-65 TPE as a minor componenthe overall change produced from this system would be reduced. The overall properof these IPN systems are directly dependent on th orated into the system. From this it can be said that the surface of the PPy/8211-65 samples are capable of producing dynamic surface properties such as surface modulus and topography. Furtests need to be conducted on these samples with the proper fluid cell tip in order to determine the true value and magnitude for the changes in surface modulus. 010152025303540E(V) versus Ag/AgClA Complex Modulus (GPa) PPy (10 micron area) PPy/821165 (25 micron area) Ppy/821165 (10 micron area) PPy/821165 (1 micron area) 5vg. -1-0.500.51

PAGE 240

215 8.4 CONCLUSIONS Hysitron nano-DMA was used to conduct dynamic electroc hemical fluid cell measureswitching podulus the characterization of these types of material. ments on the PPy/8211-65 system. The PPy/8211-65 systems were shown to produce an average surface modulus change on the order of ~20-40% when switched between its two redox states. Also dynamic changes in surface morphology were also shown to follow the redox rocess in these systems. Dynamic fluid cell nano-DMA measurements were also conducted on PPy. Polypyrrole was shown to undergo a dynamic surface mchange on the order of 90% when switched. This is the first time anybody has conducted dynamic surface modulus and topography mapping in this manner. This technique could prove to be very valuable in

PAGE 241

CHAPTER 9 rk was to develop new and advanced high performance condu surface coatings for marine and commork are ement technique was developed using commercially available strain gages to monitor the in situ strain performance of conducting polymer (CP) based bimorph actuators. This new in situ strain gage measurement technique was used to evaluate the strain performance of a commercially available CP polypyrrole (PPy) and two new CPs poly(3,4-ethylenedioxypyrrole) (PEDOP) and poly[3,6-2(2-(3,4-ethylenedioxythienyl)-carbazole] (PBEDOT-Cz). From this study it was shown that the strain performance of PPy far exceeded (~4.5 times higher) that of PEDOP and PBEDOT-Cz. Due to this PPy was chosen as the standard material for all continued work. 2. Interlayer adhesion between the CP and base electrode was determined to be a problem associated with CP based actuators. A new electrochemically deposited Au (EcAu) layer was developed that greatly CONCLUSIONS AND FUTURE WORK The initial goal of this wo cting polymer based actuator designs and evaluation techniques. After this initial work the knowledge and lessons learned were then applied to the development of conducting polymer based dynamic surface coatings. These dynamic surfaces have a potential role as dynamic non-toxic antifouling ercial applications. A quick summery of some of the major points to this wlisted below; they will be discussed in further detail in the following paragraphs: 1. A new in situ strain gage measur 216

PAGE 242

217 enhanced the interlayer adhesion between the CP and base electrode. As a result of this; a drastic improvements in the overall strain performance and lifetime of these devices was obtained. 3. A new ra the fabrication of ue can c mic on of a 5. e energy and were also pid patterning technique was developed for advanced electrode designs utilizing standard office software and printer systems. This rapid patterning technique was used to design and fabricateadvance linear actuator designs. It was also shown that this techniqalso be used to quickly produce inexpensive photomasks for lithographiprocessing. 4. Utilizing techniques obtained from the previous studies conducting polymer based dynamic surface coating were developed. The dyna surface energy of these materials was shown using contact angle measurement techniques. These films were based on the formatiPPy IPN system inside a PDMSe base material. The effects of various solvent soak systems used in the fabrication of these materials were also evaluated. Further development of CP based dynamic surface coatings was continued using Santoprene thermoplastic elastomers (TPE). By utilizing elastomers with improved mechanical properties compared to PDMSe tougher more robust dynamic surface coating were developed. These materials were shown to have a dynamic surfac shown to have dynamic surface moduli and topographies as measure by dynamic fluid cell nano-DMA mapping.

PAGE 243

218 Throughconducting polthe initial studire developed by the forming CP based lnated by electrochemsubstrates. Duredox cycling athe performancconditions is crucial to there continued development and use. A new in situ strain gage measurnt tPPy, PEDOP, conducting polThe two new conductincontenders for evaluate the strstrain gage metechnique was used. Utilizing this technique it was shown that cyclic sn recapable of prod. PPy produced an avPBEDOT-Cz prespectively. Amaterials was e largest degree d PBEDOT-Cz. out the various studies conducted the formation and evaluation of ymer based composite structures and devices was the common theme. In es conducting polymer based actuators we ami composite structures. Conducting polymer based actuators were formeical polymerization of various CP monomers on EvAu coated flexible e to the expansion and contraction of the conducting polymer layer during bending motion can be produced from the actuator. The determination of e characteristics of these devices under there intended application emeechnique was developed and utilized to evaluate the strain performance of and PBEDOT-Cz based actuators. PPy being a commercially available ymer is widely used in various conducting polymer applications. g polymers PEDOP and PBEDOT-Cz were thought to be potential the replacement of PPy in CP based actuator applications. In order to ain performance of actuator system based on these materials the in situ asurements traisponse of these materials can precisely be measured. PPy was shown to be ucing a significantly higher strain than PEDOT and PBEDOT-Czerage overall strain response on the order of 250 while PEDOP and roduced an average overall strain response of about 45 and 35 lso the degree of hysteresis present in the strain response of these also shown. While PPy produced the largest strain response it also has thof hysteresis when compared to PEDOP an

PAGE 244

219 From this study it was also determined that the interlayer adhesion betweenconducting polymer layer and the base electrode (EvAu) was a potential problem for the prolonged use of these devices. Due to the repeated swelling and deswelling of the CP layer and the resultant bending motion developed in these devices during electrochemicalcycling a large amount of interlayer strain is produced. This initiates micro-crack formation which eventually propagates and cause the complete delamination of the CP from the base electrode. This drastically reduced the potential lifetime of these devices. Also as the interlayer cracks propagate across the electrode surface the CP contactwith the base electrode is reduced. This in turn reduces the ability of CP to generate the desired strain response. This issue was remedied by the application of an electrochemically deposited Au(EcAu) layer on top of the base EvAu layer. The EcAu layer was found to greatly improve the interlayer adhesion of the CP layer to the base electrode layer. This wresult of the drastically increased electrode surface produced by the EcAu treatment alwith the improved mechanical adhesion due the growth of the CP layer into the nooks and crannies of the EcAu layer (mechanical interlocking). The application of the EcAu layer to the base electrode layer was also found to improve the performance properties such as strain, strain rate, and frequency response of the actuator systems. This believedto be due the decreased CP layer thickness due to the increased electrode surface area.This improves the diffusion rate of counter ions and associated solvent into and outCP layer. By controlling the EcAu deposition time the size of the EcAu crystals ccontrolled thus controlling the CP layer thickness. It was shown that by increasing the EcAu layer thickness up to a point that the overall strain production, strain the area as the ong of the an be rate,

PAGE 245

220 frequ s An increase in the applied driving potential (oxidation poten disrupts ency response, and lifetime was improved. The EcAu treatment produced an increase in the strain response of about 110% and improved the actuator strain performance after 2000 cycles by 20%. The in situ strain gage measurement technique was also used to evaluate the effects of counter ion type, driving potentials, electrochemical polymerization charge, andelectrochemical polymerization potential on the overall strain response of PPy/EcAu andPPy/EvAu based actuators. From these tests it was determined that the counter ion type greatly influenced the overall strain response of the actuator systems. Larger counter ionsuch as ClO4 produced about a two fold increase it the strain response than smaller counter ions such as Cl. tial) was shown to increase the overall strain response of the material up to a potential of 0.6V. It was also shown that as the electropolymerization charge was increased, resulting in thicker CP thickness, the overall strain response of the system also increased. However for a given increase in electropolymerization charge the resultant increase in the strain response decreases. This is believed to be due to a decrease in thediffusion rate of the counter ions and associated solvent in the CP layer as the layer thickness increases. The evaluation of the effects of electrochemical polymerization potential showed that as the electropolymerization potential is increased above 0.5V the overall strain response if the system is increased up to 0.7V. Above this potential the overall strain response of the system was found to decrease. The strain response of the system was decreased by ~20% when the electropolymerization potential was increased from 0.7V to 1.0V. As the electropolymerization potential is increase the ability of sidereactions to occur such as the 3,4 polymerization in pyrrole also increases. This

PAGE 246

221 the overall conjugation of the system and increases the crosslink density of the syThis results in a decrease in conductivity and in increase in the stiffness of the mThe increased material stiffness will resu stem. aterial. lt in a decrease in the diffusion rate and swellte signs. s on ign e s ability of the system decreasing the overall strain performance. In order to further the development of CP based actuators the ability to fabricanew and diverse electrode patterns on various substrates is required. A new rapid patterning technique was developed utilizing standard office printers and standard Au coating techniques. This technique allows for the rapid development and fabrication of virtually any electrode pattern of flexible substrates such as polyimide and PET. This technique was used in the development of advanced linear actuator deInitial linear actuator designs were based on a staggered placement of PPy segmentopposite sides of a conductive flexible substrate. These actuators were capable of producing linear strains of 2.5%, 12.5%, and 24% for 5, 10, and 20mm PPy segments. Advanced pattern designs were created with both the working and counter electrode being placed on the actuator surface. This eliminated the need for an external counter electrode. Both single and double sided designs were created. The double layer deswas constructed using a modified backbone type configuration where CP segments arplaced on opposite sides of the flexible substrate. As one side is oxidized the other is reduced resulting in a push-pull type strain production. The major issue with using thitechnique in the fabrication of electrode patterns is that the final fabricated electrode fidelity is only as good as that of the printer used to print it. Laser jet printers have the capability of printing on almost any substrate but they suffer from toner overspray. Random toner particles end up in the un-patterned areas of the design causing defects in

PAGE 247

222 the final electrode pattern. Ink jet printers were found to greatly reduce the amount of overspray in the unpatented areas. However; ink jet printers require the use of specially treated substrates to improve the wetting and drying characteristics of the ink on tsubstrate. This surface treatment has been found to prevent the adhes he ion of the applied Au la The produn into e ed e yer thus preventing the fabrication of the electrode pattern. However ink jet printing has been shown to be an effective way of rapidly producing inexpensive photomasks for use in standard photolithographic techniques. Conducting polymer based dynamic surface coatings were developed utilizingthe induced changes in physical property of CPs that occur during redox cycling. ction of strain seen in CP based actuator systems is produced from the dynamic change in charge developed in the CP system. As a positive charge is developed in the CP during the oxidation process counter ions and there associated solvent are drivethe CP to neutralize the developed charge. This results in the swelling of the CP thus producing the observed strain. The swelling of the CP also results in a decrease in thmodulus of the CP compared to the reduced state. The solvent associated with the counter ions acts as a plasticizer in the CP and also increases the free volume of the system this in turn decreasing the overall modulus of the material. The surface properties of dynamic PPy/PDMSe composite coatings were evaluatusing electrochemical fluid cell contact angle measurements. These coatings produced a21 change in contact angle when stepped between an applied potential of 1.0V. Thoverall dynamic properties of these composite structures are directly proportional the weight percent of the CP phase incorporated into the system.

PAGE 248

223 The loading of the CP phase in the composite structure is dependent on the solvensystem used during the fabrication of these systems. By changing the solvent system used in the formation of PPy/PDMSe composite structures from ethanol/supercritical(scCO t CO2 re e of lubility of FeCl3 and other Py Due to this; the PPy/Pre stems. Dynamic contact angle changes of 11 were produced in PPy/TPE systems when subjected to a .5V 2 ) to isopropanol/scCO 2 the weight percent of PPy incorporated into the structure was increased from ~2wt% to ~5wt%. Supercritical CO 2 based solvent systems weinitially chosen due the low environmental impacts and cost associated with the usscCO 2 ScCO 2 is very effective at swelling PDMSe however the so inorganic oxidizers is very low in scCO 2 By adding alcohol based co-solvents to the solvent system the solubility of FeCl 3 was improved. THF was found to be a bettersolvent medium than scCO 2 based systems due to the improved solubility of FeCl 3 andPDMSe and was in place of the scCO 2 based systems. The distribution of FeCl 3 in FeCl 3 /THF doped PDMSe was found to be mainly concentrated in the bulk of the material using EDS mapping. A 60-100m surface boundary layer in PDMSe was detected that was very low in concentration of FeCl 3 compared to the bulk of the material. This resulted in the formation of an unstable Psurface layer (due to oxidizer deposited directly on the sample surface) that was not directly connected to the PPy deposited in the bulk of the material. DMSe samples were found to be mainly surface conductive and not bulk conductive. Dynamic surface coatings based on Santoprene TPEs and Udel polysulfone wedeveloped in order to produce dynamic surface coating with improved mechanical properties and chemical stability compared to PDMSe based sy

PAGE 249

224 stimue system and decreased the overall densi n Se. he future development of these systems needs to focus on improved fabricn its lso lus. Dynamic changes in surface modulus were measured using Hysitron nano-DMA modulus mapping conducted in an electrochemical fluid cell. This is the first timmeasurements of this type have been conducted. A ~40% change in the average PPy/TPE surface modulus was measured when scanned over a 10m square area. Whenthe scan area was reduced to one square micron the change in the average surface modulus was determined to be ~20%. It should be noted that a Berkovich fluid cell tip was used for these experiments. The Berkovich tip design is not ideal for the mapping and measurement of polymers and can distort the actual physical property values obtained during nano-DMA measurements. During the electrochemical oxidation of PPythe material expands due to the influx of counter ions and their associated solvent (H 2 O)This swelling behavior increases the free volume of the ty of the material due to the differences in the density of PPy ( = 1.05 g/ml) andH 2 O ( = 1 g/ml). Santoprene TPEs and Udel polysulfone have also been micropatterned with aengineered biomimetic sharklet pattern. This pattern has been shown to decrease the settlement of Ulva spore by 85% on patterned PDMSe compared to un-patterned PDMBy incorporating this type of micropatterning with the dynamic surface coatings the overall performance and capabilities of these films should be enhanced. T ation techniques. The current solvent/oxidizer soak system is effective in the formation of conducting polymer IPN systems. However this technique is limited iability to form composite structures containing high volume fractions of CP due the diffusion limited process of oxidizer insertion into the base material. This process is a

PAGE 250

225 impractical as a commercially utilized process. By incorporating higher amounts of Cin the composite structure the dynamic properties of these materials should be increased. More conventional processing techniques such as melt mixing/compounding and extrusion would be more commercially viable as a fabrication technique. These techniques allow for precise control over the loading of the CP phase. One issue toconsidered in utilizing these techniques is the morphology of the resultant CP phases. During mechanical melt mixing the CP would be added into the system as a dry filler material. This would result in the formation of particulate CP phase. While the usreactive extrusion processes could result in the formation of a more uniform IPN type phase. During reactive extrusion separate conducting polymer monomer and oxidizer feeds would be pumped into the melt and would then polymerize as a secondary phase. By controlling the processing conditions the morphology of this secondary phase shoulbe able to be change P reactive be e of d d between particulate and IPN type morphologies. Also by utilizing these on of marine organrials f the techniques a variety of base materials and conducting polymers could be used allowing for the creation of various systems each with its own unique properties. Thiswould allow for the tailoring of these coating systems to there particular applications and environments. The effects of these dynamic surfaces on the settlement and adhesi isms needs to be evaluated in order to evaluate the potential use of these mateas dynamic non-toxic antifouling coatings. The wave form, magnitude, and timing oelectrical stimuli used operate these materials needs to be thoroughly evaluated. Square and sinusoidal/cyclic wave forms can be used to drive these systems. Under a squarewave stimulus the system can be instantaneously switched or incrementally stepped

PAGE 251

226 between its maximum oxidation and reduction potentials. A random walk approach could also be used to switch the applied potential as well. The length of each of these steps can also be varied. Under a sinusoidal or cyclic stimulus the applied potential is continuously ramped between the oxidation and reduction set points. The magnitthese set points and the scan/ramp rate of the applied potential can be varied. All of thevariables will affect the ability of these coating to deter the settlement and adhesion of marine organisms on these surfaces. An example of the problem being faced is for thesettlement of the Ulva spore. These spores have been shown to probe a given substrate surfaces prior to settling on that surface. Once a suitable surface has been found the spore will settle. If the dynamic surface is being driven by a square wave potential the switching time would potentially be a critical factor influencing the settlement of the spore. If the switch time is sufficiently long that the spore has time to probe and then settle on the surface before the potential is switched then the spore will probably settle and adhere to the surface. However if the switching time is shorter than the time reqfor the spore to probe and settle on the surface then the properties of that surface wouldchange during the probing process. This would most likely deter the spore from settling on the sample surfac ude of se uired e. If a cyclic wave form was used then the sample surface would const antly be in flux. This would make it difficult for the spore to determine if it likedthe surface and it would probably not settle on the surface. The scan/cycle rate applied tothe surface would affect how much change in the sample surface the spore would detect for a given amount of time. All of these variables need to be thoroughly evaluated in order to optimize these systems for the prevention of marine organism settlement and adhesion.

PAGE 252

APPENDIX A ABBREVIATIONS 271-55: Santoprene 271-55 thermoplastic elastomer 8211-65: Santoprene 8211-65 thermoplastic elastomer 8281-65: Santoprene 8281-65 thermoplastic elastomer AFM: atomic force microscopy Ag: silver Ag/AgCl: silver-silver chloride reference electrode ATR: attenuated total reflectance ATR-FTIR: attenuated total reflectance Fourier transform infrared spectrometry ATS: allyltriethoxy silane CO 2 carbon dioxide Au: gold CP: conducting polymer DI: distilled water DMA: dynamic mechanical analysis EAP: electroactive polymer EcAu: electrochemical gold EDS: energy dispersive x-ray spectroscopy EPDM: ethylene propylene diene monomer EtOH: ethanol EvAu: evaporated gold 227

PAGE 253

228 FTIR: Fourier transform infrared spectrometry I.O.: Instant Ocean artificial sea water IPA: isopropanol IPN: interpenetrating polcarbazole) Silastic T2) thylthiophene) dioxide electron microscopy py ymer network PBEDOT-Cz: poly(3,6-bis(2-(3,4-ethylenedioxy)thienyl)-NPDMSe: polydimethylsiloxane elastomer (Dow Corning PEDOP: poly(3,4-ethylenedioxypyrrole) PET: polyethylene terephthalate PMeT: poly(3-me PP: polypropylene PPP: poly(p-phenylene) PPy: polypyrrole PSU: polysulfone scCO 2 : supercritical carbon SEM: scanning SPM: scanning probe microsco THF: tetrahydrofuran TP: thermoplastic TPE: thermoplastic elastomer V: volts

PAGE 254

ENCES tension and ce in1, 13, (4), gn, actuation and control of an ism and Machine Theory 2000, 35, (7), 945-62. S.; He, K. Q.; Madou, M. J., Microactuators toward ow controlled drug delivery. Sensors and Actuators B--160. ro, .; Rodriguez, J., An electromechanical model for the ion of conducting polymers. Synthetic Metals 1996, 76, (1-., Electrochemically induced volume hangeiophene). Chemistry of Materials 1996, 8, acDiarmid, A. G., "Aritificial muscle": lyaniline films. Synthetic Metals 1995, 71, o, odriguez, J., A New Model for Electrochemical under Conformational Relaxation Control. Journal of 394, (1-2), 211-216. Shacklette, L. W.; Elsenbaumer, R. L.; Plichta, E. J.; Becht, C., anical actuators based on conducting polymers. In Molecular Electronics, Lazarev, P. I., Ed. Kluwer Academic Publishers: Dordrecht, 1991; pp 267-289. 9. Baughman, R. H., Conducting polymer artificial muscles. Synthetic Metals 1996, 78, 339-353. LIST OF REFER 1. Irvin, D. J.; Goods, S. H.; Whinnery, L. L., Direct measurement of exfor conductive polymer gel actuators. Chemistry of Materials 200 1143-1145. 2. Tuijthof, G. J. M.; Herder, J. L., Desianthropomorphic robot arm. Mechan 9 3. Low, L. M.; Seetharaman,micraves for responsive Chemical 2000, 67, (1-2), 149 4. OteT. F.; Grande, Helectrochemical oxidat 3), 293-295. 5. Chen, X. W.; Xing, K. Z.; Inganas, Ocs in poly(3,4ethylenedioxyth (10), 2439-2443. 6. Kaneto, K.; Kaneko, M.; Min, Y.; MElectromechanical actuators using po 2211-2212. 7. OterT. F.; Grande, H.; ROxidation of Polypyrrole Electroanalytical Chemistry 1995, 8. Baughman, R. H.;Micro electromech 229

PAGE 255

230 10. Della Santa, A.; De Rossi, D.; Mazzoldi, A.; Tiberi, M. In Conducting polymer actuating structures for steerable microcatheters, Proceedings of the International Symposium on Microsystems, Intelligent Materials and Robots, Sendai, Japan, Sept. 27-29, 1995; Sendai, Japan, 1995; p 275. 11. Pei, Q.; Inganas, O., Elecips built from conjugated polymers. Sy, 3718-3723. 12. Naoi, K.; Lien, M.; Smyrl, W. H., Quartz Crystal Microbalance Study Ionic 3. .; Murray, P.; Spinks, G. M.; Wallace, G. G., Mechanism of 247-256. 14. ochemically Driven Polypyrrole Freestanding Films. Journal of Intelligent 15. es Polypyrrole .2. Electrochemical Quartz-16. etric impedance techniques .2. Anion and water transport in PMPy 17. impedance techniques .1. Water transport in PPy/CuPTS films. 18. etric impedance techniques. 3. Complex charge transport in 19. s Studied by 20. in the redox processes of polypyrrole-modified electrodes. Journal of 1. nguy, J., Electrogravimetrical Study of Polypyrrole. Synthetic Metals 1989, 28, (1-2), C171-C176. trochemical muscles: Bending strnthetic metals 1993, 55-57 Motion across Conducting Polymers. Journal of the Electrochemical Society 1991, 138, (2), 440-445. Gandhi, M. R 1 electromechanical actuation in polypyrrole. Synthetic metals 1995, 73 Chiarelli, P.; Dellasanta, A.; Derossi, D.; Mazzoldi, A., Actuation Properties of Electr Material Systems and Structures 1995, 6, (1), 32-37. Torresi, R. M.; Detorresi, S. I. C.; Matencio, T.; Depaoli, M. A., Ionic Exchangin Dodecylbenzenesulfonate-Doped Crystal Microbalance Study. Synthetic Metals 1995, 72, (3), 283-287. Yang, H.; Kwak, J., Mass transport investigated with the electrochemical and electrogravim and PPy films. Journal of Physical Chemistry B 1997, 101, (23), 4656-4661 Yang, H.; Kwak, J., Mass transport investigated with the electrochemical and electrogravimetric Journal of Physical Chemistry B 1997, 101, (5), 774-781. Yang, H.; Kwak, J., Mass transport investigated with the electrochemical and electrogravim PPy/PSS films. Journal of Physical Chemistry B 1998, 102, (11), 1982-1988 Peres, R. C. D.; Depaoli, M. A.; Torresi, R. M., The Role of Ion-Exchange in theRedox Processes of Polypyrrole Dodecyl-Sulfate Films a Electrogravimetry Using a Quartz Crystal Microbalance. Synthetic Metals 1992, 48, (3), 259-270. Maia, G.; Torresi, R. M.; Ticianelli, E. A.; Nart, F. C., Charge compensation dynamics Physical Chemistry 1996, 100, (39), 15910-15916. Slama, M.; Ta 2

PAGE 256

231 22. Song, M. K.; Gong, M. S.; Rhee, H. W., Anion movement in polypyrrole films prepared in lithium dodecylsulfate aqueous solutions. Molecular Crystals and Liquid Crystals Science and Technology Section a-Molecular Crystals and LiquCrystals 1996, 280, 145-150. id hemistry B 2001, 105, (36), 8492-8497. 24. les ymers. In Handbook of Conducting Polymers, second ed.; Skotheim, T. A.; Elsenbaumer, R. L.; Reynolds, J. R., Eds. Marcel Dekker: 5. Jager, E. W. H.; Smela, E.; Inganas, O., Microfabricating conjugated polymer 26. formation of polyaniline films doped with self-existent and giant anions. Reactive & Functional Polymers 27. uctive polymer for microactuator and micromuscle applications, Micro-Mechanical d 0. Min, Y.; MacDiarmid, A. G.; Kaneto, K., Electromechanical actuators using 31. ficial muscles and examination of bending behavior. Japanese Journal of Applied Physics Part 2-Letters 1999, 38, (9AB), L1070-L1072. 32. l ing and Flexible Material Artificial Muscle. Journal of Electroanalytical Chemistry 1992, 341, (1-2), 36923. Bay, L.; Jacobsen, T.; Skaarup, S.; West, K., Mechanism of actuation in conducting polymers: osmotic expansion. Journal of Physical C Otero, T. F.; Grande, H.-J., Electrochemomechanical Devices: Artificial MuscBased on Conducting Pol 1998; pp 1015-1028. 2 actuators. Science 2000, 290, (5496), 1540-1545. Kaneko, M.; Kaneto, K., Electrochemomechanical de 1998, 37, (1-3), 155-161. Lee, A. P.; Hong, K. C.; Trevino, J.; Northrup, M. A. In Thin film cond Systems Symposium of the International Mechanical Engineering Congress anExposition, Chicago, IL, 6-11 November, 1994; Proceedings of the Dynamics, Systems, and Controls Division: Chicago, IL, 1994; pp 725-732. 28. Lewis, T. W.; Spinks, G. M.; Wallace, G. G.; De Rossi, D.; Pachetti, M., Development of an all polymer electromechanical actuator. Polym. Prep. 1997, 38, 520. 29. Madden, J. D.; Cush, R. A.; Kanigan, T. S.; Hunter, I. W., Fast contracting polypyrrole actuators. Synthetic Metals 2000, 113, (1-2), 185-192. 3 polyaniline-polyaniline films. Polym. Mater. Sci. Eng. 1994, 71, 713. Onoda, M.; Okamoto, T.; Tada, K.; Nakayama, H., Polypyrrole films with anisotropy for arti Otero, T. F.; Angulo, E.; Rodriguez, J.; Santamaria, C., ElectrochemomechanicaProperties from a Bilayer Polypyrrole Nonconduct 375.

PAGE 257

232 33. Otero, T. F., Artificial muscles, electrodissolution and redox processes in conducting polymers. In Vol. 4. Conductive Polymers: Transport, Photophysiand Applications, Nalwa, H. S., Ed. John Wiley & Sons: New York, 1997; pp 517-594. cs, m emistry 1992, 96, (25), 10507-10514. 36. ., Electrochemical Applications of the Bending Beam Method .2. Electroshrinking and Slow Relaxation in Polypyrrole. Journal of 7. Pei, Q.; Inganas, O., Conjugated Polymers as Smart Materials, Gas Sensors and 38. nyi, M., Shape Transition of Magnetic Field Sensitive Polymer Gels. Macromolecules 1998, 31, 6541-6548. 39. ne) 40. electrochemical actuator using electrochemically-deposited poly-aniline film. 41. n, R. H.; Cui, C.; Zakhidov, A. A.; Iqbal, Z.; Barisci, J. N.; Spinks, G. M.; Wallace, G. G.; Mazzoldi, A.; De Rossi, D.; Rinzler, A. G.; Jaschinski, O.; -ials Science and Engineering C 1997, 5, (2), 101-109. 43. 34. Pei, Q. B.; Inganas, O., Electrochemical Applications of the Bending BeaMethod .1. Mass-Transport and Volume Changes in Polypyrrole During Redox. Journal of Physical Ch 35. Smela, E.; Inganas, O.; Lundstrom, I., Controlled Folding of Micrometer-Size Structures. Science 1995, 268, (5218), 1735-1738. Pei, Q. B.; Inganas, O Physical Chemistry 1993, 97, (22), 6034-6041. 3 Actuators Using Bending Beams. Synthetic Metals 1993, 55-57, 3730-3735. Szabo, D.; Szeghy, G.; Zri Chen, X.; Inganas, O., Doping-induced volume changes in poly(3-octylthiophesolids and gels. Synthetic Metals 1995, 74, 159-164. Takashima, W.; Kaneko, M.; Kaneto, K.; MacDiarmid, A. G., The Synthetic Metals 1995, 71, 2265-2266. Baughma Roth, S.; Kertesz, M., Carbon Nanotube Actuators. Science 1999, 284, 13401344. 42. Della Santa, A.; Mazzoldi, A.; Tonci, C.; De Rossi, D., Passive mechanical properties of polypyrrole films: a continuum poroelastic model. Mater Ouyang, J.; Li, Y., Preperation and characterization of flexible polypyrole nitrate films. synthetic metals 1996, 79, 121-125. 44. Della Santa, A.; De Rossi, D.; Mazzoldi, A., Performance and work capacity of a polypyrrole conducting polymer linear actuator. Synthetic Metals 1997, 90, 93-100.

PAGE 258

233 45. Lewis, T. W.; Moulton, S. E.; Spinks, G. M.; Wallace, G. G., Optimisationpolypyrrole based actuator. Synthetic metals 1997, 85, 1419-1420. Mazzoldi, of a 46. A.; Degl'Innocenti, C.; Michelucci, M.; De Rossi, D., Actuative properties of polyaniline fibers under electrochemical stimulation. Materials 47. ctrochemical induced ductile-brittle transition in tosylate-doped (pTS) polypyrrole. Synthetic 8. Hutchison, A. S.; Lewis, T. W.; Moulton, S. E.; Spinks, G. M.; Wallace, G. G., etals 9. Smela, E.; Gadegaard, N., Surprising volume change in PPy(DBS): An atomic 0. Callow, M. E.; Callow, J. E., Marine biofouling: a sticky problem. Biologist 2002, 1. Phillippi, A. L.; O'Conner, N. J.; Lewis, A. F.; Kim, Y. K., Surface flocking as a 2. Brady, R. F.; Singer, I. L., Mechanical Factors Favoring Release from Fouling 53. ang, F.-S.; Sun, Z.-X.; Ye, Z.-J., Polyaniline as marine antifouling and corrosion-prevention 54. potential, and challenges. SPIE Press: Bellingham, Wash., 2001; p xiv, 671. 55. letrolytes, SPRINGER-VERLAG BERLIN: Berlin, 1999; Vol. 145, pp 57-122. 56. nolds, J. R., In-Situ Electrochemical Strain Gage Monitoring of Actuation In Conducting Polymers. Journal of the 57. hman, R. H.; Shacklette, L. W., Application of dopant-induced structure-property changes of conducting polymer. Science & Applications of conducting polymers 1990, May, 47-61. Science and Engineering C 1998, 6, 65-72. Murray, P.; Spinks, G. M.; Wallace, G. G.; Burford, R. P., Ele metals 1998, 97, 117-121. 4 Development of polypyrrole-based electromechanical actuators. Synthetic M2000, 113, (1-2), 121-127. 4 force microscopy study. Advanced Materials 1999, 11, (11), 953-+. 5 49, 1-5. 5 possible anti-biofoulant. Aquaculture 2001, 195, (3-4), 225-238. 5 Release Coatings. Biofouling 2000, 15, 7-81. Wang, X.-H.; Li, J.; Zhang, J.-Y.; Sun, Z.-C.; Yu, L.; Jing, X.-B.; W agent. Synthetic Metals 1999, 102, (1-3), 1377-1380. Bar-Cohen, Y., Electroactive polymer (EAP) actuators as artificial muscles : reality Reddinger, J. L.; Reynolds, J. R., Molecular engineering of pi-conjugated polymers. In Radical Polymerisation Polye Bohn, C. C.; Sadki, S.; Brennan, A. B.; Rey Electrochemical Society 2002, 149, (8), E281-E285. Baug

PAGE 259

234 58. Baughman, R. H., Conducting Polymers in Redox Devices and Intelligent Materials Systems. Makromolekulare Chemie-Macromolecular Sym posia 1991, 51, 193-215. 59. am, Washington USA, 2001. f a ynth. Met. 1997, 84, 795-6. 63. Chandrasekhar, P., Conducting polymers, fundamentals and applications : a practical approach. Kluwer Academic: Boston, 1999; p xxxvi, 718. 64. isms of pyrrole electropolymerization. Chemical Society Reviews 2000, 29, (5), 283-293. 65. aria, C., Artificial Muscles from Bilayer Structures. Synthetic Metals 1993, 57, (1), 3713-3717. 66. 11-414. C., Molecular Electronics. lied -L840. 0. Otero, T. F.; Sansinena, J. M., Soft and wet conducting polymers for artificial Bar-Cohen, Y., Electroactive Polymer (EAP) Actuators as Artificial Muscles. Yoseph Bar-Cohen ed.; SPIE Press: Bellingh 60. DellaSanta, A.; DeRossi, D.; Mazzoldi, A., Characterization and modelling oconducting polymer musclelike linear actuator. Smart Materials & Structures 1997, 6, (1), 23-34. 61. Diaz, A. F.; Hall, B., Mechanical-Properties of Electrochemically Prepared Polypyrrole Films. Ibm Journal of Research and Development 1983, 27, (4), 342-347. 62. Kaneko, M.; Fukui, M.; Takashima, W.; Kaneto, K., Electrolyte and strain dependences of chemomechanical deformation of polyaniline film. S Sadki, S.; Schottland, P.; Brodie, N.; Sabouraud, G., The mechan Otero, T. F.; Rodriguez, J.; Angulo, E.; Santam Otero, T. F.; Sansinena, J. M., Artificial Muscles Based on Conducting Polymers. Bioelectrochemistry and Bioenergetics 1995, 38, (2), 4 67. Baughman, R. H.; Shacklette, L. W.; Elsenbaumer, R. L.; Plichta, E. J.; Becht,Micro electromechanical actuators based on conducting polymers. P. I. Lazarev 1991, 68. Kaneto, K.; Kaneko, M.; Takashima, W., Response of chemomechanical deformation in polyaniline film on variety of anions. Japanese Journal of AppPhysics, Part 2 (Letters) 1995, 34, (7A), L837 69. Otero, T. F.; Sansiena, J. M., Bilayer dimensions and movement in artificial muscles. Bioelectrochem. Bioenerg. 1997, 42, 117. 7 muscles. Advanced Materials 1998, 10, (6), 491-+.

PAGE 260

235 71. Madden, J. D.; Cush, R. A.; Kanigan, T. S.; Brenan, C. J.; Hunter, I. W., Encapsulated polypyrrole actuators. Synthetic Metals 1999, 105, (1), 61-64. 2. Bohn, C. C.; Pyo, M. H.; Smela, E.; Reynolds, J. R.; Brennan, A. B., In-situ ls 2003, 15, (4), 916-922. use 5. Lennox, J. A Biofilm Primer. http://www.biofilmsonline.com/cgi76. d, P.; Pfiefer, B.; Rhodes, M.; Bowman, C. N., Effect of polymer surface properties on the reversibility of attachment of Pseudomonas ), they settle. Biofouling 2002, 18, (4), 269-273. 79. marine and related adhesive proteins. Colloids and Surfaces a-Physicochemical and 80. .; Brennan, A. B.; Seegert, C. E.; Gibson, A.; Wilson, L.; Feinberg, A.; Baney, R.; Callow, J. A., Microtopographic cues for g 2002, 1. Wilkerson, W. R.; Seegert, C. A.; Feinberg, A. W.; Zhao, L. C.; Callow, M. E.; mical Society 2001, 221, 253-POLY. 82. and biotechnology. Current Opinion in Biotechnology 2002, 13, (3), 244-248. 7 measurement of conducting polymers on evaportated and electrochemically deposited Au surfaces. Proc. SPIE 2002, 4695, 245-252. 73. Pyo, M.; Bohn, C. C.; Smela, E.; Reynolds, J. R.; Brennan, A. B., Direct strain measurement of polypyrrole actuators controlled by the polymer/gold interface. Chemistry of Materia 74. Dally, J. W.; Riley, W. F., Experimental Stress Analysis. 3rd ed.; College HoEnterprises: 2001; p 639. 7 bin/biofilmsonline/ed_how_primer.html. Last accessed 7/15/04. Pasmore, M.; Tod aeruginosa in the early stages of biofilm development. Biofouling 2002, 18, (165-71. 77. Melo, L. F.; Bott, T. R., Biofouling in water systems. Experimental Thermal and Fluid Science 1997, 14, (4), 375-381. 78. Gray, N. L.; Banta, W. C.; Loeb, G. I., Aquatic biofouling larvae respond to differences in the mechanical properties of the surface on which Yamamoto, H.; Nishida, A.; Ohkawa, K., Wettability and adhesion of Engineering Aspects 1999, 149, (1-3), 553-559. Callow, M. E.; Jennings, A. R settlement of zoospores of the green fouling alga Enteromorpha. Biofoulin18, (3), 237-245. 8 Baney, R.; Brennan, A. B., Bioadhesion studies on microtextured siloxane elastomers. Abstracts of Papers of the American Che de Nys, R.; Steinberg, P. D., Linking marine biology

PAGE 261

236 83. Kim, J.; Shen, W.; Latorre, L.; Kim, C.-J., A micromechanical switch withelectrostatically driven liquid-metal droplet. Sensors and Actuators A: Physic al 2002, 97-98, 672-679. 84. iew 6. Moon, H.; Cho, S. K.; Garrell, R. L.; Kim, C. J., Low voltage electrowetting-on7. Lee, J.; Moon, H.; Fowler, J.; Schoellhammer, T.; Kim, C. J., Electrowetting and s and Actuators tions 9. Peykov, V.; Quinn, A.; Ralston, J., Electrowetting: a model for contact-angle 0. Clarisse, C.; Fadda, E., Characterization of the surface modifications of 1. Liu, M. J.; Tzou, K.; Gregory, R. V., Influence of the Doping Conditions on the 92. ro charge for polypyrrole. Synthetic Metals 1999, 101, (1-3), 442-443. 93. 4. Bartlett, P. N.; Grossel, M. C.; Millan Barrios, E., Electrochemistry and contact 2-148. 95. ; Garnier, F., Stability of Conducting Polythiophene and Derivatives. Journal of the Electrochemical Society 1983, 130, (10), 2042-2044. 96. rs. 1st edition ed.; Dekker: 1986. Chou, T., Geometry-dependent electrostatics near contact lines. Physical RevLetters 2001, 8710, (10), art. no.-106101. 85. Digilov, R., Charge-induced modification of contact angle: The secondary electrocapillary effect. Langmuir 2000, 16, (16), 6719-6723. 8 dielectric. Journal of Applied Physics 2002, 92, (7), 4080-4087. 8 electrowetting-on-dielectric for microscale liquid handling. Sensora-Physical 2002, 95, (2-3), 259-268. 88. Vallet, M.; Berge, B.; Vovelle, L., Electrowetting of water and aqueous soluon poly(ethylene terephthalate) insulating films. Polymer 1996, 37, (12), 2465-2470. 8 saturation. Colloid and Polymer Science 2000, 278, (8), 789-793. 9 conducting poly(3-octylthiophene) films by contact angle measurements. Synthetic Metals 1995, 72, (1), 99-103. 9 Surface Energies of Conducting Polymers. Synthetic Metals 1994, 63, (1), 67-71. Lee, M.-H.; Kim, S. S.; Kang, Y., Potential of ze Habib, M. A., Electrogravimetric Characterization of the Electrochromic Polyaniline Film Surface. Langmuir 1988, 4, (6), 1302-1304. 9 angle measurements on polymer films of [omega]-(3-pyrrolyl)-alkanoic acids in aqueous solution. Journal of Electroanalytical Chemistry 2000, 487, (2), 14Tourillon, G. Skotheim, T. A., Handbook of Conducting Polyme

PAGE 262

237 97. Shenoy, S. L.; Kaya, I.; Erkey, C.; Weiss, R. A., Synthesis of conductive elastomeric foams by an in situ polymerization of pyrrole using supercritical carbon dioxide and etha nol cosolvents. Synthetic Metals 2001, 123, (3), 509-514. ypyrrole. Synthetic Metals 2000, 110, (1), 57-63. 99. polypyrrole composite with a polymeric co-dopant. Synthetic Metals 2000, 114, (3), 347-353. 100. C. Y., Synthesis and characterization of soluble polypyrrole. Synthetic Metals 1997, 84, 101. ns and solution-cast films of conducting polymers. Synthetic Metals 1997, 84, (1-3), 149-150. 102. als 1998, 10, (2), 93-116. ls 1993, 55, (2-3), 1418-1423. 104. by XPS. Journal of Chemical Physics 1984, 80, (1), 544-553. 105. Switching Dual Polymer Electrochromic Devices. Chemical Materials 1998, 10, 06. Somanathan, N.; Wegner, G., Mechanical properties of conducting poly(307. Sotzing, G. A.; Reddinger, J. L.; Katritzky, A. R.; Soloducho, J.; Musgrave, R.; 08. Yamato, H.; Kai, K.; Ohwa, M.; Wernet, W.; Matsumura, M., Mechanical, cta 1997, 42, (16), 2517-2523. 98. Song, K. T.; Lee, J. Y.; Kim, H. D.; Kim, D. Y.; Kim, S. Y.; Kim, C. Y., Solvent effects on the characteristics of soluble pol Lee, Y. H.; Lee, J. Y.; Lee, D. S., A novel conducting soluble Lee, J. Y.; Song, K. T.; Kim, S. Y.; Kim, Y. C.; Kim, D. Y.; Kim (1-3), 137-140. Lee, C. H.; Lee, J. Y.; Lee, H., Solutio McCullough, R. D., The Chemistry of Conducting Polythiophenes. Advanced Materi 103.Otero, T. F.; Rodriguez, J., Polypyrrole Electrogeneration at Different Potentiain Acetonitrile and Acetonitrile Water Solutions. Synthetic Metals Pfluger, P.; Street, G. B., Chemical, electronic, and structural properties ofconducting heterocyclic polymers: a view Sapp, S. A.; Sotzing, G. A.; Reynolds, J. R., High Contrast Ratio and Fast2101-2108. 1 cyclohexyl thiophene) films. Polymer 1996, 37, (10), 1891-1895. 1 Reynolds, J. R., Multiply colored electrochromic carbazole-based polymers. Chemistry of Materials 1997, 9, (7), 1578-1587. 1 electrochemical and optical properties of poly(3,4ethylenedioxythiophene) sulfated poly(beta-hydroxyethers) composite films. Electrochimica A

PAGE 263

238 109.Kovacic, P.; Jones, M. B., Dehydro Coupling of Aromatic Nuclei by CataOxidant Systems Poly(Para-Phenylene). Chemical Reviews 1987, 87, (2), 35379. lyst 7Part a-Polymer Chemistry 1978, 16, (5), 1147-1157. 111. Mechanism of Chain Propagation in Polymerization of Benzene by Aluminum emistry 12. Durham, J. E.; Kovacic, P., Polymerization of Benzene by Aluminum-Chloride. 13. Kovacic, P.; Koch, F. W., Polymerization of Benzene to P-Polyphenyl by Ferric 14. Kovacic, P.; Wu, C., Reaction of Ferric Chloride with Benzene. Journal of 115. ; Zong, K. W.; Schottland, P.; Thompson, B. C.; Thomas, C. A.; Reynolds, J. R., Poly(3,4-ethylenedioxypyrrole): Organic electrochemistry of a 1133. 116. .; Schottland, P.; Reynolds, J. R., Poly(3,4-alkylenedioxypyrrole)s as highly stable aqueous-compatible conducting polymers 117. ng, J., Effect of electrolyte concentration on the properties of the electropolymerized polypyrrole films. Journal of Applied Polymer Science 1997, 18. Thomas, C. A.; Zong, K.; Schottland, P.; Reynolds, J. R., Poly(3,4-2-+. 19. Kang, M. S.; Martin, C. R., Investigations of potential-dependent fluxes of ionic 120. K. E.; Penciner, J., Interfacial Electrochemistry. An Experimental Approach. Addison-Wesley Reading: MA, 1975. 110.Durham, J. E.; McFarland, K. N.; Kovacic, P., Polymerization of Aromatic Nuclei.18. Polymerization of Benzene to Poly-Para-Phenylene by Aluminum Chloride Air-Water. Journal of Polymer Science Engstrom, G. G.; Kovacic, P., Polymerization of Aromatic Nuclei .16. Chloride Cupric Chloride. Journal of Polymer Science Part a-Polymer Ch1977, 15, (10), 2453-2468. 1 Journal of Polymer Science Part C-Polymer Letters 1976, 14, (6), 347-349. 1 Chloride. Journal of Organic Chemistry 1963, 28, (7), 1864-&. 1 Polymer Science 1960, 47, (0149), 45-54. Gaupp, C. L. highly stable electrochromic polymer. Macromolecules 2000, 33, (4), 1132-Thomas, C. A.; Zong, K with biomedical implications. Adv. Mater 2000, 12, (3), 222-225. Li, Y. F.; Ya 65, (13), 2739-2744. 1 alkylenedioxypyrrole)s as highly stable aqueouscompatible conducting polymers,vith biomedical implications. Advanced Materials 2000, 12, (3), 22 1 permeates in gold nanotubule membranes prepared via the template method.Langmuir 2001, 17, (9), 2753-2759. Gileadi, E.; Eisner

PAGE 264

239 121. Zalewska, T.; Lisowska-Oleksiak, A.; Biallozor, S.; Jasulaitiene, V., Polypyrrofilms polymerised on a nickel substrate. Electrochimica Acta 2000, 45, (24), 4031le 4040. 23. Martin, H.; Carro, P.; Creus, A. H.; Gonzalez, S.; Salvarezza, R. C.; Arvia, A. J., isotropic 24. Schmidt, U.; Donten, M.; Osteryoung, J. G., Gold electrocrystallization on carbon lied Electrochemistry 1996, 26, (8), 843-852. 126. and Initial Growth of Pulse-Plated Gold on Crystalline and Amorphous Substrates. Journal of the Electrochemical 127. assembled monolayers. Langmuir 1998, 14, (12), 3287-3297. 128. on-Transport -Ion 32. Maw, S.; Smela, E.; Yoshida, K.; Sommer-Larsen, P.; Stein, R. B., The effects of ctuated 175-184. 122. Finot, M. O.; Braybrook, G. D.; McDermott, M. T., Characterization of electrochemically deposited gold nanocrystals on glassy carbon electrodes. Journal of Electroanalytical Chemistry 1999, 466, (2), 234-241. 1 Growth mode transition involving a potential-dependent isotropic to ansurface atom diffusion change. Gold electrodeposition on HOPG followed by STM. Langmuir 1997, 13, (1), 100-110. 1 and highly oriented pyrolytic graphite from concentrated solutions of LiCl. Journal of the Electrochemical Society 1997, 144, (6), 2013-2021. 125. Li, Y. G.; Chrzanowski, W.; Lasia, A., Nucleation and crystal growth in gold electrodeposition from acid solution .1. Soft gold. Journal of App Holmbom, G.; Jacobson, B. E., Nucleation Society 1988, 135, (11), 2720-2725. Hou, Z. Z.; Abbott, N. L.; Stroeve, P., Electroless gold as a substrate for selfHou, Z. Z.; Dante, S.; Abbott, N. L.; Stroeve, P., Self-assembled monolayers on (111) textured electroless gold. Langmuir 1999, 15, (8), 3011-3014 129. Skaarup, S.; West, K.; Gunaratne, L.; Vidanapathirana, K. P.; Careem, M. A., Determination of ionic carriers in polypyrrole. Solid State Ionics 2000, 136, 577-582. 130. Reynolds, J. R.; Pyo, M.; Qiu, Y. J., Cation and Anion Dominated IDuring Electrochemical Switching of Polypyrrole Controlled by PolymerInteractions. Synthetic Metals 1993, 55, (2-3), 1388-1395. 131. Sonoda, Y.; Takashima, W.; Kaneto, K., Characteristics of soft actuators based onpolypyrrole films. Synthetic Metals 2001, 119, (1-3), 267-268. 1 varying deposition current density on bending behaviour in PPy(DBS)-abending beams. Sensors and Actuators a-Physical 2001, 89, (3),

PAGE 265

240 133. Hohnholz, D.; MacDiarmid, A. G., Line patterning of conducting polymers: New horizons for inexpensive, disposable electronic devices. Synthetic Metals 200121, (1-3), 1 1, 327-1328. s. 35. MacDiarmid, A. G., Synthetic metals: a novel role for organic polymers. Synthetic 136. r nthetic Metals 2003, 137, (1-3), 947-948. 137. and 7), 576-577. Materials 13. r 27. ly Conductive Polyaniline. Chemistry of Materials 2004, 16, (9), 1615-1621. 141. rate larval settlement. Proceedings of the 21st Conference on Engineering in Medicine and 142. E. R.; Cannon, G.; Roberts, D.; Schmidt, A. R.; Sutherland, J. P.; Rittschof, D., The influence of initial surface chemistry on development of the 43. Hillman, A. R.; Efimov, I.; Skompska, M., Dynamics of regioregular conducting 134. MacDiarmid, A. G., "Synthetic metals": a novel role for organic polymerCurrent Applied Physics 2001, 1, (4-5), 269-279. 1 Metals 2001, 125, (1), 11-22. Okuzaki, H.; Ishihara, M.; Ashizawa, S., Characteristics of conducting polymetransistors prepared by line patterning. Sy Hara, S.; Zama, T.; Sewa, S.; Takashima, W.; Kaneto, K., Highly stretchablepowerful polypyrrole linear actuators. Chemistry Letters 2003, 32, ( 138. Bay, L.; West, K.; Sommer-Larsen, P.; Skaarup, S.; Benslimane, M., A conducting polymer artificial muscle with 12% linear strain. Advanced(Weinheim, Germany) 2003, 15, (4), 310-3 139. Qi, B.; Lu, W.; Mattes, B. R., Strain and Energy Efficiency of Polyaniline FibeElectrochemical Actuators in Aqueous Electrolytes. Journal of Physical Chemistry B 2004, 108, (20), 6222-62 140. Lu, W.; Smela, E.; Adams, P.; Zuccarello, G.; Mattes, B. R., Development of Solid-in-Hollow Electrochemical Linear Actuators Using High Rittschof, D.; Costlow, J. D., Surface determination of macroinvertib Biology 1989. Holm fouling community at Beaufort, North Carolina. Journal of Experimental Marine Biology and Ecology 1997, 215, (2), 189-203. 1 polymer electrodes in response to electrochemical stimuli. Faraday Discussions 2002, 121, 423-439. 144. Skompska, M.; Jackson, A.; Hillman, A. R., Evolution from gravimetric to viscoelastic response of poly(3-methylthiophene)-loaded acoustic wave resonators. Physical Chemistry Chemical Physics 2000, 2, (20), 4748-4757

PAGE 266

241 145. Murray, P.; Wallace, G. G.; Spinks, G. M.; Burford, R. P., In-situ mechanical properties of tosylate doped (pTS) polypyrrole. Synthetic Metals 1997, 84, (1-3),847-848. enot, roperties of conducting polymer nanofibers. Synthetic Metals 2003, 135, (1-3), 329-330. 147. ater-instigated changes in elastic modulus of polypyrrole studied by thin rod acoustic wave sensor. Analytica Chimica Acta 148. i, R. A.; De Paoli, M.-A., Chemical preparation of conductive elastomeric blends: polypyrrole/EPDM III. Electrochemical characterization. Journal of 49. Xie, H.; Yan, M.; Jiang, Z., Transition of polypyrrole from electroactive to 50. Turcu, R.; Brie, M.; Leising, G.; Niko, A., Reflection and absorption studies on 51. He, C.; Yang, C.; Li, Y., Chemical synthesis of coral-like nanowires and 52. Song, M.-K.; Kim, Y.-T.; Kim, B.-S.; Kim, J.; Char, K.; Rhee, H.-W., Synthesis 53. Cataldo, F., Iodine: A ring opening polymerization catalyst for tetrahydrofuran. 154. olution. Electrochimica Acta 1999, 44, (19), 3295-3301. 155. ecylbenzenesulfonic acid blends. Synthetic Metals 2004, 143, (2), 237-242. 156. on mer Journal 2002, 38, (12), 2459-2463. 146. Duvail, J. L.; Retho, P.; Godon, C.; Marhic, C.; Louarn, G.; Chauvet, O.; CuS.; Pra, N.; Demoustier-Champagne, S., Physical p Li, P. C. H.; Thompson, M., W 1997, 353, (2-3), 255-262. Zopp Electroanalytical Chemistry 1997, 437, (1-2), 175-182. 1 electroinactive state investigated by use of in situ FTIR spectroscopy. Electrochimica Acta 1997, 42, (15), 2361-2367. 1 polypyrrole films electrochemically prepared with different electrolyte types. Synthetic Metals 1999, 100, (2), 217-221. 1 nanowire networks of conducting polypyrrole. Synthetic Metals 2003, 139, (2)539-545. 1 and characterization of soluble polypyrrole doped with alkylbenzenesulfonic acids. Synthetic Metals 2004, 141, (3), 315-319. 1 European Polymer Journal 1996, 32, (11), 1297-1302. Avgousti, C.; Georgolios, N.; Kyriacou, G.; Ritzoulis, G., The electrochemical oxidation of tetrahydrofuran in sulphuric acid s Tsotra, P.; Friedrich, K., Thermal, mechanical, and electrical properties of epoxy resin/polyaniline-dod Faez, R.; Schuster, R. H.; De Paoli, M. A., A conductive elastomer basedEPDM and polyaniline. II. Effect of the crosslinking method. European Poly

PAGE 267

242 157. Faez, R.; De Paoli, M. A., A conductive rubber based on EPDM and polyanilinI. Doping method effect. European Polymer Journal 2001, 37, (6), 1139-1143. e 58. Faez, R.; Gazotti, W. A.; De Paoli, M. A., An elastomeric conductor based on 1 polyaniline prepared by mechanical mixing. Polymer 1999, 40, (20), 5497-5503.

PAGE 268

BIOGRAPHICAL SKETCH Clayton Bohn was born in Metairie, Louisiana, on June 6, 1974. He lived in the Metairie New Orleans area until he was 5 and then moved to the Mississippi gulf coast where he grew up. After high school he completed a Bachelor of Science in polymer pcience at the University of Southern Mississippi with a minor in chemistry in May 1998. Following graduation, he attended the University of Florida where he earned his Master of Science degree in materials science and engineering in August 2003. The author enjoys Gator football, mountain bike riding, hiking/backpacking, scuba diving, and other outdoor activities. He also enjoys disassembling just about anything he can get his hands on and then trying to put them back together. His mothers favorite day was when he first figured out how to reassemble the things he had taken apart. 243