<%BANNER%>

Template Synthesis for the Development of Three-Dimensional Nanostructured Solid-State Lithium-Ion Batteries and Investi...

Permanent Link: http://ufdc.ufl.edu/UFE0021422/00001

Material Information

Title: Template Synthesis for the Development of Three-Dimensional Nanostructured Solid-State Lithium-Ion Batteries and Investigation of Carbon-Based Nanostructured Materials
Physical Description: 1 online resource (121 p.)
Language: english
Creator: Xu, Fan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: batteries, dimensional, ion, lithium, nanostructure, solid, state, synthesis, template, three
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Template synthesis is a general and powerful approach for preparing nanomaterials that entails synthesizing the desired materials within the pores of a micro or nanoporous membrane. In our research, two aspects based on template synthesis are addressed: the use of templated nanomaterials in developing 3-dimensional (3-D) solid-state Li-ion batteries and investigation of novel structures of carbon-based nanomaterials. Due to the planar configuration, traditional 2-D parallel-plate batteries face the challenges in achieving both high energy and high power density per unit area at the same time. A proposed concept of 3-D battery architecture shows the great promise by incorporating the 3-D nonplanar electrode configuration. In our research we developed a nanofabrication strategy to create a prototype 3-D solid-state Li-ion battery in which the template-synthesized nano-electrodes and the polymer electrolyte were used. The battery showed very stable cycling performance and high coulombic efficiency when operated at room temperature. The fast capacity decay with increasing discharge currents was observed. This is mainly attributed to the low ionic conductivity of the solid polymer electrolyte at room temperature. When increasing the temperature to a reasonable extent, we observed the delayed capacity drop and dramatically improved rate capability of the prepared battery. In comparison with currently reported 3-D batteries, our prototype solid battery incorporated with template-synthesized nano-electrodes and the polymer electrolyte showed the great advantages of high cell voltage, superior rate capabilities, stable cycling performance and improved safety due to its solid-state design. Carbon-based nanomaterials are rapidly growing area of research. They showed potential applications ranging from electronic, medical therapy, energy storage and conversion. Based on the combination of template synthesis and chemical vapor deposition (CVD), we investigated two novel carbon nanostructures: nanofibrous carbon/metal composite materials and carbon nanotubes with diamond-shaped cross-sections (DCNTs). Structure characterizations such as SEM, TEM indicated the prepared materials were freestanding, compact and highly dense fibers or tubes with controlled dimensions in nano-scale. These new carbon nanostrctures are promising candidates for nanoscale electrical interconnects, battery electrodes, sensing and field-emission devices.
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 Fan Xu.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Martin, Charles R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021422:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021422/00001

Material Information

Title: Template Synthesis for the Development of Three-Dimensional Nanostructured Solid-State Lithium-Ion Batteries and Investigation of Carbon-Based Nanostructured Materials
Physical Description: 1 online resource (121 p.)
Language: english
Creator: Xu, Fan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: batteries, dimensional, ion, lithium, nanostructure, solid, state, synthesis, template, three
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Template synthesis is a general and powerful approach for preparing nanomaterials that entails synthesizing the desired materials within the pores of a micro or nanoporous membrane. In our research, two aspects based on template synthesis are addressed: the use of templated nanomaterials in developing 3-dimensional (3-D) solid-state Li-ion batteries and investigation of novel structures of carbon-based nanomaterials. Due to the planar configuration, traditional 2-D parallel-plate batteries face the challenges in achieving both high energy and high power density per unit area at the same time. A proposed concept of 3-D battery architecture shows the great promise by incorporating the 3-D nonplanar electrode configuration. In our research we developed a nanofabrication strategy to create a prototype 3-D solid-state Li-ion battery in which the template-synthesized nano-electrodes and the polymer electrolyte were used. The battery showed very stable cycling performance and high coulombic efficiency when operated at room temperature. The fast capacity decay with increasing discharge currents was observed. This is mainly attributed to the low ionic conductivity of the solid polymer electrolyte at room temperature. When increasing the temperature to a reasonable extent, we observed the delayed capacity drop and dramatically improved rate capability of the prepared battery. In comparison with currently reported 3-D batteries, our prototype solid battery incorporated with template-synthesized nano-electrodes and the polymer electrolyte showed the great advantages of high cell voltage, superior rate capabilities, stable cycling performance and improved safety due to its solid-state design. Carbon-based nanomaterials are rapidly growing area of research. They showed potential applications ranging from electronic, medical therapy, energy storage and conversion. Based on the combination of template synthesis and chemical vapor deposition (CVD), we investigated two novel carbon nanostructures: nanofibrous carbon/metal composite materials and carbon nanotubes with diamond-shaped cross-sections (DCNTs). Structure characterizations such as SEM, TEM indicated the prepared materials were freestanding, compact and highly dense fibers or tubes with controlled dimensions in nano-scale. These new carbon nanostrctures are promising candidates for nanoscale electrical interconnects, battery electrodes, sensing and field-emission devices.
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 Fan Xu.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Martin, Charles R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021422:00001


This item has the following downloads:


Full Text

PAGE 1

1 TEMPLATE SYNTHESIS FOR THE DEVE LOPMENT OF THREE-DIMENSIONAL NANOSTRUCTURED SOLID-ST ATE LITHIUM-ION BATTERIES AND INVESTIGATION OF CARBON-BASED NANOSTRUCTURED MATERIALS By FAN XU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Fan Xu

PAGE 3

3 To my beloved parents and husband

PAGE 4

4 ACKNOWLEDGMENTS During the past years of my Ph.D. study, I have been always supported, advised and encouraged by many individuals around me. Th ey are behind each progress I made along the way. First of all, I thank my advisor, Dr. Charles R. Martin, for his greatest guidance, assistance and encouragement throughout my Ph.D research. He mentored me to not only grow as an experimentalist and chemist but also as an inde pendent thinker and effective communicator. I am and will continue to be nefit from all of these. I thank Martin group including the past and pres ent members. They are the indispensable part of my research life in UF. I would never forget the great time I experienced with them. Thanks for Dr. Naichao Li and Dr. Charles Si des in helping me on th e battery research I conducted all these years in Martin group. I appr eciate Dr. Lane Baker shared the insightful advices on my research. Dr. Myungchan Kang showed great patience in training me in electrochemical deposition and plasma etchi ng. I thank Dr. John Wharton for his great contribution on our collaborated research. I appreciate the grea t friendship with Dr. Heather Hillebrenner. Her positive and humorous attitu de will be invaluable for me. I enjoy the wonderful time with Dr. Hitomi Mukaibo, Dooho Park, Mario Caicedo, Jiahai Wang, Pu Jin, Lloyd Horne, Warren Mino, Kaan Kececi, Lindsay Sexton, Funda Tongay, Peng Guo. I am truly grateful to have the opportunity to work with a ll of these talented peopl e and will miss them very much in future. I give special thanks to Major Analytical Inst rumentation Center in De partment of Material Science and Engineering in UF. Dr. Luisa Am elia Dempere, Wayne Acree and Kerry Siebein provided me the great help in instrumentati on training and then allowed me to access the instruments by myself.

PAGE 5

5 Finally, I give my sincere appreciation to my fa mily. My mother and father are always the strongest support behind me no ma tter where I am. They are also my terrific examples of braveness and perseverance. I thank my bel oved one-Dr. Tao Wang-for his great advice and unanimous support on my research and personal life.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......10 ABSTRACT....................................................................................................................... ............13 CHAPTER 1 INTRODUCTION AND BACKGROUND...........................................................................15 Introduction................................................................................................................... ..........15 Nanostructured Materials by Template Synthesis..................................................................15 Template used..................................................................................................................16 Track-etch membrane...............................................................................................16 Porous anodic aluminum oxide (AAO) membrane..................................................19 Synthetic methods...........................................................................................................20 Sol-gel deposition.....................................................................................................20 Chemical vapor deposition.......................................................................................22 Electrochemical deposition......................................................................................23 Electroless deposition...............................................................................................25 Application of Template-Synthesi zed Nanostructured Materials...........................................25 Energy storage and conversi on lithium-ion batteries....................................................25 Nanostructured battery electrodes............................................................................27 Three-dimensional battery configurations...............................................................29 Variations on a synthetic them e of carbon-based nanomaterials....................................31 Dissertation Overview.......................................................................................................... ..32 2 TEMPLATE SYNTHESIS FOR THE DE VELOPMENT OF 3-DIMENSIONAL NANOSTRUCTURED SOLID-ST ATE LI-ION BATTERIES............................................41 Introduction................................................................................................................... ..........41 Experiment..................................................................................................................... .........42 Materials...................................................................................................................... ....42 Preparation of the nanofibrous V2O5 cathode.................................................................42 Preparation of the carbon nanotube membrane anode....................................................43 Preparation of the polymer electrolyte............................................................................44 Assembling of 3-D nanostructu red solid-state batteries..................................................44 Nano-LiV2O5/PEO-LiClO4/Li solid-state battery....................................................44 Nano-V2O5/PEO-LiClO4/nano-LixC solid-state battery...........................................45 Electrochemical characterization.....................................................................................46 Scanning electron microscopy.........................................................................................46

PAGE 7

7 Results and Discussion......................................................................................................... ..47 Nano-LiV2O5 / PEO-LiClO4 / Li solid-state battery.......................................................47 Morphology of nanofibrous V2O5 cathode...............................................................47 Electrochemical characteri zation of nanofibrous V2O5 cathode..............................47 Electrochemical tests of nano-LiV2O5/PEO-LiClO4/Li solid-state battery..............49 Nano-V2O5 / PEO-LiClO4 / LixC solid-state battery.......................................................50 Morphology of carbon nanotube membrane anode..................................................50 Electrochemical characterization of carbon nanotube membrane anode.................51 Morphology of polymer electrolyte coated nanostructured electrodes....................51 Electrochemical tests of nano-V2O5/PEO-LiClO4/LixC solid-state battery.............52 Conclusion..................................................................................................................... .........56 3 TEMPLATE SYNTHESIS OF NA NOSTRUCTURED CARBON/METAL COMPOSITE MATERIALS AND THEIR A PPLICATION IN LI-ION BATTERIES.......72 Introduction................................................................................................................... ..........72 Experiment..................................................................................................................... .........73 Materials...................................................................................................................... ....73 Electrodeposition of metal fibers inside the alumina template.......................................73 Preparation of the nanostructured carbon........................................................................74 Scanning electron microscopy and EDS analysis...........................................................75 Electrochemical characterization.....................................................................................75 Results and Discussion......................................................................................................... ..76 Structure characterization and elemental analysis...........................................................76 Electrochemical characte rization of nanostructured carbon/metal composite................79 Cyclic voltammetry..................................................................................................79 Galvanostatic charge/discharge................................................................................79 Conclusion..................................................................................................................... .........80 4 TEMPLATE SYNTHESIS OF CARBON NA NOTUBES WITH DIAMOND-SHAPED CROSS-SECTIONS...............................................................................................................94 Introduction................................................................................................................... ..........94 Experiment..................................................................................................................... .........95 Materials...................................................................................................................... ....95 Preparation of track-etched mica membranes.................................................................95 Preparation of diamond shaped carbon nanotubes (DCNTs)..........................................95 Structure characterization................................................................................................96 Results and Discussion......................................................................................................... ..97 Scanning electron microscopy of the mica su rface before and after deposition of carbon......................................................................................................................... ..97 Electron microscopy of the liberated DCNTs...............................................................100 Conclusion..................................................................................................................... .......101 5 CONCLUSIONS..................................................................................................................110 LIST OF REFERENCES.............................................................................................................113

PAGE 8

8 BIOGRAPHICAL SKETCH.......................................................................................................121

PAGE 9

9 LIST OF TABLES Table page 1-1 Comparison of commerci al rechargeable batteries...............................................................39 4-1 Characteristics of the two mica temp late membranes used for these studies.......................104 4-2 Characteristics of the DCNTs.............................................................................................. .106

PAGE 10

10 LIST OF FIGURES Figure page 1-1 Scanning electron micrograph of th e surface of a commercial track-etch polycarbonate membrane with nominal pore size of 50 nm..............................................34 1-2 Scanning electron micrograph of the surface of a track-etched muscovite mica membrane....................................................................................................................... ....35 1-3 Scanning electron micrograph of the su rface of a commercial alumina template membrane with nominal pore size of 200 nm....................................................................36 1-4 Scheme of CVD synthesis of carbon in an alumina template membrane..........................37 1-5 Scheme of synthesis of nanofibrous ma terials by electroche mical deposition in a porous alumina template....................................................................................................38 1-6 A schematic of a Li-ion batte ry charge/discharge process................................................40 2-1 Scheme for the template-synthesis of a nanofibrous V2O5 electrode................................58 2-2 Scheme for preparation of a nanostr uctured solid-state Li-ion battery..............................59 2-3 SEM images of (A) a polycarbonate membra ne with nominal pore diameter of 50 nm and (B) a synthesized V2O5 cathode removal of the polycarbonate template...................60 2-4 Cyclic voltammogram of th e template-synthesized nano-V2O5 electrode. Scan rate = 0.5 mV s-1...........................................................................................................................61 2-5 Galvanostatic charge/discharge curves of the template-synthesized nano-V2O5 electrode. Current = 5 A..................................................................................................62 2-6 Galvanostatic charge/disch arge curves for the nano-LiV2O5/PEO-LiClO4/Li solidstate battery. Potential range is between 2.5 and 3.8 V. Temperature = 25 oC................63 2-7 Discharge capacity of the assembled nano-LiV2O5/PEO-LiClO4/Li solid-state battery at 25 oC as a function of galvanos tatic discharge currents....................................64 2-8 SEM images of (A) an alumina membra ne with nominal pore diameter of 200 nm and (B) a synthesized CNTs anode afte r removal of the alumina template.......................65 2-9 Galvanostatic charge/disch arge curves of the template-synthesized CNTs membrane electrode. Charge current = 100A, discharge current = 50 A.......................................66 2-10 SEM images of the PEO-LiClO4 coated nanofibrous V2O5 after removal of polycarbonate template. (A) Low magnificati on view. (B) High magnification view......67

PAGE 11

11 2-11 SEM images of the PEO-LiClO4 coated CNTs membrane with the alumina template. (A) Low magnification view. (B ) High magnification view.............................................68 2-12 Galvanostatic charge/disch arge curves for the nano-V2O5/PEO-LiClO4/nano-LixC solid-state battery. (A) Current = 14 A/cm2. (B) Current = 32 A/cm2..........................69 2-13 Galvanostatic charge/disch arge curves for the nano-V2O5/PEO-LiClO4/nano-LixC solid-state battery. Current = 119 A/cm2. Temperature = 67 oC.....................................70 2-14 Discharge capacity of the assembled nano-LiV2O5/PEO-LiClO4/nano-LixC solidstate battery at 25 oC and 67 oC as a function of discharge currents.................................71 3-1 Schematic of the catalyst-CVD-based template synthesis of nanostructured carbon/metal composite materials within an alumina template.........................................82 3-2 Schematic of the electrochemical depos ition cell used for the deposition of metal fibers inside the pores of an alumina template...................................................................83 3-3 Scanning electron micrograph of the su rface of a commercial alumina template membrane with nominal pore size of 200 nm....................................................................84 3-4 SEM and EDS of Ag/Ni fibers. (A) the Ag film. (B) EDS of the sample circled in (A). (C) the deposited Ni nanofibers. (D) EDS from th e sample circled in (C)................85 3-5 SEM and EDS of Ag/Ni/Carbon fibe rs. (A) low-magnification. (B) highmagnification. (C) EDS of sample circled in (B). (D) EDS of sample squared in (B)......87 3-6 SEM and EDS of sample in Figure 3-5 af ter removing metal fibers and template. (A) low-magnification. (B) high-magnification. (C) EDS of the sample in (B)......................89 3-7 SEM images of Cu/Ni/Carbon fibers. (A) cr oss-section of the sample before removal of the template. (B) topview of the sample after removal of the template......................91 3-8 Cyclic voltammetry of the Cu/Ni/Carbon fibe rs after removal of the template. Scan rate = 10 mV s-1. Scan potential ranges between 0 and 3 V (vs. Li+/Li)...........................92 3-9 Galavanostatic charge/discharge of the Cu/Ni/Carbon fibers after removal of the template. Current = 60 A. The potentia l ranges between 0 and 1 V (vs. Li+/Li)............93 4-1 Diagram of a diamond-shaped pore in track-etched mica. al and as represent the long and short axes, respectively. represents the included angle made by al and as.............102 4-2 SEM images of track-etched mica. (A) Etched in 10 wt.% HF for 190 min. (B) Etched in 20 wt.% HF for 10 min....................................................................................103 4-4 SEM images of the liberated DCNTs. (A) and (B) DCNTs with large equivalentdiameter pores. (C) and (D ) DCNTs with the small eq uivalent-diameter pores............107

PAGE 12

12 4-5 SEM images of the small equivalent-diame ter DCNTs protruding from the still-intact lower carbon surface film. (A) low-magnification. (B) high-magnification..................108 4-6 TEM images of the small equivalent-d iameter DCNTs. (A) low-magnification. (B) high-magnification. (C) Electron diffracti on data for an individual DCNT....................109

PAGE 13

13 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TEMPLATE SYNTHESIS FOR THE DEVE LOPMENT OF THREE-DIMENSIONAL NANOST RUCTURED SOLID-STATE LITHIU M-ION BATTERIES AND INVESTIGATION OF CARBON-BASED NANOSTRUCTURED MATERIALS By Fan Xu August 2007 Chair: Charles R. Martin Major: Chemistry Template synthesis is a general and powerful approach for preparing nanomaterials that entails synthesizing the desired materials within the pores of a micro or nanoporous membrane. In our research, two aspects based on template synthesis are addressed: the use of templated nanomaterials in developing 3-dimensional (3-D) so lid-state Li-ion batteries and investigation of novel structures of carb on-based nanomaterials. Due to the planar configuration, traditional 2-D parallel-plate batteries face the challenges in achieving both high energy and high power density per unit area at the same time. A proposed concept of -D battery architecture shows the great prom ise by incorporating the 3-D nonplanar electrode configuration. In our research we developed a nanofabrication strategy to create a prototype 3-D solid-state Li-ion battery in which the template-synthesized nanoelectrodes and the polymer electrolyte were used. The battery showed very stable cycling performance and high coulombic efficiency when operated at room temperature. The fast capacity decay with increasing discharge currents was observed. This is mainly attributed to the low ionic conductivity of the solid polymer electrolyte at room temperature. When increasing the temperature to a reasonable extent, we observe d the delayed capacity drop and dramatically

PAGE 14

14 improved rate capability of the prepared batter y. In comparison with currently reported 3-D batteries, our prototype solid battery incorporat ed with template-synthesized nano-electrodes and the polymer electrolyte showed the great adva ntages of high cell voltage, superior rate capabilities, stable cycling pe rformance and improved safety due to its solid-state design. Carbon-based nanomaterials are rapidly grow ing area of research. They showed potential applications rangi ng from electronic, medical therapy, energy stor age and conversion. Based on the combination of template synthesis and ch emical vapor deposition (CVD), we investigated two novel carbon nanostructures: nanofibrous carbon/metal composite materials and carbon nanotubes with diamond-shaped cros s-sections (DCNTs). Structur e characterizations such as SEM, TEM indicated the prepared materials we re freestanding, compact and highly dense fibers or tubes with controlled dimensions in na no-scale. These new carbon nanostrctures are promising candidates for nanoscale electrical interconnects, battery electrodes, sensing and fieldemission devices.

PAGE 15

15 CHAPTER 1 INTRODUCTION AND BACKGROUND Introduction Since firstly predicted by Richard Feynman in 1959,1 nanotechnology has been of tremendous interest from both a fundamental and an applied perspective. Currently, it is a cutting-edge technology in wide areas ranging from material scie nce to device fabrication. The impetus for nanotechnology stems from the discover y of the unique properties of materials in nanoscale compared to its bulk scale. And the esse ntial of this technology is to control a matter on the nanometer scale as well as fabricate devi ces on the same length scale. Exploration of diverse applications is a critical is sue in todays nanotechnology research. Among numerous (chemical a nd physical) methods for preparing nanomaterials,1, 2 one technique called template synthesis has been wi dely explored in Charles Martin group. With this method, a variety of nanomaterials includi ng organic and inorganic materials have been prepared and their related applic ations have been explored. On e big field explored in Martin group is the nanomaterials applied for energy storage and conversion, in particular, Li-ion batteries. Despite the commercial success of Li-i on batteries during the past decade, the more demanding and exotic application of these batteries arise the ne w challenges for both battery materials and battery systems. One possible so lution revealed by fundame ntal research is to integrate nanotechnology, for example, nanoscaled battery components. My research explored the imp lication of template synthesis especially in nanostructured Li-ion batteries and nanostructur ed carbon-based materials. Nanostructured Materials by Template Synthesis The history of template synthesis applied in nanomaterials can be traced back to 1970, in which metal nanowires were prepared by Possin.3 However, until the late 1980s and early 1990s,

PAGE 16

16 this method was re-explored and became a general method for preparing nanomaterials.4-6 As the name suggested, template synt hesis involves synthesi zing materials by using a template as a mold. The typical template is a solid membrane which contains cylindric al microor nanopores. By depositing the desired materials into each pores of the template, micro-or nanostructured materials can be obtained. Common templates in clude organic (i.e. poly mer) and inorganic (i.e. anodized alumina) membranes. In most cases, the porous templates we used contain pores in nanometer size and thickness in micrometer, whic h gives the aspect rati o on the order of 10. Depending on the deposition method and interactions between the deposited material and the pore wall, the resulting nanostructure s can be either tubula r (hollow) or fibrous (solid). The host template membrane can then be removed in or der to expose the nanostructured materials. Removal of the template can be achieved by chemical (i.e. acid/base dissolution) or physical (i.e. plasma etching) methods, which should be compatib le with the synthesized materials. Martin group has applied template synthesis method in pr eparing a variety of nanostructured materials including metal,7, 8 semiconductor,9 polymer,10 carbon,11, 12 and Li-ion battery materials.13-18 Template Used Track-Etch Membrane The term track-etch refers to the pore-produc tion process in two steps.19, 20 In the first step, a solid membrane (usually dielectric materials) is exposed to a collimated beam of high-energy nuclear fission fragments, which leaves randomly-dis tributed damage-tracks in the film. This high energy (on the order of 2 GeV) of the fragme nts ensures that the tr acks span the entire length of the membrane (typically from 6 to 10 m). After this step, the reactive chains end at the damage-tracks in the film. In the second step the latent ion tracks wi th the reactive chains are chemically etched with a strong base solu tion to form uniform pores. Usually before chemical etching, irradiation of the film is performed in that it al lows the latent ion track to be

PAGE 17

17 preferentially etched by the chemical etchant. One ion creates one track, which in turn becomes one pore. Pore densities depend on the fluence of the heavy ion beam and the duration that the film is exposed to the fission fragment beam Pore sizes depend on the chemical etching conditions such as etching time, etchant temper ature and strength. Po re geometries depend on the membrane material, etchan t composition, temperature, etc. Most micro-and nanoporous track-etched membranes are commercially available ( e.g., GE Osmonics) in a variety of pore geometries, pore sizes and membrane materials.21 Pore densities range from 104 to 108 pore cm-2 and pore diameters range from 10 nm to 20 m.21 Among various commercial track-etch membrane s, polycarbonate is one of widely used membranes in Martin group. A scanning electron micrograph of the su rface of a polycarbonate template is shown in Figure 1-1. The chem ical composition of poly carbonate is (OOC-O-C6H4C(CH3)2-C6H4). During etching, the chemical bonds on both sides of th e carbonate group (-O(C=O)-O-) are attacked by a basic etchant and ruptured to form carbonate ions CO3 2-, while diphenylol HO-C6H4-C(CH3)2-C6H4-OH is also formed at the same time. The commercial polycarbonate membrane contains cylindrical pores, but as the pore diameter is reduced to the smaller nanoscopic dimensions the shape of the por e becomes like a cigar, slightly tapered at the ends. The formation of bottleneck pores is at tributed to following two possible explanations: one is because the fission fragment which crea tes the damage track also generates secondary electrons. These secondary electrons contribute to the damage along the track too. The number of secondary electrons generated at the faces of the membrane is less than in the central region of the membrane, and leads to the formation of bot tleneck pores; the second explanation is that the surfactant protective layer adsorbed to the surface of the membrane retards the local etching process.20

PAGE 18

18 In the production of template-synthesi zed nanostructured electrodes for Li-ion batteries, track-etched polycarbonate templates are commonly used. This is because these templates can be easily removed under conditions which do not adve rsely affect the nanostructures themselves. The wide variety of commercially available pore sizes and pore densities can generate comparative structures of differing geometries that are key tools for fundamental study. However, one disadvantage of track-etch poly carbonate for their app lication to electrode materials is its low porosity (2-10%). And this low porosity represents the low ratio of battery active materials occupied in a given footprint area or volume on the current-collector surface. Other than polymer templates, another type of track-etch templates used for synthesizing nanostructured materials are inorgani c single crystal muscovite mica (KAl2(AlSiO3O10)(OH)2) (Figure 1-2). Compared to the amorphous polym er membrane, track-etched mica has several unique properties. First, the surface of etched mica template is extremely smooth. It is a good candidate to be the platfo rm for AFM imaging of DNA22 and support layer for lipid bilayers.23 Second, the track-etched muscovite mica always has diamond-shaped pores with inner angles close to 60 and 120 and all the pores have same size, shape and orientation.24 By correlating the results of scanning electron microscopy and X-ray diffraction on etched mica crystals, it was found that the orientation of the diamonds is exactly the same as that of the mica unit cell. The four sides of the diamonds are parallel to the four oxygen-terminated planes within the unit cell. These facts point out that the diamondshaped pores have thei r origin in the mica crystal structure.24 The slowest etching planes are the ones terminated by oxygen.24, 25 As a result of this etch-rate anisotropy, pores etched in mica have diamond-shaped cross-sections, rather than the circular-shaped pores produced in amorphous polymer membranes.

PAGE 19

19 Porous Anodic Aluminum Oxide (AAO) Membrane The anodic aluminum oxide (AAO) membrane is one of the widely used nanopore templates. It is fabricated by anodization of aluminum metal in an acidic environment which causes the metal to be etched into a porous structure.26-31 Unlike the track-etch process, the aluminum anodization is highly systematic, wh ich generates a highly-ordered hexagonal pore array with cylindrical pore shapes. The pore dens ity of alumina template can be prepared on the order of 1011 pore/cm2, which is about 3 orders of magnit ude higher than th e pore density of track-etched polycarbonate. The high pore density of the template results the high porosity (~50%). These alumina template are commercial available as filter membranes (Whatman) as one shown in Figure 1-3. The available nominal pore sizes are 20 nm, 100 nm and 200 nm.32 The thickness of the membrane ranges from 10-1 00 m. However, pores in commercial alumina membranes branch into many smaller pores near th e tips of one face of the membrane due to the voltage reduction technique used during the anodization.31 The inorganic alumina membranes have much greater mechanical stability and chem ical resistivity than po lycarbonate. The great mechanical stability is of a benefit for integrating the alumina template into the device assembling.33, 34 However, the chemical resistivity indicates the harsh alumina dissolution conditions, and this could be a disadvantage for preparing most of battery electrode materials with one exception, carbon. For chemical vapor deposition (CVD) of carbon within the alumina template, it is the chemical inertness of this template that allows carbon to be deposited under high temperatures. Although ther e are several other techniques35-38 for preparing nanostructured carbon, CVD-based template synthesis of carbon within the alumina template provides the unique advantages such as high density, m onodisperse pore distribution, well-alignment, simplicity and low-cost. By varying the expe rimental conditions and the pore geometries of

PAGE 20

20 alumina templates, carbon with di fferent properties such as crysta llinity and geometries can be prepared.39-41 It is also notable that the alumin a structure is electronically insulating. Our lab also synthesizes home-grown alumin a template. The detailed synthetic process is described in our previous papers.42, 43 These templates are similar to the commercial ones but without pore branching near the tips of one face of the membrane. This is often preferred for use as templates. Also, control over the synthetic conditions allows us to tailor pore diameters and thickness for our specific applicat ion. The applications of these alumina templates that have demonstrated in our lab alone are as diverse as enantiometric-separations,44 templates for carbon nanotubes,12 plasma-etching masks for prepari ng nanostructured battery materials.14 Synthetic Methods As we mentioned earlier, te mplate synthesis method invol ves two parts: template membranes and material synthesis. The strategi es of material-depositio n employed in template synthesis are usually adapted from those for prepar ing bulk materials. Based on the properties of the membrane materials, different deposition stra tegies are applied for preparing compatible nanostructured materials. Until now, the depos ition strategies used in Martin group include electrochemical 43, 45, 46 and electroless deposition,47, 48 chemical49 and electrochemical polymerization,50 sol-gel deposition,15, 51 chemical vapor deposition.12, 39, 52, 53 Sol-Gel Deposition Sol-gel chemistry is a powerful route for prep aring inorganic material s. Although firstly discovered in late 1800s, sol-ge l technique gained renewed inte rest in the early 1970s for its extensive application in the production of glasses. As the name implies, sol-gel process typica lly entails the hydrolysis of a solution of a precursor molecular to obtain firstly a suspension of colloidal practical (sol) and then form a continuous liquid-filled solid ne twork (gel) composed of aggr egated sol particles. The

PAGE 21

21 amorphous gel may then be thermally treated to yield a more crystalline product. Compared with other methods for preparing inorganic materi als, sol-gel chemistry provides several unique advantages. For example, high-purity materials can be synthesized at a lower temperature. Homogenous multicomponent materials can be obta ined by mixing their precursor solutions; this allows for easy chemical doping of the materials prepared.54 Martin group has investigated the use of sol-gel-based template synthesis within the pores of the alumina and polycarbonate membrane to prep are nanostructured inor ganic oxide materials. These include semiconductors (TiO2, WO3, ZnO, SiO2) and Li-ion battery intercalation materials (V2O5, SnO2).9, 13, 15, 16, 51 In a typical sol-gel template-synthesis process, a template membrane is firstly immersed into a precursor solution of the material for a given period of time to allow the precursor deposit in the pores. Then the pr ecursor is hydrolyzed into the sol under desired conditions, followed by the further drying and he at-treatment to conve rt into the desired materials. Same as other template synthesis methods, longer immersion yield fibers while short immersion produces tubes. The formation of tube s after a short immersion indicates that the sol particles adsorb to the pore walls of the templa te membrane. This can be explained due to the electrostatic interaction betw een the positively charged semiconductor sol particles and negatively charged pore walls. 54 Also, it has been found that the ra te of gelation is faster within the pore than in bulk solution. This is most likely due to the enhancement in the local concentration of the sol pa rticles owing to their adso rption on the pore walls. The precursors used in sol-gel chemistry consist of a metal or metalloid element surrounding by various reactive ligands. The most popular precursor is metal alkoxides because they can easily react with water. Some examples include tetraethoxysilane (TEOS), tetramethoxysilane (TMOS) and triisopropoxyvanadium (TIVO). In Chapter 2, I will discuss the

PAGE 22

22 application of sol-gel chemis try of TIVO in preparing the nanostructured cathode for Li-ion battery research. Chemical Vapor Deposition (CVD) Chemical vapor deposition (CVD) is a versatil e process in which gas-phase molecules are decomposed to reactive species, lead ing to film or particle growth. 55 CVD can be used to deposit a variety of conducting, se miconducting and insulating materi als. Two advantages can be obtained from CVD technique: the ability to controllably cr eate films of widely varying stoichiometry and to uniformly deposit thin film s of materials, even onto nonuniform shapes. The most prevalent application of CVD is the deposition of hydro carbon precursor into a porous host (i.e. zeolite, inorganic membrane) to pr epare nanostructured carbon.11, 12, 39, 56 Martin group and others have synthesized the nanotubular/nanof ibrous carbon that involves CVD of carbon within the pores of alumina templates.8, 11, 12, 39, 57-59 The heat-tolerance and high pore-density of alumina membranes make them good templates for CVD. Moreover, alumina templates can be chemically dissolved after CVD pr ocess, leaving an array of carbon tubes/fibers. The schematic of a general experimental setup of CVD is shown in Figure 1-4. This involves placing an alumina membrane in a high temperature furnace (~700 oC) and passing a carbon precursor gas such as ethylene or propylene th rough the membrane. Thermal decomposition of the carboneous gas occurs thr oughout the pores, resulting in the deposition of carbon along the length of the pore walls (i.e. carbon tubes are formed within the pores). As a characteristic of template synthesis, the outer diameter of the car bon tubes is dependent on the diameter of pores in the template, while the inne r diameter of the tubes depends on the CVD duration, precursor gas pressure and flow rate. An inert gas such as Ar flows during the he ating and cooling stages of the CVD process. During the heating process, this inert ga s ensures deposition begins only after the desired temperature. Also, this iner t gas prevents oxidizati on of freshly deposited

PAGE 23

23 carbon to CO2 during the cooling process. It is not eworthy that during CVD both faces of a template membrane are also coated with carbon, which turns to be carbon films connected on the both ends of carbon nanotubes or fibers. Typically, the CVD-synthesized carbon nanostructures have the cylindrical geometry and the circular cross-section due to the cylindrical pores in the alumina template. Recently, Masuda et al. developed a new type of alumina template s with non-circular cr oss-sections and noncylindrical geometries (i.e. tria ngular and square-shaped pores). 41, 40 By using the template with triangular-shaped pores, triangular-shaped car bon nanotubes were prepared by CVD in this template. Also, in the previous section of track-etch membrane, one type of track-etch templates, single crystal mica, was described. This template has unique diamond-shaped pores and ultra-smooth surface due to its highly crysta lline structure. Very recently, Martin group explored CVD synthesis of a diamond-shaped ca rbon nanotube within this mica template. The detailed discussion will be presented in Chap ter 4. Compared to the conventional carbon nanostructures with cylindrical geometries, these non-cylindri cal carbon nanostructures might have potential applications, fo r example, in modifying fiel d-emission properties of CNTs.60 Electrochemical Deposition Electrochemical deposition, also interchangeably named as electroplating, has thrived as an industrial technique for metal de position for several decades. An electrochemical deposition cell consists of an anode (often sacrificial), an electrolyte (t he target-ion source) and a cathode (target plating site). By passing an applied stimulus (current or potential), the target metal ions can be reduced on the cathode to form solid metal film. The thickness of th is film is governed by quantity of charge applied (current time) and the number of electrons required per reduction reaction.

PAGE 24

24 In Martin group, we adapted this establishe d process into our template synthesis for electrochemically depositing nanostr uctured materials into porous templates. The schematic of this process is shown in Figur e 1-5. Both AAO and track-etched polycarbonate have been used as the template membranes. During the depositi on process, one face of the porous membrane is firstly coated with a thin metal film via ion-sputtering or thermal evaporation. This metal film serves as the cathode in this configuration for redu ction of metal ions (deposition of metal) inside pores of the template. Generally, the synthesi zed materials grow from the bottom upwards as solid fibers. Depending on the amount of metal deposited, either short or long fibers can be obtained. In our group, a variety of nanofibrous metals has been prepared including copper, platinum, gold, silver and nickel fibers.7, 43, 61-65 However, with modest pretreatment of th e template, hollow metal tubes can also be obtained by this deposition strate gy. It is done by de rivatizing the pore wa ll with a molecular anchor, prior to electrochemical deposition. The challenges for this method are to identify chemistry for forming the metal within the pores of the template, to identify a suitable molecular anchor and to develop chemistry fo r attaching this anchor to the pore walls in the template. By using organocyanide silane-alumina chemistry, Martin group has prepared Au nanotubes with controllable inner diameter in alumina templates. 7 Other than deposition of metals, Martin gr oup also demonstrated the electrochemical deposition of nanostructured polymers within na nopore templates. Various conductive polymer materials with nanotubular or nanofibrous structur es have been prepared including polypyrrole, poly(3-methylthiophene) and polyaniline.66 At the beginning of th e deposition, polymers grow as tubes inside the template. This is due to the solvophobic interac tion between the soluble monomers and insoluble polycationic-forms, by whic h the polymers are pulled to the pore walls.

PAGE 25

25 Also, the electrostatic interaction pulls the ca tionic polymer to the anionic sites on the pore wall.66 Depending on the deposition duration, thin-w alled or thick-walled or solid-fibrous polymer can be obtained. Electroless Deposition Different with electrochemical deposition, el ectroless deposition does not require the electronically-conductive substrate. This is beca use this deposition method relies on the use of a chemical reducing agent to deposit metal from th e solution onto a surface.67 Metals can be deposited into a porous template without conductiv e coating of the templa te. Martin group has developed the methods for electroless-depositing gold and other metals into the porous alumina and polymer templates.8, 47, 68, 69 In the gold plating process, the polycarbonate membrane is firstly sensitized by Sn (II) ions, followed by being replaced with the noble Ag metal particles in a AgNO3 solution, then followed by being replaced with the noble Au metal particles in a commercial Au plating solution. As a result, a ll the exposed surfaces of the template (including the two faces and the pore walls) are coated with Au particles so that the Au tube spanning the entire length of the template is formed. Same as electrochemical deposition, the thickness of the plated metal can be controlled by the plating durati on. However, the ability to control at will of the inner diameter of a tube is a distin ct advantage over electrochemical deposition. Application of Template-Synthes ized Nanostructured Materials Energy Storage and Conversion Lithium-Ion Batteries Introduced in the early 1990s by T. Nagaur a and K. Tozawa in SonyTec Inc., Li-ion batteries have occupied the major power-source market for portable and lightweight electronic devices.70 Table 1-1 compares severa l types of commercial rechar geable batteries. Among all the batteries, Li-ion batteries show the advantages in high cell voltage, high energy (Wh L-1, Wh kg-1) and power density (W L-1, W Kg-1). Figure 1-6 shows the schematic of the basic

PAGE 26

26 configuration of a Li-ion battery. Same as any battery syst em, a Li-ion battery consists of two electrodes (cathode and anode) in contact with an ion-conductive electrolyte. The cathode and anode are the electrodes at which reductions a nd oxidations occur during the battery discharge, respectively. Regardless of th e direction of current flow, th is convention (adopted from the discharge process) is obeyed. Li-ion batteries operate by revers ibly intercalating Li ions in each of the two electrodes. Intercalation is a process by whic h guest species (Li ions) reversib le insert/extract from the host matrix (electrode material) without significantly ch anging the structure of the host. As shown in Figure 1-6, during battery discharge process, Li ions deintercalate from the anode (i.e. carbon), migrate through a Li-ion conducting electrolyte, and then intercal ate into the cathode (i.e. V2O5). The ions then must rely on solid-state diffusi on to fill the non-surface intercalation site. Meanwhile, to compensate for the movement of each Li ion, an electron must travel through the electrical circuit and provide power to the load. During battery charge pr ocess, above process is reversed. The general forms of the charge/disch arge reaction at the cathode and anode of a Liion battery are shown in Equa tion 1-1 and 1-2, respectively. Cathode: MO2 + Li+ + xeLixMO2 1-1 Anode: 6C + xLi+ + xeLixC6 1-2 For the cathode materials in Equation 1-1, M re presents a single or mixed transition metal compound which can be readily reduced/oxidize d. Co, Ni, Mn and V are commonly used Discharge Charge Charge Discharge

PAGE 27

27 transition metals. Almost all the cathode materials can be divided into two classes of materials: 1) the layered compounds with an anion close-pack ed or almost close-packed lattice in which alternate layers between the an ion layers are occupied by a re dox-active transition metal, and then Li inserts into the essentially empty remaining layers. LiCoO2, LiNiCoO2 are the examples; 2) more open structure such as V2O5. The anode material shown in Equation 1-2 is carbon. It is the discovery of carbonaceous materials that led to the commercialization of Li-ion batteries in the 1990s.70 Substituting Li metals (anode material in the early Li-i on batteries) with carbonaceous materials obviated the short-circuit problem. When th e Li metal is used as anode, Li dendrites are formed on this anode during charge process. And these Li dendrites can penetrate through the separator membrane and ca use short-circuit of batteries. Until now, various types of carbonaceous materials have been investigated ra nging from highly ordered graphites to disordered carbons.71, 72 Nanostructured Battery Electrodes For battery electrode materials, decreasing particle-size can shorten the solid-state diffusion distance for Li ions a nd increase the specific surface area for Li ions intercalation. The Li ion diffusion distance, together with Li -ion solid-state diffusion coefficient, affects the concentration polarization of Li ions during their intercala tion/deintercalation process. Concentration polarization originates from the sl ow solid-state diffusion of Li-ions inside the solid electrode materials. It is well-documented that the typical diffusion coefficients of Li-ions are extremely low, in the range of 10-11 to 10-15 cm2 s-1 although this value depends on materials and state-of-charge. Therefore, when the Li-ions intercalate/deintercalate into an electrode, they are not able to solid-state diffu se rapidly enough to compensate for the facile nature of the insertion/extraction flux, and this results in concentration polarization.73 This problem is exacerbated when high currents are applied for a demanding application. Since the diffusion

PAGE 28

28 coefficient is an intrinsic propert y of electrode materials, we work with a strategy to decrease the electrode-particles size, for example using templa te-synthesized nanostructu red electrodes. This allows the intercalation sites to be closer to the surface, shortening th e distance over which the sluggish solid-state diffusion must propagate.13, 15, 16 Other than the advantage of short solidstate diffusion distance, nanostruc tured electrode materials provi de increased surface area per total volume fraction. This can decrease the current density at any give rate so that the sluggish electronic-kinetics of the system can be improved. The general procedure for template-synthe sizing nanostructured battery electrodes involves the use of a porous membrane with cy lindrical, monodisperse nanopores. By using chemical or electrochemical depos ition routes, the electrode active material is filled into each pore. The material is then processed in a way to achieve the desired ch emical property which is capable for Li-ion intercalation. Meanwhile, the template can be removed by plasma etching or chemical dissolution. After removal of the templa te, the resulted material maintains the fibrous or tubular structure which mirrors the geometry (length, diameter and number density) of the pores in the template. Typicall y, fibers or tubes are 100-nm in diameter, 6 microns in length and 108 elements per cm2. This nanostructured electrode materi al will protrude from a lower current collector surface like the bristle of a brush. Each electrode is el ectrochemically characterized in a half-cell system. The commercial battery electrode consists of three main com ponents: the electrode active material, the polymer binder a nd the conductive material. Th e active materials are usually micrometer-sized particles with diameter of 2-20 m. These di screte particles are mixed with the polymer binder and conductive materials into sl urry and then coated onto a metal current collector. The conductive material is to improve the electronic conductivity of the system, which

PAGE 29

29 is very low in this system due to the point-contact of the particles. The polymer binder is used to hold the materials together and ensure the good p hysical contact with the current collector. However, in our template-synthesized nanostr uctured electrode, only the electrode active material as well as the lower metal current collec tor is needed while othe r inactive (not capable of Li-ion intercalation) component s are unnecessary. This is becau se 1) the parallel electronic conduction and high density of nanofirbours or nanotubular materials greatly reduce the electronic resistance of th e electrode material; 2) the structur es are made to directly protrude from the surface of the current co llector. Therefore, our templa ted electrodes do not decrease volumetric and gravimetric energy density or involve complicate analysis.74 Martin group have explored various temp lated nanostructured electrode materials including cathode13, 15, 17 and anode materials. 14, 16 All these materials have shown the superior rate capability (the ability to charge/discharge at high currents) than a traditional thin-film electrode at any given high-rate. As illustrate d above, this is undoubtly attributed to delayed concentration polarization resul ting from the reduced Li ion solid-state diffusion distance (micrometer vs. nanometer) and the increased surface area of the nano-electrodes. Three-Dimensional Battery Configurations Recently, more demanding and exotic applications of the batteries have emerging, such as the electric component in hybrid vehicl es and integrated power supply in microelectromechanical systems (MEMs). Th e battery with high energy and power density, miniaturized size and improved safety become a critical issue for leading the next-generation battery technology. However, curren t 2-D parallel-plate batteries wi th liquid (or gel) electrolytes are facing the difficulties to meet above requirements espe cially in powering devices with limited areal footprint, being integrated in to the solid-state device and the safety.75

PAGE 30

30 One way towards achieving above goals is to reconfigure current 2D battery design, for example, into 3-D configuration design.75 In traditional 2-D batteries, the thick planar electrode can provide high energy density but also increase the power loss due to the slow diffusion of ions inside the thick solid electrode materials. T hus, 2-D design trades off between the available energy density and the ability to de liver this energy without loss of the internal power. However, this problem can be mitigated in 3-D battery configuration. This is because in 3-D battery the electrodes are configured in 3-D nonplanar structure to increase th e energy density of the battery per unit area while the ion transport distance can still be maintained on 1-D at micro-or nanoscopic level to minimize internal power loss.75 Although the 3-D battery configuration is a new concept in Li-ion battery research, several strategies have been in progre ss and all of them are based on developing micro-or nano-scaled battery components. Teeters et al. assembled the arrays of Li -ion nanobatteries by infiltrating the V2O5 xerogel cathode and polymer electrolyte in to the nanopore alumina template, and then capped with a thin film of anode (Li metal or SnO2).34, 76 By combining colloidal crystal templating, nanocasting and CVD methods, 3D ordered macroporous(3DOM) carbon was prepared by Stein et al. .77, 78 And this 3DOM has been recently used to assemble an electrochemical cell by conformal coating a PPO/SPPO electrolyte layer, followed by infiltrating a V2O5 cathode.79 Another strategy in the 3-D battery research is developed based on microfabrication techniques. Unlike the nano-sc aled porous template used in the previous strategies, a photolithographic sili con or glass microchannel plate with holes of tens to hundreds nanometer in diameter was used for conforma l and sequential deposi tion of cathode, polymer electrolyte and anode.80, 81

PAGE 31

31 Although above strategies demonstrated the f easibility in assembli ng 3-D Li-ion batteries, the low cell operation voltage and poor cycling performance indicate some intrinsic problems in these battery systems. Furthermore, the requirem ent of incorporation of liquid electrolytes in most of the strategies prevents the further applications of th ese 3-D batteries in solid-state devices. Overall, the current 3-D ba ttery research is still in its ve ry early stage mainly due to the big challenge in ability of precise and controlle d assembly of battery components in micro-or nanoscopic level without affecti ng their individual functions. Variations on a Synthetic Theme of Carbon-Based Nanomaterials One-dimensional nanostructured carbon ha s attracted great interests from both fundamental and applied perspective.82, 83 Its advantageous propertie s include the high surface area, excellent thermal and chemical stability, biological compatibility, low cost, environmental benign and diverse synthesis approa ches. The investigated applica tions of nanostructured carbon range from energy storage and conversion,12, 14, 84, 85 catalyst supports,82, 83, 86, 87 nanoelectronics,88-92 chemical and bio-sensors.93-95 Currently, several types of carbon nanostructures have been explored including carbon nanotubes,11, 12, 35, 38, 39, 57, 93, 94 carbon nanofibers,39, 96-98 carbon aerogels,99-101 carbon onions.102, 103 Martin group, and others, has focused on synthesizing carbon nanofiber and nanotube ensembles using CVD-based template-synthesis methods.11, 12, 39, 49, 57-59 As the characteristic of template synthesis, the outer diameter of the prep ared fiber or tube is controlled by the diameter of the pore in the template, while the inner diamet er of the tube is controlled by CVD duration. Template-synthesized nanofibrous or tubular carbon has shown great potential for Li-ion batteries,12 fuel-cell technology12 and fundamental study of electroosmotic flow.11 However, with the more exte nsive applications of carbon na nomaterials, novel structures of carbon-based materials are needed to be explored. As we men tioned earlier in this chapter,

PAGE 32

32 template synthesis is a powerful tool in creati ng diverse nanostructured materials, which can be found for various applications. Therefore, by inve stigating a new type of the template membrane, or incorporating a new material-deposition stra tegy, or creating a new combination of deposition methods, one can continue in exploring novel ca rbon nanostructures which will be advantageous for certain applications or fundamental studies. In this research, two new types of carbon nanostructures are inve stigated: composite carbon/metal nanofibers and carbon nanotubes with non-cylindrical geometry and non-circular cross-section. Both are prepar ed by using CVD-based template-synthesis method. However, the combinations of different types of template membrane and deposition st rategies result in two different carbon-based nanostructures. These ne w carbon nanostrctures ar e promising candidates for nanoscale electrical interconnect, battery elec trodes, sensing and field-emission devices. Chapter 3 and 4 will give the thorough details of investigation of these new carbon-based nanomaterials. Dissertation Overview The ultimate goal of this research is to dem onstrate the capability of template-synthesized nanostructured materials in developing advanced miniature solid-state Li-ion batteries and to investigate new template-synthesized carbon na nostructures which ha ve great potential applications. It will begin with a discussi on of the application of template-synthesized nanostructured electrodes into assembling 3-D nanostructured solid-sta te Li-ion batteries. Compared to traditional thin-film Li-ion batterie s and other prototype 3-D Li-ion batteries, our assembled battery shows the unique advantages including high cell voltage, superior rate capability and cycling performance. Next, tw o new carbon-based nanostructures are introduced including carbon/metal composite nanofibers an d carbon nanotubes with diamond-shaped crosssections. Both structures are achieved base on variation of CVD-based template synthesis

PAGE 33

33 methods. These new carbon nanostructures show th e potential applications for electrical nanointerconnect, battery electrodes, se nsing and field-emission devices.

PAGE 34

34 Figure 1-1. Scanning electron micrograph of the surface of a commercial track-etch polycarbonate membrane with nominal pore size of 50 nm.

PAGE 35

35 Figure 1-2. Scanning electron micrograph of th e surface of a track-etched muscovite mica membrane.

PAGE 36

36 Figure 1-3. Scanning electron micrograph of th e surface of a commercial alumina template membrane with nominal pore size of 200 nm.

PAGE 37

37 Figure 1-4. Scheme of CVD synthesis of car bon in an alumina template membrane.

PAGE 38

38 Figure 1-5. Scheme of synthesis of nanofibrous materials by electroche mical deposition in a porous alumina template.

PAGE 39

39 Table 1-1 Comparis on of commercial rech argeable batteries. Lead-Acid Ni-Cd Ni-MH Li-Ion Voltage (V) 2 1.2 1.2 3.6 Energy density by weight (Wh/kg) 25 50~60 60~80 90~120 Energy density by volume (Wh/l) 80 150~170 180~220 220~300 Self-discharge (%M) 5 25 20~25 10

PAGE 40

40 Figure 1-6. A schematic of a Li-ion ba ttery charge/discharge process.

PAGE 41

41 CHAPTER 2 TEMPLATE SYNTHESIS FOR THE DE VELOPMENT OF 3-DIMENSIONAL NANOSTRUCTURED SOLID-STATE LI-ION BATTERIES Introduction Lithium ion batteries have become the most suitable power sources for aerospace, modern health care and portable electronic devices due to their significan tly high energy density and light weight.104, 105 Recently, with the rapid development of microeletromechanical system (MEMS) devices for the integrated-circuits(IC) industry, a power supply that is miniature in size and maximum in energy and power densit y per unit area is in great demand.106-109 3-D solid-state battery configurations, instead of the traditional 2-D parallel-plate configurations, can achieve high energy density per unit area wh ile keep low internal power lo ss due to the incorporation of 3-D nonplanar electrodes.75 Besides, the solid-state design fits batteries much better as the IC power supplies due to the flexible design, safe lightweight, and ease of processing into the electronic system.110-115 By using template synthesis method pioneer ed in Martin group, nanostructured battery electrodes with a variety of ma terials have been explored.12, 13, 15, 116-118 We have shown that such nanostructured (nanofibrous or nanotubules ) electrodes have improved rate capability compared with the conventional thin-film electrodes composed of the same material. This is due to the larger surface area and shor ter Li ion solid-state diffusion di stance (within nanometer scale) inside nanostructured electrodes.12, 13, 15, 16 In this study, we combined the advantages of nanostructured electr odes and 3-D solid-state battery design to develop 3-D nanostructured so lid-state Li-ion battery prototype. Such 3-D prototypes were assembled by sandwiching polymer electrolyte-coated na nostructured electrodes (cathode and anode) with current-c ollectors attached. The prelim inary results of the synthesis

PAGE 42

42 and characterization of nanostruc tured electrodes, battery fabri cation and electrochemical tests are presented below. Experiment Materials Commercial available alumina and polycarbonate filters were used as the template membranes for preparing nanostructured electro des. The alumina were purchased from Whatman Anapore (nominal pore diameter: 200 nm, thickness: 60 m, pore density: 1011 cm-2) while the polycarbonate were from Poretics (nominal pore diameter: 50 nm, thickness: 6 m, pore density: 108 cm-2). An ethylene/helium mixture (30% ethylene, Praxair) was used as the Chemical Vapor Deposition (CVD) carbon precur sor gas. Triisopropoxyvanadium (V) oxide TIVO (Aldrich) was used as the V2O5 precursor. Ethylene car bonate (Aldrich), diethyl carbonate (Aldrich) and LiClO4 salt (Aldrich) were used to pr epare the liquid electrolyte, both organic solutions were used as received. Polyethylene oxide (P EO, Mw 400,000, Aldrich), LiClO4 salt and tetrahydrofuran (THF) (H2O < 50 ppm, ACROS) were used to prepare polymer electrolytes. PEO and LiClO4 were vacuum-dried for 48 hr under 60 C and 120 C, respectively, before use. Preparation of the nanofibrous V2O5 cathode Sol-gel based template synthesis wa s used to prepare nanofibrous V2O5 cathode13, 15, and the experimental procedures are shown in Fi gure 2-1. Briefly, a piece of the polycarbonate membrane was placed on a Pt foil in a glove box filled with Ar. 0.6 L of V2O5 precursor TIVO was applied on the membrane surface so that TIVO filled the pores in the template. Next, hydrolysis of TIVO was carried out in a glove box under low water partial pressure for 12 hr at room temperature, followed by heating in air at 80 C for 2 hr. Then, the templates were removed by the oxygen plasma with a plasma re active-ion etching system (Samco, model RIE-

PAGE 43

43 1C). The following etch conditions were used: plasma power = 25 W, O2 pressure = 15 Pa, O2 flow rate = 10 standard cubic cm per min (SCCM) etch time = 2 hr. Finally, the material was heated at 400 C for 10 hr in a flowing O2 gas to obtain crystalline V2O5. Preparation of the carbon nanotube membrane anode The CVD method described in chapter 1 wa s used to prepare nanostructured carbon anode.11 Prior to CVD, the alumina membrane was preheated in order to prevent its curling during CVD. This was accomplished by sandwiching an alumina membrane (a semicircle with a diameter of 3.5 cm) between two qua rtz slides and then placing the assembly inside a quartz tube (diameter = 4.5 cm, length = 48 cm). The tube wa s then inserted into a high-temperature tube furnace (Thermolyne 21100) and heated at 720 oC for 1 hr in air. After preheat treatment, CVD carbon was initiated using the similar set-up. Th e preheated membrane was placed vertically into the quartz tube such that the plane of the membrane was perpendicular to the length of the tube. The furnace was then heated to 670 oC under Ar flow. Once the temperature stabilized, the Ar gas was replaced with the ethylene ga s mixture flowing at 20 SCCM. The ethylene thermally decomposed on the pore walls to yield the CNTs within the pores. Both faces of the membrane were also coated with thin layers of carbon and these carbon surface films were too thin to block the CNT openings at the membrane su rfaces. It is noteworthy that it is these carbon surface layers that allow us to make an elec tronic contact between the CNTs array and the attached copper tape current-collector for th e electrochemical experiments (see the Prelithiation of CNTs membrane anode section below). After 5.5 hr deposition, the furnace was turn ed off, the ethylene gas was replaced by Ar, and the furnace was allowed to cool to room temperature. The yielded carbon thickness can be controlled by CVD duration. It needs to point out that the prepared CNT membrane saved the

PAGE 44

44 underlying alumina template for the mechanical support of this synthesized anode during all experiments, except in SEM imaging. Preparation of the polymer electrolyte A typical PEO-LiClO4 polymer complex was used as the electrolyte for assembling the solid-state battery. The prep aration of the elec trolyte involved firs tly dissolving LiClO4 salt (0.3 wt%) in 7 ml THF solution, followed by the addition of PEO into LiClO4-THF solution. The ratio of EO/Li (monomeric units CH2CH2O / the number of Li) was fixed at 9:1. The mixture was then under vigorous stirri ng until a homogeneous and transp arent solution was observed. This whole procedure was carried out in a Ar-filled glove box. Assembling of 3-D nanostruct ured solid-state batteries Nano-LiV2O5/PEO-LiClO4/Li solid-state battery Pre-lithiation of the nano-V2O5 cathode: Before battery as sembly, the prepared nano-V2O5 electrode was lithiated into LiV2O5. A typical three-electrode ce ll was used in which the nanoV2O5 electrode was the working elec trode and Li foils were the re ference and counter electrode. The Pt foil used for preparation of the nano-V2O5 electrode served as the current-collector for the working electrode. 1M LiClO4 in EC:DEC (3:7 v/v) was prepared as the liquid electrolyte. The lithiation was performed by galvanostatically charging the nano-V2O5 from the open circuit potential (OCP, ~3.5 V) to 2.8 V (ver sus Li+/Li) at th e current of 5 A. The lithiation was run on a Solartron 1287 Potentiostat driven by the CorrWare software package. Battery assembling: Battery assembling was co nducted in a glove box purged with Ar. As shown in the scheme of our battery design (Figure 2-2), the 3-D nanostructured solid-state battery was assembled by sandwiching polymer electr olyte-coated cathode and anode. Here, as the first concept-proof experiment, a Li metal foil anode, instead of na nostructured anode, was used to assemble a solid battery. The prepared PEO-LiClO4 polymer solution was dropped onto

PAGE 45

45 the surfaces of the prep ared lithiated nano-V2O5 cathode and Li foil anode. Before the polymer layers became completely dry, the coated cathod e and anode were sandwi ched together between two glass slides with a metal clamp. Then, the as sembled battery was vacuum-dried for ~4 hr to remove the THF solvent. The thickness of th e polymer electrolyte layer was measured ~20 m. Nano-V2O5/PEO-LiClO4/nano-LixC solid-state battery Pre-lithiation of the CNTs membrane anode : The CNTs membrane anode was lithiated electrochemically before the battery assembling. A three-electrode cell was used in which the prepared CNTs membrane was the working electrode and Li foils were th e reference and counter electrode. The current-collector on the work ing electrode was made by applying a copper conductive tape (single adhesive surface, Product No. 16072-1, Ted, Pella, Inc.) to one end of the CNT membrane. 1M LiClO4 in EC:DEC (3:7 v/v) was used as the liquid electrolyte. The lithiation was performed by galvanostatically lowering the potential from the open circuit potential (OCP, ~2.9 V) to 0 V (versus Li +/Li) at currents ranging from 10 to 100 A. This lithiation process was run on a Solartron 1287 Potentiostat driven by the CorrWare software package. Battery assembling: Battery assembling was co nducted in a glove box purged with Ar. Here, both electrodes prepared were nanostructured ones including nano-V2O5 cathode and lithiated nano-carbon anode. Exactly same as the scheme shown in Figure 2-2, the prepared PEO-LiClO4 polymer solution was dropped onto the surfaces of the prepared nano-electrodes. Before the polymer layers became completely dry, the coated cathode and anode were sandwiched together between two glass slides with a metal clamp to complete the battery. Then, the assembled battery was vacuum-dried for ~4 hr to remove the THF solvent. The thickness of the polymer electrolyte laye r was controlled ~7 40 m.

PAGE 46

46 Electrochemical characterization Cyclic voltammetric and galvanostatic charge/d ischarge experiments were performed using a Solartron 1287 Potentiostat, driven by the CorrWa re software package. For liquid half-cell experiments, a 3-electrode system was used in which the nanostructured electrode was the working electrode and Li foils we re the reference and the counter electrode. The electrolyte was 1M LiClO4 in EC:DEC (3:7 v/v). For battery tests, nano-V2O5 and Li (or nano-carbon) was the cathode and anode, respectively. All electroche mical experiments were conducted in a glove box filled with argon. For battery testing at 67 oC, the battery was placed on a stirrer/hot plate and the temperature of the plate was kept under 67 oC for 2 hr to reach the thermal equilibrium before the tests. Scanning electron microscopy JEOL 6335F field emission scanning electron microscopy (FESEM) instrument was used to obtain SEM images of the prepared nano-V2O5 and CNTs membrane before and after polymer coating. All samples were placed onto copper ta pes adhered to standard SEM stubs. Prior to SEM imaging, the samples were sputtered with Au/Pd using a Desk Cold Sputter instrument (Denton Vacuum, LLC). The sputtering current was 45 mA, the Ar pressure was 75 mTorr, and the sputtering time was 60 sec. This yiel ded a Au/Pd film that was ~16 nm thick. It needs to mention that the polymer-coated sa mples were vacuum-dried right after coating and before SEM imaging in order to remove th e organic THF solvent in the polymer solution. The drying was performed at r oom temperature for ~4 hr.

PAGE 47

47 Results and Discussion Nano-LiV2O5 / PEO-LiClO4 / Li solid-state battery Morphology of nanofibrous V2O5 cathode The commercial polycarbonate membrane with a nominal pore diameter of 50 nm was used to prepare the nanostructured V2O5, and the surface image of th e template is shown in Figure 2-3A. It can be seen that all the pores are randomly dispersed over the entire membrane surface and the average pore size was measured as 50 nm. Figure 2-3B shows the typical SEM image of the resulting nanostructured V2O5 after removal of the template membrane. These template-synthesized V2O5 nanofibers were protruding from th e Pt foil substrate. The fibers show highly dense and freestanding structures which replicate the structure of pore array in the template. The porous structure can be seen in each fiber. Same as our previous observation,15 the average diameter of the V2O5 fiber was measured as 100 nm, which is larger than nominal one. This can be explained due to two possibl e reasons: the pore expa nds during formation of the nanofibers and the pore diameter is smaller at the surface of the membrane than in the middle. It is noteworthy that the lower Pt foil used acts as the substrate and current-collector for the V2O5 cathode. The V2O5 prepared via this solgel based method yields cr ystalline orthorhombic structure. 15 Electrochemical characteri zation of nanofibrous V2O5 cathode Cyclic voltammetry (CV): Cyclic volta mmetry is one of the commonly used electrochemical techniques. It is used to c onfirm the activity of the electrode, identify the potential region of electroactivit y, and comment on the electron-ki netics and diffusion regimes of the process. In this study, we investigat ed the Li-ion intercalation properties of V2O5 with CV. The charge/discharge reactions for V2O5 can be written as follows:

PAGE 48

48 xLi+ + xe+ V2O5 LixV2O5 (0
PAGE 49

49 Electrochemical tests of nano-LiV2O5/PEO-LiClO4/Li solid-state battery The OCP of the assembled nano-LiV2O5/PEO-LiClO4/Li solid-state battery was measured as 2.7 V. The battery was then cycled in the potential ranged from 2.5 to 3.8 V at 25 C. Figure 2-6 shows the galvanosta tic charge/discharge curves of the nano-LiV2O5/PEO-LiClO4/Li solidstate battery. The similar charge/discharge profile was observed during each cycle. This indicates the good cycling perfor mance of the battery at 25 oC. The coulombic efficiency (ratio of a discharge capacity versus a charge capacity) in the first cycle was calculated as 49%, and this represents a large irreversible capacity. Ho wever, the irreversible capacities are greatly decreased in the subsequent charge/discharge cycles, which are corresponding to the increased coulombic efficiencies (~60%). The rate capability of this solid Li-ion batte ry was also evaluated by measuring discharge capacitiy as a function of discha rge current (Figure 2-7). In this study, the reported capacities were normalized by unit area (per cm2) to present electrochemical performance of the 3-D battery. In this battery, the capacity of lithiated V2O5 cathode is much less than that of the lithium anode. The battery capacity is then limite d by the capacity of the cathode. Therefore, we apply the theoretically maximum capacity per area (Cmax,c) of this nano-V2O5 cathode as the maximum capacity of our assembled battery. The maximum capacity of the cathode was calculated based on the V2O5 theoretically maximum capacity pe r gram and the theoretical mass per area of V2O5. As we mentioned before, the theore tical gravimetric-capacity at which V2O5 is lithiated into LiV2O5 is 147 mAhg-1. The mass per area (g/cm2) of V2O5 is the product of its density (g/cm3), the volume (cm3) of each pore and the pore density (cm-2) of the template. Based on above parameters, Cmax,c is calculated as 14 Ah/cm2 for the nano-V2O5 cathode. This theoretical capacity is relati vely low compared to typical thin -film batteries, and this is due to the low pore density of the polycarbona te template used for preparing nano-V2O5 electrodes.

PAGE 50

50 By increasing the pore density of the template membrane, the enhanced theoretical capacity can be expected. Turning now to Figure 2-7, as typi cally seen in any batter y, the highest discharge capacity was obtained at the lowest current. And the highest capacity we obtained is ~6 Ah/cm2 which represents 42% of the maximum cap acity. With increasing currents, the discharge capacities of the battery fell off fast. One main reason for this fast capacity decay is attributed to the high resistance of the solid polymer electrolyte. In the following text we will give more detailed discussion for this issue. Nano-V2O5 / PEO-LiClO4 / LixC solid-state battery Morphology of carbon nanotube membrane anode In this study, the commercial nanopore alum ina membrane was used to prepare CNTs membrane anode. Figure 2-8A shows the surface image of an alumina template membrane. In contrast to the randomly distri buted pores in polycarbonate s hown in Figure 2-3A, the highly ordered pores were observed in the alumina template. In addi tion, the pore density of the alumina is much higher than that of polycarbonat e (Figure 2-8A vs. Figure 2-3A). The average pore size was 200 nm. The well-aligned, highly compact and freesta nding carbon tubes were obtained after CVD and dissolution of the alumina template (Figure 28B). The outside diameter and the length of each tube are same as those of the pores in the alumina template. The inside diameter of CNT was measured ~120 nm. Thus, the wall thickness of deposited CNT was calculated as ~40 nm. In our previous paper,12 we demonstrated that the CNTs membrane prepared by alumina template showed the great promising as the anode material for Li-ion batteri es. In this study, this nanotubular structure of carbon ag ain provides several unique adva ntages for our 3-D battery design: first, it allows for the easy penetration of the polymer electrolyte into each pore; second, it ensures the minimized solid-state diffusion dist ance of the intercalating Li ions inside each

PAGE 51

51 nanopore; third, it yields the extremely high surface area. Th is large surface area can greatly decrease the effective current density, whic h lowers the interfacial kinetic overpotential.75 Electrochemical characterization of carbon nanotube membrane anode To characterize the electrochemical property of the synthesized CNTs membrane anode, we performed the galvanostatic charge/discharge tests and the results are shown in Figure 2-9. It can be seen that Li ions can reversibly (d e)intercalate into the synthesized carbon material. However, a very large irreversible capacity was ob served during the charge process. It is mainly attributed to the formation of solid electrolyte interface (SEI) and the redu ction of surface oxides (formed during CVD) at the large surface of CNTs.120, 121 The sloping profile of the charge curve is the characteristic of disordered carbon.122 Morphology of polymer electrolyte coated nanostructured electrodes Figure 2-10 shows SEM images of the crosssections of polymer-coated nanostructured V2O5 cathode. The polycarbonate template wa s removed prior to imaging. The low magnification image in Figure 2-10A shows that th e polymer electrolyte was mainly coated as a thin film on the surface of nano-V2O5. The thickness of the film is ~8 m. Taking a close view in Figure 2-10B, we can see that th e polymer did coat onto individual V2O5 nanofibers. This can be confirmed by two aspects: as-prepared porous V2O5 fibers (Figure 2-3B) became the nonporous one after polymer coating (Fi gure 2-10B); the diameter of the V2O5 fiber increased from ~100 nm before polymer coating (Figure 23B) to ~130 nm after pol ymer coating ( Figure 2-10B). Moreover, due to the large surface area of the synthesized nano-V2O5, this polymer coating greatly increases the effective contact area between the polymer electrolyte and the cathode material. The lithiated CNTs membrane was likewise co ated with the polymer electrolyte to assemble the battery. However, during the transf er of the sample into the SEM chamber, the

PAGE 52

52 sample will be inevitably exposed to air. The lithiated carbon is extremely reactive and tends to react with explosive results when exposed to air. Therefore, the unlithiated CNTs membrane was used for SEM imaging. Figure 2-11 shows the cross-sectional images of the coated CNT sample. The low magnification image (Figure 211A) shows the surface of the CNTs membrane was completely covered with a thin polymer la yer. The average thickness of the layer was measured ~5 m. From the high magnification imag e (Figure 2-11B), we observed the undiscriminable boundary between the polymer la yer and the ends of carbon nanotubes, which indicates the impregnation of the polymer electrolyte into each carbon tube. However, the penetration depth of the polymer electroly te is not clear at this point. Electrochemical tests of nano-V2O5/PEO-LiClO4/LixC solid-state battery Electrochemical lithiation of CNTs anode: The V2O5 cathode and CNTs anode used in our battery were prepared by poly carbonate and alumina template, respectively. Due to the much higher pore density of alumina (1011 cm-2) than that of polycarbonate (108 cm-2), the density of carbon tubes is much higher than that of V2O5 fibers. Thus, the capacity of the lithiated CNTs membrane anode is much larger than that of lithiated V2O5 cathode. In this study, to ensure the enough amounts of Li ions during the battery operation, the carbon anode was chosen to be lithiated. Lithiation of CNTs membrane anode wa s performed by decreasing the potential to 0 V (vs. Li+/Li) at currents ranging from 10 to 100 A. At the beginning of lithiation, a large cathodic current (100 A) was applied to reduce the residual oxygenated species on the surface of carbon nanotube and then allow the intercalation of Li ions.123 However, we observed that this large current caused large potential polariza tion which greatly decreased the capacity of the anode. In addition, we found that after first lithiation with 100 A, the OCP of the anode immediately jumped from 0 V to ~0.2 V and kept increasing in a rate of ~0.1 V/hr. A similar phenomenon was also observed by Ergang et. al..79 Therefore, after first lithiation, multiple

PAGE 53

53 lithiation cycles with reduced currents were applied. The post-lithiation OCP of the CNT membrane was measured as 0.09 V and its increase rate was ~0.01 V/hr. Electrochemical tests of nano-V2O5 / PEO-LiClO4 / nano-LixC6 solid-state Li-ion battery: Before assembling the battery, the OCPs of as-prepared nano-V2O5 cathode and lithiated CNTs membrane anode were measured as 3.3 V a nd 0.1 V, respectively. After assembling and vacuum-drying the battery, the OCP of the battery was measured ~2.9 V. The slight decrease in battery OCP is attributed to two reasons: the te mporary short circuit of the battery; the selfdischarge of lithiated CNTs membrane anode duri ng the battery assembling and drying processes. Beginning with a discharge process, the battery was cycled at 25 oC between 2.6 V and 3.8 V and then terminated after a charge process. Th e current densities applie d during each cycle range from 5.88 ~ 178.57 A/cm2. In this chapter the shown batte ry cycling data under each specific current always begins with the second charge cycle, rather than the first di scharge cycle. This is because the capacity in the fi rst discharge cycle was corresponding to the capacity obtained under a different current. As we mentioned earlier, the cap acity of our battery is lim ited by the capacity of the nanoV2O5 cathode, and the calculated maximum capacity of the battery is 14 Ah/cm2. Therefore, based on this limiting capacity, we determined the C-rate for our following battery tests. Figure 2-12 shows galvanostatic charge/d ischarge curves of the assemb led 3-D battery under the current of 14 A/cm2 (1C) and 32 A/cm2 (2.3C). The battery testing was performed at 25 oC. Under each current, our 3-D batteries were rechargeable and showed the relatively stable reversible capacity for each cycle. At 14 A/cm2 (1C), the discharge capacity decreased 3% and 12% after the second and ninth cycle, resp ectively. The discharge capacity maintained almost constant during the nine cycles under the current of 32 A/cm2 (2.3C). However, in the 3-D cell prepared

PAGE 54

54 by Ergang et.al. ,79 the discharge capacity decreased 76% after the second cycle and the cell behaved more like a pseudocapacitor with rapid polarization during later cycles. From Figure 212, the calculated coulombic effi ciencies were 87% and 84% under the discharge current of 14 A/cm2 and 32 A/cm2, respectively. Thus, it can be c oncluded that our synthesized 3-D nanostructured solid-state battery showed th e very stable cycling performance and good reversibility under applied rates. Comparing the cycling data at low and high cu rrents in Figure 2-12, we noticed that the reversible capacity of the battery in the first cycl e fell off very fast with the increased currents. The capacity of the first cycle dropped 95% (4.5 to 0.2 Ah/cm2) for the current increase factor of 2.3 (14 to 32 A/cm2). One main reason for this fast capacity drop is due to the high resistance resulting from the PEO-LiClO4 polymer electrolyte at 25 oC. It is studied that PEO is a highly crystalline polymer with a melting point of 65C. Above 65 oC, PEO is in a totally amorphous state and shows good flexibility and pol ymer segmental motion which contribute to the efficient ionic conductivity of the polymer complex.124-126 However, below 65oC, the ion transport greatly slow down due to the highly crys talline and rigid PEO ma trix. It is reported that the room-temperature ionic conductivity of PEO-LiClO4 (EO:Li = 8) complex is about 10-8 ~10-7 S cm-1 which is several orders of magnitude less than that exhibited by liquid electrolytes.127 In comparison with reported nanostructured batteries,34, 79, 80 our template-synthesis-based 3-D solid-state battery functioned well at room te mperature. Within the potential window used for testing commercial Li-ion batte ries, our battery showed relativ ely stable cycling performance under each current. Also, it needs to mention th at common micro-batteri es with the similar electrolyte system can not function under room te mperature due to the low ionic conductivity of

PAGE 55

55 the electrolyte and th eir micro-structured battery components.128, 129 This suggests that our nanostructured electrodes showed superior char acteristic in our 3-D battery configuration. It is studied that the efficien t ionic conductivities of PEO-LiClO4 can be achieved when the temperature is above the melting temperature (65 oC) of PEO. It is also reported that when the temperature increases from 25 oC to 67 oC, the ionic conductivity of the polymer electrolyte increased from ~10-8 S cm-1 to ~10-5 S cm-1, which is three orders of magnitude difference.124 Therefore, we can expect that the delayed capa city fade and improved rate capability should be observed if the assembled 3-D solid-s tate battery is operated above 65 oC. Figure 2-13 shows the galvanostatic charge/disch arge curves of the 3-D battery cycled at 67 oC. The current was 120 A/cm2 (8.6C). The similar charge/discharge profiles were obtained at 67 oC as those at 25 oC. The stable cycling and good coulom bic efficiency of the battery was also observed at 67 oC. The discharge capacity dropped 1% a nd 5% after second and third cycle. The average coulombic efficiency was 75%. However, the batteries showed dramatically improved rate capability at 67 oC comparing to the results obtained at 25 oC. For instance, the discharge capacity of the first cycle was 0.2Ah/cm2 at the current of 32 A/cm2 (2.3C) and 25 oC. When the temperature was increased to 67 oC and the current increased nearly 4 times to 120 A/cm2, instead of observing a capacity drop, we found the discharge cap acity increased 15 times to 3 Ah/cm2 (8.6C, Figure 2-13). These results st rongly support our hypothesis that the increased ionic conductivities of the polymer electrolyte can great ly improve the rate capability of our battery. Another possi ble reason for this improved ra te capability is the improved interface contact between the polymer and el ectrodes at the elevated temperature. Pesserini et. al. reported that at 90 oC, their Li/PEO-LiBETI/V2O5 solid battery delivered ~90% theoretical capacity at 0.05 C, but the capacity dropped to 20% when the rate increased to

PAGE 56

56 1.1 C. 130 Salomon et. al. reported that their solid-state Li/LiFePO4 polymer battery can not be cycled at 76 oC when charge/discharge rate was ~ 0.17 C, although more than 90% reversible capacity was delivered at 100 oC.128 In contrast, our battery stil l can deliver ~20% theoretical capacity at 8.6C under 67 oC and showed very good rate capability compared to 2-D solid-state battery. This suggests that our nanostructured electrodes and 3-D battery design have great advantage in rate capability over th e traditional micro-batteries. A more direct comparison of the temper ature effects on the rate capability for the assembled 3-D solid battery is shown in Figur e 2-14. For each low and high temperature test, two batteries were assembled and evaluated at the same testing conditions. The overlapping of two curves for the batteries tested at each temperature indicates good reproducibility of our assembled 3-D solid Li-ion battery. As typically observed in Li-ion batte ries, capacity fell off with increasing discharge currents for each test However, the extent of capacity loss was dramatically reduced in the high temperature (67 oC) tests. When the current increased from 11.90 A/cm2 (0.9C) to 60 A/cm2 (4.3 C), the discharge capacity dropped 33% at 67 oC vs.95% at 25 oC. Also, if noticing the C-rate labeled in this figure, our assembled ba ttery still delivered a certain amount of capacity up to ~13C, while co mmon 2-D thin-film Li-ion batteries were not operated under this large current density even at higher temperature (>>67 oC). This rate capability advantage and delayed capacity decay undoubtedly inherits our nanostructured electrodes (short lithium-ion solid-state diffusi on distance and large surface area of the active material) and our 3-D batte ry architecture design. Conclusion In our previous studies, we demonstrated that the template-synthesized nanostructured (nanofibrous or nanotubules) el ectrodes have improved rate ca pability compared with the conventional thin-film electrodes composed of the same material This is due to the larger

PAGE 57

57 surface area and shorter Li ion so lid state diffusion distance (with in nanometer scale) inside nanostructured electrodes. Based on these studies, herein, we developed a simple nanofabrication strategy to use such nanostructu red electrodes to assemble 3-D nanostrctured solid-state Li-ion batteries. Such batterie s were assembled by sandwiching nanostructured electrodes (cathode and anode) coated with thin layers of polymer electrolyte. The 3-D nano-batteries showed very stable cycling performance and high coulombic efficiency when operated at 25 oC. At 14 A/cm2 (1C), the discharge cap acity decreased 3% and 12% after the second and ninth cy cle, respectively. The discharg e capacity maintained almost constant during the nine cycl es under the current of 32 A/cm2 (2.3C). The coulombic efficiencies were calculated as 87% a nd 84% under the discharge current of 14 A/cm2 and 32 A/cm2, respectively. With the increasing currents, the fast capacity decay was observed at 25 oC. The capacity in the first cycle decreased 95% for the curre nt increase factor of 2.3. By increasing the battery operation temperature abov e the melting temperature of the polymer, the capacity loss was greatly delayed. When the current increased from 11.90 A/cm2 (0.9C) to 60 A/cm2 (4.3 C), the discharge cap acity decreased 33% at 67 oC vs .95% at 25 oC. Our assembled battery showed the dramatically improved rate capability at 67 oC. The battery can be operated up to ~13C at which the commercial 2-D Li-ion batte ries failed. This delayed capacity drop and improved rate capability are mainly attributed to the improved ionic conductivity of the polymer complex at the elevated temperature. In compar ison with currently reporte d 3-D Li-ion batteries, our prototype 3-D solid battery showed the uni que advantages including high cell voltage, superior rate capabilities and st able cycling performance and impr oved safety. These advantages benefit from our template-synthesized nanostruc tured electrodes as well as this 3-D battery design.

PAGE 58

58 Figure 2-1. Scheme for the template-synthesis of a nanofibrous V2O5 electrode.

PAGE 59

59 Figure 2-2. Scheme for preparation of a na nostructured solid-state Li-ion battery.

PAGE 60

60 Figure 2-3. FESEM images of (A ) the surface of a commercial polycarbonate membrane with nominal pore diameter of 50 nm and (B ) the sol-gel template-synthesized V2O5 cathode after removal of the polycarbonate template. A B

PAGE 61

61 Figure 2-4. Cyclic voltammogram of the template-synthesized nano-V2O5 electrode. Scan rate = 0.5 mV s-1.

PAGE 62

62 Figure 2-5. Galvanostatic char ge/discharge curves of the template-synthesized nano-V2O5 electrode. Current = 5 A.

PAGE 63

63 Figure 2-6. Galvanostatic charge/d ischarge curves for the nano-LiV2O5/PEO-LiClO4/Li solidstate battery. Current = 1 A. Pote ntial range is between 2.5 and 3.8 V. Temperature = 25 oC.

PAGE 64

64 Figure 2-7. Discharge capacity of the assembled nano-LiV2O5/PEO-LiClO4/Li solid-state battery at 25 oC as a function of galvanos tatic discharge currents.

PAGE 65

65 Figure 2-8. FESEM images of (A) the surface of a commercial alumina membrane with nominal pore diameter of 200 nm and (B) the C VD template-synthesized CNTs membrane anode after removal of the alumina template. B A

PAGE 66

66 Figure 2-9. Galvanostatic charge/discharge curves of the template-synthesized CNTs membrane electrode. Charge current = 100 A, discharge current = 50 A.

PAGE 67

67 Figure 2-10. FESEM images of the cross-section of the PEO-LiClO4 coated nanofibrous V2O5 after removal of polycarbonate templa te. (A) Low magnification view. (B) High magnification view. B A

PAGE 68

68 Figure 2-11. FESEM images of th e cross-section of PEO-LiClO4 coated CNTs membrane with the alumina template. (A) Low magnifi cation view. (B) High magnification view. A B

PAGE 69

69 Figure 2-12. Galvanostatic charge/d ischarge curves for the nano-V2O5/PEO-LiClO4/nano-LixC solid-state battery. (A) Current = 14 A/cm2. (B) Current = 32 A/cm2. Potential range is between 2.6 and 3.8 V. Temperature = 25 oC. B A

PAGE 70

70 Figure 2-13. Galvanostatic charge/d ischarge curves for the nano-V2O5/PEO-LiClO4/nano-LixC solid-state battery. Current = 119 A/cm2. Potential range is between 2.6 and 3.8 V. Temperature = 67 oC.

PAGE 71

71 Figure 2-14. Discharge capacity of the assembled nano-LiV2O5/PEO-LiClO4/nano-LixC solidstate battery at 25 oC and 67 oC as a function of galvanos tatic discharge currents. 67 oC 25 oC

PAGE 72

72 CHAPTER 3 TEMPLATE SYNTHESIS OF NANOSTRUC TURED CARBON/M ETAL COMPOSITE MATERIALS AND THEIR APPLICATION IN LI-ION BATTERIES Introduction Nanostructured carbon have drawn much a ttention from fundamental and applied perspectives.82 83 Their diverse applications have been investigated in th e fields of hydrogen storage84, catalyst suppport,82, 83, 86, 87 nanoelectronics,88-92 lithium-ion batteries,12, 14 chemical and bio-sensors.93-95 Martin group and others have explored the synthesis of nanostru ctured carbon based on template synthesis method.11, 12, 14, 39, 57 With the monodisperse pores and uniform pore-size in the porous template, well-aligned carbon nanotubes or nanofibers are obtained by using chemical vapor deposition (CVD) within the te mplate. As the characteristic of template synthesis, the outer diameters of prepared fibe rs or tubes are controlled by th e diameter of the pore in the template. And the inner diameters of the t ubes are controlled by C VD duration. Templatesynthesized nanofibrous or t ubular carbon has shown great potentials for Li-ion batteries,12 fuelcell technology,12 and fundamental study of electroosmotic flow.11 In chapter 2, we applied the templated carbon nanotubes as the anode material s for our 3-D nanostructu red solid-state Li-ion battery research. Despite the extensive applicati ons of the nanostructured carb on, there are limitations in the conventional CVD-based template synthesis met hod for nano-carbon synthesis. First, the CVD temperature is high (~700 oC), under which the commercial alum ina template occurs physical deformation such as curling. Second, by direc tly depositing carbon with in the pores in the template, the length of carbon nanotubes or nanofibe rs is only determined by the thickness of the template. Since it is hard to prepare and handl e thin alumina templates (less than 10 m), the nanostructured carbon with shorter le ngths is difficult to obtain.

PAGE 73

73 Hence, in this study we proposed a new me thod to overcome above limitations. This method incorporates electrochemi cal deposition into CVD-based te mplate synthesis. With this method, new fibrous carbon/metal composite struct ures were obtained within pores of the template. More important, this new method can allow us to effectively decrease CVD temperature and control the length of the synthesized nanostructured carbon. Experiment Materials The template membranes used in this study were commercially available alumina membranes (200 nm diameter-pores, 60 m thick) purchased from Whatman Anapore. Commercial Ag and Au plating solutions were obtained from Technic Inc. The Ni plating solution was made of 180g/L nick el sulfate, 8g/L ammonium chlo ride, 30g/L Boric Acid. Cu plating solution was made of 0.05 M CuSO4 and 0.5 M H2SO4. Ethylene (30% balanced with Helium, from Praxair) was used as CVD carbon pr ecursor gas. Ethylen e carbonate (Aldrich), diethyl carbonate (A ldrich) and LiClO4 (Aldrich) were used to prepare the liquid electrolyte. LiClO4 was vacuum-dried under 120 C for 48 hr before use. Electrodeposition of metal fibers inside the alumina template Figure 3-1 outlines the entire fabrication proce ss of the nanofibrous metal/carbon within an alumina template. In detail, a thin layer of Au /Pd was firstly sputtered onto one face of an alumina membrane (Figure 3-1a) using a Desk II Cold Sputter instrument (Denton Vacuum, LLC). The sputtering current was 45 mA, the Ar pressure was 75 m Torr, and the sputtering time was 180 sec. A ~48 nm thick Au/Pd film was yi elded. This layer was too thin to block the pores in the template membrane but converted this surface into a conductive electrode for the following electrochemical deposition. Next, a Ag (or Cu) fiber was bottom-up electrodeposited within each pore from this metal-sputtered surfac e. A thin film of Ag (or Cu) metal was also

PAGE 74

74 simultaneously formed on this surface of the me mbrane to totally cover the Au/Pd sputtered layer (Figure 3-1b). After that, Ni nanofibers were likewise elec trodeposited on top of Ag (or Cu) fibers to introduce the catalyst for the follo wing CVD carbon (Figure 3-1c). The applied electrodeposition current ranged from 1 to 5 mAcm-2. The thickness of the deposited metal was controlled by the deposition time. After e ach deposition experiment, the membrane was dissembled from the cell, rinsed with copious purifie d water, then left to dry before use. It is noteworthy that above electrodepos ition procedure can be modified in order to deposit different metal fibers for specific applications. Figure 3-2 shows the schematic set-up for electr odeposition of metals inside the template membrane. The electrodeposition cell consists of the metal-spu ttered alumina membrane as the cathode and the metal coiled wire which is to be deposited (Ag or Cu) as the anode. The nonmetal-sputtered side of the alumina membrane wa s exposed to the plating solution and the anode, while the opposite side of the membrane remained intact. Ag or Cu was deposited galvanostatically in a plating solution which contains a Ag or Cu salt. Preparation of the nanostructured carbon The CVD method described in detail previously 11 was used to deposit the carbon nanofibers within the pores of the alumina me mbrane. Briefly, the alumina membrane with electrodeposited metal nanofibers and thin film wa s fixed between two pieces of ceramic plates (diameter = 3.7 cm). A hole (diameter =1.4 cm) was made into one of the plates; this hole defined the area of the membrane that was ex posed to the CVD carbon precursor gas. The assembly was placed vertically into a quartz tube (diameter = 4.5 cm, length = 48 cm) such that the plane of the membrane was perpendicular to the length of the tube. The tube was then inserted into a high-temperature tube furnace (Thermolyne 21100) and heated to 545 oC under Ar

PAGE 75

75 flow. When the temperature stabilized, the Ar gas was replaced with the ethylene gas mixture flowing at 10 standard cubic cm per min (SCCM). Due to the presence of the fibrous Ni catalys t as well as the sealing of one surface of the membrane with the non-catalyst metal, ethylene preferentially decomposed on the Ni nanofiber within each pore (Figure 3-1d). After the desire d deposition time, the furnace was turned off, the ethylene gas was replaced by Ar, and the furnace wa s allowed to cool to room temperature. Scanning electron microscopy and EDS analysis Samples were imaged by a scanning electr on microscopy (JEOL JSM 6400). Also, this instrument was equipped with an energy dispersi ve x-ray spectrometer (EDS, Oxford Link ISIS), which was used to obtain elemental compositi on data of the samples. The prepared carbon/metal-containing alumina template was imme rsed in 50 wt% HF solution for 20 hr to dissolve the template (Figure 31e). To image the indivi dual carbon fibers, the prepared carbon/metal samples were further immersed into 69.2 wt% HNO3 solution to dissolve metal fibers. Prior to imaging, the samples were a dhered to SEM stubs usi ng conductive copper tapes, then sputtered with Au/Pd to improve the resolu tion of the images. The sputtering current was 45 mA, the Ar pressure was 75 m Torr, and the s puttering time was 60 sec. This yielded a Au/Pd film that was ~16 nm thick. Electrochemical characterization Cyclic voltammetry and galvanostatic charge/d ischarge experiments were performed using a Solartron 1287 Potentiostat, driven by the Co rrWare software package. For half-cell experiments, a 3-electrode cell was used in which the nanostructured carbon/metal composite was the working electrode and Li foils were the reference and the c ounter electrode. The electrolyte was 1M LiClO4 in EC:DEC (3:7 v/v). All electrochemical experiments were conducted at room temperature in a glove box filled with argon.

PAGE 76

76 Results and Discussion Structure characterization and elemental analysis Figure 3-3 shows the images of an alumina template used for preparing nanostructured carbon/metal samples. The alumina membrane is 200 nm in pore diameter and 60 m in thickness. The pore density of the membranes is approximately109 pores/cm2. Figure 3-4 shows the SEM images and EDS data of the electrodeposited Ag/Ni composite metal fibers. The Ag film deposited on the Au/P d-sputtered surface of the membrane is shown in Figure 3-4A. It can be seen that the original porous surface (see Figure 3-3) is totally covered with a dense film. EDS data (Fi gure 3-4B) confirms that this de nse layer consists of Ag. The deposited Ag film shows dual functions in this study: 1) it serves as the electronic conductive substrate for the following electrodeposition of metal segments; 2) it is used as the currentcollector for the later electrochemical tests of this composite carbon/metal material. As illustrated in the scheme in Figure 3-1, the Ag/Ni fibers grow bottom-up from an electrodeposited Ag film on one face of the template And these fibers can be clearly seen in Figure 3-4A. These fibers show the freestandin g, highly-dense structure after removal of the template. By using the electrodeposition me thod, good physical contact between the nanofibres and the thin film unde rneath can be obtained. The Ni fibers were electrodeposited on top of the Ag fibers, and the top view of the resulted Ni nanofibers was show n in Figure 3-4C. Due to the replicated morphology from the high-pore-density alumina template, the highly dense nanofibers were observed. The average diameter of these fibers was measured as 200 nm. The high-intensity Ni peaks shown in the EDS analysis (Figure 3-4D) confirm the deposited Ni element. Ag peaks are also observed in the EDS data and this is probably due to the d eep penetration depth (~1 m) of the X-ray beam compared to the length (~0.6 m) of Ni fibers.

PAGE 77

77 After characterizing the prepar ed Ag/Ni composite fibers, we performed the CVD carbon on these fibers. Traditionally, in Martin group, preparation of nanostructured (fibrous and tubular) carbon is realized by us ing CVD of carbon precursor gas w ithin the alumina template. Although this technique is quite useful and show s excellent reproducibilit y, the alumina template always becomes curling under the high CVD temperature (~700 oC). One way to solve this problem is the preheat treatment of the alum ina template as we described previously. 11 However, preheat treatment is time-consuming and complicates the experiments. Here, we introduce a catalyzed-CVD method in order to lowe r the CVD temperature and totally avoid the curling problem of the template. The catalyst we used here is electrodepos ited Ni nanofibers. It has been studied that Ni, as a catalyst, can efficiently catalyze the decomposition of hydrocarbons gas during CVD to form carbon or ev en highly ordered graphite nanofibers at rather low temperatures.39, 131-133 Moreover, with this catalyzed-CVD method, a new nanostructure of carbon-based materials can be obtained by incorporating template-synthesis method. The prepared carbon/metal composite nanofibers are shown in Figure 3-5, and these fibers maintain high density and freestanding stru cture after dissolution of the template. Based on the image contrast, the different materials can be observed along each fiber. The upper segment with the bright color represents the de posited metals, while the lower segment with the dark color represents the deposited carbon. The interface between carbon and metal segments can be more clearly seen in the high magnifica tion image (Figure 3-5B). The composition of different segments along each fiber was also c onfirmed by EDS data shown in Figure 3-5C and Figure 3-5D. It needs to mention that it is hard to observe the interface of two metal (Ni and Ag) segments in both images due to the short Ni fi bers. To further investigate the deposited nanofibrous carbon, the metal segments (Ni and Ag) were dissolved using concentrated (69.2

PAGE 78

78 wt%) HNO3. The SEM images of the resulted sample are shown in Figure 3-6. From both low(Figure 3-6A) and high-magnification images (Figure 3-6B), it can be seen that the straight and freestanding fibers remain intact and show highl y ordered structure whic h replicates the porous structure of the template. The average diameter of the carbon fiber is ~200 nm. The length of the nanofibers is measured ~ 5 m. EDS results shown in Fi gure 3-6C confirm the metal segments and the alumina template were to tally removed. Previously, our group also investigated the use of Ni catalyst for CVD carbon within alumina templates.39 However, due to the different catalyst-loading method used, the cat alyst loaded in the pr evious study existed as nanoparticles randomly sticking to the inside wall s of the pores. Therefore, the synthesized carbon tubes or fibers always mixed with these metal catalyst nanoparticle s. Furthermore, the length of the nanostructured carbon was hard to alter with the conventio nal CVD-based template synthesis. However, in this study, by appl ying the electrodeposited Ni fibers as the CVD catalysts, the carbon nanofibers with a uniform composition and controllable length were obtained. It is well-known that carbon can be used as the anode material in Li -ion batteries. In the composite metal/carbon structure, the deposited me tal fibers serve as the substrate and currentcollector for the fibrous carbon. In Li-ion battery industry, Cu is commonly used currentcollector for anode carbon due to its low-cost and good electrochemical stability in organic electrolytes.73 Thus, in this study we also invest igated the prepara tion of Cu/Ni/Carbon composite nanofibers for their application in Li -ion batteries. Figure 3-7 shows the images of synthesized composite fibers. Three different segments of ma terials along each fiber can be clearly visualized in Figure 37A. By increasing the electrode position time, longer Ni nanofibers were obtained. Also, we observe d a very interesting phenomenon that the length of the metal

PAGE 79

79 substrate affects the CVD rate in this experiment For example, the short metal fiber substrate results in the shorte r carbon fiber (1.5 m) as shown in Figure 3-7A. However, as shown in Figure 3-5, the longer metal fiber pr oduces the longer carbon fiber (5 m) even with a short CVD time. One possible reason is due to the mass transfer limitation of the carboneous gas molecular inside the nan opore of the template. However, fu rther investigati on will be required before a definitive conclusion can be reached. Figure 3-7B indicates that all the fibers have solid, freestanding and highly compact structur e after dissolution of the template. Electrochemical characterization of na nostructured carbon/metal composite Cyclic voltammetry As one of the applications of the synthesized carbon/metal co mposite nanofibers, we also characterized their electrochemical properties. Figure 3-8 shows the cyclic voltammetry curves obtained from Cu/Ni/Carbon composite nanof ibers. The scan rate is 10 mV s-1 and the scanned potential window ranges from 0 to 3 V. From the first two and half cycles shown in this figure, it can be seen that the synthesized nanofibrous carbon shows the electrochemical activity and can reversibly intercalate Li ions in the half-cel l test. No oxidation and reduction peaks of the deposited metals are observed. The well-overlap ped two cycles represen ts the good reversibility and stability of the synthesized na nomaterials. Also, the shape of CV curve is characteristic of Li ions intercalation into disordered carbon.134 Galvanostatic charge/discharge Figure 3-9 shows the galvanostatic charge/d ischarge curves of the nanostructured Cu/Ni/Carbon electrode. The curre nt applied is 60 A and potenti al ranges are between 0 and 1V. Two charge/discharge cycles are presented in this figure. The following features can be noted from these curves. First, in consistent with the cyclic voltammetry results, the charge/discharge curves indi cate our synthesized nanofibrous carbon are electrochemically

PAGE 80

80 active and shows the revers ibility in (de)inte rcalation of Li ions. Second, the overlapping of the first and second discharge curves suggests the good reversible capabilities of the nanomaterial. Third, the sloping discharge prof ile indicates the amorphous structure of the synthesized carbon,122 which is corresponding to the result in cy clic voltammetry. A broad peak starting from 0.6 V in the first charge process is corres ponding to the solid elect rolyte interface (SEI) formation as well as reduction reactio n of the carbon surface oxygenated groups.123 Following the SEI formation, the potential declines continuous ly and the majority of Li ion intercalation occurs below 0.25 V. We also found there is a large irreversible capac ity (difference between charge and discharge capacity) in the first cycles. This is possible due to the large quantity of SEI formation and the reduction of surface oxides due to the large surface area of the synthesized nanostructured carbon material. Conclusion In this study, a new method was develope d to create nanostructured carbon/metal composite materials. The method incorporated electrodeposition, CVD and template synthesis methods. By using electrodeposition method, the nanofibrous metals ar e firstly introduced within the pores of the template. These metal fibers serve as the cata lyst and substrate for subsequent carbon deposition inside the pores of the template. SEM and EDS analysis indicate the prepared metal/carbo n composite are freestanding, compact and highly dense fibers with the controlled diameters in nano-scale. Electroch emical tests show Li ions can reversibly intercalate/deintercalate in nanofibrous carbon materials with out adding any additives. Compared to the conventional non-catalyst CVD-based template synthesis, this new method provides the following advantages: CVD temp erature is decreased with the presence of catalyst metal fibers so that the physical deformation of the alumina template can be totally avoid; the length of carbon fibers can be controlled by altering the length of electrodeposited metal

PAGE 81

81 fibers substrate. Moreover, these new nanostr uctured carbon/metal composite materials can provide the improved contact betw een carbon and metal for the possi ble applications in battery electrodes, nanoscale elec trical interconnects.

PAGE 82

82 Cross-section of alumina template Sputter Au/Pd Film Electroplate Cu or Ag Film and Fibers Electroplate Ni Catalyst CVD Carbon Anode Fibers Dissolve TemplateAg or Cu Ni Carbon(b) (a) (d) (c) (e) Cross-section of alumina template Sputter Au/Pd Film Electroplate Cu or Ag Film and Fibers Electroplate Ni Catalyst CVD Carbon Anode Fibers Dissolve TemplateAg or Cu Ni Carbon(b) (a) (d) (c) (e) Cross-section of alumina template Cross-section of alumina template Sputter Au/Pd Film Sputter Au/Pd Film Electroplate Cu or Ag Film and Fibers Electroplate Cu or Ag Film and Fibers Electroplate Ni Catalyst Electroplate Ni Catalyst CVD Carbon Anode Fibers CVD Carbon Anode Fibers Dissolve TemplateAg or Cu Ni Carbon Dissolve Template Dissolve TemplateAg or Cu Ni Carbon(b) (a) (d) (c) (e) Figure 3-1. Schematic of the catalyst-CVD-ba sed template synthesis of nanostructured carbon/metal composite materials within an alumina template.

PAGE 83

83 Figure 3-2. Schematic of the electrochemical de position cell used for the deposition of metal fibers inside the pores of an alumina template.

PAGE 84

84 Figure 3-3. Scanning electron micrograph of th e surface of a commercial alumina template membrane with nominal pore size of 200 nm.

PAGE 85

85 Figure 3-4. SEM images and EDS data of the co mposite Ag/Ni nanofibers after removal of the alumina template. (A) top view of the deposited Ag thin film. (B) EDS data taken from the sample circled in (A). (C) top vi ew of the deposited Ni nanofibers. (D) EDS data taken from the sample circled in (C). A B

PAGE 86

86 Figure 3-4. Continued D C

PAGE 87

87 Figure 3-5. SEM images and EDS data of the pr epared Ag/Ni/Carbon com posite nanofibers after removal of the alumina template. (A) low-magnification image. (B) highmagnification image. (C) EDS data taken from the sample circled in (B). (D) EDS data taken from the sample squared in (B). B A

PAGE 88

88 Figure 3-5. Continued C D

PAGE 89

89 Figure 3-6. SEM images and EDS data of the samp le shown in Figure 3-5 after removal of the Ag and Ni nanofibers and alumina templa te. (A) low-magnification image. (B) highmagnification image. (C) EDS data taken from the sample in (B). A B

PAGE 90

90 Figure 3-6. Continued C

PAGE 91

91 Figure 3-7. SEM images of prepared Cu/Ni/Car bon composite nanofibers. (A) cross-section image of the sample before removal of the template. (B) top-view image after removal of the alumina template. Cu Ni C A B

PAGE 92

92 Figure 3-8. Cyclic voltammetry of the prepar ed Cu/Ni/Carbon composite nanofibers after removal of the alumina template. Scan rate = 10 mV s-1. Scan potential ranges between 0 and 3 V (vs. Li+/Li).

PAGE 93

93 0200040006000800010000 0.0 0.4 0.8 1.2 Potential (V vs.Li/Li+)Time(sec) 1st charge 1st discharge Figure 3-9. Galavanostatic charge/d ischarge of the prepared Cu/Ni/Carbon composite nanofibers after removal of the alumina template. Curre nt = 60 A. The potenti al ranges between 0 and 1 V (vs. Li+/Li). 2nd discharge 2nd charge

PAGE 94

94 CHAPTER 4 TEMPLATE SYNTHESIS OF CARBON NA NOTUBES WITH DIAMOND-SHAPED CROSS-SECTIONS Introduction Carbon nanotubes are of great current research interest 12, 39 11, 35, 57, 58, 135, 136 and have been proposed for applications in technologies as diverse as lithium-ion batteries,12, 39, 137 fuel cells,12 actuators,57 and membrane-based separations.11 There are two primary types of carbon nanotubes fullerene tubes prepared by arc discharge35, 36 and laser vaporization37, 38 methods, and template-synthesized carbon nanotubes prepar ed by chemical-vapor deposition of carbon within the pores of a nanopore template.11, 12, 39, 57, 58 Fullerene nanotubes always have a circular cross-section and cylindrical geometry, and with one recent exception,41 so do templatesynthesized nanotubes. As disc ussed in that chapter, it might be useful to prepare carbon nanotubes with non-circular crosssections and non-cylindrical geometries, for example, to modify their field-emission properties.41 This realization led these authors to develop a template-synthesis technology for preparing car bon nanotubes with a tria ngular cross-section. This was made possible by the use of a novel template membrane that had pores with a correspondingly triangu lar cross-section.41 While template membranes that have pores w ith non-circular geometries are unusual, there is one technology by which such pores can be routinely prepared. Th is is the track-etch method,19, 138 provided the material being tracked and etched is crystalline and not amorphous. The primary example is track-etched mica, which fo r reasons that will be di scussed below, yields membranes that have pores with diamond-sh aped cross-sections (Figure 1-2). Sun et al. electrodeposited Ni into the diamond-shaped pores in track-etched mica to obtain Ni nanowires with a corresponding diamond-shaped cross-section.24, 25 However, there have been no reports of using this approach to prepare carbon nanotubes with diamond-shaped cross-sections. We report

PAGE 95

95 the first examples of such carbon nanotubes here A simple chemical vapor deposition method was used,11, 12, 39, 57, 58 and we have shown that with th is method diamond-cross-section carbon nanotubes (DCNTs) with various outside and in side diameters and wa ll thicknesses can be prepared. Experiment Materials Muscovite mica, 10 m thick, was obtained from Spruce Pine Co (Spruce Pine, NC). Damage tracks were prepared in the mica by irra diation with U25+ ions (kinetic en ergy = 2.2 GeV, fluence = 106 or 108 cm-2) using the cyclotron at GSI (Darmsta dt, Germany). An ethylene/helium mixture (30% ethylene, Praxair) was used as the CVD-carbon precurs or gas. Hydrofluoric acid (~50 wt%, ACROS) was used to dissolve the mi ca membrane. Purified water was prepared by passing house-distilled water th rough a Millipore Milli-Q water purification system. Preparation of track-etched mica membranes Mica membranes with damage-track densities of 106 cm-2 and 108 cm-2 were used for these studies. Aqueous solutions of HF were used to etch these damage tracks to create the pores in the mica membranes. Membrane samples that were ~1.5 cm ~1.5 cm were etched. The membranes with 106 tracks cm-2 were etched using 10 wt.% HF for 190 min, and the 108 cm-2 membranes were etched using 20 wt.% HF for 10 min. Etching was terminated by immersing the membrane in purified water for 10 min. This process was repeated twice, and then the membrane was immersed for one hour in purified water. The membranes were then rinsed and left to dry in air overnight before use. Preparation of diamond shaped carbon nanotubes (DCNTs) The CVD method described in detail previously 11 was used to prepare the DCNTs within the pores of the mica membranes. Briefly, the et ched mica membrane was placed vertically into

PAGE 96

96 a quartz tube (diameter = 4.5 cm, length = 48 cm) such that the plane of the membrane was perpendicular to the length of the tube. The tube was then inserted into a high-temperature tube furnace (Thermolyne 21100) and heated to 670 oC under Ar flow. When the temperature stabilized, the Ar gas was replaced with the et hylene gas mixture flowing at 20 standard cubic cm per min (SCCM). During CVD, the ethylene thermally decompos ed on the pore walls to yield the DCNTs within the pores. Both faces of the membrane were also coated with carbon, but these carbon surface films were too thin to block the DCNT openings at the membrane surfaces. After the desired deposition time, the furnace was turned o ff, the ethylene gas was replaced by Ar, and the furnace was allowed to cool to room temperature. It is worth to mention that previously when using alumina templates in CVD, we always had to preheat the alumina to prevent its unwanted curling during CVD.11 However, here, we found that mica has much better thermal resistance than alumina since mica can withst and temperatures of at least 900 oC without any physical deformation. So the heat pretreatment of mica is unnecessa ry in this study. Structure characterization To characterize the structure and morphology of prepared samples, electron microscopy was employed including field-emission sca nning electron microscopy (FESEM, JEOL 6335F) and transmission electron micros copy (TEM, JEOL 2010F). For imaging DCNTs, two methods were used to prepare nanotube samples. The first entailed immersion of the nanotube-containing mica template in 50 wt% HF solution for 16 hr in order to dissolve the template. The HF was then pipetted away leaving bundles of nanotubes c onnected by the carbon films that had covered the mica surface. The nanotube bundles were rinsed with, and then suspended in methanol. The suspension was ultrasonicated for 20 min to br eak up the bundles and liberate the individual nanotubes. For FESEM imaging, a drop of the ul trasonicated suspension was deposited onto a

PAGE 97

97 piece of copper tape adhered to a standard SEM stub. For TEM imaging, a drop of the ultrasonicated suspension was deposited on a car bon TEM grid. Prior to FESEM imaging, the sample was sputtered with Au/Pd using a Desk Cold Sputter instrume nt (Denton Vacuum, LLC). The sputtering current was 45 mA, the Ar pressure was 75 mTorr, and the sputtering time was 60 sec. This yielded a Au/Pd film that was ~16 nm thick. In the second sample-preparation method, an oxygen plasma was used to remove the carbon surface film from one face of the mi ca membrane. This was accomplished by sandwiching the membrane sample between two 3 cm 3 cm pieces of aluminum foil. Prior to assembly, a 1 cm 1 cm hole was cut into one of the foils; this hole defined the area of the membrane that was exposed to the oxygen plasma. The assembly was placed with the holecontaining Al foil facing up in the center of the vacuum chamber of a plasma reactive-ion etching system (Samco, model RIE-1C). The following etch conditions were used: plasma power = 100 W, O2 pressure = 300 Pa, O2 flow rate = 30 SCCM, etch time = 30 sec. After etching away the carbon surface film, the membrane was immersed into 50 wt% HF to dissolve the mica template. This yielded a collection of carbon nanotubes protruding from the still-intact lower carbon surface film. The sample was then rinsed with purified wate r, air-dried, sputtered with Au/Pd and imaged with the FESEM. Results and Discussion Scanning electron microscopy of the mica su rface before and after deposition of carbon As mentioned in Chapter 1, the track-etch pro cess entails bombarding a thin membrane of a dielectric material with high energy particles fr om a cyclotron or nuclear reactor, to create damage tracks in the material, followed by a chem ical etch to convert the damage tracks into pores.19, 138 When amorphous or partially crystalline materials are track-etc hed, cylindrical pores

PAGE 98

98 are typically obtained. This can be seen in our polycarbonate template introduced in chapter 1 and 2 (Figure 1-1 and Figure 2-3A). However, Mi ca is a crystalline materi al, and the etch rates are different for different directi ons relative to the crystal axes. The slowest etching planes are the ones terminated by oxygen.24, 25 As a result of this etch-rat e anisotropy, pores etched in mica have a diamond-shaped cross-section (Figure 12). We obtained our tracked Muscovite mica membranes from GSI (Darmstadt, Germany) and et ched the pores using aqueous HF solution. As shown in Figure 4-1, diamond-shaped pores ca n be characterized by the lengths of their long and short axes, which we call al, and as, respectively. In this study, all these dimensions were obtained from FESEM images of the memb rane surface before car bon deposition (Figure 4-2). The cross-sectional area of the pore, Ap, can be calculated from al, as and measured included angle as shown in the following Equation 4-1. ) sin ( 2 1 s l pa a A (4-1) Ap can then be used to calcul ate an equivalent diameter (dequiv.) for a cylindrical pore having the same cross-sectional area as the diamond-shaped pore, as shown in Equation 4-2 p equivA d 4. (4-2) The advantage of using dequiv is that it allows for the size of the diamond-shaped pore to be described by a single number. The pores can also be characterized by the values of their major and minor angles (Table 4-1). We used two different etchant (HF) concentrat ions, and two different etch times, for these studies. The membranes that were etched with 10% HF for 190 min (Figure 4-2A) yielded pores with an equivalent pore diameter of dequiv = 451 nm. The other relevant parameters for these pores are shown in Table 4-1. The membranes th at were etched with 20% HF for 10 min (Figure

PAGE 99

99 4-2B) yielded pores with dequiv = 92 nm. The other relevant parameters for these pores are shown in Table 4-1. The angles for both sets of pores (Table 4-1) are in good agreement with values reported in the literature for track-etched mica.24, 25 A CVD method described in detail previously11 was used to deposit the DCNTs within the pores of the mica membranes. We and others have used this CVD method to deposit carbon nanotubes within the cylindrical pores of na nopore alumina templates. When the alumina template is used, carbon nanotubes with the conv entional cylindrical cros s-section are obtained.11 In analogy to the alumina-membrane case, th e CVD method yields the DCNTs lining the pore walls as well as thin carbon surface layers c overing both faces of the mica membrane. Figure 43 shows FESEM images of mica membranes afte r CVD synthesis of the DCNTs. The carbon surface layers can be seen as the enhanced surfac e roughness; this is particularly evident in the higher magnification images, Figure 4-3B vs. Fi gure 4-2B. The same parameters used to characterize the pores before carbon deposition can be measured from the images after deposition (Table 4-2). The DCNTs deposited within th e pores can be characterized by two key dimensions the equivalent nanotube inside diameter, dequiv,id, and the equivalent na notube outside diameter, dequiv,od, which is simply the equivalent diameter of the pore into which the nanotube was deposited; i.e., dequiv,od = dequiv (Table 4-2). A longer carbon deposition time (5.5 hr) was used for the membranes having the larger equivalent-d iameter pores. As shown in Table 4-2, these nanotubes had dequiv,od = 451 nm and dequiv,id = 316 nm. The difference between these numbers provides the effective nanotube wa ll thickness (Table 4-2). A s horter carbon depos ition time (53 min) was used for the membranes having the sma ller equivalent-diameter pores. As shown in Table 4-2, these nanotubes had dequiv,od = 92 nm and dequiv,id = 58 nm.

PAGE 100

100 Dividing the wall thickness by the CVD deposition time provides the average carbon deposition rate (Table 4-2). We found that the average rate of deposition is higher for the nanotubes that had thinner walls. This suggest s that the rate of de position decreases with increasing deposition time. One explanation for this observation is that the mica pore wall is acting as a catalyst for carbon deposition. Howeve r, further research w ill be required before a definitive conclusion can be reached. Electron microscopy of the liberated DCNTs. As discussed in the Experimental section, after CVD synthesis of the DCNTs, the mica membranes were dissolved in HF to yield bund les of liberated tubes connected by their carbon surface films. A suspension of the liberated DCNTs was then ultrasonicated to break up the bundles. Figure 4-4 shows the FESEM images of th e dried suspension. From the images, it can be seen that some of the DCNTs remained c onnected by the carbon surf ace films (Figure 4-4A and Figure 4-4C), whereas some were completely separated from the surface films (Figure 4-4B and Figure 4-4D). In both case, however, the we ll-defined diamond-shaped cross-section of the nanotubes is clearly seen Furthermore, these images s how that the DCNTs are open on both ends. An alternative approach for obtaining FESEM images of the DCNTs entailed using an oxygen plasma to remove one of the carbon surfa ce films and then dissolv ing away the mica. This yielded samples in which the DCNTs were protruding from the still-intact lower carbon surface film (Figure 4-5). These images show that even with the carbon surface film removed the DCNTs bundle. We have observed this phenomenon for nanotube s composed of gold47 and conductive polymers.10 Bundling is driven by Van der Waal interactions between the nanotubes. Lower magnification images (Figure 4-5A) show that the DCNTs extende d through the entire 10 m thickness of the mica template. Higher magn ification images (Figure 4-5B) again clearly

PAGE 101

101 show the diamond-shaped cross-section of the tubes. We also confirmed the tubular structure of synthesis nanostructured carbon by TEM. Low (F igure 4-6A) and high ma gnification (Figure 46B) images in Figure 4-6 indicate that the prep ared DCNTs are hollow, and that they have a uniform outside diameter and wall thickness dow n their entire lengths. Finally, electron diffraction data (Figure 4-6C) show that, like the cylindrical carbon nanotubes prepared by template synthesis in the nanopore alumina membranes,39 the DCNTs prepared here are composed of disordered, graphitic carbon. Conclusion We have shown that like nanopore alumina, tr ack-etched mica can be used as a template material for chemical vapor deposition synthe sis of carbon nanotubes. However, unlike the cylindrical pores found in nanopore alumina, the pores in the track-etched mica have a diamondshaped cross-section. As a result, unusua l carbon nanotubes having a correspondingly diamondshaped cross-section are obtained. These nanot ubes can be characterized by an equivalent outside diameter and an equivalent inside diameter, which were m easured using electron microscopy. As is characteristic of template synthesis, th e equivalent outside diameter can be varied by varying the equivalent pore diameter in the mica template. The equivalent inside diameter can be varied by varying the carbon deposition time. The smallest nanotubes prepared in these studies had an equivalent outside diameter of 92 nm, an equiva lent inside diameter of 58 nm and an equivalent wall thickness of 17 nm. Even smaller nanotubes should be possible with smaller pore-diameter mica templates. One possible advant age of using mica as a template membrane is that the thickness of the template can be controlle d by cleaving the mica. This should, in turn, allow for control of the lengths of the DCNTs. This could prove important in applications of these nanotubes to re sistive-pulse sensing.139, 140 We are currently expl oring this application.

PAGE 102

102 Figure 4-1. Diagram of a diamond-shaped pore in track-etched muscovite mica membrane. al and as represent the long and short axes, respectively. represents the included angle made by al and as.

PAGE 103

103 Figure 4-2. FESEM images of the surfaces of tr ack-etched mica membranes. (A) Track density was 106 cm-2. Etched in 10 wt.% HF for 190 min to yield large equivalent-diameter pores (see text). (B) Track density was 108 cm-2. Etched in 20 wt.% HF for 10 min to yield small equivalent-diameter pores. B A

PAGE 104

104 Table 4-1. Characteristics of the two mica template membranes used for these studies. Designation al (nm)[a] as (nm)[a] Major Angle[a] Minor Angle[a] Included Angle ( )[a] Ap (nm 2 )[b] d equiv (nm)[c] Large Pore 749 427 124 56 87 1.65 451 Small Pore 151 88 122 59 84 6.63 92 [a] Measured from FESEM image of mica me mbrane surface. The included angle ( ) was made by al and as. [b] Calculated via Ap= 1/2 (al as sin). [c] Calculated via dequiv= (4Ap/ )1/2. dequiv represents the equivalent diameter of a circular pore with the same cross-sectiona l area as a diamond-shaped pore.

PAGE 105

105 Figure 4-3. FESEM images of the surfaces of track-etched mica membranes after carbon deposition within the pores and on the memb rane faces. (A) A mica membrane with large equivalent-diameter pores (Figure 4-2A) was used. The carbon deposition time was 5.5 hr. (B) A mica membrane with sma ll equivalent-diameter pores (Figure 4-2B) was used. The carbon deposition time was 53 min. B A

PAGE 106

106 Table 4-2. Characteristics of the DCNTs. dequiv,od (nm)[a] al (nm)[b] as (nm)[b] Major Angle[b ] Minor Angle[b ] Included Angle( )[b] Ap (nm2)[c] dequiv,id (nm)[d] Effective Wall Thicknes s (nm)[e] Deposition Rate (nm min-1)[f] 451 533 294 125 55 91 7.84 316 68 0.20 92 96 56 123 57 93 2.73 58 17 0.32 [a] From Table 4-1. [b] Measured from FESEM image of mica memb rane surface after carbon deposition. The included angle ( ) was made by al and as. [c] Calculated via Ap= 1/2 (al as sin). [d] Calculated via dequiv= (4Ap/ )1/2. dequiv represents the equivalent diameter of a circular pore with the same cross-sectional area as a diamond-shaped pore. [e] Calculated from dequiv,od and dequiv,id. [f] Calculated from the effective wall thickness and the carbon deposition time.

PAGE 107

107 Figure 4-4. FESEM images of the liberated DCNT s. (A) and (B) DCNTs prepared in the large equivalent-diameter pores (451 nm). (C ) and (D) DCNTs prepared in the small equivalent-diameter pores (92 nm). A B C D

PAGE 108

108 Figure 4-5. FESEM images of the small equivalent-diameter DCNTs (dequiv,od = 92 nm) protruding from the still-intact lower carbon surface film. (A) low-magnification view. (B) high-magnification view. B A

PAGE 109

109 Figure 4-6. TEM images of the small equivale nt-diameter DCNTs (dequiv,od = 92 nm). (A) low-magnification image. (B) high-magnifica tion image. (C) Electron diffraction data for an individual DCNT. A C B

PAGE 110

110 CHAPTER 5 CONCLUSIONS Template synthesis method is a general and powerful tool for preparing nanostructured materials. The objective of th is dissertation is to demonstrate the capability of templatesynthesized nanostructured materials in develo ping advanced miniature solid-state Li-ion batteries and to investigate new template-synt hesized carbon nanostructures which have great potential applications. In chapter 2, we discussed the application of template-synthesized nano-electrode materials in assembling 3-D nanostructured solid-stat e Li-ion batteries. Both nanofibrous V2O5 cathode and carbon nanotube anode were prep ared and coated with thin la yers of polymer electrolyte. Two electrodes were then sandwiched to assemb le a complete nanostructured Li-ion battery. This 3-D nano-battery showed very stable cycl ing performance and high coulombic efficiency when operated at 25 oC. At 14 A/cm2 (1C), the discharge capacity decreased 3% and 12% after the second and ninth cycle, respect ively. The discharge capacity almost maintained constant during the nine cycles under the current of 32 A/cm2 (2.3C). The coulombic efficiency was calculated as 87% and 84 % under the current of 14 A/cm2 and 32 A/cm2, respectively. With the increasing currents, the fast capacity decay was observed at 25 oC. The capacity in the first cycle decreased 95% for the curren t increase factor of 2.3. By increasing the battery operation temperature above the melting temperature of th e polymer, the capacity loss was greatly delayed. When the current increased from 11.90 A/cm2 (0.9C) to 60 A/cm2 (4.3 C), the discharge capacity decreased 33% at 67 oC vs .95% at 25 oC. Our assembled 3-D battery showed the dramatically improved rate capability at 67 oC. The battery can be operated up to ~13C at which the commercial 2-D Li-ion batteries failed. This delayed capacity drop and improved rate capability are mainly attributed to the improved ionic conductivity of the polymer complex at the

PAGE 111

111 elevated temperature. In general, in comp arison with the traditional 2-D thin-film Li-ion batteries and other prototype 3-D Li-ion batteri es, our assembled battery showed the unique advantages including high cell voltage, superi or rate capability and cycling performance especially at high temperature. These adva ntages benefit form our template-synthesized nanostructured electrodes as well as the 3-D battery design. Two new carbon-based nanostructures were intr oduced in this dissertation. They are carbon/metal composite nanofibers and carbon nanot ubes with diamond-shaped cross-sections. Both nanostructured materials were achieved ba se on the variation of CVD-template synthesis methods. Chapter 3 described a catalyzed-CVD-bas ed template synthesis for preparing nanostructured carbon/metal composite materials. The metal nanosegments were introduced by electrodeposition within pores of the alumina templa te. These metal fibers serve as the catalyst and substrate for subsequent ca rbon deposition inside the pores of the template. SEM and EDS analysis indicated that after removal of the template, prepared materials were freestanding, compact and highly dense fibers with the contro lled diameters in nano-scale. Electrochemical tests showed Li ions reversib ly intercalated/deintercalated in nanofibrous carbon materials without adding any additives. Compared to th e conventional non-catalys t CVD-based template synthesis, this new method provides the following advantages: relatively low CVD temperature can be obtained so that the physical deformation of the alumina template is totally avoid; the length of carbon fibers can be cont rolled by altering the le ngth of electrodeposited metal fibers. Moreover, by incorporating the electrodepositio n method, the varied metal components can be prepared for the specific applications. This composite carbon/metal na nostructure can provide

PAGE 112

112 the improved electronic contact be tween carbon and metal for the possible applications in battery electrodes and nanoscale elec trical interconnects. A new carbon nanostructure with a non-circ ular cross-section and non-cylindrical geometry was synthesized and char acterized in chapter 4. This structure was realized by using a new type of template membranes which is track-etched muscovite mica. Due to the high crystallinity of muscovite mica, the track-etched pores in mica always showed diamond-shaped. As a result, unusual carbon nanotubes having a correspondingly diamond-shaped cross-section were obtained when CVD was performed on these mica templates. The dimensions of nanotubes were characterized by an equivale nt outside diameter and an equi valent inside diameter, which were measured using electron microscopy. As is characteristic of te mplate synthesis, the equivalent outside diameter can be varied by va rying the equivalent pore diameter in the mica template. The equivalent inside diameter can be varied by varying the carbon deposition time. The smallest nanotubes prepared in these studies ha d an equivalent outside diameter of 92 nm, an equivalent inside diameter of 58 nm and an e quivalent wall thickness of 17 nm. Even smaller nanotubes should be possible with smaller por e-diameter mica templates. One possible advantage of using mica as a template membrane is that the thickness of the template can be controlled by cleaving the mica. This should, in turn, allow for control of the lengths of the DCNTs. This could prove important in applications of these nanotubes to re sistive-pulse sensing. We are currently exploring this application.

PAGE 113

113 LIST OF REFERENCES (1) Science 1991, 254 1300-1301. (2) Ozin, G. A. Advanced Materials (Weinheim, Germany) 1992, 4 612-649. (3) Possin, G. E. Review of Scientific Instruments 1970, 41 772-774. (4) Martin, C. R. Science (Washington, D. C.) 1994, 266 1961-1966. (5) Penner, R. M.; Martin, C. R. Analytical Chemistry 1987, 59 2625-2630. (6) Penner, R. M.; Martin, C. R. Journal of the Electrochemical Society 1986, 133 310-315. (7) Brumlik, C. J.; Martin, C. R. Journal of the American Chemical Society 1991, 113 31743175. (8) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Science (Washington, D. C.) 1997, 278 655-658. (9) Lakshmi, B. B.; Patrissi, C. J.; Martin, C. R. Chemistry of Materials 1997, 9 2544-2550. (10) Martin, C. R. Accounts of Chemical Research 1995, 28 61-68. (11) Miller, S. A.; Young, V. Y.; Martin, C. R. Journal of the American Chemical Society 2001, 123 12335-12342. (12) Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature (London) 1998, 393 346-349. (13) Sides, C. R.; Martin, C. R. Advanced Materials (Weinheim, Germany) 2005, 17 125-128. (14) Li, N.; Mitchell, D. T.; Lee, K.-P.; Martin, C. R. Journal of the Electrochemical Society 2003, 150 A979-A984. (15) Patrissi, C. J.; Martin, C. R. Journal of the Electrochemical Society 1999, 146 3176-3180. (16) Li, N.; Martin, C. R. Journal of the Electrochemical Society 2001, 148 A164-A170. (17) Sides, C. R.; Croce, F.; Young, V. Y.; Martin, C. R.; Scrosati, B. Electrochemical and Solid-State Letters 2005, 8 A484-A487. (18) Patrissi, C. J.; Martin, C. R. Journal of the Electrochemical Society 2001, 148 A1247A1253. (19) Fleisher, R. L. P., P.B.; Walker, R.M. Nuclear Tracks in Solids: Principles and Applications. ; University of Californi a Press: Berkeley, 1975. (20) Apel, P. Radiation Measurements 2001, 34 559-566.

PAGE 114

114 (21) Poretics, 1995. (22) Sasou, M.; Sugiyama, S.; Yoshino, T.; Ohtani, T. Langmuir 2003, 19 9845-9849. (23) Heim, M.; Cevc, G.; Guckenberger, R.; Knapp, H. F.; Wiegraebe, W. Biophysical Journal 1995, 69 489-497. (24) Chien, C. L.; Sun, L.; Tanase, M.; Bauer, L. A.; Hultgren, A.; Silevitch, D. M.; Meyer, G. J.; Searson, P. C.; Reich, D. H. Journal of Magnetism and Magnetic Materials 2002, 249 146-155. (25) Sun, L.; Searson, P. C.; Chien, C. L. Applied Physics Letters 1999, 74 2803-2805. (26) Furneaux, R. C.; Rigby, W. R.; Davidson, A. P. Nature (London, United Kingdom) 1989, 337 147-149. (27) Jessensky, O.; Muller, F.; Gosele, U. Applied Physics Letters 1998, 72 1173-1175. (28) Li, A.-P.; Muller, F.; Birner, A.; Nielsch, K.; Gosele, U. Advanced Materials (Weinheim, Germany) 1999, 11 483-487. (29) Thompson, G. E. Thin Solid Films 1997, 297 192-201. (30) Yakovleva, N. M.; Yakovlev A. N.; Chupakhina, E. A. Thin Solid Films 2000, 366 37-42. (31) Masuda, H.; Fukuda, K. Science (Washington, D. C.) 1995, 268 1466-1468. (32) Whatman, 2004. (33) Lux, K. W.; Rodriguez, K. J. Nano Letters 2006, 6 288-295. (34) Vullum, F.; Teeters, D. Journal of Power Sources 2005, 146 804-808. (35) Colbert, D. T.; Zhang, J.; McClure, S. M.; Nikolaev, P.; Chen, Z.; Hafner, J. H.; Owens, D. W.; Kotula, P. G.; Carter, C. B.; et al. Science (Washington, D. C.) 1994, 266 1218-1222. (36) Journet, C.; Maser, W. K.; Bernier, P.; Lo iseau, A.; Lamy de la Chapells, M.; Lefrant, S.; Deniard, P.; Lee, R.; Fischer, J. E. Nature (London) 1997, 388 756-758. (37) Guo, T.; Nikolaev, P.; Thess, A.; Colbert, D. T.; Smalley, R. E. Chemical Physics Letters 1995, 243 49-54. (38) Thess, A.; Lee, R.; Nikolaev, P.; Dai, H.; Petit, P.; Robert, J.; Xu, C.; Lee, Y. H.; Kim, S. G.; et al. Science (Washington, D. C.) 1996, 273 483-487. (39) Che, G.; Lakshmi, B. B.; Martin, C. R.; Fisher, E. R.; Ruoff, R. S. Chemistry of Materials 1998, 10 260-267.

PAGE 115

115 (40) Masuda, H.; Asoh, H.; Watanabe, M. ; Nishio, K.; Nakao, M.; Tamamura, T. Advanced Materials (Weinheim, Germany) 2001, 13 189-192. (41) Yanagishita, T.; Sasaki M.; Nishio, K.; Masuda, H. Advanced Materials (Weinheim, Germany) 2004, 16 429-432. (42) Hornyak, G. L.; Patrissi, C. J.; Martin, C. R. Journal of Physical Chemistry B 1997, 101 1548-1555. (43) Foss, C. A., Jr.; Hornyak, G. L.; Stockert, J. A.; Martin, C. R. Journal of Physical Chemistry 1994, 98 2963-2971. (44) Lee, S. B.; Mitchell, D. T.; Trofin, L.; Nevanen, T. K.; Soederlund, H.; Martin, C. R. Science (Washington, DC, United States) 2002, 296 2198-2200. (45) Blondel, A.; Meier, J. P.; Doudin, B.; Ansermet, J. P. Applied Physics Letters 1994, 65 3019-3021. (46) Schwarzacher, W. Electrochemical Society Interface 1999, 8 20-24. (47) Menon, V. P.; Martin, C. R. Analytical Chemistry 1995, 67 1920-1928. (48) Ang, L.-M.; Hor, T. S. A.; Xu, G.-Q.; Tung, C.-h.; Zhao, S.; Wang, J. L. S. Chemistry of Materials 1999, 11 2115-2118. (49) Parthasarathy, R. V.; Phani, K. L. N.; Martin, C. R. Advanced Materials (Weinheim, Germany) 1995, 7 896-897. (50) Cepak, V. M.; Martin, C. R. Chemistry of Materials 1999, 11 1363-1367. (51) Lakshmi, B. B.; Dorhout, P. K.; Martin, C. R. Chemistry of Materials 1997, 9 857-862. (52) Bowes, C. L.; Malek, A.; Ozin, G. A. Chemical Vapor Deposition 1996, 2 97-103. (53) Shelimov, K. B.; Moskovits, M. Chemistry of Materials 2000, 12 250-254. (54) Cepak, V. M.; Hulteen, J. C.; Che, G.; Jirage, K. B.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R.; Yoneyama, H. Chemistry of Materials 1997, 9 1065-1067. (55) John, P. The Chemistry of the Semiconductor Industry ; Chapman and Hall: New York, 1987. (56) Kyotani, T.; Ma, Z.; Tomita, A. Carbon 2003, 41 1451-1459. (57) Che, G.; Miller, S. A.; Fisher, E. R.; Martin, C. R. Analytical Chemistry 1999, 71 31873191. (58) Kyotani, T.; Tsai, L.-f.; Tomita, A. Chemistry of Materials 1996, 8 2109-2113.

PAGE 116

116 (59) Kyotani, T.; Pradhan, B. K.; Tomita, A. Bulletin of the Chemical Society of Japan 1999, 72 1957-1970. (60) Saito, Y.; Hata, K.; Murata, T. Japanese Journal of Applied Physics, Part 2: Letters 2000, 39 L271-L272. (61) Brumlik, C. J.; Martin, C. R.; Tokuda, K. Analytical Chemistry 1992, 64 1201-1203. (62) Brumlik, C. J.; Menon, V. P.; Martin, C. R. Journal of Materials Research 1994, 9 11741183. (63) Foss, C. A., Jr.; Hornyak, G. L.; Stockert, J. A.; Martin, C. R. Journal of Physical Chemistry 1992, 96 7497-7499. (64) Foss, C. A., Jr.; Hornyak, G. L.; Stockert, J. A.; Martin, C. R. Advanced Materials (Weinheim, Germany) 1993, 5 135-136. (65) Tierney, M. J.; Martin, C. R. Journal of Physical Chemistry 1989, 93 2878-2880. (66) Martin, C. R. Advanced Materials (Weinheim, Germany) 1991, 3 457-459. (67) G.O. Mallory, J. B. H., Eds. Electroless plating: Fundam entals and Applications ; William Andrew Publishing: Noyes, 1990. (68) Nishizawa, M.; Menon, V. P.; Martin, C. R. Science (Washington, D. C.) 1995, 268 700702. (69) Hulteen, J. C.; Jirage, K. B.; Martin, C. R. Journal of the American Chemical Society 1998, 120 6603-6604. (70) Nagaura, T.; Tozawa, K. Progress in Batteries & Solar Cells 1990, 9 209-217. (71) Dahn, J. R.; Zheng, T.; Liu, Y.; Xue, J. S. Science (Washington, D. C.) 1995, 270 590-593. (72) Sato, K.; Noguchi, M.; Demachi, A.; Oki, N.; Endo, M. Science (Washington, DC, United States) 1994, 264 556-558. (73) Johnson, B. A.; White, R. E. Journal of Power Sources 1998, 70 48-54. (74) Fransson, L.; Eriksson, T.; Edstrom, K.; Gustafsson, T.; Thomas, J. O. Journal of Power Sources 2001, 101 1-9. (75) Long Jeffrey, W.; Dunn, B.; Rolison Debra, R.; White Henry, S. Chem Rev FIELD Full Journal Title:Chemical reviews 2004, 104 4463-4492. (76) Vullum, F.; Teeters, D.; Nyten, A.; Thomas, J. Solid State Ionics 2006, 177 2833-2838. (77) Wang, Z.; Li, F.; Ergang, N. S.; Stein, A. Chemistry of Materials 2006, 18 5543-5553.

PAGE 117

117 (78) Lee, K. T.; Lytle, J. C.; Ergang, N. S.; Oh, S. M.; Stein, A. Advanced Functional Materials 2005, 15 547-556. (79) Ergang, N. S.; Lytle, J. C.; Lee, K. T.; Oh, S. M.; Smyrl, W. H.; Stein, A. Advanced Materials (Weinheim, Germany) 2006, 18 1750-1753. (80) Golodnitsky, D.; Yufit, V.; Nathan, M.; Shechtman, I.; Ripenbein, T.; Strauss, E.; Menkin, S.; Peled, E. Journal of Power Sources 2006, 153 281-287. (81) Nathan, M.; Golodnitsky, D.; Yufit, V.; St rauss, E.; Ripenbein, T.; Shechtman, L.; Menkin, S.; Peled, E. Materials Research Society Symposium Proceedings 2005, 835 367-372. (82) Freemantle, M. Chem. Eng. News 1996, 74 (83) Baum, R. M. Chem. Eng. News 1997, 75 39-41. (84) Dillon, A. C.; Jones, K. M.; Bekkedahl, T. A.; Kiang, C. H.; Bethune, D. S.; Heben, M. J. Nature (London) 1997, 386 377-379. (85) Gao, B.; Kleinhammes, A.; Tang, X. P.; Bower, C.; Fleming, L.; Wu, Y.; Zhou, O. Chemical Physics Letters 1999, 307 153-157. (86) Planeix, J. M.; Coustel, N.; Coq, B.; Brotons V.; Kumbhar, P. S.; Dutartre, R.; Geneste, P.; Bernier, P.; Ajayan, P. M. Journal of the American Chemical Society 1994, 116 79357936. (87) Sloan, J.; Cook, J.; Green, M. L. H.; Hutchison, J. L.; Tenne, R. Journal of Materials Chemistry 1997, 7 1089-1095. (88) Tseng, G. Y.; Ellenbogen, J. C. Science (Washington, DC, United States) 2001, 294 12931294. (89) McEuen, P. L. Nature (London) 1998, 393 15, 17. (90) Tans, S. J.; Verschueren, A. R. M.; Dekker, C. Nature (London) 1998, 393 49-52. (91) Ouyang, M.; Huang, J.-L.; Cheung, C. L.; Lieber, C. M. Science (Washington, DC, United States) 2001, 292 702-705. (92) Rueckes, T.; Kim, K.; Joselevich, E.; Tseng, G. Y.; Cheung, C.-L.; Lieber, C. M. Science (Washington, D. C.) 2000, 289 94-97. (93) Davis, J. J.; Coleman, K. S.; Azamian, B. R.; Bagshaw, C. B.; Green, M. L. H. Chemistry-A European Journal 2003, 9 3732-3739. (94) Azamian, B. R.; Davis, J. J.; Coleman, K. S.; Bagshaw, C. B.; Green, M. L. H. Journal of the American Chemical Society 2002, 124 12664-12665.

PAGE 118

118 (95) Kong, J.; Franklin, N. R.; Zhou, C.; Chapline, M. G.; Peng, S.; Cho, K.; Dailt, H. Science (Washington, D. C.) 2000, 287 622-625. (96) Tang, Y. H.; Wang, N.; Zhang, Y. F.; Lee, C. S.; Bello, I.; Lee, S. T. Applied Physics Letters 1999, 75 2921-2923. (97) Liao, K.-H.; Ting, J.-M. Carbon 2004, 42 509-514. (98) Tsai, S. H.; Chao, C. W.; Lee, C. L.; Liu, X. W.; Lin, I. N.; Shih, H. C. Electrochemical and Solid-State Letters 1999, 2 247-250. (99) Fang, B.; Binder, L. Journal of Power Sources 2006, 163 616-622. (100) Kabbour, H.; Baumann, T. F.; Satcher, J. H., Jr.; Saulnier, A.; Ahn, C. C. Chemistry of Materials 2006, 18 6085-6087. (101) Fairen-Jimenez, D.; Carrasco -Marin, F.; Moreno-Castilla, C. Carbon 2006, 44 2301-2307. (102) Kroto, H. W. Nature (London) 1992, 359 670 671 (103) Ugarte, D. Nature (London, United Kingdom) 1992, 359 707-709. (104) Brodd, R. J. Electrochemical Society Interface 1999, 8 20-23. (105) Halpert, G.; Frank, H.; Surampudi, S. Electrochemical Society Interface 1999, 8 25-30. (106) Lin, L. Y., Wu, M.C., Sawada, R., Mohr, J.A. J. Lightwave Technol. 2003, 21 582-583. (107) Niarchos, D. Sensors and Actuators, A: Physical 2003, A106 255-262. (108) Tilmans, H. A. C., Raedt, W.D., Beyne, E. J. Micromech. Microeng. 2003, 13 S139-S163. (109) Mehregany, M.; Zorman, C. A. MRS Bulletin 2001, 26 289-290. (110) Bates, J. B.; Lubben, D.; Dudney, N. J.; Hart, F. X. Journal of the Electrochemical Society 1995, 142 L149-L151. (111) Bates, J. B.; Gruzalski, G. R.; Dudney, N. J.; Luck, C. F.; Yu, X. Solid State Ionics 1994, 70-71 619-628. (112) Jeon, E. J.; Shin, Y. W.; Nam, S. C.; Cho, W. I.; Yoon, Y. S. Journal of the Electrochemical Society 2001, 148 A318-A322. (113) Lee, S.-H.; Liu, P.; Tracy, C. E.; Benson, D. K. Electrochemical and So lid-State Letters 1999, 2 425-427. (114) Baba, M.; Kumagai, N.; Kobayashi, H.; Nakano, O.; Nishidate, K. Electrochemical and Solid-State Letters 1999, 2 320-322.

PAGE 119

119 (115)Park, Y.-S.; Lee, S.-H .; Lee, B.-I.; Joo, S.-K. Electrochemical and Sol id-State Letters 1999, 2 58-59. (116) Hulteen, J. C.; Martin, C. R. Journal of Materials Chemistry 1997, 7 1075-1087. (117) Li, N.; Patrissi, C. J.; Che, G.; Martin, C. R. Journal of the Electrochemical Society 2000, 147 2044-2049. (118) Nishizawa, M.; Mukai, K.; Kuwaba ta, S.; Martin, C. R.; Yoneyama, H. Journal of the Electrochemical Society 1997, 144 1923-1927. (119) Delmas, C.; Cognac-Auradou, H.; Cocciantel li, J. M.; Menetrier, M.; Doumerc, J. P. Solid State Ionics 1994, 69 257-264. (120) Endo, M.; Kim, C.; Nishimura, K.; Fujino, T.; Miyashita, K. Carbon 2000, 38 183-197. (121) Fong, R.; Von Sacken, U.; Dahn, J. R. Journal of the Electrochemical Society 1990, 137 2009-2013. (122) Kinoshita, K.; Zaghib, K. Journal of Power Sources 2002, 110 416-423. (123) Wang, T.; Liu, X.; Zhao, D.; Jiang, Z. Chemical Physics Letters 2004, 389 327-331. (124) Croce, F.; Appetecchi, G. B.; Persi, L.; Scrosati, B. Nature (London) 1998, 394 456-458. (125) Meyer, W. H. Advanced Materials (Weinheim, Germany) 1998, 10 439-448. (126) Lightfoot, P.; Mehta, M. A.; Bruce, P. G. Science (Washington, DC, United States) 1993, 262 883-885. (127) Gray, F. M. Solid Polymer Electrolytes. Fundam entals and Technological applications 1991. (128) Appetecchi, G. B.; Hassoun, J.; Scrosati, B.; Croce, F.; Cassel, F.; Salomon, M. Journal of Power Sources 2003, 124 246-253. (129) Lemont, S.; Billaud, D. Journal of Power Sources 1995, 54 338-341. (130) Appetecchi, G. B.; Shin, J. H.; Alessandrini, F.; Passerini, S. Journal of Power Sources 2005, 143 236-242. (131) Johansson, A.-S.; Carlsson, J.-O. Thin Solid Films 1995, 261 52-58. (132) Jeong, S.-H.; Hwang, H.-Y.; Lee, K.-H.; Jeong, Y. Applied Physics Letters 2001, 78 20522054. (133) Wu, Q. B.; Ren, S.; Deng, S. Z.; Chen, J.; She, J. C.; Xu, N. S. Surface and Interface Analysis 2004, 36 493-496.

PAGE 120

120 (134) Jean, M.; Desnoyer, C.; Tranchant, A.; Messina, R. Journal of the Electrochemical Society 1995, 142 2122-2125. (135) Miyazawa, K. i.; Minato, J.-i.; Yoshii, T.; Fujino, M.; Suga, T. Journal of Materials Research 2005, 20 688-695. (136) Mintmire, J. W.; Dunlap, B. I.; White, C. T. Physical Review Letters 1992, 68 631-634. (137) Martin, C. R., Sides, C. R., Xu. F. Elelctrochemical Society Transactions 2005, 13-36. (138) Apel, P. Y.; Korchev, Y. E.; Siwy, Z.; Spohr, R.; Yoshida, M. Nuclear Instruments & Methods in Physics Research, Section B: Beam Interactio ns with Materials and Atoms 2001, 184 337-346. (139) Harrell, C. C.; Choi, Y.; Horne, L. P.; Baker, L. A.; Siwy, Z. S.; Martin, C. R. Langmuir 2006, 22 10837-10843. (140) Heins Elizabeth, A.; Siwy Zuzanna, S.; Baker Lane, A.; Martin Charles, R. Nano Lett FIELD Full Journal Title:Nano letters 2005, 5 1824-1829.

PAGE 121

121 BIOGRAPHICAL SKETCH Fan Xu was born as a single child in a teachers family. She spent her 4-year undergraduate study in East China of Science and Technology in Shanghai and obtained a B.S. in Chemical Engineering in June 2000. After th at, she went to Fudan Univeristy in Shanghai and studied in the fields of electro chemistry and batteries under the guidance of Dr. Zhiyu Jiang. In August 2003, after she obtained her M.S. in Physic al Chemistry, Fan Xu joined Dr. Charles R. Martin group in the Department of Chemistry at the University of Florida and continued her study in the fields of electroche mistry, Li-ion batteries and nano materials. She completed her research in July 2007, obtaining a doctor of philosophy degree.