Sie sind auf Seite 1von 149

HIGH PERFORMANCE SYNTHESIS AND PURIFICATION OF CARBON NANOTUBES

Caterina Leone

This Ph. D. thesis has been realized with the financial support of the Department of Chemical and Food Engineering of the University of Salerno

U n i o n e E u r o p e a UNIVERSITÀ DEGL

Unione Europea

U n i o n e E u r o p e a UNIVERSITÀ DEGL STUDI
U n i o n e E u r o p e a UNIVERSITÀ DEGL STUDI

UNIVERSITÀ DEGL STUDI DI SALERNO

FONDO SOCIALE EUROPEO

Programma Operativo Nazionale 2000/2006

“Ricerca Scientifica, Sviluppo Tecnologico, Alta Formazione” Regioni dell’Obiettivo 1 – Misura III.4

“Formazione superiore ed universitaria”

Department of Chemical and Food Engineering

Ph.D. Course in Chemical Engineering (VI Cycle-New Series)

HIGH PERFORMANCE SYNTHESIS AND PURIFICATION OF CARBON NANOTUBES

Supervisor Prof. Paolo Ciambelli

Scientific Referees Prof. D. Sannino Ing. M. Sarno

Ph.D. Course Coordinator Prof. Paolo Ciambelli

Ph.D. student Caterina Leone

To my family

Acknowledgments

Per la realizzazione di questo mio lavoro di tesi ringrazio il prof. Paolo Ciambelli per la collaborazione, la prof. ssa Diana Sannino per il suo contributo, l’ing. Maria Sarno che mi ha seguito con costanza e disponibilità.

Vorrei inoltre ringraziare i miei compagni di laboratorio Antonietta, Arianna, Claudia, Elvirosa, Emma, Giovanna, Giusy, Massimo, Paola, Roberto e Vincenzo, che hanno condiviso con me la movimentata attività di ricerca; e insieme ai quali ho collaborato con armonia e serietà.

Ringrazio infine i miei familiari sempre presenti con il loro affetto e con la loro attenzione.

Publications list

1. Ciambelli, P.; Sannino, D.; Sarno, M.; Leone, C. and Lafont, U.;

Effects of alumina phases and process parameters on the multiwalled carbon

nanotubes growth (2007) Diamond & Related Materials 16 11441149.

2. Ciambelli, P.; Sannino, D.; Sarno, M.; Leone, C. and Lafont, U.;

Effects of alumina phases and process parameters on the multiwalled carbon nanotubes growth. 16 th European Conference on Diamond, Diamond-Like Materials, Carbon Nanotubes, and Nitrides Estoril, Portugal, September 3-8, 2006, section 5.8.36 3. Ciambelli, P.; Sannino, D.; Sarno, M.; Leone, C. and Lafont, U.; CVD synthesis and characterization of carbon nanotubes produced on a monolith support. XXII National Congress Italian Chemical Society, SCI 2006, Florence September 10-15, 2006, p.284.

Abstract

Carbon nanotubes, constituted by rolled up graphene sheets, have become a popular topic of discussion, thanks to their exceptional electronic, mechanical and thermal properties, some of which derive from the close relation between carbon nanotubes and graphite, others from their high aspect ratio. Catalytic chemical vapour deposition (CCVD) is a widely employed method to synthesise carbon nanotubes, at controlled growth level. It requires much lower temperatures than other methods of synthesis, then it is more suitable for large-scale production at a lower cost. In this light, this thesis work has been focused to improve multi-wall carbon nanotubes (MWNT) growth by means of CCVD and to explore innovative purification methods. The first part of investigation has been addressed to enhance catalyst performance for the synthesis of CNT, investigating different aluminium hydrates, hydroxides, oxides (bohemite, gibbsite, bayerite, γ-alumina) as supports for Co and Fe metal particles. Subsequently, a large range of operative conditions to optimize the synthesis process on the selected “best catalyst” have been explored. Research activity was devoted to the determination for catalytic performances and lifetime of Co-Fe-supported gibbsite catalyst, developing a cross monitoring of MWNT growth by gas-phase and by solid-phase. The synthesis laboratory plant was implemented with on-line specific analysers for the continuous examination of reactor outlet stream, while discontinuous evaluations by means of physic-chemical characterisation of produced solid phase, was performed at set synthesis times. Parallel, innovative purification of as-grown CNT from by ethylene catalytic chemical vapour deposition (CCVD) on Co/Fe-zeolite, selected with respect to the high level of impurities, was investigated. Carbon nanotubes impurities were dissolved by a microwave assisted system, in the presence of inorganic acids such as HF and a HNO 3 /H 2 SO 4 mixture, founding that a much lower concentration of acids and acid immersing time were sufficient in comparison to more conventional methods. High performance purification permitted also functionalisation of MWNT.

On the synthesised nanotubes and on samples at a progressive purification degree, a wide characterization approach was based on the combined use of various techniques, such as transmission electron microscopy (TEM) coupled with EDX analysis, field emission scanning electron microscopy (FESEM), Raman and FT-IR Spectroscopy, X-ray diffraction analysis, N 2 adsorption-desorption analysis, quantitative evaluations on termogravimetric (TG) and inductive coupled plasma-mass spectrometer (ICP-MS) analyses. High performance synthesis of MWNT arose from very high carbon and reaction yield obtained by tuning operative parameters and catalysts selection and optimisation. The catalyst activity was kept constant for a longer period compared to literature previous findings. 100% selectivity to MWNT was reached by choosing the most appropriate support and by varying active metals amount. Finally, a kinetic investigation of carbon nanotubes synthesis has been performed.

Thesis outline

In chapter I, after an introduction, providing some historical background and a brief summary of the subject matter, the structure and properties of CNTs are reviewed in. In Chapter II and in the appendix the most common techniques employed for carbon nanotubes and characterisation are, in turn, discussed. Chapter III, IV, V, VI and VII deal with the experimental part of the thesis. The laboratory plant, for CNT synthesis is described in Chapter III.

Chapter IV reports the results obtained through different support methods for the CNT synthesis, the choice of the best materials and report the effects of several operating conditions on the synthesis performance. With the aim of obtaining high performance synthesis, in terms of CNT reaction yield and MWNTs selectivity, both catalytic activity and lifetime have been determined: the results are reported in the Chapter V. The experimental results of the innovative purification procedure, obtained with the microwave apparatus in comparison with a more conventional method, are described in Chapter VI. Finally, the results of the kinetic investigation are summarised in Chapter

VII.

Objectives

The target of the research has been to improve performance in catalytic synthesis and purification to obtain high quality carbon nanotubes, considering techniques at lower costs. To this purpose, a catalytic, chemical vapour deposition (CCVD) technique has been used to synthesise carbon nanotubes. The objectives have been achieved through the optimisation of the catalyst formulation, the operative conditions for the synthesis, the purification procedure.

Collaborations

Transmission electron microscopy (TEM) analyses have been conducted with the support of Delft Chem Tech.

Contents

3IChapter I Structure and properties of carbon nanotubes

1

I.1Structure

1

I.2 Characterization of carbon nanotubes

4

I.3 Mechanics of carbon nanotubes

6

I.3.1 Single-walled carbon nanotubes

6

I.3.2. Multi-walled carbon nanotubes…………………………………… 9

I.4 Thermal and electrical properties

10

Chapter II: Carbon nanotubes synthesis technique

13

II.1 Synthesis

13

II.2 Arc discharge

13

II.2.1 Synthesis of SWNT

15

II.2.1.1 Inert gas

15

II.2.1.2 Optical plasma control

15

II.2.I.3 Catalyst

16

II.2.I.4 Improvement of oxidation resistance

16

II.2.2 Synthesis of MWNT

17

II.2.2.1 Synthesis in liquid nitrogen

18

II.2.2.2 Magnetic field synthesis

18

II.2.2.3 Plasma rotating arc discharge

19

II.3 Laser ablation

20

II.3.1SWNT versus MWNT

22

II.3.2 Large scale synthesis of SWNT

23

II.3.3 Ultra fast Pulses from a free electron laser (FEL)method

23

II.3.4 Continuous wave laser-powder method

23

II.4 Chemical vapour deposition

25

II.4.1 Plasma enhanced chemical vapour deposition

25

II.4.2 Thermal chemical vapour deposition

26

II.4.3 Alcohol catalytic chemical vapour deposition………

27

II. 4.4 Vapour phase growth……………………………………………

28

II.4.5 Aero gel-supported chemical vapour deposition…………………29

II.4.6 CoMoCat process

29

Chapter III: Experimental I:Experimental Plant

33

III.1 Introduction

33

III.2 Experimental Plant for nanotubes synthesis

33

Chapter IV: Experimental II Effect of alumina phases and process

parameters in the MWNT growth

37

IV.1 Introduction

37

IV.2 Experimental

38

IV.2.1TG, DTG evaluation

38

IV.2.2 SEM characterization

42

IV.2.3 Transmission electron microscopy characterization

44

IV.2.4 Raman spectroscopy

45

Chapter V: Experimental III On-line analysis

47

V.I Introduction

47

V.2Experimental

48

V.3 Results and discussion

48

V.3.1 Catalyst characterization

48

V.3.2 On-line analysis of effluent gas

………………………………… 55

V.3.3 TG, DTG evaluation

 

61

V.3.4 SEM characterisation

64

V.3.5 Transmission electron microscopy characterization

65

V.3.6 Raman spectroscopy

 

67

Chapter

VI:

Experimental

IV

Microwave

purification

and

functionalization of MWNTs

 

69

VI.I Introduction

 

69

VI.2Experimental Part

 

70

VI.2.1 Carbon nanotubes synthesis

70

VI.2.1 Purification treatment

….………………………………

70

VI.2.3 Methods

 

.73

VI.3 Result and discussion

73

VI.3.1 Microscopy images

73

VI.3.2 TG-DTG evaluation

.78

VI.3.3 ICP-MS analysis

80

VI.3.4 Raman spectroscopy analysis

.81

VI.3.5 FT-IR

83

VI.3.6 X-ray analysis

84

VI.3.7 Surface area evaluation

86

VII Kinetic evaluation

87

VII.1Introduction

87

VII.2Experimental

87

VII.3 Results and discussion

88

VII.3.1 Reaction rate

88

VII.3.2 External and internal diffusion evaluation

88

VII.3.3 Kinetic study

92

Conclusion

97

References

99

Appendix: Investigation method

107

1 Electron microscopy

107

1.1 Scanning electron microscopy (SEM)

108

1.2

Transmission Electron Microscopy (TEM)

110

2 Raman spectroscopy

112

2.1 Introduction

112

2.2 Raman spectra of carbon

114

2.2.1 Raman spectra of graphite

114

2.2.2 Raman spectra of carbon nanotubes

115

Figures content

Figure I.1. Schematic diagram showing how a hexagonal sheet of graphite is

‘rolled’ to form a carbon nanotube

2

Figure I.2

zig-zag nanotube

Illustrations of the atomic structure of (a) an armchair and (b) a

2

Figure I.3 The Stone-Wales transformation occurring in an armchair nanotube under axial tensione 3

Figure I.4 TEM micrograph showing the layered structure of a multiwalled

4

carbon nanotube

Figure I.5 Micrographs showing (a) the apparatus for tensile loading of

MWCNTs and (b) the telescoping, ‘‘sword and sheath’’ fracture behavior of

the MWCNT

….5

Figure I.6 TEM micrograph and computer simulation of nanotube buckling 7

Figure I.7 Band-gap values vs. nanotube diameters define nanotubes as

metallic or

11

Figure II.1 Experimental set-up of an arc discharge apparatus

14

Figure II.2 Schematic drawings of the electrode set-ups for (a) the conventional and (b) the new arc discharge electrodes 16

Figure II.3 Experimental set-up of the torch arc method in open air

17

Figure II.4 Schematic drawings of the arc discharge apparatus in liquid nitrogen 18

Figure II.5 Schematic diagram of the synthesis system for MWNTs in a

19

magnetic field

Figure II.6 TEM micrograph and computer simulation of nanotube buckling 19

Figure II.7 Schematic rotating electrode system

20

Figure II.8 Schematic drawings of a laser ablation apparatus

21

Figure II.9 TEM images of a bundle of SWNTs catalysed by Ni/Y (2:0.5 at.

22

%) mixture, produced with a continuous

Figure II.10 Schematic drawings of the ultra fast-pulsed laser ablation

24

apparatus

Figure II.11(Left) The principle scheme of the set-up for carbon SWNT production by continuous wave laserpowder method (Right) HRTEM of a SWNT-bundle cross-section 24

Figure II.12 Schematic diagram of plasma CVD apparatus

 

26

Figure II.13

Schematic

diagram

of

thermal

CVD

apparatus

267

Figure II.14 ACCVD experimental apparatus.…………………………

28

Figure II.16 . Experimental set-up for laser-assisted CVD

 

30

Figure II.17 Schematic diagram of a CoMoCat apparatus

31

Figure III.1 Experimental plant

 

35

Figure III.2 Uras 14 and Caldos 17 ABB analysers

 

36

Figure IV.1 TG-DTG curves of samples obtained by supported synthesys

with: (a) bohemite; (b) gibbsite; (c) γ-alumina ; (d)

42

Figure IV.3 SEM images of: GDl60200 78 (a); GDl60300 78 (b); GDl10200 78

44

(c); GDl30200 78 (d), GDl120200 78 (e, f)

Figure IV.4 TEM images of: BWs30100 200 (a); TG residue at 510°C of

BWs30100 200 (b); GDl60200 78 (c); GDl30200 78 (d, e)

45

Figure IV.5 Raman Spectra of samples BWs30100 200 , TG residue at 510°C

46

of BWs30100 200 and BWs60300 200

Figure V.I FESEM image of impregnated and dried catalyst, the inserts show the EDX maps of the particles agglomerate in the image 50

Figure V.2. FESEM image of catalyst after synthesis pretreatment, the inserts show the EDX maps of the image zone 51

Figure V.3 FESEM image of catalyst after synthesis and successive oxidation of CNT, the inserts show the EDX maps of the image zone 52

Figure V.4 TGDTG curves of samples G5_30, G10_30, G15_30, G20_30

(a), TGDTG curves of gibbsite after a blank test

54

Figure V.5: X-ray diffraction patterns of gibbsite, catalyst after

pretreatment, catalyst after synthesis, γ-Al 2 O

54

Figure V.6 :Raman spectrum of support (gibbsite)

55

Figure V.7: Profiles of C 2 H 4 , C 2 H 2 , CH 4 and H 2 concentrations during CNT

growth tests for samples G15_30

56

Figure V.8: Profiles of C 2 H 4 , C 2 H 2 , CH 4 and H 2 concentrations during blank test 56

Figure V.9: Ethylene conversion and hydrogen yield for G5_30, G10_30

8

Figure V.10: Ethylene conversion and hydrogen yield for G15_30,G20_30 59

Figure V.11: Ethylene conversion and hydrogen yield for G5_30, G5_70, and G5_140…… 60

Figure V 12: TGDTG curves of samples G5_30, G10_30, G15_30,

test

G20_30 (a), TGDTG

curves

of

gibbsite

after

a

blank

62

Figure V.13: TGDTG curves c) of sample G5_30 taken at different heights

of the as grown bed a) A catalyst image b)

63

Figure. V.14: SEM images of G5_140 and G5_30 at 500X

64

Figure V.15 TEM images of G15_30 at low resolution a); higher resolution b), c), d) and f). EDX analysis of a catalytic particle inside the hallow core of a CNT e), and EDX analysis of catalytic particles covered of no-nanotubes

carbon g), evidencing Fe and Co presence

66

Figure V.16 Raman spectra (514 nm excitation wavelength) of samples

prepared at different ethylene partial pressure

67

Fig.V.17 Raman spectra (785 nm excitation wavelength) of G5_30, G10_30

68

and G20_30

Figure VI.1 TEM images of MW1(a, b) and EDX spectra of MW1(c )… 73

Figure VI.2 TEM images of MW2( a,b) and MW3 (c)

74

Figure VI.3 TEM images of MW4 (a), MW5 (b,c), MW6 (d,e)

75

Figure VI.4 Histogram of The inner and outer diameter of the raw and

treated materials

76

Figure VI.5 FESEM of MW1 (a,b)

77

Figure VI.6 MW1(a), MW3(b), MW6(c)………

……….…

79

Figure VI.7 TG-DTG curves of different

heating rate…………………………………………………………………81

samples and of MW1 at several

Figure VI.8 Raman spectra (785 nm excitation wavelength) of MW1, MW2, MW3, MW5, MW6…………………………………………………82

Figure VI.9 FTIR spectra of samples MW3 and MW6………….……….84

Figure VI.10 X-ray diffraction spectra of the raw and treated materials

85

Figure VI. 11 adsorption-desorption isotherms of N 2 at 77 K for MW1, MW2,MW3, MW5 and MW6…………………… …………… 86

Figure VII.1 Example (sample G5_30 120600 ) of hydrogen production curve as a function of time and determination of the global observed rate of CNT synthesis 91

Figure VII.2 Kinetic constant evaluation at 600°C, varying the amount of

catalyst at a constant feed rate, composition, pressure

96

Figure VII.3 Kinetic constant evaluation at 650°C, varying the amount of

catalyst at a constant feed rate, composition, pressure

97

Figure VII.4 Kinetic constant evaluation at 700°C, varying the amount of

catalyst at a constant feed rate, composition, pressure

Figure VII.5 Kinetic constant evaluation at 700°C, varying gas flow at given amount of catalyst at a constant feed composition and pressure 97

Figure VII.6 Plot of the natural logarithm of growth rate against the reciprocal of absolute temperature 97

97

Figure A.I Signal generated when a high-energy electron beam interacts with a thin specimen. Most signals can be detected in a SEM or TEM 107

Figure A.2 The two basic operations of a TEM: projecting the diffraction pattern on the viewing screen (left) or projecting the image on the screen

(right). The intermediate lens selects the back focal- or the image plane of

the objective lens as its object

111

Figure A.3 Signal generated when a high-energy electron beam interacts with a thin specimen. Most signals can be detected in a SEM or TEM 112

Figure A.4 The two basic operations of a TEM: projecting the diffraction pattern on the viewing screen (left) or projecting the image on the screen

(right). The intermediate lens selects the back focal- or the image plane of

the objective lens as its object

115

Figure A.5 Polarized molecule in an electric field (left), energy diagram of the different light scattering processes. The elastic Rayleigh scattering and the inelastic Raman scattering (Stokes/Anti-Stokes) (right)… 116

Figure A.6 Normalised Raman spectra of HOPG, glassy carbon, CVD MWNT and CVD SWNT. The laser wavelength was λ = 514.5 nm (left). The inverse crystallite size vs.the ratio of the D- and G-band; according to (right) 117

Tables content

Table IV.I Synthesis conditions with the different supports, carbon yield, MWNT selectivity and reaction yield obtained by TG evaluation 39

Table IV.2 Synthesis conditions with the different supports, carbon yield, MWNT selectivity and reaction yield obtained by TG evaluation 40

Table IV.3 Carbon yield obtained by TG evaluation for samples prepared

with different metal loading

41

Table V.1 carbon yield, MWNT selectivity and reaction yield obtained by TG evaluation 61

Table V.2 I D /I G intensity ratio at two different excitation wavelength for CNT produced at increased ethylene partial pressure and at different times 65

Table VI.1 TG residue % samples MW1, MW2, MW3, MW5, MW6

78

Table VI.2 Results of the ICP-MS analysis for the raw and treated materials, performed with particular regards to the Co and Fe content 80

Table VI.3 Raman spectroscopy analysis: I D /I G ratio, D and G wavenumbers for raw samples and treated materials 83

Table VI.4 Effects of the treatments on the nanotubes geometrical characteristics, XRD evaluation 85

Table VII.1 External diffusion evaluation before and after syntheses

90

Table VII.2 Internal diffusion evaluation before and after synthesis.,

93

Table VII. 3 Calculated results for the reaction C 2 H 4 2C+2H 2

94

Chapter I Structure and properties of CNT

I.1 Structure

Carbon nanotubes can be visualized as a sheet of graphite that has been rolled into a tube. Unlike diamond, where a 3-D diamond cubic crystal structure is formed with each carbon atom having four nearest neighbors arranged in a tetrahedron, graphite is formed as a 2-D sheet of carbon atoms arranged in a hexagonal array. In this case, each carbon atom has three nearest neighbors. ‘Rolling’ sheets of graphite into cylinders forms carbon nanotubes. The properties of nanotubes depend on the atomic arrangement (how the sheets of graphite are ‘rolled’), the diameter and length of the tubes, and the morphology, or nano structure. Nanotubes exist as either single-walled or multi-walled structures; multi-walled carbon nanotubes (MWCNTs) are simply composed of concentric single-walled carbon nanotubes of progressive larger diameters.“Thostenson et al.(2001)”. The atomic structure of nanotubes is described in terms of tube chirality, or helicity, which is defined by the chiral vector, C h , and the chiral angle, θ. In Fig. I.1, we can visualize cutting the graphite sheet along the dotted lines and rolling the tube so that the tip of the chiral vector touches its tail. The chiral vector, often known as the roll-up vector, can be described by the following equation:

r

C

h

=

na r

1

+

ma r

2

where the integers (n, m) are the number of steps along the ziz-zag carbon bonds of the hexagonal lattice and a 1 and a 2 are unit vectors, shown in Fig. I.1.

Chapter I

Chapter I Figure I.1 : Schematic diagram showing how a hexagonal sheet of graphite is ‘rolled’

Figure I.1: Schematic diagram showing how a hexagonal sheet of graphite is ‘rolled’ to form a carbon nanotube.

The chiral angle determines the amount of ‘twist’ in the tube. There are two limiting cases where the chiral angle is at 0° and 30°. These limiting cases are referred to as ziz-zag (0°) and armchair (30°) based on the geometry of the carbon bonds around the circumference of the nanotube. The difference in armchair and zig-zag nanotube structures is shown in Fig.

I.2.

and zig-zag nanotube structures is shown in Fig. I.2. Figure I.2 : Illustrations of the atomic

Figure I.2: Illustrations of the atomic structure of (a) an armchair and (b) a zig-zag nanotube.

In relation to the roll-up vector, the ziz-zag nanotube is (n, 0) and the armchair nanotube is (n, n). The roll-up vector of the nanotube also defines the nanotube diameter since the inter-atomic spacing of the carbon atoms is known. The chirality of the carbon nanotube has significant implications on its material properties. In particular, tube chirality is known to have a strong

Structure and properties of CNT

impact on the electronic properties of carbon nanotubes. Graphite is considered to be a semi-metal, but it has been shown that nanotubes can be either metallic or semiconducting, depending on tube chirality “Dresselhaus et al. (1996)”. Investigations on the influence of chirality on the mechanical properties have also been reported. The analytical work of Yakobson et al. “Yakobson et al. (1996a), Yakobson et al. (2000b)” examined the instability of carbon nanotubes beyond linear response. Their simulations show that carbon nanotubes are remarkably resilient, sustaining extreme strain with no signs of brittleness or plasticity. Although the chirality has a relatively small influence on the elastic stiffness, they concluded that the Stone-Wales transformation, a reversible diatomic interchange where the resulting structure is two pentagons and two heptagons in pairs, plays a key role in the nanotube plastic deformation under tension. The Stone-Wales transformation, shown in Fig. I.3, occurs when an armchair nanotube is stressed in the axial direction. Nardelli et al.(1998) theorized that the Stone- Wales transformation results in a ductile fracture for armchair nanotubes.

results in a ductile fracture for armchair nanotubes. Figure I.3: The Stone-Wales transformation occurring in an

Figure I.3: The Stone-Wales transformation occurring in an armchair nanotube under axial tension.

As mentioned before, fullerenes are closed, convex cages that are composed of pentagons and hexagons. The Stone-Wales transformation introduces a new defect into the nanotube structure, the heptagon. Heptagons allow for concave areas within the nanotube. Thus, the heptagonal defects in nanotubes can result in many possible equilibrium shapes. Indeed, most nanotubes are not straight cylinders with hemispherical caps. In addition to different tube morphologies caused by defects, carbon nanotubes can be single walled or multi-walled structures. Fig. I.4 shows a transmission electron microscope (TEM) image showing the nanostructure

Chapter I

of a multi-walled carbon nanotube where several layers of graphitic carbon and a hollow core are evident. Multi-walled carbon nanotubes are essentially concentric single walled tubes, where each individual tube can have different chirality. These concentric nanotubes are held together by secondary, van der Waals bonding. Single-walled nanotubes are essential for fundamental investigations of the structure/property relationships in carbon nanotubes, since the intra-tube interactions further complicate the properties of carbon nanotubes. Indeed, both single and multi-walled nanotubes show unique properties that can be exploited for use in composite materials.

that can be exploited for use in composite materials. Figure I.4: TEM micrograph showing the layered

Figure I.4: TEM micrograph showing the layered structure of a multiwalled carbon nanotube. “Thostenson et al.(2001)”

I.2 Characterization of carbon nanotubes

Significant challenges exist in both the micromechanical characterization of nanotubes and the modelling of the elastic and fracture behaviour at the nano-scale. Challenges in characterization of nanotubes and their composites include: (a) complete lack of micromechanical characterization techniques for direct property measurement, (b) tremendous limitations on specimen size, (c) uncertainty in data obtained from indirect measurements, and (d) inadequacy in test specimen preparation techniques and lack of control in nanotube alignment and distribution. In order to understand the mechanical properties of carbon nanotubes, a number of investigators have attempted to characterize them directly. Treacy et al.”Treacy et al. (1996)” first investigated the elastic modulus of isolated multi-walled nanotubes measuring, by transmission electron microscopy, the amplitude of their intrinsic thermal vibration. The average value obtained over 11 samples was 1.8 TPa. Direct measurement of the stiffness and strength of individual, structurally isolated multi-wall carbon nanotubes has been made with an atomic-force microscope (AFM). Wong et al. “Wong et al.(1997)” were the first to perform direct measurement of the stiffness and strength of individual, structurally isolated multi-wall carbon nanotubes using atomic force microscopy. The nanotube was pinned at one end to molybdenum

Structure and properties of CNT

disulfide surfaces and load was applied to the tube by a AFM tip. The bending force was measured as a function of displacement along the unpinned length, and a value of 1.26 TPa was obtained for the elastic modulus. The bending force was measured as a function of displacement along the unpinned length, and a value of 1.26 TPa was obtained for the elastic modulus. The average bending strength measured was 14.2_8 GPa. Single-walled nanotubes tend to assemble in the ‘ropes’ of nanotubes. Salvetat et al. “Salvetat et al.(1999)” measured the properties of these nanotube bundles with the AFM. As the diameter of the tube bundles increases, the axial and shear module decrease significantly. This suggests slipping of the nanotubes within the bundle. Walters et al. “Walters et al. (1999)” further investigated the elastic strain of nanotube bundles with the AFM. On the basis of their experimental strain measurements and an assumed elastic modulus of 1.25 TPa, they calculated a yield strength of 45±7 GPa for the nanotube ropes. Indeed, their calculated value for strength would have been much lower if the elastic modulus of the nanotube bundle had been decreased as a consequence of slipping within the bundle, as suggested by Salvetat et al. “Salvetat et al. (1999).

b

a

by Salvetat et al. “Salvetat et al . (1999). ” b a Figure I 5 :
by Salvetat et al. “Salvetat et al . (1999). ” b a Figure I 5 :

Figure I 5: Micrographs showing (a) the apparatus for tensile loading of MWCNTs and (b) the telescoping, ‘‘sword and sheath’’ fracture behavior of the MWCNT

Chapter I

Yu et al. “ Yu et al. (2000a), Yu et al. (2000b)” have investigated the tensile loading of multi-walled nanotubes and single-walled nanotube ropes. In their work, the nanotubes were attached between two opposing AFM tips and loaded under tension. Their experimental set-up is shown in Fig. I.5. For multi-walled carbon nanotubes “Yu et al.(2000b)” the outermost tube failed followed by the pullout of the inner nanotubes. This ‘sword and sheath’ telescoping failure mechanism of multi-walled carbon nanotubes in tension is also shown in Fig. I.5. The experimentally calculated tensile strengths of the outermost layer ranged from 11 to 63 GPa and the elastic modulus ranged from 270 to 950 GPa. In their subsequent investigation of single- walled nanotube ropes Yu et al. “Yu et al.(2000a)”, assumed that only the outermost tubes assembled in the rope carried the load during the experiment, and they calculated tensile strengths of 13 to 52 GPa and average elastic module of 320 to 1470 GPa. Xie et al. “Xie et al.(2000)” also tested ropes of multi-walled nanotubes in tension. In their experiments, the obtained tensile strength and modulus were 3.6 and 450 GPa, respectively. It was suggested that the lower values for strength and stiffness may be a consequence of defects in the CVD-grown nanotubes.

I.3 Mechanics of carbon nanotubes

As discussed in the previous section, nanotube deformation has been examined experimentally “Thostenson et al.(2001)”. Recent investigations have shown that carbon nanotubes possess remarkable mechanical properties, such as exceptionally high elastic modulus “Treacy et al., (1996), Wong et al.(1997)”, large elastic strain and fracture strain sustaining capability “Falvo et al.(1997), Bower et al.(1999)”. Similar conclusions have also been reached through some theoretical studies “Overney et al.(1993), Lu (1997), Yakobson et al.(1997), Bernholc et al. (1998)” although very few correlations between theoretical predictions and experimental studies have been made. In this section we examine the mechanics of both single walled and multi-walled nanotubes.

I.3.1 Single-walled carbon nanotubes

Theoretical studies concerning the mechanical properties of single-walled nanotubes have been pursued extensively. Overney et al. “Overney et al. (1993)” studied the low-frequency vibrational modes and structural rigidity of long nanotubes consisting of 100, 200 and 400 atoms. The calculations were based on an empirical Keating Hamiltonian with parameters determined from first principles. A comparison of the bending stiffnesses of single-walled nanotubes and an iridium beam was presented. The bending stiffness of the iridium beam was deduced by using the continuum Bernoulli- Euler theory of beam bending. Overney and his co-workers concluded that

Structure and properties of CNT

the beam bending rigidity of a nanotube exceeds the highest values found in any other presently available materials. In addition to their experimental observations, Iijima et al. “Iijima et al., (1996)” examined response of nanotubes under compression using molecular dynamics simulations. They simulated the deformation properties of single- and multi-walled nanotubes bent to large angles. Their experimental and theoretical results show that nanotubes are remarkably flexible. The bending is completely reversible up to angles in excess of 110°, despite the formation of complex kink shapes. Fig. I.6 shows their numerical and experimental results, demonstrating the exceptional resilience of carbon nanotubes under high strain. Ru “Ru et al.(2000)” noticed that the actual bending stiffness of singlewalled nanotubes is much lower than that given by the elastic-continuum shell model if the commonly defined representative thickness is used. Ru proposed the use of an effective nanotube bending stiffness as a material parameter not related to the representative thickness. With the aid of this concept, the elastic shell equations can be readily modified and then applied to single-walled nanotubes. The computational results based on this concept substantiate the results from molecular dynamics simulations.

the results from molecular dynamics simulations. Figure I.6 : TEM micrograph and computer simulation of

Figure I.6: TEM micrograph and computer simulation of nanotube buckling

Vaccarini et al. “Vaccarini et al.(2000)” investigated the influence of nanotube structure and chirality on the elastic properties in tension, bending, and torsion. They found that the chirality had little influence on the nanotube tensile modulus. However, the chiral tubes exhibit asymmetric torsional behavior with respect to left and right twist, whereas the armchair and ziz-

Chapter I

zag tubes do not exhibit this asymmetric torsional behaviour. A relatively comprehensive study of the elastic properties of single-walled nanotubes was reported by Lu “Lu (1997)”. In this study, Lu adopted an empirical lattice- dynamics model “Al-Jishi et al. (1982)” which has been successfully adopted in calculating the phonon spectrum and elastic properties of graphite. In this lattice-dynamics model, atomic interactions in a single carbon layer are approximated by a sum of pair-wise harmonic potentials between atoms. The local structure of a nanotube layer is constructed from conformal mapping of a graphite sheet on to a cylindrical surface. Lu’s work attempted to answer such basic questions as: (a) how do elastic properties of nanotubes depend on the structural details, such as size and chirality? and (b) how do elastic properties of nanotubes compare with those of graphite and diamond? Lu concluded that the elastic properties of nanotubes are insensitive to size and chirality. Young’s predicted modulus (~1 TPa), shear modulus (~0.45 TPa), and bulk modulus (~0.74 TPa) are comparable to those of diamond. Hernandez and his co-workers “Hernandez et al.(1998)” performed calculations similar to those of Lu and found slightly higher values (~1.24 TPa) for Young’s tubes modulus. But unlike Lu, they found that elastic modulus are sensitive to both tube diameter and structure. Besides their unique elastic properties, the inelastic behaviour of nanotubes has also received considerable attention. Yakobson and co-workers “Yakobson et al.(1996), Bernholc et al. (1998)” examined the instability behaviour of carbon nanotubes beyond linear response by using realistic many-body Tersoff-Brenner potential and molecular dynamics simulations. Their molecular-dynamics simulations show that carbon nanotubes, when subjected to large deformations, reversibly switch into different morphological patterns. Each shape change corresponds to an abrupt release of energy and a singularity in the stress/strain curve. These transformations are explained well by a continuum shell model. With properly chosen parameters, their model provided a very accurate ‘roadmap’ of nanotube behavior beyond the linear elastic regime. They also carried out molecular dynamics simulations to single- and double-walled nanotubes of different chirality and at different temperatures “Yakobson et al.(1997b)”. Their simulations show that nanotubes have an extremely large breaking strain (in the range 30–40%) with decreases according to temperature. Yakobson “Yakobson et al.(2000c)” also applied dislocation theory to carbon nanotubes to describe their main routes of mechanical relaxation under tension. It was concluded that the yield strength of a nanotube depends on its symmetry and it was believed that there exists an intra-molecular plastic flow. Under high stress, this plastic flow corresponds to a motion of dislocations along helical paths within the nanotube wall and causes a stepwise necking, a well-defined new symmetry, as the domains of different chiral symmetry are formed. As a result, both the mechanical and electronic

Structure and properties of CNT

properties of carbon nanotubes are changed. The single walled nanotubes produced by laser ablation and arc-discharge techniques have a greater tendency to form ‘ropes’ or aligned bundles “Journet et al.(1995), Thess et al.(1996)”. Thus, theoretical studies have been made to investigate the mechanical properties of these nanotube bundles. Ru “Ru (2000) presented a modified elastic-honeycomb model to study elastic buckling of nanotube ropes under high pressure. Ru gave a simple formula for the critical pressure as a function of nanotube Young’s modulus and wall thickness-to-radius ratio. It was concluded that single-walled ropes are susceptible to elastic buckling under high pressure and elastic buckling is responsible for the pressure-induced abnormalities of vibration modes and electrical resistivity of single walled nanotubes. Popov et al. “Popov et al. (2000b)” studied the elastic properties of triangular crystal lattices formed by single-walled nanotubes by using analytical expressions based on a force constant lattice dynamics model “Popov et al.(1999a)”. They calculated various elastic constants of nanotube crystals for nanotube types, such as the armchair and zigzag. It was shown that the elastic modulus, Poisson’s ratio and bulk modulus clearly exhibit a strong dependence on the tube radius. The bulk modulus was found to have a maximum value of 38 GPa for crystals composed of singlewalled nanotubes with ~0.6 nm radius.

I.3.2. Multi-walled carbon nanotubes

Multi-walled nanotubes are composed of a number of concentric single walled nanotubes held together with relatively weak van der Waals forces. The multi-layered structure of these nanotubes further complicates the modelling of their properties. Ruoff and Lorents “Ruoff et al.(1995)” derived the tensile and bending stiffness constants of ideal multi-walled nanotubes in terms of the known elastic properties of graphite. It has been suggested that unlike the strongly anisotropic thermal expansion in conventional carbon fibres and graphite, the thermal expansion of carbon nanotubes is essentially isotropic. However, the thermal conductivity of nanotubes is believed to be highly anisotropic and its magnitude along the axial direction is perhaps higher than that of any other material. Lu “Lu et al.(1997)” also calculated the elastic properties of many multi- walled nanotubes formed by single-layer tubes by means of the empirical- lattice dynamics model. It was found that elastic properties are insensitive to different combinations of parameters, such as chirality, tube radius and numbers of layers, and the elastic properties are the same for all nanotubes with a radius larger than one nm. Interlayer van der Waals interaction has a negligible contribution to both the tensile and shear stiffness.

Chapter I

Govindjee and Sackman “Govindjee et al.(1999)” were the first to examine the use of continuum mechanics to estimate the properties of multi- walled nanotubes. They investigated the validity of the continuum approach by using Bernoulli-Euler bending to infer the Young’s modulus. They used a simple elastic sheet model and showed that at the nanotube scale, the assumptions of continuum mechanics must be carefully respected in order to obtain reasonable results. They showed the explicit dependence of ‘material properties’ on system size when a continuum cross-section assumption was used. Ru “Ru et al.(2000a)” used the elastic-shell model to study the effect of van der Waals forces on the axial buckling of a double-walled carbon nanotube. The analysis showed that the van der Waals forces do not increase the critical axial buckling strain of a double-walled nanotube. Ru “Ru et al.(2000b), Ru et al.(2001c)” thereafter also proposed a multiple column model that considers the interlayer radial displacements coupled through the van der Waals forces. This model was used to study the effect of interlayer displacements on column buckling. It was concluded that the effect of interlayer displacements could not be neglected unless the van der Waals forces are extremely strong. Kolmogorov and Crespi “Kolmogorov et al.(2000)” investigated the interlayer interaction in two-walled nanotubes. A registry-dependent two-body graphite potential was developed. It was demonstrated that the tightly constrained geometry of a multi-walled nanotube could produce an extremely smooth solid-solid interface wherein the corrugation against sliding does not grow in relation to system size. The energetic barrier to interlayer sliding in defect-free nanotubes containing thousands of atoms can be comparable to that for a single unit cell of crystalline graphite.

I.4 Thermal and electrical properties

Thermal properties including specific heat and thermal conductivity of carbon nanotubes are determined primarily by the phonons “Cui et al.(2000)”. A phonon is a quantum acoustic energy similar to the photon. Phonons are a result of lattice vibrations observed in the Raman spectra “Collins et al.(2000).”. The phonon contribution to these quantities dominates especially at low temperatures and is due to the acoustic phonons. The measurements of thermoelectric power of nanotube systems give direct information for the type of carriers and conductivity mechanisms. Theoretical and experimental results have shown the superior electrical properties of carbon nanotubes. They can produce electric current carrying capacity 1000 times higher than copper wires “Salvetat et al. (1999)”. For 1D systems, cylindrical surface and translational symmetry with a screw axis could affect the electronic structures and related properties. The electronic capabilities possessed by carbon nanotubes are seen to arise predominately

Structure and properties of CNT

from interlayer interactions, rather than from interlayer interactions between multi layers within a single carbon nanotube or between different nanotubes “ Walters et al.(1999)”. These optical properties have proved to be especially unique with the capabilities of acting either as a metallic or semiconductor, which depends on tubule diameter and chiral angle. Studies have shown that metallic conduction can be achieved without the introduction of doping effects. For semiconducting nanotubes the band gaps have been found to be proportional to a fraction of the diameter and without relation to the tubule chirality “Walters et al.(1999)”.

to the tubule chirality “Walters et al. (1999)”. Figure I.7 : Band-gap values vs. nanotube diameters

Figure I.7: Band-gap values vs. nanotube diameters define nanotubes as metallic or semiconducting.

The I-tight-binding model within the zone folding scheme shows, one third of carbon nanotubes are found to be metallic while two thirds are semiconducting, depending on their indices “Cui et al.(2000)”. Calculations based on the use of r and P bands, due to curvature induced mixing of these bands, are used to predict that some metallic nanotubes are very-small-band- gap semiconducting nanotubes “Yu et al.(2000b)” (Fig. I.7). The symmetry of the structures basically links all the calculations in both single and multi- wall carbon nanotubes. The electronic properties of bundles of single wall nanotubes can be derived, assuming the intertube interactions are not strong enough to change the band structure. Broken symmetry caused by interactions between tubes in a bundle create a pseudogap of about 0.2 and 0.1 eV “Yu et al.(2000a)” This pseudo-gap, which is created can modify

Chapter I

electronic properties such as semimetal-like temperature dependence of the electrical conductivity and a finite gap in the infrared absorption spectrum is predicted “Cui et al.(2000)”.

Chapter II Carbon nanotubes synthesis technique

II.1 Synthesis

Since carbon nanotubes were discovered nearly a decade ago, there have been a variety of techniques developed for their preparation. Primary synthesis methods for single and multi-walled carbon nanotubes include arc-discharge “Keidar et al.(2004), Lee et al (2002)”, laser ablation “Assovskii et al.(2003)”, gas-phase catalytic growth from carbon monoxide “Khassin et al.(1998)”, and chemical vapour deposition (CVD) of hydrocarbons “Zhang et al.(2005), Du et al.(2005)”. In order to apply carbon nanotubes, large quantities are required, and the scale-up limitations of the arc discharge and laser ablation techniques would make the cost of nanotube prohibitive. During nanotube synthesis, impurities in the form of catalyst particles, amorphous carbon, and non-tubular fullerenes are also produced. Thus, subsequent purification steps are required to separate the tubes. The gas-phase processes tend to produce nanotubes with fewer impurities and are more amenable to large-scale processing. It is considered that gas phase techniques, such as CVD, for nanotube growth offer the greatest potential for the scaling-up of nanotube production for the processing of composites and applications in different fields. In this chapter, we summarize the progress made in recent years in growing carbon nanotubes by various methods to produce carbon nanotubes and review some of the benefits and drawbacks of each technique.

II.2 Arc discharge

The carbon arc discharge method, initially used for producing C 60 fullerenes, is the most common and perhaps easiest way to produce carbon nanotubes as it is rather simple to undertake. However, it is a technique that

Chapter II

produces a mixture of components and requires separating nanotubes from the soot and the catalytic metals present in the raw product. This method creates nanotubes through arc-vaporisation of two carbon rods placed end to end, separated by approximately 1mm, in an enclosure that is usually filled with inert gas (helium, argon) at low pressure (between 50 and 700 mbar). Recent investigations have shown that it is also possible to create nanotubes with the arc method in liquid nitrogen “Anazawa et al.(2002)”. A direct current of 50 to 100 A driven by approximately 20 V creates a high temperature discharge between the two electrodes. The discharge vaporises one of the carbon rods and forms a small rod shaped deposit on the other rod. Producing nanotubes in high yield depends on the uniformity of the plasma arc and the temperature of the deposit form on the carbon electrode “Lee et al. (2002)” Insight into the growth mechanism is increasing and measurements have shown that the different diameter distributions that have been found are dependant on the mixture of helium and argon. These mixtures have different diffusions coefficients and thermal conductivities. These properties affect the speed with which the carbon and catalyst molecules diffuse and cool, affecting nanotube diameter in the arc process. This implies that single- layer tubules nucleate and grow on metal particles in different sizes depending on the quenching rate in the plasma and it suggests that temperature and carbon and metal catalyst densities affect the diameter distribution of nanotubes “Lee et al. (2002)” Depending on the preparation technique it is possible to selectively grow SWNTs or MWNTs, as shown in Figure II.1.

selectively grow SWNTs or MWNTs, as shown in Figure II.1. Figure II.1 : Experimental set-up of

Figure II.1: Experimental set-up of an arc discharge apparatus

Carbon nanotubes synthesis technique

II.2.1 Synthesis of SWNT

If SWNTs are to be obtained, the anode has to be doped with metal catalyst, such as Fe, Co, Ni, Y or Mo.

A lot of elements and mixtures of elements have been tested by various

authors “Journet et al.(1998)” and it is noted that the results vary a lot, even

though they use the same elements. This is not surprising as experimental conditions differ. The quantity and quality of the nanotubes obtained depend on various parameters such as the metal concentration, inert gas pressure, type of gas, current and system geometry. Usually the diameter is in the range of 1.2 to 1.4 nm. Some ways of improving the process of arc discharge are described below.

II.2.1.1 Inert gas

The most common problems with SWNT synthesis are that the product

contains a lot of metal catalyst, SWNTs have defects and purification is hard

to perform. On the other hand, one advantage is that the diameter can be

slightly controlled by changing thermal transfer and diffusion, hence condensation of atomic carbon and metals between the plasma and the vicinity of the cathode can control nanotube diameter in the arc process. This was shown in an experiment in which different mixtures of inert gases were

used “Farhat et al.(2001a)”. It appeared that argon, with a lower thermal conductivity and diffusion

coefficient, gave SWNTs with a smaller diameter of approximately 1.2 nm.

A linear fit of the average nanotube diameter showed a 0.2 nm diameter

decrease per 10 % increase in argon helium ratio, when nickel/yttrium was

used (C/Ni/Y was 94.8:4.2:1) as catalyst.

II.2.1.2 Optical plasma control

A second way of controlling plasma is to change the anode to cathode

distance (ACD). The ACD is adjusted in order to obtain strong visible vortices around the cathode. This enhances anode vaporisation, which improves nanotubes formation. Combined with controlling the argon-helium mixture, one can simultaneously control the macroscopic and microscopic parameters of the nanotubes formed “Farhat et al.(2001b)”. With a nickel and yttrium catalyst (C/Ni/Y is 94.8:4.2:1) the optimum nanotube yield was found at a pressure of 660 mbar for pure helium and 100 mbar for pure argon. The nanotube diameter ranges from 1.27 to 1.37 nanometre.

Chapter II

II.2.I.3 Catalyst

Since chemical vapour deposition (CVD) could give SWNTs with a diameter of 0.6–1.2 nm, researchers tried the same catalyst as used in CVD on arc discharge. Not all of the catalysts used appeared to give nanotubes for both methods. But there seemed to be a correlation of diameter of SWNTs synthesised by CVD and arc discharge. As a result, the diameter can be controlled to a range of 0.6-1.2 nm with arc-discharge.

II.2.I.4 Improvement of oxidation resistance

There has also been progress in developing methods to improve the oxidation resistance of the SWNTs, which is a consequence of the defects present in nanotubes. Strong oxidation resistance is needed for the nanotubes if they have to be used for applications such as field emission displays. Recent research has indicated that a modified arc-discharge method using a bowl-like cathode (see Figure II.2) decreases the defects and gives cleaner nanotubes, and thus improves the oxidation resistance “Huang et al.(2002)” The Raman spectrum of the newly synthesised nanotubes shows that the nanotubes formed are cleaner and less defective compared to those synthesised by conventional methods. The anode rod contained Ni and Y catalyst (C /Ni/Y is 94.8:4.2:1). No information is given about the diameter size.

No information is given about the diameter size. Figure II.2 :Schematic drawings of the electrode set-ups

Figure II.2:Schematic drawings of the electrode set-ups for (a) the conventional and (b) the new arc discharge electrodes.

It was only recently discovered that it is possible to form MWNTs in the open air “Takikawa et al.(2001b)”; by welding arc torch, the process was shielded with argon gas flow.The anode and cathode were made of graphite containing Ni and Y (Ni/Y is 4.2:1 at. %).

Carbon nanotubes synthesis technique

Carbon nanotubes synthesis technique Figure II.3 : Experimental set-up of the torch arc method in open

Figure II.3: Experimental set-up of the torch arc method in open air.

This method was modified for preparing SWNTs “Takikawa et al.(2001a)” A plate target made of graphite containing Ni and Y , was fixed at the sidewall of a water–cooled, steel based electrode. The torch arc aimed at the edge of the target and the soot was deposited on the substrate behind the target (see Figure II.3). The arc was operated at a direct current of 100 A. and shielding argon gas flowed through the torch, enhancing the arc jet formation beyond the target. Carbon nanohorns (CNHs) and bundles of SWNT with an average diameter of 1.32 nm were found in the soot. However, the yield was much lower than for the conventional low-pressure arc discharge method. There are two reasons for this; first of all, because of the open air, the lighter soot will escape into the atmosphere, secondly, the carbon vapour might be oxidised and emitted as carbon dioxide gas. In order to improve the yield in this method, a means for collecting soot and developing an appropriate target are required. This method seems to be convenient and inexpensive once the conditions for higher yield are optimised. With a Ni/Y catalyst (Ni/Y is 3.6:0.8), SWNT bundles and CNHs are formed. In this case the SWNTs have a diameter of approximately 1.32 nm “Takikawa et al.(2002b)”.

II.2.2 Synthesis of MWNT

If both electrodes are graphite the main product will be MWNTs. But next to MWNTs a lot of side products are formed such as fullerenes, amorphous carbon, and some graphite sheets. The purification of the MWNTs causes loss of structure and disorders the walls. Therefore, scientists are developing ways to gain pure MWNTs in a large-scale process without purification.Typical sizes for MWNTs are 1-3 nm for inner diameter

Chapter II

and approximately 10 nm for outer diameter Because no catalyst is involved in this process, there is no need for a heavy acidic purification step. This means the MWNT can be synthesised with few defects.

II.2.2.1 Synthesis in liquid nitrogen

One possible economical route to highly crystalline MWNTs is the arc- discharge method in liquid nitrogen “Jung et al.(2003)”. By this route, indeed, mass production is also possible. For this option low pressures and expensive inert gasses are not needed (Fig. II.4).

and expensive inert gasses are not needed (Fig. II.4). Figure II.4 :Schematic drawings of the arc

Figure II.4:Schematic drawings of the arc discharge apparatus in liquid nitrogen.

The content of MWNTs can be as high as 70 % of the reaction product. Analysis with Auger spectroscopy revealed that no nitrogen was incorporated in the MWNTs. There is a strong possibility that SWNTs can be produced with the same apparatus under different conditions.

II.2.2.2 Magnetic field synthesis

Synthesis of MWNTs in a magnetic field “Anazawa et al. (2002)” gives defect-free, high purity MWNTs that can be applied as nanosized electric wires for device fabrication. In this case the arc discharge synthesis was controlled by a magnetic field around the arc plasma (figure II.5).

Carbon nanotubes synthesis technique

Extremely pure graphite rods (purity > 99.999 %) were used as electrodes. Highly pure MWNTs (purity > 95 %) were obtained without further purification, which disorders walls of MWNTs.

further purification, which disorders walls of MWNTs. Figure II.5 : Schematic diagram of the synthesis system

Figure II.5: Schematic diagram of the synthesis system for MWNTs in a magnetic field.

of the synthesis system for MWNTs in a magnetic field. Figure II.6 : SEM images of

Figure II.6: SEM images of MWNTs synthesised with (a) and without (b) the magnetic field.

II.2.2.3 Plasma rotating arc discharge

A second possible economical route to mass production of MWNTs is synthesis by plasma rotating arc discharge technique “Lee et al.(2002)”.

Chapter II

The centrifugal force caused by the rotation generates turbulence and accelerates the carbon vapour perpendicular to the anode. In addition, the rotation distributes the micro discharges uniformly and generates stable plasma. Consequently, it increases the plasma volume and raises the plasma temperature. At a rotation speed of 5000 rpm a yield of 60 % was found at a temperature of 1025 °C without the use of a catalyst. The yield increases up to 90% after purification if the rotation speed is increased and the temperature is raised to 1150 °C. The diameter size was not mentioned in this publication.

The diameter size was not mentioned in this publication. Figure II.7: Schematic rotating electrode system. II.3

Figure II.7: Schematic rotating electrode system.

II.3 Laser ablation

In 1995, the Smalley's group “Guo et al.(1995)” at Rice University reported the synthesis of carbon nanotubes by laser vaporisation., with the apparatus shown in Figure II.8. A pulsed “Yudasaka et al.(1999), Eklund et al.(2002)” or continuous “Maser et al.(1998), Bolshakov et al.(2002)” laser is used to vaporise a graphite target in an oven at 1200 °C. The main difference between a continuous and a pulsed laser, is that the pulsed laser demands a much higher light intensity (100 kW/cm 2 compared to 12 kW/cm 2 ). The oven is filled with helium or argon gas in order to keep the pressure at 500 Torr. A very hot vapour plume forms, then expands and cools rapidly. As the vaporised species cool, small carbon molecules and atoms quickly condense to form larger clusters, possibly including fullerenes. The catalysts also begin to condense, but more slowly at first, and attach themselves to the carbon clusters and prevent their closing into cage structures “Scott et al.(2001)”. Catalysts may even open cage structures when they attach to them. From these initial clusters, tubular molecules grow into single-wall

Carbon nanotubes synthesis technique

carbon nanotubes until the catalyst particles become too large, or until conditions have cooled sufficiently that carbon no longer can diffuse through or over the surface of the catalyst particles. It is also possible that the particles become well coated with a carbon layer that they cannot absorb more and the nanotube stops growing. The SWNTs formed in this case are bundled. There are some striking, but not exact similarities, in the comparison of the spectral emission of excited species in laser ablation of a composite graphite target to that of laser-irradiated C60 vapour. This suggests that fullerenes are also produced by laser ablation of catalyst-filled graphite, as is the case when no catalysts are included in the target. However, subsequent laser pulses excite fullerenes to emit C2 that adsorbs on catalyst particles and feeds SWNT growth. However, there is insufficient evidence to conclude this with certainty “Scott et al.(2001)” Laser ablation is almost similar to arc discharge, since the optimum background gas and catalyst mix is the same as in the arc discharge process. This might be due to very similar reaction conditions required, and the reactions probably occur with the same mechanism.

and the reacti ons probably occur with the same mechanism. Figure II.8 : Schematic drawings of

Figure II.8: Schematic drawings of a laser ablation apparatus.

Chapter II

II.3.1 SWNT versus MWNT

The condensates obtained by laser ablation contain carbon nanotubes and carbon nanoparticles. In the case of pure graphite electrodes, MWNTs would be synthesised, but uniform SWNTs could be synthesised if a mixture of graphite with Co, Ni, Fe or Y was used instead of pure graphite. SWNTs synthesised this way exist as 'ropes', see Figure II.9 “Maser et al.(1998), Scott et al.(2001)”. Laser vaporisation results in a higher yield for SWNT synthesis and the nanotubes have better properties and a narrower size distribution than SWNTs produced by arc-discharge. Nanotubes produced by laser ablation are purer (up to about 90 % purity) than those produced in the arc discharge process. The Ni/Y mixture catalyst (Ni/Y is 4.2/1) gave the best yield.

Ni/Y mixture catalyst (Ni/Y is 4.2/1) gave the best yield. Figure II.9 : TEM images of

Figure II.9: TEM images of a bundle of SWNTs catalysed by Ni/Y (2:0.5 at. %) mixture, produced with a continuous laser.

The size of the SWNTs ranges from 1-2 nm, for example the Ni/Co catalyst with a pulsed laser at 1470 °C gives SWNTs with a diameter of 1.3-1.4 nm “Yudasaka et al.(1999)” In case of a continuous laser at 1200 °C and Ni/Y catalyst (Ni/Y is 2:0.5 at. %), SWNTs with an average diameter of 1.4 nm were formed with 20-30 % yield (see FigureII.9. “Maser et al.(1998)”.

Carbon nanotubes synthesis technique

II.3.2 Large scale synthesis of SWNT

Because of the good quality of nanotubes produced by this method, scientists are trying to scale up laser ablation. However the results are not yet as good as for the arc-discharge method, but they are still promising. In the next two sections, two of the newest developments on large-scale synthesis of SWNTs will be discussed. The first is the ‘ultra fast pulses from a free electron laser’ “Eklund et al.(2002)” method, the second is ‘continuous wave laser-powder’ method " Bolshakov et al.(2002)”. Scaling up is possible, but the technique is rather expensive due to the laser and the large amount of power required.

II.3.3 Ultra fast Pulses from a free electron laser (FEL)method

While usually the pulses in an Nd:YAG system have width of approximately 10 ns, in this FEL system the pulse width is ~ 400 fs. The repetition rate of the pulse is enormously increased from 10 Hz to 75 MHz. To give the beam the same amount of energy as the pulse in an Nd:YAG system, the pulse has to be focused. The intensity of the laser bundle behind the lens reaches ~5 x 10 “Damnjanovic et al.(1999)” W/cm 2 , which is about 1000 times greater than in Nd:YAG systems. “Eklund et al.(2002)”. A jet of preheated (1000 °C) argon through a nozzle tip is situated close to the rotating graphite target, which contains the catalyst. The argon gas deflects the ablation plume approximately 90° away from the incident FEL beam direction, clearing away the carbon vapour from the region in front of the target. The produced SWNT soot is collected in a cold finger. This process can be seen in Figure II.10. The yield is 1,5 g/h, which is at 20 % of the maximum power of the not yet upgraded FEL. If the FEL is upgraded to full power and is working at 100 % power, a yield of 45 g/h could be reached since the yield was not limited by the laser power. Using this method the maximum achievable yield with the current lasers is 45 g/h, with a NiCo or NiY catalyst, in an argon atmosphere at 1000 °C and a wavelength of ~3000 nm. The SWNTs produced in bundles of 8-200 nm and a length of 5-20 microns has a diameter range 1-1.4 nm. “Eklund et

al.(2002)”

II.3.4 Continuous wave laser-powder method

This method is a novel continuous, highly productive laser-powder method of SWNT synthesis based on the laser ablation of mixed graphite and metallic catalyst powders by a 2-kW continuous wave CO2 laser in an argon or nitrogen stream. Because of the introduction of micron-size particle powders, thermal conductivity losses are significantly decreased compared to laser heating of the bulk solid targets in known laser techniques. As a result, more effective utilisation of the absorbed laser power for material

Chapter II

evaporation was achieved. The set-up of the laser apparatus is shown in Figure II.11 “Bolshakov et al.(2002)”.

is shown in Figure II.11 “Bolshakov et al. (2002)”. Figure II.10: Schematic drawings of the ultra

Figure II.10: Schematic drawings of the ultra fast-pulsed laser ablation apparatus.

of the ultra fast-pulsed laser ablation apparatus. Figure II.11 : (Left) The principle scheme of the

Figure II.11: (Left) The principle scheme of the set-up for carbon SWNT production by continuous wave laser powder method (Right) HRTEM of a SWNT-bundle cross-section.

Carbon nanotubes synthesis technique

II.4 Chemical vapour deposition

Chemical vapour deposition (CVD) synthesis is achieved by putting a carbon source in the gas phase and using an energy source, such as a plasma

or a resistively heated coil, to transfer energy to a gaseous carbon molecule. Commonly used gaseous carbon sources include methane, carbon monoxide and acetylene. The energy source is used to “crack” the molecule into reactive atomic carbon. Then, the carbon diffuses towards the supported metal catalyst (usually a first row transition metal such as Ni, Fe or Co)

Carbon nanotubes are formed if the proper parameters are maintained. Excellent alignment, as well as positional control on nanometre scale, can be achieved by using CVD. Control of the diameter, as well as the growth rate

of the nanotubes can also be maintained. The appropriate metal catalyst can

preferentially grow single rather than multi-walled nanotubes “Sinnot et

al.(1999)”.

CVD carbon nanotube synthesis is essentially a two-step process consisting of a catalyst preparation step followed by the actual synthesis of the

nanotube. The catalyst is generally prepared by sputtering a transition metal onto a substrate and then using either chemical etching or thermal annealing

to induce catalyst particle nucleation. Thermal annealing results in cluster

formation on the substrate, from which the nanotubes will grow. Ammonia may be used as the etchant “Ren et al.(1998a), Ren et al.(1999b), Yudasaka et al.(1995a), Yudasaka et al.(1997b)”. The temperatures for the synthesis of nanotubes by CVD are generally within the 650–900 °C range “Ren et al.(1998a), Ren et al.(1999b), Yudasaka et al.(1995a), Yudasaka et al.(1997b)”. Typical yields for CVD are approximately 30%. These are the basic principles of the CVD process. In the last decade, different techniques for the carbon nanotubes synthesis with CVD have been developed, such as plasma enhanced CVD, thermal chemical CVD, alcohol catalytic CVD, vapour phase growth, aero gel-supported CVD and laser assisted.

II.4.1 Plasma enhanced chemical vapour deposition

The plasma enhanced CVD method generates a glow discharge in a chamber or a reaction furnace by means of high frequency voltage being applied to both electrodes. Figure II.12 shows a schematic diagram of a

typical plasma CVD apparatus with a parallel plate electrode structure.

A substrate is placed on the earthed electrode. In order to form a uniform

film, the reaction gas is supplied from the opposite plate. Catalytic metals, such as Fe, Ni and Co, are used typically on Si, SiO2, or glass substrates

using thermal CVD or sputtering. After nanoscopic fine metal particles are formed, carbon nanotubes will be grown on the metal particles on the substrate by the glow discharge generated from high frequency power.

Chapter II

Chapter II Figure II.12: Schematic diagram of plasma CVD apparatus. A carbon containing reaction gas, such

Figure II.12: Schematic diagram of plasma CVD apparatus.

A carbon containing reaction gas, such as C2H2, CH4, C2H4, C2H6, CO is supplied to the chamber during the discharge “Chen et al.(2002)”. The catalyst has a strong effect on the nanotube diameter, growth rate, wall thickness, morphology and microstructure. Ni seems to be the most suitable pure-metal catalyst for the growth of aligned multiwalled carbon nanotubes (MWNTs) “Huang et al.(2002)”. The diameter of the MWNTs is approximately 15 nm. The highest yield of carbon nanotubes achieved was about 50% and was obtained at relatively low temperatures (below 330° C) “Chen et al.(2002)”

II.4.2 Thermal chemical vapour deposition

In this method Fe, Ni, Co or an alloy of the three catalytic metals is initially deposited on a substrate. After the substrate is etched in a diluted HF solution with distilled water, the specimen is placed in a quartz boat. The boat is positioned in a CVD reaction furnace, and nanometre-sized catalytic metal particles are formed after an additional etching of the catalytic metal film using NH3 gas at a temperature of 750 to 1050 °° C. As carbon nanotubes are grown on these fine catalytic metal particles in CVD synthesis, forming these fine catalytic metal particles is the most important process. Figure II.13 shows a schematic diagram of thermal CVD apparatus in the synthesis of carbon nanotubes.

Carbon nanotubes synthesis technique

Carbon nanotubes synthesis technique Figure II.13: Schematic diagram of thermal CVD apparatus. When growing carbon

Figure II.13: Schematic diagram of thermal CVD apparatus.

When growing carbon nanotubes on a Fe catalytic film by thermal CVD, the diameter range of the carbon nanotubes depends on the thickness of the catalytic film. By using a thickness of 13 nm, the diameter distribution lies between 30 and 40 nm. When a thickness of 27 nm is used, the diameter range is between 100 and 200 nm. The carbon nanotubes formed are multiwalled “Park et al.(2002)”.

II.4.3 Alcohol catalytic chemical vapour deposition

Alcohol catalytic CVD (ACCVD) is a technique that is being intensively developed for the possibility of large-scale production of high quality single wall nanotubes SWNTs at low cost. In this technique, evaporated alcohols, methanol and ethanol, are been utilised over iron and cobalt catalytic metal particles supported in zeolite. CNT generation is observed at a relatively low minimum temperature of about 550 °C. It seems that hydroxyl radicals, coming from reacting alcohol to catalytic metal particles, remove carbon atoms with dangling bonds, which are obstacles in creating high-purity SWNTs. The diameter of the SWNTs is about 1 nm. Figure II.14 shows the ACCVD experimental apparatus. The lower reaction temperature and the high-purity material obtained by ACCVD technique guarantee the possibility of scaling production up at low cost. Furthermore, the reaction temperature, which is lower than 600 °C, ensures that this technique is easily applicable for the direct growth of SWNTs on semiconductor devices already patterned with aluminium “Maruyama et al.(2003)”.

Chapter II

Chapter II Figure II.14: ACCVD experimental apparatus . II. 4.4 Vapour phase growth Vapour phase growth

Figure II.14: ACCVD experimental apparatus.

II. 4.4 Vapour phase growth

Vapour phase growth is a synthesis method of carbon nanotubes, directly supplying reaction gas and catalytic metal in the chamber without a substrate “Lee et al.(2002)”. Figure II.15 shows a schematic diagram of a vapour phase growth apparatus. Two furnaces are placed in the reaction chamber. Ferrocene is used as catalyst. In the first furnace, vaporisation of catalytic carbon is maintained at a relatively low temperature. Fine catalytic particles are formed and when they reach the second furnace, decomposed carbon is absorbed and diffused to the catalytic metal particles. Here, the growth of carbon nanotubes occurs The diameter of the carbon nanotubes by using the vapour phase growth are in the range of 2 – 4 nm for SWNTs “Ge et al.(1994)” and between 70 and 100 nm for MWNTs “Lee et al.(2002)”.

between 70 and 100 nm for MWNTs “Lee et al. (2002)”. Figure II.15: S chematic diagram

Figure II.15: Schematic diagram of a vapour phase growth apparatus

Carbon nanotubes synthesis technique

II.4.5 Aero gel-supported chemical vapour deposition

In this method SWNTs are synthesised by the disintegration of carbon monoxide on an aero gel-supported Fe/Mo catalyst. There are many important factors that affect the yield and quality of SWNTs, including the surface area of the supporting material, reaction temperature and feeding gas. Because of the high surface area, the porosity and ultra-light density of the aero gels, the productivity of the catalyst is much higher than in other methods “Su et al.(2000)” . After a simple acidic treatment and oxidation process high purity (>99%) SWNTs can be obtained. When using CO as the feeding gas the yield of the nanotubes is lower but the overall purity of the materials is very good. The diameter distribution of the nanotubes is between 1.0 nm and 1.5 nm. The optimal reaction temperature is approximately 860 °C. In laser-assisted thermal CVD (LCVD) a medium power, continuous wave CO2 laser, which was perpendicularly directed onto a substrate, pyrolyses sensitised mixtures of Fe(CO)5 vapour and acetylene in a flow reactor. The carbon nanotubes are formed by the catalysing action of the very small iron particles. Figure II.16 shows the experimental set-up for laser-assisted CVD “Alexandrescu et al.(2003)”. By using a reactant gas mixture of iron pentacarbonyl vapour, ethylene and acetylene both single- and multi-walled carbon nanotubes are produced. Silica is used as substrate. The diameters of the SWNTs range from 0.7 to 2.5 nm. The diameter range of the MWNTs is 30 to 80 nm . “Alexandrescu et al.(2003)”.

II.4.6 CoMoCat process

In this method, SWNTs are grown by CO disproportionateness at 700 – 950 °C. The technique is based on a unique Co-Mo catalyst formulation that inhibits the sintering of Co particles and therefore inhibits the formation of undesired forms of carbon that lower the selectivity. During the SWNT reaction, cobalt is progressively reduced from the oxidic state to the metallic form. Simultaneously, molybdenum is converted to the carbidic form (Mo 2 C). Co acts as the active species in the activation of CO, while the role of the Mo is possibly dual.

Chapter II

Chapter II Figure II.16 : Experimental set-up for laser-assisted CVD It would stabilise Co as a

Figure II.16: Experimental set-up for laser-assisted CVD

It would stabilise Co as a well-dispersed Co 2+ avoiding its reduction and would act as a carbon reservoir to moderate the growth of carbon inhibiting the formation of undesirable forms of carbon “Bachilo et al. (2003)”. It has been found that one of the critical conditions for an effective reactor operation is that the space velocity has to be high enough to keep the CO conversion as low as possible. Figure II.17 shows a fluidised bed reactor for a CoMoCat process. The most important advantage of fluidised bed reactors is that they permit continuous addition and removal of solid particles from the reactor, without stopping the operation. This method can be scaled-up without losses in SWNT quality. By varying the operation conditions, SWNTs can be produced with different diameter ranges. The CoMoCat catalyst has a high selectivity towards SWNTs, namely 80 – 90 %.

Carbon nanotubes synthesis technique

Carbon nanotubes synthesis technique Figure II.17: Schematic diagram of a CoMoCat apparatus 31

Figure II.17: Schematic diagram of a CoMoCat apparatus

Chapter III Experimental I Experimental Plant

III.1 Introduction

To study carbon nanotubes catalytic growth a semi-continuous experimental plant was implemented and set up to allow the continuous analysis of gaseous streams containing the carbon source. In the difference from literature, a vertical quartz micro-reactor where a catalytic fixed bed is crossed by reaction feed was employed. Continuous analysis of gaseous streams, realised with NDIR continuous analysers, permitted to perform carbon mass balance, to evaluate hydrocarbons conversion, the presence of by-products, catalysts performances and lifetime.

III.2 Experimental Plant for nanotubes synthesis

A diagram of

the experimental plant for CNT syntheses is shown in Fig

III.1. It consists of three sections:

feed section

reaction section

analysis section

All the gas pipes (¼’’ ed) are of Teflon, connections are made with Swagelok union and two, three and four way Nupro valves. All the connections are in stainless steel to avoid any corrosion due to the presence of water in the reaction products. The feed section allows the feeding of the carbon source using N 2 as the carrier gas. For each gas a mass flow controller (MFC) was used, in order to assure a constant flow rate. Each gas came from cylinders (SOL SPA) with a high purity degree.

Chapter III

Brooks measured flow controller (MFC) are used, able to operate with a maximum pressure drop of 3 atm. The working principle of the MFC is heat transport: the temperature difference in a capillary, where a part of the gas is split, it is measured. This temperature difference is proportional to the amount of heat adsorbed by the mass gas for the equation:

ΔT = K • C p Φ m where:

ΔT = temperature difference. C p = specific heat of the gas. K = dimensional constant. Φ m = mass flow. The instrument’s temperature detector produces an electrical signal from 0 to 5 V (c.c.); this signal is sent to the control unit (MFC C.U.) which converts the signal in volumetric flow. This control unit allows the mass flow of the gases to be regulated. The reaction section allows the preparation of carbon nanotubes in a reactor, consisting of a quartz tube of 300 mm length and 16 mm internal diameter. A portion of the reactor, placed in a vertical furnace, was filled with a thin layer of catalyst particles. An external quartz tube, internal diametre 35 mm, permitted the preheating of the reactants stream. In order to measure the temperature inside the reactor, a thermocouple is placed on a 4 mm internal diameter quartz shield, coaxial to the reactor. The reactor is set in a vertical electrically heated oven. The oven temperature and the control parameters are adjusted by a temperature programmer-controller, connected to a type K thermocouple located inside the reactor. A temperature reader connected to a second thermocouple measures the temperature of the catalytic bed inside the reactor. In the reaction section a system of valves allows the reactants to go to the reactor, and the products to the analysis section, or, in the bypass position, the reactants to the analysis section to verify the reactant composition. In bypass mode an independent line is used to send air or nitrogen to the reactor for the thermal pretreatment of the catalysts, or to wash the reactor. For catalytic test in the presence of water, vapour is added into the reactants by a glass saturator closed with a porous set immersed in distilled H 2 O. He is sent to the reactor through the saturator, and by changing temperature and He flow, it is possible to obtain different concentrations in the reaction feed. In the analysis section continuous analysers (ABB) Fig. III.2 mean that C 2 H 4 , C 2 H 2 , CH 4 (URAS 14) and H 2 (Caldos 17) concentrations in the effluent stream on line during the reaction can be monitored. The URAS 14 (Infrared analyser module) uses a NDIR absorption process, which is based on resonance absorption at the characteristic vibration rotation spectrum bands of non-elemental gases in the middle

Experimental I – Experimental plant

infrared range. Because of their bipolar moment, the gas molecules interact with infrared emissions. The photometer is characterized by high stability and selectivity. The analyzer has gas-filled opto-pneumatic detectors which have been optimized for each application. This enables higher sensitivity, a wider range of sample components and reduced cross-sensitivity to interfering components. Detector filling corresponds to the gas being measured. This means that the detector provides optimal sensitivity and high selectivity for the component of interest. Calibration is possible via the internal calibration cells which do not require expensive bottled test gas mixtures. This greatly reduces operation and maintenance costs.

This greatly reduces operation and maintenance costs. BYPASS M.F.C He ANALYZERS C 2 H 4 /He
BYPASS M.F.C He ANALYZERS C 2 H 4 /He C 2 H 2 /He C
BYPASS
M.F.C
He
ANALYZERS
C 2 H 4 /He
C 2 H 2 /He
C H
CH 4 /He
VENT
C 2 H 2
H 2
T.C
o
O O
O o
o
1 2
o O
O o
o
O o
3
4
CH4
MFC control unit
H 2 O
N2/ Air
H2
COMPUTER

Figure III.1: Experimental plant

The Thermal conductivity analyzer module Caldos 17 exploits the varying thermal conductivities of different gases. Measurement is carried out by means of a silicon sensor that allows especially rapid measurement and extremely small measuring ranges.

Chapter III

Chapter III Figure III.2: Uras 14 and Caldos 17 ABB analysers 36

Figure III.2: Uras 14 and Caldos 17 ABB analysers

Chapter IV Experimental II of alumina phases and

process parameters in the

MWNT growth

Effect

IV.1 Introduction

The CCVD of hydrocarbons over supported metal catalysts is an effective way to synthesise carbon nanotubes. The peculiar ability of metals to promote the CNT growth is related to their catalytic activity for the decomposition of the carbon feedstock (typically a hydrocarbon), that precedes the CNT growth “Ebbesen et al.(1992), Ciambelli et al.(2005), Kumar et al.(2005), Das et al.(2006), Yudasaka et al.(1995), Li et al.(2005)”. The role of the support material is to disperse the active phase, provide the porous structure, prevent sintering, improve mechanical strength and promote catalysis. Although these properties have been partially investigated their influence is not yet clear “Dupuis et al.(2005)”, since the active phase can yield completely different results upon CCVD when supported on different materials. Alumina and aluminium hydroxides have been recently reported as very effective catalytic supports for CNT preparation “Nagaraju et al.(2002), Kathyayini et al.(2006), Buang et al.(2006), Aghababazadeh et al.(2006), Kibria et al.(2001)”. In this perspective, different aluminium hydroxides have been investigated as supports of Co, Fe for the synthesis of CNT, to find out the most effective catalyst and to investigate the influence of the support nature on the catalyst activity. CNT characterization by thermogravimetry, Raman spectroscopy, scanning and transmission electron microscopy, has been performed. Very high CNT selectivity at high carbon and reaction yields has been obtained with gibbsite and bayerite.

Chapter IV

IV.2 Experimental

Co, Fe catalysts (2.5 wt.% of each metal) were prepared by dry (D) or wet (W) impregnation with cobalt acetate and iron acetate solution of (Ba)

bayerite (α-Al(OH) 3 ), (B) boehmite (γ-AlO(OH)), (G) gibbsite (γ-Al(OH) 3 ) and for comparison (γA) γ-alumina (γ-Al 2 O 3 ). The samples were successively dried at 393 K for 20 or 720 min. For CNT synthesis, a mixture of ethylene in helium (10% v/v) was fed to a continuous flow microreactor for 10, 30, 60 and 120 min. The gas flow rate was 78 or 200 (stp) cm 3 /min, the catalyst mass 100, 200, 250 or 300 mg. Gradual heating and fast heating were used as procedures for catalysts preheating. The composite samples obtained from the synthesis are denominated as indicated in Table IV.1. In gradual heating (GH): reactor temperature increases from 298 K up to

973 K (10 K/min) under helium flow, then reaction gas is introduced for 10,

30, 60 and/or 120 min. In fast heating (FH) reactor temperature is kept constant at 298 K for 4 min and at 973 K for 6 min under helium flow. Four reaction times were set:

10, 30, 60 and/or 120 min. The composite samples obtained from the synthesis are denominated as indicated in Table IV.1.

Thermogravimetric analysis (TGDTG) at 10 K/min heating rate in flowing air was performed with a SDTQ 500 Analyser (TA Instruments). Scanning electron microscopy (SEM) pictures were obtained with a LEO

420 microscope.

Transmission electron microscopy (TEM) images were performed using a Philips CM30T electron microscopy with a LaB6 filament as source of

electrons operated at 300 kV. Raman spectra were obtained at room temperature with a microRaman spectrometer Renishaw inVia (514 nm excitation wavelength).

IV.2.1 TG, DTG evaluation

Deposited carbon yield, reaction yield and CNT selectivity (Tab.IV.2) were calculated “Ciambelli et al.(2004b), (2005c)” with the help of TG- DTG curves reported in Fig.IV.1. In fact, in air flow, carbon combustion occurs in the range 350-700°C. The TG oxidation profiles show mainly two weight losses: the first due to amorphous carbon, that burns at lower temperature, the second attributed to CNT combustion. In particular, carbon deposit yield and reaction yield were calculated:

% carbon deposit yield = 100x

m

cd

m

cat

(1)

where m cat is the dry catalyst weight and m cd the deposited carbon mass.

Experimental II

The amount of deposited carbon was evaluated as the weight loss in the temperature range where carbon burns in air, and m cat is the weight left at 800 °C.

% reaction yield = 100 x

m

cd

m

ci

(2)

where m cd is the deposited carbon mass (TG evaluated) and m ci the mass of fed carbon.

MWNT selectivity=100·m 2 / (m 2 +m 1 )

(3)

where m 1 and m 2 are respectively the mass loss in the first and second oxidation steps during TG experiments. The amounts m 1 and m 2 were evaluated by fitting the DTG profiles with two Gaussian curves centred on the DTG peaks and measuring the areas under the curves, that are proportional to the mass losses. For the TG experiments, 5 mg of reaction product was used.

Table IV.I: Synthesis conditions with the different supports, carbon yield, MWNT selectivity and reaction yield obtained by TG evaluation.

Sample

Catalyst

Drying time

Reaction

Catalyst

Feed

preparation a

(min) d

time

mass

flow

- Reaction

(min)

(mg)

(Ncc/min)

pretreatment

(b) (c)

BWs30100

200

W-GH W-GH W-GH W-GH D-GH W-GH D - FH W-GH D-FH D-FH D-FH D-FH D-FH D-FH D-FH D-FH W-GH D-FH W-GH D-FH D-FH

S

30

100

200

BWs60100

200

S

60

100

200

BWs30200

200

S

30

200

200

BWs60200

200

S

60

200

200

BDl60200 200 BWs60200 78 BDl60200 78 BWs60300 200 e

L

60

200

200

S

60

200

78

L

60

200

78

S

30

300

200

GDl60100

78

L

60

100

78

GDl60200

78

l

60

200

78

GDl60250

78

l

60

200

78

GDl60300

78

l

60

300

78

GDl10200

78

l

10

200

78

GDl30200

78

l

30

200

78

GDl120200 78

l

120

200

78

GDl60200 200

l

60

200

200

GWs60200 200

s

60

200

200

BaDl60200 78

l

60

200

85

BaWs60200 78

s

60

200

200

γADl60200

78

l

60

200

85

γADl60200

200

l

60

200

200

γAWs60200 200 W-GH

s

60

200

200

a) Wet impregnation (W); Dry impregnation (D).

b) Gradual heating (GH):

c) Fast heating (FH)

d) 15 min drying catalyst time (s), 720 min drying catalyst time (l).

e) BWs60300200 has been prepared by adding water 10% v/v to the feed flow.

Chapter IV

Table IV.2: Synthesis conditions with the different supports, carbon yield, MWNT selectivity and reaction yield obtained by TG evaluation

Sample

Carbon yield (%)

MWNT

Reaction

 

Selectivity

Yield

(%)

(eq.3)

(%)

(eq.2)

BWs30100

200

66,7

61,6

7,8

BWs60100

200

86,0

65,8

5,0

BWs30200

200

116,9

69,1

27,3

BWs60200

200

252,6

73,5

29,5

BDl60200 200 BWs60200 78 BDl60200 78 BWs60300 200 e

89,4

61,1

10,4

135,2

68,8

40,4

40.3

39.2

12,1

71.0

-

12.4

GDl60100

78

189.9

~ 100

24.6

GDl60200

78

278.4

~ 100

72.2

GDl60250

78

226.5

~ 100

73.4

GDl60300

78

190.4

~ 100

74.0

GDl10200

78

47.8

~ 100

74.3

GDl30200

78

150

~ 100

77.8

GDl120200 78

282.1

~ 100

36.6

GDl60200 200

163.2

~ 100

16.5

GWs60200 200

105

~ 100

10.6

BaDl60200 78

260.1

~ 100

69.3

BaWs60200 78

107.2

~ 100

11.7

γADl60200

78

71.8

29.7

25.8

γADl60200

200

79.5

30.5

11.1

γAWs60200 200

34.5

53.8

4.8

For the different prepared catalysts results are summarised in Table IV.2. Boehmite samples. For B samples obtained on wet impregnated catalyst (BWs) both carbon yield and CNT selectivity increase by increasing catalyst mass and synthesis time, while reaction yield decreases. Larger catalyst mass gives higher CNT selectivity, due to better use of hydrocarbon fed, that lowers its partial pressure and homogeneous reaction rate. CNT selectivity increases with time as result of decreased amorphous carbon deposition rate. Reaction yield decrease is likely due to higher hydrocarbon decomposition rate than CNT growing rate. Sample BWs60300 200 , obtained by adding water 10% v/v in the gas stream, showing a single DTG peak at low temperature (Fig.IV.1a-insert), has been used as a carbon sample material which doesn't contains nanotubes. Gibbsite samples. Carbon oxidation occurs in a single loss step (Fig.IV.1b). The DTG peaks are sharper and shifted to higher temperatures compared to B samples, due to the high purity and quality of CNTs. The reaction yield (Tab.IV.1) is higher than for B samples. Increasing of reaction time results in enhanced carbon yield and higher degree of graphitization, which could be explained by up-shifted DTG peak temperature. A reaction time of 120 minutes causes abrupt reduction of reaction yield, indicating decreased catalytic activity. Moreover in table IV.2 we can observe that the carbon yield was higher at a catalyst mass of 200 mg while it was reduced by further mass increase.

Experimental II

Bayerite

samples.

Very

good

performances

are

given

by

bayerite

supported syntheses showing similar yields and selectivity to those obtained with gibbsite supported ones. TG results (Fig. IV.1c) and Raman spectra (not reported) of Ba samples indicate a CNT purity quite equivalent to G samples. Г-alumina samples. Performances comparison and characterization of the reaction products show that aluminum hydroxides are better supports than aluminum oxide for CNT CCVD. The higher performances of catalyst based on aluminium hydroxides are probably due to the effective interaction of the -OH rich surface with metal particles precursor during preparation phase that enhances the grafting of metal particles to the support surface, after catalyst pretreatment. The comparison of the results relevant to gibbsite and boehmite (Table IV.1) indicates that starting from gibbsite, longer catalyst drying time, dry impregnation and fast heating are preferable operating conditions. However, for boehmite the best operating conditions are wet impregnation of the support, a short catalyst drying time and gradual heating reaction pretreatment. These different conditions are clear evidence of the role of support in the catalyst for CNT synthesis. Further work is necessary to confirm the promising performance of bayerite. A further observation is that all hydroxides are transformed to oxides under CNT synthesis conditions. A systematic investigation of the three catalytic systems is currently in progress. Our syntheses, obtained through gibbsite and bayerite supported Co and Fe, exhibit high values of carbon and reaction yield, in particular considering the use of less reactive ethylene as carbon source in comparison to aluminium hydroxide Co and Fe supported synthesis under acetylene flow “Nagaraju et al.(2002), Kathyayini et al.(2006)”. In scientific publications there are several examples of CVD CNT synthesis on similar catalysts “Kumar et al.(2005), Das et al.(2006), Yudasaka et al.(1995), Li et al.(2005), Dupuis et al.(2005), Nagaraju et al.(2002), Kathyayini et al.(2006)”, also characterized by high carbon nanotubes yield. However, a comparison between the different results is not immediate because they were obtained with different carbon feedstocks, metal loadings, synthesis temperatures, reaction times, etc. The comparison of the results relevant to gibbsite impregnated with different metals loading is reported in Table IV.3.

Table IV.3: Carbon yield obtained by TG evaluation for samples prepared with different metal loading.

sample

T (°C)

% Fe

% Co

Carbon

 

yield (%)

GDl30200

GDl30200

GDl30200

78

78

78

1

700

2.5

0

1

2

700

2.5

2.5

150

4

700

5

5

35

Chapter IV

GDl30200 78 2 catalyts, at 2.5 wt% of iron and 2.5 wt% of cobalt, results in the highest carbon yield.

IV.2.2 SEM characterization

SEM images show the morphology of the tubes, allowing to measure their geometric characteristics. BWs30100 200 (Fig.IV.2a) is mostly constituted of CNT bundles. Moreover, the surface of some boehmite particles is covered by a small number of CNT and not by bundles (areas in Fig.IV.4a indicated by arrows), suggesting a partial lack of uniformity of boehmite impregnation, which cause also the deposition of amorphous carbon, as indicated by oxidation

deposition of amorphous carbon, as indicated by oxidation Figure IV.1: TG-DTG curves of samples obtained by

Figure IV.1: TG-DTG curves of samples obtained by supported synthesys with: (a) bohemite; (b) gibbsite; (c) γ-alumina; (d) bayerite.

profiles in TG tests. Increasing reaction time and catalyst mass longer CNT bundles were obtained. The image of BWs60200 200 (Fig.IV.2d) shows catalyst particles, indicated by arrows, located at the end of the nanotubes, indicating a weak interaction between active phase and boehmite. On BWs60300 200 (Fig.IV.2c) no CNT were observed, in agreement with TG results.

Experimental II

Experimental II Figure IV.2: SEM images of : BWs30100 2 0 0 (a, b); BWs60300 2

Figure IV.2: SEM images of: BWs30100 200 (a, b); BWs60300 200 (c); BWs60200 200 (d).

Fig.IV.3 of G samples shows CNT homogeneously distributed; nanotubes bundles starting from the catalyst are grown in the direction of gas flow. Catalyst particles (support + metals) are not visible. Higher resolution SEM images give evidence of the bundles organisation, consisting in a dense network of tangled carbon nanotubes. Increasing catalyst mass from 100 to 200 mg the bundle length increase up to 100 μm. A further mass increase until to 300 mg doesn’t lead to longer carbon nanotubes bundles. By comparing SEM images of Fig.IV.3c and Fig.IV.3d we can observe that starting from 10 min, a further increase of reaction time results in longer carbon nanotubes, while reaction time longer than 60 min determines only a small increase in the nanotubes length (Fig.IV.3e). The observed formation of a few agglomerates of carbon comes from homogeneous decomposition

(Fig.IV.3f).

The coiled morphology of the ends of nanotubes bundle, suggests a different catalyst deactivation mechanism . Samples from bayerite show similar morphology observed on samples from gibbsite. By contrast γ-alumina doesn’t exhibit CNT bundles.

Chapter IV

IV.2.3 Transmission electron microscopy characterization

Fig.IV.4a shows a typical TEM picture of BWs30100 200 . Nanotubes with a diameter ranging from 9 to 23 nm are visible together with amorphous carbon indicated by white arrows.

together with amorphous carbon indicated by white arrows. Figure IV.3: SEM images of: GDl60200 7 8

Figure IV.3: SEM images of: GDl60200 78 (a); GDl60300 78 (b); GDl10200 78 (c); GDl30200 78 (d), GDl120200 78 (e, f).

If we compare Fig.IV.4a and Fig.IV.4b relevant to a TG residue collected at an intermediate temperature between the two DTG peaks, we can notice that the amorphous carbon areas, observed in the first image, are absent in the second picture, giving experimental support to the attribution of the two

Experimental II

weight loss steps of TG curves. GDl60200 78 and GDl30200 78 are shown in Fig.IV.4c and Fig.IV.4d, 4e respectively. The tube have an entangled structure, a small number of catalytic particles are present in the hallow core of the nanotubes. The external diameter of GDl60200 78 is in the range 9-22 nm, while the internal diameter varies from 5 to 10 nm. Nanotubes synthesised in shorter time show a reduced outer diameter in the range 6-16 nm, as previously observed “ Ciambelli et al. (2005)”.

nm, as previously observed “ Ciambelli et al . (2005)”. Figure IV.4: TEM images of: BWs30100

Figure IV.4: TEM images of: BWs30100 200 (a); TG residue at 510°C of BWs30100 200 (b); GDl60200 78 (c); GDl30200 78 (d, e).

IV.2.4 Raman spectroscopy

Raman spectra of BW and GD samples show the typical D, G, 2×D peaks, centred at 1364, 1594 and 2702 cm -1 , respectively. The intense G peak indicates the presence of crystalline graphite carbon, while the D peak can be attributed to amorphous carbon, defects in curved graphene sheet and finite size of tubes crystalline domains. The 2×D peak, which is indicative of a clean sample “Chuang et al.(2004)”, appears at the position of doubling frequency of the D-line peak. The Raman spectra support the findings from TG analysis, as shown for example in Fig.IV.5: the I D /I G intensity ratio, a measure of disorder amount, decreases with catalyst mass increase and reaction time from 1 (BWs30100 200 ) to 0.74 (BWs60200 200 ), due to the CNT selectivity increase (Table IV.1). Moreover, after TG stopped at 510°C (i.e., after amorphous carbon burning) BWs30100 200 has an I D /I G ratio of 0.67. BWs60300 200 , that burns at low temperature, and shows (Fig.IV.5) I D /I G = 1, without the 2×D band “Chuang et al. (2004)” and a down-shift of D-line

Chapter IV

together with an up-shift of G band “Jawhari et al.(1995)”, indicating a low crystalline quality carbon (see TableIV.3 and selectivity). The I D /I G intensity ratio of G samples is very similar (~0.8) for all samples and not affected by changes in synthesis time and catalyst mass, confirming the TG-DTG evaluations.

time and catalyst mass, confirming the TG-DTG evaluations. Figure IV.5: Raman Spectra of samples BWs30100 2

Figure IV.5: Raman Spectra of samples BWs30100 200 , TG residue at 510°C of BWs30100 200 and BWs60300 200 .

Chapter V Experimental III On-line analysis

V.1 Introduction

The most of literature on chemical vapour deposition (CCVD) method for synthesis of CNT was devoted to investigate catalyst composition and carbon sources, while few paper deal with operating conditions of CNT synthesis reactor. Many efforts have been dedicated to understand the best active phases, support, carbon precursor, flow rates, operative conditions, but in-situ diagnostic for CNT growth remains difficult to implement in a CVD reaction owing high temperatures and reactive atmospheres. To explore CNT kinetics growth a common method is to track the CNT weight or height at specific time. Several limitations are due to changing of growth rate during the time so only an average growth rate can be evaluated. In addition, overall kinetics arises also from transport effects, and must be taken into account for kinetics parameters calculations. Horizontal furnaces, with catalyst bed placed in quartz boat is the more diffused configuration for scientific investigations, but suffers of diffusive limitations, both longitudinal, along the reactor axes, and vertical, diffusion into the bed height “Gommes et al.(2004)”. With the aim to optimise the process, a large range of operative conditions have been explored in a more “classic” fixed bed with an orthogonal gas flow. Particular attention has been devoted to parameters such as catalytic activity and the lifetime of Co-Fe-supported gibbsite, which emerged as high promising catalyst “Ciambelli et al.(2007)”, for the synthesis of MWNTs, by monitoring the effluent stream with handy and by the immediate response on-line ABB analysers. This system permits that the rection rate could be inferred from exhaust gas composition. Very high carbon and reaction yield have been obtained. Catalyst activity by employing a fixed bed with an orthogonal gas flow, was

Chapter V

higher and more stable for longer period as compared with previous published works “Ago et al., (2006); Pirard et al., (2007)”. On the nanotubes produced, a characterisation to correlate their properties with the reactor operative conditions was also performed and showed that our system results in high quality carbon nanotubes being obtained.

V.2 Experimental

The syntheses were performed in the experimental plant described in Chapter IV, equipped also with the on-line ABB analysers that permit the monitoring of C 2 H 4 , C 2 H