Materials made of carbon that can be synthesised and characterised at the nano level have become a mainstay in the nanotechnology arena. These carbon materials can have a remarkable range of morphologies. They can have structures that are either hollow or filled and can take many shapes, as evidenced by the well-documented families of fullerenes and carbon nanotubes. However, these are but two of the shapes that carbon can form at the nano level. In this review we outline the types of shaped carbons that can be produced by simple synthetic procedures, focusing on spheres, tubes or fibres, and helices. Their mechanisms of formation and uses are also described.

Carbon is a remarkable element and has been described as ‘the key element of living substances’.¹ It is the ability of carbon to bond to itself to form oligomers and polymers that allows carbon to play this important role in life processes. This property can also be used to produce the myriad of structures that makes carbon such an important commodity element; for example, it is the element that leads to the basis of the Fischer–Tropsch process used by Sasol in South Africa to make fuels and chemicals. But the ability of carbon to form strong bonds to oxygen to generate CO₂ and lead to a carbon sink also reveals the ‘dark side’ of carbon. The control and understanding of the bonding properties of carbon thus becomes crucial if the chemistry of carbon is to be harnessed for the good of the world’s population.

Carbon has four electrons that can be used for bonding and this determines the structural chemistry that is associated with the element. In the classical valence bond model, these four electrons (called sp³ electrons) are used to form four bonds to other atoms. In the simplest case, when the bonds only occur between carbon atoms, C-C bonds are formed and the classical structure of diamond is produced (Figure 1a). However, it is the ability of carbon to form multiple bonds between elements that gives carbon many of its unique features. In this way carbon can also link to another carbon atom to give C=C (as found in graphite, Figure 1b) and C≡C bonds (as found in acetylene, Figure 1c). The chemical and physical properties associated with the C-C, C=C and C≡C interactions are all different and the ability to control the synthesis of structures containing these units has led to an exploitation of the chemistry of carbon.

This ability to make all-carbon containing nanomaterials, in particular those containing networks of C=C double bonds, has been one of the key events that has led to the current nanotechnology revolution. The discovery of fullerene in 1985 (Figure 2a)² and the subsequent studies by Iijima³ on carbon nanotubes in 1991 were key events that have spurred the study of nanostructures in general and nanocarbon structures in particular. Through these discoveries a third allotrope of carbon, following from the graphite (sp² hybridised carbon) and diamond (sp³ hybridised carbon) allotropes, was recognised – this allotrope is based on a bent sp² hybridised carbon. These discoveries coincided with attempts to miniaturise devices (such as cell phones and computers) and the use and development of new characterisation tools (such as scanning probe microscopes and electron microscopes) to visualise these new structures. The outcome has been the emergence of the field of nanotechnology.

The most common form of oligomerised or polymerised carbon is soot (Figure 2b). Soot, produced by burning carbonaceous materials, has an amorphous structure with little long-range order. But, by controlled decomposition of carbon-containing reactants under appropriate conditions, it has been possible to make these carbons with long-range order. Control of the experimental conditions permits morphology control (shape, length, diameter, etc.) of the carbon products at the nano level and this has generated a wide range of variously shaped carbon nanomaterials (SCNMs). The synthetic approach is based on templating and self-assembly principles, similar to the processes used to grow NaCl (salt) crystals from a salt solution, or raindrops in clouds. Some of these SCNMs are shown in Figure 3 and discussed in more detail below. As can be seen in Figure 3, a wide range of structures can be made (tubes, spheres, helices and Y-junctions). Each shape influences the property of the carbon material and it is this carbon shape–property relationship that is the key to the manufacture of new devices. These new properties will also be influenced by the size of the SCNMs.

Nanocarbons have a number of remarkable properties:

1. Carbon is a light element and structures made from carbon tend to be lightweight (carbon is about six times lighter than iron).
2. Carbon in tubular form has been shown to be the strongest material synthesised to date (Young’s modulus for a single-walled carbon nanotube (SWCNT) is about 1 TPa).⁴
3. Carbon can act as conductor, semi-conductor or insulator, depending on the carbon–carbon bonding and the carbon structure.
4. Carbon in the form of diamond or SWCNTs is the hardest material known.
5. The optical studies of carbon have shown that SWCNTs have absorbances of 0.98 a.u. – 0.99 a.u. over a wide range of wavelengths, making them a near perfect black body.
6. The thermal conductivity of carbon is variable. In SWCNTs the thermal conductivity along the tube axis is ‘ballistic’ (10 times that of copper) but when perpendicular to the axis, a SWCNT is an insulator.⁵
7. The surface of carbon materials can be chemically modified (functionalised), leading to a new generation of reagents that can be used in new applications, for example, in composite materials.

Our own involvement in SCNMs dates back to the exploitation of using carbon as a catalyst support i.e. a material used to spread a metal catalyst and increase the number of metal atoms available at the surface for reaction. The finding that gold supported on carbon could be used to catalyse the reaction between ethyne (acetylene) and HCl to give CH₂=CHCl (vinyl chloride) as a monomer for polyvinylchloride synthesis⁵ led us to investigate carbon nanotubes as gold supports in the mid-1990s. This required that we develop a programme for the synthesis of carbon nanotubes, studies that have since led us to investigate the synthesis of other shaped carbons as generic catalyst support materials.

A consideration of work on carbons in South Africa (and elsewhere in Africa) has revealed little research in this area (with the exception of diamonds) prior to the mid-1990s. In South Africa, the company SA Carbide (operating out of Newcastle), has made carbon spheres for the battery market for decades. The carbon in pulp technology, for concentrating gold, has also been exploited in the mining industry since the 1990s.⁷ More recently, the possibility of making C@U core-shell spheres for a proposed Pebble Bed Modular Reactor to generate nuclear energy in South Africa has also seen an exploitation of the carbon market in the country.⁸

Other than these examples the main development of carbons in South Africa has been in the research arena, with activities focused in academic chemistry and physics departments. For example, our research group, CATOMAT, based in the School of Chemistry at the University of the Witwatersrand, is working on the synthesis of SCNMs and their application as strong materials in sensors, as catalyst supports, and in solar and fuel cells. Other groups at this university are working on new reactor designs for making SCNMs (Chemical Engineering), and studying their physical properties e.g. mechanical and electronic properties (School of Physics) or exploring their use as neuropharmaceuticals (Medical School). Research is also being performed on carbon nanotubes at many other South African universities (such as the University of Johannesburg, Tshwane University of Technology, the University of KwaZulu-Natal at Westville and Vaal University of Technology), as well as at other South African research organisations (the Centre for Nanostructured Materials at the Council for Scientific and Industrial Research, and MINTEK). The programmes range from pure synthesis to studies on the properties of SCNMs.

Research in South Africa in this area is largely funded by the South African Department of Science and Technology while nanotechnology activities in South Africa are coordinated by the South African Nanotechnology Initiative (SANi).

In the sections below we describe some of the new SCNMs that are being studied in our group. The descriptions have been categorised according to the shape of the carbon material (tubes or fibres, spheres, helices and ‘other’, which includes graphene). Indeed all the SCNMs to be discussed can be viewed as being generated from a single graphene sheet. For example, Figure 4 shows how a fullerene, a carbon nanotube or a graphite (layer or layers) can be made from a graphene sheet.⁹ The review should thus provide an introduction to this new area of solid carbon chemistry.

Figure 1: Schematic diagrams of different carbon types: (a) diamond, (b) graphite and (c) acetylene.

Figure 2: (a) A schematic representation of a fullerene model and (b) a photograph of soot.

Figure 3: A variety of shaped carbon materials: (a) a solid fibre growing from a catalyst particle, (b) a tubular structure with hollow inner, (c) a scanning electron microscopy (SEM) image of branched carbon fibres, (d) a SEM image of coiled carbon fibres, (e) a transmission electron microscopy image of a spring-like fibre and (f) a spherical carbon material.

Figure 4: Graphene is a two-dimensional building material for carbon materials of all other dimensionalities. It can be wrapped up into zero-dimensional buckyballs, rolled into one-dimensional nanotubes or stacked into three-dimensional graphite.

Carbon has the ability to form carbon nanotubes (CNTs) and carbon nanofibres (CNFs) in which the diameters are typically 1 nm – 100 nm while the lengths can range from 10 nm to a few centimetres. In addition, although typically represented as a material showing good alignment, CNTs and CNFs are generally synthesised with an appearance more like ‘cooked spaghetti’, made of interwoven strands of carbon as shown in Figure 5. It is also possible to make carbon tubes that have Y-junctions and T-junctions (Figure 2d).

Many variations of both filled and hollow tubes exist. Thus, the carbon tube or fibre can be made of V-shaped cups stacked on top of each other, or of flakes that generate a herringbone structure. Finally, the tubes can be partially layered in the tube hollow, generating a bamboo structure. (Figures 2 and 6). The ability to visualise the structures of carbon at the nano level has permitted the exploration of the synthesis and morphology of nanocarbons. Clearly, variations in the structural morphology must be related to the properties and uses of the tubes or fibres.

CNTs can be described as seamless cylinders of rolled up graphene sheets of carbon atoms.¹⁰ Depending on the number of rolled up graphene sheets, three types of CNTs have been observed: SWCNTs, double-walled CNTs (DWCNTs) and multi-walled CNTs (MWCNTs). On the other hand, CNFs are cylindrical nanostructures (Figure 6c) with graphene layers arranged as stacked cones, cups or plates, as mentioned above. The history of CNFs dates back to 1889¹¹ when CNFs were reported to be grown from carbon-containing gases using a metallic crucible as the catalyst. In contrast, the history of CNTs dates to the 1950s¹² but with the explosive study of these materials only commencing in the 1990s.³ In general, CNFs tend to be wider (> 100 nm) than CNTs (typically having internal diameters of < 50 nm) and can be synthesised at lower temperatures than required for CNT synthesis. Indeed fibres can be formed at temperatures below 250 °C; these CNFs tend to be highly amorphous.

Figure 5: Scanning electron microscopy images showing (a) aligned carbon nanofibres and (b) 'cooked spaghetti type' carbon nanofibres.

Figure 6: Animated representations of: (a) carbon nanotubes, (b) carbon nanofibres and (c) bamboo shapes.

Synthesis of CNTs and CNFs

Typically, the CCVD approach requires a catalyst or template and a carbon source to produce SCNMs, in particular CNTs. The reaction is generally performed in a horizontal reactor such as that shown in Figure 7. The reactor can also be arranged in a vertical geometry. The reactor system comprises a quartz tube inserted into a hot oven. The carbon source is then passed through the quartz tube using an appropriate carrier gas at high temperature (typically 600 °C – 1100 °C), resulting in the decomposition of the reactants and production of the SCNMs. The reaction is usually performed in the presence of a catalyst. The catalyst can either be placed in the reactor (typically supported on an inert carrier to increase surface area)¹³ or passed as a gas through the reactor (as a floating catalyst).¹⁴ While the supported catalyst method involves a catalyst dispersed on a support, the floating catalyst method normally uses volatile organometallic compounds as precursors, with Fe(CO)₅ and ferrocene typically being used in the catalytic synthesis procedure.^15,16 The floating catalyst method is advantageous because no support removal procedures are required after the reaction and the catalyst can also be introduced into the CCVD reactor as either a liquid or a gas.

Figure 7:A floating catalyst chemical vapour deposition reactor for the synthesis of shaped carbon nanomaterials.

Modification of the CNT structure

Functionalisation of CNTs

Dispersion of CNTs can be achieved by sonication. The dispersion produced can be very stable and the CNTs can remain in solution for weeks or months. Functionalisation methods, such as oxidation of the CNTs, can create more active bonding sites on the CNT surface. For biological uses, CNTs can be functionalised by attaching biological molecules such as lipids and proteins to surfaces. The CNT–biomolecule materials can be used to mimic biological processes, such as protein adsorption, the binding of DNA and drug molecules and the fixing of red blood cells. These reactions are very useful in medicine (and pharmaceutics), in particular in drug-delivery systems.

Doping of CNTs

The incorporation of heteroatoms into a CNT modifies the tube characteristics. Microscopy studies reveal that nitrogen-doped CNTs (N-CNTs) are hollow inside with a likely occurrence of bamboo compartments¹⁹ (Figure 6 and Figure 8). Incorporation of nitrogen atoms favours the formation of pentagons and heptagons and increases the reactivity of the neighbouring carbon atoms resulting in a higher degree of disorder in a N-CNT relative to a ‘pure’ CNT. Thus, considerable effort has been made to study the fabrication and the growth mechanism of the related carbon nanostructures with nitrogen as the dopant.

Figure 8: A transmission electron microscopy image of a carbon nanotube grown from ferrocene-aniline mixtures.

Growth mechanisms of CNTs and CNFs

The use of a floating catalyst also generates metal particles leading to CNT or CNF growth similar to that described above. The type of carbon formed will depend on the control of the catalyst particle morphology and catalyst particle size distribution during the reaction. For instance, a low iron to carbon ratio generates more carbon spheres and less amorphous material, while a high iron to carbon ratio produces high yields of CNTs and less amorphous material.²¹ No spheres are formed when the ratio is high, reflecting a strong iron catalyst influence during the growth. Further, large diameter CNTs originate from large catalytic nanoparticles and small diameter CNTs originate from small catalytic species. To complicate mechanistic studies for carbon growth, carbon structures can also be formed without the use of a catalyst, although more typically carbon spheres and amorphous carbon are produced in this instance.

Figure 9: Carbon nanotube mechanistic models: (a) root-growth mechanism and (b) tip-growth mechanism.

Other types of tubular carbons

Properties and uses of tubular carbons

CNTs and CNFs are also being used as catalyst supports for heterogeneous catalytic reactions. Because of their high surface area and diverse morphologies (sizes and shapes), tubular carbons have been used as supports in various catalytic reactions, for example, in Fischer–Tropsch synthesis.^26,27 These metal–carbon materials display excellent activity and selectivity when compared to conventional catalyst supports such as alumina or silica. CNTs have high aspect ratios and small tip radii of curvature as well as a two-dimensional array which is suitable for their use in electron field emission.²⁸ Because of their flexibility, SWCNTs are used in the tips of atomic force microscopy (AFM) probes to make the tip sharper, which allows for an increase in the atomic resolution of the surface being studied.²⁹ The high strength of CNFs has promoted their incorporation in polymer nanocomposites and their use as polymer fillers.³⁰

Doping of CNTs with nitrogen or boron induces novel electronic properties into the CNTs, and provides a means for tuning the field emission of CNT emitters. It has also been shown that nitrogen doping can enhance the mechanical, conducting, energy storage and electron transport properties of CNTs.

The branched CNTs have potential applications as building elements in nanoelectronic devices. For example, they can be used in a nanoscale three-terminal device or in other transistor applications.

Fullerenes

Synthesis of fullerenes
The fullerene molecule requires that the C-C bonds interact through bent sp² hybridised carbon atoms. This leads to a strained structure with good reactivity; for example, a fullerene can act as an electron acceptor. For this reason, many derivatives of fullerenes have been made; one method of making these derivatives is the Prato reaction,³³ which involves the reaction of azomethine ylides generated in situ by decarboxylation of ammonium salts derived from the thermal condensation of amino acids and aldehydes (or ketones).

In one of our own studies we have functionalised C₆₀ using the Prato reaction and reacted this functionalised C₆₀ with thiophene to generate C₆₀-polythiophene composites (via a ring-opening metathesis polymerisation reaction) for possible use in solar cell devices.³⁴

Growth mechanisms of fullerenes
The mechanism for the formation of fullerenes is still not known. A number of proposals have been made in which carbon atoms coalesce to give small structures which then grow into a fullerene by addition of more carbon atoms, dimers, etc. Some possibilities include: (1) a graphene to fullerene transformation (Figure 4), (2) the pentagon road mechanism based on growth from a curved small carbonaceous material such as corranulene, (3) ring stacking in which rings of carbons build a fullerene and (4) the fullerene road mechanism in which small fullerenes grow into larger fullerenes by addition of mainly C₂ units.³⁵

Uses of fullerenes
Some potential applications of fullerenes include use in optical devices, hardening agents for carbides, chemical sensors, gas separation devices, thermal insulators, batteries, catalysts, hydrogen storage media, polymers and polymer additives, and medical applications (e.g. fullerenes and their derivatives have potential antiviral activity, which has strong implications for the treatment of HIV infection). Various researchers in the United States are manufacturing and developing proprietary fullerene derivatives for use in diagnostics and therapeutics.

A carbon sphere (CS) generally refers to a spherical form of carbon that can be either semi-crystalline or crystalline (graphitic) and can have a solid, hollow or core shell morphology (Figure 10). The CS differs from a fullerene in that the outer carbon layer is more than one carbon layer thick. Solid carbon spheres, especially with diameters smaller than 100 nm tend to accrete and form bead- or necklace-like structures (Figure 11) The spherical carbons include carbon black, carbon onions, carbon microbeads and mesoporous carbon microbeads.³⁶

Historically, carbon black is one of the oldest forms of the spherical carbons known. Early documentation reveals that it was used for writing letters on papyrus in ancient Egypt and writing on bamboo strips in ancient China.³⁷ Carbon black is largely produced by the partial combustion and thermal decomposition of hydrocarbons such as oil or natural gas. It is an important chemical commodity used in various applications, from the black pigment of newspaper inks to the electrically conductive agent used in high-technology materials.³⁶

CSs have been categorised in four ways. Firstly, CSs, like all spherical bodies, can be classified as solid, core-shell or hollow (Figure 10).³⁸ Secondly, they can be classified according to whether the spheres are made of concentric, radial or random carbon layers (Figure 12).³⁹ Thirdly, spheres can be categorised in terms of their diameter.⁴⁰ In this method three types are recognised: (1) well-graphitised spheres (2 nm – 20 nm), (2) less graphitised spheres (50 nm – 1000 nm) and carbon beads (> 1000 nm). Fourthly, it is possible to classify spheres in terms of strategies used in their synthesis.³⁶

Figure 10: A cartoon of (a) hollow, (b) core shell and (c) filled spheres.

Figure 11: A transmission electron microscopy image of accreted filled carbon spheres.

Figure 12: A cartoon indicating (a) random, (b) concentric and (c) radial layer orientations in carbon spheres.

Synthesis of CSs

CSs are synthesised using procedures similar to those that generate CNTs and CNFs. Thus, CSs can be synthesised using arc discharge, laser ablation and plasma processes, shock compression techniques, chemical vapour deposition (catalytic and non-catalytic), autoclave processes (catalytic and non-catalytic), and by carbonisation routes.³⁶ Their diameters can range from 10 nm to 10 µm and their surface areas from 2 m²/g to > 1500 m²/g. Key issues when making CSs is to ensure monodispersity of size.

The rim of a CS with concentric structure is made up of carbon flakes. These carbon flakes are comprised predominantly of aromatic structures, made up of condensed benzene rings. These flakes can be small (< 10 nm), but after graphitisation at high temperatures these grow in size leading to structures that contain curved graphite planes. The sp² carbon atoms at the edge of the flakes will react with other atoms, such as hydrogen and oxygen atoms, to give OH and COOH groups to satisfy valence requirements. The oxygen content typically varies between 1% and 5% for low surface area CSs. CSs like CNTs have curved structures because of the presence of C5 and C7 rings. The CSs once formed can be characterised by particle size, structure and surface chemistry (Figure 13).

Figure 13: The structure of carbon black showing some functional groups on the surface of the spherical carbon structure.

Hollow and core-shell spherical CSs

Modification of the CS structure

The surface modification of CSs has been shown to increase the reactivity of these materials (e.g. solubility and wetability), thus widening their applications. Many chemical treatments have been performed on CSs such as oxidation with potassium persulphate, severe air oxidation, halogenations and grafting reactions. All the chemical procedures modify the CS surface properties and the structural changes correlate with their reactivity.⁴² It was observed that carbon blacks have different behaviour depending on their mode of preparation but they generally perform similarly, after a given modification procedure.

As with other forms of carbon, such as nanotubes, CSs can be doped with heteroatoms such as nitrogen and boron.^43,44 As expected, the doping of CSs with a heteroatom modifies their electronic and chemical properties making them suitable for a wide range of applications.

Growth mechanisms of CSs

Wang et al.⁴⁶ proposed that the sphere is nucleated from a pentagonal carbon ring followed by a spiral shell growth (Figure 14). This type of mechanism was also proposed by Kroto and McKay³¹ to explain the formation of large fullerenes.

Figure 14: (a) Nucleation of a pentagon, (b) growth of a quasi-icosahedral shell, (c) formation of a spiral shell carbon particle proposed by Kroto and McKay31 and (d) growth of a large carbon sphere.

Properties and uses of spherical carbons

CSs exhibit properties such as blackness and dispersibility when they are mixed with inks, paints, or resins. Recent studies have shown that ‘ultrastrong’ spherically shaped materials can be made. Pol et al.⁴⁷ synthesised CSs from polyethylene terephthalate in an autoclave at 700 °C. The 2 mm – 10 mm solid spheres reportedly ‘broke one diamond knife and damaged a second’ during microtome measurements. The tensile strength of a CS was measured under a compressive load and showed linear behaviour with time until the sphere fractured.^36,39,47

Hollow spheres have also been exploited and have been used in rechargeable batteries, in the protection of proteins and enzymes, gene and drug delivery, hydrogen storage, catalysis, sensing, and fuel cell electrodes.⁴⁸

First observed as an ‘unusual form of carbon’ by Davis et al.⁴⁹ in 1953, the study and synthesis of carbon helices (CHs) has grown considerably within the past decade. This is in part because of the development of new synthetic methodologies and the unique properties and use of the helical carbons in versatile applications.^50,51

Synthesis of CHs

Several distinct forms of helical carbon can be produced by employing a specific set of reaction conditions. The different individual structures and morphologies (tubular, filamentous, graphitic or amorphous) are shown in transmission electron microscopy (TEM) images (Figure 15) and reflect the morphological diversity of the family of helical carbon materials: spring-like, single, double or triple helices, twisted belts, spirals, etc. The shape and type of materials can be tailored by adjusting the temperature. For example, using an iron–copper catalyst supported on CaCO₃, single helix multi-walled carbon nanocoils (with tubular structure) can be grown at 750 °C (Figure 15a), while twisted ribbons are formed at 650 °C (Figure 15b). Lowering the temperature to 550 °C produces solid fibres with tight spirals (Figure 15c) and if iron is replaced by nickel, branched spirals (non-tubular) can be produced (Figure 15d). The addition of silver to an iron/nickel catalyst produces tightly twisted ribbons (Figure 15e). The current lack of understanding with regard to the mechanism of carbon nanomaterial formation makes it difficult to propose a specific reason as to why certain morphologies are preferred over others.

The dimensions of these materials are typically reported by measuring the pitch and diameter of the helices or coils (Figure 16).⁵¹ Such dimensional analysis is used to determine the regularity and tightness, as well as the axial and radial diameters, of a helical material. It has also been suggested that a statistical study of these dimensions can be used to indirectly determine whether the formation of helical material correlates to internal or external stresses.⁵⁸

Figure 15: Transmission electron microscopy images of helical carbon material: (a) tubular spring-like (formed using Fe/Cu - 750 °C), (b) twisted ribbon (Fe/Cu - 650 °C), (c) spiralled solid fibres (Fe/Cu - 550 °C), (d) branched spirals (Ni/Cu - 750 °C) and (e) tightly twisted ribbon (Fe/Ni/Ag - 650 °C).

Figure 16: Schematic representation of dimensions used to describe helical morphology: coil pitch and coil diameter.

Growth mechanisms of CHs

The morphology of the helical carbons formed should be related to the shape of the catalyst particle used and this can be evaluated from the dimensions of the regularly shaped metal particles found within or at the ends of the helical products.^52,53,56 In order to investigate the growth mechanism, Qin et al.⁵⁶ carried out the synthesis of helical carbon at low temperature (220 °C) over a copper catalyst; they observed that two helical fibres grew symmetrically from a single catalyst nanoparticle (Figure 18). These helical fibres had opposite helical senses, identical coil pitch and diameter, and the fibre diameters were approximately equal to the size of the nanoparticle.

Qin et al.⁵⁶ concluded that the size of the metal particle played a key role, and that catalyst particles with regular shapes and small sizes (≤ 50 nm) gave helical material. However, Du et al.⁶⁰ prepared copper particles with octahedral geometry, and found no helical material formed, presumably because of the large copper particle sizes (> 100 nm) used. However, helical material several microns in diameter has been synthesised, indicating that metal particle size was not the only factor that determined carbon morphology.^50,51

The catalyst particle shape has also been proposed to control the carbon growth. Motojima⁵⁰ showed that different crystal faces of a metal particle could show variable coil growth. From these results it was proposed that coiling could be related to unequal extrusion rates of carbon from different catalyst faces (Figure 17b). We have also investigated the effect of the shape of copper particles on the carbons formed by analysing TEM images of the copper particles at various tilt angles (Figures 18) and observed that there is a relationship between the particle morphology and the type of helical carbon extruded.

Thus both size and shape of the catalyst particles impact upon the type of helical material extruded. The stability of certain particle morphologies is only possible under a certain set of conditions and this is a likely source of the variability in carbon fibre morphology observed. The change in cross-section shape of a coil, as reported by Cheng et al.⁵³, and observations made by Hanson et al.⁶¹ suggest that particle morphology can also be altered by the gas environment.

Figure 17: Schematic representations of: (a) a coiled nanotube with insertion of pentagon-heptagon rings and (b) a catalyst particle with different facets, showing curvature in extrusion.

Figure 18: (a)Transmission electron microscopy (TEM) image of a spiralled carbon fibre with symmetrical growth from a copper particle and tilted TEM images of a copper particle from different angles: (b) A - 0° and B - 0°, (c) A - 24° and B - 29° and (d) A - 42° and B - 21°.

Properties and uses of CHs

Given these properties, it has been suggested that helical fibres could increase the fracture toughness of polymer-based nanocomposites. The coiled shape could improve the problem of polymer reinforcement that exists with straight carbon nanotubes, thus reducing the possibility of polymer fracture.⁵⁸ They can also be incorporated into nanodevices that require nanosprings as resonating elements. Volodin et al.⁶² reported that, when nanomaterials with helical morphology were excited electrically or acoustically, they were able to detect resonances ranging from 100 MHz to 400 MHz. It has been suggested that such a device could detect mass changes as small as a few attograms.

Given the helical morphology of carbon microcoils (CMCs), Motojima et al.⁶⁴ were able to show that when CMC-polymethylmethacrylate composite beads were prepared and studied, a 1–2 wt.% addition of CMCs resulted in strong electromagnetic wave absorbtivity. This absorbtivity was attributed in part to the chiral character of the CMCs. More recently, Tang et al.⁶⁵ explored the electric properties of carbon nanocoils by developing nanodevices based on these materials. It was found that the conduction mechanism could be described as three-dimensional electron hopping with robust Coulomb interaction, with Coulomb gaps ranging from several mega electrons to 20 Me. Helical materials are also considered a promising candidate for making novel catalyst supports. It is expected that bends or junctions in the carbon may be effective in stabilising deposited metal particles that could influence the selectivity of reactants as a result of possible sterric interferences.

Figure 19: Coiled carbon material at various elongation percentages: (a) 0%, (b) 20% and (c) 33%.

While the sections above have related to the three most popular and common types of SCNMs made and studied in laboratories throughout the world, including South Africa, carbon can also form other morphologies. These morphologies include nanocones, nanohorns, nanofoam and graphene.

Nanocones and nanohorns

Depending on the synthesis conditions (e.g. CO₂ laser ablation of graphite at room temperature,⁶⁶ low-temperature Ar + H₂ + CH₄ plasma-enhanced CVD⁶⁷ or microwave plasma-assisted CVD⁶⁸), carbon nanocones with different opening angles with respect to the apex of the cone can be made. For example, tubular carbon nanocones with single crystal nanotips were synthesised by Shang et al.⁶⁸ by means of microwave plasma-assisted CVD using in situ evaporated iron as a catalyst (Figure 21). The nanocones were stable even at high temperatures, making them potential candidates for scanning probes in high temperature, oxygen-containing environments. Carbon nanohorns also have potential applications in nanocage applications, such as for gas storage and as drug carriers.⁶⁹

Figure 20:Transmission electron microscopy image of single-walled nanohorns.

Figure 21:Scanning electron microscopy image of tubular graphite cones grown on silicon substrates.

Carbon nanofoam

Figure 22: Scanning electron microscopy images of carbon nanofoam produced at 700 °C. Both straight and helical fibres can be clearly seen from the images. Aggregation of the fibres became significant with increasing deposition temperature. Carbon particles with a relatively uniform particle size, produced at 700 °C, are bonded together, forming an interconnected structure.

Graphene and other related carbons

According to Geim and Novoselov⁹, the study of graphene has led to the emergence of a new paradigm of ‘relativistic’ condensed-matter physics, where quantum relativistic phenomena, some of which are unobservable in high-energy physics, can now be mimicked and tested in table-top experiments. This is attributed to the unusual electron spectrum of graphene.

More recently, Li and co-workers⁷⁴ discovered a new form of carbon called graphdiyne. This material was synthesised and predicted to be the most stable of the non-natural carbon allotropes. It is a two-dimensional layer with one-atom thickness and a strongly bonded carbon network that shows remarkable chemical stability and electrical conductivity. In the report, a novel methodology was used to generate large area graphdiyne films (3.61 cm²) on the surface of copper via a cross-coupling reaction of hexaethynylbenzene. A device based on graphdiyne films for measurement of electrical property was fabricated and exhibited semi-conducting properties.⁷⁴

Working at the nano level raises issues regarding human health. Studies to investigate the toxicity of these new SCNMs are in their infancy, with the impact of the materials on human cells being unknown. Studies are further complicated by the belief that effects will be influenced by the size, shape and agglomeration, as well as the concentration of the SCNMs. Care must thus be exercised by workers in this area of nanotechnology.⁷⁵

Although SCNMs have been known for many years, it is still difficult to understand why different shapes of carbon can be formed under certain conditions. In this review, an overview of the types of shaped carbons that can be produced by simple production techniques was outlined with a focus on tubes or fibres, spheres and helices and with more emphasis on the work carried out in South Africa. The synthesis, properties or applications and mechanisms of formation of the different shaped carbons were also described. In addition, we demonstrated that it is possible to manipulate the morphology or carbon nanostructure of SCNMs by functionalisation and doping, which help enhance the solubility and dispersion characteristics and electronic structures of the new carbons. In the last part of the review, we discussed briefly the health implications associated with the use of the nanomaterials.

We thank all the members (including postgraduates, post-doctoral fellows and previous members) of the CATOMAT group who have contributed over the last decade to the development of this new area of chemistry at the University of the Witwatersrand. We also gratefully acknowledge the financial support given by the Department of Science and Technology (South Africa), the DST/NRF Centre of Excellence in Strong Materials and the University of the Witwatersrand.

1. Levi P. The periodic table. New York: Shocken Books, 1984; p. 227.

2. Dresselhaus MS, Dressalhaus G, Eklund PC. Science of fullerenes and carbon nanotubes. New York: Academic Press; 1996.

3. Iijima S. Helical microtubules of graphitic carbon. Nature. 1991;354:56–58. doi:10.1038/354056a0

4. Salvetat J-P, Briggs GAD, Bonard J-M, et al. Elastic and shear moduli of single walled carbon nanotube ropes. Phys Rev Lett. 1999;82(5):944–947. doi:10.1103/PhysRevLett.82.944

5. Pop E, Mann D, Wang Q, Goodson K, Dai H. Thermal conductance of an individual single-wall carbon nanotube above room temperature. Nano Lett. 2005;6(1):96–100. doi:10.1021/nl052145f , PMid:16402794

6. Nkosi BS, Coville NJ, Adams MD, Hutchings GJ. Hydrochlorination of acetylene using carbon supported gold catalysts: A study of catalyst reactivation. J Catal. 1991;128:378–386. doi:10.1016/0021-9517(91)90296-G

7. Adams MD. The elution of gold from activated carbon at room temperature using sulfide solutions. J S Afr Inst Mining Metallurgy. 1994;Aug:187–198.

8. McKune C. Pebble bed modular reactor demonstration plant is funded but not constructed. S Afr J Sci. 2010;106(5/6):1–3. doi:10.4102/sajs.v106i5/6.287

9. Geim AK, Novoselov KS. The rise of graphene. Nature Mater. 2007;6:183–191. doi:10.1038/nmat1849 , PMid:17330084

10. Iijima S, Ichihashi T. Single-shell carbon nanotubes of 1-nm diameter. Nature. 1993;363:603–605. doi:10.1038/363603a0

11. De J, Krijn P, Geus JW. Carbon nanofibres: Catalytic synthesis and applications. Catal Rev. 2000;42(4):481–510. doi:10.1081/CR-100101954

12. Monthioux M, Kuznetsov VL. Who should be given the credit for the discovery of carbon nanotubes? Carbon. 2006;44:1621–1623. doi:10.1016/j. carbon.2006.03.019

13. Mhlanga SD, Mondal KC, Carter R, Witcomb MJ, Coville NJ. The effect of catalyst preparation method on the catalytic synthesis of multiwalled carbon nanotubes using Fe-Co/CaCO₃ catalysts. S Afr J Chem. 2009;62:67–76.

14. Nyamori VO, Mhlanga SD, Coville NJ. The use of organometallic transition metal complexes in the synthesis of shaped carbon nanomaterials. J Organomet Chem. 2008;693:2205–2222. doi:10.1016/j.jorganchem.2008.04.003

15. Nyamori VO, Nxumalo EN, Coville NJ. The effect of arylferrocene ring substituents on the synthesis of multi-walled carbon nanotubes. J Organomet Chem. 2009;694(14):2222–2227. doi:10.1016/j.jorganchem.2009.02.031

16. Mohlala MS, Liu XY, Robinson JM, Coville NJ. Organometallic precursors for use as catalysts in carbon nanotube synthesis. Organometallics. 2005;24:972–976. doi:10.1021/om049242o

17. Balasubramanian K, Burghard M. Chemically functionalised carbon nanotubes. Small. 2005;1(2):180–192. doi:10.1002/smll.200400118 , PMid:17193428

18. Taylor R. Lecture notes on fullerenes. London: Imperial College Press; 1999. doi:10.1142/9781848160675

19. Nxumalo EN, Nyamori VO, Coville NJ. CVD synthesis of nitrogen doped carbon nanotubes using ferrocene/aniline mixtures. J Organomet Chem. 2008;693:2942–2948. doi:10.1016/j.jorganchem.2008.06.015

20. Endo M, Hayashi T, Kim YA, Terrones M, Dresselhaus MS. Applications of carbon nanotubes in the twenty-first century. Phil Trans Roy Soc Lond A. 2004;362:2223–2238. doi:10.1098/rsta.2004.1437 , PMid:15370479

21. Nyamori VO, Coville NJ. Effect of ferrocene/carbon ratio on the size and shape of carbon nanotubes and microspheres. Organometallics. 2007;26:4083–4085. doi:10.1021/om7003628

22. Osváth Z, Koós AA, Horváth ZE, et al. Arc-grown Y-branched carbon nanotubes observed by scanning tunneling microscopy (STM). Chem Phys Lett. 2002;365(3–4):338–342. doi:10.1016/S0009-2614(02)01483-5

23. Durbach SH, Krause RW, Witcomb MJ, Coville NJ. Synthesis of branched carbon nanotubes (BCNTs) using copper catalysts in a hydrogen-filled DC arc discharger. Carbon. 2009;43:635–644. doi:10.1016/j.carbon.2008.10.037

24. Ebbesen TW, editor. Carbon nanotubes: Preparation and properties. Boca Raton: CRC Press; 1997.

25. Tomanek D, Enbody RJ. Science and application of nanotubes. New York: Springer-Verlag; 2000.

26. Bahome MC, Jewell LL, Hildebrandt D, Glasser D, Coville NJ. Fischer-Tropsch synthesis over iron catalysts supported on carbon nanotubes. Appl Catal General A. 2005;287:60–67. doi:10.1016/j.apcata.2005.03.029

27. Bahome MC, Jewell LL, Padayachy K, et al. Fe:Ru small particle bimetallic catalysts supported on carbon nanotubes for use in Fischer-Tropsch synthesis. Appl Catal General A. 2007;328:243–251. doi:10.1016/j.apcata.2007.06.018

28. Nakayama Y, Akita S. Field-emission device with carbon nanotubes for a flat panel display. Synth Met. 2001;117:207–210. doi:10.1016/S0379-6779(00)00365-9

29. Wang QH, Setlur AA, Lauerhaas, JM, Dai JY, Seelig EW, Chang RPH. A nanotube-based field-emission flat panel display. Appl Phys Lett. 1998;72(22):2912–2913. doi:10.1063/1.121493

30. Mordkovich VZ. Carbon nanofibres: A new ultrahigh-strength material for chemical technology. Theor Found Chem Eng. 2003;37(5):429–438. doi:10.1023/A:1026082323244

31. Kroto HW, McKay K. The formation of quasi-icosahedral spiral shell carbon particles. Nature. 1988;331:328–331. doi:10.1038/331328a0

32. Krätschmer W, Fostiropoulos K, Huffman DR. The infrared and ultraviolet absorption spectra of laboratory-produced carbon dust: Evidence for the presence of the CbO molecule. Chem Phys Lett. 1990;170:167–170. doi:10.1016/0009-2614(90)87109-5

33. Guldi DM, Prato M. Excited-state of C₆₀ fullerene derivatives. Acc Chem Res. 2000;33:695–703. doi:10.1021/ar990144m , PMid:11041834

34. Mamo MA, Machado WS, van Otterlo WAL, Coville NJ, Hümmelgen IA. Simple write-once-read-many-times memory device base on carbon spheres-poly(vynilphenol) composite. Organ Electron. In press.

35. Khan SD, Ahmad S. Modelling of C₂ to the addition of C₆₀. Nanotechnology. 2006;17:4654–4658. doi:10.1088/0957-4484/17/18/021

36. Deshmukh AA, Mhlanga SD, Coville NJ. Carbon spheres: A review. Mater Sci Eng R. 2010;70(1–2):1–28. doi:10.1016/j.mser.2010.06.017

37. Fine PM, Cass GR, Simonet BR. Environ Sci Technol. 1999;33:2352–2355. doi:10.1021/es981039v

38. Xia Y, Gates B, Yin Y, Lu Y. Monodispersed colloidal spheres: Old materials with new applications. Adv Mater. 2000;12:693–713. doi:10.1002/(SICI)1521- 4095(200005)12:10<693::AID-ADMA693>3.3.CO;2-A

39. Inagaki M. Discussion of the formation of nanometric texture in spherical carbon bodies. Carbon. 1997;31:711–713. doi:10.1016/S0008-6223(97)86645-6

40. Serp PH, Feurer R, Kalck PH, Kihn Y, Faria JL, Figueiredo JL. A chemical vapour deposition process for the production of carbon nanospheres. Carbon. 2001;39:621–626. doi:10.1016/S0008-6223(00)00324-9

41. Ma Y, Hu Z, Huo K, et al. A practical route to the production of carbon nanocages. Carbon. 2005;43:1667–1672. doi:10.1016/j.carbon.2005.02.004

42. Papirer E, Lacroix R, Donnet J-B. Chemical modification and surface properties of carbon blacks. Carbon. 1996;34:1521–1529. doi:10.1016/S0008-6223(96)00103 0

43. Xiong H, Moyo M, Rayner MK, Jewell LL, Billing DG, Coville NJ. Autoreduction and catalytic performance of a cobalt Fischer–Tropsch synthesis catalyst supported on nitrogen-doped carbon spheres. ChemCatChem. 2010;2:514–518. doi:10.1002/cctc.200900309

44. Mondal KC, Strydom AM, Tetana Z, et al. Boron doped carbon microspheres. Mater Chem Phys. 2009;114:973–977. doi:10.1016/j.matchemphys.2008.11.008

45. Lahaye J, Ehrburger-Dolle F. Mechanisms of carbon black formation. Correlation with the morphology of aggregates. Carbon. 1994;32:1319–1324. doi:10.1016/0008-6223(94)90118-X

46. Wang ZL, Wang ZC. Pairing of pentagonal and heptagonal carbon rings in the growth of nanosize carbon spheres synthesised by a mixed-valent oxide-catalytic carbonization process. J Phys Chem. 1996;100:17725–17731. doi:10.1021/jp962762f

47. Pol SV, Pol VG, Sherman D, Gedanken A. A solvent free process for the generation of strong, conducting carbon spheres by the thermal degradation of waste polyethylene terephthalate. Green Chem. 2009;11:448–451. doi:10.1039/b819494g

48. Donath E, Sukhorukov GB, Caruso F, Davis SA, Mőhwald H. Novel hollow polymer shells by colloid-templated assembly of polyelectrolytes. Angew Chem Int Ed. 1998;37(16):2201–2205. doi:10.1002/(SICI)1521-3773(19980904)37:16<2201::AID-ANIE2201>3.3.CO;2-5

49. Davis WR, Slawson RJ, Rigby GR. An unusual form of carbon. Nature. 1953;171:756. doi:10.1038/171756a0

50. Motojima S. Development of ceramic microcoils with 3D-herical/spiral structures. J Cer Soc Jpn. 2008;116(9):921–927. doi:10.2109/jcersj2.116.921

51. Lau TK, Lu M, Hui D. Coiled carbon nanotubes: Synthesis and their potential applications in advanced composite structures. Compos Part B Eng. 2006;37:437–448. doi:10.1016/j.compositesb.2006.02.008

52. Kawaguchi M, Nozaki K, Motojima S, Iwanaga H. A growth mechanism of regularly coiled carbon fibres through acetylene pyrolysis. J Cryst Growth. 1992;118:309–313. doi:10.1016/0022-0248(92)90077-V

53. Cheng J-B, Du J-H, Bai S. Growth mechanism of carbon microcoils with changing fibre cross-section shape. New Carbon Mater. 2009;24(4):354–358. doi:10.1016/S1872-5805(08)60057-8

54. Motojima S, Kagiya S, Iwanaga H. Vapour-phase formation of micro-coiled carbon fibres using Ni catalyst and PH₃ impurity. Mater Sci Eng. 1995;B34:47–52. doi:10.1016/0921-5107(95)01234-6

55. Qin Y, Zhang Y, Sun X. Synthesis of helical and straight carbon nanofibres by chemical vapor deposition using alkali chloride catalysts. Microchim Acta. 2009;164:425–430. doi:10.1007/s00604-008-0078-2

56. Qin Y, Yu L, Wang Y, Li G, Cui Z. Amorphous helical carbon nanofibres synthesised at low temperature and their elasticity and processability. Solid State Commun. 2006;138:5–8. doi:10.1016/j.ssc.2006.01.048

57. Yang S, Chen X, Katsuno T, Motojima S. Controllable synthesis of carbon microcoils/nanocoils by catalysts supported on ceramics using catalyzed chemical vapor deposition process. Mater Res Bull. 2007;42(3):465–473. doi:10.1016/j.materresbull.2006.06.026

58. Szabó A, Fonseca A, Nagy JB, Lambin PH, Biró LP. Structural origin of coiling in coiled carbon nanotubes. Carbon. 2005;43:1628–1633. doi:10.1016/j.carbon.2005.01.025

59. Hernadi K, Thiên-Nga L, Forró L. Growth and microstructure of catalytically produced coiled carbon nanotubes. J Phys Chem B. 2001;105:12464–12468. doi:10.1021/jp011208p

60. Du F, Liu J, Guo Z. Shape controlled synthesis of Cu₂O and its catalytic application to synthesise amorphous carbon nanofibres. Mater Res Bull. 2009;44:25–29. doi:10.1016/j.materresbull.2008.04.011

61. Hanson PL, Wagner JB, Helveg S, Rostrup-Nielsen JR, Clausen BS, Topsøe H. Atom-resolved imaging of dynamic shape changes in supported copper nanocrystals. Science. 2002;295(5562):2053–2055. doi:10.1126/science.1069325 , PMid:11896271

62. Volodin A, Buntinx D, Aslskog M, Fonseca A, Nagy JB, van Haesendonck C. Coiled carbon nanotubes as self-sensing mechanical resonators. Nano Lett. 2004;4(9):1775–1779. doi:10.1021/nl0491576

63. Chen X, Zhnag S, Dikin DA, et al. Mechanics of a carbon nanocoil. Nano Lett. 2003;3(9):1299–1304. doi:10.1021/nl034367o

64. Motojima S, Hoshiya S, Hishikawa Y. Electromagnetic wave absorption properties of carbon microcoils/PMMA composite beads in W bands. Carbon. 2003;41:2653–2689. doi:10.1016/S0008-6223(03)00292-6

65. Tang N, Kuo W, Jeng C, Wang L, Lin K, Du Y. Coil-in-coil carbon nanocoils: 11 gram-scale synthesis, single nanocoil electrical properties, and electrical contact improvement. ACS Nano. 2010;4(2):781–788. doi:10.1021/nn901417z , PMid:20092354

66. Iijima S, Yudasaka M, Yamada R, et al. Nano-aggregates of single-walled graphitic carbon nano-horns. Chem Phys Lett. 1999;309(3-4):165–170. doi:10.1016/S0009-2614(99)00642-9

67. Tsakadze ZL, Levchenko I, Ostrikov K, Xu S. Plasma-assisted self-organized growth of uniform carbon nanocone arrays. Carbon. 2007;45:2022–2030. doi:10.1016/j.carbon.2007.05.030

68. Shang N, Milne WI, Jiang X. Tubular graphite cones with single-crystal nanotips and their antioxygenic properties. J Am Chem Soc. 2007;129:8907–8911. doi:10.1021/ja071830g , PMid:17589995

69. Ajima K, Yudasaka M, Murakami T, Maigne A, Shiba K, Iijima S. Carbon nanohorns as anticancer drug carriers. Mol Pharm. 2005;2(6):475–480. doi:10.1021/mp0500566 , PMid:16323954

70. Rode AV, Gamaly EG, Luther-Davies B. Formation of cluster-assembled carbon nano-foam by high-repetition-rate laser ablation. Appl Phys A Mater Sci Process. 2000;70(2):35–144. doi:10.1007/s003390050025

71. Zhang Y, Sun X. Synthesis of carbon nanofibres and foam by catalytic chemical vapour deposition using a water-soluble alkali salt catalyst. Adv Mater. 2007;19:961–964. doi:10.1002/adma.200602084

72. Jacoby M. Graphene: Carbon as this as can be. Chem Eng News. 2009;87(9):14–20. doi:10.1021/cen-v087n009.p014

73. Schedin F, Geim AK, Morozov SV, Hill EW, Blake P, Novoselov KS. Detection of individual gas molecules adsorbed on graphene. Nature Mater. 2007;6:652– 655. doi:10.1038/nmat1967 , PMid:17660825

74. Li G, Li Y, Liu H, Guo Y, Li Y, Zhu D. Architecture of graphdiyne nanoscale films. Chem Commun. 2010;46:3256–3258. doi:10.1039/b922733d , PMid:20442882

75. Hester RE, Harrison RM. Nanotechnology: Consequences for human health and the environment. Issues in environmental science and technology. Cambridge: Royal Society of Chemistry; 2007.