<%BANNER%>

Effect of gas composition and carbon activity on the growth of carbon nanotubes

University of Florida Institutional Repository

PAGE 1

EFFECT OF GAS COMPOSITION AND CARBON ACTIVITY ON THE GROWTH OF CARBON NANOTUBES By ARCHIT LAL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Archit Lal

PAGE 3

This document is dedicated to my grandmother and grandfather Mr. and Mrs. Malhotra.

PAGE 4

ACKNOWLEDGMENTS This thesis is the fruit of the enduring support, love and cooperation of my mother Mrs. Neera Lal, father Dr. Jagdish Lal and uncle Mr. Mandeep Singh Bains. Though, being thousands of miles away they have been unshakable pillars of support, helping me fight and win every battle. Their faith in my abilities has been my strength and their visions my inspiration. All those words of wisdom have been my driving force through all those hard times. No words said can express how indebted I am to them for what I am today and for what I shall accomplish in the future. I would also like to thank my sister, Ruchika, for being there for me and being my motivation. I would also like to thank my fianc Indu who talking me through things whenever I felt down and out. Dr. Sigmund, my advisor, my teacher, my supervisor and my mentor taught me more than I ever hoped to learn here at graduate school. He taught me how to think individually and gave me all the independence to do my research in a very professional manner. His work has been my inspiration. This work has been a product of his patience and endurance. He has inspired me to be a better researcher and also a better person. He understood my problems and has helped me to succeed in spite of them. My success is and will be a reflection of his outstanding abilities as a teacher. I hope to do my best in whatever task I undertake and always strive for perfection, as a tribute to him. Nothing short of this will be adequate to express my gratitude to him. iv

PAGE 5

I would like to thank Dr. Singh for all the support he has given me through the entire project. His support and cooperation all through the project helped me immensely. I am grateful to Dr. Butt for agreeing to be on my committee and giving me valuable suggestions. I would also like to thank Nagraj Kulkarni for his guidance and helping me with a lot of stuff that was not known to me. I would also like to thank all the students in our research group Dr. Sigmunds Research group for being the greatest group ever-helpful, supportive and understanding. I would also like to thank the entire Materials Science and Engineering department for making this possible. v

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 2 GENERAL THEORIES AND FABRICATION METHODS FOR SYNTHESIS OF CARBON NANOSTRUCTURES................................................................................3 History........................................................................................................................ ..3 Fabrication Methods.....................................................................................................3 Applications of carbon nanotubes................................................................................5 3 WHY CHOOSE THE TERNARY PHASE DIAGRAM.............................................6 4 GAS EQULIBRIA ANALYSIS OF THE CARBON-HYDROGEN-OXYGEN(C-HO) SYSTEM USING PREDOMINANCE DIAGRAM.............................................10 SOLGASMIX (HSC Chemistry).........................................................................11 Constrained Chemical Po tentials Method (CCPM)............................................12 CCPM Software: Thermodynamic Basis............................................................14 5 RAW MATERIALS AND EXPERIMENTAL TECHNIQUES................................17 Catalyst Loading and Generation...............................................................................17 Deposition of Ni Film.................................................................................................18 Characterization of the Catalysts................................................................................19 Carbon Nanotube and Nanofiber Synthesis................................................................22 Characterization of the Carbon Nanotubes and Fibers...............................................24 Scanning Electron Microscopy............................................................................24 Transmission Electron Microscopy.....................................................................25 Raman Shift.........................................................................................................25

PAGE 7

6 RESULTS AND DISCUSSION.................................................................................27 Results for Composition 40CH4/20CO2/41H2/Ni.....................................................28 Results for Composition 15 CH 4 /50 CO 2 /15 H 2 /Ni....................................................36 Results for Composition 30 CH 4 /30 CO 2 /30H 2 ..........................................................43 Results for Composition 45CH 4 /50 H 2 /Ni..................................................................50 Results for Composition 50 sccm CO 2 .......................................................................55 Experiments Using Iron Nanoparticles as Catalyst....................................................60 7 CONCLUSIONS........................................................................................................63 LIST OF REFERENCES...................................................................................................65 BIOGRAPHICAL SKETCH.............................................................................................68 vii

PAGE 8

LIST OF TABLES Table page 2-1 Applications of carbon nanotubes.................................................................................5 5-1. Raw materials used for the experiments.....................................................................17 5-2 The various gas compositions that were used using Nickel as catalyst. nX(O), nX(C), nX(H) being the atom fractions for the various gases respectively.........................22 5-3 The various gas composition using iron nanoparticles as a catalyst...........................24 viii

PAGE 9

LIST OF FIGURES Figure page 2-1 Experimental sets up showing the CVD reactor............................................................5 3-1 The Carbon-H-O ternary phase diagram showing the different compositions and activities of carbon.....................................................................................................7 3-2 Non-diamond growth region, the gas phase region. (App. Phys.A 67 225)..................9 4-1 C-H-O diagram with the partial pressures of O 2 and activity of carbon. {Program developed by Nagraj Kulkarni (University of Florida) now at ORNL}..................11 5-1 Set up for making nanoparticles..................................................................................18 5-2 Average particle size distribution of the iron nanoparticles using Zeta Plus..............19 5-3 (A)TEM image of iron nanoparticles at 200 X.well dispersed Iron Nanoparticles.(B) TEM image of iron nanoparticles at 200X. Agglomerated iron nanoparticles can be seen in the right hand corner....................................................................................20 5-4 Growth of carbon nanotubes over annealed Ni catalyst film......................................21 5-5 EDS of Ni film over silica substrate. The peaks are oxygen, nickel and silicon respectively...............................................................................................................21 5-6 The C-H-O ternary phase diagram showing the different compositions and activities of carbon...................................................................................................................23 6-1 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 40sccm CH 4 20sccmCO 2 and 41sccm H 2 ..........................................................................................................29 6-2 SEM micrograph showing carbon nanostructures at 40CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.................................................31 6-3 TEM image of nanotubes, which were highly entwined. 40CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.................................................32 6-4 TEM micrograph showing a nanotube with some particles inside the tube due to the presence of CO 2. 40CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.................................................................................................32 ix

PAGE 10

6-5 HRTEM showing MWCNTs with graphitic fringes and a lattice spacing of 0.34nm. 40 CH4/20 CO2/41 H2/Ni composition at 1040 C and 1 atmosphere pressure.....33 6-7 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 40 CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.....................................................................................................................34 6-8 Raman spectra of as prepared sample showing second order peaks (632.8-nm helium-neon laser). 40 CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.....................................................................................................................34 6-9 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 15sccm CH 4 50sccmCO 2 and 15sccm H 2 ..........................................................................................................35 6-10 SEM micrograph showing carbon nanostructures at 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure.................................................38 6-11 SEM micrograph showing carbon nanostructures at 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure.................................................38 6-12 HRTEM image of a closed cap MWCNT. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure.........................................................................39 6-13 HRTEM image of a carbon nano-onion. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure.........................................................................39 6-18 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 30sccm CH 4 30sccmCO 2 and 30sccm H 2 ..........................................................................................................42 6-23 HRTEM showing the synthesis of carbon nanofiber along with the nanotubes30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure............47 6-25 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.....................................................................................................................48 6-26 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 45sccm CH 4 and 50sccm H 2 ..............................................................................................................................49 6-27 SEM image of carbon nanostructures. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure..............................................................................................51 6-28 HRTEM showing MWCNTs with graphitic fringes. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure......................................................................51 x

PAGE 11

6-29 HRTEM showing a bunch MWCNTs with graphitic fringes and a lattice spacing of 0.34nm. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure......52 6-30 HRTEM showing a carbon nanotube in the shape of a balloon. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.................................................52 6-31 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.......53 6-32 The C-H-O ternary phase diagram which shows the different gas compositions and activities of carbon. The one highlighted corresponds 50sccm CO 2 50sccm CO 2 ....54 6-33 SEM image of nanostructures. 50 sccm CO 2 ............................................................56 6-34 SEM image of nanostructures. 50 sccm CO 2 ............................................................56 6-35 HRTEM image showing amorphous Sio X 50 sccm CO 2 ..........................................57 636 HRTEM image showing amorphous SiO X nanowire with catalyst particle attached at the tip. 50 sccm CO 2 .............................................................................................57 6-37 HRTEM image showing some graphitic layers with a lattice spacing of 0.34nm which were also present. 50 sccm CO 2 ....................................................................58 6-38 Raman spectra of as prepared sample (632.8-nm helium-neon laser). 50 sccm CO 2 58 6-39 Raman spectra of as prepared sample showing the silica peak (632.8-nm helium-neon laser). 50 sccm CO 2 .........................................................................................59 6-40 EDS Spectra showing Si, O and Ni peaks. 50 sccm CO 2 ..........................................59 641 80 CH 4 /17 CO 2 SEM image showing nanostructures...............................................60 642 50 CH 4 /50 CO 2 SEM image showing nanostructures..............................................60 643 15 CH 4 /50 CO 2 SEM image showing nanostructures..............................................61 644 50 CO 2 SEM image showing nanostructures...........................................................61 645 EDS at 50 CO 2 ..........................................................................................................62 xi

PAGE 12

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EFFECT OF GAS COMPOSITION AND CARBON ACTIVITY ON THE GROWTH OF CARBON NANOTUBES By Archit Lal August 2003 Chair: Dr Wolfgang Sigmund Major Department: Materials Science and Engineering This thesis tests the hypothesis that carbon nanostructures deposited, using a mixture of various gases that contain carbon, hydrogen and oxygen are a part of the C-H-O ternary phase diagram system. Synthesis of various forms of carbon by using different gas compositions keeping the temperatures and pressure constant and changing the carbon activity is reported. It is generally accepted that the activity of carbon greater than unity is a necessary condition for the deposition of carbon. In contrast to this, it was found that carbon deposition occurred even when the carbon activity was less than unity. Furthermore, the ternary phase diagram provides general guidelines about suitable conditions for growth of different carbon species. Graphitic carbon nanotubes, nanofibers, onions and amorphous carbon nanowires were synthesized by using equilibrated C-H-O gases, in the presence of a Ni catalyst, at 1040 C in a CVD chamber. The main objective of this thesis is to demonstrate the influence of initial gas composition, activity and to an extent temperature and pressure on the growth of xii

PAGE 13

nanotubes. TEM, Raman spectroscopy and SEM show that these carbon nanostructures consist of graphite layers and some impurities. xiii

PAGE 14

CHAPTER 1 INTRODUCTION This thesis tests the hypothesis that carbon nanostructures deposited, using a mixture of various gases that contain carbon, hydrogen and oxygen are a part of the C-H-O ternary phase diagram system, which was previously published for diamond [1]. It has been found that the diamond synthesis in a CVD is restricted not only by the gas phase composition, but also the temperature and pressure [2, 3]. While investigating the materials deposited on an arc-evaporation cathode, Japanese microscopist Sumio Iijima [4] discovered carbon nanotubes. In this thesis project, study of the C-H-O ternary phase diagram and how this diagram will help us in selecting the starting gas compositions for growing carbon nanotubes is demonstrated. It also describes the influence of activity of carbon and its importance for the growth of carbon nanotubes. Different characterization techniques like SEM, TEM and Raman were used to characterize the carbon nanostructures. Chapter 2 deals with an introduction to carbon nanotubes. A brief history, discovery and fabrication methods are described. The applications of nanotubes for the various engineering majors are mentioned in a table. A schematic of the CVD process used in our experiments is also drawn. Chapter 3 describes the motivation to explore the C-H-O ternary phase diagram and the various hypotheses in the process of fabrication of nanostructures. Many examples of various researchers have been illustrated showing the dependence on the C-H-O phase diagram. 1

PAGE 15

2 Chapter 4 deals with the predominance diagram. The compositions of various gases in the C-H-O regime are plotted on the Gibbs triangle. In chapter 5, the various characterizations on the raw materials taken are explained. The methodology for the experiments and the equipments used for the various experiments are also mentioned here. Plots of the C-H-O phase diagram with the various experimental conditions are also plotted in this chapter. Chapter 6 deals with the results and discussion of the various compositions used. Chapter 7 gives a brief conclusion to the results of the experiments.

PAGE 16

CHAPTER 2 GENERAL THEORIES AND FABRICATION METHODS FOR SYNTHESIS OF CARBON NANOSTRUCTURES History The earliest report on vapor growth of carbon fibers is as early as 1889 patent [5] describing the growth of hair like carbon filaments from methane decomposition in an iron crucible. It was not until the invention of the electron microscope that Davis et al. [6] almost 50 years ago observed minute vermicular growth (from 10nm to 200nm in thickness) that penetrated the brickwork of blast furnaces. A plethora of studies using filamentous carbon was later reviewed by Baker and Harris [7]. Carbon is a very unique element because it can assume various forms and structures. Carbon nanotubes are byproducts in the fullerene synthesis and were discovered by Sumio Iijima in 1991[4]. The first to be discovered in 1991 by Iijima were the multi-walled carbon nanotubes (MWCNTs) made of a few to a few tens of concentric cylinders placed around a common central hollow, with the interlayer spacing close to that of graphite (0.34 nm).The other nanotubes are called as single walled carbon nanotubes (SWCNTs), which are composed of only one graphite sheet and are in the range of 1-2 nm [8]. Fabrication Methods Various techniques are capable of synthesizing carbon nanotubes. The carbon arc method [4], used initially for producing C 60 fullerenes is the most common and perhaps easiest way to produce carbon nanotubes. Chemical vapor deposition, [9] an apparatus 3

PAGE 17

4 used for creating vapor grown carbon fibers, has also produced carbon nanotubes. Finally, laser vaporization of a carbon block has produced the most uniform single wall carbon nanotubes. Laser ablation [10] has emerged as an alternative method to produced SWNT. In this process; a target consisting of graphite mixed with small amount of transition metal particles (as catalysts) is placed at the end of a quartz tube enclosed in a furnace. The target is exposed to an argon ion laser beam, which vaporizes graphite and nucleates carbon nanotubes in the short wave just in front of the furnace, carries the vapor and nucleated nanotubes which continue to grow. The nanotubes get deposited on the cooler walls of the tube downstream of the furnace. This method produces about 60-70% SWNTs with the rest being catalyst particle and graphitic soot. In this work the CVD method is used for growing carbon nanotubes. Previous results demonstrate how such a system can be used to control nanotube diameter and length [11]. CVD has been very effective for growth of pattern substrates. Thus CVD can be practiced either as a continuous gas phase process with floating catalyst or a substrate growth process with heterogeneous catalyst. Figure 2-1 shows us a schematic of the CVD reactor used. It consisted of a quartz tube which is 1.5 inches in diameter and 48 inches in length. The gases were passed through the Rota meters which measured the flow before the gases enter into the chamber. A temperature controller was used to set the cycles for heating and cooling.

PAGE 18

5 FLOW METERS GAS INLET SAMPLE QUARTZ BOAT GAS OUTLET H 2 CO 2 CH 4 OVEN TEMP 1040C Figure 2-1 Experimental sets up showing the CVD reactor. Applications of carbon nanotubes The high elastic modulus of the graphene sheets makes carbon nanotubes one of the toughest materials ever known [12]. The density of MWCNTs is 2.6 gm/cm 3 as compared to that of steel, which is 7.8 gm/cm 3 [13]. They have application in a variety of engineering majors. Some of the important applications are listed in the table below. Table 2-1 Applications of carbon nanotubes S.No Engineering majors Application 1) Electrical engineering Nanotube field effect transistors[14] CNT Inter and intermolecular logic gates [15] Vacuum microelectronics[16] 2) Biomedical engineering Nanoprobes and sensors[16] 3) Electrochemical engineering CNT-Lithium rechargeable batteries[16] 4) Mechanical engineering Air/space craft body parts 5) Construction engineering Lower density as compared to steel[13] 6) General engineering CNTs can store liquid and gas Hydrogen storage

PAGE 19

CHAPTER 3 WHY CHOOSE THE TERNARY PHASE DIAGRAM During the last decade, CVD has been used to synthesize carbon nanotubes; however, most of the deposition conditions have been involving gases like C 2 H 2 CH 4 ferrocene and xylene mixtures which are on the C-H side of the ternary phase diagram. Peter E. Nolan et al. was one of the first groups to describe a thermodynamic model explaining how hydrogen in low concentration controls filament morphology and why equilibrium is shifted from that for graphite during carbon deposition. Carbon deposition reaction rates at low carbon activity in the absence of hydrogen are reported [17]. He also stated that the thermodynamics of the carbon deposition system determines the size and structure of the carbons formed, and thus the size and shape of the catalyst particle. [17]. J.A.Libera[18] et al have synthesized various forms of carbon from the C-H-O system under hydrothermal conditions. Some other groups [19, 20] used CH 4 and H 2 in a PECVD system to which they added small quantities of O 2 were added (0-12%). They also show that small O 2 additives improved the purity of carbon nanotubes and surprisingly led to the high quality even growth of carbon nanotubes [19] they also showed that when the O 2 content was increased to 8% the ID/ IG relative intensity ratio increased drastically. The ID/IG ratio is the ratio of the intensities of the G and D band in the Raman respectively. The G-line peak at 1596 cm -1 which is a characteristic of graphite sheets [21].The broad peak near 1337 cm -1 indicates the existence of defective graphitic layers on the wall surfaces, the so called D band[22] which is related to structural disorder of sp 2 bonded nanocrystalline and /or amorphous carbon species 6

PAGE 20

7 Figure3-1 The Carbon-H-O ternary phase diagram showing the different compositions and activities of carbon. T = 800 K, Ptotal = 1 atm.C O H a(C) carbon isoactivity lines compositions of selected gaseous species isoactivity line for carbon, a(C) = 1 aC PO2 10 24 10 310 2710 265, 1025, 5010 2CH4C2H6CH2CO2H2OCOa(C) = 110 -210 2810 -1 However, a common scheme connecting all the experimental methods, gas composition and growth of nanotubes is still lacking. To develop a strategy to explore carbon deposition is necessary to have some hypothesis. Some of the hypothesis that were taken into consideration before the experiments were, 1. Which carbon carrier gases and gas mixtures are suitable for nanotube growth in CVD?( A gas mixture containing 3 gases CH 4 CO 2 and H 2 are used) 2. Is there a specific region in the ternary phase diagram where the nanotube growth can be controlled and longer nanotubes can be produced? 3. Why are certain CVD methods producing carbon nanotubes at higher growth rates?

PAGE 21

8 Checking the growth of carbon nanostructures using a mixture of gases and thereby changing the carbon activity and the oxygen and hydrogen partial pressures are studied in this work. We used 3 gases CH 4 CO 2 and H 2 in our experiments. Thus by changing the initial gas compositions of the 3 gases we found ourselves at different points in the gas composition triangle. In the past decade many gases and gas mixtures have been used to grow carbon nanostructures in the CVD system. Out of them CH 4 C 2 H 2 CO and a mixture of CH 4 and H 2 C 2 H 2 and H 2 have been used. In the floating catalyst method Andrews et al. reported a continuous production of aligned MWCNTs in bulk quantities through the catalytic decomposition of ferrocene-xylene mixture. CO 2 has been used in the past for the synthesis of carbon nanofibres. Carbon nanofibers and nanotubes were deposited on the Si substrate with CO 2 as the carrier gas through ethanol [23] but the gas ratio was ambiguous. Previously nanotubes were grown grew on a substrate using methane as a feeding gas, but less than 5 percent of the catalyst particles sprout nanotubes, "By switching the feeding gas from methane to a carbon monoxide and hydrogen mixture Liu at al significantly improved the amount of nanotubes grown on a surface. Such nanotubes have a range of favorable physical properties. Apart from all these examples the C-H-O Gibbs triangle has been used for the synthesis of diamond and a region has been described by P.K. Bachmann et al. [1] and the region has been confirmed by many other researchers. Figure 3-2 below shows us the region of the diamond domain, the non diamond growth region and the gas phase region. All these examples stated above demonstrate that the same is possible for the growth of carbon nanostructures as well.

PAGE 22

9 Figure 3-2 Non-diamond growth region, the gas phase region. (App. Phys.A 67 225)

PAGE 23

CHAPTER 4 GAS EQULIBRIA ANALYSIS OF THE CARBON-HYDROGEN-OXYGEN(C-H-O) SYSTEM USING PREDOMINANCE DIAGRAM The predominance diagram described above does not provide direct information regarding the detailed gas composition. This is because in addition to the gases that are represented on the axes of the predominance diagram (e.g., O 2 and H 2 ), a number of other gaseous species are also present. The mixture of all these gaseous species determines the gas composition [24, 25], which is best specified by the atom fractions of the elemental gases. For example, in the O-C-H system, a unique gas composition is given by the atom fractions, X C X O and X H This composition plots as a point on a Gibbs triangle (Figure41). For example, a gas mixture consisting of 20 % H 2 and 80 % CO 2 has a gas composition given by: 143.0)21)(8.0(2)2.0(2)2.0(HX 286.0)21)(8.0(2)2.0(1)8.0(CX 571.0)21)(8.0(2)2.0(2)8.0(OX (1) 1 OCHXXX The partial pressures of the various gases that exist in this system can be determined for this composition or for other compositions, from thermodynamics of the gases under equilibrium conditions. Different combinations of gas mixtures may be utilized to obtain the same gas composition (Figure 4-1). While the composition does not vary with temperature or total pressure in the system (conservation law), the partial 10

PAGE 24

11 pressures do vary. Knowledge of the variation in the partial pressures of the gases, such as oxygen, with composition and temperature, is obviously very important for controlling the results. This information can be obtained using two different thermodynamic approaches for determination of gaseous equilibria (or partial pressures). These are discussed below. T = 1000 K, Ptotal= 1 atm.C O H p(O2) oxygen isobars compositions of selected gaseous species isoactivity line for carbon, a(C) = 1 CH4C2H6CH2CO2COH2O aCPO20.90.80.70.60.50.40.30.210 -110 -3-10 -1510 -1710 -1910 -27 10-29 10-3010 -2310 -2510 -3110 -3310 -3510 -21a(C) = 110 -2010 -22 Figure 4-1 C-H-O diagram with the partial pressures of O 2 and activity of carbon. {Program developed by Nagraj Kulkarni (University of Florida) now at ORNL} SOLGASMIX (HSC Chemistry) The commercially available HSC Chemistry software [26] contains several modules for thermodynamic analysis. An important module is the SOLGASMIX code, which is capable of analyzing gaseous equilibria. The input to the SOLGASMIX

PAGE 25

12 calculation is the initial composition of the gas mixture, e.g., 10 % H 2 90 % CO 2 Then from thermodynamics of all the gaseous and condensed phase species that may exist in that relevant system, the variation in the partial pressures of all gaseous species is computed as a function of temperature. If precipitation of condensed phases is thermodynamically favored, the calculation also provides the equilibrium compositions and amounts of all species in that system. To accomplish this, a sophisticated equilibrium solver, which is the critical component of the HSC software, is necessary. Usually rapid convergence to the correct solution is desired from the equilibrium codes developed. The results are given in a tabular form but more commonly in the form of a graph that gives the partial pressure of a particular gaseous species as a function of temperature, see Figure 4-1. However, the calculation has to be repeated for a different choice of the initial gas composition. This is the main drawback of this technique, since one does not know a priori, which gas composition or system will provide the desired partial pressure at a given temperature. Nevertheless, with some experience and intuition, it is possible to use the SOLGASMIX code in a more predictive capacity. The HSC Chemistry software contains an exhaustive thermodynamic database, that is updated frequently, but is not critically evaluated, which renders its usage for an uninitiated user rather difficult. For gases, ideal behavior is usually assumed although it does permit non-ideal terms to be added. It also allows the user to utilize his or her own database should the need arise. Its user friendliness and robust equilibrium solver are its major advantages. Constrained Chemical Potentials Method (CCPM) In many practical problems of atmosphere control including the present, it is desired to predict the possible atmospheres that can provide a fixed value of partial pressures of gases at equilibrium. The approach is the exact opposite of that given above.

PAGE 26

13 For e.g., if one wants to fix a particular set of values for pO 2 and pH 2 at equilibrium, what initial mixtures of say H 2 + H 2 O + or CO 2 + H 2 if any, will produce the desired equilibrium? It is quite possible that a number of different gas mixtures may provide the desired composition. Obviously the previous equilibrium approach can still be utilized but it can prove to be quite cumbersome. Also the initial choice of the gas mixtures for the equilibrium approach has to be provided by the user and is hence a function of the experience and intuition of the user. Having calculated the gas phase equilibria, how can one readily assess the influence of composition, temperature or total pressure upon the gaseous equilibria? What are the practical advantages of, say a H 2 + H 2 O + mixture over a H 2 +CO 2 mixture? Is a mixture of CO 2 + H 2 a poor or a good chemical buffer for pO 2 and activity of carbon, i.e., is the gas displaced from the desired equilibrium point by a small addition of impurities? Will C deposition occur at lower temperatures? Although such questions could be systematically answered with the aid of computerized tabulations available through equilibrium solvers such as SOLGASMIX, the problem often is that the influence of any one particular variable is not always clear since the properties are interdependent. The "constrained chemical potential method" developed by Bale et al. [24, 25], for calculating gaseous equilibria allows one to compute the partial pressures of gaseous species and display them on a Gibbs composition triangle. It is this graphical output of gaseous phase equilibria on a Gibbs composition triangle which is the most efficient way of answering many of the questions posed above (Figure 4-1). It was thus decided to develop a code based on this technique and apply it to the C-O-H and the H-O-N gas

PAGE 27

14 systems. The results obtained for these systems were indeed very useful in deciding appropriate compositions of gas mixtures that can achieve the desired conditions during the sintering of additive fuel pellets. The computational scheme for developing this software in MathCad is discussed below. CCPM Software: Thermodynamic Basis The thermodynamic basis for the CCPM approach is discussed using the C-H-O system as an example. (1) Initially the thermodynamic data for all the gaseous species in this system was obtained from the book by Kubaschewski and Spencer [27] inputted into an Excel spreadsheet and imported into MathCad. The thermodynamic data included the standard enthalpies and entropies of formation as well as the specific heat (Cp) data for the various gaseous species. The reference elements selected were solid carbon, oxygen and hydrogen gas. (2) The gas phase reactions were balanced and the equilibrium constants were computed at a specified temperature and pressure (1 atm.) A generic reaction for this system is )()(2)(2)(22gOHCgOkgHjsiCkji ;lnKRTG 222)2(.)(.)()(kjikjiOPHPCaOHCPK (2) where i, j and k are non-negative integers, G is the standard free energy of formation for the reaction at the temperature T (K) and K is the equilibrium constant. a(C) is the activity of carbon, P(H 2 ) and P(O 2 ) are the partial pressures of hydrogen and oxygen respectively. (3) The partial pressure (or activity) of any gaseous species is obtained from the above equation. KOPHPCaOHCPkjikji.)(.)(.)()(2222 (3)

PAGE 28

15 From the Gibbs phase rule, the three component gas phase system, has three degrees of freedom (f = c-p+2 = 3-1+2 =3). If the temperature and the total pressure are fixed, then the gas phase is defined, if any two of the three variables, a(C), P(H 2 ) or P(O 2 ) are also defined. Hence, it is possible to reduce the system of non-linear equations containing the equilibrium constant, into one non-linear equation in one unknown. (4) In the present instance, choose P(H 2 ) as the dependent variable. For a given value of P(0 2 ), choose a value for a(C) and compute P(H 2 ) using a non-linear equation solver based on the "Half-Interval Technique" [28]. (5) Compute the partial pressures of all gaseous species and hence determine the compositions (X O X C and X H ) for each value of a(C). (6) Vary a(C) incrementally and repeat the calculation for the same oxygen partial pressure hence obtaining the entire isobar. (7) To obtain different oxygen isobars, repeat the calculation for a different value of P(O 2 ). (8) Plot the gas compositions for each isobar on a Gibbs Composition Triangle. A similar procedure may be used to compute isoactivity lines for carbon or isobars for hydrogen. In general, it took about 1000-3000 calculations for each isobar and the time involved was about 10-12 minutes on a Pentium 266. Thus approx. 6-8 hrs of computer time was needed for an entire set of isobars for a given temperature and pressure. However, a rapid survey for a portion of a specified isobar could be done in 15-30 seconds in MathCad. A typical graphical output from the CCPM software, that contains oxygen isobars on a Gibbs Composition Triangle, is shown in Figure 4-1. The partial pressures for the oxygen isobars are indicated in the Figure.

PAGE 29

16 Similar plots can be obtained at different temperatures and pressures for other gaseous species in the system. [29,30,31,32,33,34].

PAGE 30

CHAPTER 5 RAW MATERIALS AND EXPERIMENTAL TECHNIQUES Table 5-1. Raw Materials used for the Experiments S.No Material Company 1) (Fe(CO) 5 ) 99.9% Aldrich 2) (TOPO) 99.9% Aldrich 3) CO 2 gas Aproximately 99% Strate Welding Supply 4) CH 4 gas (99.99%) Strate Welding Supply 5) H 2 gas (99.99%) Strate Welding Supply 6) Ar gas (99.99%) Strate Welding Supply Catalyst Loading and Generation The experiments were performed using two types of catalysts, iron and nickel. An n type (resistivity 1-10 cm, thickness from 254-305 m) silicon (100) wafer purchased from Montco Silicon Technologies, Inc. with naturally grown oxide layer was used as a substrate for the growth of carbon nanotubes. The spherical iron nanoparticles were prepared by the thermal decomposition of organometallic precursor (Fe(CO) 5 ) in the presence of a stabilizing surfactant, which is the modified procedure of the previously reported method (Figure5-1)[33,34]. A 0.2 mL portion of Fe(CO) 5 was added to 5.0 g of trioctylphosphine oxide (TOPO) at 320C under argon atmosphere, and the resulting solution was aged for 30 min at 320C. The preparation temperature was set to over 320C so that a complete thermal decomposition of Fe (CO) 5 could be achieved to generate metallic iron atoms. The reaction mixture was added into excess acetone to produce black precipitate. The precipitate was retrieved by centrifuging and this process 17

PAGE 31

18 was repeated several times. The synthesized powder form of spherical nanoparticles can be easily re-dispersed in pyridine to yield a clear homogeneous solution. [35, 36] The transmission electron micrograph (TEM) revealed that the iron nanoparticles are 20 nm in diameter but most of them are agglomerated. Iron nanoparticles have a strong tendency to agglomerate because of their magnetic characteristics and high specific surface energy as can be seen from the TEM image. Particle sizing was performed using the Zeta plus, which showed that the average size to the particles is about 110nm due to agglomeration. Figure 5-1 Set up for making nanoparticles. Deposition of Ni Film The silicon substrate with the naturally grown oxide was cleaned in acetone, isopropyl alcohol and deionized water bath prior to the Ni film deposition. The deposition was done using an E beam evaporation system. The base pressure was 1.5* 10 -6 and current was 0.12A.The thickness of the film as measured by the crystal monitor was 38 0 A .It is seen that Ni breaks into small islands due to surface tension, as well as the compressive stress due to the mismatch of the thermal expansion coefficients of Si and Ni. [37,38].

PAGE 32

19 Characterization of the Catalysts The catalysts were characterized using TEM, SEM, particle sizer and EDS. Figure 5-2 showing the average size distribution of the iron nanoparticles. The average size is 118nm as shown in the figure. Figure 5-3a and b show the TEM images of the iron catalyst particles. The size of the paricles is about 20nm but as iron as a strong tendency to agglomerate, hence the average size is 118nm which is shown by the Zeta plus. Figure 5-4 shows the SEM image of the annealed Ni film. Figure 5-5 shows the EDS spectra of a Ni coated Silica substrate. Figure 5-4 shows the growth of carbon nanostructures over an annealed Ni film. 020406080100120050100150200250300Diameter(nm)Number of particlesAverage Size=118.5nm Figure 5-2 Average particle size distribution of the iron nanoparticles using Zeta Plus.

PAGE 33

20 Figure 5-3 (A)TEM image of iron nanoparticles at 200 X.well dispersed Iron Nanoparticles.(B) TEM image of iron nanoparticles at 200X. Agglomerated iron nanoparticles can be seen in the right hand corner.

PAGE 34

21 Figure 5-4 Growth of carbon nanotubes over annealed Ni catalyst film Figure 5-5 EDS of Ni film over silica substrate. The peaks are oxygen, nickel and silicon respectively

PAGE 35

22 Carbon Nanotube and Nanofiber Synthesis A quartz tube that is 1.5 in. in diameter and 48 in. in length was used for the CVD furnace. The furnace temperature was increased by 6 C/min up to the growth temperature in argon atmosphere with a flowing speed of 1000 sccm. The sample was placed in a quartz boat and placed carefully in the centre of the heating zone. On reaching growth temperature, the various gas mixtures were introduced into the chamber simultaneously, the annealing was done in argon (1000 sccm) before growth of the CNTs. After growing the CNTs, the furnace was cooled to room temperature in argon (1000 sccm). Graphitic nanostructures have been synthesized using various conditions during the past decade. In our experiments we used a mild oxidizing agent in CO 2 The activity of carbon was decrease using the mild oxidizing agent with higher flow rates. Figure 5-6 shows us a C-H-O ternary phase diagram with the different experimental composition plotted at 1040 C and 1 atm pressure. These compositions are only for the Ni catalyst. Table 5-2 The various gas compositions that were used using Nickel as catalyst. nX(O), nX(C), nX(H) being the atom fractions for the various gases respectively. Flow in sccm nX(O) nX(C) nX(H) nX(O)% nX(C)% nX(H)% 40CH 4 /20 CO 2 /41H 2 0.1169 0.1754 0.7077 11.69 17.54 70.77 15CH 4 /50 CO 2 /15H 2 0.39216 0.2549 0.35294 39.216 25.49 35.294 30CH 4 /30 CO 2 /30H 2 0.2 0.2 0.6 20 20 60 45CH 4 /50H 2 0 0.1384 0.8616 0 13.84 86.16 50 CO 2 0.6667 0.3333 0 66.67 33.33 0

PAGE 36

23 T = 1040oC, Ptotal = 1 atmC O H PO2 H2OCO2COCH4C2H6CH2aC 10 1610 1510 210-14,10 510 310 410 110310250251051102 50 sccm CO215CH4/50CO2/15H230CH4/30CO2/30H240CH4/20CO2/41H245CH4/50H2 Figure 5-6 The C-H-O ternary phase diagram showing the different compositions and activities of carbon.

PAGE 37

24 Table 5-3 The various gas composition using iron nanoparticles as a catalyst. Gas Mixtures (sccm) Atom fraction Atom fraction Atom fraction Atom fraction% Atom fraction% Atom fraction% nX(O) nXC nX(H) nX(O)% nXC% nX(H)% 80CH4/17 CO 2 0.0755 0.215 0.7095 7.55 21.5 70.95 50CH4/50 CO 2 0.25 0.25 0.5 25 25 50 15CH4/50 CO 2 0.4444 0.2888 0.2668 44.44 28.88 26.68 50 CO 2 0.6667 0.3333 0 66.67 33.33 0 Characterization of the Carbon Nanotubes and Fibers Scanning Electron Microscopy The SEM imaging was carried out using the JEOL JSM6330F and the SEM 6400. The JEOL 6330 is a cold field emission scanning electron microscope. The cold field emission has the advantage of high brightness (large current density) and small beam diameter (high resolution) at low accelerating voltages to allow imaging of nanotubes without causing damage to the sample. Resolution of the instrument is around 1.5nm depending on the sample. The instrument has an energy dispersive X-ray spectrometer (EDS) for elemental analysis. The JEOL JSM-6400 imaging modes include secondary electrons, backscattered electrons and x-rays. Under ideal conditions, resolution in secondary electron imaging mode is 5 nm. Images can be recorded with Polaroid film and/or digital image capture through the XL system with subsequent output to a 1200 dpi laser printer.

PAGE 38

25 The sample was directly taken out of the furnace and attached onto the SEM sample holder by means of a double sided carbon tape. Transmission Electron Microscopy Transmission electron microscopy images of the carbon nanostructures were obtained using TEM 2010F microscope. The carbon nanostructures were scratched out from the substrate and dispersed in ethanol. The solution containing the carbon nanostructures were then ultrasonicated for 10 min before the Alicia copper grid was dipped in the solution. The sample was dried for about 36 hrs before observation. The freed nanostructures adhere well to the TEM grid and were easily imaged. Raman Shift Raman shift was performed using two machines. 1) The Raman shift was performed using a 647nm excitation using a krypton laser. Both the radial breathing modes and the second order bands were observed. The carbon nanostructures that were deposited on the silica substrate were directly observed sample as it is. 2) The second machine was used to get the second order peaks .A confocal micro-Raman spectrometer system (LabRam Infinity, Jobin Yvon) was used for all measurements. The excitation source was a 632.8-nm helium-neon laser operating with nominal output energy of 15 mW. The laser beam was focused on the sample using a 10x objective resulting in a focused beam spot of 35 mm on the corneal sample surface. The laser beam was attenuated with a neutral density filter prior to entering the microscope objective to yield a laser beam power of about 3 mW on the sample surface. Raman scattered light was collected in the backscatter mode through the microscope objective and dispersed onto a 1024-pixel CCD detector array using a 600 grooves/mm grating.

PAGE 39

26 The effective spectral dispersion was 0.12 nm/pixel, equal to about 2. cm-1/pixel over the specified wavelength range. The Raman microscope was configured for confocal microscopy. The confocal aperture was set to a diameter of 500 mm, which resulted in an effective integration depth of 300 mm.

PAGE 40

CHAPTER 6 RESULTS AND DISCUSSION These experiments are designed to establish broad limits on the synthesis conditions resulting in carbon tube growth changing the initial gas compositions and thus activity and partial pressures of oxygen. The results are presented in terms of three precursors namely CO 2 CH 4 and H 2 The synthesis of nanotubes, fibers, graphitic particles, onions and non planar graphitic structures is shown. The activities for carbon where these structures have grown are ranging from 100 to approximately 10 -2 which is several orders of magnitude less than one and the necessary condition for deposition of carbon. 99% CO 2 used in the experiments yields SiOx nanowires and very few carbon structures can be noticed. The response of rates to the partial pressures of oxygen and hydrogen is not studied in detail. Deposition rates have been plotted with respect to activity by Nolan et al. [17], showing a generally linear response. In all the Raman spectra except the one for CO 2 the degree of graphitization by the ID/IG ratio is almost the same and in agreement with the CVD growth of carbon nanostructures using a CH 4 and H 2 mixture. Various forms of carbon nanostructures were synthesis using different gas compositions. At the temperature and pressure reported here the equilibrium species of more importance are CO, CO 2 H 2 O, H 2 and CH 4 Using different gas composition syntheses of various forms of carbon like carbon nanotubes, nanofibers, onions, spheres, films and silica nanowires are demonstrated. Experiments conducted here support the role of C-H-O Gibbs phase triangle and suggest that the synthesis is possible by using injection of a mixture of various gases like CO, CO 2 H 2 O, 27

PAGE 41

28 CH 4 and H 2 Here experiments further elucidate that at low pressures the carbon deposition is not only related to the carbon activity but also depends on the partial pressures of O 2 and H 2 We deliberately change the carbon activity by changing the initial gas composition of the gases. Graphitic nanostructures have been synthesized using various conditions during the past decade. A mild oxidizing agent in CO 2 is used. CO 2 has been used in the past for the synthesis of carbon nanofibres. Carbon nanofibers and nanotubes were deposited on the silicon substrate with CO 2 as the carrier gas through ethanol [24]. The flow rate yielded nanofibers which had high aspect ratio, implying good field emission characteristics were shown but the ratio of CO 2 to ethanol was ambiguous. A ternary diagram is presented based on three radical species compositions showing the difference in deposition. Results for Composition 40CH4/20CO2/41H2/Ni A mixture of 40 sccm CH 4 20 sccm of CO 2 and 41sccm of H 2 were passed simultaneously in the CVD chamber. The point is plotted on the C-H-O system as shown in Figure 6-1. The SEM image of the carbon nanostructures deposited on the substrate is shown in Figure 6-2. Figure 6-3 shows the TEM image of nanotubes which were highly entwined, similar to the nanotubes produced catalytically using floating catalyst method. Many of the nanotubes have a darker center cavity then normal. The nanotube on the left corner stretched a long way, but typically most nanotubes started and ended in the same aggregate with sections coming out of the aggregate.

PAGE 42

29 T = 1040oC, Ptotal = 1 atmC O H PO2 H2OCO2COCH4C2H6CH2aC 10 1610 1510 210-14,10 510 310 410 110310250251051102 We are here now Figure 6-1 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 40sccm CH4, 20sccmCO2 and 41sccm H2.

PAGE 43

30 The diameter of the tube is ranging from 50 to 80 nm. Figure 6-4 shows us that the nanotubes grow via the tip-growth mode with the catalytic nanoparticle 'seeds' remaining anchored on the top of the nanotube during growth [40, 41]. Figure 6-5 shows a bunch of closed end multiwalled carbon nanotubes with well ordered graphitic layers with a lattice spacing of 0.34 between the layers. The activity of carbon at this point is 50 which is well above unity as can be seen in Figure 6-1. Figure 6-7 shows the Raman spectra of carbon nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica substrate show the G-line peak at 1596-9 cm -1 which is a characteristic of graphite sheets [21].The broad peak near 1337.9 cm -1 indicates the existence of defective graphitic layers on the wall surfaces, the so called D band [22] which is related to structural disorder of sp 2 bonded nanocrystalline and /or amorphous carbon species. This figure also shows the radial breathing mode for this composition. At first glance the spectrum may be interpreted as being noisy, but virtually all the features are reproduced in the spectra from CH 4 and H 2 mixture. X. Zhao et al[23]. show the various radial breathing modes of the MWCNTs .The peaks at 202.62 and the peak at 279 can be attributed for all of the Raman spectra to MWCNTs. The second Raman machine used a 632.8-nm helium-neon laser. Two second order peaks are observed at 2650 cm -1 and 2890 cm -1 which is a feature for graphite. Pure graphite also has a peak at 2650 cm -1 The gap that can be seen at about 515 cm -1 in the Raman is due to the substrate which was silicon with naturally grown oxide layer. At this concentration of species most of the nanostructures are carbon nanotubes with some exceptions being nanofibers and nanoparticles. As can be seen in the Figure 6-3 and 6-4 Ni particles were found in the tube. The likelihood of finding Ni

PAGE 44

31 particles in the tube was in direct proportion to tube diameter and also the presence of CO 2 Some of the tubes had some graphitic fringes running inside the tube. Corrosion of the tubes occurs due to the introduction of CO 2 [42] Usually CO 2 attacks the cap of the nanotubes exposing the internal cavity which when exposed, would allow foreign particles enter the inner cavities or inter-planar spacing of nanotubes. The graphitic layers of carbon nanotubes are susceptible to the attack of oxidation media at the regions of greater curvature. CNTs prepared by CVD method always have non perfect graphitic layers. The CO 2 present selectively attach the tube walls during the growth process. This attach can be seen very clearly in Figure 6-6 where the graphitic layers have been stripped off. In conclusion at this concentration of species Ni and graphitic filled nanotubes have been synthesized. The experimental products contained mostly carbon nanotubes with some exceptions and most of them being filled with Ni and graphitic particles. Figure 6-2 SEM micrograph showing carbon nanostructures at 40CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.

PAGE 45

32 Figure 6-3 TEM image of nanotubes, which were highly entwined. 40CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure. Figure 6-4 TEM micrograph showing a nanotube with some particles inside the tube due to the presence of CO 2. 40CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.

PAGE 46

33 Figure 6-5 HRTEM showing MWCNTs with graphitic fringes and a lattice spacing of 0.34nm. 40 CH4/20 CO2/41 H2/Ni composition at 1040 C and 1 atmosphere pressure. Figure 6-6 HRTEM showing outer portion of MWCNTs with graphitic fringes and a lattice spacing of 0.33nm. 40 CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.

PAGE 47

34 12532552572592511251325152517251925Raman shift (1/cm)Intensity (a.u.)1609 G-band (for "graphite")1344 D-band (for "disorder") Figure 6-7 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 40 CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure. 0204060801001201401602200230024002500260027002800290030003100Raman Shift (1/cm)Intensity (a.u.) Figure 6-8 Raman spectra of as prepared sample showing second order peaks (632.8-nm helium-neon laser). 40 CH 4 /20 CO 2 /41 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 48

35 T = 1040oC, Ptotal = 1 atmC O H PO2 H2OCO2COCH4C2H6CH2aC 10 1610 1510 210-14,10 510 310 410 110310250251051102 We are here now Figure 6-9 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 15sccm CH 4 50sccmCO 2 and 15sccm H 2

PAGE 49

36 Results for Composition 15 CH 4 /50 CO 2 /15 H 2 /Ni A mixture of 15 sccm of methane, 50 sccm of CO 2 and 15 sccm of H 2 were passed simultaneously in the CVD chamber as shown in Figure 6-9. The SEM image of the carbon nanostructures are shown in Figure 6-10 and 6-11.The nanostructures were deposited almost on the entire substrate. Figure 6-12 to 6-15 shows the HRTEM image of the different nanostructures deposited. Figure 6-12 shows the HRTEM image of a typical closed cap MWCNT, and the graphene sheets in the wall are aligned parallel to the tube axis. The outer and inner diameter of this tube is 12.5 and 2.5 respectively. The Figure also shows well ordered graphitic layers with a gap of 0.33-0.34nm between the layers. Figure 6-11 shows a TEM micrograph of carbon nano-onion synthesized during the process .The onion was poorly crystallized. Figure 6-14 and 6-15 show the same region with Figure 6-14 being at a lower resolution. The Figure 6-15 shows graphitic particles or sheets of graphite with a clear spacing of 0.33-0.34nm between the layers. The lattice spacing was observed in 3 regions where the lattice was clear. Here it can be observed that the activity of carbon is lower than unity and synthesis of not only carbon nanotubes but also onions and graphitic sheets or particles has taken place. The activity of carbon at this point is 10 -2 as seen in Figure 6-9. Figure 6-16 shows the Raman spectra of carbon nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica substrate show the G-line peak at 1596cm -1 which is a characteristic of graphite sheets. [21].The broad peak near 1337cm -1 indicates the existence of defective graphitic layers on the wall surfaces, the so called D band [22] which is related to structural disorder of sp 2 bonded nanocrystalline and /or amorphous carbon species. This figure also shows the radial breathing mode for this composition. At first glance the spectrum may be

PAGE 50

37 interpreted as being noisy, but virtually all the features are reproduced in the spectra from CH 4 and H 2 mixture. X. Zhao et al[23]. show the various radial breathing modes of the MWCNTs. The peaks at 202 and the peak at 279 can be attributed for all of the Raman spectra to MWCNTs. The second order peak at 2650cm -1 has very small intensity as compared to the CH 4 and H 2 mixture. In conclusion, at this composition the degree of graphitization is less because of decreasing activity of carbon. The Raman shows that the carbon deposition at this activity is almost the same but different forms of carbon have been formed in this region, majority being graphitic sheets and fibers with graphitic fringes without a cavity that is typical feature of the nanotube. Very few nanotubes were observed for this composition. The high flow rate of CO 2 restricts the growth of nanotubes and the tubes are no longer aligned. As seen in Figure 6-12, there could also be a possibility that during the synthesis of carbon nanostructures, if the reaction is allowed to take place for a long time, methane pyrolysis at 1040C leads to an appreciable thickness of carbon coating over a nanotube core [43, 44]. The important part of this particular composition is that the activity of carbon is two orders of magnitude less than one and deposition of different forms of carbon can be observed. It would be interesting to see what happens at this particular point in the phase diagram with the use of other gas species in the phase diagram.

PAGE 51

38 Figure 6-10 SEM micrograph showing carbon nanostructures at 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-11 SEM micrograph showing carbon nanostructures at 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 52

39 Figure 6-12 HRTEM image of a closed cap MWCNT. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-13 HRTEM image of a carbon nano-onion. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 53

40 Figure 6-14 TEM image of a graphitic sheet. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-15 HRTEM image showing graphitic fringes for the figure 6-12. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 54

41 12532552572592511251325152517251925Raman Shift (1/cm) 1339D-band (for "disorder")1592G-band (for "graphite") ) a.u ( it y ns te In Figure 6-16 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure 1001021041061081101121141161182100220023002400250026002700280029003000Raman shift(1/cm)Intensity (a.u.) Figure 6-17 Raman spectra of as prepared sample showing second order peak (632.8-nm helium-neon laser). 15CH 4 /50CO 2 /15H 2 /Ni composition at 1040 C and 1 atmosphere pressure.

PAGE 55

42 T = 1040oC, Ptotal = 1 atmC O H PO2 H2OCO2COCH4C2H6CH2aC 10 1610 1510 210-14,10 510 310 410 110310250251051102 We are here now Figure 6-18 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 30sccm CH 4 30sccm CO 2 and 30sccm H 2

PAGE 56

43 Results for Composition 30 CH 4 /30 CO 2 /30H 2 A mixture of 30 sccm of methane, 30 sccm of CO 2 and 30 sccm of H 2 were passed simultaneously in the CVD chamber. The point is highlighted in Figure 6-18. SEM image of the carbon nanostructures are shown in Figure 6-19 and 6-20.The nanostructures were deposited more on the edges of the substrate. Figure 6-21 to 6-24 shows the HRTEM image of the different nanostructures deposited. Figure 6-21 shows a defective nanotube with the graphitic walls uneven and some of the fringes entering the capsule of the tube as can be seen in Figure 6-22. Figure 6-23 shows a nanostructure which is close to a carbon nanofiber with some fringes but all the fringes are discontinuous and are not present in most of the regions. Figure 6-24 shows a top part of a closed end carbon nanotube with lattice spacing of 0.34nm. Figure 6-25 shows the Raman spectra of carbon nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica substrate show the G-line peak at 1596 cm -1 which is a characteristic of graphite sheets. [21] .The broad peak near 1337cm -1 indicates the existence of defective graphitic layers on the wall surfaces, the so called D band [22] which are related to structural disorder of sp 2 bonded nanocrystalline and /or amorphous carbon species. This figure also shows the radial breathing mode for this composition. At first glance the spectrum may be interpreted as being noisy, but virtually all the features are reproduced in the spectra from CH 4 and H 2 mixture. X. Zhao et al[23]. show the various radial breathing modes of the MWCNTs. The peaks at 202and the peak at 279 can be attributed for all of the Raman spectra to MWCNTs. This composition is very interesting as many experiments have been done in the past using CO as the feed gas for the synthesis of single walled carbon nanoubes at

PAGE 57

44 approximately the same temperature, pressure as the one reported here [8, 45, 46]. The activity of carbon at this point is exactly unity for both experiments but different gases have been used. It is very interesting to note that past experiments yielded single walled carbon nanotubes by using CO gas alone at a temperature very close to the one used here at (1040C). The reaction that takes place when CO is used is 2CO(g)C(s)+O2(g)[45]. At this particular composition and temperature, it is most likely that a mixture of H 2 and CO will be produced. Bronikowski et al[46] synthesized large quantity ropes of SWCNTs at 1050 C and different pressures (1-10atmosphere). The activity for both these experiments the ones from Smalleys lab[8, 45, 46] and the ones described here have the same activity of carbon but different partial pressures of oxygen and hydrogen. It can be seen that CO is placed on the C-O line from Figure 5.6. From the above conclusions can be made that the synthesis of carbon nanotubes not only depends on the composition of the various gases but also which gas mixtures are fed into the reactor and the partial pressures of oxygen and hydrogen. It can be seen that even though this particular composition corresponds to the activity one line there is no sign of single walled carbon nanotubes present and neither were the nanotubes produced in large quantities as done by R.E Smalley et al.[8, 45, 46].

PAGE 58

45 Figure 6-19 SEM micrograph showing carbon nanostructures at 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-20 SEM micrograph showing carbon nanostructures at 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 59

46 Figure 6-21 HRTEM image of a defective nanotube with an uneven cavity. 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-22 HRTEM image of the walls of the nanotube shown above. 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 60

47 Figure 6-23 HRTEM showing the synthesis of carbon nanofiber along with the nanotubes30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-24 HRTEM image of a closed cap MWCNT. 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 61

48 ns 12532552572592511251325152517251925Raman shift (1/cm)Inteity1334.9 D-band (for "disorder")1601.4 G-band (for "graphite") Figure 6-25 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 30 CH 4 /30 CO 2 /30 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 62

49 T = 1040oC, Ptotal = 1 atmC O H PO2 H2OCO2COCH4C2H6CH2aC 10 1610 1510 210-14,10 510 310 410 110310250251051102 We are here now Figure 6-26 The C-H-O ternary phase diagram, which shows the different gas compositions and activities of carbon. The one highlighted corresponds to 45sccm CH 4 and 50sccm H 2

PAGE 63

50 Results for Composition 45CH 4 /50 H 2 /Ni A mixture of 45 sccm of methane, 50 sccm of H 2 were passed simultaneously in the CVD. The point can be clearly seen on the C-H side of the C-H-O diagram Figure 6-26SEM images of the carbon nanostructures are shown in Figure 6-27. The HRTEM shows us that most of the nanostructures on this substrate are nanotubes. Figure 6-28 shows a typical nanotube with an inner diameter of about 5 nanometers and outer diameter of about 28nm. All the carbon nanotubes have no encapsulated catalytic particles at the closed tip. Figure 6-30 shows a balloon shaped carbon nanotube, it is not typical to grow balloon shaped nanotubes at this composition. Figure 6-29 shows a bunch of carbon nanotubes with well ordered graphitic layers with a gap of 0.34 between the layers. The activity of carbon at this point is 10 2 as seen in Figure 6-26. Figure 6-31 shows the Raman spectra of carbon nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica substrate show the G-line peak at 1596-9 cm -1 which is a characteristic of graphite sheets [21].The broad peak near 1337cm -1 indicates the existence of defective graphitic layers on the wall surfaces, the so called D band [22] which is related to structural disorder of sp 2 bonded nanocrystalline and /or amorphous carbon species. This figure also shows the radial breathing mode for this composition. X. Zhao et al[23]. show the various radial breathing modes of the MWCNTs. The peaks at 202 and the peak at 279 can be attributed for all of the Raman spectra to MWCNTs .In conclusion most of the nanostructures are carbon nanotubes with very few exceptions being nanofibers. The nanotubes had highly oriented graphitic walls and only MWCNTs are observed. It is observed that many nanotubes are covered with amorphous carbon as seen in Figure 6-29, there could also be a possibility that during the synthesis of carbon

PAGE 64

51 nanostructures, if the reaction is allowed to take place for a long time, methane pyrolysis at 1040C leads to an appreciable thickness of carbon coating over a nanotube core [43,44]. Figure 6-27 SEM image of carbon nanostructures. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-28 HRTEM showing MWCNTs with graphitic fringes. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 65

52 Figure 6-29 HRTEM showing a bunch MWCNTs with graphitic fringes and a lattice spacing of 0.34nm. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure Figure 6-30 HRTEM showing a carbon nanotube in the shape of a balloon. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure.

PAGE 66

53 12532552572592511251325152517251925Raman Shift (1/cm)1340D-band (for "disorder")1601G-band (for "graphite") .) .u (a ity ns te In Figure 6-31 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and D band. 45 CH 4 /50 H 2 /Ni composition at 1040 C and 1 atmosphere pressure

PAGE 67

54 T = 1040oC, Ptotal = 1 atmC O H PO2 H2OCO2COCH4C2H6CH2aC 10 1610 1510 210-14,10 510 310 410 110310250251051102 We are here now Figure 6-32 The C-H-O ternary phase diagram which shows the different gas compositions and activities of carbon. The one highlighted corresponds 50sccm CO 2 50sccm CO 2

PAGE 68

55 Results for Composition 50 sccm CO 2 Fifty standard cubic centimeter of CO 2 and 1000sccm Ar was passed simultaneously in the CVD chamber. The point can be seen clearly on the C-O side of the C-H-O ternary phase diagram. The SEM image of a large quantity of SiO X nanowires are shown in Figure 6-33 and 6-34. The SiO X nanowires are several micrometers in length and the diameter of the SiO X nanowires is about 15-50 nm approximately. TEM was employed to study the structure of the nanowires in detail. Figure 6-35 shows the high resolution TEM image of SiO X nanowire. The TEM reveals that the nanowires are in perfect amorphous state. Figure 6-36 shows that the catalyst particle is attached to the tip of the nanowires showing tip growth of the nanowires. Figure 6-37 shows that apart from the SiO X nanowires some graphitic features could also be seen. The lattice spacing between the layers is 0.34 which exactly matches with graphite but these features were very scarce. Raman spectra were observed using an excitation source with a 632.8-nm helium-neon laser operating with nominal output energy of 15 mW. The G and D line peaks which are a characteristic of graphitic peaks and carbonaceous particles have disappeared. The Raman spectra on the bulk of the SiOx nanowires Figure 6-39 shows an asymmetrical sharp peak at 508 cm -1 with line width of 10 cm -1. Previously, Yu et al. [47] reported that the Raman spectrum of amorphous SiO 2 nanowires is the same as that of bulk non-crystalline SiO 2 solids; whereas Wang et al. thought that there is no scattering peak in the stoichiometric SiO 2 nanowires [48]. It has been reported that there is a peak at 480 cm -1 for bulk amorphous silicon [49] and a peak at 520 cm -1 for crystalline silicon. The above investigation verified that there is no trace of crystalline phase in the silicon oxide nanowires here. The activity of carbon at this point is 10 -5 Figure 6-40 shows the EDS analysis showing a clear Silicon, oxygen and nickel peak. In conclusion synthesis of

PAGE 69

56 bulk quantities of SiOx nanowires using CO 2 and Ar is observed; apart from the nanowires some graphitic fringes could also be seen. Figure 6-33 SEM image of nanostructures. 50 sccm CO 2 Figure 6-34 SEM image of nanostructures. 50 sccm CO 2

PAGE 70

57 Figure 6-35 HRTEM image showing amorphous Sio X 50 sccm CO 2 Figure 636 HRTEM image showing amorphous SiO X nanowire with catalyst particle attached at the tip. 50 sccm CO 2

PAGE 71

58 Figure 6-37 HRTEM image showing some graphitic layers with a lattice spacing of 0.34nm which were also present. 50 sccm CO 2 01020304050607010001100120013001400150016001700180019002000Raman Shift (cm-1)Intensity (a.u.) Figure 6-38 Raman spectra of as prepared sample (632.8-nm helium-neon laser). 50 sccm CO 2

PAGE 72

59 020040060080010001200400450500550600650700750800Raman Shift (cm-1)Intensity (au) Figure 6-39 Raman spectra of as prepared sample showing the silica peak (632.8-nm helium-neon laser). 50 sccm CO 2 Figure 6-40 EDS Spectra showing Si, O and Ni peaks. 50 sccm CO 2

PAGE 73

60 Experiments Using Iron Nanoparticles as Catalyst Various compositions have been used for the growth of carbon nanostructures using iron nanoparticles as catalyst. The various compositions are given in the table below in table 2.1. Only SEM has been used to characterize the various nanostructures and thus a more detailed study is required to confirm the type of nanostructures. The SEM images for the various compositions have been shown below. Figure 641 80 CH 4 /17 CO 2 SEM image showing nanostructures Figure 642 50 CH 4 /50 CO 2 SEM image showing nanostructures

PAGE 74

61 Figure 643 15 CH 4 /50 CO 2 SEM image showing nanostructures Figure 644 50 CO 2 SEM image showing nanostructures

PAGE 75

62 Figure 645 EDS at 50 CO 2

PAGE 76

CHAPTER 7 CONCLUSIONS In all the Raman spectra except the one for CO 2 the degree of graphitization by the ID/IG ratio is almost the same and in agreement with the CVD growth of carbon nanostructures. Many forms of carbon nanostructures were synthesis using different gas compositions. At the temperature and pressure reported here the equilibrium species of more importance are CO, CO 2 H 2 O, H 2 and CH 4 Using various gas composition syntheses of various forms of carbon like carbon nanotubes, nanofibers, onions, spheres, films and silica nanowires are reported. These experiments support the role of C-H-O Gibbs phase triangle and suggesting that the synthesis is possible by using injection of a mixture of various gases like CO, CO 2 H 2 O, CH 4 and H 2 Here experiments further elucidate that the conditions for termination of carbon deposition is anywhere between 10 -2 and 10 -3 activity which is well below the deposition condition for carbon. A proper understanding as to why this is happening has to be studied. It can also be seen that the carbon deposition is not only related to the carbon activity but also depends on the partial pressures of O 2 and H 2 A detailed study using other gas species in the C-H-O system has to be done in order to get a particular phase where high purity of carbon nanostructures of a particular choice can be obtained. A better understanding for the mechanism of nanostructure formation under C-H-O conditions should make it possible to optimize the conditions for the growth of large quantities of nanotubes. The eventual goal of this work is to explain the synthesis of carbon nanotubes and to provide an analysis of gas 63

PAGE 77

64 compositions on different stages of the synthesis equilibrium thermodynamics, since in-situ CVD experiments are extremely difficult to analyze.

PAGE 78

LIST OF REFERENCES 1. P K Bachmann, D. Leers and H. Lydtin, Diamond and Related Materials 1 (1991) 1. 2. N. A. Prijaya, J. C .Angus and P. K. Bachmann, Diamond and Related Materials 39 (1993) 129. 3. Y. Z Wan, D.W. Zhang, Z.J. Lai, J. T. Wang Appl. Phys. A 67. (1998) 225. 4. S. Iijima, Nature 354 (1991) 56. 5. T.V. Hughes, C. R. Chambers, US Patent 405 480 (1889). 6. W.R. Davis, R.J. Slawson and G.R. Rigby, Nature 171 (1953) 756. 7. R.T.K. Baker and P.S Harris, Chem. Phys. Carbon 14 (1978) 83. 8. P. Nikolaev, M. J. Bronikowski, R. K Bradley, F. Rohmund, D. T. Colbert, K. A. Smith and R. E. Smalley, Materials Research Bulletin 1-2 (1999) 91. 9. C. Journet and P Bernier Appl, Phys. Chem. 96 (1992) 6941. 10. H.W Kroto, J. R. Heath, S. C. O. Brien, R. F. Curl, R.E. Smalley, Nature 318 (1985) 1629. 11. Ch. Emmenegger, P. Mauron, A. Zuttel, Ch. Nutzenadel, A. Schneuwly, R Gallay and L. Schlapbach, Appl. Surf. Science 452 (2000) 162. 12. C.M. Lieber, Solid State Commun, 107 (1998) 607. 13. Basic Properties of Carbon Nanotubes, http://www.applied-nanotech.com/cntproperties.htm (accessed June 2003) 14. IBM Research, http;// www.research.ibm.com/nanoscience/fet.html (accessed June 2003) 15. V. Derycke, R. Martel, J. Appenzeller and P.H. Avouris,Carbon Nanotube Interand Intramolecular Logic Gates. IBM Research Division, New York. 16. P. M. Ajayan and O. Z Zhoum, Applications of Carbon Nanotubes. Department of Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, NY. 65

PAGE 79

66 17. P. E Nolan, D. C. Lynch and Andrew Hall Cutler, J. Phys. Chem. B, 102 (1998) 4165 18. J. A. Libera, Y. Gogotsi, G. Yazicioglu and C. Megaridis, Appl. Phys. Lett. 79 (2001) 1021. 19. H. Lee, Y-S. Kang, P. S. Lee and J.-Y. Lee, Journal of alloys and compounds 330-332 (2002) 569. 20. A. Cao, X, Zhang, C. Xu, J. Liang, and D. Wu, Journal Mater. Res.16 No 11 (2001). 21. P.C. Eklund, J.M. Holden and RA Jishi, Carbon 33 (1995) 959. 22. R. T. K. Baker, Carbon 27, (1989) 315. 23. X. Zhao, Y. Ando, L.-C. Qin, H.-Kataura, Y. Maniwa, R. Saito, Physica B 323 (2002) 265. 24. C.-F. Chen, C.-L. Lin and C. Ming Wang and N. Jiang, App Phy Lett 82, 15 (2003) 2515. 25. J. Melancon and C. W. Bale, Oxid. Met. 18 (1982) 147. 26. C. W. Bale, J. Melancon and A. Pinho, "High Temperature Gaseous Equilibria-Computation and Graphical Representation," Can. Met. Quat. 19 (1981), 363. 27. HSC Chemistry for Windows (ver. 4.1): Chemical Reaction and Equilibrium Software with Extensive Thermochemical Database, Outokumpu Research, Oy, Finland. 28. O. Kubaschewski, C. B. Alcock and P. J. Spencer. Materials Thermochemistry, 6 th ed., Pergamon Press, New York 1993. 29. R. L. Burden and J. D. Faires, Numerical Analysis, 4 th ed., PWS-Kent, Boston, 1989. 30. Chapter 4 of Thesis N.Kulkarni, part of Ph.D. proposal, University of Florida, 2002, Now at ORNL. 31. Facility for the Analysis of Chemical Thermodynamics (F*A*C*T), Ecole Polytechnique, Montreal, Quebec, Canada. 32. NIST-JANAF Thermochemical Tables, 4 th ed., edited by M. W. Chase, Jr., pub. Am. Chem. Soc. & Am. Inst. Phys.Washington, D.C., 1998. 33. I. Barin, Thermochemical Data of Pure Substances, V.C.H., Weinheim, Germany, 1989.

PAGE 80

67 34. C. W. Bale, A. D. Pelton and W. T. Thompson, Can. Met. Quat. 25 (1986) 107. 35. R. T. DeHoff, Thermodynamics in Materials Science, McGraw-Hill, New York, N.Y. (1993). 36. C. Petit, A.Taleb and M. P. Pileni, J. Phys. Chem. B. 103 (1999) 1805. 37. S.-J. Park, S. Kim and S. Lee, J.Am.Chem. Soc 2000,122, (2000) 8581. 38. C Bower, O. Zhou, D.J. Werder, and S Jin, Appl. Phys. Lett. 77 (2000) 2767. 39. V.I. Merkulov, D.H. Lowndis,Y. Y Wei, G. Eres, and E. Voelkl, Appl. Phys. Lett 76, (2000) 3555. 40. S. Amelinckx, X. B. Zhang, D. Bernaerts, X. F. Zhang, V. Ivanov, J.B. Nagy, Science 265 (1994) 635. 41. G. G.Tibbetts, M. G. Devour, and E.J. Rodda, Carbon 25 (1987) 367. 42. S C Tsang, P.J .F Harris and M. L. H. Green, Nature 362 (1993) 520. 43. Tibbetts, G. G. in Carbon Fibers, Filaments and Composites 73-94 Kluwer Academic, Amsterdam (1990). 44. L. C. Qin and S. Iijima, Mater. Lett. 30 (1997) 311. 45. H. Dai, A. G. Rinzler, P. Nikolaev, A. Thess, D. T. Colbert and R. E. Smalley Chemical Physics Letters 260 Issues 3-4 (1996) 471. 46. M. J. Bronikowski, P. A Willis, D. T. Colbert, K. A. Smith and R.E. Smalley J. Vac. Sci. Technol. A 19 No 4 (2001) 1800. 47. D.P. Yu, Q.L. Houg, Y. Ding, H.Z. Zhang, Z.G. Bai, J.J. Wang, Y.H. Zou, W. Qian, G.C. Xiong and S.Q. Feng, Appl. Phys. Lett. 73 (1998) 3076. 48. N. Wang, Y.H. Tang, Y.F. Zhang, C.S. Lee, I. Bello and S.T. Lee, Chem. Phys. Lett. 299 (1999) 237. 49. A.T. Voutsaa, M.K. Hatilis and J. Boyce, J. Appl. Phys. 78 12 (1995) 6999.

PAGE 81

BIOGRAPHICAL SKETCH Archit Lal was born in India, Jabalpur. He came to Pune when he was one year old and has been brought up there. He finished his schooling from Loyola High School, Pune, one of the premier schools in India. Apart from being excellent throughout his academic career, Archit is a natural athlete and has represented the state of Maharashtra in badminton and table tennis; he has also represented India in field hockey. Even after scoring a good percentage in the 10 th standard examination, he continued his schooling in Loyola. Doing well in his 12 th standard Maharashtra board exam he got admission into one of the top universities and engineering colleges. His undergraduate degree was in Petrochemical Engineering from the University of Pune. He worked in the National Chemical Laboratory after graduation. The National Chemical Laboratory is one of the premier institutes in India. As he enjoyed his research in NCL he decided to go to graduate school. Getting admission and a research assistantship into University of Florida in the Materials Science and Engineering Department, which is currently ranked in the top 10 departments in the nation, made his decision easy. Ever since Archit has been at the University of Florida, Gainesville, and will be graduating with a Master of Science degree in August 2003. 68


Permanent Link: http://ufdc.ufl.edu/UFE0001151/00001

Material Information

Title: Effect of gas composition and carbon activity on the growth of carbon nanotubes
Physical Description: Mixed Material
Creator: Lal, Archit ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001151:00001

Permanent Link: http://ufdc.ufl.edu/UFE0001151/00001

Material Information

Title: Effect of gas composition and carbon activity on the growth of carbon nanotubes
Physical Description: Mixed Material
Creator: Lal, Archit ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001151:00001


This item has the following downloads:


Full Text












EFFECT OF GAS COMPOSITION AND CARBON ACTIVITY ON THE GROWTH
OF CARBON NANOTUBES















By

ARCHIT LAL


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Archit Lal

































This document is dedicated to my grandmother and grandfather Mr. and Mrs. Malhotra.















ACKNOWLEDGMENTS

This thesis is the fruit of the enduring support, love and cooperation of my mother

Mrs. Neera Lal, father Dr. Jagdish Lal and uncle Mr. Mandeep Singh Bains. Though,

being thousands of miles away they have been unshakable pillars of support, helping me

fight and win every battle. Their faith in my abilities has been my strength and their

visions my inspiration. All those words of wisdom have been my driving force through

all those hard times. No words said can express how indebted I am to them for what I am

today and for what I shall accomplish in the future.

I would also like to thank my sister, Ruchika, for being there for me and being my

motivation. I would also like to thank my fiance Indu who talking me through things

whenever I felt down and out.

Dr. Sigmund, my advisor, my teacher, my supervisor and my mentor taught me

more than I ever hoped to learn here at graduate school. He taught me how to think

individually and gave me all the independence to do my research in a very professional

manner. His work has been my inspiration. This work has been a product of his patience

and endurance. He has inspired me to be a better researcher and also a better person. He

understood my problems and has helped me to succeed in spite of them. My success is

and will be a reflection of his outstanding abilities as a teacher. I hope to do my best in

whatever task I undertake and always strive for perfection, as a tribute to him. Nothing

short of this will be adequate to express my gratitude to him.









I would like to thank Dr. Singh for all the support he has given me through the

entire project. His support and cooperation all through the project helped me immensely.

I am grateful to Dr. Butt for agreeing to be on my committee and giving me valuable

suggestions. I would also like to thank Nagraj Kulkarni for his guidance and helping me

with a lot of stuff that was not known to me.

I would also like to thank all the students in our research group "Dr. Sigmund's

Research group" for being the greatest group ever-helpful, supportive and understanding.

I would also like to thank the entire Materials Science and Engineering department for

making this possible.
















TABLE OF CONTENTS
page

A C K N O W L ED G M EN T S ............................... ......... ................................................... iv

LIST OF TABLES ....... .. ................................................... ......... viii

LIST OF FIGURES ......... ........................................... ............ ix

AB STRA CT ........... ........................................................... xii

CHAPTER

1 IN T R O D U C T IO N ............................................................................. .............. ...

2 GENERAL THEORIES AND FABRICATION METHODS FOR SYNTHESIS OF
CARBON NANOSTRUCTURES ....... .... ..................................... ..................... 3

H history ........................................................................... . 3
Fabrication M methods .......... ..... ....... ............ ............ .... ........ .. ..
A applications of carbon nanotubes ........................................ .......................... 5

3 WHY CHOOSE THE TERNARY PHASE DIAGRAM ............................................6

4 GAS EQULIBRIA ANALYSIS OF THE CARBON-HYDROGEN-OXYGEN(C-H-
O) SYSTEM USING PREDOMINANCE DIAGRAM .............. ................ 10

SOLG A SM IX (H SC Chem istry) ...................................................... ...............11
Constrained Chemical Potentials Method (CCPM) .........................................12
CCPM Software: Thermodynamic Basis .................................... ............... 14

5 RAW MATERIALS AND EXPERIMENTAL TECHNIQUES.............................17

C atalyst L oading and G generation .................................................... ..................... 17
D position of N i F ilm ........... ... ...... .... .. ...... ...... .................................. 1
Characterization of the Catalysts ................................................... .. ................. 19
Carbon Nanotube and Nanofiber Synthesis..................................... ............... 22
Characterization of the Carbon Nanotubes and Fibers.............................................24
Scanning Electron M icroscopy....................................... ......................... 24
Transmission Electron M icroscopy ................................. .. ............. 25
R am an Shift ............. .. ........... ........... ...... ... ........ ...... ...... 25









6 RESULTS AND DISCU SSION ........................................... .......................... 27

Results for Composition 40CH4/20CO2/41H2/Ni................................................28
Results for Composition 15 CH4/50 CO2/15 H2/Ni...................................................36
Results for Composition 30 CH4/30 CO2/30H2 ............... ...................................43
Results for Composition 45CH4/50 H2/Ni............................................................. 50
Results for Com position 50 seem CO2 ............................................ ............... 55
Experiments Using Iron Nanoparticles as Catalyst.......................................... 60

7 CON CLU SION S ........................................ .............. .. .............63

LIST OF REFEREN CES ............................................................................. 65

BIOGRAPH ICAL SKETCH ...................................................... 68
















LIST OF TABLES

Table pge

2-1 A applications of carbon nanotubes ........................................... .......................... 5

5-1. Raw materials used for the experiments................................................ ............... 17

5-2 The various gas compositions that were used using Nickel as catalyst. nX(O), nX(C),
nX(H) being the atom fractions for the various gases respectively .......................22

5-3 The various gas composition using iron nanoparticles as a catalyst.........................24















LIST OF FIGURES


Figure pge

2-1 Experimental sets up showing the CVD reactor.............. ...........................................5

3-1 The Carbon-H-O ternary phase diagram showing the different compositions and
activities of carbon. ................................................... ................ .. 7

3-2 Non-diamond growth region, the gas phase region. (App. Phys.A 67 225)...............9

4-1 C-H-O diagram with the partial pressures of 02 and activity of carbon. {Program
developed by Nagraj Kulkarni (University of Florida) now at ORNL} .................. 11

5-1 Set up for m making nanoparticles. ........................................................................... 18

5-2 Average particle size distribution of the iron nanoparticles using Zeta Plus .............19

5-3 (A)TEM image of iron nanoparticles at 200 X.well dispersed Iron Nanoparticles.(B)
TEM image of iron nanoparticles at 200X. Agglomerated iron nanoparticles can be
seen in the right hand corner. ............................................................................. 20

5-4 Growth of carbon nanotubes over annealed Ni catalyst film ....................................21

5-5 EDS ofNi film over silica substrate. The peaks are oxygen, nickel and silicon
respectively ............... ........... ................. .......... 21

5-6 The C-H-O ternary phase diagram showing the different compositions and activities
of carb on. ............................................................................. 23

6-1 The C-H-O ternary phase diagram, which shows the different gas compositions and
activities of carbon. The one highlighted corresponds to 40sccm CH4, 20sccmCO2
and 41sccm H2............... ........... .. ...... ...................... 29

6-2 SEM micrograph showing carbon nanostructures at 40CH4/20 C02/41 H2/Ni
composition at 10400 C and 1 atmosphere pressure........................................... 31

6-3 TEM image of nanotubes, which were highly entwined. 40CH4/20 C02/41 H2/Ni
composition at 10400 C and 1 atmosphere pressure........................................... 32

6-4 TEM micrograph showing a nanotube with some particles inside the tube due to the
presence of CO2. 40CH4/20 C02/41 H2/Ni composition at 10400 C and 1
atm sphere pressure. .................................................. ............. 32









6-5 HRTEM showing MWCNTs with graphitic fringes and a lattice spacing of 0.34nm.
40 CH4/20 C02/41 H2/Ni composition at 10400 C and 1 atmosphere pressure ....33

6-7 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and
D band. 40 CH4/20 C02/41 H2/Ni composition at 10400 C and 1 atmosphere
pressure..................................... ................................... ......... 34

6-8 Raman spectra of as prepared sample showing second order peaks (632.8-nm helium-
neon laser). 40 CH4/20 C02/41 H2/Ni composition at 10400 C and 1 atmosphere
pressure..................................... ................................... ......... 34

6-9 The C-H-O ternary phase diagram, which shows the different gas compositions and
activities of carbon. The one highlighted corresponds to 15sccm CH4, 50sccmCO2
an d 15 sccm H 2 ......... ................................................. ................................. 3 5

6-10 SEM micrograph showing carbon nanostructures at 15CH4/50C02/15H2/Ni
composition at 10400 C and 1 atmosphere pressure.............................................. 38

6-11 SEM micrograph showing carbon nanostructures at 15CH4/50C02/15H2/Ni
composition at 10400 C and 1 atmosphere pressure.............................................. 38

6-12 HRTEM image of a closed cap MWCNT. 15CH4/50C02/15H2/Ni composition at
10400 C and 1 atm osphere pressure .............................................. ............... 39

6-13 HRTEM image of a carbon nano-onion. 15CH4/50C02/15H2/Ni composition at
10400 C and 1 atm osphere pressure .............................................. ............... 39

6-18 The C-H-O ternary phase diagram, which shows the different gas compositions and
activities of carbon. The one highlighted corresponds to 30sccm CH4, 30sccmCO2
an d 3 0 sccm H 2 .................................................... ................ 4 2

6-23 HRTEM showing the synthesis of carbon nanofiber along with the nanotubes30
CH4/30 C02/30 H2/Ni composition at 10400 C and 1 atmosphere pressure ............47

6-25 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and
D band. 30 CH4/30 C02/30 H2/Ni composition at 10400 C and 1 atmosphere
pressure..................................... ................................... ......... 48

6-26 The C-H-O ternary phase diagram, which shows the different gas compositions and
activities of carbon. The one highlighted corresponds to 45sccm CH4, and 50sccm
H 2 ............. .................... ..................................... ......... ...... 4 9

6-27 SEM image of carbon nanostructures. 45 CH4/50 H2/Ni composition at 10400 C and
1 atm sphere pressure .................. .............................. ........ .. ........ .. 51

6-28 HRTEM showing MWCNTs with graphitic fringes. 45 CH4/50 H2/Ni composition
at 10400 C and 1 atm osphere pressure........................................... .....................51









6-29 HRTEM showing a bunch MWCNTs with graphitic fringes and a lattice spacing of
0.34nm. 45 CH4/50 H2/Ni composition at 10400 C and 1 atmosphere pressure ......52

6-30 HRTEM showing a carbon nanotube in the shape of a balloon. 45 CH4/50 H2/Ni
composition at 10400 C and 1 atmosphere pressure .............................................52

6-31 Raman spectra of as prepared sample (647nm of a krypton laser) showing the G and
D band. 45 CH4/50 H2/Ni composition at 10400 C and 1 atmosphere pressure.......53

6-32 The C-H-O ternary phase diagram which shows the different gas compositions and
activities of carbon. The one highlighted corresponds 50sccm CO250sccm C02....54

6-33 SEM image of nanostructures. 50 sccm CO2 .................. ................56

6-34 SEM image of nanostructures. 50 sccm CO2 .................. ................56

6-35 HRTEM image showing amorphous Siox. 50 sccm CO2............................... 57

6- 36 HRTEM image showing amorphous SiOx nanowire with catalyst particle attached
at the tip. 50 sccm CO 2 ................................................ .. .... .. ........ .... 57

6-37 HRTEM image showing some graphitic layers with a lattice spacing of 0.34nm
which were also present. 50 sccm CO2 ........................................ ............... 58

6-38 Raman spectra of as prepared sample (632.8-nm helium-neon laser). 50 sccm C0258

6-39 Raman spectra of as prepared sample showing the silica peak (632.8-nm helium-
n eon laser). 50 sccm C O 2 .............................................................. .....................59

6-40 EDS Spectra showing Si, 0 and Ni peaks. 50 sccm CO2 ..........................................59

6- 41 80 CH4/17 CO2 SEM image showing nanostructures.................... .....................60

6- 42 50 CH4/50 CO2 SEM image showing nanostructures ...................................... 60

6- 43 15 CH4/50 CO2 SEM image showing nanostructures .............. ................ 61

6- 44 50 CO2 SEM image showing nanostructures ................................. ............... 61

6- 45 ED S at 50 CO2 ................. ................. ............. ............... ......... 62















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

EFFECT OF GAS COMPOSITION AND CARBON ACTIVITY ON THE GROWTH
OF CARBON NANOTUBES

By

Archit Lal

August 2003

Chair: Dr Wolfgang Sigmund
Major Department: Materials Science and Engineering

This thesis tests the hypothesis that carbon nanostructures deposited, using a

mixture of various gases that contain carbon, hydrogen and oxygen are a part of the C-H-

O ternary phase diagram system. Synthesis of various forms of carbon by using different

gas compositions keeping the temperatures and pressure constant and changing the

carbon activity is reported. It is generally accepted that the activity of carbon greater than

unity is a necessary condition for the deposition of carbon. In contrast to this, it was

found that carbon deposition occurred even when the carbon activity was less than unity.

Furthermore, the ternary phase diagram provides general guidelines about suitable

conditions for growth of different carbon species. Graphitic carbon nanotubes,

nanofibers, onions and amorphous carbon nanowires were synthesized by using

equilibrated C-H-O gases, in the presence of a Ni catalyst, at 1040 oC in a CVD chamber.

The main objective of this thesis is to demonstrate the influence of initial gas

composition, activity and to an extent temperature and pressure on the growth of









nanotubes. TEM, Raman spectroscopy and SEM show that these carbon nanostructures

consist of graphite layers and some impurities.














CHAPTER 1
INTRODUCTION

This thesis tests the hypothesis that carbon nanostructures deposited, using a

mixture of various gases that contain carbon, hydrogen and oxygen are a part of the C-H-

O ternary phase diagram system, which was previously published for diamond [1]. It has

been found that the diamond synthesis in a CVD is restricted not only by the gas phase

composition, but also the temperature and pressure [2, 3].

While investigating the materials deposited on an arc-evaporation cathode,

Japanese microscopist Sumio lijima [4] discovered carbon nanotubes. In this thesis

project, study of the C-H-O ternary phase diagram and how this diagram will help us in

selecting the starting gas compositions for growing carbon nanotubes is demonstrated. It

also describes the influence of activity of carbon and its importance for the growth of

carbon nanotubes. Different characterization techniques like SEM, TEM and Raman were

used to characterize the carbon nanostructures.

Chapter 2 deals with an introduction to carbon nanotubes. A brief history,

discovery and fabrication methods are described. The applications of nanotubes for the

various engineering majors are mentioned in a table. A schematic of the CVD process

used in our experiments is also drawn.

Chapter 3 describes the motivation to explore the C-H-O ternary phase diagram and

the various hypotheses in the process of fabrication of nanostructures.

Many examples of various researchers have been illustrated showing the

dependence on the C-H-O phase diagram.









Chapter 4 deals with the predominance diagram. The compositions of various

gases in the C-H-O regime are plotted on the Gibbs triangle.

In chapter 5, the various characterizations on the raw materials taken are explained.

The methodology for the experiments and the equipment used for the various

experiments are also mentioned here. Plots of the C-H-O phase diagram with the various

experimental conditions are also plotted in this chapter.

Chapter 6 deals with the results and discussion of the various compositions used.

Chapter 7 gives a brief conclusion to the results of the experiments.














CHAPTER 2
GENERAL THEORIES AND FABRICATION METHODS FOR SYNTHESIS OF
CARBON NANOSTRUCTURES

History

The earliest report on vapor growth of carbon fibers is as early as 1889 patent [5]

describing the growth of hair like carbon filaments from methane decomposition in an

iron crucible. It was not until the invention of the electron microscope that Davis et al. [6]

almost 50 years ago observed minute vermicular growth (from 10nm to 200nm in

thickness) that penetrated the brickwork of blast furnaces. A plethora of studies using

filamentous carbon was later reviewed by Baker and Harris [7].

Carbon is a very unique element because it can assume various forms and

structures. Carbon nanotubes are byproducts in the fullerene synthesis and were

discovered by Sumio lijima in 1991[4]. The first to be discovered in 1991 by lijima were

the multi-walled carbon nanotubes (MWCNTs) made of a few to a few tens of concentric

cylinders placed around a common central hollow, with the interlayer spacing close to

that of graphite (0.34 nm).The other nanotubes are called as single walled carbon

nanotubes (SWCNTs), which are composed of only one graphite sheet and are in the

range of 1-2 nm [8].

Fabrication Methods

Various techniques are capable of synthesizing carbon nanotubes. The carbon arc

method [4], used initially for producing C60 fullerenes is the most common and perhaps

easiest way to produce carbon nanotubes. Chemical vapor deposition, [9] an apparatus









used for creating vapor grown carbon fibers, has also produced carbon nanotubes.

Finally, laser vaporization of a carbon block has produced the most uniform single wall

carbon nanotubes.

Laser ablation [10] has emerged as an alternative method to produced SWNT. In

this process; a target consisting of graphite mixed with small amount of transition metal

particles (as catalysts) is placed at the end of a quartz tube enclosed in a furnace. The

target is exposed to an argon ion laser beam, which vaporizes graphite and nucleates

carbon nanotubes in the short wave just in front of the furnace, carries the vapor and

nucleated nanotubes which continue to grow. The nanotubes get deposited on the cooler

walls of the tube downstream of the furnace. This method produces about 60-70%

SWNT's with the rest being catalyst particle and graphitic soot.

In this work the CVD method is used for growing carbon nanotubes. Previous

results demonstrate how such a system can be used to control nanotube diameter and

length [11]. CVD has been very effective for growth of pattern substrates. Thus CVD can

be practiced either as a continuous gas phase process with floating catalyst or a substrate

growth process with heterogeneous catalyst. Figure 2-1 shows us a schematic of the CVD

reactor used. It consisted of a quartz tube which is 1.5 inches in diameter and 48 inches

in length. The gases were passed through the Rota meters which measured the flow

before the gases enter into the chamber. A temperature controller was used to set the

cycles for heating and cooling.









FLOW
METERS






GAS
OUTLET


GAS INLET


CO, CHd H? QUARTZ BOAT I SAMPLE

OVEN TEMP 10400C

Figure 2-1 Experimental sets up showing the CVD reactor.

Applications of carbon nanotubes

The high elastic modulus of the graphene sheets makes carbon nanotubes one of

the toughest materials ever known [12]. The density of MWCNTs is 2.6 gm/cm3 as

compared to that of steel, which is 7.8 gm/cm3 [13]. They have application in a variety of

engineering majors. Some of the important applications are listed in the table below.

Table 2-1 Applications of carbon nanotubes
S.No Engineering majors Application
1) Electrical engineering Nanotube field effect transistors[14]
CNT Inter and intermolecular logic gates [15]
Vacuum microelectronics[16]
2) Biomedical engineering Nanoprobes and sensors[16]


3) Electrochemical engineering CNT-Lithium rechargeable batteries[16]

4) Mechanical engineering Air/space craft body parts

5) Construction engineering Lower density as compared to steel[13]

6) General engineering CNTs can store liquid and gas
Hydrogen storage














CHAPTER 3
WHY CHOOSE THE TERNARY PHASE DIAGRAM

During the last decade, CVD has been used to synthesize carbon nanotubes;

however, most of the deposition conditions have been involving gases like C2H2, CH4,

ferrocene and xylene mixtures which are on the C-H side of the ternary phase diagram.

Peter E. Nolan et al. was one of the first groups to describe a thermodynamic model

explaining how hydrogen in low concentration controls filament morphology and why

equilibrium is shifted from that for graphite during carbon deposition. Carbon deposition

reaction rates at low carbon activity in the absence of hydrogen are reported [17]. He also

stated that the thermodynamics of the carbon deposition system determines the size and

structure of the carbons formed, and thus the size and shape of the catalyst particle. [17].

J.A.Libera[18] et al have synthesized various forms of carbon from the C-H-O

system under hydrothermal conditions. Some other groups [19, 20] used CH4 and H2 in a

PECVD system to which they added small quantities of 02 were added (0-12%). They

also show that small 02 additives improved the purity of carbon nanotubes and

surprisingly led to the high quality even growth of carbon nanotubes [19] they also

showed that when the 02 content was increased to 8% the ID/ IG relative intensity ratio

increased drastically. The ID/IG ratio is the ratio of the intensities of the G and D band in

the Raman respectively. The G-line peak at 1596 cm-1 which is a characteristic of

graphite sheets [21].The broad peak near 1337 cm-1 indicates the existence of defective

graphitic layers on the wall surfaces, the so called "D band"[22] which is related to

structural disorder of sp2 bonded nanocrystalline and /or amorphous carbon species










T = 800 K, Ptotal = 1 atm.



0





2 C02
co







/ Z



H C
10 2 10
CH4 C2H6 CH2

a(C) carbon isoactivity lines
compositions of selected gaseous species
-- isoactivity line for carbon, a(C) = 1

Figure3-1 The Carbon-H-O ternary phase diagram showing the different compositions
and activities of carbon.

However, a common scheme connecting all the experimental methods, gas

composition and growth of nanotubes is still lacking. To develop a strategy to explore

carbon deposition is necessary to have some hypothesis. Some of the hypothesis that

were taken into consideration before the experiments were,

1. Which carbon carrier gases and gas mixtures are suitable for nanotube growth in
CVD?( A gas mixture containing 3 gases CH4, CO2 and H2 are used)

2. Is there a specific region in the ternary phase diagram where the nanotube growth
can be controlled and longer nanotubes can be produced?

3. Why are certain CVD methods producing carbon nanotubes at higher growth rates?









Checking the growth of carbon nanostructures using a mixture of gases and thereby

changing the carbon activity and the oxygen and hydrogen partial pressures are studied in

this work. We used 3 gases CH4, CO2 and H2 in our experiments. Thus by changing the

initial gas compositions of the 3 gases we found ourselves at different points in the gas

composition triangle.

In the past decade many gases and gas mixtures have been used to grow carbon

nanostructures in the CVD system. Out of them CH4, C2H2, CO and a mixture of CH4 and

H2, C2H2 and H2 have been used. In the floating catalyst method Andrews et al. reported

a continuous production of aligned MWCNTs in bulk quantities through the catalytic

decomposition of ferrocene-xylene mixture. CO2 has been used in the past for the

synthesis of carbon nanofibres. Carbon nanofibers and nanotubes were deposited on the

Si substrate with CO2 as the carrier gas through ethanol [23] but the gas ratio was

ambiguous. Previously nanotubes were grown grew on a substrate using methane as a

feeding gas, but less than 5 percent of the catalyst particles sprout nanotubes, "By

switching the feeding gas from methane to a carbon monoxide and hydrogen mixture Liu

at al significantly improved the amount of nanotubes grown on a surface. Such nanotubes

have a range of favorable physical properties.

Apart from all these examples the C-H-O Gibbs triangle has been used for the

synthesis of diamond and a region has been described by P.K. Bachmann et al. [1] and

the region has been confirmed by many other researchers. Figure 3-2 below shows us the

region of the diamond domain, the non diamond growth region and the gas phase region.

All these examples stated above demonstrate that the same is possible for the growth of

carbon nanostructures as well.














(d) P=100 kPa


Figure 3-2 Non-diamond growth region, the gas phase region. (App. Phys.A 67 225)


o_1 0-2 0/(0-+ H) --. 0o.9














CHAPTER 4
GAS EQULIBRIA ANALYSIS OF THE CARBON-HYDROGEN-OXYGEN(C-H-O)
SYSTEM USING PREDOMINANCE DIAGRAM

The predominance diagram described above does not provide direct information

regarding the detailed gas composition. This is because in addition to the gases that are

represented on the axes of the predominance diagram (e.g., 02 and H2), a number of other

gaseous species are also present. The mixture of all these gaseous species determines the

gas composition [24, 25], which is best specified by the atom fractions of the elemental

gases. For example, in the O-C-H system, a unique gas composition is given by the atom

fractions, Xc, Xo and XH. This composition plots as a point on a Gibbs triangle (Figure4-

1). For example, a gas mixture consisting of 20 % H2 and 80 % CO2 has a gas

composition given by:

(0.2)2
X (02)2 0.143
(0.2)2 + (0.8)(1+2)

(0.8)1
X = = 0.286
(0.2)2 + (0.8)(1+ 2)

(0.8)2
Xo = 0.571 (1)
(0.2)2+ (0.8)(1+ 2)

XH + x +Xo = 1
The partial pressures of the various gases that exist in this system can be

determined for this composition or for other compositions, from thermodynamics of the

gases under equilibrium conditions. Different combinations of gas mixtures may be

utilized to obtain the same gas composition (Figure 4-1). While the composition does not

vary with temperature or total pressure in the system (conservation law), the partial










pressures do vary. Knowledge of the variation in the partial pressures of the gases, such

as oxygen, with composition and temperature, is obviously very important for controlling

the results. This information can be obtained using two different thermodynamic

approaches for determination of gaseous equilibria (or partial pressures). These are

discussed below.

T = 1000 K, Ptotal = 1 atm.

0


1029 CH4 C2H6 CH2
1030
_. p(02) oxygen isobars
compositions of selected gaseous species
- isoactivity line for carbon, a(C) = 1


Figure 4-1 C-H-O diagram with the partial pressures of 02 and activity of carbon.
{Program developed by Nagraj Kulkarni (University of Florida) now at
ORNL}

SOLGASMIX (HSC Chemistry)

The commercially available HSC Chemistry software [26] contains several

modules for thermodynamic analysis. An important module is the SOLGASMIX code,

which is capable of analyzing gaseous equilibria. The input to the SOLGASMIX









calculation is the initial composition of the gas mixture, e.g., 10 % H2, 90 % CO2. Then

from thermodynamics of all the gaseous and condensed phase species that may exist in

that relevant system, the variation in the partial pressures of all gaseous species is

computed as a function of temperature. If precipitation of condensed phases is

thermodynamically favored, the calculation also provides the equilibrium compositions

and amounts of all species in that system. To accomplish this, a sophisticated equilibrium

solver, which is the critical component of the HSC software, is necessary. Usually rapid

convergence to the correct solution is desired from the equilibrium codes developed.

The results are given in a tabular form but more commonly in the form of a graph

that gives the partial pressure of a particular gaseous species as a function of temperature,

see Figure 4-1. However, the calculation has to be repeated for a different choice of the

initial gas composition. This is the main drawback of this technique, since one does not

know a priori, which gas composition or system will provide the desired partial pressure

at a given temperature. Nevertheless, with some experience and intuition, it is possible to

use the SOLGASMIX code in a more predictive capacity. The HSC Chemistry software

contains an exhaustive thermodynamic database, that is updated frequently, but is not

critically evaluated, which renders its usage for an uninitiated user rather difficult. For

gases, ideal behavior is usually assumed although it does permit non-ideal terms to be

added. It also allows the user to utilize his or her own database should the need arise. Its

user friendliness and robust equilibrium solver are its major advantages.

Constrained Chemical Potentials Method (CCPM)

In many practical problems of atmosphere control including the present, it is

desired to predict the possible atmospheres that can provide a fixed value of partial

pressures of gases at equilibrium. The approach is the exact opposite of that given above.









For e.g., if one wants to fix a particular set of values for pO2 and pH2 at equilibrium, what

initial mixtures of say H2 + H20 + or CO2 + H2, if any, will produce the desired

equilibrium? It is quite possible that a number of different gas mixtures may provide the

desired composition. Obviously the previous equilibrium approach can still be utilized

but it can prove to be quite cumbersome. Also the initial choice of the gas mixtures for

the equilibrium approach has to be provided by the user and is hence a function of the

experience and intuition of the user.

Having calculated the gas phase equilibria, how can one readily assess the

influence of composition, temperature or total pressure upon the gaseous equilibria?

What are the practical advantages of, say a H2 + H20 + mixture over a H2 +CO2 mixture?

Is a mixture of CO2 + H2 a poor or a good chemical buffer for pO2 and activity of carbon,

i.e., is the gas displaced from the desired equilibrium point by a small addition of

impurities? Will C deposition occur at lower temperatures? Although such questions

could be systematically answered with the aid of computerized tabulations available

through equilibrium solvers such as SOLGASMIX, the problem often is that the

influence of any one particular variable is not always clear since the properties are

interdependent.

The "constrained chemical potential method" developed by Bale et al. [24, 25], for

calculating gaseous equilibria allows one to compute the partial pressures of gaseous

species and display them on a Gibbs composition triangle. It is this graphical output of

gaseous phase equilibria on a Gibbs composition triangle which is the most efficient way

of answering many of the questions posed above (Figure 4-1). It was thus decided to

develop a code based on this technique and apply it to the C-O-H and the H-O-N gas









systems. The results obtained for these systems were indeed very useful in deciding

appropriate compositions of gas mixtures that can achieve the desired conditions during

the sintering of additive fuel pellets. The computational scheme for developing this

software in MathCad is discussed below.

CCPM Software: Thermodynamic Basis

The thermodynamic basis for the CCPM approach is discussed using the C-H-O

system as an example.

(1) Initially the thermodynamic data for all the gaseous species in this system was
obtained from the book by Kubaschewski and Spencer [27] inputted into an Excel
spreadsheet and imported into MathCad. The thermodynamic data included the
standard enthalpies and entropies of formation as well as the specific heat (Cp) data
for the various gaseous species. The reference elements selected were solid carbon,
oxygen and hydrogen gas.
(2) The gas phase reactions were balanced and the equilibrium constants were computed
at a specified temperature and pressure (1 atm.) A generic reaction for this system is

iC(s) + H2(g)+ O,(g) = CHO(g)
2 2
P(C Hj O, )
AG = -RTlnK; K = P(CH k (2)
a(C) .P(H )2 .P(02)2

where i, j and k are non-negative integers, AG is the standard free energy of formation

for the reaction at the temperature T (K) and K is the equilibrium constant. a(C) is the

activity of carbon, P(H2) and P(02) are the partial pressures of hydrogen and oxygen

respectively.

(3) The partial pressure (or activity) of any gaseous species is obtained from the above
equation.
K
P(CHOk) = a(C)'.P(H2)2 .P(02)2.K (3)









From the Gibbs phase rule, the three component gas phase system, has three

degrees of freedom (f = c-p+2 = 3-1+2 =3). If the temperature and the total pressure are

fixed, then the gas phase is defined, if any two of the three variables, a(C), P(H2) or P(O2)

are also defined. Hence, it is possible to reduce the system of non-linear equations

containing the equilibrium constant, into one non-linear equation in one unknown.

(4) In the present instance, choose P(H2) as the dependent variable. For a given value of

P(02), choose a value for a(C) and compute P(H2) using a non-linear equation solver

based on the "Half-Interval Technique" [28].

(5) Compute the partial pressures of all gaseous species and hence determine the

compositions (Xo, Xc and XH) for each value of a(C).

(6) Vary a(C) incrementally and repeat the calculation for the same oxygen partial

pressure hence obtaining the entire isobar.

(7) To obtain different oxygen isobars, repeat the calculation for a different value of

P(02).

(8) Plot the gas compositions for each isobar on a Gibbs Composition Triangle.

A similar procedure may be used to compute isoactivity lines for carbon or

isobars for hydrogen. In general, it took about 1000-3000 calculations for each isobar

and the time involved was about 10-12 minutes on a Pentium 266. Thus approx. 6-8

hrs of computer time was needed for an entire set of isobars for a given temperature

and pressure. However, a rapid survey for a portion of a specified isobar could be

done in 15-30 seconds in MathCad. A typical graphical output from the CCPM

software, that contains oxygen isobars on a Gibbs Composition Triangle, is shown in

Figure 4-1. The partial pressures for the oxygen isobars are indicated in the Figure.






16


Similar plots can be obtained at different temperatures and pressures for other gaseous

species in the system. [29,30,31,32,33,34].














CHAPTER 5
RAW MATERIALS AND EXPERIMENTAL TECHNIQUES

Table 5-1. Raw Materials used for the Experiments
S.No Material Company
1) (Fe(CO)5) 99.9% Aldrich

2) (TOPO) 99.9% Aldrich
3) CO2 gas Aproximately 99% Strate Welding Supply
4) CH4 gas (99.99%) Strate Welding Supply
5) H2 gas (99.99%) Strate Welding Supply
6) Ar gas (99.99%) Strate Welding Supply


Catalyst Loading and Generation

The experiments were performed using two types of catalysts, iron and nickel. An n

type (resistivity 1-10 Q cm, thickness from 254-305 [tm) silicon (100) wafer purchased

from Montco Silicon Technologies, Inc. with naturally grown oxide layer was used as a

substrate for the growth of carbon nanotubes. The spherical iron nanoparticles were

prepared by the thermal decomposition of organometallic precursor (Fe(CO)5) in the

presence of a stabilizing surfactant, which is the modified procedure of the previously

reported method (Figure5-1)[33,34]. A 0.2 mL portion of Fe(CO)5 was added to 5.0 g of

trioctylphosphine oxide (TOPO) at 3200C under argon atmosphere, and the resulting

solution was aged for 30 min at 3200C. The preparation temperature was set to over

3200C so that a complete thermal decomposition of Fe (CO)5 could be achieved to

generate metallic iron atoms. The reaction mixture was added into excess acetone to

produce black precipitate. The precipitate was retrieved by centrifuging and this process









was repeated several times. The synthesized powder form of spherical nanoparticles can

be easily re-dispersed in pyridine to yield a clear homogeneous solution. [35, 36] The

transmission electron micrograph (TEM) revealed that the iron nanoparticles are 20 nm in

diameter but most of them are agglomerated. Iron nanoparticles have a strong tendency to

agglomerate because of their magnetic characteristics and high specific surface energy as

can be seen from the TEM image. Particle sizing was performed using the Zeta plus,

which showed that the average size to the particles is about 11lOnm due to agglomeration.


Fe(C00)5
Ar I














Figure 5-1 Set up for making nanoparticles.

Deposition of Ni Film

The silicon substrate with the naturally grown oxide was cleaned in acetone,

isopropyl alcohol and deionized water bath prior to the Ni film deposition. The

deposition was done using an E beam evaporation system. The base pressure was 1.5* 10-

6 and current was 0.12A.The thickness of the film as measured by the crystal monitor was

380 A .It is seen that Ni breaks into small islands due to surface tension, as well as the

compressive stress due to the mismatch of the thermal expansion coefficients of Si and

Ni. [37,38].









Characterization of the Catalysts

The catalysts were characterized using TEM, SEM, particle sizer and EDS. Figure

5-2 showing the average size distribution of the iron nanoparticles. The average size is

118nm as shown in the figure. Figure 5-3a and b show the TEM images of the iron

catalyst particles. The size of the paricles is about 20nm but as iron as a strong tendency

to agglomerate, hence the average size is 118nm which is shown by the Zeta plus. Figure

5-4 shows the SEM image of the annealed Ni film. Figure 5-5 shows the EDS spectra of a

Ni coated Silica substrate. Figure 5-4 shows the growth of carbon nanostructures over an

annealed Ni film.


120

100

80

60

40

20

0


Average
Size=118.5nm


0 50


100 150 200
Diameter(nm)


250 300


Figure 5-2 Average particle size distribution of the iron nanoparticles using Zeta Plus.















































Figure 5-3 (A)TEM image of iron nanoparticles at 200 X.well dispersed Iron
Nanoparticles.(B) TEM image of iron nanoparticles at 200X. Agglomerated
iron nanoparticles can be seen in the right hand corer.


































Figure 5-4 Growth of carbon nanotubes over annealed Ni catalyst film


Figure 5-5 EDS ofNi film over silica substrate. The peaks are oxygen, nickel and silicon
respectively









Carbon Nanotube and Nanofiber Synthesis

A quartz tube that is 1.5 in. in diameter and 48 in. in length was used for the CVD

furnace. The furnace temperature was increased by 6 C/min up to the growth

temperature in argon atmosphere with a flowing speed of 1000 sccm. The sample was

placed in a quartz boat and placed carefully in the centre of the heating zone. On reaching

growth temperature, the various gas mixtures were introduced into the chamber

simultaneously, the annealing was done in argon (1000 sccm) before growth of the

CNTs. After growing the CNTs, the furnace was cooled to room temperature in argon

(1000 sccm).

Graphitic nanostructures have been synthesized using various conditions during the

past decade. In our experiments we used a mild oxidizing agent in CO2. The activity of

carbon was decrease using the mild oxidizing agent with higher flow rates. Figure 5-6

shows us a C-H-O ternary phase diagram with the different experimental composition

plotted at 1040 C and 1 atm pressure. These compositions are only for the Ni catalyst.

Table 5-2 The various gas compositions that were used using Nickel as catalyst. nX(O),
nX(C), nX(H) being the atom fractions for the various gases respectively.

Flow in sccm nX(O) nX(C) nX(H) nX(O)% nX(C)% nX(H)%


40CH4/20 C02/41H2 0.1169 0.1754 0.7077 11.69 17.54 70.77

15CH4/50 C02/15H2 0.39216 0.2549 0.35294 39.216 25.49 35.294

30CH4/30 C02/30H2 0.2 0.2 0.6 20 20 60

45CH4/50H2 0 0.1384 0.8616 0 13.84 86.16


0 66.67 33.33 0


50 CO2


0.6667 0.3333







T=10400C, Ptotal =1 atm
0


CH4 C2H6 CH2 40CH4/20CO2/41H2


45CH4/50H2


Figure 5-6 The C-H-O ternary phase diagram showing the different compositions and
activities of carbon.


carbon isoactivity lines
isoactivity line for carbon, a(C) = 1
compositions of selected gaseous species *
Experimental Compositions +


a(











Table 5-3 The various gas composition using iron nanoparticles as a catalyst.

Gas Mixtures Atom Atom Atom Atom Atom Atom
(sccm) fraction fraction fraction fraction% fraction% fraction%


nX(O) nXC nX(H) nX(O)% nXC% nX(H)%


80CH4/17 CO2 0.0755 0.215 0.7095 7.55 21.5 70.95

50CH4/50 CO2 0.25 0.25 0.5 25 25 50

15CH4/50 CO2 0.4444 0.2888 0.2668 44.44 28.88 26.68

50 CO2 0.6667 0.3333 0 66.67 33.33 0

Characterization of the Carbon Nanotubes and Fibers

Scanning Electron Microscopy

The SEM imaging was carried out using the JEOL JSM6330F and the SEM 6400.

The JEOL 6330 is a cold field emission scanning electron microscope. The cold field

emission has the advantage of high brightness (large current density) and small beam

diameter (high resolution) at low accelerating voltages to allow imaging of nanotubes

without causing damage to the sample. Resolution of the instrument is around 1.5nm

depending on the sample. The instrument has an energy dispersive X-ray spectrometer

(EDS) for elemental analysis. The JEOL JSM-6400 imaging modes include secondary

electrons, backscattered electrons and x-rays. Under ideal conditions, resolution in

secondary electron imaging mode is 5 nm. Images can be recorded with Polaroid film

and/or digital image capture through the XL system with subsequent output to a 1200 dpi

laser printer.









The sample was directly taken out of the furnace and attached onto the SEM

sample holder by means of a double sided carbon tape.

Transmission Electron Microscopy

Transmission electron microscopy images of the carbon nanostructures were

obtained using TEM 201OF microscope. The carbon nanostructures were scratched out

from the substrate and dispersed in ethanol. The solution containing the carbon

nanostructures were then ultrasonicated for 10 min before the Alicia copper grid was

dipped in the solution. The sample was dried for about 36 hrs before observation. The

freed nanostructures adhere well to the TEM grid and were easily imaged.

Raman Shift

Raman shift was performed using two machines.

1) The Raman shift was performed using a 647nm excitation using a krypton laser. Both

the radial breathing modes and the second order bands were observed. The carbon

nanostructures that were deposited on the silica substrate were directly observed sample

as it is.

2) The second machine was used to get the second order peaks .A confocal micro-Raman

spectrometer system (LabRam Infinity, Jobin Yvon) was used for all measurements. The

excitation source was a 632.8-nm helium-neon laser operating with nominal output

energy of 15 mW. The laser beam was focused on the sample using a 10x objective

resulting in a focused beam spot of 35 mm on the corneal sample surface. The laser

beam was attenuated with a neutral density filter prior to entering the microscope

objective to yield a laser beam power of about 3 mW on the sample surface. Raman

scattered light was collected in the backscatter mode through the microscope objective

and dispersed onto a 1024-pixel CCD detector array using a 600 grooves/mm grating.






26


The effective spectral dispersion was 0.12 nm/pixel, equal to about 2. cm-1/pixel over the

specified wavelength range. The Raman microscope was configured for confocal

microscopy. The confocal aperture was set to a diameter of 500 mm, which resulted in

an effective integration depth of 300 mm.














CHAPTER 6
RESULTS AND DISCUSSION

These experiments are designed to establish broad limits on the synthesis

conditions resulting in carbon tube growth changing the initial gas compositions and thus

activity and partial pressures of oxygen. The results are presented in terms of three

precursors namely C02, CH4 and H2. The synthesis of nanotubes, fibers, graphitic

particles, onions and non planar graphitic structures is shown. The activities for carbon

where these structures have grown are ranging from 100 to approximately 10-2 which is

several orders of magnitude less than one and the necessary condition for deposition of

carbon. 99% CO2 used in the experiments yields SiOx nanowires and very few carbon

structures can be noticed. The response of rates to the partial pressures of oxygen and

hydrogen is not studied in detail. Deposition rates have been plotted with respect to

activity by Nolan et al. [17], showing a generally linear response.

In all the Raman spectra except the one for CO2 the degree of graphitization by the

ID/IG ratio is almost the same and in agreement with the CVD growth of carbon

nanostructures using a CH4 and H2 mixture. Various forms of carbon nanostructures were

synthesis using different gas compositions. At the temperature and pressure reported here

the equilibrium species of more importance are CO, C02, H20, H2 and CH4. Using

different gas composition syntheses of various forms of carbon like carbon nanotubes,

nanofibers, onions, spheres, films and silica nanowires are demonstrated. Experiments

conducted here support the role of C-H-O Gibbs phase triangle and suggest that the

synthesis is possible by using injection of a mixture of various gases like CO, C02, H20,









CH4 and H2. Here experiments further elucidate that at low pressures the carbon

deposition is not only related to the carbon activity but also depends on the partial

pressures of 02 and H2. We deliberately change the carbon activity by changing the initial

gas composition of the gases.

Graphitic nanostructures have been synthesized using various conditions during the

past decade. A mild oxidizing agent in CO2 is used. CO2 has been used in the past for the

synthesis of carbon nanofibres. Carbon nanofibers and nanotubes were deposited on the

silicon substrate with CO2 as the carrier gas through ethanol [24]. The flow rate yielded

nanofibers which had high aspect ratio, implying good field emission characteristics

were shown but the ratio of CO2 to ethanol was ambiguous. A ternary diagram is

presented based on three radical species compositions showing the difference in

deposition.

Results for Composition 40CH4/20C02/41H2/Ni

A mixture of 40 sccm CH4, 20 sccm of CO2 and 41sccm of H2 were passed

simultaneously in the CVD chamber. The point is plotted on the C-H-O system as shown

in Figure 6-1. The SEM image of the carbon nanostructures deposited on the substrate is

shown in Figure 6-2. Figure 6-3 shows the TEM image of nanotubes which were highly

entwined, similar to the nanotubes produced catalytically using floating catalyst method.

Many of the nanotubes have a darker center cavity then normal. The nanotube on the left

corner stretched a long way, but typically most nanotubes started and ended in the same

aggregate with sections coming out of the aggregate.









T=10400C, Ptotal = 1 atm

0


CH4 C2H6 CH2


Figure 6-1 The C-H-O ternary phase diagram, which shows the different gas
compositions and activities of carbon. The one highlighted corresponds to
40sccm CH4, 20sccmCO2 and 41sccm H2.


carbon isoactivity lines
isoactivity line for carbon, a(C) = 1
compositions of selected gaseous species *
Experimental Compositions +









The diameter of the tube is ranging from 50 to 80 nm. Figure 6-4 shows us that the

nanotubes grow via the tip-growth mode with the catalytic nanoparticle 'seeds' remaining

anchored on the top of the nanotube during growth [40, 41]. Figure 6-5 shows a bunch of

closed end multiwalled carbon nanotubes with well ordered graphitic layers with a lattice

spacing of 0.34 between the layers. The activity of carbon at this point is 50 which is well

above unity as can be seen in Figure 6-1. Figure 6-7 shows the Raman spectra of carbon

nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm

of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica

substrate show the G-line peak at 1596-9 cm-1 which is a characteristic of graphite sheets

[21].The broad peak near 1337.9 cm-1 indicates the existence of defective graphitic layers

on the wall surfaces, the so called "D band" [22] which is related to structural disorder of

sp2 bonded nanocrystalline and /or amorphous carbon species. This figure also shows the

radial breathing mode for this composition. At first glance the spectrum may be

interpreted as being noisy, but virtually all the features are reproduced in the spectra from

CH4 and H2 mixture. X. Zhao et al[23]. show the various radial breathing modes of the

MWCNTs .The peaks at 202.62 and the peak at 279 can be attributed for all of the

Raman spectra to MWCNTs. The second Raman machine used a 632.8-nm helium-neon

laser. Two second order peaks are observed at 2650 cm-1 and 2890 cm-1 which is a

feature for graphite. Pure graphite also has a peak at 2650 cm-1. The gap that can be seen

at about 515 cm-1 in the Raman is due to the substrate which was silicon with naturally

grown oxide layer. At this concentration of species most of the nanostructures are carbon

nanotubes with some exceptions being nanofibers and nanoparticles. As can be seen in

the Figure 6-3 and 6-4 Ni particles were found in the tube. The likelihood of finding Ni









particles in the tube was in direct proportion to tube diameter and also the presence of

CO2. Some of the tubes had some graphitic fringes running inside the tube. Corrosion of

the tubes occurs due to the introduction of CO2 [42] Usually CO2 attacks the cap of the

nanotubes exposing the internal cavity which when exposed, would allow foreign

particles enter the inner cavities or inter-planar spacing of nanotubes. The graphitic layers

of carbon nanotubes are susceptible to the attack of oxidation media at the regions of

greater curvature. CNTs prepared by CVD method always have non perfect graphitic

layers. The CO2 present selectively attach the tube walls during the growth process. This

attach can be seen very clearly in Figure 6-6 where the graphitic layers have been

stripped off. In conclusion at this concentration of species Ni and graphitic filled

nanotubes have been synthesized. The experimental products contained mostly carbon

nanotubes with some exceptions and most of them being filled with Ni and graphitic

particles.





















Figure 6-2 SEM micrograph showing carbon nanostructures at 40CH4/20 CO2/41 H2/Ni
composition at 10400 C and 1 atmosphere pressure.
A W, -









comp n at 1d ad 1.
We- 011L A,
WMP 11... -4 IO









composition at 1040' C and I atmosphere pressure.


































Figure 6-3 TEM image of nanotubes, which were highly entwined. 40CH4/20 C02/41
H2/Ni composition at 10400 C and 1 atmosphere pressure.


lii,~


Figure 6-4 TEM micrograph showing a nanotube with some particles inside the tube due
to the presence of CO2. 40CH4/20 C02/41 H2/Ni composition at 10400 C and
1 atmosphere pressure.






















.-L-










10 rim


Figure 6-5 HRTEM showing MWCNTs with graphitic fringes and a lattice spacing of
0.34nm. 40 CH4/20 C02/41 H2/Ni composition at 10400 C and 1 atmosphere
pressure.


Figure 6-6 HRTEM showing outer portion of MWCNTs with graphitic fringes and a
lattice spacing of 0.33nm. 40 CH4/20 CO2/41 H2/Ni composition at 10400 C and 1
atmosphere pressure.














1344
D-band (for "disorder")
1609
G-band
S(for "graphite")


125 325 525 725 925 1125 1325 1525 1725 1925
Raman shift (1/cm)


Figure 6-7 Raman spectra of as prepared sample (647nm of a krypton laser) showing the
G and D band. 40 CH4/20 C02/41 H2/Ni composition at 10400 C and 1 atmosphere
pressure.


S 2300 2400 2500 2600 2700 2800 2900 3000 3100
Raman Shift (1/cm)


Figure 6-8 Raman spectra of as prepared sample showing second order peaks (632.8-nm
helium-neon laser). 40 CH4/20 C02/41 H2/Ni composition at 10400 C and 1
atmosphere pressure










T=10400C, Ptotal = 1 atm

0


CH4 C2H CH2


Figure 6-9 The C-H-O ternary phase diagram, which shows the different gas
compositions and activities of carbon. The one highlighted corresponds to
15sccm CH4, 50sccmCO2 and 15sccm H2.


carbon isoactivity lines
isoactivity line for carbon, a(C) = 1
compositions of selected gaseous species *
Experimental Compositions









Results for Composition 15 CH4/50 C02/15 H2/Ni

A mixture of 15 seem of methane, 50 seem of CO2 and 15 seem of H2 were passed

simultaneously in the CVD chamber as shown in Figure 6-9. The SEM image of the

carbon nanostructures are shown in Figure 6-10 and 6-11.The nanostructures were

deposited almost on the entire substrate. Figure 6-12 to 6-15 shows the HRTEM image of

the different nanostructures deposited. Figure 6-12 shows the HRTEM image of a typical

closed cap MWCNT, and the graphene sheets in the wall are aligned parallel to the tube

axis. The outer and inner diameter of this tube is 12.5 and 2.5 respectively. The Figure

also shows well ordered graphitic layers with a gap of 0.33-0.34nm between the layers.

Figure 6-11 shows a TEM micrograph of carbon nano-onion synthesized during the

process .The onion was poorly crystallized. Figure 6-14 and 6-15 show the same region

with Figure 6-14 being at a lower resolution. The Figure 6-15 shows graphitic particles or

sheets of graphite with a clear spacing of 0.33-0.34nm between the layers. The lattice

spacing was observed in 3 regions where the lattice was clear. Here it can be observed

that the activity of carbon is lower than unity and synthesis of not only carbon nanotubes

but also onions and graphitic sheets or particles has taken place. The activity of carbon at

this point is 10-2 as seen in Figure 6-9. Figure 6-16 shows the Raman spectra of carbon

nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm

of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica

substrate show the G-line peak at 1596cm-1 which is a characteristic of graphite sheets.

[21].The broad peak near 1337cm-1 indicates the existence of defective graphitic layers on

the wall surfaces, the so called "D band [22] which is related to structural disorder of sp2

bonded nanocrystalline and /or amorphous carbon species. This figure also shows the

radial breathing mode for this composition. At first glance the spectrum may be









interpreted as being noisy, but virtually all the features are reproduced in the spectra from

CH4 and H2 mixture. X. Zhao et al[23]. show the various radial breathing modes of the

MWCNTs. The peaks at 202 and the peak at 279 can be attributed for all of the Raman

spectra to MWCNTs. The second order peak at 2650cm-1 has very small intensity as

compared to the CH4 and H2 mixture. In conclusion, at this composition the degree of

graphitization is less because of decreasing activity of carbon. The Raman shows that the

carbon deposition at this activity is almost the same but different forms of carbon have

been formed in this region, majority being graphitic sheets and fibers with graphitic

fringes without a cavity that is typical feature of the nanotube. Very few nanotubes were

observed for this composition. The high flow rate of CO2 restricts the growth of

nanotubes and the tubes are no longer aligned. As seen in Figure 6-12, there could also be

a possibility that during the synthesis of carbon nanostructures, if the reaction is allowed

to take place for a long time, methane pyrolysis at 10400C leads to an appreciable

thickness of carbon coating over a nanotube core [43, 44]. The important part of this

particular composition is that the activity of carbon is two orders of magnitude less than

one and deposition of different forms of carbon can be observed. It would be interesting

to see what happens at this particular point in the phase diagram with the use of other gas

species in the phase diagram.






























Figure 6-10 SEM micrograph showing carbon nanostructures at 15CH4/50C02/15H2/Ni
composition at 10400 C and 1 atmosphere pressure


Figure 6-11 SEM micrograph showing carbon nanostructures at 15CH4/50C02/15H2/Ni
composition at 10400 C and 1 atmosphere pressure
































Figure 6-12 HRTEM image of a closed cap MWCNT. 15CH4/50CO2/15H2/Ni
composition at 10400 C and 1 atmosphere pressure


Figure 6-13 HRTEM image of a carbon nano-onion. 15CH4/50C02/15H2/Ni composition
at 10400 C and 1 atmosphere pressure




































Figure 6-14 TEM image of a graphitic sheet. 15CH4/50C02/15H2/Ni composition at
10400 C and 1 atmosphere pressure


Figure 6-15 HRTEM image showing graphitic fringes for the figure 6-12.
15CH4/50C02/15H2/Ni composition at 10400 C and 1 atmosphere pressure


1~






























125 325 525 725 925 1125 1325
Raman Shift (1/cm)


1525 1725 1925


Figure 6-16 Raman spectra of as prepared sample (647nm of a krypton laser) showing the
G and D band. 15CH4/50CO2/15H2/Ni composition at 10400 C and 1 atmosphere pressure

118

116

114

112

110

108

106

104

102

100
2100 2200 2300 2400 2500 2600 2700 2800 2900 3000
Raman shift(1/cm)


Figure 6-17 Raman spectra of as prepared sample showing second order peak (632.8-nm
helium-neon laser). 15CH4/50CO2/15H2/Ni composition at 10400 C and 1 atmosphere
pressure.


1339
D-band (for "disorder")
1592
S G-band
(for "graphite")


h











T=10400C, Ptotal =1 atm

0


CH4 C2H6 CH2


Figure 6-18 The C-H-O ternary phase diagram, which shows the different gas
compositions and activities of carbon. The one highlighted corresponds to
30sccm CH4, 30sccm CO2 and 30sccm H2


carbon isoactivity lines
isoactivity line for carbon, a(C) = 1
compositions of selected gaseous species *
Experimental Compositions *









Results for Composition 30 CH4/30 CO2/30H2

A mixture of 30 seem of methane, 30 seem of CO2 and 30 seem of H2 were passed

simultaneously in the CVD chamber. The point is highlighted in Figure 6-18. SEM image

of the carbon nanostructures are shown in Figure 6-19 and 6-20.The nanostructures were

deposited more on the edges of the substrate. Figure 6-21 to 6-24 shows the HRTEM

image of the different nanostructures deposited. Figure 6-21 shows a defective nanotube

with the graphitic walls uneven and some of the fringes entering the capsule of the tube

as can be seen in Figure 6-22. Figure 6-23 shows a nanostructure which is close to a

carbon nanofiber with some fringes but all the fringes are discontinuous and are not

present in most of the regions. Figure 6-24 shows a top part of a closed end carbon

nanotube with lattice spacing of 0.34nm. Figure 6-25 shows the Raman spectra of carbon

nanotubes. Micro Raman spectroscopy was used with an excitation wavelength of 647nm

of a krypton laser. The Raman spectra of a nanotube grown by CVD on Ni coated silica

substrate show the G-line peak at 1596 cm-1 which is a characteristic of graphite sheets.

[21] .The broad peak near 1337cm-1 indicates the existence of defective graphitic layers

on the wall surfaces, the so called "D band" [22] which are related to structural disorder

of sp2 bonded nanocrystalline and /or amorphous carbon species. This figure also shows

the radial breathing mode for this composition. At first glance the spectrum may be

interpreted as being noisy, but virtually all the features are reproduced in the spectra from

CH4 and H2 mixture. X. Zhao et al[23]. show the various radial breathing modes of the

MWCNTs. The peaks at 202and the peak at 279 can be attributed for all of the Raman

spectra to MWCNTs.

This composition is very interesting as many experiments have been done in the

past using CO as the feed gas for the synthesis of single walled carbon nanoubes at









approximately the same temperature, pressure as the one reported here [8, 45, 46]. The

activity of carbon at this point is exactly unity for both experiments but different gases

have been used. It is very interesting to note that past experiments yielded single walled

carbon nanotubes by using CO gas alone at a temperature very close to the one used here

at (10400C). The reaction that takes place when CO is used is 2CO(g)-C(s)+02(g)[45].

At this particular composition and temperature, it is most likely that a mixture of H2 and

CO will be produced. Bronikowski et al[46] synthesized large quantity ropes of SWCNTs

at 1050 o C and different pressures (1-lOatmosphere).

The activity for both these experiments the ones from Smalley's lab[8, 45, 46] and

the ones described here have the same activity of carbon but different partial pressures of

oxygen and hydrogen. It can be seen that CO is placed on the C-O line from Figure 5.6.

From the above conclusions can be made that the synthesis of carbon nanotubes not only

depends on the composition of the various gases but also which gas mixtures are fed into

the reactor and the partial pressures of oxygen and hydrogen. It can be seen that even

though this particular composition corresponds to the activity one line there is no sign of

single walled carbon nanotubes present and neither were the nanotubes produced in large

quantities as done by R.E Smalley et al.[8, 45, 46].































Figure 6-19 SEM micrograph showing carbon nanostructures at 30 CH4/30 C02/30 H2/Ni
composition at 10400 C and 1 atmosphere pressure


Figure 6-20 SEM micrograph showing carbon nanostructures at 30 CH4/30 C02/30 H2/Ni
composition at 10400 C and 1 atmosphere pressure
























0


Figure 6-21 HRTEM image of a defective nanotube with an uneven cavity. 30 CH4/30
C02/30 H2/Ni composition at 10400 C and 1 atmosphere pressure


Figure 6-22 HRTEM image of the walls of the nanotube shown above. 30 CH4/30
C02/30 H2/Ni composition at 10400 C and 1 atmosphere pressure































Figure 6-23 HRTEM showing the synthesis of carbon nanofiber along with the
nanotubes30 CH4/30 C02/30 H2/Ni composition at 10400 C and 1 atmosphere
pressure


Figure 6-24 HRTEM image of a closed cap MWCNT. 30 CH4/30 C02/30 H2/Ni
composition at 10400 C and 1 atmosphere pressure






48




1334.9
D-band (for "disorder")

1601.4
G-band
(for "graphite")




l-1










125 325 525 725 925 1125 1325 1525 1725 1925
Raman shift (1/cm)
Figure 6-25 Raman spectra of as prepared sample (647nm of a krypton laser) showing the
G and D band. 30 CH4/30 C02/30 H2/Ni composition at 10400 C and 1
atmosphere pressure













T=10400C, Ptotal = 1 atm

O


10.-1 CO2
104,10-4
10


P02


CH4 C2H6 CH2


We are here now


Figure 6-26 The C-H-O ternary phase diagram, which shows the different gas
compositions and activities of carbon. The one highlighted corresponds to
45sccm CH4, and 50sccm H2


carbon isoactivity lines
isoactivity line for carbon, a(C) = 1
compositions of selected gaseous species *
Experimental Compositions









Results for Composition 45CH4/50 H2/Ni

A mixture of 45 seem of methane, 50 seem of H2 were passed simultaneously in the

CVD. The point can be clearly seen on the C-H side of the C-H-O diagram Figure 6-26-

SEM images of the carbon nanostructures are shown in Figure 6-27. The HRTEM shows

us that most of the nanostructures on this substrate are nanotubes. Figure 6-28 shows a

typical nanotube with an inner diameter of about 5 nanometers and outer diameter of

about 28nm. All the carbon nanotubes have no encapsulated catalytic particles at the

closed tip. Figure 6-30 shows a balloon shaped carbon nanotube, it is not typical to grow

balloon shaped nanotubes at this composition. Figure 6-29 shows a bunch of carbon

nanotubes with well ordered graphitic layers with a gap of 0.34 between the layers. The

activity of carbon at this point is 102 as seen in Figure 6-26. Figure 6-31 shows the

Raman spectra of carbon nanotubes. Micro Raman spectroscopy was used with an

excitation wavelength of 647nm of a krypton laser. The Raman spectra of a nanotube

grown by CVD on Ni coated silica substrate show the G-line peak at 1596-9 cm-1 which

is a characteristic of graphite sheets [21].The broad peak near 1337cm-1 indicates the

existence of defective graphitic layers on the wall surfaces, the so called "D band" [22]

which is related to structural disorder of sp2 bonded nanocrystalline and /or amorphous

carbon species. This figure also shows the radial breathing mode for this composition. X.

Zhao et al[23]. show the various radial breathing modes of the MWCNTs. The peaks at

202 and the peak at 279 can be attributed for all of the Raman spectra to MWCNTs .In

conclusion most of the nanostructures are carbon nanotubes with very few exceptions

being nanofibers. The nanotubes had highly oriented graphitic walls and only MWCNTs

are observed. It is observed that many nanotubes are covered with amorphous carbon as

seen in Figure 6-29, there could also be a possibility that during the synthesis of carbon









nanostructures, if the reaction is allowed to take place for a long time, methane pyrolysis

at 10400C leads to an appreciable thickness of carbon coating over a nanotube core

[43,44].


Figure 6-27 SEM image of carbon nanostructures. 45 CH4/50 H2/Ni composition at 10400
C and 1 atmosphere pressure


Figure 6-28 HRTEM showing MWCNTs with graphitic fringes. 45 CH4/50 H2/Ni
composition at 10400 C and 1 atmosphere pressure































Figure 6-29 HRTEM showing a bunch MWCNTs with graphitic fringes and a lattice
spacing of 0.34nm. 45 CH4/50 H2/Ni composition at 10400 C and 1
atmosphere pressure


Figure 6-30 HRTEM showing a carbon nanotube in the shape of a balloon. 45 CH4/50
H2/Ni composition at 10400 C and 1 atmosphere pressure.






53







1340
D-band (for "disorder")

SI1601
m G-band
(for "graphite")

C

-lI








125 325 525 725 925 1125 1325 1525 1725 1925
Raman Shift (1/cm)


Figure 6-31 Raman spectra of as prepared sample (647nm of a krypton laser) showing the
G and D band. 45 CH4/50 H2/Ni composition at 10400 C and 1 atmosphere
pressure










T= 10400C, Ptotal =1 atm

O


We are here now




10- C 2
10-14,10 5
10 5

10 -2
S 10-1 CO
1


CH4 2H6 CH2


Figure 6-32 The C-H-O ternary phase diagram which shows the different gas
compositions and activities of carbon. The one highlighted corresponds
50sccm CO250sccm CO2


carbon isoactivity lines
isoactivity line for carbon, a(C) = 1
compositions of selected gaseous species *
Experimental Compositions









Results for Composition 50 seem CO2

Fifty standard cubic centimeter of CO2 and 1000sccm Ar was passed

simultaneously in the CVD chamber. The point can be seen clearly on the C-O side of the

C-H-O ternary phase diagram. The SEM image of a large quantity of SiOx nanowires are

shown in Figure 6-33 and 6-34. The SiOx nanowires are several micrometers in length

and the diameter of the SiOx nanowires is about 15-50 nm approximately. TEM was

employed to study the structure of the nanowires in detail. Figure 6-35 shows the high

resolution TEM image of SiOx nanowire. The TEM reveals that the nanowires are in

perfect amorphous state. Figure 6-36 shows that the catalyst particle is attached to the tip

of the nanowires showing tip growth of the nanowires. Figure 6-37 shows that apart from

the SiOx nanowires some graphitic features could also be seen. The lattice spacing

between the layers is 0.34 which exactly matches with graphite but these features were

very scarce. Raman spectra were observed using an excitation source with a 632.8-nm

helium-neon laser operating with nominal output energy of 15 mW. The G and D line

peaks which are a characteristic of graphitic peaks and carbonaceous particles have

disappeared. The Raman spectra on the bulk of the SiOx nanowires Figure 6-39 shows an

asymmetrical sharp peak at 508 cm-1 with line width of 10 cm-1 Previously, Yu et al. [47]

reported that the Raman spectrum of amorphous SiO2 nanowires is the same as that of

bulk non-crystalline SiO2 solids; whereas Wang et al. thought that there is no scattering

peak in the stoichiometric SiO2 nanowires [48]. It has been reported that there is a peak at

480 cm-1 for bulk amorphous silicon [49] and a peak at 520 cm-1 for crystalline silicon.

The above investigation verified that there is no trace of crystalline phase in the silicon

oxide nanowires here. The activity of carbon at this point is 10-5. Figure 6-40 shows the

EDS analysis showing a clear Silicon, oxygen and nickel peak. In conclusion synthesis of









bulk quantities of SiOx nanowires using CO2 and Ar is observed; apart from the

nanowires some graphitic fringes could also be seen.


Figure 6-33 SEM image of nanostructures. 50 seem CO2


Figure 6-34 SEM image of nanostructures. 50 seem CO2


























Figure 6-35 HRTEM image showing amorphous Siox. 50 seem CO2


Figure 6- 36 HRTEM image showing amorphous SiOx nanowire with catalyst particle
attached at the tip. 50 sccm CO2


















0-1;


2 nm


Figure 6-37 HRTEM image showing some graphitic layers with a lattice spacing of
0.34nm which were also present. 50 sccm CO2


1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
Raman Shift (cm-1)

Figure 6-38 Raman spectra of as prepared sample (632.8-nm helium-neon laser). 50 sccm
C02








1200


400 450 500 550 600
Raman Shift (cm-1)


650 700 750 800


Figure 6-39 Raman spectra of as prepared sample showing the silica peak (632.8-nm
helium-neon laser). 50 sccm CO2


Figure 6-40 EDS Spectra showing Si, O and Ni peaks. 50 sccm CO2


\___


__,









Experiments Using Iron Nanoparticles as Catalyst

Various compositions have been used for the growth of carbon nanostructures

using iron nanoparticles as catalyst. The various compositions are given in the table

below in table 2.1. Only SEM has been used to characterize the various nanostructures

and thus a more detailed study is required to confirm the type of nanostructures. The

SEM images for the various compositions have been shown below.


Figure 6- 41 80 CH4/17 CO2 SEM image showing nanostructures


Figure 6- 42 50 CH4/50 CO2 SEM image showing nanostructures































Figure 6- 43 15 CH4/50 CO2 SEM image showing nanostructures


Figure 6- 44 50 CO2 SEM image showing nanostructures




























Figure 6- 45 EDS at 50 CO2
Figure 6- 45 EDS at 50 CO2














CHAPTER 7
CONCLUSIONS

In all the Raman spectra except the one for CO2 the degree of graphitization by the

ID/IG ratio is almost the same and in agreement with the CVD growth of carbon

nanostructures. Many forms of carbon nanostructures were synthesis using different gas

compositions. At the temperature and pressure reported here the equilibrium species of

more importance are CO, C02, H20, H2 and CH4. Using various gas composition

syntheses of various forms of carbon like carbon nanotubes, nanofibers, onions, spheres,

films and silica nanowires are reported. These experiments support the role of C-H-O

Gibbs phase triangle and suggesting that the synthesis is possible by using injection of a

mixture of various gases like CO, C02, H20, CH4 and H2. Here experiments further

elucidate that the conditions for termination of carbon deposition is anywhere between

10-2 and 10-3 activity which is well below the deposition condition for carbon. A proper

understanding as to why this is happening has to be studied. It can also be seen that the

carbon deposition is not only related to the carbon activity but also depends on the partial

pressures of 02 and H2. A detailed study using other gas species in the C-H-O system has

to be done in order to get a particular phase where high purity of carbon nanostructures of

a particular choice can be obtained. A better understanding for the mechanism of

nanostructure formation under C-H-O conditions should make it possible to optimize the

conditions for the growth of large quantities of nanotubes. The eventual goal of this work

is to explain the synthesis of carbon nanotubes and to provide an analysis of gas






64


compositions on different stages of the synthesis equilibrium thermodynamics, since in-

situ CVD experiments are extremely difficult to analyze.
















LIST OF REFERENCES

1. P K Bachmann, D. Leers and H. Lydtin, Diamond and Related Materials 1 (1991)
1.

2. N. A. Prijaya, J. C .Angus and P. K. Bachmann, Diamond and Related Materials 39
(1993) 129.

3. Y. Z Wan, D.W. Zhang, Z.J. Lai, J. T. Wang Appl. Phys. A 67. (1998) 225.

4. S. Iijima, Nature 354 (1991) 56.

5. T.V. Hughes, C. R. Chambers, US Patent 405 480 (1889).

6. W.R. Davis, R.J. Slawson and G.R. Rigby, Nature 171 (1953) 756.

7. R.T.K. Baker and P.S Harris, Chem. Phys. Carbon 14 (1978) 83.

8. P. Nikolaev, M. J. Bronikowski, R. K Bradley, F. Rohmund, D. T. Colbert, K. A.
Smith and R. E. Smalley, Materials Research Bulletin 1-2 (1999) 91.

9. C. Journet and P Bernier Appl, Phys. Chem. 96 (1992) 6941.

10. H.W Kroto, J. R. Heath, S. C. O. Brien, R. F. Curl, R.E. Smalley, Nature 318
(1985) 1629.

11. Ch. Emmenegger, P. Mauron, A. Zuttel, Ch. Nutzenadel, A. Schneuwly, R Gallay
and L. Schlapbach, Appl. Surf. Science 452 (2000) 162.

12. C.M. Lieber, Solid State Commun, 107 (1998) 607.

13. Basic Properties of Carbon Nanotubes, http://www.applied-
nanotech.com/cntproperties.htm (accessed June 2003)

14. IBM Research, http;// www.research.ibm.com/nanoscience/fet.html. (accessed June
2003)

15. V. Derycke, R. Martel, J. Appenzeller and P.H. Avouris,Carbon Nanotube Inter-
and Intramolecular Logic Gates. IBM Research Division, New York.

16. P. M. Ajayan and O. Z Zhoum, Applications of Carbon Nanotubes. Department of
Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, NY.






66


17. P. E Nolan, D. C. Lynch and Andrew Hall Cutler, J. Phys. Chem. B, 102 (1998)
4165

18. J. A. Libera, Y. Gogotsi, G. Yazicioglu and C. Megaridis, Appl. Phys. Lett. 79
(2001) 1021.

19. H. Lee, Y-S. Kang, P. S. Lee and J.-Y. Lee, Journal of alloys and compounds 330-
332 (2002) 569.

20. A. Cao, X, Zhang, C. Xu, J. Liang, and D. Wu, Journal Mater. Res. 16 No 11
(2001).

21. P.C. Eklund, J.M. Holden and RA Jishi, Carbon 33 (1995) 959.

22. R. T. K. Baker, Carbon 27, (1989) 315.

23. X. Zhao, Y. Ando, L.-C. Qin, H.-Kataura, Y. Maniwa, R. Saito, Physica B 323
(2002) 265.

24. C.-F. Chen, C.-L. Lin and C. -Ming Wang and N. Jiang, App Phy Lett 82, 15
(2003) 2515.

25. J. Melancon and C. W. Bale, Oxid. Met. 18 (1982) 147.

26. C. W. Bale, J. Melancon and A. Pinho, "High Temperature Gaseous Equilibria-
Computation and Graphical Representation," Can. Met. Quat. 19 (1981), 363.

27. HSC Chemistry for Windows (ver. 4.1): Chemical Reaction and Equilibrium
Software with Extensive Thermochemical Database, Outokumpu Research, Oy,
Finland.

28. 0. Kubaschewski, C. B. Alcock and P. J. Spencer. Materials Thermochemistry,
6th ed., Pergamon Press, New York 1993.

29. R. L. Burden and J. D. Faires, Numerical Analysis, 4th ed., PWS-Kent, Boston,
1989.

30. Chapter 4 of Thesis N.Kulkarni, part of Ph.D. proposal, University of Florida,
2002, Now at ORNL.

31. Facility for the Analysis of Chemical Thermodynamics (F*A*C*T), Ecole
Polytechnique, Montreal, Quebec, Canada.

32. NIST-JANAF Thermochemical Tables, 4th ed., edited by M. W. Chase, Jr., pub.
Am. Chem. Soc. & Am. Inst. Phys.Washington, D.C., 1998.

33. I. Barin, Thermochemical Data of Pure Substances, V.C.H., Weinheim, Germany,
1989.









34. C. W. Bale, A. D. Pelton and W. T. Thompson, Can. Met. Quat. 25 (1986) 107.

35. R. T. DeHoff, Thermodynamics in Materials Science, McGraw-Hill, New
York, N.Y. (1993).

36. C. Petit, A.Taleb and M. P. Pileni, J. Phys. Chem. B. 103 (1999) 1805.

37. S.-J. Park, S. Kim and S. Lee, J.Am.Chem. Soc 2000,122, (2000) 8581.

38. C Bower, O. Zhou, D.J. Werder, and S Jin, Appl. Phys. Lett. 77 (2000) 2767.

39. V.I. Merkulov, D.H. Lowndis,Y. Y Wei, G. Eres, and E. Voelkl, Appl. Phys. Lett
76, (2000) 3555.

40. S. Amelinckx, X. B. Zhang, D. Bernaerts, X. F. Zhang, V. Ivanov, J.B. Nagy,
Science 265 (1994) 635.

41. G. G.Tibbetts, M. G. Devour, and E.J. Rodda, Carbon 25 (1987) 367.

42. S C Tsang, P.J .F Harris and M. L. H. Green, Nature 362 (1993) 520.

43. Tibbetts, G. G. in Carbon Fibers, Filaments and Composites 73-94 Kluwer
Academic, Amsterdam (1990).

44. L. C. Qin and S. Iijima, Mater. Lett. 30 (1997) 311.

45. H. Dai, A. G. Rinzler, P. Nikolaev, A. Thess, D. T. Colbert and R. E. Smalley
Chemical Physics Letters 260 Issues 3-4 (1996) 471.

46. M. J. Bronikowski, P. A Willis, D. T. Colbert, K. A. Smith and R.E. Smalley J.
Vac. Sci. Technol. A 19 No 4 (2001) 1800.

47. D.P. Yu, Q.L. Houg, Y. Ding, H.Z. Zhang, Z.G. Bai, J.J. Wang, Y.H. Zou, W.
Qian, G.C. Xiong and S.Q. Feng, Appl. Phys. Lett. 73 (1998) 3076.

48. N. Wang, Y.H. Tang, Y.F. Zhang, C.S. Lee, I. Bello and S.T. Lee, Chem. Phys.
Lett. 299 (1999) 237.

49. A.T. Voutsaa, M.K. Hatilis and J. Boyce, J. Appl. Phys. 78 12 (1995) 6999.















BIOGRAPHICAL SKETCH

Archit Lal was born in India, Jabalpur. He came to Pune when he was one year old

and has been brought up there. He finished his schooling from Loyola High School,

Pune, one of the premier schools in India. Apart from being excellent throughout his

academic career, Archit is a natural athlete and has represented the state of Maharashtra

in badminton and table tennis; he has also represented India in field hockey. Even after

scoring a good percentage in the 10th standard examination, he continued his schooling in

Loyola. Doing well in his 12th standard Maharashtra board exam he got admission into

one of the top universities and engineering colleges. His undergraduate degree was in

Petrochemical Engineering from the University of Pune. He worked in the National

Chemical Laboratory after graduation. The National Chemical Laboratory is one of the

premier institutes in India. As he enjoyed his research in NCL he decided to go to

graduate school. Getting admission and a research assistantship into University of

Florida in the Materials Science and Engineering Department, which is currently ranked

in the top 10 departments in the nation, made his decision easy. Ever since Archit has

been at the University of Florida, Gainesville, and will be graduating with a Master of

Science degree in August 2003.