<%BANNER%>

Electroosomotic Flow in Template Prepared Carbon Nanotubes

Permanent Link: http://ufdc.ufl.edu/UFE0043807/00001

Material Information

Title: Electroosomotic Flow in Template Prepared Carbon Nanotubes
Physical Description: 1 online resource (72 p.)
Language: english
Creator: Hardy, William C
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: carbon -- electroosmotic -- flow -- nanotubes -- template
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The fabrication of carbon nanotube membranes (CNMs) is an area of emerging research. The strength, small dimensions and the remarkable physical properties of carbon nanotubes have made them a material with a wide range of promising applications. Many of these applications will require the use of uniform dense arrays of free-standing, open-ended nanotubes. In the research presented in this thesis, CNMs are prepared by template synthesis. Template synthesis entails chemical vapor deposition of ethylene gas onto a commercial alumina membrane. By utilizing this method carbon nanotubes can be grown with tailored properties, dimensions, and shapes. The objective of this research was to customize the chemical, physical, and structural properties of carbon nanotube membranes and to study the fundamental transport processes occurring at the interface of the carbon nanotube membranes and the electrolyte by monitoring and controlling electroosmotic flow. In Chapter 2, we describe the utilization of the template method to fabricate carbon nanotube membranes. This method involves the synthesis of carbon nanotube membranes by chemical vapor deposition within the pores of high pore density alumina membrane templates. We thoroughly explain the advantage of using template synthesis to customize nanostructures with specific characteristics. We demonstrate that by controlling the parameters of chemical vapor deposition, we can tune the properties of the carbon nanotube membrane. Lastly, we discuss the methods of characterization for the CNM. In chapter 3, we describe EOF in carbon nanotube array membranes. EOF was driven across the CNM by placing the membrane between two electrolyte solutions and placing an electrode in each solution to pass a constant current through the nanochannels of the membrane. EOF was monitored by measuring the flux of a probe molecule across the membrane. The CNMs have an anionic surface charge, which results in EOF in the direction of cation migration across the membrane. EOF was used to control the flux of the probe molecule. When EOF was driven in the direction of diffusion there was an increase in the flux of the probe molecule across the membrane. Reversing EOF decreases flux, and at high reverse EOF velocities the transport is halted and eventually produces transport in the opposite direction. EOF through the CNM was modified by changing the pH, ionic strength and the current density applied to the membrane. These changes resulted in the exhibited control of EOF velocity and direction. Results from this work will lay a solid foundation for the further development of selective molecular transport in CNMs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by William C Hardy.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Martin, Charles R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043807:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043807/00001

Material Information

Title: Electroosomotic Flow in Template Prepared Carbon Nanotubes
Physical Description: 1 online resource (72 p.)
Language: english
Creator: Hardy, William C
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: carbon -- electroosmotic -- flow -- nanotubes -- template
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The fabrication of carbon nanotube membranes (CNMs) is an area of emerging research. The strength, small dimensions and the remarkable physical properties of carbon nanotubes have made them a material with a wide range of promising applications. Many of these applications will require the use of uniform dense arrays of free-standing, open-ended nanotubes. In the research presented in this thesis, CNMs are prepared by template synthesis. Template synthesis entails chemical vapor deposition of ethylene gas onto a commercial alumina membrane. By utilizing this method carbon nanotubes can be grown with tailored properties, dimensions, and shapes. The objective of this research was to customize the chemical, physical, and structural properties of carbon nanotube membranes and to study the fundamental transport processes occurring at the interface of the carbon nanotube membranes and the electrolyte by monitoring and controlling electroosmotic flow. In Chapter 2, we describe the utilization of the template method to fabricate carbon nanotube membranes. This method involves the synthesis of carbon nanotube membranes by chemical vapor deposition within the pores of high pore density alumina membrane templates. We thoroughly explain the advantage of using template synthesis to customize nanostructures with specific characteristics. We demonstrate that by controlling the parameters of chemical vapor deposition, we can tune the properties of the carbon nanotube membrane. Lastly, we discuss the methods of characterization for the CNM. In chapter 3, we describe EOF in carbon nanotube array membranes. EOF was driven across the CNM by placing the membrane between two electrolyte solutions and placing an electrode in each solution to pass a constant current through the nanochannels of the membrane. EOF was monitored by measuring the flux of a probe molecule across the membrane. The CNMs have an anionic surface charge, which results in EOF in the direction of cation migration across the membrane. EOF was used to control the flux of the probe molecule. When EOF was driven in the direction of diffusion there was an increase in the flux of the probe molecule across the membrane. Reversing EOF decreases flux, and at high reverse EOF velocities the transport is halted and eventually produces transport in the opposite direction. EOF through the CNM was modified by changing the pH, ionic strength and the current density applied to the membrane. These changes resulted in the exhibited control of EOF velocity and direction. Results from this work will lay a solid foundation for the further development of selective molecular transport in CNMs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by William C Hardy.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Martin, Charles R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043807:00001


This item has the following downloads:


Full Text

PAGE 1

1 ELECTRO OSOMOTIC FLOW IN TEMPLATE PREPARED CARBON NANOTUBES By WILLIAM C. HARDY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 201 1

PAGE 2

2 201 1 William Cornelius Hardy

PAGE 3

3 To my parents, my brother, my lab associates and God Almighty

PAGE 4

4 ACKNOWLEDGMENTS My graduate years at the University of Florida have had mentorship from numerous outstanding individuals both from within the university and outside of it. It is to these individuals, who m have invested time, support and insight to my studies, which my heartfelt gratitude and thanks go out to, for without their help, this thesis would not have been possible. I would like to thank Dr. Charles R. Marti n for his patience, guidance, support and encouragement throughout my graduate career at the University of Florida. He has been an excellent mentor by providing scientific discussion suggestions, knowledge about my research and constructive criticism about how to more effective ly communicate science. His dedication from the initial to the final stage has enabled me to develop a s ubstantial understanding of electrochemistry and fluid dynamics. The Martin group members have been very supportive. I am especially gratef ul to Jillian Perry, Lloyd Horne, Greg Bishop, Hitomi Mukaibo Dooho Park Kaan Kececi, Pu Jing, Fonda Tongay, and Otoyne Braide, for providing insightful discussions about my experiments and for valuable training in surface modification, analytical detection techniques and instrumentation. Li Zhao Lindsey Sexton and Marcos Lopez shared their research experiences and provided helpful experimental method suggestions. I want to thank Dr David Wei, Dr. Rick Yost Dr. Danyell Wilson, and Dr. Anne Donnelly for their valued advice on my research life, and career. I am also grateful to Karen Kelly and Kim from the ICBR for their help with SEM and TEM, to Eric Lambers from the MAIC for his help with XPS, and for their help with the four point probe measurements

PAGE 5

5 Finally, my gratitude goes to my family and friends who provided enor mous help and moral support during my graduate career. My parents Lathen and Katherine Hardy Jr., and my brother Lathen Hardy, III deserve most of the credit for my success as they were constant sources of encouragement and guidance before and during my gr aduate career. I want to also thank members of the Black Graduate Student Organization and the Office of Min ority Programs especially Dr. Lawerence Alexander, Sarah Cuningham and Janet Broiles f or providing a very friendly, atmosphere and helping me mak e the transition from Jackson, Mississippi to Gainesville, Florida smooth. Lastly, I offer my regards and blessings to all of those who supported me in any resp ect during the completion of this project.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTR ODUCTION AND BACKGROUND ................................ ................................ 14 General Concept of a Battery ................................ ................................ ................. 15 Battery Separators ................................ ................................ ................................ .. 18 Carbon Nan otubes ................................ ................................ ................................ .. 20 Carbon Properties ................................ ................................ ............................ 21 Carbon Nanotube Synthesis ................................ ................................ ............. 22 Template Synthesis Method ................................ ................................ ............. 23 Anodic Aluminum Oxide Template Membranes ................................ ............... 25 Electrokinetics ................................ ................................ ................................ .. 27 Electrical Double Layer ................................ ................................ .................... 28 Electroosmotic Flow ................................ ................................ ......................... 31 Electroo smotic flow versus pressure driven flow ................................ ........ 32 Modulation of electroosmotic flow ................................ .............................. 34 Thesis Overview ................................ ................................ ................................ ..... 35 2 SYNTHESIS OF CARBON NANOTUBE MEMBRANES BY THE TEMPLATE METHOD ................................ ................................ ................................ ................ 37 Introduction ................................ ................................ ................................ ............. 37 Experimental ................................ ................................ ................................ ........... 39 Materials ................................ ................................ ................................ ........... 39 Preparation of Carbon Nanotubes ................................ ................................ .... 39 Characterization of Carbon Nanotube Membranes ................................ .......... 40 X Ray Photoelectric Spectroscopy ................................ ................................ .. 40 Electron microscopy ................................ ................................ ................... 41 Sheet resistance ................................ ................................ ........................ 41 Results and Discussion ................................ ................................ ........................... 43 Effect of CVD Rate on Carbon Nanotube Growth ................................ ............ 44 Effect of Temperature on CVD Rate ................................ ................................ 46 Conclusion ................................ ................................ ................................ .............. 47

PAGE 7

7 3 ELECTROOSMOTIC F LOW IN CARBON NANOTUBE MEMBRANES PREPARED BY THE TEMPLATE SYNTHESIS METHOD ................................ ..... 49 Introduction ................................ ................................ ................................ ............. 49 Experimental ................................ ................................ ................................ ........... 50 Materials ................................ ................................ ................................ ........... 50 Preparation of CNM ................................ ................................ .......................... 51 EOF Measurements ................................ ................................ ......................... 51 Results and Discussion ................................ ................................ ........................... 53 Effect of Applied Current Density on Permeate Flux ................................ ........ 53 Calculation of the Electroosmotic Velocity ................................ ........................ 55 Effect of ionic strength of electrolyte on v eof ................................ ............... 57 Effect of solution pH on v eof ................................ ................................ ........ 59 Conclusions ................................ ................................ ................................ ............ 61 4 DISCUSSIONS AND CONCLUSIONS ................................ ................................ ... 63 LIST OF REFERENCES ................................ ................................ ............................... 66 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 72

PAGE 8

8 LIST OF TABLES Table page 1 1 Classification of electrokinetic phenomena ................................ ......................... 28 3 1 Effect of applied current density on the flux, enhancement factor, peclet number and the electroosmotic flow velocity of phenol through the carbon nanotube membrane. ................................ ................................ ......................... 55

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Schematic representation of a general electrochemical battery cell Li Ion batteries ................................ ................................ ................................ .............. 16 1 2 Schematic representation of the charge discharge process of a lithium ion cell ................................ ................................ ................................ ...................... 17 1 3 The general reactions taking place at the electrodes and the overall reaction for a typical Li ion cell ................................ ................................ ......................... 18 1 4 Structure of Lithium Ion battery with battery separator ................................ ....... 19 1 5 Chemical vapor deposition of carbon onto the anodic alumina membrane ........ 24 1 6 Electron micrograp hs of 200nm Whatman Anodisc TM alumina template membrane. ................................ ................................ ................................ ......... 25 1 7 Illustration of anodic alumina membrane with branched pores ........................... 26 1 8 Alumina Fabrication ................................ ................................ ............................ 26 1 9 Illustration of the Electrical Double Layer ................................ ........................... 29 1 10 Schematic representation of electro osmotic f low ................................ ............... 32 1 11 Comparison of fluid velocity profiles. ................................ ................................ .. 33 2 1 Basic setup for sheet resistance measurements ................................ ................ 42 2 2 Electron micrograph of t he cross section of carbon nanotube membrane .......... 44 2 3 Electron micrograph of individual carbon nanotubes ................................ .......... 44 2 4 Plot of % carbon on the alumina template vs.CVD time ................................ ..... 45 2 5 Plot of % carbon vs. CVD temperature ................................ ............................... 47 3 1 Illustration of the carbon nanotube membrane assembly ................................ ... 52 3 2 Schematic of permeation cell used in EOF experiments ................................ .... 53 3 3 Plot of phenol transport vs. time at different current densities ............................ 54

PAGE 10

10 3 4 Plot of electroosmotic flow velocity vs. ionic strength of permeate and feed solutions ................................ ................................ ................................ ............ 58 3 5 Plot of electroosmotic flow velocity vs. pH of permeate and feed solutions ....... 60

PAGE 11

11 LIST OF ABBREVIATION S CNM c arbon n anopore m embrane CNT c arbon n anotubes EOF e lectroosmotic f low LIB l ithium i on b atteries SEM scanning electron microscopy TEM transmission electron microscopy XPS x ray p hotoelectric s pectroscopy

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the U niversity of Florida i n Partial Fulfillment of the Requirements for the Degree of Master of Science ELECTROOSOMOTIC FLOW IN TEMPLATE PREPARED CARBON NANOTUBES By William Hardy December 2011 Chair: Charles R. Martin Major Chemistry The fabrication of carbon nanotube membranes (CNMs) is an area of emerging research. The strength, small dimensions and the remarkable physical properties of carbon nanotubes have made them a material with a wide range of promising applications. Many of t hese applications will require the use of uniform dense arrays of free standing, open ended nanotubes. In the research presented in this thesis, CNMs are prepared by template synthesis. Template synthesis entails chemical vapor deposition of ethylene gas o nto a commercial alumina membrane. By utilizing this method carbon nanotubes can be grown with tailored properties, dimensions, and shapes. The objective of this research was to customize the chemical, physical, and structural properties of carbon nanotube membranes and to study the fundamental transport processes occurring at the interface of the carbon nanotube membranes and the electrolyte by monitoring and controlling electroosmotic flow. In Chapter 2, we describe the utilization of the template meth od to fabricate carbon nanotube membranes. This method involves the synthesis of carbon nanotube membranes by chemical vapor deposition within the pores of high pore density alumina

PAGE 13

13 membrane templates. We thoroughly explain the advantage of using template synthesis to customize nanostructures with specif ic characteristic s We demonstrate that by controlling the parameters of chemical vapor deposition we can tune the properties of the carbon nanotube membrane. Lastly, we discuss the methods of characterizat ion for the CNM. In C hapter 3, we describe EOF in carbon nanotube array membranes. EOF was driven across the CNM by placing the membrane between two electrolyte solutions and placing an electrode in each solution to pass a constant current through the nano channels of the membrane. EOF was monitored by measuring the flux of a probe molecule across the membrane. The CNMs have an anionic surface charge, which results in EOF in the direction of cation migration across the membrane. EOF was used to control the f lux of the probe molecule. When EOF was driven in the direction of diffusion there was an increase in the flux of the probe molecule across the membrane. Reversing EOF decreases flux and at high reverse EOF veloci ties the transport is halted and eventuall y produces transport in the opposite direction. EOF through the CNM was modified by changing the pH, ionic strength and the current density applied to the membrane. These changes resulted in the exhibited control of EOF velocity and direction. Results fr om this work will lay a solid foundation for the further development of selective molecular transport in CNMs.

PAGE 14

14 CHAPTER 1 INTRODUCTION AND B ACKGROUND Energy storage is more vital today than at any other time in human history. Applications such as low emission hybrid electric vehicles, portable electronic devices (cell phones, laptop computers etc.), and the need for power storage from renewable sources put even more stringent requiremen ts on energy storage devices. 1 In all of the applications mentioned above high performance batteries can be applied. Rechargeable lithium ion batteries have become the dominant power source choice for the fulfillment of such energy demands. The first prototypes of l i thium batteries used Li metal as the anode material. The motivation for producing a battery technology based on Li metal as an an ode was based on the fact that l i thium is the most electropositive ( 3.04 V ver sus standard hydrogen electrode) and the lightest (equivalent weight M = 6.94 g mol 1 specific gravity 0.53 g cm 3 ) metal, which promotes the design of storage systems with very high energy density. 1 2 These first lithium based batteries were unsuccessful because Li metal is highly unstable and led to many safety concerns Thus, Sony in 1991 decided to use lithium ion moving electrodes instead of lithium metal in their rechargeable battery design. Although the Li ion battery is slightly lower in energy density it is much safer and offers considerable environmental value. ( Oxidized lithium ions are non toxic. They can easily be extracted from a battery, neutralized, and used as feed stock for new Li i on batteries. ) For these pur poses intensive research efforts have been directed at creating both efficient and safe lithium ion cells by many universities, government agencies, and battery companies. However, some problems still have to be solved before the

PAGE 15

15 complete success of the li thium ion battery is achieved. Among the concerns are long term stability of cell components, heat control of the cell, higher quality product development, and better cost performance of materials. General Concept of a Battery A battery consists of a group of interconnected electrochemical cells. The quantity of cells grouped together and the method by which they are connected depends on the specific application for which the battery is designed. An electrochemical cell is composed of a negative electrode ( anode), a positive electrode (cathode) and an electrolyte. These compon ents are the essential b uilding blocks of any battery. 4 5 Electric energy is released or discharged from a battery when it is connected to a work load. The anode reacts with the ions in the electrolyte and releases electrons through an oxidation reaction while the cathode reacts with the ions in the electrolyte and incorporates electrons through a reduction reaction. When these reactions are reversible, the battery is a rechargeable batt ery, meaning the electric energy is restored or charged when the battery is connected to an external power supply and the opposite reactions are induced. The energy density of an electrode is defined by the amount of electrons that can be stored per unit volume or weight. 1 4 The anode of a lithium ion cell is usually a solid reductant and the cathode is usually a solid oxidant. A typical lithium ion cell used today is commonly composed of a carbon based material as the anode and a lithiated transition me tal oxide as the cathode. During the charge cycle of the cell, t he cathode material releases Li ions to the electrolyte (deintercalation) and electrons are removed from the cathode by an external field a nd transferred to the anode. Li ions then rush to th e carbon based anode to

PAGE 16

16 compensate for the charge (intercalation). The chemical reactions at the anode and cathode of a Li ion cell are reversible. On discharge, the anode supplies interca lated Li ions to the electrolyte and electrons to the external circ uit, the cathode is an electronically conducting medium into which the Li ions intercalate from the electrolyte and compensate for the charge of the electrons from the external circuit. Since both anode and cathode are hosts for the reversible intercalatio n/deintercalation of Li ions 4 5 Figure 1 1 Schematic representation of a general electrochemical battery cell Li Ion b atteries Although different applications have different requirements, the most common carbon material for the anode is either graphite coke type, or a combination of both. Common cathode materials include LiMn 2 O 4 LiCoO 2 and LiNiO 2 Electrolyte can be

PAGE 17

17 either liquid or solid. Liquid electrolyte is us ually a non aqueous solution of a lithium salt and various solvents including different types of esters, ethers and carbonat es. A schematic of a typical Li ion cell is shown in Fig ure 1 2 5 and the general reactions taking place at the electrode s in a Li io n cell are depicted in Figure 1 3. Figure 1 2 Schematic representation of the charge discharge process of a lithium ion cell The lithium ion battery has shown superior characteristics in both energy density and cell voltage when compared to other battery systems. These advantages have given lithium ion batteries success in the small application market and a great deal of interest in possible large scale applications.

PAGE 18

18 Figure 1 3. The general reactions taking place at the electrodes and the over all reaction for a typical Li ion cell Battery Separators As stated earlier rechargeable lithium ion batteries have achieved a great deal of commercial success and established domi nance in small cell applications Their high energy density and intrinsi c discharge voltage also make them desirable candidates for larger applications, including electronic vehicles and satellites. 6 7 Such applications require a higher performance standard for lithium ion batteries. An ideal battery would be reliable, robust, and inexpensive. It would supply a constant power supply under heavy loads and at high and low temperatures. It would be safe under cr itical environmental conditions and b e eas y to mass produce. 6 8 One way to universally improve lithium ion battery performance is through the use of advanced battery separators. A battery separator is a porous membrane placed between the anode and cathode of a battery. The main function of a separator is to keep the po sitive and negative electrodes apart in order to prevent electrical short

PAGE 19

19 circuits, while also allowing the transport of ionic charge carriers which are needed to complete the circuit during the passage of current in an electrochemical cell 9 10 Figure 1 4 Structure of L ithium Ion b attery with battery separator Separators have been manufactured using a variety of different materials including cellulosic papers, cellophane, nonwoven fabrics, foams, ion exchange membranes, and micro porous flat sheet memb ranes made from polymeric materials. As batteries have become more complex, separator function has also become more demanding, sophisticated and multifaceted. 10 11 There are many properties that are required for separators used in electrochemical batteries Battery separators shoul d be good electronic insulators, have the ability to conduct ions by either ionic condu ction or by soaking electrolyte, and minimize any processes or reactions that adversely affect the electrochemical ener gy efficiency of the bat teries. 9 Relative to the amount of research on electrode materials and electrolytes, very little attention has been aimed at characterizing and developing new separators. Likewise, there has been modest interest given to separators in publications reviewin g battery technologies. In recent years there ha ve been numerous reviews on battery cell

PAGE 20

20 fabrication, their performance, and application in real life applications, but very few reviews have discussed separators in detail. 9 19 The majority of the separators currently used in batteries were originally designed for applications in other existing technologies. They were usually not developed specifically for the batteries in which they are utilized and thus there is a need for optimization. 8, 2 0 21 The research results presented in this thesis have been aimed at studying the transport processes occurring at the interface of a porous carbon nanotube membrane and an electrolyte through different electrochemical approaches. The carbon nanotube membrane is synthesiz ed using the template method. The advantage of this approach is the structural, physical, and chemical properties of the carbon nanotube membrane can be modified through stringent control of the experimental parameter s of the chemical vapor deposition of c arbon. The flexibility of the template method will enable these membranes to exhibit optimized permeability and selectivity. The results from this research should also be helpful in the practical application of using a carbon nanotube membrane as a possib le new material for advance battery separators. Carbon Nanotubes The discovery of carbon nanotubes was credited to Sumio Iijima in 1991. Iijima was studying the structure of buckmisterfullerenes created by the arc discharge process, when he noticed an ext remely thin needle like material observed under an electron microscope. The material proved to be graphite in structure and was a few hundred nanometers in length so he called the material, carbon nanotubes. 22 23 Carbon nanotubes can generally be describe d as a graphite sheet rolled up into a nanoscale tube. There are several types of nanotubes; but they can be divided in two

PAGE 21

21 main categories: single wall ed carbon nanotubes (SWCNTs), and multi wall carbon nanotubes (MWCNTs). Single walled carbon nanotubes (SWCNT) typically have a diameter of about 1 nanometer, with a tube length that can be many millions of times longer. The structure of a SWNT can be conceptualized as one atom thick layer of graphite called graphene wrapped i nto a seamless cylinder. SWCNTs are often considered to be structurally perfect because of the high structural order that they possess. 22 23 Multi walled carbon nanotubes (MWCNT) consist of multiple rolled layers of concentric tubes of graphite. MWCNT ca n be conceptualized using the Parchment model. In the P archment model, a single sheet of graphite is rolled around itself, resembling a scroll of parchment or a rolled newspaper. The interlayer distance in multi walled nanotubes is close to the distance be tween graphene layers in graphite, which is approximately 3.4 MWCNTs are generally prepared with different levels of graphitization which means MWCNTs are typically less ordered than SWCNTs. 22 Carbon Properties The physical properties of carbon nanotubes are still being discovered and disputed. CNTs have a broad range of electric, thermal, and structural properties that change depending on the dimensions and unique structure of the nanotube. Many studies have confirmed that the sp carbon carbon bonds in CNTs supply amazing mechanical properties to the structure of the nanotube. The stiffness of a material is measured in terms of its Young's modulus, which is the rate of change of stress with applied strain. The Young's modulus of CNTs can be as high as 1000 GPa which is approximately 5x higher than steel. The tensile strength, or breaking strain of CNTs can be up to 63 GPa, which is about 50x higher than steel. Research has also

PAGE 22

22 indicated that CNTs are considered to be great conductors of heat an d may someday find applications as miniature heat conduits in a h ost of devices and materials. 24 27 The electronic properties of carbon nanotubes are also extraordinary. The fact that CNTs can be metallic or semiconducting depending on their structure is particularly remarkable Thus, some CNTs have conductivities higher than that of copper, while others behave more like silicon. There is great interest in the possibility of constructing nanoscale electronic devices from CNTs, and some progress is being ma de in this area. However, in order to construct a useful device we need to arrange many thousands of CNTs in a defined pattern. 28 30 Carbon Nanotube Synthesis Carbon nanotubes are commonly produced by three main techniques, laser ablation, arc discharge a nd chemical vapor deposition. In the laser ablation technique, a high power laser beam impinges on a volume of carbon, while simultaneously an inert carrier gas is bled though the chamber. Nanotubes subsequently develop on the cooler surfaces of the reacto r as the vaporized carbon condenses. In the arc discharge technique, a direct electrical current is applied between two carbon electrodes in a gas atmosphere. As the high energy current passes between the carbon electrodes a carbon vapor is produced. Carbo n nanotubes can self assemble from the resulting vapor with or without a catalyst. 22, 31 35 Both of these techniques can produce high quality nanotubes but producing large quantities of uniform high purity carbon nanotubes is still an arduous task. Using these techniques, to grow large amounts of CNTs typically produces low quality masses of CNTs that are randomly mixed with a variety of by products. High purity SWCNTs are produced on a small scale but they are moderately expensive. MWCNTs are

PAGE 23

23 synthesized on the large scale but size uniformity is difficult to control using these techniques. 36 37 Chemical Vapor Deposition (CVD) is a process in which precursor gases typically diluted in carrier gases are introduced into a reaction chamber at ambient tempera tures. As the gas comes into contact with a heated substrate, a reaction takes place causing the decomposition of the source gas which forms a solid phase. This solid phase is preferentially deposited onto the substrate. The substrate temperature is a crit ical parameter, which can influence what reactions will take place. 38 CVD has several advantages over other methods. First, the CVD method allows opportunities to easily control the growth of CNTs which is an option not readily available in the laser abl ation or arc discharge methods. Second, the CVD is a method that is easily scaled up for mass production of CNTs. Third, because in CVD the gas flows in a certain direction the nanotubes that are grown are anchored to the substrate material at one end, whi ch allows for directional, aligned growth controlled strategies. These strategies can include controlling gas flow rate, van der Waals interactions, and electric fields. Lastly the CVD method is easily manipulated and is highly compatible with the template synthesis method for fabricating controlled arrays of nanostructures. 36, 39 47 Template Synthesis M ethod The Martin laboratory and others have illustrated the versatility of this method by growing carbon nanotubes by CVD along the outer surface and inner pore walls of nanoporous templates. This method has resulted in films of high density, free standing defe ct free arrays of carbon nanot ubes embedded in the pores of the template. 4 8 5 2

PAGE 24

24 Figure 1 5 Chemical v apor d eposition of c arbon onto the a nodic a lumina m embrane The template synthesis method is a technique for preparing one dimension al nanostructures. The Martin group pioneered this method in the late 1980s and early 1990s, 47, 5 3 5 4 although the method originated with Possin 5 5 who applied it for the fabrication of metal nanowires. In the template method a material is deposited into the pores of a mask called the template. Typically this deposition results in mono dispersed arrays of dense parallel nanostructures in the shapes and dimensions of the predisposed template pores. By controlling the depositio n conditions, one can vary the wall thickness and thus produce hollow or solid nanostructures. The template synthesis method is a universal technique for producing nanostructures with various geometries and dimensions composed of a variety of different ma terials 46 47 5 6 59 It has been use d to produce nanowires, nanotubes, nanopillars, and even drug molecular vehicles. The broad utility of this method is illustrated by considering the examples of fabrication of biodegradable nano test tubes for drug deliv ery 6 0 and conducting polymer nanostructures for energy storage 6 1

PAGE 25

25 Anodic Aluminum Oxide Template Membranes Anodic aluminum oxide (AAO) nanoporous membranes (Figure 1 5 ) are used as the templates for the carbon nanotube membranes discussed in this thesis 51 6 2 AAO membranes are inorganic films composed of a high purity alumina matrix manufactured electrochemically. AAO membranes make remarkable templates for this project for many reasons. AAO membranes are thermally and chemically stable at high temperatur es and in many environments. They are mechanically rigid, and can be easily dissolved in high or low pHs. Figur e 1 6. Electron m icrograph s of 200nm Whatman Anodis c TM a lumina template m embrane (A) Unbranched face with ~250 nm pores. (B) Branc hed face with 100 200 nm pores The AAO membranes involved in this study were purchased commercially from Whatman Anodisc TM possess a very high pore density (1.8x10 9 pores/cm 2 ) of parallel, monodisperse, virtual ly cylindrical pores. Anodisc TM membranes are manufactured in 3 nominal pore size diameters: 0.02 m, 0.1 m, and 0.2 m. The pores are called nominal because their diameters are actually much larger (20 0 250nm) through about ~98% of the membrane thicknes s and changes to a much smaller diameter for the last 2% of the

PAGE 26

26 membrane thickness. The original large diameter pores branch into several smaller pores for the last 2% of the membrane thickness. Figure 1 7 Illustration of a nodic alumina m embrane with branched pores This branching is a result of the method by which the manufacturer fabricates these membranes. Aluminum oxide (alumina) is a product of the oxidation of aluminum foil in an acid at a particular voltage. The a lumina grows as a porous honey c omb film attached to the aluminum foil. Once the alumina film has reached a desired thickness the oxidation is discontinued and the alumina is separated from the aluminum. Figure 1 8 Alumina Fabrication

PAGE 27

27 There are two methods generally used to separate porous alumina film from aluminum foil. These methods are chemical etching and voltage step reduction separation. In the chemical etching technique, anodization is discontinued once the porous film reaches the desired thickness. The aluminum template is t hen placed in a chemical solution where the aluminum and the alumina barrier layer is chemically etched away producing a free standing porous alumina membrane. Alternatively, in voltage reduction separation once the membrane has nearly reached the desired thickness the voltage is slowly reduced at periodic intervals. As the voltage is reduced the pores growing in the film branch off into smaller pores with thinner alumina walls. Anodization is discontinued once the alumina walls are thin enough to quickl y dissolve in an acidic solution. The layer of the alumina with the narrowest walls will dissolve most quickly, leaving a free standing defect free porous film separated from the aluminum foil. The manufacturer uses the voltage step reduction separation me thod to fabricate the alumina membranes which e xplains the branched pores. In this study we chose to use Whatman Anodisc TM particularly because they are readily available in large quantities, inexpensive, easy to handle, and are of consistent quality. El ectrokinetics Electrokinetics defines those processes where a charged fluid solid interface reacts to external influences. Electrokinetics is often catego rized into four distinct areas, as shown in Table (1 1). Streaming potential and sedimentation potential are concerned with the generation of an electric field due to forced fluid or solid motions, where charged fluid solid interfaces are present. These phenomena are not the focus of this research and the

PAGE 28

28 intere sted reader should seek elsewhere for more in depth treatment of these occurrences. 63 64 The remaining two electrokinetic phenomena, electroosmosis and electrophoresis, are of significance to the research herein but only electroosmosis will be discussed in detail. Table 1 1 Classification of electrokinetic phenomena The idea of electroosmosis dates back to the research of F.F. Reuss who disc overed the phenomena in 1809 6 5 Electroosmotic flow operates by moving a liquid past a stationary surface by an electric field. To truly understand and appreciate the electroosmotic process we must discuss the electrical double layer. Electrical Double Layer When an aqueous medium comes into contact with solid surfaces, mechanisms such as ionization, ion adsorption and ion dissolution typically take place. These mechanisms result in the surface acquiring a charge. Depending on the polarity of the surface charge, a predominantly positive or negative ch arge near the surface will exist. Ions of opposite charge to the wall, (counter ions), are attracted to the surface of the wall and form a layer several Angstroms thick called the compact layer (also referred to as the Stern layer). These counter ions in the compact layer are stationary and do not move relative to the surface. However, beyond the compact layer is another layer of ions, called the diffuse layer (also called the Gouy Chapman layer). The diffuse Name Movement Description Action Electroosmosis Liquid moves relative to a stationary liquid Applied electric used to induce movement Electrophoresis Charged surface moves relative to a stationary liquid Applied electric used to induce movement Streaming potential Liquid moves relative to a stationary charged surface Movement used to create an electric field Sedimentation potential Charged surface moves relative to a stationary liquid Movement used to create an electric field

PAGE 29

29 layer is comprised generally of counter ions bu t co ions (ions of the same charge as the wall) are also present in this region. The diffuse layer is thicker than that of the compact layer and the ions in the diffuse layer are mobile. In general the thickness of the diffuse layer will be approximately 3 to 5 times that of the compact layer. The imaginary interface between the compact and diffuse layer is called the shear plane. The compact layer and the diffuse layer together are known as the electrical double layer (EDL) as shown in Figure 1 9 Figure 1 9 Illustration of the Electrical Double Layer One of the major characteristics of the electrical double layer is the zeta potential ( ). The zeta potential is the measure of the electrical potential at the plane of shear. It is characterized by E quation 1 1 (1 1)

PAGE 30

30 Where k is the Boltzmann constant, T is the temperature, z is the counter ion charge, e is the electrolyte charge, is the permittivity of water, 1 is the debye length and is the surface charge density at the plane of shear between the capillary wall and the electrolyte. The zeta potential is used as a boundary condition for the potential at the point of zero velocity and is thus essential in calculations of electro kineti c flows. E quation 1 1 simplifies to E quation 1 2 when solutions with re latively small zeta potentials are implemented. (1 2 ) When the electrolyte is symmetrical and is at 25 o C, 1 is related to the concentration of the electrolyte by E q uation 1 3 where z is the charge of the electrolyte and c the concentration of the electrolyte. (1 3 ) The last important feature of the EDL that will be discussed in this thesis is the length of the EDL, which is defined by the Debye length, or Debye shielding distance, denoted by D The Debye length is an estimate of the approximate thickness of the EDL. It is derived from the Poisson equation and is shown symbolically as E quation 1 4 where k B permittivity of the fluid, T is the absolute temperature of the fluid; e is the elementary charge and n is the ionic strength of the fluid. (1 4 ) From Equations 1 1 though 1 4, we observe that the size of the electrical double layer increases as the ionic strength of the electrolyte decreases. We also recognize the

PAGE 31

31 relationship between the zeta potential and the concentration of the electrolyte. Thus we can make the following con clusions: First, the electrical double layer becomes thicker in lower concentration electrolyte solutions ; second the zeta potential in an aqueous environment increases with increasing surface charge and decreasing electrolyte concentration. Thus the size of the electrical double layer is a very important parameter when determining the influence of electroosmotic flow in fluidic channels Electroosmotic Flow When an electric field is applied across the length of a fluidic channel, the charged molecules of a fluid will experience an electrostatic force. In the bulk fluid where the concentration of counter ions and co ions are equal, there is no induced movement from the electric field. However, the mobile counter ions in the diffuse layer experience that sam e electrostatic force which causes them to migrate toward the oppositely charged electrode. Due to viscous interaction, the mobile ions in the diffuse layer drag the fluid in the bulk layer. In this way a fluid flow is created which is called e lectroosmoti c flow (EOF). The velocity v e of of the EOF is described by the Helmholtz Smochulowski equation which is describe in Equation 1 5 (1 5 ) where t The electroosmotic mobility e of is given by the Equation 1 6. (1 6 )

PAGE 32

32 Figure 1 10 Schematic r epresentation of e lectro o smotic f low Thus, the electroosmotic volume flow q e of through the channel is defined by Equation 1 7. (1 7 ) Here, A is the cross sectional area of the channel, L its length, and U is the potential difference along the channel length. Equations, 1 5, 1 6 and 1 7 show that the EOF is strongly dependent on the zeta potential. The zeta potential is not a constant ; it is influenced by several factors which will be discussed in further detail in a later section of this thesis. Electroosmotic f low v ersus p ressure d rive n f low To truly understand and appreciate the transport properties of EOF, there is a need to compare EOF to other flow transport mechanisms such as pressure driven or hydraulic flow systems. One of the distinguishing characteristics among fluid flow systems are fluid flow profiles. Typically in EOF the fluid flow profile across the cross

PAGE 33

33 section of a pore is flat or plug like. This is because the force that induces EOF originates in the diffus e layer of the EDL, which is very close to the pore wall. In contrast, in a pressure driven flow system the fluid flow profile is parabolic or laminar. The wall of the pore induces a frictional force upon the flow causing the velocity of the flow to be zer o at the surface of the pore wall and to be slowed near the wall. The fluid velocity increases as the flow moves to the center of the fluidic channel and reaches maximum in the center of the pore, where fluid flow velocity is unhindered. Figure 1 1 1 Comparison of f luid velocity profiles A ) EOF B ) Pressure d riven f low Another important characteristic that distinguishes EOF from pressure driven flow phenomenon is observed in the relative efficiency of each flow system as a function of pore radius. 4 v hy (where a is the pore radius and v hy 2 v e of These equations imply that when the pore radius is large (>10cm) pressure driven flo w

PAGE 34

34 systems are most effective but as the pore becomes smaller (on the order of the nano and micro scale) EOF becomes relatively more effective. 6 3 In the Helmholtz Smoluchowski e quation (Equation 1 5) there is a limitation based on pore radius. In the Helmholtz Smoluchowski equation the assumption is made that most of the solution in a pore or capillary is always bulk solution which is not always valid. When the diameter of a pore or capillary approaches the dimensions of the electrical double layer ( 1 In this situation, the electrical double layer overlaps the pore diameter and the charged pore wall is no longer completely balanced by the counter ions. Often in this case the pore will ex hibit a certain level of permselectivity. Other conditions that make cause deviations from the Helmholtz Smoluchowski theory are large zeta potentials, increases in surface conductance, changes in permittivity, and changes in viscosity within the pore or 1/z 1. In this environment the driving force of EOF is extended throughout the entire pore cross section, the flow will experience a frictional force at the pore wall. As a consequence of this frictional force the flow velocity will be slow near the pore wall and the fluid flow velocity profile will be parabolic instead of plug like. Thus the EOF will be will less 1/z 1. Modulation of e lectroosmotic f low As stated earlier, the zeta potential is an important parameter in EOF systems. From the Helmholtz Smoluchowski equation we observe that EOF scales with the magnitude of this potential. Therefore it is important to be able to influence and control this potential. In general, for a solid surface the zeta potential is dependent on the bulk ionic concentration, valence of the ions, and pH of the solution (pH has the ability to

PAGE 35

35 change the zeta potential from positive to negative). However, the zeta potential can also be altered by changing the microchannel surface material (e.g. from glass to PDMS) or by placing coatings on the microchannel wall. In addition, the zeta potential is influenced by the ionic strength of th e electrolyte solution since a higher ionic strength leads to a compression of the electric double layer. As a result of this compression the zeta potential will be reduced. Other parameters that may influence the zeta potential are the chemical modificat ion of the capillary wall, the addition of surface active species or organic modifiers to the buffer, and the application of an external voltage to the system to regulate surface charge. Thesis Overview In Chapter 2, we describe the utilization of the template method to fabricate carbon nanotube membranes. This method entails synthesis of carbon nanotube membranes by chemical vapor deposition within the pores of high pore density alumina membrane templates. We thoroughly explain the advantage of exploi ting the flexibility in customizing nanostructures with specific characteristic s by using the template method. We illustrate methods of controlling specific parameters of chemical vapor deposition which produce different properties in the carbon nanotube m embrane. Lastly, we discuss the methods of characterization for the CNM. In C hapter 3, we describe EOF in carbon nanotube array membranes. EOF was driven across the CNM by placing the membrane between two electrolyte solutions and placing an electrode in e ach solution to pass a constant current through the nanochannels of the membrane. EOF was monitored by measuring the flux of a probe molecule across the membrane. The CNMs have an anionic surface charge, which

PAGE 36

36 results in EOF in the direction of cation migr ation across the membrane. EOF was used to control the flux of the probe molecule (phenol) When EOF was driven in the direction of diffusion there was an increase in the flux of the probe molecule across the membrane. Reversing EOF decreases flux in the d irection of diffusion At very high reversal EOF velocities transport in the direction of diffusion becomes negli gi ble and net transport proceeds in the opposite direction. EOF through the CNM was modified by changing the pH, ionic strength and the current density applied to the membrane. These changes resulted in changes in the zeta potential. As stated earlier, the zeta potential is an important parameter in EOF systems. From the Helmholtz Smoluchowski equation (Equation 1 5), we observe that EOF s cales with the magnitude of this potential. Therefore it is important to be able to influence and control this potential.

PAGE 37

37 CHAPTER 2 SYNTHESIS OF CARBON NANOTUBE MEMBRANES B Y THE TEMPLATE METHO D Introduction Sinc e their initial discovery in 1991 by Iijima 22 23 CNTs have generated much interest for their unique properties and possible applications. CNTs are a promising material for: field emitters 6 6 molecular electronics 6 7 strength enhancing filler materials 6 8 hydrogen storage 69 7 1 and molecular sieves 7 2 N ew applications for CNTs are continuously being investigated and new discoveries are made every day. 7 3 Recently, c arbon nanotubes have been fabricated in arrays and utilized as membrane materials. As a me mbrane material carbon nanotubes are of significant interest due to the phenomenal properties that they possess These properties include a high surface area, chemical and thermal stability, low production cost, and electrical conductivity. When designin g a membrane fabrication strategy there are several parameters to consider. These parameters include permeability, perm selectivity, porosity, pore size distribution, material absorbance, and conformability. Many of these parameters are typically at odds w ith each other, thus it is important to have a fabrication strategy that can easily be customized for a given application. 7 4 There are many advanced membrane design strategies that have been employed to control pore length, diminish pore size distributions and maximize pore density. One such strategy is chemical vapor deposition of carbon on to the pore walls of a membrane. This method has been used to narrow and homogenize pore diameter and length. Althoug h considerable progress has been achieved by utilizing membrane design strategies producing a mechanically strong membrane of defect free, homogeneous,

PAGE 38

38 aligned arrays of carbon nanotubes is still considered a daunting task. In the literature Sun and Crooks developed an inventiv e method of fabricating a CNM but with only a single multi walled carbon nanotube. 7 5 Alternatively, the Martin group and others have demonstrated the ability to grow CNM by CVD of carbon along the pore walls of nanoporous templates using a method called t emplate synthesi s ( Figure 1 4) 4 8 52 Template synthesis offers the flexibility of using membranes on the order of 10 11 nanopores per cm 2 4 5 4 6 The distinctive feature of template synthesis of CNTs by CVD is that the inner diameters of the CNT are tunable Membranes have been prepared with nanotube diameters ranging from >100nm to <10nm. Here we discuss the advantages of template synthesis. We also illustrate the utility of this m ethod by using it to fabricate CNM s. This is achieved by chemical vapor deposition of carbon into the nanopores of an anodic aluminum oxide membrane template The key advantage of template synthesis is the ability to customize nanostructures with specific characteristics. In this method n anotubes of carbon conform to the sizes and shapes of the respective host pores in which they are grown. Thus, there are a variety of design options available from which to choose. By selecting specific design parameters we have extensive control of the c arbon morphology and membrane characteristics. The carbon layer growth was controlled as a function of the CVD rate and time. At low CVD rates, carbon deposition is homogeneous and easily controlled as a function of time. However, at high CVD rates

PAGE 39

39 carbon deposition becomes diffusion limited and preferentially deposits at the mouths of the pores. Experimental Materials Two types of Whatman Anodisc TM Filter Membranes were used in diameters of 20nm or 200nm as templates. Ethylene (30% in helium) and argon were used as received. All other chemicals were reagent grade and used as received from Fisher Scientific (Fairlawn, NJ). All solutions were prepared using water tha t was purified by passing house distilled water through a Barnstead (Dubuque, IA) E pure water purification system. The carbon nanotubes discussed in this work were prepa r ed in 200 nm templates unless otherwise stated. Preparation of Carbon Nanotubes The preparation of carbon nanotubes by template synthesis has been previously reported. 4 8 5 0 5 1 The first step in CNM fabrication is heat pretreatment. Prior to CVD synthesis, the alumina template membranes were pressed between two quartz plates and heated t o a temperature slightly above the CVD temperature for 1 hour in air. If this heat pretreatment is not done, the alumina membranes curl into a tight cylinder during the CVD synthesis of the carbon nanotube membrane. Heat pretreatment between the quartz pla tes prevent this unwanted curling, and allow for synthesis of planar carbon nanotube membranes. After heat pretreatment a (2.2 cm x 1.5cm) piece of the heat treated alumina template membrane was placed vertically in the CVD reactor. The CVD reactor consist ed of a quartz tube (diameter = 4.2 cm, length= 52 cm) that was placed within a tube furnace. The reacto r temperature was increased to the designated deposition

PAGE 40

40 temperature ( 650 C 770C ) under constant argon flow. When the temperature stabilized, the argon flow was terminated, and simultaneously a 20 sccm flow of ethylene gas mixture was initiated. Under these conditions ethylene pyrolyzes to yield carbon nanotubes within the pores of the template as well as a thin carbon film on both faces of the memb rane. Deposition was continued for a prescribed amount of time. After the deposition the ethylene flow was terminated, the argon flow was resumed, and the furnace was turned off and allowed to cool to room temperature. After synthesis the membranes were st ored in air for at least 24 h ours prior to use. The CVD temperature and time were varied for several reasons. The CVD tem perature controls CVD kinetics and determine s many of the properties of the carbon The deposition time regulates the thickness of the carbon nanotube wall and surface layer of carbon on the membrane. Characterization of Carbon Nanotube Membranes X Ray Photoelectric Spectroscopy X r ay P hotoelectric S pectroscopy (XPS) studies of the samples were performed on a Perkin Elmer 5100 XPS sy The samples were mounted onto a stainless steel sample stub by means of a 7 mm diameter carbon sheet disk (SPI Supplies), so as to completely cover both the carbon disk and the sample stub. The sample s were inserted into the sample analyzer chamber by means of a quick insertion probe, and spectral acquisition with the fixed analyzer transmission mode commenced after the pressure decreased to 5 x10 9 Torr. Survey scans, high resolution C 1s spectra, and high resolution O 1s spectra were recorded at a takeoff angle of 20 relative to the sample surface. The information depth is estimated

PAGE 41

41 to be 2.5 nm for C 1s. Data analysis was done by using the AugerScan software package. Peak positions were all referenc ed to 285.0 eV for the main carbon peak. Electron m icroscopy Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) were used to visualize the carbon nanotubes prepared within the pores of the alumina template membrane. The carbon na notubes were imaged both as a cross section while still inside the alumina membrane and also as liberated individual nanotubes. This was achieved by first dissolving the alumina in a solution of 10% H 3 PO 4 for 16 hours. The H 3 PO 4 was removed by pipette leaving the liberated nanotubes, still connected together by the carbon surface films. The nanotube array was carefully rinsed with methanol and then suspended in methanol. This suspension was then ultra sonicated for 20 minutes, which produced individual carbon nanotubes separated from the carbon surface film. For the preparation of TEM samples a drop of the sonicated membrane suspension was applied to a Formvar Layered Carbon Type A TEM grid. The carbon nanotubes adhered well to the grid and were easily visualized with use of a Hitachi H 7000 tran s mission electron microscope. For SEM imaging a drop of the suspension was applied to a piece of copper tape attached to a SEM stub. The solvent of the suspension was allowed to dry and gold was sputte re d onto t he sample to improve electrical conductivity. The images were acquired using a Hitachi S 4000 field emission scanning electron microscope. Sheet r esistance Sheet resistance is a measurement of the resistance for thin films that are uniform in thickness. The sheet resistance was measured for the CNMs using a Lucas

PAGE 42

42 Signatone SP4 4 point probe and a Keithley 2100 multimeter. The basic model for these measurements is illustrated in Figure 2 1. Figure 2 1. Basic setup for sheet resistance measurements The sample is placed on a flat surface and four sharp probes make contact with the material to be measured. A constant input current is passed through the two outer electrodes, and the changes in voltages are measured across the inner pair of electrodes. 76 77 The spacing of the probes plays key role in the sheet resistance measurement. If the spacing between the probe points is constant, and the conducting film thickness is less than 40% of the spacing, and the e dges of the film are more than four times the

PAGE 43

43 spacing distance from the measurement point, the average resistance of the film or the sheet resistance is given by E quation 2 1 76 77 ( 2 1) cm) are related to s b y Equation 2 2. / thickness (cm) (2 2) Results and Discussion As discussed previously, the Anodisc TM aluminum oxide membrane has an unbr anched face and a branched face ( Figure 1 5). At the branched face of the template the pore openings are irregularly shaped with diameters from 100~200 nm (Figure 1 5 ). At the unbranched face of the template the pore openings are circular with an averag e approximate pore diameter of 25 0 nm. Each of the pores in the membrane serve as a cast for which the carbon nanotubes are grown. Thin layers of carbon film are deposited by CVD onto the pore walls of the membranes which create nanotubes. As stated earlier the morphology and dimensions of each nanotu be is determined by the profile of the host pore in which it is grown Figure 2 2 displays a cross section of the carbon nanotube membrane. W e measure the pore diameter and calculate the pore density f rom these electron micrographs. Figur e 2 3 shows a TEM image of individual carbon nanotubes that have been detached from the carbon surface film after alumina dissolution following ultrasonication of the nanotube suspension. These images illustrate the consistency of the wall thickness throughout the entire le ngth of nanotube. From these images a consistent wall thickness is obtained.

PAGE 44

44 Figure 2 2 Electron m icrograph of the cross section of carbon nanotube membrane Figure 2 3 Electron m icrograph of individual carbon nanotubes Effect of CVD Rate on Carbon Nanotube Growth Exercising precise control of the CVD rate of carbon nanotube growth inside the pores of the alumina membrane is of utmost importance. There are several factors that

PAGE 45

45 affect the CVD rate including the diameter of the host pores, the deposit ion temperature and the gas flow rate. In this section we will very briefly discuss these parameters. Chemical vapor deposition of carbon in a nanoporous alumina membrane is gas diffusion limited; as the pores of the nanotubes become narrower the gas is more impeded and has a harder time passing through the whole length of the pore. As a result of this hindrance the CVD rate decreases. We define the ratio of the mass of carbon to the mass of the alumina membrane (AAO) as % carbon in Equation 2 3. ( 2 3 ) Here, m int is the mass of the alumina template before carbon deposition and m fin is the mass of the membrane after carbon nanotube growth. The CVD rate is defined as the amount of carbon deposited onto a substrate per unit time of depositio n at a constant temperature. F igure 2 4 displays a plot of percent carbon versus time of carbon deposition grown at 725 o C. Figure 2 4 Plot of % c arbon on the a lumina t emplate vs.CVD t ime 0 10 20 30 40 50 60 70 0 100 200 300 400 500 600 % Carbon ( gC/gAAO) CVD Time (min)

PAGE 46

46 Initially there seems to be an approximant linear relationship between the percent carbon and the CVD time (CVD T ) but as the CVD T is extended the percent carbon plateaus. This ca n be explained by investigating the relationship between the mass of the carbo n deposited and the surface area of the substrate. According to the literature, 7 8 the pore walls of the 200 nm Anodisc TM membrane comprise more than 99 % of the total surface area of these membranes. As the carbon is deposited onto the pore walls of the memb rane the diameter of the pores becomes narrower, which produces a significant drop in surface area of the membrane. As seen in the plot, it is hypothesized that at some point in the deposition the carbon has essentially closed the pore, which means deposition would only continue on the template surface and not inside the pore. From the slope of the plot displayed in F igure 2 4 the CVD rate can be predicted for short duration depositions at a given temperature. A n average CVD rate can be calculated from E quation 2 4 ( 2 4 ) Effect of Temperature on CVD Rate It has been concluded that at high temperatures CV D reactions occur more rapidly. 79 When the temperature is very high (>850) the vapor deposition reactions occur so rapidly, that the carbon deposition is non uniform along the length of the nanotube. Diffusion is considered to be a slow process, when the temperature is high the CVD reactions occur so fast that diffusion can no longer maintain equil ibrium throughout the thickness of the membrane; thus a carbon concentration gradient develops along the length of the nanopore. In Figure 2 5 we display that as the temperature increases the average CVD rate also increases. When a low CVD temperature (< 750) is selected the

PAGE 47

47 result is usually a CNM with uniform wall thickness. The duration of low temperature CVD typically results in thicker or thinner nan o tube walls. This utility allows for the customization of the CNMs with a variety of pore diameters. Figure 2 5 Plot of % c arbon vs CVD t emperature Conclusion Carbon nanotubes are of great interest in both fundamental and applied science. Our group has created template synthesized carbon nanotubes by chemical vapor deposition of ethylene gas onto co mmercial alumina membranes. By utilizing this method carbon nanotubes can be grown with tailored properties, dimensions, and shapes. The key feature of template synthesis is that each carbon nanotube conforms to the pore morphology of the template membrane Each nanotube has an inner diameter defined by the thickness of the tube wall and the diameter of the host pore. 0 10 20 30 40 50 60 70 640 660 680 700 720 740 760 780 800 % Carbon (gC/gAAO) CVD Temperature

PAGE 48

48 The CNMs prepared in the alumina templates contain parallel, homogeneous, dense, arrays of aligned monodisperse carbon nanotubes. We have em ployed a number of design strategies to modify and control carbon nanotube growth inside the alumina membrane. Controlling the CDV rate by modifying certain CVD conditions have been investigated. From the results we conclude that conducting CVD at a low te mperature (<770 o C) and for durations less than 7 hours will produce the most consist ent and reproducible CNM. The carbon nanotube membranes (CNM) discuss ed in this work are used for fundamental electrochemical investigations. They are used specifically as tools to monitor and control electroosmotic flow (EOF). 10,43 Although nanotubes grown in the 200 nm Anodisc TM templates are not exactly perfect cylinders, they are inexpensive, commercially available and of consistent quality. They are ideal for electroosmotic flow p ermeation experiments and have been u sed in past research endeavors. 50 ,6 2

PAGE 49

49 CHAPTER 3 ELECTROOSMOTIC FLOW IN CARBON NANOTUBE M EMBRANES PREPARED BY THE TEMPLATE SYNTHESIS METHOD Introduction Carbon nanotubes (CNTs) have attracted great interest due to their remarkable electronic, optical, and mechanical properties. Potential applications for carbon nanotubes range from fundamental research, to the development of new devices. The Martin group and others have created template synthesized carbon nanotubes by chemical vapor deposition of ethylene gas onto commercial alumina membranes. By utilizing this method carbon nanotubes can be grown with tailored properties, dimensions, and shapes. The key f eature of template synthesis is that each carbon nanotube conforms to the pore morphology of the template membrane. Each nanotube has an inner diameter defined by the thickness of the tube wall and the diameter of the host pore. The CNMs prepared in the al umina templates contain parallel dense arrays of aligned monodisperse carbon nanotubes. It has recently been suggested that carbon nanotube membranes could possibly be used as separator membranes in Li ion batteries. A battery separator is a porous membran e placed between the anode and cathode of a battery. The main function of a separator is to keep the positive and negative electrodes apart in order to prevent electrical short circuits, while also allowing the transport of ionic charge carriers. 8 0 Relative to the amount of research on electro de materials and electrolytes, very little attention has been aimed at characterizing and developing new separator materials. The most important requirements for Li ion battery separators are permeability, porosity, chemical and thermal stability, pore siz e, and electrical

PAGE 50

50 resistance. Many of these characteristics are available and tunable in template prepared CNMs. The research results presented in this thesis have been aimed at studying the fundamental transport processes occurring at the interface of a porous carbon nanotube membrane and an electrolyte by monitoring and controlling electroosmotic flow (EOF). EOF refers to the movement of solution past a stationary surface due to an externally applied electric field. B uilding upon previous investigations 5 0 6 2 we demonstrate here, the effects of pH, ionic strength, and applied transm embrane current on the EOF by monitoring and measuring the flux of a probe molecule across the cylindrical pores of carbon nanotube membranes 5 0 6 2 81 83 The as synthesi zed CNMs have anionic surface charge, and as a result, EOF is in the direction of cation migration across the membrane. The results from this research should provide some of the fundamental information needed to understand a practical approach for using a carbon nanotube membrane as a possible new material for advance battery separators. Experimental Materials Whatman Anodisc TM Filter Membranes with 200 nm pores were used as temp late membranes. Ethylene (30% in helium) and argon were used as received. Buffe r solutions were prepared from monobasic sodium phosphate and the pH was adjusted with 1 M NaOH and 1 M HCl. Phenol was obtained from Sigma Aldrich (St. Louis, MO). All other chemicals were reagent grade and used as received from Fisher Scientific (Fairla wn, NJ). All solutions were prepared using water that was purified by passing house distilled water through a Barnstead (Dubuque, IA) E pure water purification system.

PAGE 51

51 Preparation of CNM The CVD method used to prepare these carbon nanotubes has been desc ribed previously in Chapter 2 of this thesis a nd elsewhere in the literature 4 8 5 0 5 1 Briefly, the alumina template membrane was pressed between two quartz plates and heated to 750 o C for 1 hour in air prior to chemical vapor deposition. This heat pretreatment is done to prevent unwanted curling of the alumina membrane during the CVD synthesis of the carbon nanotubes. After heat pretreatment a (2.2 cm x 1.5cm) piece of the alumina t emplate membrane was placed vertically in the CVD reactor in the tube furnace. The reactor temperature was increased to 725C, under argon flow. When the temperature stabilized, the argon flow was terminated, and simultaneously a 20 sccm flow of an ethylen e gas mixture was initiated for 4 hours. After deposition the argon flow was resumed, and the ethylene flow was terminated After carbon nanotube synthesis the membrane was stored in air for at least 24 h ours CNMs prepared in this way are classified as membranes and have an anionic surface charge. EOF Measurements The permeation measurements in this study were conducted with a CNM assembly. The CNM assembly as shown in Figure 3 1 is a modified assembly that was originally fabricated by Mille r et al. 90, 88 5 0 6 2 The assembly is composed of two acrylic slides, two pieces of parafilm, two parafilm washers, the carbon nanotube membrane and a piece of conductive copper tape attached to the membrane. The various components of the assembly were secured by heating the membrane assembly in an oven at 125C for 5 min. The heat caused the p arafilm to melt, which acted as a glue to hold the various pieces together.

PAGE 52

52 Figure 3 1 Illustration of the c arbon n anotube m embrane a ssembly Electroosmotic flow (EOF) measurements were made in simple U tube permeation cells 5 0 ,6 2 8 3 This assembly exposed a 0.495 cm 2 area of the CNM to the electrolyte solution. As with previous reports, 5 0 ,6 2 81 8 3 the rate of EOF was determined by measuring the flux of a neutral probe molecule across the membrane or in this case the membrane assembly. The carbon nanotube membrane assembly was clamped between the two halves of a U tube cell. The membrane assembly was e xposed to the electrolyte solution on one side (permeate side) and electrolyte solution with 10 mM phenol added to it on the other side (feed side). A platinum wire electrode was placed in each solution, and a constant transmembrane current was supplied us ing an EG&G 263A galvanostat. As stated in the previous section, the as synthesized membranes have an anionic surface charge, which means the direction of EOF will initially be toward the cathode. The galvanostat was configured as illustrated in Figure 3 2 such that the reference/counter electrode

PAGE 53

53 (cathode) was positioned on the permeate side and the working electrode (anode) was located in the feed side Figure 3 2 Schematic of p ermeation c ell used in EOF e xperiments Results and Discussion Effect of Applied Current Density on Permeate Flux The flux of the probe molecule is directed related to the magnitude and direction of the current. Figure 3 5 shows a plot of micromoles of phenol transported across the carbon nanotube membrane as a function of ap plied current density. This plot consists of five linear curves and the slope of each curve is the flux of phenol ( N J ) at that applied current density as seen in T able 3 1. The dark blue linear curve in Figure 3 1 was diffusive flux ( N diff ) which means these measurements were obtained with no applied current. The flux in this case is only from the diffusion of phenol molecules from the feed side across the membrane and into the permeate half ce ll. The purple linear curve was obtained wi th an applied current

PAGE 54

54 density of J app = 6 .0 6 mA/ cm 2 The flux of phenol across the membrane in this linear curve was observed to decrease below that of the diffusion curve indicating that EOF is in the opposite direction of the diffusive flux (EOF from permeate solution to feed solution). This concept is confirmed by data in the red linear curve which were obtained with J app = +6 .0 6 mA/ cm 2 This data display a net flux that is higher than the diffusive flux which signifies EOF in the direction of diffusive transport. Th e concluding two linear curves were obtained at higher positive and negative applied current densities and f urther enhancement and diminution in the flux were respectively observed as a result. F igure 3 3 Plot of p henol t ransport vs. time at differen t c urrent d ensities 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 100 200 300 400 500 600 700 Phenol Transported ( mol) Time (seconds) 12.12mA/cm^2 6.06mA/cm^2 Diffusion -6.06mA/cm^2 -12.12mA/cm^2

PAGE 55

55 Table 3 1. Effect of applied current density on the flux, enhancement factor, peclet number and the electroosmotic flow velocity of phenol through the carbon nanotube membrane J app mA/cm 2 Flux nmol/(s*cm 2 ) E N J /N Diff Peclet Number v eof m/s 12.12 0.53 0.65 .081 13.07 6.06 0.67 0.83 0.36 5.81 0.00 0.81 1.00 0.00 0.00 6.06 0.98 1.21 0.39 6.29 12.12 1.21 1.49 0.86 13.87 The geometric area of the membrane was 0.495 cm 2 Calculation of the E lectroosmotic V elocit y The Nernst Plank equation was used to describe the steady state flux of permeante molecules across a membrane in the presence of an applied transmembrane current. ( 3 1) I n Equation (3 1), D is the diffusion coefficient, C is the concentration, and z is the charge of the permeate molecule, respectively C(x)/ x is the concentration gradient in the x direction across the membrane, is the potential gradient of the electrolyte withi n the nanotubes, and v eof is the electroosmotic velocity. The Nernst Plank equation separates molecule transport into three distinct terms that describe the diffusive, migrational and EOF convective transport processes. Since phenol is a neutral molecule at the pH used in these experiments E q uation 3 1 simplifies to Equation 3 2. C v eof ( 3 2 ) When there is no applied current ( J app = 0) the flux is only generated by diffusion so only the first term in E quation 3 2 becomes operative ( N diff ). A parameter called the enhancement factor, E can be calculated from the experimental flux data in Table 3 1.

PAGE 56

56 The enhancement factor can then be used to calculate the electroosmotic velocity as seen in E quation 3 3 82 ( 3 3 ) E is the ratio of the flux in the presence of an applied current, to the diffusive flux (flux at J app = 0). The enhancement factor ( E) values for the data in Figure 3 5 are shown in Table 3 1. E is greater than unity when a positive current is applied. This result indicates that EOF is in the direction of the diffusive flux. Analogous results were obtained by Miller et al. for carbon nanotube membranes, wh ere EOF is due to the membranes fixed carboxylate sites. 8 3 From t hese results we confirm that for air treated carbon, the pore walls on the carbon nanotubes have fixed negatively charged sites. 8 5 8 7 The peclet numb er is the ratio of pure EOF flux to diffusive flux. It can be calculated from the equation Srinivasan and Higuchi applied to demonstrate the relationship between the enhancement fact ( E ) and peclet number ( Pe ) 84 ( 3 4 ) The peclet number can then be used to calculate v eo f using E quation 3 5 ( 3 5 ) In Equation 3 5, l is the thickness of the membrane, and D is the bulk solution diffusion coefficient. From Table 3 1, we observe d that current densities of the same magnitude but opposite sign give EOF velocity values of approximately the same value but opposite signs. T he Helmholtz Smoluchowski equation 63 depicts the relationship between the electroosmotic velocity and the linear e lectric field gradient ( E ( x ), V cm 1 ) across the membrane

PAGE 57

57 ( 3 6 ) In E quation 3 6 electrical field gradient (E/l), and product of the applied current density and the resistivity of the electrolyte within the for E(x) in E quation 3 6 we derive Equation 3 7. ( 3 7 ) Equation 3 7 reveals the linear relationship between the applied current density and the v eof From the Figure 3 3 we observe that the flux increases with increasing positive applied current density (J a p p ). This result indicat es that EOF plays a substantial role in the transport of permeate across the membrane. At sufficiently negative J app values, the net flux approaches zero. In this case, EOF from the permeate half cell to the feed half cell is canceling out the diffusive flux from feed to permeate. 8 1 Effect of i o nic s trength of e lectrolyte on v eof Data was collected from the transport of phenol at a const ant applied current density of 10 mA/cm 2 and at a various ionic strengths. As seen in F igure 3 4 the ionic s trength ranged from 0.01M to 0.6 0M. The pH of the electrolyte was maintained at 7. The general tread of the plot shows that as the ionic strength increases the v eof decreases as is expected. We observe from Equation 3 7 that veof is directly related to the zeta potential. The zeta potent ial is related to the Debye length k 1 by Equation (3 8) 88 ( 3 8 )

PAGE 58

58 Figure 3 4 Plot of electroosmotic flow velocity vs. ionic strength of permeate and feed solutions Applied current density was + 10 mA/ cm 2 Here, k is the Boltzmann constant, T is the temperature, z is the counter ion charge, e is the electrolyte charge, is the permittivity of water, 1 is the Debye length and is the surface charge density at the plane of shear between the capillary wal l and the electrolyte. The zeta potential is used as a boundary condition for the potential at the point of zero velocity and is thus essential in calculations of electro kinetic flows. When solutions with relatively small zeta pote ntials are utilized in permeate experiments E quation 3 8 simplifies to Equation 3 9. ( 3 9 ) 0 10 20 30 40 50 60 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 v eof (m/s) Ionic Strength (M)

PAGE 59

59 When the electrolyte solution is symmetrical and is at 25 o 1 is related to the concentration of the electrolyte by Equation 3 10. 63 ( 3 10 ) In Equation 3 10, z is the charge of the electrolyte and c is the concentration of the electrolyte From this equation we observe that the double layer thickness decreases potential decreases with increasing ionic strength (Equation 3 9). This information explains one of the reasons why v e of decreases with increasing ionic strength, which is experimentally observed in F igure 3 4 From Equation 3 7 we observe that another factor contributes to the observed decrease in v eof w proportional to the v eof The resistivity is inversely proportional to the number of charge carrying i decreases with increasing ionic strength of the electrolyte, and Equation 3 7 indicates that this would also contribute to the observed decrease in v eof with increasing ionic strength as observed in Figure 3 4 For these reason s we observe that when the highest ionic strength electrolytes are employed the electroosmotic velocities decrease close to zero. Effect of s olution pH on v eof The v eof v ersus solution pH data are shown in Figure 3 5 The transport of phenol was monitored at applied current densities of positive 1 5 mA cm 2 in electrolyte solutions that were 10 mM in NaCl and 0.5 mM in phosphate buffer and with the pH adjusted to values between 3 and 7 The NaCl served to maintain the ionic strength at a constant value.

PAGE 60

60 The plot followed an expected trend considering EOF is a charge dependent phenomenon and the surfaces of the CNMs have anionic surface charge. At high pH values where the negative surface charge is enhanced the v eof is also enhanced A lternatively at low pH values where the negative surface charge is diminished, the v eof is also diminished. Figure 3 5 .Plot of electroosmotic flow velocity vs. pH of permeate and feed solutions. Applied current density was + 15 mA/ cm 2 From the data in Figure 3 5 we can calculate both the number density and p K a of the surface acidic sites on the carbon nanotube membranes. 88 We apply the Boltzmann equation to first calculate the surface hydronium ion concentration ([H + ] s ) from the bul k solution hydronium ion concentration ([H + ] o ). ( 3 1 1 ) 0 5 10 15 20 25 30 3 4 5 6 7 8 v eof (m/s) pH

PAGE 61

61 The zeta potential values needed at each pH can be obtained by utilizing E quation 3 7 employing the known values for the permittivity and viscosity of water (6.95 x 10 10 C 2 J 1 m 1 and 0.890 cP ) 89 cm). From the literature Jimbo 88 determined the surface charge density is related to the number density of acidic sites ( N a ) and the surface p K a via E quation 3 12: (3 12) We solve E quation 3 9 for and set it to equal E quation 3 12. By solving also for zeta potential using Equations 3 9 and 3 12 we derive E quation 90 3 13. ( 3 13 ) Z eta potential may be plotted against pH and t hough beyond the scope of this thesis, a nonlinear least squares regression curve of E quation 3 13 with surface pKa and N a as adjustable parameters can be used to obtain the number density and p K a of the surface acidic si tes on the carbon nanotube membranes a s described by Miller et al. 90, 88 Conclusions EOF is an important analytical phenomenon that can be used to elucidate information for a given system. Th is research has illustrated that the EOF rate and direction can be monitored and controlled across template prepared carbon nanotube membranes through a variety of methods. EOF is proportional to the current density applied to a particular system. Thus the rate and direction of EOF can be controlled by changing the mag nitude and the sign of the transmembrane current. The rate and direction of EOF can also be controlled by varying the pH of the contacting solution

PAGE 62

62 phase at a constant applied current density. Ionic strength of the electrolyte solution and surface charge o f the membrane play significant roles in EOF measurements. In the future we hope to capitalize on the electronic conductivity of CNMs. We have previously shown that the sign and magnitude of the excess surface charge can be controlled potentiostatically in gold nanotube membranes by simply applying a potential to the membrane in an electrolyte solution. 9 1 We endeavor to control surface charge potentiostatically and simultaneously monitor EOF in CNMs. In this system we expect that the CNMs will exhibit ex cess negative surface charge when negative potentials are applied, and excess positive surface charge when positive potentials are applied.

PAGE 63

63 CHAPTER 4 DISCUSSIONS AND C ONCLUSIONS Carbon nanotubes have attracted the interest of many scientists throughout the world. The strength, small dimensions and remarkable physical properties of carbon nanotubes have made them a material with a range of promising applications. Many of these applications will require the use of uniform dense arrays of free standing, open ended nanotubes. One way to achieve such nanotubes is template synthesis. The Martin group and others have created template synthesized carbon nanotubes by chemical vapor deposition of ethylene gas onto commercial alumina membranes. The key feature of template synthesis is that each carbon nanotube conforms to the pore morphology of the template membrane. By utilizing this method carbon nanotubes can be grown with tailored properties, dimensions, and shapes. The goals of the research present ed in this thesis were to customize the chemical, physical, and structural properties of carbon nanotube membranes and to study the fundamental transport processes occurring at the interface of the carbon nanotube membranes and the electrolyte by monitorin g and controlling electroosmotic flow. We have briefly introduced carbon nanotubes, their discovery, properties, the need for carbon nanotube membranes and some of the possible applications in which carbon nanotubes membranes may be utilized. One such possible application is as a new material for battery separators. A battery separator is a porous membrane placed between the anode and cathode of a battery. The main function of a separator is to keep the positive and negative electrodes apart in order to prevent electrical short circuits, while also allowing the transport of ionic charge carriers which are needed to

PAGE 64

64 complete the circuit during the passage of current in an electrochemical cell The most important requirements for Li io n battery separators are permeability, porosity, chemical and thermal stability, pore size, and electrical resistance. Many of these characteristics are available and tunable in template prepared CNMs. We also explain in detail electro osmotic flow and the principles necessary to understand this phenomenon. We de tailed the utilization of the template method to fabricate carbon nanotube membranes. This method entails synthesis of carbon nanotube membranes by chemical vapor deposition within the pores of alu mina membrane templates with high pore density. We investigated and defined the growth rate of carbon on to alumina and discovered the necessary conditions for producing consistent, conductive, defect free membranes. We described the advantages of using the template method to customize nanostructures with specific characteristics. We illustrated methods of controlling certain parameters of chemical vapor deposition to produce and enhance different properties in the carbon nanotube membrane. Lastly, we dis cuss ed the methods of characterization for the CNM as they related to the specific aim of fluidic transport. Electroosmotic flow is an important analytical phenomenon that can be used to elucidate information for a given system. The research in this thesis has illustrated that the EOF rate and direction can be monitored and controlled across template prepared carbon nanotube membranes through a variety of methods. The work presented here has been introductory in many ways. Much of this work has been a repr oduction of previous studies 5 0 5 8 6 2 8 1 86 and has laid a necessary framework for new research frontiers. In the future we hope to capitalize on the

PAGE 65

65 electronic conductivity of CNMs. We have previously shown that the sign and magnitude of the excess su rface charge can be controlled potentiostatically in gold nanotube membranes by simply applying a potential to the membrane in an electrolyte solution. 91 We endeavor to control surface charge potentiostatically and simultaneously monitor EOF in CNMs. In th is system we expect that the CNMs will exhibit excess negative surface charge when negative potentials are applied, and excess positive surface charge when positive potentials are applied thus providing another avenue for controlling and modulating EOF.

PAGE 66

66 LIST OF REFERENCES (1) Tarascon, J. M.; Armand, M. Nature 2001, 414, 359 367. (2) Winter, M.; Besenhard, J.O. Electrochim. Acta 1999, 45, 31 50. (3) Nishi, Y. Lithium Ion Batteries, Wakihara, M.Yamamoto, O.,Eds.; Kodansha: Tokyo, Japan, 1998; pp 181. (4) Hossain S. Handbook of Batteries, 2nd ed.;, Linden, D., Ed.; Mcgraw Hill: New York; 1995; pp 12 35. (5) Rand, D. A. J; Woods, R.; Dell R. M. Batteries for Electric Vehicles; Bagshaw, N. E., Ed.; Research Studies Press: Somerset, England 1998; p p 427. (6) Choi, S.; Lim, H. Factors that affect cycle life and possible degradation mechanisms of lithium ion batteries based LiCoO2. J. Power Sources 2002, 111, 130 136 (7) Linden, D.; Reddy, T. B. Handbook of Batteries, 3rd ed.; McGraw Hill: New York 2 002. (8) Besenhard, J. O., Ed. Handbook of Battery Materials; Wiley VCH: Weinheim, Germany, 1999. (9) Arora, P; Zhang, Z. Chem. Rev. 2004, 104, 4419 4462. (10) Benett, J.; Choi, W. M. Developments in small cell separators. In Proceedings of the 10th Annual Battery Conference on Applications & Advances; IEEE: New York, 1995; p 265. (11) Kinoshita, K.; Yeo, R. Survey on Separators for Electrochemical Systems; Lawrence Berkeley National Laboratory. January 1985. (12) Berndt, D. Maintenance Free Batteries, 3rd ed., Research Studies Press: Somerset, England, 2003. (13) Bode, H. Lead Acid Batteries; John Wiley: New York, 1977. (14) Falk, S. U.; Salkind, A. J. Alkaline Storage Batteries; John Wiley: New York, 1969. (15) Fleischer, A.; Lander, J. J. Zinc Silver Oxi de Batteries; John Wiley: New York, 1971.

PAGE 67

67 (16) Brodd, R. J., Friend, H. M., Nardi, J. C., Eds. Lithium Ion Battery Technology: ITE JEC Press: Brunswick, OH, 1995. (17) Wakihara, M., Yamamoto, O., Eds. Lithium Ion Batteries, Fundamentals and Performance; Wi ley VCH: New York, 1998. (18) Yoshino, A. Chem. Ind. 1995, 146, 870. (19) Schalkwijk, W. A. V., Ed. Advances in Lithium Ion Batteries; Kluwer Academic: New York, 2002 (20) Bhnstedt, W. Separators. In Handbook of Battery Materials; Besenhard, J. O., Ed.; W iley VCH: New York, 1999. pp 245 292 (21) Spotnitz, R. Separators for Lithium Ion Batteries In Handbook of Battery Materials; Besenhard, J. O., Ed.; Wiley VCH: New York 1999. pp 553 563 (22) Iijima, S. Nature 1991, 354, 56 58. (23) Iijima, S.; Ichihashi, T. Nature 1993, 8, 3233 3250. (24) Gao, G.; Cagin, T.; Goddard, W. A., III. Nanotechnology, 1998, 9, 183 191 (25) Yu, M.; Files, B. S.; Arepalli, S.; Ruoff, R.S. Phys. Rev. Lett. 2000, 84, 5552 5555. (26) Hone, J.; Whitney, M.; Zettle, A. Synthetic Met 1999, 103 2498 2499. (27) Odom, T.W.; Huang, J.; Kim, P.; Lieber, C. M; Nature 1998, 391, 62 64. (28) Satishkumar, B.C.; Govindaraj, A.; Sen, R.; Rao, C.N.R. Chem. Phys. Lett, 1998, 293, 47 52. (29) Frank, S.; Poncharal, P.; Wang, Z.L.; Science 1998, 280, 1744 1746 (30) Thess, A.; Lee, R.; Nikolaev, P.; Dai, H.; Petit, P.; Robert, J.; Xu, C.; Lee Y. H.; Kim, S. G.; Rinzler, A. G.; Colbert, D. T.; Scuseria G.; Tomnek, D.; Fischer, J. E.; Smalley, R. E. Science 1996, 273, 483 487. (31) Ebbesen, T.W.; A jayan P.M. Nature 1992, 358, 220 222 (32) Lange, H.; Huczko, A.; Byszewski, P.; Mizera E.; Shinohara, H. Chem. Phys. Lett. 1998 289, 174 180. (33) Chernozatonskii, L.A.; Kosakovskaya, Z.Y.; Fedorov, E.A.; Panov, V.I. Phys. Lett. A 1995, 197, 40 46.

PAGE 68

68 (34 ) Ge, M.; Sattler, K. Science 1993, 260, 515 518. (35) Ge, M.; Sattler, K. Appl. Phys. Lett. 1994 64, 710 711. (36) Dai, H. Accounts of Chemical Research 2002, 35, 1035 1044 (37) Baughman, R. H.; Zakhidov, A.; de Heer, W. A. Science 2002, 297 787 792. (38 ) Campbell, S.A. The Science and Engineering of Microelectronic Fabrication; Oxford University Press: Oxford U.K., 1996. (39) Dai, H Sur. Sci. 2002, 500, 218 241. (40) Fan, S.S.; Chapline, M.G.; Franklin, N. R.; Tombler, T. W.; Cassell, A.M.; Dai, H. J. Science 1999, 283, 512 514. (41) Teo, K. B. K.; Chhowalla, M.; Amaratunga, G.A.J.; Milne, W. I.; Hasko, D. G.; Pirio, G.; Legagneux, P.; Wyczisk, F.; Pribat, D. Appl. Phys. Lett. 2001, 79, 1534 1536. (42) Franklin, N. R.; Dai, H. J. Adv. Mater. 2000, 12, 890 894. (43) Cassell, A.M; Franklin, N.R.; Tombler, T.W.; Chan, E.M.; Han, J.; Dai, H. J. J. Am. Chem. Soc. 1999, 121, 7975 7976. (44) Zhang, Y.G.; Chang, A.L.; Cao, J.; Wang, Q.; Kim, W.; Li Y. M.,; Morris, N.; Yenilmez E.; Kong, J.; Dai, H. J. Appl. Phys. Lett. 2001, 79, 3155 3157. (45) Hulteen, J.C.; Martin, C.R. J. Mater. Chem. 1997, 7, 1075 1087. (46) Martin, C. R. Science 1994, 266, 1961 1966. (47) Martin, C. R.; Mitchell, D. T. Anal. Chem. 1998, 70, 322A 327A. (48) Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature 1998, 393, 346 349. (49) Kyotani, T.; Tsai, L. F; Tomita, A. Chem.Matter. 1996, 8, 2109 2113. (50) Miller, S. A.; Young, V. Y.; Martin, C.R. J. Am. Chem. Soc. 2001, 123, 12335 12342. (51) Che, G. Lakshmi, B. B.; Martin, C. R.; Fisher, E. R.; Ruoff, R.S. Chem. Mater. 1998, 10, 269 267. (52) Kyotani, T.; Tsai, L. F.;Tominta, A. Chem. Commun. 1997, 701 702.

PAGE 69

69 (53) Penner, R.M.; Martin, C.R. Anal Chem. 1987, 59, 2625 2630. (54) Cheng, I. F.; Mart in, C. R. Anal. Chem. 1988, 60, 2163 2165. (55) Possin, G.E. Rev. Sci. Instrum. 1970, 41, 772 774 (56) Huczko, A, Appl. Phys A Mater. 2000, 70, 365 376. (57) Martin, C. R. Accounts of Chemical Research 1995, 28, 61 68 (58) Wirtz M.; Miller, S. A.; Martin, C. R. Int. J. Nanosci. Ser. 2002, 1, 255 268. (59) Liang, Z.; Susha, A. S.; Caruso, F. Adv. Mater. 2003, 15, 1849 1853. (60) Perry, J.L.; Martin, C. R.; Stewart, J. D. Chem. Eur. J. 2011, 17, 6296 6302. (61) Pan, L.; Qiu, H.; Dou, C.; Li, Y.; Pu, L.; Xu, J.; Shi, Y. Int. J. Mol. Sci. 2010, 11, 2636 2657. (62) Miller, S.A.; Martin, C. R. Carbon Nanotubes: Electroosmotic Flow Control in Membranes. In Dekker Encyclopedia of Nanoscience and Nanotechnology 1st ed.; Schwarz, J. A., Contescu, C. K. P., Eds.; Marcel Dekker, Inc: New York. 2004; pp 519 528. (63) Probstein, R.F. Physicochemical Hydrodynamics: An Introduction. Wiley: New York, 1994. (64) Shaw, D. Introduction to Colloid and Surface Chemistry: 3rd ed.; Butterworth: B oston, Mass., 1980. (65) Reuss, F. F. Proceedings of the Imperialist Society of Naturalists of Moscow 1809, 3,327 344. (66) Zhu, W.; Bower, C.; Kochanski, G.; Jin, S. Appl. Phys. Lett. 1999, 75, 873 875. (67) Bachtold, A.; Hadley, P.; Nakanishi, T.; Dekker C. Science, 2001, 294, 1317 1320. (68) Frankland, S. J. V.; Caglar, A.; Brenner, D.W.; Griebel, M. J. Phys. Chem. B. 2002, 106, 3046 3048. (69) Darkrim, L. F.; Malbrunot, P.; Tartaglia, G.P. Int. J. Hydrogen Energ. 2002, 27, 193 202. (70) Dillon, A. C.; Heben, M. J. Appl. Phys. A Mater. 2001, 72, 133 142.

PAGE 70

70 (71) Poirier, E.; Chahine, R.; Bnard, P.; Cossement, D.; Lafi, L. Appl. Phys. A Mater. 2004, 78, 961 967. (72) Wang, Q.; Challa, S. R.; Sholl, D.S.; Johnson, J.K. Phys. Rev. Lett. 1999, 82, 956 95 9. (73) Baddour, C.E.; Briens, C. Int. J. Chem. React. Eng. 2005, 3, p. R3. (74) Koros, W. Chem. Eng. Prog. 1995, 91, 68. (75) Sun, L.; Crooks, R. M. J. Am. Chem. Soc. 2000, 122, 12340 12345. (76) Smits F M Bell Syst. Tech. J. 1958 5 711 718 (77) Valdes, L. G. Proc. I.R. 1954 42, 420 427 (78) Crawford, G. P.; Steele, L. M.; Ondriscrawford, R.; Iannacchione,G. S.; Yeager, C. J.; Doane, J. W.; Finotello, D. J. Chem Phys. 1992, 96, 7788 7796. (79) Morosanu, C. E. Thin Films by Chemical Vapour De position; Elsevier: Amsterdam, 1990; 7. (80) Choi, J. Developments in small cell separators. In Proceedings of the 10th Annual Battery Conference on Applications & Advances; IEEE: New York, 1995; p 265. (81) Bath, B. D.; Lee, R. D.; White, H. S.; Scott, E. R. Anal. Chem. 1998, 70, 1047 1058. (82) Bath, B. D.; White, H. S.; Scott, E. R. Anal. Chem. 2000, 72, 433 442. (83) Miller, S. A.; Martin, C. R. J. Am. Chem. Soc. 2004, 126, 6226 6227. (84) Srinivasan, V.; Higuchi, W. I. Int. J. Pharm. 1990, 60, 133 138. (85) Panzer, R. E.; Elving, P. J. Electrochim. Acta 1975, 20, 635 647. (86) Bismarck, A.; Springer, J. Colloids Surf. A: Physicochem. Eng. Asp. 1999, 159, 331 339. (87) Garcia, A. B.; Cuesta, A.; Montes Moran, M. A.; Martinez Alonso, A.; Tascon, J. M. D. J. Colloid Interface Sci. 1997, 192, 363 367. (88) Jimbo, T.; Tanioka, A.; Minoura, N. Langmuir 1998 14, 7112 7118.

PAGE 71

71 (89) Weast, R. C.; Astle, M. J.; Beyer, W. H., Eds. CRC Handbook of Chemistry and Physics 66th ed.; CRC Press, Inc.: Boca Raton, FL, 1985 (90) Miller, S.A. Nanofludics in Tailored Carbon Nanotube Membranes. Ph. D. Dissertation, University of Florida, U. S., May 2011. (91) Nishizawa, M.; Menon, V. P.; Martin, C. R. Science 1995, 268, 700 702.

PAGE 72

72 BIOGRAPHICAL SKETCH William Cornelius Har dy was born in Jackson, MS. He graduated from Jackson State University in 2007 with a B.S. in chemistry and 2008 with a M.S in chemistry. While attending JSU, William disc overed he had a passion for nan oscience. After graduating, from JSU, William continue d his education by joining the research group of Dr. Charles R. Martin at the University of Florida in August 2008. In his free time William enjoys music, reading, and cooking.