CVD synthesis of nitrogen doped carbon nanotubes using iron pentacarbonyl as catalyst By Nafise Ghadimi Submitted in fulfillment of the requirements for the degree of Masters of Science in the Faculty of Science Department of Chemistry University of the Witwatersrand Private Bag X03 WITS 2050 Supervisor: Professor Neil. J. Coville August 2011 ii DECLARATION I declare that this is my own unaided work under the supervision of Professor Neil. J. Coville. It is being submitted to the Faculty of Science at the University of the Witwatersrand for the degree of Masters of Science. It has not been submitted for any degree or examination in any other University. Name: Nafise Ghadimi Signature of candidate ……………. ……………day of…………….2011 iii DEDICATION Lovingly dedicated to my magnificent parents and my supportive sister Hadise iv ACKNOWLEGMENTS I would like to express my gratitude to my Almighty God for his support, kindness, strength, and giving me a supportive family. I would like to thank my supervisor Professor Neil. J. Coville for his infinite support and assistance and also for giving me enough confidence to run this project. I also want to express my great appreciation to my beloved sister Hadiseh and my fantastic parents. Without their support I would not have achieved this stage in my life. I would like to thank, Zikhona, Messai, and Ahmed for their consultation and Manoko and George for doing TGA analysis. I am also thankful to Mr. Rudolph Erasmus for doing Raman analysis. I am also thankful to the Microscopy and Microanalysis Unit, in particular Professor Michael Witcomb, Professor Alexander Ziegler, Caroline Lalkhan and Abe Seema for their assistance. The financial support for this work by the DST/NRF centre of excellence in strong materials is highly appreciated. v ABSTRACT In this dissertation, the synthesis of nitrogen doped carbon nanotubes (N-CNTs) was performed successfully, using a floating catalyst chemical vapour deposition (CVD) method. Fe(CO)5 was utilized as the catalyst and acetonitrile and toluene as nitrogen and carbon sources respectively. Two different procedures were used to add reagents to the reactor: an injection method and a bubbling method. The effect of nitrogen concentration and physical parameters such as reaction temperature, gas flow rate on the morphology, crystallinity and thermal stability of the tubes was studied. The synthesized materials were characterized by means of Raman spectroscopy, TGA and TEM analyses. The presence of nitrogen was confirmed by the presence of the bamboo formations in the tubes by TEM. A comparison of the data from the numerous reactions revealed that N-CNTs can be made from Fe(CO)5 and acetonitrile. Further the main conclusions achieved using the injection method were: i) the maximum number of tubes with bamboo structure were made using on acetonitrile concentration of 15%, ii) The best growth temperature to make N-CNTs was 850 oC, iii) An increase in acetonitrile concentration decreased the yield of N- CNTs and iv) Tubes with the narrowest outer diameters were made using an acetonitrile concentration of 15%. TABLE OF CONTENTS vi SECTION PAGE DECLARATION ii DEDICATION iii ACKNOWLEGMENTS iv ABSTRACT v TABLE OF CONTENTS vi LIST OF FIGURES xii LIST OF TABLES xv LIST OF ABBREVIATIONS xvii CHAPTER 1 1 1. INTRODUCTION 1 1.1 Background to the study 1 1.2 Objectives 2 1.3 Outline of the dissertation 3 1.4 References 4 CHAPTER 2 6 2. LITERATURE REVIEW 6 2.1 Carbon nanomaterials 6 2.1.1 Allotropes of carbon 6 2.1.2 Fullerene 7 2.2 Carbon nanotubes 8 2.2.1 Historical introduction 8 vii 2.2.2 Structural features of CNTs 9 2.2.3 Types of CNTs 10 2.2.4 Properties of CNTs 11 2.2.4.1 Mechanical properties of CNTs 11 (a) Stress 11 (b) Young’s modulus 12 (c) Elastic modulus of carbon nanotubes 12 2.2.4.2 Thermal properties 13 (a) Coefficient of thermal expansion 13 (b) Thermal conductivity 13 (c) Electronic properties 13 2.2.4.3 Chemical properties 13 2.2.4.4 Graphite structure 14 (a) Graphite 3D system 14 (b) Graphite 2D system 14 2.2.4.5 Doping of graphene 14 2.2.4.6 Physical properties 15 (a) Chirality of CNTs 15 2.2.5 Synthesis methods to make CNTs 16 2.2.5.1 Synthesis method to make SWCNTs 16 2.2.5.2 Synthesis methods to make MWCNTs 16 (a) Arc discharge 16 (b) Laser ablation 16 viii (c) Chemical vapour deposition (CVD) 17 2.2.6 CNT purification processes 19 (a) Gas phase oxidation 19 (b) Liquid phase oxidation 19 2.2.7 Doping of carbon nanotubes 19 2.2.7.1 Doping configuration 20 2.2.7.2 Defects induces by nitrogen doping 21 2.2.7.3 N-CNT structure 22 2.2.7.4 Growth mechanism 23 (a) Growth mechanism for undoped CNTs 23 (b) Growth of N-CNTs 24 2.2.7.5 Synthesis of nitrogen doped CNTs 25 2.2.7.6 Consequences of N-insertion 26 2.2.7.7 CNTs in biological applications 27 (a) Functionalisation of N-CNTs with proteins 28 (b) N-CNTs in biosensing 28 2.2.7.8 Role of CNTs (or) N-CNTs in catalysis 28 (a) Metal supported CNTs 28 (b) Metal-free carbon catalysts 30 (c) CNTs in fuel cells 31 2.2.7.9 Characterization techniques used for carbon nanotubes 33 (a) Transmission electron microscopy 33 ix (b) Raman spectroscopy 34 (c) Thermogravimetric analysis 34 2.3 Toxicity of carbon nanotubes 35 2.4 Conclusion 36 2.5 References 36 CHAPTER 3 51 3. CVD SYNTHESIS OF NITROGEN DOPED CARBON NANOTUBES USING IRON PENTACARBONYL AS CATALYST 51 3.1 Introduction 51 3.1.1 Carbon nanotubes (CNTs) 51 3.2 Experimental 54 3.2.1 Chemicals and materials 54 3.2.2 Characterization 54 3.2.2.1 Thermogravimetric analysis 54 3.2.2.2 Raman spectroscopy 54 3.2.2.3 Transmission electron microscope 54 3.2.3 Synthetic procedures: Injection method 55 3.2.3.1 The synthesis of nitrogen doped carbon nanotubes (N-CNTs) 55 3.2.3.2 Purification of N-CNTs 58 3.3 Results and discussion 58 3.3.1 Synthesis of nitrogen doped carbon nanotubes 58 3.3.2 Effect of % CH3CN in toluene 59 3.3.2.1 Yield study 59 3.3.2.2 TEM studies 60 x (a) Sample analysis 60 (b) Outer diameter distribution of the N-CNTs 62 (c) Inner diameter distribution of the N-CNTs 65 (d) Compartment distance study 66 (e) Size of spheres 67 3.3.2.2 TGA analysis 68 3.3.2.3 Raman spectroscopy 71 3.3.3 Effect of temperature 73 i) Effect of temperature on the outer diameter distribution 76 ii) Effect of temperature on the bamboo compartments 77 3.3.4 Effect of carrier gas 78 i) The effect of hydrogen on carbon yield 78 ii) The effect of hydrogen on type of production 79 iii) The effect of hydrogen on the CNT outer diameters 81 3.2.4 Synthesis procedure: Bubbling method 82 i) Synthesis of N-CNTs 82 ii) Results and discussion 82 iii) Yield study 83 iv) TEM studies 84 v) Outer diameter distribution 87 3.4 Conclusion 90 3.5 References 91 CHPATER 4 95 xi 4. CONCLUSIONS AND RECOMMENDATIONS 95 4.1 Conclusions 95 4.2 Recommendations for future study 97 4.3 References 97 xii LIST OF FIGURES FIGURES PAGE Figure 2.1 Some allotropes of carbon: a) diamond; b) graphite; c) lonsdaleite; d–f) fullerenes (C60, C540, C70); g) amorphous carbon; h) carbon nanotube 8 Figure 2.2 The structure of graphite 10 Figure 2.3 The different shapes of CNTs. (a) SWCNT, (b) DWCNT, (c) MWCNT 11 Figure 2.4 The different shapes of single walled carbon nanotubes 15 Figure 2.5 Catalyst used for CNTs synthesis; (a) ferrocene and (b) Fe(CO)5 18 Figure 2.6 Possible bonding configurations for N in graphitic networks: (a) pyridine-like N, (b) pyrrole-like N, (c) substitutional N, (d) nitrile CN, (e) amine NH2, (f) single N pyridinic vacancy, (g) triple N pyridinic vacancy, and (h) interstitial N 21 Figure 2.7 TEM images of MWCNTS doped with nitrogen, produced by thermolysis of ferrocene-benzylamine solutions at 850 oC: (a) low- resolution image of compartmentalized structure of N-CNTs. (b) high- resolution image of an individual bamboo compartment 22 Figure 2.8 The position of pentagons in a CNT cap 23 Figure 2.9 Schematic of tip and base growth 24 Figure 2.10 Growth mechanism proposed for N-doped MWCNT bamboo or ‘nanobell’ structures. (a) before precipitating carbon layers matching the particle shape (b). The addition of new precipitate layers (b) increases pressure on the catalyst particle (c) until it is ejected from the inside of the cup (d). When the strain is relieved the process can repeatedly (e-f) 25 Figure 2.11 ODH of ethylbenzene 31 xiii Figure 2.12 A 4 e- ORR process 32 Figure 3.1 A furnace and syringe attachment used for synthesis the N- CNTS 57 Figure 3.2 The TEM image of bamboo structure doped with N 61 Figure 3.3 : TEM images of the N-CNT samples produced using CH3CN: 2% (a, b ,c), 5% (d, e, f), 15% (g, h, i) respectively (T: 850 oC, flow rate: 300 mL/min) 62 Figure 3.4 Graphs of diameter distribution made from 2% (a). 5% (b), 15% (c) of CH3CN, (T: 850 oC, flow rate 300 mL/min) 65 Figure 3.5 TEM images of bamboo compartments and the distance between bamboo caps 66 Figure 3.6 The average distance between the compartment caps for N- CNTs made from different %CH3CN (T: 900 oC, flow rate: 200 mL/min) 67 Figure 3.7 Synthesized spheres from the 15% sample (T: 850 oC, flow rate: 300 mL/min) 68 Figure 3.8 TGA curves of the purified N-CNTs with different N content (A) 0% , (B) 5%, (C) 15%, (D) 25% 69 Figure 3.9 The derivative curve of 0% CH3CN 70 Figure 3.10 Raman spectra of the N-CNTs with different N content (A) 2%, (B) 5%, (C) 15%, (D) 20%, (E) 25% 72 Figure 3.11 TEM images of the samples synthesized at 750 oC (a-b), 800 oC (c-d), 850 oC (e-f), 900 oC (g-h) ( % CH3CN: 15% flow rate: 200 mL/min) 76 Figure 3.12 Graph of diameter distribution of N-CNTs synthesised at a) 850 oC, b) 900 oC 77 Figure 3.13 TEM images of N-CNTs produced using different carrier gases: Ar (a, b, c), Ar + 5% H2 (d, e, f) (T: 850 oC, flow rate: 200 mL/min) 80 Figure 3.14 Graphs of diameter distributions using different carrier gas a) Ar, b) Ar + 5% H2 (15% CH3CN in toluene, T: 850 oC, flow rate: 200 mL/min) 81 xiv Figure 3.15 TEM images of the N-CNTs produced under different conditions. Figures a, b (T: 800 oC, flow rate: 200 mL/min, bubbled at RT), c, d (T: 800 oC, flow rate: 300 mL/min, bubbled at 75 oC), e, f (T: 850 oC, flow rate: 200 mL/min, bubbled at 40 oC), g, h (T: 850 oC, flow rate: 300 mL/min, bubbled at RT), i, j (T: 900 oC, flow rate: 300 mL/min, bubbled at RT), k, l (T: 900 oC, flow rate: 300 mL/min, bubbled at RT), m, n (T: 900 oC, flow rate: 100 mL/min, bubbled at RT) 87 Figure 3.16: Graphs of diameter distributions for the N-CNTs produced under different conditions, a ( T: 800 oC, flow rate: 200 mL/min), b (T: 800 oC, flow rate: 300 mL/min, bubbled at 75 oC), c (T: 850 oC, flow rate: 200 mL/min bubbled at 40 oC), d (T: 900 oC, flow rate: 300 mL/min) 90 xv LIST OF TABLES TABLES PAGE Table 2.1 Corresponds to Young’s modulus of some materials 12 Table 2.2 Physical properties of carbon, nitrogen and oxygen. aaccording to Pauling, bin aromatic compounds 27 Table 3.1 Use of Fe(CO)5 as catalyst to make doped and undoped CNTs 53 Table 3.2 Experimental conditions used for the N-CNTs produced 56 Table 3.3 Samples content as determined by using TEM images 61 Table 3.4 The average outer diameter distributions 63 Table 3.5 Average inner diameter distribution of the different CH3CN% concentration 65 Table 3.6 Distance between N-CNT bamboo compartments 67 Table 3.7 Thermogravimetric analysis of N-CNTs and CNTs (A-D) 70 Table 3.8 Raman analysis of the N-CNTs 72 Table 3.9 The content of the samples synthesized at different temperature 73 Table 3.10 showing the use of different C/N sources and catalyst on fabrication of N-CNTs at different temperatures 74 Table 3.11 The avgerage distance between individual bamboo compartment at 850 and 900 oC 78 Table 3.12 The experimental conditions used for N-CNT production 79 xvi Table 3.13 Shows the effect of hydrogen on the quality of sample content 80 Table 3.14 The experimental conditions used for the synthesis of N-CNTs 82 Table 3.15 Displays the average outer diameter under different conditions 88 xvii LIST OF ABBREVIATIONS SCNM(s) Shaped carbon nanomaterial(s) CNT(s) Carbon nanotube(s) N-CNT(s) Nitrogen doped carbon nanotube(s) CVD Chemical vapour deposition TGA Thermal gravimetric analysis TEM Transmission electron microscopy XPS X-ray photoelectron spectroscopy SWCNT(s) Single walled carbon nanotube(s) DWCNT(s) Double walled carbon nanotube(s) MWCNT(s) Multi walled carbon nanotube(s) PVD Physical vapour deposition AACVD Aerosol assisted chemical vapour deposition DLICVD Direct liquid injection chemical vapour deposition MPCVD Microwave plasma assisted chemical vapour deposition ORR Oxygen reduction reaction ODH Oxidative dehydrogenation ID/IG The intensity ratio of the D to G band (Raman peaks) CCD camera Charge coupled device camera 1 CHAPTER 1 1. INTRODUCTION 1.1 Background to the study A majority of inventions at the nano scale, that have been described, arise from research in academic and government laboratories. This includes both their discovery and their product development. Nanotechnology products can be found in any field, for example, they can be used in building materials and also in biological materials. Nanomaterials can be used in scratch resistance artefacts and catalysts. They have also been added to polymers to give stronger lighter polymer nanocomposites. The study of a shaped carbon nanomaterial (SCNM) such as a carbon nanotube is a hot research topic. Many studies were carried out on these types of materials by several researchers during the time period 1952-1989 [1-17]. However it is generally accepted that it is Iijima who brought CNTs to the awareness of the scientific community [18]. He linked their structure to that of fullerenes. CNTs possess many extraordinary properties: they are stronger that steel, harder than diamond, their electrical conductivity is higher than copper and their thermal conductivity is higher than diamond. Consequently, CNTs have become a material, of much interest today. Although there has been much progress in the synthesis of CNT over the years, many challenges still remain in their development. For example, methods to produce CNTs in high purity with controlled morphology, the actual mechanism of carbon growth, and the influence of defects on these physical properties are issues that are still open to study and debate. Due to the remarkable properties associated with CNTs, scientists have devised strategies to both understand and more importantly to improve on these properties. For example, in 1994 Stephen et al. doped CNTs with N and B atoms using an arc discharge method [19]. Doping not only modified the CNT morphology, but also 2 changed their electronic properties of the CNTs [20]. The development (synthesis study) of doped CNTs, in particular with nitrogen, is described in chapter 2. Among the techniques, used to make CNTs, the CVD method has been proven to be the most popular and widely used because of its low cost, high production yield and it is an easy method to scale-up. In this dissertation a CVD method was used to make nitrogen doped CNTs (N-CNTs) using a floating catalyst process. In the floating catalyst method both catalyst and carbon source are introduced in the gaseous phase. The method requires a volatile catalyst. Organometallic complexes such as ferrocene (FcH) and substituted ferrocene are typically used. Many studies have been done using ferrocene since the first reports on its first use in 1985 by Komatsu and Endo to fabricate nano shaped materials. But other readily available iron containing organometallic complexes have been used to make CNTs, one of them is iron pentacarbonyl (Fe(CO)5). To date only a few reports exist on the use of Fe(CO)5 to make CNTs, even though it is used commonly in the HiPco process for CNT synthesis. In this dissertation N containing carbon sources have been used in the presence of Fe(CO)5 for the synthesis of N-CNTs. The growth of the N-CNTs was carried out in a tubular furnace in a horizontal quartz tube. The quartz tube was heated (typically 900-1200 oC) in the presence of Fe(CO)5 and nitrogen containing compounds by a carrier gas at a controlled flow rate using one of the two methods: (the bubbling and the injection methods). The black soot material formed in the reaction (i.e. the N-CNT containing material) was characterized by transmission electron microscopy (TEM), Raman spectroscopy, and thermogravimetric analysis (TGA). 1.2 Objectives The objectives of this project are as follows: 3 (i) To synthesis N-doped CNTs by floating catalyst and injection methods using Fe(CO)5 as catalyst. The N source will be nitrogen containing reagents such as acetonitrile. (ii) To analyse the effect of gas flow rate, temperature and N concentration on N- CNT yield and quality. (iii) To determine the most productive conditions for N-CNT synthesis using the above methods. (iv) To characterize the synthesized N-CNTS using Raman spectroscopy, TEM, and TGA analyses procedures. 1.3 Outline of the dissertation Below is an outline of the content of this dissertation: Chapter 1: This chapter gives a review of the motivation for this study and the objectives of the project. Chapter 2: This chapter gives a review of the literature related to carbon nanomaterials such as carbon nanotubes. The structure, physical and chemical properties, synthetic methods, purification procedures and applications of CNTs and in particular N-CNTS will be discussed. Chapter 3: This chapter describes the procedures used to make the N-CNTs. The chapter contains information on the experiments performed and the characterization procedures used to evaluate the yield and quality of the N-CNTs produced. The results are discussed in terms of mechanisms. Chapter 4: In this chapter, conclusions are summarized based on the results obtained with respect to initial objectives. It also includes some recommendation for future studies. 4 1.4 References 1. L. V. Radushkevich, V. M. Lukyanovich, J. Phys. Chem., 1952, 26, 88. 2. P.A. Tenser, I. Echeistova, D. A. Nauk, USSR, 1952, 87, 1029. 3. W. R. Davis, R. J. Slawson, G. R. Rigby, Nature, 1953, 171, 756. 4. L. J. E. Hofer, E. Streling, J. T. McCartney, J. Phys. Chem., 1955, 59, 1153. 5. P. L. Walker, J. F. Rakszawski, G. R. Imperial, J. Phys. Chem., 1959, 63, 133. 6. T. Baird, J. R. Fryer, B. Grant, Nature, 1971, 223, 329. 7. T. Baird, J. R. Fryer, B. Grant, Carbon, 1974, 12, 591. 8. R. T. K. Baker, M. A. Barber, P. S. Harris, F. S. Feates, R. J. Waite, J. Catalysis, 1972, 26, 51. 9. R. T. K. Bake, P. S. Harris, R. B. Thomas, R. J. Wiate, J. Catalysis, 1973, 30, 86. 10. R. T. Baker, R. J. Waite, J. Catalysis, 1975, 37, 101. 11. T. Koyama, M. Endo, Y. Onuma, J. Appl. Phys., 1972, 11, 445. 12. A. Oberlin, M. Endo, T. Koyama, J. Cryst. Growth, 1976, 32, 335. 13. R. T. K. Bake, P. S. Harris, Chemistry and Physics of Carbon, 1978, 14, 83. 14. R. T. K. Baker, Carbon, 1989, 27, 315. 15. M. Endo, Chem. Tech, 1988, 18, 568. 16. M. S. Dresselhaus, G. Dresselhaus, K. Sugihara. I. L. Spain, H. A. Goldberg, Graphite Fibers and Filaments, Springer, Berlin, Germany, 1988. 5 17. J. S. Speck, M. Endo, M. S. Dresselhaus, J. Cryst. Growth, 1989, 94, 834. 18. S. Iijima, T. Ichihashi, Nature, 1993, 363, 603. 19. O. Stephen, P.M. Ajayan, C. Colliex, Ph Redlich, J.M. Lambert, P. Beirnier, P. Letin, Science, 1994, 266, 1683. 20. E. N. Nxumalo, N. J. Coville, Materials, 2010, 3, 2141. 6 CHAPTER 2 2. LITERATURE REVIEW 2.1 Carbon nanomaterials Carbon is the chemical element with atomic number 6 and symbol C. Carbon is placed in group 14 in the periodic table. Thus carbon is a nonmetallic element. It contains 4 electrons in its valence shell, to form covalent chemical bonds. Carbon can either bond to three atoms with sp2 hybridization or bond to 4 other atoms with sp3 hybridization. Sources of carbon can be either organic (oil, coal) or inorganic (dolomites, limestones). Carbon resists oxidation more significantly when compared to iron, nitrogen, and copper at room temperature. At high temperature, metal oxides form carbon monoxide and metal. An example is shown as below: Fe3O4 + 4C(s) → 3Fe + 4CO(g) 2.1.1 Allotropes of carbon Carbon allotropes can be found naturally in the form of, graphite, diamond, amorphous carbon, and carbon spheres as shown in Figure 2.1 [1]. The physical properties of carbon depends on its allotropic form. Under normal pressure, graphite is the thermodynamically stable form of carbon. In graphite, each C atom is bonded trigonally to three others in a plane composed of hexagonal rings, and this gives rise to a two-dimensional network. The forces between each carbon layer are van der Waals forces. Thus graphite has softness and cleaving properties. Electrical conductivity only occurs in the plane of each covalently bonded sheet. Graphite conducts electricity due to the π bond (the π bond forms due to delocalization of one 7 of the outer electrons of each atom). At higher pressure, carbon forms a more compact structure having a higher density than graphite. This allotrope is called diamond. In the diamond structure each C atom is bonded tetrahedrally to four others and it forms a three-dimensional network. Its hardness is due to the strength of the C- C bonds. In terms of resistance to scratching it is the hardest material known. Diamond has different characteristics compared to graphite. Thus diamond is transparent and it is hard whereas, graphite is opaque and very soft. In terms of electrical conductivity diamond has a low conductivity in contrast to graphite which is a good conductor. In addition diamond has a high thermal conductivity under normal conditions. Among all the carbon allotropes graphite has the highest thermodynamic stability. Amorphous carbon is completely anisotropic. 2.1.2 Fullerene Buckminsterfullerene C60 (fullerene) was first discovered in 1985 by Harold Kroto, James Heath, Sean O'Brien, Robert Curl and Richard Smalley at Rice University [2]. The C60 molecule was called a fullerene and named after Buckminster Fuller. He was an American engineer, inventor and designer. One of his well-known inventions was the geodesic dome. A geodesic dome is a spherical shell structure or lattice shell based on a network of great circles lying on the surface of a sphere. Buckminsterfullerene was named due to its resemblance to geodesic spheres. Figure 2.1: Some allotropes of carbon: a) fullerenes (C60, C540, C70 2.2 Carbon nanotubes 2.2.1 Historical introduction In 1952 a clear image of a CNT with Journal of Physical Chemistry [ language that it was published in (Russian). published in 1976 clearly depicted a hollow carbon fiber with a nanometer scale diameter, using a vapour that a single layer of graphene was detected (TEM) images. Later on Endo referred to this image as a single walled nanotub (SWNT). In 1979 the evidence conference of Carbon by John Abrahamson. He described which were produced on carbon anodes during arc nanoparticles were charact synthesized carbon nanoparticle monoxide. The results from X diamond; b) graphite; c) lonsdaleite ); g) amorphous carbon; h) carbon nanotube [1]. a 50 nm diameter was published in the 3]. The discovery remained unknown due to the A paper by Orbelin, Endo and Kayama -growth technique [4]. Additionally the authors mentioned from transmission electron microscopy for CNTs was presented at the 14 CNTs as carbon nanofib discharge. In 1981, ca erized and published by a group of Soviet scientists. They s by the thermocatalytic disproportionation of carbon -ray photoelectron spectroscopy (XPS) and TEM 8 ; d, e, f) Soviet e th Biennial ers rbon 9 confirmed that the multilayer tubes were made from carbon atoms. They suggested that their carbon multilayer tubular crystals were formed by rolling graphene layers into cylinders. In 1987, Tennett of Hyperion Catalysis was issued a U.S patent for the synthesis of cylindrical carbon nanofibres. The carbons had a constant diameter between about 3.5 and about 70 nanometers. In 1991 Iijima reported that he viewed multilayer CNTs in the insoluble material of arc-discharge graphite rods. This discovery of CNTs accelerated independent discoveries [5,6] by Bethune at IBM [7] and Iijima at NEC of SWCNTs, and methods to synthesis them by adding transition metal-catalysts to the graphite rods in the arc discharge synthesis process. The discovery of CNTs remains a contentious issue. Many believe it is Iijima who is the discoverer of CNTs because he brought CNTs to the awareness of the scientific community [8], since he linked their structure to that of fullerenes. 2.2.2 Structural features of CNTs Carbon nanotubes (CNTs) are basically allotropes of carbon with a cylindrical structure. The cylindrical structure of CNTs generates novel properties which make them useful in many applications. Individual nanotubes can be applied in electronic devices, such as biosensors, scanning probe tips, sensitive nanobalances etc [9]. As the price of production decreases, the use of macroscopic nanotube derived materials becomes more feasible. Nanotube materials can be available in many forms: e.g. as compressed random mats of raw or purified soot, filter deposited foils [10], spin coated or solvent cast films etc [11]. It is also possible to tailor the physical and chemical properties of CNTs by modifying their structure for use in various applications. Carbon nanotubes are generally synthesized by the arc evaporation of graphite [12] or by the pyrolysis of hydrocarbons in the presence of metal particles, particularly using 10 the CVD method [13]. The CVD method is particularly suitable for the synthesis of CNTs in large scale. 2.2.3 Types of CNTs The CNT shape, is based on a plane graphite sheet containing a layer of hexagonal carbons, which shows sp2 hybridizations, that has been rolled into a tube Figure 2.2 [14]. Figure 2.2: The structure of graphite [14]. There are generally three types of CNTs have been made Figure 2.3 [15]. (a) Single-walled carbon nanotubes (SWCNTs): SWCNTs have remarkable properties. But SWCNTs are difficult to synthesize in large yield and in high purity. (b) Double-walled carbon nanotubes (DWCNTs): DWCNTs consist of two rolled layers of graphite. (c) Multi-walled carbon nanotubes (MWCNTs): MWCNTs consist of multiple rolled layers of CNTs. 11 a b c Figure 2.3: The different shapes of CNTs. (a) SWCNT, (b) DWCNT, (c) MWCNT [15]. 2.2.4 Properties of CNTs CNTs have attracted the attention of scientists due to their remarkable properties. Some examples are given below: 2.2.4.1 Mechanical properties of CNTs Mechanical properties of bulk materials at the macro level can be transposed directly to the nano level via scaling effects. What gives CNTs their high stiffness and strength? The strength is due to the strength and adaptability of the C-C bonds. The stiffness is due to the phenomenon of hybridization of a C-C bond in hexagonal carbon rings. In graphite the C atoms are all sp2 hybridized. In CNTs, the curvature is due to a blend of sp2 and sp3 hybridization states in the hexagonal C atoms. When the carbon atoms are under stress they can accommodate the stress by increasing bond angles by assuming more sp3 character. Once the stress is removed, recovery to the original configuration occurs [16]. (a) Stress: Stress is the measurement of an average force (F) applied per area (F.m-2 ) The unit of stress is defined as the Pascal (usually associated with Tera, giga, nano units). There are several types of stress: tensile, sheer, compressive and bending. 12 (b) Young’s modulus: Young’s modulus is the measurement of stiffness. A Young’s modulus is a numerical evaluation of Hook's law, i.e. the ratio of stress over strain. In other words it is the measurement of resistance to elastic deformation. Young's modulus is used to determine the length that a material stretches under tension or the load that will make a material buckle under compression. Some examples are depicted in Table 2.1 [17]. Table 2.1: Corresponds to Young’s modulus of some materials [17]. General engineering material Example Young's modulus (GPa) Metal Steel Cu Al 200 115 69 Carbon materials Diamond Graphite Single walled CNTs 1100 1100 1000+ polymers Epoxy Polyester Nylon 6.9 6.9 3-7 Fibers C-glass Graphite fiber Boron filament 69 340-380 410 (c) Elastic modulus of carbon nanotubes Basically, the theoretical Young's modulus of CNTs is predicted to be in the range from 1 to 5 TPa, whereas the best stainless steel and Kevlar values range from 200 and 250 GPa. Respectively, SWCNTs show a ca. 24% elongation of the material at 13 breaking as compared to 2% for Kevlar and 50% maximum for steel. The high value of a SWCNT Young's modulus correlates with the SWCNT density. Because CNTs have a low density, 1.3-1.4 g.cm-3, the specific strength of CNTs can be as high as 4.8 N.m.kg-1 much greater than that of high carbon steel (1.3-1.4 N.m.kg-1 ). 2.2.4.2 Thermal properties (a) Coefficient of thermal expansion: the coefficient of thermal expansion can be explained as a change in the dimension of a material in response to a change in temperature. Carbon nanomaterials normally are characterised by a low co-efficient of thermal expansion. Carbon is used in space craft manufacture where the carbon materials are exposed to temperature extremes [18]. (b) Thermal conductivity: thermal conductivity is the ability of a material to conduct heat. The thermal conductivity of CNTs is in the range of 1800-6000 (W.m1.K-1) which is 10 times more than that of stainless steel [19]. (c) Electronic properties: An electronic property is the ability of a material to conduct electricity. CNTs can in theory be both metallic and semiconducting. Metallic nanotubes can carry an electrical density of 4 x 109 A/cm-2 which is 1000 times greater than that of metals e.g. copper [20]. 2.2.4.3 Chemical properties Nanomaterials have a high surface energy, thus they seek stabilization by many methods. Some nanomaterials, such as SWCNTs are kinetically stable under room temperature conditions. The graphene structure determines the chemical properties of CNTs. 14 2.2.4.4 Graphite structure (a) Graphite 3D system Graphite can be defined as an infinite 3D crystal made of several stacked layers with sp2 hybrididzation. Each layer interacts weakly via van der Waals forces. (b) Graphene 2D system A graphene crystal is an infinite two-dimensional layer consisting of sp2 hybridized carbon atoms. Highly crystalline graphene surfaces usually react with other molecules via physical adsorption (π-π interactions). However, some functional groups such as carboxyl (COOH), amines (NH2), carbonyl (CO) can be anchored at the edges of graphene sheets, which are more the chemically reactive sites. Furthermore, chemical reactivity depends on the termination of carbon at the edges which can either have a zigzag or armchair arrangement. On the other hand to make a graphene surface more reactive, doping with heteroatoms or addition of a high amount of defects or a high degree of curvature is needed. 2.2.4.5 Doping of graphene Doping-induced defects, arising from the introduction of a foreign atom into the graphene sheet is also possible. Doping occurs via substitution of heteroatoms (e.g. N, B) for C atoms. Addition of both dopants increases the chemical reactivity of the graphene surface. Insertion of N adds an extra electron to the surface. The type of defect influences the kind of conduction generated ranging from n-type transport (N substitutional doping) to p-type conduction (substitution of boron atoms into the lattice) [21]. Studies have shown that other elements like Si, P, S can also be introduced into the hexagonal lattice of carbon tubules [22-26]. In summary, the insertion of non-carbon atoms into graphene-like materials can tailor the chemical reactivity and electronic transport of the carbons. 2.2.4.6 Physical properties a) Chirality of CNTs There are generally three different ways to roll graphite sheets to generate a tube. Some of the formed CNTs have mirror symmetry both parallel and perpendicular to the nanotube axis (zigzag and armchair configurati chiral nanotube which d [27]. Zigzag (n,0) Figure 2.4: The different shapes of s on). The other type of CNT is a oes not have any mirror symmetry. It is shown in Armchair(n,n) Chiral (n,m) ingle walled carbon nanotubes [27]. 15 Figure 2.4 16 2.2.5 Synthesis methods to make CNTs 2.2.5.1 Synthesis methods to make SWCNTs There are a range of methods that have been developed to make SWCNTs. One of them is the HiPco process. It is a high pressure method that uses CO as a carbon source. Large scale SWCNT synthesis is possible by using this method [28]. HiPco is a gas phase CVD process operating at 30-50 atm and 900-1100 oC. The carbon source material used is carbon monoxide and iron pentacarbonyl is the catalyst precursor material. Fe clusters are formed in situ from decomposition of the Fe(CO)5 above 300 oC. Condensation of the reduced iron occurs in the gas stream. CO decomposes as shown below: 2CO → CO2+ Cswcnt The SWCNTs can be produced with a purity of 97-mol% SWCNTs and contain 3- mol% Fe [28]. The standard reported production conditions are: 30 atm, 1050 oC, CO volume flow rate at 8.4 L.min-1 and 0.25 torr Fe(CO)5 at 1.4 - 8.4 L.min-1 for 24-72 h. The rate of production of SWNTs under these conditions is ca.10.8 g.day-1 [28] . 2.2.5.2 Synthesis methods to make MWCNTs There are several ways in which MWCNTs can be synthesized. (a) Arc discharge: Arc discharge entails the use of two carbon electrodes in an evacuated chamber. Current passes through the two rods. Ultimately one of the rods evaporates (the anode) and condenses on the other rod (cathode). Carbon is vaporised from a graphite rod which condenses on the opposite rod (which has opposite charge) [29]. This method is used to mainly make MWCNTs. (b) Laser ablation: Laser ablation entails the vaporization of carbon. During vaporization the substrate material absorbs energy which is delivered by a laser beam. 17 The absorbed energy is converted into thermal energy. When evaporation occurs carbon soot will form [30]. Pure CNTs can be formed using this process. (c) Chemical vapour deposition (CVD): The CVD method can easily allow scale up. CVD is used to produce high purity, solid materials. It can make both SWCNTs and/or MWCNTs. The CVD approach involves a chemical reaction, contrary to the physical vapour deposition (PVD) process. CVD is used industrially due to its low cost, flexibility and simplicity. CVD also has the ability to produce high purity ceramics, metallic and semi conducting film at high deposition rates. CVD is a versatile and cost effective system for making CNTs, because the feedstock, a hydrocarbon source, can be used in the solid, liquid or gaseous phase. Further, the reaction can be done on various substrates. Thus, the product can be expected to be available as a thin film, powder or thick coating. Generally the type of CNTs made depends on the synthesis temperature. For instance production of MWCNTs requires lower temperatures (600-900 oC) when compared with SWCNT synthesis (900-1200 oC). A drawback of the CVD techniques is that they produce carbons with structural defects. This implies that the structure of the so-called MWCNTs is far from the ideal rolled-up hexagonal carbon ring lattice shown in figures and cartoons. There are a range of CVD processes that have been used to make CNTs. These include high or low pressure aerosol assisted CVD (AACVD). Here the precursors are transported to the substrate by means of liquid injection. In the direct liquid injection CVD (DLICVD) method, the precursor is a liquid or a solid (which is dissolved in a liquid) and the liquid is injected into a reactor. The precursor vapour is transported to the substrate. A microwave plasma-assisted CVD (MPCVD) procedure has also been used to make CNTs. There are two generic approaches to the CVD synthesis methodology. The one is to pass the carbon source in the gas phase over a supported catalyst. In the second approach, both the catalyst and carbon source are gaseous. This procedure is called the floating catalyst method. In the second method volatile organometallic catalysts 18 can be used. The prototype catalysts are ferrocene Figure 2.5 a and Fe(CO)5 Figure 2.5 b [31]. Figure 2.5: Catalyst used for CNTs synthesis; (a) ferrocene and (b) Fe(CO)5 [31]. The first use of ferrocene in the synthesis of shaped carbon nanomaterials appeared in 1985 described by Komatsu and Endo [32]. In 1993, a report by Tibbets discribed the use of ferrocene in carbon nanofiber growth [33]. Since then many reports have been written on the use of ferrocene to catalyze carbon shaped nanomaterials from a different range of hydrocarbons. The factors that make ferrocene popular to apply are; price, accessibility, volatility, and stability in air. Other organometallic catalysts have also been used to make CNTs, but their use has not been as extensively studied. An example is Fe(CO)5. Fe(CO)5 is used in the HiPco process to produce CNTs [34]. Fe(CO)5 like ferrocene contains Fe and carbon. A major difference is in their oxidation states. Fe(CO)5 has a zero oxidation state whereas, ferrocene formally has the +2 oxidation state. They also differ in their physical states at room temperature: Fe(CO)5 is in a liquid while ferrocene is a solid. 19 2.2.6 CNT purification processes: CNTs prepared by arc discharge, laser ablation and CVD methods contain many impurities such as amorphous carbon, fullerenes, graphite particles and the metal catalyst. The presence of impurities might affect the performance of CNTs. Hence CNTs must be purified, typically by gas phase oxidation [35-37], liquid phase oxidation [38-44] or physical separation procedures [45-48]. (a) Gas phase oxidation: Gas phase oxidation is achieved by the CNTs reacting in air, pure oxygen or chlorine atmospheres at 500 oC [36]. Gas oxidative treatments can burn off more than 95% of the nanotube materials [35]. (b) Liquid phase oxidation: Liquid phase oxidation can be carried out simply by dipping the sample into strong acids such as concentrated HCl, HNO3 and H2SO4. To improve this procedure, mixtures of H2SO4, HNO3 (3:1) or other strong oxidation agent like KMnO4, HClO4 and H2O2 can be used. Experimental data shows that oxidation of nanotubes in either HNO3 or H2SO4 is slow. Once a mixture of both is used it improves the yields. The best oxidation of CNTs is achieved with KMnO4 in an acidic solution. It was demonstrated by Hernadi et al. [41] that oxidation by KMnO4 in a suspension of acid provides nanotubes free of amorphous carbon. The purification process also opens up the tips of nanotubes leading to the formation of carbonyl and carboxyl groups on the surface of the CNTs. This can improve the dispersion of CNTs in polymers, because such functional group enhances interfacial interaction between the tubes and the polymer [49]. 2.2.7 Doping of carbon nanotubes Shaped carbon nanomaterials (SCNMs) have remarkable properties. They are used in the computer industry due to their small size [50-52]. In addition, carbon nanomaterials can be used in catalysis as catalyst support materials or as a catalyst 20 [53-56]. The range of their applications, can be enhanced by altering their physical- chemical properties. Altering their properties has become a hot topic nowadays. To achieve this goal, SCNMs can be doped by heteroatoms like nitrogen and boron. Doping is the introduction of a foreign atom into the backbone of a CNT. A CNT doping reaction with N was first reported by Stephen et.al. in 1994 [57]. They doped CNTs with nitrogen using the arc discharge method. The incorporation of nitrogen, with an extra electron compared to carbon, can enhance field emission properties of the carbons [58]. Doping also leads to n-type semiconductor behaviour due to the incorporation of N into CNTs. This leads to an improvement in conductivity [59] and an improvement of field emission properties of CNTs [58]. These improvements are due to the electron donor ability of the nitrogen. For instance, the lone pair on N leads to the presence of an electron in the conduction band and results in n-type semiconduction [60]. When CNTs are doped with N they usually form bamboo compartment in the tube structure and the size of the compartments are a function of the N content. N insertion also causes defects in the walls of the tubes which is desirable in terms of functionalisation. To date, many applications of N-CNTs have been reported such as their use in bio sensors [61], in biological and chemical applications [62-64] and in Li-ion batteries [65]. 2.2.7.1 Doping configurations Incorporation of nitrogen into CNTs can occur in several different bonding configurations, in a graphitic network, as shown in Figure 2.6 [66]. The three usual bonding configurations are: (a) pyridinic-like sp2 hybridized 6-fold ring arrangement, (b) pyrrolic-like sp2 hybridized 5-fold ring arrangement and (c) a quaternary substitutional, sp3 hybridized 6-fold ring arrangement. Experimental evidence on the type of N configuration can be achieved via Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). In a Raman spectrum, the prydinic configuration can be confirmed by the intensity ratio of the G to D bands of 21 N-MWCNTs [69]. From XPS analysis both prydinic and substitutional nitrogen atoms in the hexagonal structure have been revealed [70]. Other configurations can be formed in CNT structures e.g. CN (Figure 2.7 d), an amine NH2 (Figure 2.7 e), or as N2 intercalated between the graphitic layers[69-72]. Depending on the CNT application, a decision as to which form is most suitable must be made. In the catalytic oxygen reduction reaction (ORR) process, graphitic type N doping showed a better performance compared to others [73,74]. Figure 2.6: Possible bonding configurations for N in graphitic networks: (a) pyridine-like N, (b) pyrrole-like N, (c) substitutional N, (d) nitrile - CN, (e) amine - NH2, (f) single N pyridinic vacancy, (g) triple N pyridinic vacancy, and (h) interstitial N [66]. 2.2.7.2 Defects induced by nitrogen doping Nitrogen insertion generates defects in the graphitic structure of CNTs. This is shown in Figure 2.7 [66]: there are vacancies or defects in the lattice [Figure (2.6 f) or (2.6 g)]. In addition there is a correlation between diameter and type of defect with strain energy induced from curvature of the graphene sheet. Two ranges of diameter have 22 been studied to provide information on this effect i) a CNT with a diameter less than 8 nm, ii) a CNT with a diameter greater than 8 nm. When a nitrogen defect occurs, the C atoms will move towards the vacancies and form two pentagons sharing the same vertex. For diameters greater than 8 nm, the reconstruction occurs more easily [75]. 2.2.7.3 N-CNT structure The structure of N-CNTs have been studied by transmission electron microscopy (TEM). TEM revealed compartmentalized formation within a tube that is referred to as a bamboo structure. This unique morphology associated with the distortion and bending of graphene layers around nitrogen defects, gives a bamboo like structure Figure 2.7 [76]. It is also proposed, that the outer diameter of undoped CNTs are smaller in diameter when compared to N-CNTs. Figure 2.7: TEM images of MWCNTS doped with nitrogen, produced by thermolysis of ferrocene-benzylamine solutions at 850 oC: (a) low-resolution image of compartmentalized structure of N-CNTs, (b) high-resolution image of an individual bamboo compartment [76]. 23 2.2.7.4 Growth mechanism (a) Growth mechanism for undoped CNTs Undoped MWCNTs consist of several graphene layers that pack around one another. They form a hollow cylindrical structure and terminate by end caps [77]. To satisfy Euler’s theorem these caps must include at least six pentagons in a hexagonal network Figure 2.8 [77]. Figure 2.8: The position of pentagons in a CNT cap [77]. In some cases, tubes with open ends can be observed. This relates to a sudden growth termination of the tubes due to a sudden change in conditions. These species are unable to form hexagons [78]. A number of studies have been reported on the growth mechanism of CNTs on a supported catalyst [79-94]. Ultimately they all postulate that the mechanism can occur in two different ways: either base or tip growth as shown in Figure 2.9 [95], depending on the strength of the interaction between catalyst particles and support (metal-support interactions). When a strong interaction exists base growth occurs and vice versa [96]. It is also believed that both mechanisms can occur simultaneously providing both interactions exist in the same catalyst. 24 In the tip growth mechanism (where the catalyst sits at the tip of CNT) typically the precursor molecules adsorbs on the surface of the catalyst. C atoms are attached to the surface of the catalyst and C atoms can diffuse over the surface. Ultimately C atoms precipitate on the opposite surface to form a CNT wall. Figure 2.9: Schematic of tip and base growth [95]. (b) Growth of N-CNTs To date, a number of mechanistic models for N-CNT growth have been proposed [97]. No conventional nanotube growth models can explain the regular internal bamboo cavities of nitrogen doped MWCNTs. An alternatively growth model proposed for N-doped tubes is shown in Figure 2.10 [98]. In this method the metal catalyst ejects C and N atoms from the metal particle due to stress accumulated in the Tip growth Base growth 25 catalyst surface. The formation of the cavities is due to the presence of nitrogen. As the number of layers increases this pressurizes the catalyst particles until material is again ejected from the cavity and the process once gain starts up. This model is supported by the observation that a decrease in the size of catalyst particles decreases the bamboo periodicity [99]. Figure 2.10: Growth mechanism proposed for N-doped MCWNT bamboo or ‘nanobell’ structures. (a) before precipitating carbon layers matching the particle shape (b). The addition of new precipitate layers (b) increases pressure on the catalyst particle (c) until it is ejected from the inside of the cup (d). When the strain is relieved the process can repeatedly (e-f) [98]. 2.2.7.5 Synthesis of nitrogen doped CNTs Two synthetic methods have been proposed to make N-CNTs i) insertion of nitrogen into CNTs during the reaction [100-102] and ii) post functionalization of CNTs with nitrogen using organic compounds [103-105]. In the first method, arc discharge [106,107] and laser ablation [102,108] procedures have been used. These techniques 26 often require a high temperature, typically > 1000 oC. An alternative method, which can function at lower temperatures in the presence of an organometallic catalyst and numerous C/N sources is chemical vapour deposition [101,109,110]. The quantity of nitrogen incorporated into the graphene layers of the CNT is challenging. High C/N ratios have been obtained using laser ablation, arc discharge methods and magnetron sputtering (high temperature) techniques (up to 33% N incorporation) [102,106- 108,111,112]. Carbons, with up to 20% N incorporation have been achieved with CVD and pyrolysis methods [113-125]. The high temperature techniques to make N-CNTs gives poor quantitative performance compared to these produced by the CVD method [126]. The CVD method can give high yields of both CNTs [127,128] and N-CNTs [129], typically 20-25 g per gram of catalyst. In the second method N is added to an already made CNT. Thus, the N is deposited on the outer walls of an already made CNT. Regarding the second method, the major drawback is its complexity due to the multi step synthesis required. To attach an amine group to CNTs requires to be oxidized first. Once it is oxidized it then reacts with amines to obtain N-CNTs. 2.2.7.6 Consequences of N-insertion Incorporation of nitrogen into CNTs influences the electrical and structural properties of the CNTs. When a CNT is doped with N it generates defects in the graphene layer of the CNTs. The surface is not as smooth as before and attaching other functional groups to CNTs can take place more easily when compared to undoped CNTs. Physical properties of C, N and O atoms are compared in Table 2.2 [130]. 27 Table 2.2: Physical properties of carbon, nitrogen and oxygen [130]. aaccording to Pauling, bin aromatic compounds. element Atomic radius (Å) Electronegativitya C-bond lengthb (Å) Bond type C 0.70 2.55 1.38 C-C (benzene) N 0.65 3.04 1.34 C-N (pyridine) O 0.60 3.44 1.36 C-O (phenol) From Table 2.2 it can be seen that the bond lengths in aromatic structures containing C-C, C-O and C-N bonds do not differ significantly. The table also indicates that incorporation of N and O atoms in the graphene layers of CNTs is suitable. On the other hand the bond length of a C-N bond is shorter than C-O and C-C bond lengths. The N atom distorts the perfect order in a graphitic matrix. It has also been shown that a low N content (N/C < 0.17) has little effect on a graphene layer and no major change in graphitic properties and morphology of the tube are observed in this case [130]. 2.2.7.7 CNTs in biological applications Carbon nanotubes are a useful material to immobilize biomolecules [131]. CNTs can be used in biosensors, drug delivery systems and anti cancer agents, as most biomolecules can adsorbe on CNTs. The great compatibility of CNTs with biomolecules suggested that they can be functionalized with proteins. Each protein can adsorb individually, strongly and noncovalently along the length of CNTs [132]. Proteins such as, ferritin [133-135], cytochrome [136-137] and polysaccharides such as starch [138,139] have been successfully immobilized onto CNTs. 28 (a) Functionalisation of N-CNTs with proteins N-CNTs possess some distinguishable properties compared to CNTs. For example, they exhibit conductance that is independent of tube chirality [140]. The N doped CNTs are less toxic than CNTs [141], and contain defects which make them provide hydrophobic nitrogen sites [142]. The nitrogen groups can exhibit electron transfer between mettaloproteins and Au electrodes [143-144]. (b) N-CNTs in biosensing The whole potential of N-CNTs for biosensing has not been explored yet. But, it is predicted that the N-CNTs ability to be used will be due to its defects. The nitrogen is distributed inhomogeneously along a tube. Pyridinic nitrogen is useful in terms of binding to incoming species [142-145] and it also creates active sites along the tube wall [146]. In addition it can facilitate the electron transfer between mettaloproteins and an Au electrode. 2.2.7.8 Role of CNTs and N-CNTs in catalysis (a) Metal supported CNTs In general, a good catalyst must possess surface active sites essential for adsorbtion of the reactants, bond breaking, bond formation and desorption of the products. In addition, a catalyst must be sufficiently stable to show efficient catalytic activity over a long period of time. In industry metals and metal oxides are widely used as catalysts. Nowadays, the availability of new carbon nanomaterials with their peculiar molecular structures and optoelectronic properties, including carbon nanotubes (CNTs), nanodiamonds, graphene sheets and fullerenes offer new ways of developing advanced carbon-based support. To generate metallic catalysts with high catalytic performance [147-150]. Further, it has been shown that the introduction of 29 heteroatoms (e.g. nitrogen) to carbon nanomaterials leads to electron modulation and this can provide desirable electronic structures for many catalytic processes [147]. There is a report on the use of N-CNTs and N-free CNTs as the support for palladium in the liquid hydrogenation of cinnamaldehyde [151]. It has been reported in this study, that incorporation of nitrogen into the carbon matrix, modified the chemical properties of the support compared to N-free CNTs, leading to a better metal dispersion. The presence of N also significantly improved the selectivity towards the C=C bond in the hydrogenation reaction. Different particle sizes on CNT and N-CNT catalyst supports were observed. Nitrogen atoms on the surface of N-CNTs can provide a surface to increase the anchoring sites on carbon for adsorption of palladium precursor salts. Therefore these anchoring sites could improve the nucleation of the active phase and prevent the metal atom surface mobility resulting in particle sintering during thermal treatments. Liu et al. [152] have been reported that platinum adsorption was favoured on N-CNT materials compared to the pristine CNTs. They also observed that Pt was adsorbed on those carbons neighboring the N atom. It implies that N atoms did not bond directly to the metal, but the N atoms only acted as donor like atoms and not directly as binding sites for the deposited metal active phase. The nitrogen atoms located inside the carbon network are likely to play an intermediate activator role for the metal adsorption on the carbon site. In summary, the catalytic behavior of metals has been tested on both CNT and N- CNT supports to evaluate the role of the support on the catalytic activity. The results showed that N-free CNTs demonstrated the lower hydrogenation activity compared to N-CNTs. This could be attributed to different dispersion of the metal particles on the support surfaces. It is noteworthy, to consider that even the type of N species in the N-CNTs play an important role in catalyst binding and it is believed that graphitic type N-CNTs show the best results [153]. 30 (b) Metal-free carbon catalysts A metal based catalyst suffers from several disadvantages, including high cost, low selectivity, poor durability and detrimental environmental effects caused by catalyst residue and/or undesirable side-products [147]. Thus it is highly desirable to design inexpensive, metal free catalysts that have good activity and high performance over a long period of time. CNTs have been proposed as ideal catalysts for some chemical reactions due to their remarkable properties these include corrosion resistance, environmental acceptability, and unique surface properties. Glassy carbon and activated carbons have been used for certain chemical and electrochemical processes for a long time [154,155]. Carbon-based metal-free catalysts, can be involved in reduction and oxidation reactions [147,156]. In the case of nanostructured carbon nanomaterials (e.g. CNTs) the weakest adsorption sites are associated with the basal planes and consist of the hexagonal arrangement of carbon atoms. The high energy sites exist at the CNT edges and are often saturated with hydrogen atoms that can be replaced by other heteroatoms like, oxygen or nitrogen. These atoms provide strong chemical reactivities for redox or acid-base reactions. A carbon catalyst showed a good potential in ODH reactions of saturated hydrocarbons [157-162]. In particular many carbon nanostructures such as CNTs [159], carbon nanodiamonds [162] and CNFs [159,160] have been found suitable to catalyze the ODH reaction of ethylbenzene to styrene Figure 2.11 [163]. Some nanocarbon catalysts have demonstrated better catalytic performance and long- term stability when compared with metal oxide catalysts [164]. 31 Figure 2.11: ODH of ethylbenzene [163]. (c) CNTs in fuel cells Fuel cells convert chemical energy directly to electricity [165]. At the anode, a catalyst oxidizes the fuel, usually hydrogen, generates a positive charge and a negative charge electron. The electrolyte which is placed between the cathode and the anode can let ions pass through it. But electrons pass through the wire and generates an electric current. When the ions reach the cathode they reunite with the electrons and the two react with a third chemical, usually oxygen, to form H2O. The catalyst that is used in this system is Pt. During this process a slow oxygen reduction reaction (ORR) occurrs on the Pt cathode and a large amount of Pt is reduced at the anode, due to H2 oxidation. The ORR occurs in two ways, either through a 4e- process or a less efficient two step 2e- pathway. The Pt-based electrode suffers from its susceptibility to time dependent drift and CO deactivation. Nowadays replacement/reduction of the expensive Pt-based electrodes is under investigation. Some examples include electrocatalysts made from a Pt-based alloy, transition metal chalcogenides, carbon nanotube-supported metal particles etc. [166-168]. The performance of N-CNTs or N-CNFs has demonstrated some good electrocatalytic activity [167]. A synthesized N-CNF electrode prepared by the floating catalyst CVD method using ferrocene, and either xylene or pyridine has demonstrated good electrocatalytic activities for ORR via a two-step two-electron pathway [169]. In addition N-CNF electrodes showed a 100 fold increase in catalytic activities for H2O2 + H2O+1/2 O2 (1) CnH2n+2 + 1 2 O2 CnH2n + H2O (2) 32 decomposition under both neutral and alkaline conditions [170]. In a recent report Gong et al. [171] found that vertically aligned N-CNTs could act as efficiently as metal-free ORR electro catalysts in fuel cells. The metal-free N-CNTs were shown to catalyze a 4e- ORR process free from CO poisoning and with greater electrocatalytic activity and better long-term stability than Pt-based electrodes in alkaline electrolytes Figure 2.12 [163]. Aligned N-CNTs provides the high surface area, good electrical and mechanical properties and superb thermal stability for the nanotube electrode to be used in fuel cells under either room temperature, or harsher temperature conditions. Figure 2.12: A 4 e- ORR process [163]. Gong et al. [171] attributed the improvement of catalytic performance to the electron- accepting ability of the nitrogen atoms. Thus nitrogen forms a net positive charge on adjacent carbon atoms in the nanotube plane of the NCNTs. Hence, nitrogen attracts more electrons from the anode to facilitate the ORR. The nitrogen delocalization changes the chemisorptions of O2 (at the n-free CNT surface) to a side-on adsorption on N-CNT electrodes. However N-induced charge transfer from adjacent carbon atoms might lower the ORR potential and the parallel diatomic adsorption could significantly weaken O-O bonding, facilitating ORR at the vertically aligned N-CNTs electrodes. The discovery of this ORR mechanism has led to the development of various other metal-free efficient ORR catalysts for fuel cell applications. O2 + 2H2O + 2e- 4OH- (Four-electron process) O2 + 2H2O + 2e- 4OH2 - + OH- (Two-electron process) H2O + HO2- + 2e- 3OH- 33 Furthermore, the role of nitrogen influences the design/development of new catalytic materials for fuel cell and other applications [172-179]. It is noticeable that nitrogen plays an important role as an active site for nanocarbon ORR catalysts and this nitrogen can be found in three different forms (pyrrolic, pyridinc, graphitic). Chen et al. [173] produced N-CNTs from different N-sources using a single step CVD process. They found that N-CNTs with higher nitrogen content and more distortion exhibited better ORR performance in acidic electrolytes. On the other hand Kundu et.al. [174] prepared N-CNTs at different temperatures. They observed, more pyridinic nitrogen at 550 oC which showed better ORR electrocatalytic activity that the one prepared at 750 oC. This study revealed that more graphitic than pyridinic nitrogen atoms are necessary for the ORR. It is still a challenge to determine the exact location of nitrogen atom in carbon nanotubes. 2.2.7.9 Characterization techniques used for carbon nanotubes (a) Transmission electron microscopy Transmission electron microscopy (TEM) can be used to understand matter at dimensions of roughly 1 to 100 nm. TEM experiments can assist on understanding of carbon in terms of modeling, imaging, measuring and manipulating matter at the nano scale. TEM can be used to analyze, metals, alloys, ceramics, polymers, semiconductors and composite mixture of these materials [180]. TEM operates on the same basic principle as light microscopy. The major difference is an electron is used instead of light. A light microscope is limited due to the wavelength of light. TEM uses electrons with shorter wavelengths. Therefore it is plausible to get a thousand times greater resolution with electron microscopy than using normal light microscopy [181]. The limitations of TEM are, i) interpreting transmission images, TEM represents an image with 2D system obtained from a 3D specimen. One of the problems is that when a 2D image is viewed the image does not reveal depth ii) electron beam damage 34 of the sample due to ionization. Ionization has a detrimental effect on a specimen and it could damage the sample, particularly polymers. Nowadays one the solutions is to use a minimum energy dose, cooling procedures and, charge coupled device (CCD) cameras [180]. The sample should be electron transparent. It means that the specimen must be thin enough to transmit electrons, until enough electrons can be detected. (b) Raman spectroscopy Raman spectroscopy is a useful and non-destructive technique. It is widely used to identify the graphitic nature of CNTs. Raman spectroscopy is very sensitive to structural disorder that breaks the graphitic lattice translational symmetry as it measures features in small crystals [182]. It is also sensitive to strength, length and the arrangement of chemical bonds in a sample but less sensitive to chemical composition. A Raman effect arises when a beam of intense monochromic light passes through a sample which contains molecules that change their polarizability as they vibrate. The position of the bands change according to the structures. For CNTs characteristic peaks are found in three spectral regions. The radial breathing mode (RBM) peaks occur around 120-250 cm-1 and indicate the diameter of SWNCNTs. The G-band is characterized by peaks around 1580 cm-1 and indicates the graphitic nature of CNTs. The D-band peak occurs around 1350 cm-1 and corresponds to disorder features in graphitic sheets. The overall quality of a carbon material can be identified by the ID/IG ratio, where ID = intensity of the D band and IG = intensity of the G band. The higher the ID/IG ratio, the lower the graphitic structural quality of the material [183]. (c) Thermogravimetric analysis The TGA technique investigates changes in the weight of a sample as a function of temperature. Such a technique relies on three factors: weight loss, temperature, 35 temperature change [184]. Typically all weight loss curves look similar. To avoid any misinterpretation, derivative weight loss curves can be used. The derivative curve identifies the exact point at which weight loss is most apparent. From TGA curves, data on the kinetics, thermodynamics of chemical reactions, reaction mechanisms, reaction intermediates and final reaction products are obtained [185]. TGA is used to determine the thermal stability of CNTs. Furthermore the presence of a catalyst residue and the absorbed moisture content in a sample can be checked by TGA. 2.3 Toxicity of carbon nanotubes The toxicity of CNTs remains as a controversial subject and more research is still needed. Due to the great role of CNTs in industry it is vital to consider the health hazards of these new developed materials. It is believed that the needle like fiber shaped CNTs are similar to asbestos fibers. Thus, exposure to CNTs for a long time would cause mesothelima, which is a cancer of the lining of the lungs [186]. But the toxicity of these material depends on parameter data, structure, size distribution, surface area, surface charge and purity of samples. The presented data suggest that under some conditions nanotubes can pass membrane barriers. When raw materials reach the organs they generate some harmful effects e.g. inflammatory and fibrotic reactions [186]. Warheit et al. studied the toxicity of SWCNTs in rats by examining the ability of these nano particles to generate pulmonary inflammation as well as alter lung cellular proliferation [187]. They observed pulmonary granuloma formation following SWCNT exposure. In contrast to quartz particles, SWCNTs resulted in multifocal pulmonary granumola formation without any evidence of ongoing pulmonary inflammation or cellular proliferation. Warheit et al. found that the ability of SWCNTs to cause granulo matus lesions by a unique mechanism within the lung may relate to the unique physicochemical properties of SWCNTs [187]. In addition, the unique structural properties of SWCNTs make them resistant in biological and 36 ecological systems. The bio-persistancy of SWCNTs may become a significant health safety issue. 2.4 Conclusion This chapter has indicated the background information to the experimental study performed in this dissertation. What is clear from the study is that the use of organometallic floating catalysts to make CNTs has been dominated by the use of ferrocene and its derivatives. Further, while Fe(CO)5 is used in the HiPco process to successfully make CNTs, its use to make N-CNTs has not be studied. Only two reports have appeared in the literature in which Fe(CO)5 and pyridine were used [188,189]. The study that follows is based on the following questions: 1) Can N-CNTs be made using Fe(CO)5 as a floating catalyst and acetonitrile as nitrogen source? 2) If the N-CNTs can be made, can their purity and morphology be controlled? It is to be noted that this study is a synthesis study. The N-CNTs synthesized have not been used (in this study) as catalyst supports. This will be done in later work. 2.5 References 1. www.wikipedia.org/wiki/carbon, 29 May 2011. 2. H. W. Kroto, J. R. Heath, S. C. O'Brien, R. F. Curl, R. E. Smalley, Nature, 1985, 318, 162. 3. I. B. Radushkevich, J. Phys. Chem., 1952, 26, 88. 4. A. Oberlin, J. Cryst. Growth Des., 1976, 32, 335. 5. D. S. Bethune, C. H. Klang, M.S. de Vries, G. Gorman, R. Savoy, J. Vazquez, R. Beyers, Nature, 1993, 363, 605. 37 6. S. Iijima, T. Ichihashi, Nature, 1993, 363, 603. 7. www.almaden.ibm.com/st/past_projects/nanotubes, 29 May 2011. 8. M. Monthioux, V. Kuznetsov, Carbon, 2006, 44, 1621. 9. R. H. Baughman, A. A. Zakhidov, W. A. De Heer, Science, 2002, 297, 787. 10. A. G. Rinzler, J. Liu, P. Nikolaev, C. B. Huffman, F. J. Rodriguez-Macias, P. J. Boul, A. H. Lu, D. Heymann, D. T. Colbert, R. S. Lee, J. E. Fischer, A. M. Rao, P. C. Eklund, R. E. Smallery, J. App. Phys., 1998, 67,29. 11. T. V. Sreekumar. T. Liu, S. Kumar, L. M. Ericson, R. H. Hauge, R. E. Smalley, Chem. Mater., 2003, 15, 175. 12. S. Iijima, T. Ichihashi, Nature, 1993, 363, 603. 13. A. Cassell, J. A. Rraymarkers, J. Kong, H. Dai, J. Phys. Chem., 1999, 103, 6484. 14. www.wikipedia.org/ Carbon _nano_tubes. 03 March 2010. 15. www.nano.org.uk/Carbon_nano_tubes. 22 May 2010. 16. E. W. Wong, P.E. Sheehan, C. M. Lieber, Science, 1997, 277, 1971. 17. G. L. Hornyak, J. J. Moore, H. F. Tibbals, J. Dutta, Fundamental of nanotechnology, CRC, New York, USA, 2009 . 18. www.stormcable.com/thermal_ expansion_data_table. 28 February 2010. 19. J. Hone, M. Whitney, C. Piskoti, A. Zettl, Phys. Rev., 1999, 59, 2514. 20. A. Bilusic, J. Lukatela, J. Dolinsek, Strojarstvo 2006, 48, 13. 21. M. S. Dresselhaus, M. Terrones, A. Jorio, M. Endo, A. Rao, Y. Kim, T. Hayashi, H. Terrones, J. Charlier, G. Dresselhaus, Materials today magazine, 2004, 7, 30. 38 22. J. Campos-Delgado, I. O. Maciel, D. A. Cullen, D. J. Smith, A. Jorio, M. A. Pimenta, H. Terrones, M. Terrones, ACS. Nano, 2010, 4, 1696. 23. E. Cruz-Silva, F. López-Urías, E. Mu˜noz-Sandoval, B. G. Sumpter, H. Terrones, J. C. Charlier, V. Meunier, M. Terrones, ACS. Nano, 2009, 3, 1913. 24. M. Terrones, A. Jorio, I.O. Maciel, J. Campos-Delgado, E. Cruz-Silva, M. A. Pimenta, B. G. Sumpter, V. Meunier, F. López-Urías, E. Mu˜noz-Sandoval, H. Terrones, Nano Lett., 2009, 9, 2267. 25. J. M. Romo-Herrera, D. A. Cullen, E. Cruz-Silva, D. Ramírez, B. G. Sumpter, V. Meunier, H. Terrones, D. J. Smith, M. Ter-rones, Adv. Funct. Mater, 2009, 94, 1193. 26. L. Novotny, A. Jorio, I. O. Maciel, N. Anderson, M. A. Pimenta, A. Hartschuh, H. Qian, M. Terrones, H. Terrones, J. Campos-Delgado, A. M. Rao, Nature. Mater., 2008, 7, 878. 27. www.wikipedia.org/carbon _nano _tubes. 03 March 2010. 28. M. J. Bronikowski, P. A Willis, D. T. Colbert, K. A. Smith, R. E. Smalley. J. Vac. Sci. Tech., 2001, 19, 1800. 29. Ph. Redlich, M. J. Loeffler, P. M Ajayan, J. Bill, F. Aldinger. J. Chem. Phys., 1996, 260, 465. 30. A. Thess, R. Lee, P. Nikolaev, H. J. Dai. Science, 1996, 273, 483. 31. S. D. Mhlanga, K. C. Mondal, R. Carter, M. J. Witcomb, J. Organo. Chem., 2008, 693, 2942. 32. Y. Komastsu, M. Endo, Japanese patent 60-32818, 1985. 33. G. G. Tibbetts, D. W. Gorkiewicz, Carbon, 1993, 31, 809. 39 34. P. Nikolaev, M. J. Bronikowski, R. K. Bradly, F. Rohmund, D. T. Colbert, K. A. Smith, R. E. Smalley, Chem. Phys. Lett., 1999, 313, 91. 35. T. W. Ebbesen, P. M. Ajayan, H. Hiura, K. Tanigaki, Nature, 1994, 367, 319. 36. J. L. Zimmerman, R. K. Bradley, C. B. Huffman, R. H. Hauge, J. L. Margrave, J. Mater. Chem., 2000, 12, 1361. 37. Y. S. Park, Y. C. Choi, K. S. Kim, Carbon, 2001, 39, 655. 38. H. Hiura, T. W. Ebbesen, K. Tanigaki, Advacnced Material, 1995, 7, 275. 39. A. Rinzler, J. Liu, H. Dai, P. Nikolaev, C. B. Huffman, F.J. Rodriguez, P. J. Boul, D. Heymann, D. T. Colbert, R.S. Lee, E. Fischer, A. M. Rao, P. C. Eklund, R. E. Smalley, J. App. Phys., 1998, 67, 29. 40. I. W. Chiang, B.E. Brinson, R. E. Smalley, J. Phys. Chem., 2001, 105, 1157. 41. K. Herandi, A. Siska, L. Thien-Nga, L. Forro, I. Kiricsi, Solid State Ionic, 2001, 141, 203. 42. C. M. Chen, M. Chen, Y. W. Peng, H. W. Yu, C. F. Chen, Thin Solid Films, 2006, 498, 202. 43. S. Porro, S. Musso, M. Vinante, L. Vanzetti, M. Anderle, F. Trotta, A. Tagliaferro, Physica, 2007, 37, 58. 44. C. M. Chen, M. Chen, Y. W. Peng, C.H. Lin, L. W. Chang, C. F. Chen, Diam. Relat. Mater, 2005, 14, 798. 45. S. Bandow, A. M. Rao, K. A. Williams, A. Thess, R. E. Smalley, P. C. Eklund, J. Phys. Chem, 1999, 101, 8839 46. K. B. Shelimov, R. O. Esenaliev, A. G. Rinzler, C. B. Huffman, R. E. Smalley Phys. Chem. Lett., 1998, 282, 429. 40 47. F. Li, H. M. Cheng, Y. T. Xing, P. H. Tan, G. Su, Carbon, 2000, 38, 2041. 48. A. Yu, E. Bekyarova, M. E. Itkis, D. Fakhrutdinov, R. Webster, R. C. Haddon, ACS. Nano, 2006, 128, 9902. 49. S. C. Tjong, Mater. Sci. Eng., 2006, 53, 73. 50. R. H. Baughman, A. A. Zakhidov, W. A. de Heer, Science, 2002, 297, 787. 51. P. Avouris, J. Chem. Phys., 2002, 281, 429. 52. F. Kreupl, A. P. Graham, G. S. Duesberg, W. SteinHögl, M. Liebau, E. Unger, W. Hönlein, Microelectron. Eng., 2002, 64, 399. 53. K. P. de Jong, J. W. Geus, Catal. Rev. Sci. Eng, 2000, 42, 481. 54. P. Serp, M. Corrias, P. Kalck, Appl. Catal, 2003, 253, 337. 55. J. H. Bitter, M. K. van der Lee, A. G. T. Slotboom, A. J. van Dillen, K. P. de Jong, Cat. Let, 2003, 89, 139. 56. N. Keller, N. I. Maksimova, V. V. Roddatis, M. Schur, G. Mestl, Y. V. Butenko, V. L. Kuznetsov, R. Schlögl, Angew. Chem. Int. Ed, 2002, 41, 1885. 57. O. Stephen, P.M. Ajayan, C. Colliex, Ph. Redlich, J. M. Lambert, P. Beirnier, P. Letin. Science, 1994, 266, 1683. 58. R. B. Sharma, D. J. Late, D. S. Joag, A. Govindaraj, C. N. R. Rao, Chem. Phys. Lett., 2006, 428, 102. 59. D. Golberg, Y. Bando, C. C. Tang, T. E. Zhi, Adv. Mater, 2007, 19, 2413. 60. M. Terrones, P. M. Ajayan, F. Banhart, X. Blasé, D. L. Carroll, J. C. Charlier, R. Czerw, B. Foley, N. Grobert, R. Kamalakaran, P. Kolhler-Redlich, M. Ruhle, T. Seeger, H. Terrones, J. Appl. Phys., 2002, 74, 355. 41 61. P. W. Barone’s Baik, D. A. Heller, and M. S. Strano, Nature Materials, 2005, 4, 86. 62. J. C. Carrero-Sanchez, A. L. Elias, R. Mancilla, G. Arrelin, H. Terrones, J. P. Laclette, M. Terrones, Nano. Lett, 2006, 6, 1609. 63. A. L. Elias, J. C. TCarrero-Sanchez, H. Terrones, M. Endo, J. P. Laclette, M. Terrones, Small, 2007, 3, 1723. 64. B. Fragneaud, K. Masenelli-Varlot, A. Gonzalez-Montiel, M. Terrones, J. Cavaille, Chem. Phys. Lett., 2006, 419, 567. 65. Z. Zhou, X. Gao, J. Yan, D. Song, M. A. Morinaga, Carbon, 2004, 42, 2677. 66. Y. T. Lee, N. S. Kim, J. Park, J. B. Han, Y. S. Choi, H. Ryu, H. J. Lee, Chem. Phys. Lett., 2003, 372, 853. 67. J. H. Yang, B. J. Kim, Y. H. Kim, Y. J. Lee, B. H. Ha, Y. S. Shin, S. Y. Park, H. S. Kim, C.Y. Park, C. W. Yang, J. B. Yoo, M. H. Kwon, K. Ihm, H. J. Song, T. H. Kang, H. J. Shin, J. Y. Park, J. M. Kim.J. Vac. Sci. Techol., 2005, 23, 930. 68. M. Nath, B. C. Satishkumar, A. Govindaraj, C. P. Vinod, C. N. R. Rao, Chem. Phys. Lett., 2000, 322, 333. 69. M. Reyes, N. Grobert, R. Kamalakaran, T. Seeger, D. Golberg, M. R¨uhle, Y. Bando, H. Terrones, M. Terrones, Chem. Phys. Lett., 2004, 396, 167. 70. K. Suenaga, M. Yudasaka, C. Colliex, and S. Iijima, Chem. Phys. Lett., 2000, 316, 365. 71. J. Kotakoski, A. V. Krasheninnikov, Y. Ma, A. S. Foster, K. Nordlund, R. M. Nieminen, Phys. Rev., 2005, 71, 205408. 72. H. C. Choi, S. Y. Bae, J. Park, K. Seo, C. Kim, B. Kim, H. J. Song, and H. J. Shin, Appl. Phys. Lett., 2004, 85, 574. 42 73. R. Liu, D. Wu, X. Feng, K. Mullen, Angew. Chem., Int. Ed, 2010, 49, 2565. 74. H. Niwa, K. Horiba, Y. Harada, M. Oshima, T. Ikeda, K. Terakura, J. Ozaki, S. Miyata, J. Power. Sources, 2009, 187, 93. 75. D. Srivastava, M. Menon, C. Daraio, S. Jin, B. Sadanadan, A. M. Rao. Phys. Rev, 2004, 69, 153414. 76. B. G. Sumpter, V. Meunier, J. M. Romo-Herrera, E. Cruz-Silva, D. A. Cullen, H. Terrones, D. J. Smith, M. Terrones, ACS. Nano, 2007, 1, 369. 77. D. Zhou, L. Chow, Appl. Phys, 2009, 93, 9972. 78. S. Iijima, P. M. Ajayan, T. Ichihashi. Phys. Rev. Lett., 1992, 69, 3101. 79. A. C. Dupuis, P. Mater. Sci, 2005, 50, 929. 80. H. Liu, D. Takagi, H. Ohno, S. Chiashi, T. Chokan, Y. Homma, J. Appl. Phys., 2008, 1, 014001. 81. M. Endo, T. Hayashi, Y.A. Kim, M. Terrones, M.S. Dresselhaus, Phil. Trans. Roy. Soc. Lond, 2004, 362, 2223. 82. R.S. Wagner, W.C. Ellis, Trans. AIME, 1965, 233, 1053. 83. R.T.K Baker, Carbon, 1989, 27, 315. 84. H. Kanzow, A. Ding, J. Phys. Rev, 1999, 60, 11180. 85. Y. Ayumu, K. Noboru, M. Watari, J. Phys. Chem, 2002, 106, 51. 86. S.B. Sinnot, R. Andrews, D. Qian, A.M. Rao, Z. Mao, E.C. Dickey, F. Derbyshire, Chem. Phys. Lett., 1999, 315, 25. 87. K. Liu, K. Jiang, C. Feng, Z. Chen, S. Fan, Carbon, 2005, 43, 2850. 43 88. R. Seidel, G.S. Duesberg, E. Unger, A.P. Graham, M. Liebau, F. Kreupl, J. Phys. Chem, 2004, 108, 1888. 89. M. Pérez-Cabero, E. Romeo, C. Royo, A. Monzón, A. Guerrero-Ruíz, I. Rodríguez-Ramos, Catal, 2004, 224, 197. 90. A. Reina, M. Hofmann, D. Zhu, J. Kong, J. Phys. Chem, 2007, 111, 7292. 91. C. Jin, K. Suenaga, S. Iijima, ACS. Nano, 2008, 2, 1275. 92. R. Brukh, S. Mitra, Chem. Phys. Lett., 2006, 424, 126. 93. M. Grujicic, G. Cao, B. Gersten, Appl. Surf. Sci., 2002, 191, 223. 94. Y. Homma, Y. Kobayashi, T. Ogino, D. Takagi, R. Ito, Y.J. Jung, P.M. Ajayan, J. Phys. Chem. 2003, 107, 12161. 95. V. O. Nyamori, S. D. Mhlanga, N. J. Coville, J. Organomet. Chem. 2008, 693, 2205. 96. http://students.chem.tue.nl/World of Carbon_Nanotube. 18 July 2008. 97. K. Suenaga, M. Yudasaka, C. Colliex, S. Iijima, Chem. Phys. Lett., 2000, 316, 365. 98. M. Terrones, A. M. Benito, C. Manteca-Diego, W. K. Hsu, O. I. Osman, J. P. Hare, D. G. Reid, H. Terrones, A. K. Cheetham, K. Prassides, H. W. Kroto, D. R. M. Walton, Chem. Phys. Lett. 1996, 257, 576. 99. S. Trasobares, O. Stéphan, C. Colliex, W. K. Hsu, H. W. Kroto, D. R. M. Walton, J. Chem. Phys. 2002, 116, 8966. 100. M. Terrones, R. Kamalakarana, T. Seegera, M. Rühlea, M. Rühle, H. W. Kroto, D. R. M. Walton, Adv. Mater., 1999, 11, 655. 44 101. H. C. Choi, J. Park, B. Kim, J. Phys. Chem. 2005, 109, 4333. 102. J. Hu, P. Yang, C. M. Lieber, Phys. Rev. 1998, 57, 3185. 103. J. Li, M. J. Vergne, E. D. Mowles, W-H Zhong, D. M. Hercules, C. M. Lukehart, Carbon, 2005, 43, 2883. 104. E. N. Konyushenko, J. Stejskal, M. Trchova, J. Hradil, J. Kovarova, J. Prokes, M. Cieslar, Polymer, 2006, 47, 5715. 105. T. Ramanathan, F. T. Fisher, R. S. Ruoff, L. C. Brinson, Chem. Mater. 2005, 17, 1290. 106. F. Le Normand, J. Hommet, T. Szörényi, C. Fuchs, E. Fogarassy, Phys. Rev., 2001, 64, 2354161. 107. R. Droppa Jr., P. Hammer, A. C. M. Carvalho, M. C. dos Santos, F. Alvarez, J. Non-Cryst. Solids, 2002, 299, 874. 108. S. E. Rodil, W. I. Milne, J. Robertson, L. M. Brown, Appl. Phys. Lett., 2000, 77, 1458. 109. Y. T. Lee, N. S. Kim, J. Park, J. B. Han, Y. S. Choi, H. Ryu, H. J. Lee, Chem. Phys. Lett., 2003, 372, 853. 110. M. Terrones, A. M. Benito, C. Manteca-Diego, W. K. Hsu, O. I. Osman, J. P. Hare, D. G. Reid, H. Terrones, A. K. Cheetham, K. Prassides, H. W. Kroto, D. R. M. Walton, Chem. Phys. Lett., 1996, 257, 576. 111. K. Suenaga, M.P. Johansson, N. Hellgren, E. Broitman, L.R. Wallenberg, C. Colliex, J.E. Sundgren, L. Hultman, Chem. Phys. Lett., 1999, 300, 695. 112. N. Hellgren M. P. Johansson, E. Broitman, L. Hultman, J. E. Sundgren, Phys. Rev., 1999, 59, 5162. 45 113. M. He, S. Zhou, J. Zhang, Z. Liu, C. Robinson, J. Phys. Chem., 2005, 109, 9275. 114. R. Sen, B. C. Satishkumar, A. Govindaraj, K. R. Harikumar, G. Raina Chem. Phys. Lett., 1998, 287, 671. 115. R. Sen, B. C. Satishkumar, A. Govindaraj, K. R. Harikumar, M. K. Renganathan, C. N. R. Rao, J. Mat. Chem., 1997, 7, 2335. 116. X. Wang, H. Liu, Q. Zhao, Y. Li, Y. Liu, F. Lu, J. Zhang, S. Wang, L. Jiang, D. Zhu, D. Yu, L. Chi, J. Phys. Chem., 2002, 106, 2186. 117. M. Glerup, M Glerup, M Castignolles, M Holzinger., Chem. Commun., 2003, 20, 2542. 118. T. Matsui, M. Yudasaka, R. Kikuchi, Y. Ohki and S. Yoshimura, Appl. Phys. Lett., 1994, 65, 2145. 119. A. G. Kudashov, A. V. Okotrub, N. F. Yudanov, A. L. Romanenko, L. G. Bulusheva, O. G. Abrosimov, A. L. Chuvilin, E. M. Pazhetnov, J. Phys. Chem., 2004, 108, 9048. 120. T. Y. Kim, K. R Lee, K. Y. Eun, K. H Oh, Chem. Phys. Lett., 2003, 372, 603. 121. T. Nakajima, M. Koh, Carbon, 1997, 35, 203. 122. R. I. Kvon, G. N. Il'nich, A. L. Chuvilin, V. A. Likholobov, J. Molec. J. Mol. Catal., 2000, 158, 413. 123. Z. Yang, Y. Xia, R. Mokaya, Chem. Mater., 2005, 17, 4502. 124. S. Trasobares, O. Stephan, C. Colliex, W. K. Hsu, H. W. Kroto, J. Phys. Chem., 2002, 116, 8966. 46 125. M. Nath, B. C. Satishkumar, A. Govindaraj, C. P. Vinod, C. N. R. Rao, Chem. Phys. Lett., 2000, 322, 333. 126. H. Dai, Surf. Sci., 2002, 500, 218. 127. M. L. Toebes, J. H. Bitter, A. J. van Dillen, K. P. de Jong, Catal. Today, 2002, 76, 33. 128. M. K. van der Lee, A. J. van Dillen, J. W. Geus, K. P. de Jong, J. H. Bitter, Carbon, 2006, 44, 629. 129. C. Tang, Y. Bando, D. Goldberg, F. Xu, Carbon, 2004, 42, 2625. 130. D. R. Lide, Handbook of Chemistry and Physics 85th edition, CRC, Florida, USA, 2004,. 131. S. C. Tsang, Z. Guo, Y. K. Chen, M. L. H. Green, H. A. O. Hill, T. W. Hambley, P. J. Sadler. Angew. Chem. Int. Ed., 1997, 36, 2197. 132. J. J. Davis, K. S, Coleman, B. R Azamian, C. B. Bagshaw, and M. L. H. Green. Chem. Eur. J, 2003, 9, 3732. 133. R. J. Chen, S. Bangsaruntip, K. A. Drouvalakis, N. Wong Shi Kam, M. Shim, Y. Li, W. Kim, P. J. Utz, and H. Dai. Proc. Natl. Acad. Sci. USA, 2003, 100, 4984. 134. S. Bhattacharyya, C. Sinturel, J. P. Salvetat, M. L. Saboungi, Appl. Phys. Lett., 2003, 86, 113104. 135. R. Prakash, R. Superfine, S. Washburn, M. R. Falvo. Appl. Phys. Lett., 2006, 88, 63102. 136. B. R Azamian, J. J. Davis, K. S. Coleman, C. B. Bagshaw, and M. L. H. Green. J. Am. Chem. Soc., 2002, 124, 12664. 137. Y. Lin, L. F. Allard, and Y. P. Sun. J. Phys. Chem., 2004, 108, 3760. 47 138. A. Star, D. W. Steuerman, J. R. Heath, and J. F. Stoddart. Angew. Chem. Int. Ed., 2002, 41, 2508. 139. O. K. Kim, J. Je, J. W. Baldwin, S. Kooi, P. E. Pehrsson, and L. J. Buckley. J. Am. Chem. Soc., 2003, 125, 4426. 140. H. J. Burch, J. A. Davies, E. Brown, L. Hao, S. Antoranz Contera, N. Grobert, and J. F. Ryan. Appl. Phys. Lett., 2006, 89, 143110. 141. J. C. Carrero-S´anchez, A. L. El´ıas, R. Mancilla, G. Arrell´ın, H. Terrones, J. P. Laclette, M. Terrones, Nano Lett., 2006, 6, 1609. 142. F. Villalpando-P´aez, A. H. Romero, E. Munoz Sandoval, L. M. Mart´ınez, H. Terrones, M. Terrones, Chem. Phys. Lett., 2004, 386, 134. 143. M. J. Eddowes, H. A. O. Hill, J. Chem. Soc. Chem. Comm., 1977, 21, 771. 144. F. A. Armstrong, H. A. O. Hill, and N. J. Walton, Acc. Chem. Res., 1988, 21, 407. 145. S. Peng and K. Cho, Nano Lett., 2003, 3, 513. 146. C. P. Ewles and M. Glerup, J. Nanosci. Nanotechnol., 2005, 5, 1345. 147. D. S. Sue, J, Zhang, B, Frank, A. Thomas, X. Wang, J. Paraknowitsch, R. Schlögl, ChemSusChem, 2010, 3, 169. 148. L. Dai, Carbon Nanotechnology Recent Developments in Chemistry, Physics, Materials Science and Device Applications, Elsevier, Amsterdam, Netherlands, 2006. 149. J. Wu, W. Pisula, K. Müllen, Chem. Rev., 2007, 107, 718. 150. M. J. Allen, V. C. Tung, R. B. Kaner, Chem. Rev., 2010, 110, 132 . 48 151. K. Chizari, I. Janowska, M. Houllé, I. Florea, O. Ersen, T. Romero, P. Bernhardt, M. J. Ledoux, C. Pham-Huu, Appl. Catal., 2010, 380, 72. 152. R. Liu, D. Wu, X. Feng, K. Müllen, Angew. Chem., Int. Ed. 2010, 49, 2565. 153. H. Niwa, K. Horiba, Y. Harada, M. Oshima, T. Ikeda, K. Terakura, J. Ozaki, S. Miyata, J. Power Sources, 2009, 187, 93. 154. J. Muniz, G. Marban, A. B. Fuertes, Appl. Catal, 2000, 27, 27. 155. A. Sarapuua, K. Vaika, D. J. Schiffrinb, K. Tammeveski, J. Electroanal. Chem. 2003, 541, 23. 156. P. R. Schreiner, Chem. Soc. Rev., 2003, 32, 289. 157. A. Schraut, G. Emig, H.G. Sockel, Appl. Catal., 1987, 29, 311. 158. D.S. Su, N. Maksimova, J.J. Delgado, N. Keller, G. Mestl, M. J. Ledoux, R. Schlögl, Catal. Today, 2005, 102, 110. 159. T. J. Zhao, W. Z. Sun, X. Y. Gu, M. Rønning, D. Chen, Y. C. Dai, W. K. Yuan, A. Holmen, Appl. Catal., 2007, 323, 135. 160. G. Mestl, N. I. Maksimova, N. Keller, V.V. Roddatis, R. Schlögl, Angew. Chem., 2001, 40, 2066. 161. N. Keller, N.I. IMaksimova, V.V. Roddatis, M. Schur, G. Mestl, Y. V. Butenko, V. L. Kuznetsov, R. Schlögl, Angew. Chem., 2002, 41, 1885. 162. D. S. Su, N. I. Maksimova, G. Mestl, V. L. Kuznetsov, R. Schlögl, N. Keller, Carbon 2007, 45, 2145. 163. D. Yu, E. Nagelli, F. Du, L. Dai, Phys. Chem. Lett., 2010, 1, 2165. 49 164. J. Zhang, D. A. Sua. Zhang, D.Wang, RC. Hѐrbert, Angew. Chem., 2007, 46, 7319. 165. S. Basu, Recent Trends in Fuel Cell Science and Technology, Anamaya Publishers, New Delhi, India, 2007. 166. G. Che, B. B. Lakshmi, E. R. Fisher, C. R. Martin, Nature, 1998, 393, 346. 167. J. Yang, D. Liu, N. N. Kariuki, L. X. Chen, Chem. Commun. 2008, 3, 329. 168. A. Kongkanand, S. Kuwabata, G. Girishkumar, P. Kamat, Langmuir 2006, 22, 2392. 169. K. Maldonado, K. J. Stevenson, J. Phys. Chem., 2004, 108, 11375. 170. S. Maldonado, K.J. Stevenson, J. Phys. Chem., 2005, 109, 4707. 171. K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Science, 2009, 323, 760. 172. Y. Tang, B. L. Allen, D. R. Kauffman, A. Star, J. Am. Chem. Soc., 2009, 131, 13200. 173. Z. Chen, D. Higgins, H. Tao, R. S. Hsu, Z. Chen, J. Phys. Chem., 2009, 113, 21008. 174. S. Kundu, T. C. Nagaiah, W. Xia, Y. Wang, S. Van Dommele, J. H. Bitter, M. Santa, G. Grundmeier, M. Bron, W. M. Schuhmann, J. Phys. Chem., 2009, 113, 14302. 175. R. Liu, D. Wu, X. Feng, K. Müllen, Angew. Chem., Int. Ed., 2010, 49, 2565. 176. L. Qu, Y. Liu, J. Baek, L. Dai, ACS Nano, 2010, 4, 1321. 177. C. Jin, T. C. Nagaiah, WXia, B. Spliethoff, S. Wang, M. Bron, W. Schuhmann, M. Muhler, NanoScale, 2010, 2, 981. 50 178. X. Wang, J. S. Lee, Q. Zhu, J. Liu, Y. Wang, S. Dai, Chem. Mater., 2010, 22, 2178. 179. J. I. Ozaki, S. Tanifuji, A. Furuichi, K. Yabutsuka, Acta, 2010, 55, 1864. 180. D. B. Williams, C. B. Carter, Transmission electron microscopy a text book for materials science seconnd edition, Springer, New York, USA, 2009. 181. www.Nobleprize.org/education/Physics/microscopes/TEMindex.htm. 03 July 2011. 182. V. Mennella, G. Monaco, L. Colangeli, E. Bussoletti, Carbon, 1195, 33, 115. 183. C. J. Lee, J. Park, J. A. Dean, F. A. Settle, Instrumental method of analysis, sixth edition, New York, USA, 1981. 184. www.Wikipedia.org/Wiki/thermogravimetric_analysis. 03 July 2011. 185. H. H. Willard, L. L. Merritt, J. A. Dean, F. A. Settle, Instrumental methods of analysis, Sixth edition, Wadworth publishing company, California, USA, 1981. 186. www.Wikipedia.org/Wiki/carbon_nanotube. 26 June 2011. 187. D. B. Warheit, B. R. Lawrence, K. L. Reed, D. H. Roach, G. A. M. Reynolds, T. R. Webb, Toxicol. Sci., 2004, 77, 117. 188. V. Bajpai, L. Dai, T. Ohashi, J. Am. Chem. Soc. 2004, 126, 5070. 189. H. Hou, Z. Jun, F. Weller, A. Greiner, Chem. Mater., 2003, 15, 3170. 51 CHAPTER 3 3. CVD SYNTHESIS OF NITROGEN DOPED CARBON NANOTUBES USING IRON PENTACARBONYL AS A CATALYST 3.1 Introduction 3.1.1 Carbon nanotubes (CNTs) A carbon nanotube is a hexagonal array of carbon atoms rolled up into a thin, hollow, long cylinder [1]. Carbon nanotubes exist in two forms i) single walled CNTs (SWCNTs) and ii) multi walled CNTs (MWCNTs). SWCNTs have a diameter range from 1 to 10 nm and are normally capped at their ends. In contrast, MWCNTs are larger in diameter and range from 5 nm to a many tens of nanometers. SWCNTs can also form bundles which are held together by van der Waals forces [2,3]. CNTs have remarkable physical and electrical properties, which were discussed in chapter 2, section 2.2.4. In addition, CNTs possess high surface areas, high aspect ratios and good mechanical properties [4,5]. Despite their phenomenal properties, the hydrophobicity and chemical inertness of CNTs impede their commercial use. To resolve this problem, scientists have come up with methods to modify the surfaces of CNTs. Their surface properties can then be tailored either by covalent or by non-covalent surface modifications. Covalent modification involves introduction of either new elements (e.g. O, N) onto the CNT or by incorporation of organic functional groups (e.g. bio molecules) onto CNT sidewalls. Noncovalent modification, can be done via adsorption of surfactants, biological macromolecules, and polymers without any intrinsic structural change to the CNTs [6]. Both strategies have been used to fabricate modified CNTs. In the covalent approach, incorporation of the elements, such as fluorine, nitrogen and boron leads to a modification of the CNT density of state which changes their semiconducting properties [7]. In the non-covalent approach, only the CNT surface is 52 targeted and their surfaces can be modified by addition of polar functional groups that can improve the dispersion and cross linking capabilities of CNTs during polymerization. In this study the modification of the CNT properties has been achieved by introducing nitrogen atoms into the CNT structure. This process is called doping. The materials once doped are called N-CNTs (nitrogen doped carbon nanotubes). N-CNTs can be prepared by different approaches such as, laser ablation, arc discharge and chemical vapour deposition (CVD). The method which was used in this study was CVD using a floating catalyst method. A floating catalyst method is a common method to fabricate CNTs on a large scale [8,9]. In a floating catalyst method there is no need to utilize a catalyst support, which reduces the cost of CNT production. The type of catalyst required to achieve this method typically is a volatile organometallic complex such as ferrocene (FcH), a substituted FcH (with or without N substituents) or Fe(CO)5 [10]. In this study Fe(CO)5 was used as the floating catalyst. Fe(CO)5 has the advantage than the Fe is in the zero oxidation state and hence, unlike ferrocene and ferrocene derivatives, does not require the presence of a reducing gas such as H2. Further, Fe(CO)5 has been known to be an excellent catalyst for the production of CNTs that are not doped. Indeed the HiPco process is based on the use of Fe(CO)5 in a reactor at high pressure [11]. There are numerous reports on the use of Fe(CO)5 as a catalyst to produce CNTs, but little is known about its ability to produce N-CNTs. An example is a study by Banjapai et.al. [12] who synthesized aligned helical N-CNTs by the co-pyrolysis of Fe(CO)5 and pyridine on a quartz substrate at high temperature (900-1100 oC). A summary of some of the studies that have been reported in the literature is which Fe(CO)5 has been used to make CNTs and N-CNTs is given in Table 3.1. As will be seen, Fe(CO)5 has been successfully used to make both SWCNTs and MWCNTs. 53 Table 3.1: Use of Fe(CO)5 as catalyst to make doped and undoped CNTs [13]. SNCM Reactor design Carbon source Catalyst Temperature Pressure and flow rate References SWCNTs Horizontal CVD reactor; two stage furnace system Acetylene Fe(CO)5 350 oC (furnace 1) 1100 o C (furnace 2) C2H2 flow rate 50 sccm Ar flow rate 1000 sccm (Fe(CO)5 -1 ml) [14] [15] SWCNTs Horizontal CVD reactor CO Fe(CO)5 800-1200 o C CO (1-10 atm) Fe(CO)5 (1-25 mTorr) [16] [17] SWCNTs Horizontal CVD reactor; two stage furnace system CO Fe(CO)5 700 o C (furnace 1) 900 o C (furnace 2) CO flow rate: 800 sccm H2 flow rate: 200sccm [18] MWCNTs Horizontal CVD reactor; two stage furnace system with vapourization chamber and an ice cooler 1:1, Pyridine Toluene Fe(CO)5 1050-1150 o C H2 flow rate: 350-400 ml/min [19] MWCNTs Horizontal CVD reactor pyridine Fe(CO)5 900-1100 o C Ar and H2 [12] MWCNTs Vertical CVD reactor Methane Fe(CO)5 1050-1150 o C CH4 flow rate 125-250 sccm N2 flow rate: 1000-2000 sccm Fe(CO)5 feeding rate 0.5-2.5 mL/h [20] 54 3.2 Experimental 3.2.1 Chemicals and materials Toluene and acetonitrile were purchased from Merck, and iron pentacarbnyl was purchased from Fluka. Acetonitrile was distilled before use. All reactions were performed under an Ar atmosphere. 3.2.2 Characterization 3.2.2.1 Thermogravimetric analysis TGA experiments were performed on a Perkin Elmer Pyris 1 TGA analyzer under an air atmosphere in the temperature range 25-900 oC at a heating rate of 10 oC/min. Peak positions are recorded in units of Celsius. 3.2.2.2 Raman spectroscopy Raman spectra were acquired using a Jobin-Yvon T64000 Raman spectrograph operated in single spectrograph mode. The excitation source was the 514.5 nm line of an argon ion laser. This was focused onto the sample using a long-working distance 20x objective with an Olympus BX40 microscope attachment. The power at the sample was kept low (< 1.2 mW) to minimise possible localised heating of the sample. The backscattered light was collected using the microscope and dispersed via a 600 lines/mm grating onto a liquid nitrogen cooled CCD detector. The data was acquired using Labspec v4.18 software. Accumulation times varied between 2 and 3 minutes, and the confocal pinhole was set to either 2 mm or 0.5 mm, depending on the degree of background fluorescence. 3.2.2.3 Transmission electron microscopy TEM studies were performed by using a FEI Technai G2 Spirit electron microscope at 120 kV. 55 3.2.3 Synthetic procedures: Injection method 3.2.3.1 The synthesis of nitrogen doped carbon nanotubes (N-CNTs) A typical experiment is described below. Synthesis of N-CNTs was carried out at 900 oC under an Ar atmosphere at atmospheric pressure. The flow rate of Ar was kept constant at 300 mL/min. All the precursors were kept under an Ar atmosphere before use. A mixture of toluene (17.5 mL), acetonitrile (3.50 mL) and Fe(CO)5 (1.34 mL) was placed in a 25 mL syringe driven by a SAGE syringe pump. The solution was injected at 0.80 ml/min into a quartz tube reactor (32 cm x 1 m) via a special quartz delivery system cooled by water as shown in Figure 3.1. This cooling system allowed the solution to be injected into the hot region of the quartz tube. The reaction took about 20-25 min. After the reaction was done, Ar gas was passed through the tube at approximately 50 mL/min while the reactor cooled to room temperature, carbon soot was removed from the tube, weighed and kept in an air tight vial prior to characterization. Figure 3.1 shows the reactor which was used. In related studies the acetonitrile to toluene ratio was varied while the Fe(CO)5 content was maintained at 1.3 mL. In these studies the acetonitrile content ranged from 0% to 25%. A summary of the experimental condition used is shown in Table 3.2. Many other variation to the initial reaction were also performed. This included the use of a different temperature (900 oC), flow rate (200 mL/min), and a different carrier gas (Ar + 5% H2). All the condition used are shown in Table 3.2. 56 Table 3.2: Experimental conditions used for the N-CNTs produced. Experiment number CH3CN (%) Temperature (oC) Gas flow rate (mL/min) Carrier gas Yield (g) 1 0 900 200 Ar 0.65 2 0 900 300 Ar 0.21 3 2 850 200 Ar 0.21 4 2 850 200 Ar + 5% H2 0.47 5a 2 850 300 Ar 0.07 6a 2 900 200 Ar 0.06 7 2 900 200 Ar + 5% H2 0.27 8 2 900 300 Ar 0.97 9 5 850 200 Ar 0.85 10 5 850 200 Ar + 5% H2 0.38 11a 5 850 300 Ar 0.05 12 5 900 200 Ar 0.74 13 5 900 200 Ar + 5% H2 0.53 14 5 900 300 Ar 0.76 15 15 750 200 Ar 0.05 16 15 800 200 Ar 0.05 17 15 850 200 Ar 0.26 57 18 15 850 200 Ar + 5% H2 0.38 19a 15 850 300 Ar 0.03 20 15 900 200 Ar 0.66 21 15 900 200 Ar + 5% H2 0.58 22 15 900 250 Ar + 5% H2 0.26 23 15 900 300 Ar 0.40 24 20 900 200 Ar 0.78 25 25 900 200 Ar 0.71 a indicates a 10 mL solution. Figure 3.1 : A furnace and syringe attachment used for the synthesis the N-CNTS. Oven Quartz tube Syringe 58 3.2.3.2 Purification of N-CNTs The synthesized N-CNTs were purified by using either HNO3 (55%) or HCl (35%). The purification was carried out at 80 oC for 6 hours. The solution was cooled to room temperature, filtered and the black powder washed with distilled water several times until the pH of the solution was around 7. The sample was then dried in an oven at 100 oC overnight. 3.3 Results and discussion 3.3.1 Synthesis of nitrogen doped carbon nanotubes Nitrogen doped CNTs with varying degree of abundance and purity were synthesized according to the procedure outlined in the experimental section 3.2.3.1. All the samples were washed with acid prior to TGA and Raman analyses. The advantage of using HNO3 rather than HCl is that, HNO3 leads to surface functionalization, specifically at high temperatures. Further, the use of HCl is not desirable when the CNTs are to be used in catalysis applications. It has been reported that the presence of KMnO4 in acidic solution works better and it generates a pure material [21], however this reagent was not used in this study. TEM studies have shown that the N content affects the tube morphology and bamboo compartment formation. The presence of bamboo compartments is not a sufficient contribution to indicate N incorporation since bamboo-like structures have also been observed in Y-junction CNTs that do not contain N [22]. A TEM study revealed that N incorporation in the CNT structure did form separated compartments. It was reported by Nxumalo et al. [10] that an increase in N % source decreased the compartment distance. This is expected and it agrees with the mechanism proposed for N-CNT growth [23]. N-CNT bamboo compartment separation also increases with an increase in growth temperature [24]. TGA studies confirmed the presence of nitrogen in the CNTs as indicated by the change of decomposition temperatures 59 observed when the CNTs were removed with O2 in TGA experiments. The more nitrogen incorporated the less stable the N-CNTs that were formed. From a Raman study, an intense D band was observed that indicated the presence of disorder in the tubes. The level of disorder can be determined by measuring the ID/IG ratio (see below). Many variables were changed in the experiments to optimize the growth of the N- CNTs. In the sections below the effect of each of these variables on N-CNT yield, morphology and purity are discussed. In each set of data an attempt has been made to vary only one parameter at a time. 3.3.2 Effect of %CH3CN in toluene In this study the following conditions were used: reaction temperature-850 oC, gas flow rate-300 mL/min, gas used Ar. 3.3.2.1 Yield study The effect of the reaction mixture in which the N content was varied by changing the amount of CH3CN used in a CH3CN/toluene mixture on the yield was studied. It was observed, that an increase in N content, decreased the yield. The CH3CN/toluene ratios used are shown in Table 3.2 (enteries: 5, 11, 19, entries: 8, 14, 23 and entries: 1, 12, 20). It was also noted that the gas flow rate affected in the yield (Table 3.2, entries: 20, 23): the yield decreased with an increase in the gas flow rate. The effect of nitrogen containing organic compounds on N-CNT growth was reported by Nxumalo et al. [25]. They used different N compounds (e.g. amines, aromatic compounds). It was revealed that after an increase in amine concentration, the yield of CNTs generally decreased. Similar results were found for aromatic reagents (pyridine) and amides. 60 It is believed that the N concentration plays an important role in terms of yield production. Obviously with an increase in N concentration the yield drops. The reason given is nitrogen slows down the rate of tube growth and when N atoms sit on the surface of the catalyst it retards carbon growth on the catalyst. 3.3.2.2 TEM studies TEM analysis was used to examine the type, quality, size distribution and quantity of N-CNTs produced. TEM analysis allowed a facile measurement of N doping to be determined since doping CNTs with nitrogen leads to the formation of bamboo compartments. An example of a bamboo structure is shown in Figure 3.2. The formation of bamboo compartment can be rationalized by its growth mechanism. Generally, what happens is C and N atoms both deposit on the catalyst surface and when they accumulate on the catalyst surface they will both be ejected from the surface and form a separate layer which leads to the bamboo structure as shown in Figure 3.2. The formation of the cavities is due to the presence of nitrogen on the catalyst surface. (a) Sample analysis TEM images were recorded of samples produced (unpurified) from the three different ratios of CH3CN/toluene studied (temperature-850 oC, flow rate-300 mL/min). The images reflect the type of materials produced. The % of each type of sample was determined by considering many tens of such images. A fair estimate of the material could be established. A summary of the data is given in Table 3.3. As is shown in Table 3.3, the 5% CH3CN sample contains the highest amount of tubes compared to the others. Based on the observation from TEM images it is assumed that the shortest tubes were obtained from the 15% sample (Table 3.2, entry: 11) and theoretically it can be also confirmed because as the amount of nitrogen increases the length of the tubes decreases due to the presence of the N 61 atoms. Incorporation of N in the CNT reduces the rate of tube growth and it also terminates the reaction. Similar observations were seen when FcH/aniline was used to make N-CNTs [26]. Generally the lengths shorten with increased N incorporation. The 2% sample (Table 3.2, entry: 5) resulted in more straight aligned tubes, whereas the 5% sample (Table 3.2, entry: 18) contained bent tubes with a few open tips as shown in Figure 3.3 e, f. The details of the TEM images shown in Figure 3.3 are as follows: a: tubes with a very small amount of amorphous carbon, b: carbon fibers, c: tubes, d: amorphous carbon plus a few tubes, e: tubes, f: tubes with open tips, g: carbon spheres, h: tubes with open tips, i: tubes with bamboo compartments. This variation in morphology is due to the different C:N ratios. Figure 3.2: The TEM image of bamboo structure doped with N. Table 3.3: Samples content as determined by using TEM images. CH3CN content (%) Spheres (%) Amorphous carbon (%) NCNT/CNT (%) 2 0 60 40 5 0 20 80 15 1 29 70 62 Figure 3.3: TEM images of the N-CNT samples produced using CH3CN: 2% (a, b ,c), 5% (d, e, f), 15% (g, h, i) respectively (T: 850 oC, flow rate: 300 mL/min). (b) Outer diameter distribution of the N-CNTs A focus of this project was on the determination of the yield and morphology of the CNTs. The CNT content (only) of the products produced by varying % CH3CN used a b c d e f g h i 63 were studied by TEM. The tubes were viewed by TEM and the outer diameters were measured using the Image J programme. For each sample about 70 tube diameters were measured and these distributions were plotted as shown in Figure 3.4. The trends can be seen in Figure 3.4 and Table 3.4. (a) As the N content increases the tube diameters become smaller. This is in agreement with other literature studies. It has been reported by Nxumalo et al. [25] that the N incorporated in nanotubes made from FcH in benzylamine/toluene had smaller outer diameters but larger inner diameters when compared to undoped CNTs made from FcH/toluene solution under analogous experimental conditions. The same observations were noted when aniline/FcH was used to make N-CNTs [26]. Larger concentrations of the N-source yielded CNTs with smaller diameters. Similar effects were also observed when benzylamine was used as a nitrogen source [27]. (b) At all CH3CN concentration there is a wide range of outer tube diameters. However, an increase in CH3CN content gives a narrower distribution of tubes. Table 3.4: The average outer diameter distributions. CH3CN% Avg. outer diameter distributions (nm) 2 84 ± 30 5 62 ± 30 15 47 ± 10 64 0 1 2 3 4 5 6 7 10 20 30 40 50 60 70 80 90 10 0 12 0 13 0 14 0 16 0 18 0 22 0 Nu m be r o f t ub e Outer diameter of the tubes (nm) 0 5 10 15 20 25 10 20 40 70 100 120 140 170 200 280 Nu m be r o f t ub e Outer diameter of the tubes (nm) a b a 65 Figure 3.4: Graphs of diameter distribution made from 2% (a). 5% (b), 15% (c) of CH3CN, (T: 850 oC, flow rate-300 mL/min). (c) Inner diameter distribution of the N-CNTs The average inner diameters were measured from TEM micrographs (using Image J) shown in Table 3.5 As explained, an increase in acetonitrile concentration did not give a change in the inner diameters of the tubes. It was reported by Nxumalo et al. [25], that the N-CNTs made from FcH in benzylamine/touene had smaller outer diameters but larger inner diameters when compared to undoped CNTs grown from a FcH/toluene under the same conditions. Table 3.5: Average inner diameter distribution of the different CH3CN% concentration. CH3CN% Avg. inner diameter distribution (nm) 2 26 ± 4 5 29 ± 5 15 20 ± 10 0 2 4 6 8 10 12 14 16 18 10 30 50 70 100 120 140 160 190 210 240 Nu m be r o f t ub e Outer diameter of the tubes (nm) C 66 (d) Compartment distance study The average distance between bamboo caps was measured from the TEM images (with Image J) shown in Figure 3.5 for the different samples and average distance are presented in Table 3.6. It was revealed that an increase in CH3CN concentration decreased the compartment distance. This is also shown in Figure 3.6. This result is expected and can be explained by the N-CNT growth mechanism which was described in chapter 2, section 2.2.7.4 (b) [23]. Similar observations by Nxumalo et al. were also seen [26], when they used different concentration of PhNH2 and FcPhNH2 in their study of N doped CNTs. Figure 3.5: TEM images of bamboo compartments and the distance between bamboo caps. Distance between bamboo caps 67 Table 3.6: Distance between N-CNT bamboo compartments. CH3CN% Avg. distance between individual bamboo compartments (nm) 2 28 ± 8 5 22 ± 10 15 9 ± 8 Figure 3.6: The average distance between the compartment caps for N-CNTs made from different CH3CN% (T: 900 oC, flow rate: 200 mL/min). e) Size of spheres The size of spheres were measured using Image J as it shown in Figure 3.7. The average size of the spheres was 330 nm for the 15% sample (Table 3.2, entry: 19). 0 5 10 15 20 25 30 35 40 0 5 10 15 20 Av g di st an ce b et w ee n ea ch co m pa rt m en t ( nm ) CH3CN% 68 Figure 3.7: Synthesized spheres from the 15% sample (T: 850 oC, flow rate: 300 mL/min). 3.3.2.2 TGA analysis Thermal stability tests were performed by thermogravimetric analysis. The initial temperature of the oxidative process can be correlated with the presence of lattice defects. The less the defects, the higher the decomposition temperature. The shape of the derivative curve can give some information about the presence of the carbon byproducts. For example, the presence of amorphous carbonaceous particles are burned off first due to the selective etching leading to a faster oxidative etching rate when compared to CNTs. In addition, if a sample contains water or carboxyl groups then a peak around 100 or 300 oC would appear respectively. If the rapid oxidation of carbon is fast, the mass loss occurs rapidly giving a smooth curve and it also implies a more perfect crystallinity of the sample. Figure 3.8 and Table 3.7 show a comparison of the mass losses of N-CNTs (A = 0%, B = 5%, C = 15%, D = 25%) on heating in an air atmosphere at a rate of 10 oC min-1. In each case the decomposition temperature, determined by measuring the derivatives of the TGA curves, was determined. The example of a derivative curve is shown in Figure 3.9. Diameter of the sphere 69 As the incorporation of N in the CNTs increased the stability of the sample decreased compared to the pristine CNTs. This instability is attributed to the structural disorder introduced by the presence of N into the carbon lattice [28,29]. The TGA data revealed that the N-CNTs thermal stability is less than the undoped one. Similar observation by Nxumalo et al. on the stability of N-CNTs have been reported in the literature [26]. The TGA curves also showed a residual weight (5%). This may relate to the presence of FeOx residues formed in air from oxidation of the residual catalyst contained in the tubes. 0 200 400 600 800 1000 0 20 40 60 80 100 w e ig ht lo ss (% ) T em peratu re (oC ) A B C D Figure 3.8: TGA curves of the purified N-CNTs with different N content (A) 0% , (B) 5%, (C) 15%, (D) 25%. 70 0 200 400 600 800 1000 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 W e ig ht de riv a tiv e (% /o C ) Tem perature (oC) Figure 3.9: The derivative curve of 0% CH3CN. Table 3.7: Thermogravimetric analysis of N-CNTs and CNTs (A-D). CH3CN% Decomposition temperature oC 0 741 5 641 15 693 25 572 71 3.3.2.3 Raman spectroscopy Raman spectroscopy is an excellent tool to investigate the graphitic structure of carbon materials. Raman spectra generally show two prominent peaks: a G band at approximately 1590 cm-1 that corresponds to the Raman active E2g mode and a D band at round about 1390 cm-1 that corresponds to disorder features in graphitic sheets. The overall carbon quality can be determined by the ID/IG ratio. The higher the ID/IG ratio, the lower the graphitic structural quality of the material [26]. It was reported by Koós et al. [30] that the defects increased in the carbon nanotube structure as the amount of benzylamine increased. It was also confirmed in their study, that for ID/IG intensity ratios close to one the incorporation of N corresponded to a decreased crystallinty of the CNTs. In a study performed by Nxumalo et al. [26], they observed N-CNTs, produced from ferrocenylaniline/toluene were more disordered than CNTs produced from FcH/toluene alone. It was revealed from Raman analysis that as the concentration of aniline increased in FcH/aniline/toluene mixtures, the degree of disorder also increased. The impact of nitrogen on the structure of CNT reported in the literature suggests that as the concentration of nitrogen increased, the degree of disorder also increased. This implies that N-CNTs produced from 25% CH3CN should have been more disordered than the CNTs produced from 2% CH3CN. In this study the ID/IG ratio difference between the samples with different acetonitrile concentration is very small. Figure 3.10 and Table 3.8 and a comparison between Raman spectra for the different CH3CN concentrations does not give useful information on the topic of disorder. 72 500 1000 1500 2000 2500 3000 0 2000 4000 6000 8000 10000 In te n si ty Ram an shift (cm -1) A B C D E Figure 3.10: Raman spectra of the N-CNTs with different N content (A) 2%, (B) 5%, (C) 15%, (D) 20%, (E) 25%. Table 3.8: Raman analysis of the N-CNTs. sample ID IG ID/IG 2% 1236 1459 0.84 ± 0.05 5% 2140 2705 0.79 ± 0.05 15% 1261 1660 0.75 ± 0.05 20% 4273 5623 0.75 ± 0.05 25% 2999 3239 0.92 ± 0.05 73 3.3.3 Effect of temperature There are a number of experimental parameters that need to be taken into consideration in terms of N-CNT growth. Temperature is a key parameter for the fabrication of N-CNTs. The N-CNT growth temperatures used in this study to make N-CNTs was varied from 750 oC to 900 oC. In this study, the effect of four different temperatures were studied. These four reactions were carried out under the same conditions (Table 3.2 entries: 15, 16, 17, 20). TEM analysis revealed that at 750 and 800 oC no tubes were synthesized. At 750 oC only fibers (Figure 3.11 a) and agglomerates (Figure 3.11 b) were observed. At 800 oC nanopearl/beads (Figure 3.11 c) and nanoparticles embedded into a polymer (Figure 3.11 d) were observed. Good quality tubes were observed at both 850 and 900 oC (Figure 3.11 e, 3.11 f, 3.11 g, 3.11 h). A sample analysis is displayed in Table 3.9. The materials produced when T < 850 oC were not studied further. It has been reported in some papers that N-CNT synthesis is feasible between 700- 800 oC using other methods or using different catalyst like FcH. Some of these studies are shown in Table 3.10 [22]. Table 3.9: The content of the samples synthesized at different temperature. CH3CN (%) content Temperature (oC) Spheres (%) Amorphous carbon (%) NCNT/CNT (%) 2 750 0 100 0 2 800 0 100 0 2 850 0 20 80 2 900 0 70 30 74 Table 3.10: Showing the use of different C/N sources and catalyst on fabrication of N-CNTs at different temperatures [22]. N/C sources Catalyst T (oC) Method references 4-tert-butylpyridine FcH 700 CVD [31] NH3/pyridine FcH 700- 1000 CVD [32] Triphenylphosphine/benzylamine FcH 720 CVD [33] C2H2/NH3 Fe(CO)5 750- 950 CVD [34] Pyridine, pyrimidine FcH 750 CVD [35] Ethylenediamine Co, FcH 780 CVD [36] Ethanol/benzylamine FcH 800 CVD [37] 75 a b c d 76 Figure 3.11: TEM images of the samples synthesized at 750 oC (a-b), 800 oC (c-d), 850 oC (e-f), 900 oC (g-h) (% CH3CN: 15% flow rate: 200 mL/min). i) Effect of temperature on the outer diameter distribution The study of outer diameter distribution at different temperatures revealed that at both temperatures there is a wide range of tube outer diameters Figure. 3.12. It is noticeable that material synthesized at 850 oC (graph a) shows a better trend of outer diameter distribution and contained narrower tubes compared to tubes made at 900 oC (graph b). e f g h 77 Figure 3.12: Graph of diameter distribution of N-CNTs synthesised at a) 850 oC, b) 900 oC. ii) Effect of temperature on the bamboo compartments From TEM images the average distance between the bamboo caps was measured (100 compartments per sample) using Image J. It was revealed that no major 0 2 4 6 8 10 12 14 16 18 20 10 20 30 40 50 60 75 90 100 110 120 150 190 Nu m be r o f tu be s Outer diameter of the tubes (nm) 0 2 4 6 8 10 12 10 20 30 40 50 60 70 80 100 110 130 240 260 270 340 460 Nu m be r o f t ub es Outer diameter of the tubes (nm) a b 78 difference can be seen as a function of temperature at 850 oC and 900 oC in Table 3.11. Note: no tubes were synthesized at 800 oC. Table 3.11: The avgerage distance between individual bamboo compartment at 850 and 900 oC. CH3CN% Temperature oC Avg. distance between individual bamboo compartment (nm) 15 850 6 15 900 5 3.3.4 Effect of carrier gas Some experiments were performed in order to evaluate the role of the carrier gas type (Ar or Ar + 5% H2) on the product formation. In this instance H2 was added to the Ar carrier gas to evaluate the effect of H2 on the CNTs produced. The presence of H2 should in principle not play an important role when Fe(CO)5 is used. H2 is a reducing agent and is used to keep the metal in its lowest oxidation state. For example in terms of using FcH hydrogen reduces Fe+2 Fe0 but Fe(CO)5 contains Fe in the zero oxidation state so there is no need for a reducing agent. i) The effect of hydrogen on carbon yield The effect of H2 on the reaction was to give a higher yield of product (Table: 3.12). More importantly the addition of H2 also resulted in a change in the morphology of the products produced. As can be seen in Table 3.12 a higher yield of CNTs were produced. 79 Table 3.12: The experimental conditions used for N-CNT production. sample Carrier gas Flow rate (mL/min) Temperature (oC) Yield (g) 15% Ar 200 850 0.26 15% Ar + 5% H2 200 850 0.38 ii) The effect of hydrogen on type of carbon production As mentioned, it was thought that hydrogen would not play an important role in N- CNT production. However the presence of H2 influenced the yield and also the quality of the carbon produced. In this case it reduced the amount of amorphous carbon as shown in Table 3.13. Wasal et al. have carried out a study on the CVD synthesis of MWCNTs and they investigated the role of hydrogen on CNT synthesis utilizing FcH as catalyst and xylene as the carbon source [38]. The study revealed that hydrogen caused a competition between the formation of different compounds (soot, carbon fibers and CNTs) and it also reduced the rate of carbon production by dehydrogenation. Consequently more ordered and more thermodynamically stable MWCNTs were produced. Presumably the same effect is observed here. 80 Table 3.13: Shows the effect of hydrogen on the quality of sample content. Sample Carrier gas Sphere (%) Amorphous (%) N-CNTs (%) 15% Ar 0 20 80 15% Ar + H2 0 10 90 Figure 3.13: TEM images of N-CNTs produced using different carrier gases: Ar (a, b, c), Ar + 5% H2 (d, e, f) (15% CH3CN in toluene, T: 850 oC, flow rate: 200 mL/min). a b c d e f 81 iii) The effect of hydrogen on the CNT outer diameters This analysis revealed that Ar gas without hydrogen produced tubes with a narrower range of diameters as shown in Figure 3.14. This figure also indicates that most of the tubes have a diameter of 25 nm when no H2 is present whereas in the second graph (b) (H2 now added to Ar) most tubes have a larger diameter and show a broader range of diameters. Figure 3.14: Graphs of diameter distributions using different carrier gas a) Ar, b) Ar + 5% H2 (15% CH3CN in toluene, T: 850 oC, flow rate 200 mL/min). 0 5 10 15 20 10 20 30 40 50 60 75 90 100 110 120 150 190 Nu m be r o f tu be s Outer diameter of the tubes (nm) 0 2 4 6 8 10 12 14 16 10 20 30 40 50 60 70 80 90 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 20 0 26 0 32 0 34 0 Nu m be r o f t ub es Outer diameter of the tubes (nm) a b 82 3.2.4 Synthesis procedure: Bubbling method i) Synthesis of N-CNTs A bubbling method was also used to produce N-CNTs. Table 3.14 show the reaction conditions used for production of the N-CNTs using the bubbling method. In this study only one parameter was changed at a time and the ratio of Fe(CO)5/acetonitrile was kept constant at 90%. Table 3.14: The experimental conditions used for the synthesis of N-CNTs. Sample number CH3CN (mL) Fe(CO)5 (mL) Temperature (oC) Gas flow rate (mL/min) Carrier gas solution temperature (oC) Reactio n time (min) Yield (g) 1 5.1 0.67 800 200 Ar RT 60 0.03 2 5.1 0.67 800 300 Ar 75 30 0.01 3 5.1 0.67 850 200 Ar 40 30 0.01 4 10.2 1.34 850 300 Ar RT 60 0.08 5 10.2 1.34 900 300 Ar RT 30 0.06 6 10.2 1.34 900 300 Ar RT 60 0.06 7 10.2 1.34 900 100 Ar RT 30 0.03 ii) Results and discussion One of the more common problems which occurs in the injection method is blockage of the injector. This is due to decomposition of the Fe(CO)5 in the tip of the syringe. 83 To overcome this problem a bubbling method was used to generate N-CNTs. In this method argon was bubbled through a solution of Fe(CO)5 and acetonitrile. The Fe(CO)5 and the acetonitrile will have different vapour pressures and hence their volatility of the two ligands will be different. Thus, in these experiments control in terms of N-CNT formation is more difficult. Notwithstanding this issue, a study was undertaken to evaluate the effect of certain variables (flow rate, temperature, reagent ratios) on the products formed. It was observed that CH3CN and Fe(CO)5 mixed well together in the absence of the solvent (toluene). The bubbling method resulted in the production of a very good quality of N-CNTs with only a very small amount of amorphous carbon produced. However the bubbling method gave a poor yield of products, even after 30 min bubbling time at room temperature (2 mL out of the 10 mL solution was evaporated). To increase the yield the solution was heated up to 40- 75 oC (below the b.p of Fe(CO)5 (Table 3.14, entries: 2, 3), This did not resolve the yield problem. The time of reaction was increased from 30 min to 60 min but again no significant change in yield was observed. iii) Yield study The effect of different conditions on the product yield was studied and it was observed that the highest yield produced was 0.08 g under the conditions: temperature: 850 oC, gas flow rate: 300 mL/min at a reaction time of 60 min. The lowest yield obtained under the different conditions occurred: when using the condition of temperature-800 oC; gas flow rate-300 mL/min bubbled at 75 oC and at temperature-850 oC; gas flow rate-200 mL/min bubbled at 40 oC. It also noticed from the comparison between these experiments (Table 3.14 entries: 5 and 7), that when the flow rate decreased (from 300 mL/min to 100 mL/min) the yield also decreased. The effect of temperature on the yeild can also be seen in Table 3.14 (entries: 4 and 6). As the temperature increased the yeild decreased although the change was minor. 84 iv) TEM studies TEM was the only characterization technique which was used to study the products of these reactions. Based on TEM images it is clear that the bubbling method produced better quantity N-CNTs. The method produced N-CNTs with more corrugated compartments than bamboo like structures. The TEM images revealed that the N- CNTs produced from the solution heated to 40 oC contained the highest number of tubes Figure 3.15 e, f, whereas when the solution was heated to 75 oC tubes with lots of amorphous materials were observed Figure 3.15 c, d. When the gas flow rate dropped to 100 mL/min very few tubes were formed Figure 3.15 m, n. a b 85 c d e f g h 86 i j k l 87 Figure 3.15: TEM images of the N-CNTs produced under different conditions. Figures a, b (T: 800 oC, flow rate: 200 mL/min, bubbled at RT), c, d (T: 800 oC, flow rate: 300 mL/min, bubbled at 75 oC), e, f (T: 850 oC, flow rate: 200 mL/min, bubbled at 40 oC), g, h (T: 850 oC, flow rate: 300 mL/min, bubbled at RT), i, j (T: 900 oC, flow rate: 300 mL/min, bubbled at RT), k, l (T: 900 oC, flow rate: 300 mL/min, bubbled at RT), m, n (T: 900 oC, flow rate: 100 mL/min, bubbled at RT). v) Outer diameter distribution The samples were viewed by TEM and the outer diameters were measured using Image J as shown in Figure 3.16. The average of the outer diameter was measured and given in Table: 3.15. It is noticed that the samples recorded in graphs a and d (Table 3.14, entries: 1, 6) show narrower tubes when compared to samples recorded as entries: 2, 3. m n 88 Table 3.15: Displays the average outer diameter under different conditions. Sample number Avg diameter (nm) 1 49 ± 20 6 46 ± 20 2 58 ± 10 3 74 ± 10 0 2 4 6 8 10 12 14 16 18 20 20 30 40 50 60 70 80 90 110 120 140 150 160 220 Nu m be r o f t ub es Outer diameter of the tubes (nm) a 89 0 2 4 6 8 10 12 14 20 30 40 50 60 70 90 100110120130150160170180210230 Nu m be r o f t ub es Outer diameter of the tubes (nm) 0 5 10 15 20 25 10 20 30 40 50 60 80 90 130150240250270280420570720 Nu m be r o f t ub es Outer diameter of the tubes (nm) b c 90 Figure 3.16: Graphs of diameter distributions for the N-CNTs produced under different conditions, a ( T: 800 oC, flow rate: 200 mL/min), b (T: 800 oC, flow rate: 300 mL/min, bubbled at 75 oC), c (T: 850 oC, flow rate: 200 mL/min bubbled at 40 oC), d (T: 900 oC, flow rate: 300 mL/min). 3.4 Conclusion The study using Fe(CO)5 as catalyst to make N-doped CNTs has shown that this can be achieved. In this study variables such as methods to produce N-CNTs, Fe(CO)5/acetonitrile ratios, flow rates, temperatures and the addition of H2 to the Ar carrir gas were studied. Our results showed that: the bubbling method resulted in the production of a very good quality of N-CNTs with formation of a very small amount of carbon byproducts. However, the injection method gave a higher yield of products compared to the bubbling method. The injection method also required less reaction time than the bubbling method to produce more material. The variation of Fe(CO)5/acetonitrile ratios indicated that an increase in acetonitrile concentration led to formation of more tubes with bamboo structures with narrower outer diameters. It was revealed by TGA 0 2 4 6 8 10 12 14 16 18 20 10 20 30 40 50 60 80 90 100 120 210 250 Nu m be r o f t ub es Outer diameter of the tubes (nm) d 91 that an increase in N concentration decreased the thermal stability of the samples compared to the undoped CNTs. This instability is atributted to the structural disorder introduced by the presence of nitrogen. The impact of temperature on the tube outer diameter revealed that at 850 oC a better trend of outer diameter with narrower tubes was achieved compared to tubes produced at 900 oC. When H2 added to the Ar it improved the yield and the formation of more tubes with small amount of amourphous carbon. In conclusion, the study showed that Fe(CO)5 can be used as a catalyst in the presence of acetonitrile as a nitrogen source to produce N-CNTs. 3.5 References 1. T. Masciangioli, W. X. Zhang, Environ. Sci., 2003, 37, 102. 2. X. Kitipornchai, K. M. Leiw, J. Solid. Struct., 2005, 42, 6032. 3. M. Terrones, W. K. Hsu, H. W. Kroto, D. R. M. Walton, A revolution in materials science and electronics, Springer, Berlin, Germany, 1999. 4. T. W. Ebbesen, H. J. Lezec, H. Hiura, J. W. Bennett, H. F. Ghaemi, T. Thio, Nature, 1996, 381, 678. 5. J. M. Gibson, T. W.Ebbensen, M. M. J. Treacy, Nature, 1996, 381, 678. 6. J. Cheng, K. A. S. Fermando, L. M. Veca, Y. P. Sun, A. I. Lamond, Y. W. Lam, S. H. Cheng, ACS. Nano., 2008, 2, 2085. 7. B. Chandra, J. Bhattacharjee M. Purewal, Y. W. Son , Y. Wu, M. Huang, H. Yan, T.F. Heinz, P. Kim, J. B. Neaton , J. Hone, Nano Lett., 2009, 9, 1544. 8. P. Nikolaev, J. Bronikowski, R. K. Bradley, F. Rohmund, D. T. Colbert, K. A. Smith, R. E. Smalley, Chem. Phys. Lett., 1999, 313, 91. 9. O. A. Nerushev, M. Sveningsson, L. K. L. Falk, F. J. Rohmund, J. Mat. Chem., 2001, 11, 1122. 92 10. E. N. Nxumalo N. J. Coville, Materials, 2010, 21, 2141 11. M. J. Bronikowski, P. A. Willis, D.T.Colbert, K. A. Smith, R. E. Smalley. J. Vac. Sci. Tech., 2001, 19, 1800. 12. V. Bajpai, L. Dai, T. Ohashi, J. Am. Chem. Soc., 2004, 126, 5070. 13. V. O. Nyamori, S. D. Mhlanga, N. J. Coville, J. Organomet. Chem., 2008, 693, 2205. 14. B. C. Satishkumar, A. Govindaraj, R. Sen, C. N. R. Rao, Chem. Phys. Lett., 1998, 293, 47. 15. C. N. R. Rao, A. Govindaraj, Acc. Chem. Res., 2002, 35, 998. 16. P. Nikolaev, M. J. Bronikowski, R. K. Bradley, F. Rohmund, D. T. Colbert, K. A. Smith, R. E. Smalley, Chem. Phys. Lett., 1999, 313, 91. 17. M. J. Bronikowski, P. A. Willis, D. T. Colbert, K. A. Smith, R. E. Smalley, J. Vac. Sci. Tech., 2001, 19, 1800. 18. S. Huang, X. Cai, C. Du, J. Liu, J. Phys. Chem., 2003, 107, 13251. 19. H. Hou, Z. Jun, F. Weller, A. Greiner, Chem. Mater., 2003, 15, 3170. 20. C. S. Kuo, A. Bai, C. M. Huang, Y. Y. Li, C. C. Hu, C. C. Chen, Carbon, 2005, 43, 2760. 21. K. Herandi, A. Siska, L. Thien-Nga, L. Forro, I. Kiricsi, Solid State Ionic, 2001, 141, 203. 22. R. M. Yadav, D. P. Singh, T. Shripathi, O. N. Srivastava, J. Nanopart. Res., 2008, 10, 1349. 23. J. Feng, Y. Li, F. Hou, X. Zhong, Mater. Sci. Eng., 2008, 473, 238. 93 24. J. W. Jang, C. E. Lee, S. C. Lyu, T. J. Lee, C. J. Lee, Appl. Phys. Lett., 2004, 84, 2877. 25. E. N. Nxumalo, P. J. Letsoalo, L. M. Cele, N. J. Coville, J. Organomet. Chem., 2010, 695, 2596. 26. E. N. Nxumalo, V. O. Nyamori, N. J. Coville, J. Organomet. Chem., 2008, 693, 2942. 27. X.Y. Tao, X.B. Zhang, F.Y. Sun, J.P. Cheng, F. Liu, Z.Q. Luo, Diamond Relat. Mater., 2004, 16, 425. 28. R. Kurt, A. Karimi, Chem. Phys. Chem., 2001, 2, 388. 29. X. Ma, E. G. Wang, Appl. Phys. Lett., 2001, 78, 978. 30. A. A. Koós, F. Dillon, E. A. Obraztsova, A. Crossley, N. Grobert, Carbon, 2010, 48, 3033. 31. P. Ghosh, T. Soga, K. Ghosh, R. A. Afre, T. Jimbo, Y. Ando, J. Non Cryst. Solids. 2008, 354, 4101 32. J. Liu, S. Webster, D. L. Carroll, J. Phys. Chem., 2005, 109, 15769. 33. E. Cruz-Silva, D. A. Cullen, L. Gu, J. M. Romo-Herrera, E. Muñoz-Sandoval, F. López-Urías, B. G. Sumpter, V. Meunier, J. C. Charlier, D. J. Smith, H. Terrones, M. Terrones, ACS Nano, 2008, 2, 441. 34. C. J. Lee, S. C. Lyu, H. W. Kim, J. H. Lee, I. C. Cho, Chem. Phys. Lett., 2002, 359, 115. 35. J. Liu, R. Czerw, D. L. Carroll, Mater. Res., 2005, 20, 538. 36. E. J. Liang, P. Ding, H. R. Zhang, X. Y. Guo, Z. L. Du, Diamond Relat. Mater., 2004, 13, 69. 37. M. Terrones, N. Grobert, H. Terrones, Carbon, 2002, 40, 1665. 94 38. L. Ci, H. Zhu, B. Wei, J. Liang, C. Xu, D. Wu, Carbon, 1999, 37, 1652. 95 CHAPTER 4 4. CONCLUSIONS AND RECOMMENDATIONS 4.1 Conclusions The aim of this study was to synthesize nitrogen doped carbon nanotubes (N-CNTs) using the floating catalyst chemical vapour deposition (CVD) method utilizing iron pentacarbonyl as the catalyst and acetonitrile as a nitrogen source. In this study the N- CNTs were successfully synthesized and fully studied by means of Raman spectroscopy, thermogravimetric analysis (TGA) and transmission electron microscopy (TEM) analyses. Two different procedures were used to add the reagent to the reactor, an injection and a bubbling method. Most of the results for this study were achieved using the injection method. The influence of different parameters on the fabrication of N-CNTs was studied. These parameters include the physical parameters such as gas flow rate, growth temperature and the Fe(CO)5/acetonitrile ratio. Their influence on the outer and inner diameters of N-CNTs, the size of bamboo compartment, yield, the thermal stability and crystallinity of the product were determined. It was revealed that, the ratio of Fe(CO)5/acetonitrile played a great role in determining the yield, morphology, crystallinity and thermal stability of the N-CNTs. It was observed from yield studies that an increase in acetonitrile concentration decreased the yield due to the role of nitrogen. The TEM studies showed that an increase in acetonitrile concentration decreased the outer diameters and as expected had no significant effect on the inner diameters. It was noticed that as the concentration of acetonitrile was increased, the distance between each bamboo cap was reduced. Further, when a solution contained 15% acetonitrile was used the tubes showed predominantly bamboo compartments. The thermal stability of the samples was analysed by TGA analysis and it was confirmed that, as the incorporation of N in the CNTs increased the stability of the sample decreased. Raman spectroscopy did 96 not reveal the results which were consistent with other literature studies. An increase in level of disorder only was observed from samples produced from 2% and 25% of actonitrile were used. However, it was assumed that the disorder ratio increased as the concentration of acetonitrile went up even though no increase was observed in samples made with 5%, 15% and 20% acetonitrile concentrations. Reaction temperature was found to be a key parameter in the fabrication of N-CNTs. The effect of four different growth temperatures were studied as the temperature was varied from 750 oC to 900 oC. TEM analysis revealed that at 750 and 800 oC no tubes were synthesized and only fibers, agglomerates and nanobeads/pearls were observed. Good carbon products were achieved at 850 and 900 oC, and it was noticed that more tubes were formed at 850 oC compared to 900 oC. The effect of temperature on the outer diameter distribution revealed that at both temperatures (850 oC and 900 oC) there was a wide range of tube outer diameters but the material synthesized at 850 oC contained narrower tubes. No major temperature effect was observed on the average distance between the bamboo caps. Some experiments were performed in which 5% hydrogen was added to an Ar atmosphere. It was reported in the literature [1] that H2 can play an important role in changing the quality and quantity of CNT production. When hydrogen was added, it increased the yield of the product. It also reduced the amount of amorphous carbon since hydrogen reduced the rate of carbon production by hydrogenation to give more ordered and thermodynamically stable multi-walled carbon nanotubes (MWCNTs). Interestingly, it was observed that tubes with larger outer diameter were synthesized in the presence of hydrogen. The bubbling method was applied to make a comparison between the injection and the bubbling approach to make CNTs. When both methods were compared in this study it was concluded that the bubbling method appeared as a more simple method to make better quality tubes. The yield however was very low. 97 4.2 Recommendations for future study In this study Fe(CO)5 was used as the catalyst and it created some problems due to its volatility. This included blockage of the tip of the injector. It is suggested for future study to use a solid compound as the catalyst viz Fe(CO)4CH3CN. This can be synthesized from a mixture of acetonitrile and Fe(CO)5. This catalyst can be either used in solid form or be dissolved in a solvent. Another issue to be explored is the role of pressure on the reaction. The HiPco process is known to occur at pressure > 1 atmosphere. This may be an important parameter to make N-CNTs in high purity and yield. 4.3 References 1. V. O. Nyamori, S. D. Mhlanga, N. J. Coville, Organomet. Chem., 2008, 693, 2205.