Production of Carbon Nanotube

The growth of carbon nanotubes during synthesis and production is believed to commence from the recombination of carbon atoms split by heat from its precursor. Due to the overwhelming interest, enormous progress is being made in the synthesis of carbon nanotubes. Although a number of newer production techniques are being invented, three main methods are the laser ablation, electric arc discharge and the chemical vapour deposition. The last is becoming very popular because of its potential for scale up production, hence it is discussed first in this section.

Table 10.2. Selected electrical and mechanical properties of carbon nanotubes (Hoenlein et al. 2003)

Characteristics

Measure

Electrical conductivity

Metallic or semiconducting

Electrical transport

Ballistic, no scattering

Energy gap (semiconducting)

Eg(eV) *1/d (nm)

Maximum current density

~1010 A/cm2

Maximum strain

0.11% at 1 V

Thermal conductivity

6 kW/Km

Diameter

1-100nm

Length

Up to millimetres

Gravimetric surface

> 1500 m2/g

E-modulus

1000 GPa, harder than steel

10.5.1 Chemical Vapour Deposition

Chemical vapour deposition (CVD) methodology in producing carbon nanotubes bears little or no difference with the conventional vapour grown carbon fibre technology. As the latter technology improved into producing thinner carbon fibres from several micrometers to less than 100 nm, it has emerged with the advent of carbon nanotubes synthesised from catalytic CVD. In both cases, carbon fibres and carbon nanotubes may be grown from the decomposition of hydrocarbons at temperature range of 500 to 12000C. They could grow on substrates such as carbon, quartz, silicon, etc or on floating fine catalyst particles, e.g. Fe, Ni, Co, etc. from numerous hydrocarbons e.g. benzene, xylene, natural gas, acetylene, to mention but few. The CVD method has been receiving continuous improvement since Yacaman et al. first used it in 1993, and in 1994, Ivanov et al. produced MWNTs. Patterned silicon wafers of porous n+ and p type plain types were used to grow regular arrays of MWNTs as shown in Fig. 10.5(a) & (b) (Fan et al. 2000), and SMNTs were grown as low and high whiskers population density fastened to the filament fibrils for carbon fibre surface treatment shown in Fig. 10.5(c & d) (Iyuke et al. 2004). SWNTs also produced from floating catalyst CVD was presented earlier (Fakhru'l et al. 2003).

The schematic of typical catalytic chemical vapour deposition system, shown in Fig. 10.6 was equipped with a horizontal tubular furnace as the reactor. The tube was made of quartz tube, 30 mm in diameter and 1000 mm in length. Ferrocene and Benzene vapours acting as the catalyst (Fe) and carbon atom precursors respectively were transported either by argon, hydrogen or mixture of both into the reaction chamber, and decomposed into the respective ions of Fe and carbon atoms, resulting into carbon nanostructures. The growth of the nanostructures occurred in either the heating zone, before or after the heating zone, which is normally operated between 5 000C and 11500C for about 30 min. The flow of hydrogen gas was 200ml/min, while Argon gas was used to cool the reactor as reported elsewhere (Dana 2004).

Fig. 10.5. MWNTs arrays grown on (a) n+ type porous silicon and (b) p type plain silicon substrates (Fan et al., 2004); while whiskers of SWNTs were grown on the surfaceof carbon fibrils (C) sparsely growing whiskers, (d) highly populated whiskers (Iyuke et al. 2004)

Fig. 10.5. MWNTs arrays grown on (a) n+ type porous silicon and (b) p type plain silicon substrates (Fan et al., 2004); while whiskers of SWNTs were grown on the surfaceof carbon fibrils (C) sparsely growing whiskers, (d) highly populated whiskers (Iyuke et al. 2004)

Substrates or samples

Fig. 10.6. Schematic of catalytic CVD operated either as floating catalyst or substrate catalyst

Substrates or samples

Fig. 10.6. Schematic of catalytic CVD operated either as floating catalyst or substrate catalyst

10.5.2 Arc Discharge

The arc discharge method produces a number of carbon nanostructures such as fullerenes, whiskers, soot and highly graphitised carbon nanotubes from high temperature-plasma that approaches 37000C (Ebbesen 1997). The first ever produced nanotube was fabricated with the DC arc discharge method between two carbon electrodes, anode and the cathode in a noble gas (helium or argon) environment (Kroto et al. 1985). Schematic representation of a typical arc discharge unit is presented in Fig. 10.7. Relatively large scale yield of carbon nanotubes of about 75% was produced by Ebbesen and Ajayan with diameter between 2 to 30 nm and length 1 |im deposited on the cathode at 100 to 500 Torr He and about 18 V DC. It has conveniently been used to produce both SWNTs and MWNTs as revealed by Transmission Electron Microscope (TEM) analysis. Typical nanotubes deposition rate is around 1mm/min and the incorporation of transition metals such as Co, Ni or Fe into the electrodes as catalyst favours nanotubes formation against other nanoparticles, and low operating temperature. The arc discharge unit must be provided with cooling mechanism whether catalyst is used or not, because overheating would not only results into safety hazards, but also into coalescence of the nanotube structure.

10.5.3 Laser Ablation

Laser ablation technique involves the use of laser beam to vaporise a target of a mixture of graphite and metal catalyst, such as Co or Ni at temperature approximately 1200oC in a flow of controlled inert gas (argon) and pressure (Fig. 10.8), where the nanotube deposits are recovered at a water-cooled collector at much lower and convenient temperature. The method was used in early report (Thess et al 1996) to produce ropes of SWNTs with remarkably uniform narrow diameters ranging from 5-20 nm, and high yield with graphite conversion greater than 70-90%.

The bundles entangled into a 2-D triangular lattice via the van der Waals bonding to achieve lattice constant equal to 1.7 nm. The metal atom (catalyst) due to its high electronegetivity, deprived the growth of fullerenes and thus a selective growths of carbon nanotubes with open ends were obtained. Changing the reaction temperature can control the tubes diameters, while the growth conditions may be maintained over a higher volume and time, when two laser pulses are employed. However, by the virtue of relative operational complexity, the laser ablation method appears to be economically disadvantageous, which in effect hampers its scale up potentials as compared to the CVD method.

Target

Fig. 10.8. Schematic of laser ablation method

Target

Fig. 10.8. Schematic of laser ablation method

10.5.4 Mechanisms of Growth

Several mechanistic pathways are being proposed for carbon nanotube growth depending on the production techniques. However, a huge ambiguity exists in an effort to adopt a generalised mechanism for products of the continuously changing synthetic methods, talk less of the type of nanotubes (SWNTs or MWNTs). Saito, Dresselhaus and co-worker referred to this scenario as different school of thoughts, where one assumes that the growth mechanism involves C2 dimers absorption to close the tube with cap that is assisted by its pentagonal defects. Another assumption is that the nanotubes get opened up during the growth while carbon atoms are added at the open ends for length propagation. In their discussion (Saito et al. 2001) specifically on arc discharge method, the proposition is that the nanotube grows axially at the open ends, and if chiral, the additions of one hexagon continuously to the open ends occur as C2 dimers are absorbed at the active dangling bond edge sites. But if the addition of carbon atoms is not according to the discussed order, the C2 dimer absorption may result into capping the tube with pentagon, while if the carbon addition involves C3 trimer, hexagon could merely be added. In SWNTs growth, armchair type would result from the absorption of a single C2 dimer to add a hexagon, while the zigzag edge requires one C3 trimer to initiate the growth. The thickening mechanism of the nanotube has been related to the corresponding vapour grown carbon fibre mechanism (Ebbesen 1997).

Moreover, in the study of the growth mechanism of carbon nanotubes synthesised by hot-filament CVD by Chen et al (2004) it was reported that after formation of metallic island from the catalyst, a stretching force elongated the metallic nanoparticles until it was broken into two parts, where one part stays at the base of the nanotube while the other remains at the tip. A relatively similar phenomenon can be induced from the work of Danna 2004 that the SWNT was nucleated on the metal catalyst nanoparticle, which broke into two parts, while one remains at the base, the other continue to propagates the elongation of the nanotube until it drops from the tip to terminate the growth process (Fig. 10.9). Fig. 10.9(a) presents with arrow 1, the non-growing open tube end, and arrow 2 points at the still growing tube tip holding the catalyst for growth propagation. While Fig. 10.9(b) presents the tube growing phase and the open terminal end when the metal catalyst drops off. The maximum length of carbon nanotube obtained was in the average of 12 |im. As proposed earlier (Chen et al. 2004), during the growth period the portion of metal catalyst at the base of the tube keeps liquid state, and thus it is easily broken into parts, which are located in spots within the length of the tubes as seen with TEM images (Dana 2004).

C2 dimer radicals addition

C2 dimer radicals addition

Fig. 10.9. Depiction of SWNTs growth mechanism (a) arrow 1 shows open-ended terminal end, while arrow 2 points at a growing end of a nanotube carrying the metal catalyst (Danna 2004) (b) presents the addition of C2 dimers to the growing end containing the metal catalyst preceding the growth termination with an open end of the nanotube.

metal catalyst

^^^^^^ Growing

^-

phase

metal catalyst

/—

j Open s-terminal end a b

Fig. 10.9. Depiction of SWNTs growth mechanism (a) arrow 1 shows open-ended terminal end, while arrow 2 points at a growing end of a nanotube carrying the metal catalyst (Danna 2004) (b) presents the addition of C2 dimers to the growing end containing the metal catalyst preceding the growth termination with an open end of the nanotube.

10.5.5 Purification of Carbon Nanotube

Most often than none, carbon nanotubes are found in the midst of other carbon materials, such as soot, other amorphous carbon, carbon nanoparticles, etc during production. In order to obtain good quality nanotubes for specific applications therefore, isolation, separation or post production purification process must be employed. Thus a post pre-treatment technique must be mastered to accompany any promising large scale production method. The basic purification methods that have been used with appreciative success are gas phase, liquid phase and intercalation techniques (Ebbessen 1997), while others that have been tried and still being developed are filtration, chromatography and centrifugation processes. The contaminants with large particle sizes or the ones agglomerated to the nanotubes can easily be treated in an ultrasonic bath to disengage the tubes from the particles (Ebbesen 1992). They could also be separated easily due to their weight differences from the nanotubes by dispersing the powder in a solvent, such as alcohol or ether and subsequently centrifuged. The gas phase treatment, commonly known as oxidation, involves treating the samples in air or oxygen at temperatures 650 to 7500C. This has been successful because it is believed that the nanotubes survive the oxidation much longer than other carbon nanoparticles, because they are longer and are consumed from the tips inwards. The liquid phase oxidation has been used, dispersing the samples in standards oxidants e.g. H2NO3, H2SO4 or KMnO4. The acidified solution of potassium permanganate has been found to be most successful in purifying and opening the end cap of the tubes (Ebbesen 1997). Intercalation post-production treatment of carbon nanotubes applies the difference in oxidation rates of graphite and intercalated graphite to remove impurities from open graphite (Ebbesen 1997; Ikazaki et al. 1994). Thus close ended nanotubes and other nanoparticles, which are not easily intercalated can be separated from graphitic flakes or other open graphitic structures. Typical treatment steps (Ikazaki et al. 1994) for an arc discharge nanotubes involved immersion of the sample in a CuCl2-KCl / molten salt mixture at 4000C for 168 hours. Excess salt was removed with deionised water after cooling and the cupper salt reduced to the metal in He/H2 medium at 5000C, heating at 100/min for an hour. Other purification techniques involve refluxing the sample in acid, centrifuged and followed with cross flow membrane filtration to remove catalyst particles and amorphous carbon. Colloidal suspension has been used as a function of particle size to purify the tube without physical damage and size exclusion chromatography was used for purification and size selection for MWNTs. Annealing at about 30000C has been employed to remove defects from carbon nanotube and improve graphitisation; and to distinguish the chiral types of SWNTs. Current can be passed between metal electrodes, whereby conducting nanotubes are burnt while the semiconducting ones remain purified (Mamalis et al. 2004).

Was this article helpful?

0 0
Brain Blaster

Brain Blaster

Have you ever been envious of people who seem to have no end of clever ideas, who are able to think quickly in any situation, or who seem to have flawless memories? Could it be that they're just born smarter or quicker than the rest of us? Or are there some secrets that they might know that we don't?

Get My Free Ebook


Post a comment