10.1098/rsta.2004.

1437

Applications of carbon nanotubes in the twenty-rst century

By M o r i n o b u E n d o1 , Takuaya Hayashi1 , Y o o n g A h m K i m1 , Mauricio Terrones2 a n d M i l r e d S. Dresselhaus3 Faculty of Engineering, Shinshu University, 4-17-1 Wakasato, Nagano-shi 380-8553, Japan (endo@endomoribu.shinshu-u.ac.jp) 2 Advanced Materials Department, IPICyT Camino a la Presa, San Jos 2055, Lomas 4a seccin 78216, e o San Luis Potos SLP, Mexico , 3 Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139-4307, USA
Published online 13 August 2004
1

Carbon nanotubes, which consist of rolled graphene sheets built from sp2 hybridized carbon atoms, are now attracting scientists from various disciplines due to their fascinating physico-chemical properties. In this account, we will review the recent progress on the synthetic techniques for the large-scale production of carbon nanotubes, especially focusing on the oating-catalyst method used in the chemical vapour deposition (CVD) process. We will also describe eective purication methods avoiding structural damage, and discuss the electrochemical applications of these systems including the fabrication of: (i) lithium-ion secondary batteries; (ii) lead-acid batteries; (iii) electric double-layer capacitors; (iv) fuel cells; and (v) multifunctional llers in polymer composites. We foresee that carbon nanotubes will nd numerous applications and take an important place in the development of emerging technologies in the near future.
Keywords: carbon nanotubes; chemical vapour deposition; growth; size-eects

1. Introduction
We will dene nanocarbons as materials built at a nanometre scale from sp2 hybridized carbon atoms. These include various forms ranging from fullerenes (Kroto et al . 1985) and carbon nanotubes (Oberlin et al . 1976; Iijima 1991) to nanoporous materials (Tanaka et al . 1999). Among them, one-dimensional carbon nanotubes have attracted particular interest due to their unique morphology, nanosized scale, novel physico-chemical properties and, furthermore, their versatile applications (Dresselhaus et al . 1996; Saito et al . 1998; Harris 1999; Ajayan & Zhou 2001; Baughman et al . 2002; Yasuda et al . 2003). Carbon nanotubes could be visualized as rolled sheets of graphene (sp2 carbon arranged in a honeycomb lattice), which are sometimes capped (by a fullerene-like hemisphere) at each end (see gure 1). They could be either single walled with diameters as small as 0.4 nm, or multi-walled consisting of
One contribution of 12 to a Theme Nanotechnology of carbon and related materials. Phil. Trans. R. Soc. Lond. A (2004) 362, 22232238 c 2004 The Royal Society

2223

2224

M. Endo and others

Figure 1. Schematic of an individual layer of honeycomb-like carbon called graphene, and how this could be rolled in order to form a carbon nanotube.

1.5

band gap (eV)

1.0

0.5

0.5 1.0 nanotube radius (nm)

1.5

Figure 2. Band gap as a function of nanotube radius calculated using an empirical tight-binding Hamiltonian.

nested tubes (e.g. 230 concentric tubes positioned one inside the other) with outer diameters ranging from 5 to 100 nm. The constituent cylinders within multi-walled carbon nanotubes (MWNTs) may possess dierent chiral structures. In particular, it has been predicted that the electronic properties of single-walled carbon nanotubes (SWNTs) may vary from semiconductors to metals, depending upon the chiral angle (the way the hexagons are positioned with respect to the tube axis) (Dresselhaus et al . 1996; Saito et al . 1998). Figure 2 shows the band gap, estimated by a tight-binding approach, as a function of the nanotube radius. The plot indicates that as the nanotubes diameter becomes larger (the curvature of a graphene cylinder becomes less
Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

(a) (b) (c)

2225

200 nm

5 nm

5 nm

Figure 3. (a) SEM image exhibiting web-like SWNT structures on the surface of zeolite template. (b), (c) TEM images of CVD derived SWNTs in an isolated and a bundle state, respectively.

pronounced), the value of the band gap is closer to that of a planar graphene sheet (Mintmire & White 1995). In particular, three typical chiral structures of SWNTs have been proposed (armchair, zigzag and helical), and their structures have been conrmed using scanning tunnelling microscopy (STM) and electron diraction studies (Ge & Sattler 1993; Dresselhaus et al . 1996; Wilder et al . 1998). For example, if the C60 molecule is bisected normally to the vefold axis, the cap for an armchair nanotube is formed, and if the C60 molecule is bisected normally to the threefold axis, the cap for a zigzag nanotube is generated. In addition to the armchair and zigzag nanotubes, a large number of chiral or helical carbon nanotubes could be obtained, exhibiting a screw-like hexagonal pattern along the nanotube axis, which is terminated by a large variety of hemispherical but asymmetric caps. It is important to emphasize that three-dimensional stacking does not exist in MWNTs, even though each individual shell consists of a perfect rolled hexagonal graphene sheet. In MWNTs, the curvature of the graphene sheets is responsible for weaker van der Waals interactions, thus resulting in a larger inter-layer spacing (0.34 nm) when compared with three-dimensional crystalline graphite (0.335 nm). In this paper, we will review large-scale synthetic methods for producing carbon nanotubes, their purication technique, and nally we will discuss some applications of these nanomaterials in electrochemical systems and as multifunctional llers in the fabrication of polymer composites.

2. Large-scale production of carbon nanotubes

In order to use carbon nanotubes in novel devices, it is necessary to produce these materials with a high crystallinity on a large scale economically. In this context, the catalytic chemical vapour deposition (CVD) method is considered to be the optimum for producing large amounts of carbon nanotubes, particularly with the use of a oating-catalyst method (Endo 1988; Tibbetts et al . 1994; Lake 2001). This technique is more controllable and cost ecient when compared with arc-discharge and laser ablation methods (Ebbesen & Ajayan 1992; Thess et al . 1996; Journet et al . 1997; Liu et al . 1998). Various groups have been able to produce SWNTs for laboratory experiments (Cheng et al . 1999; Colomer et al . 2000). Recently, experimental groups have decided to use SWNTs produced using the high-pressure carbon monoxide (HIPCO) process developed by Smalley and his Rice University colleaguesthis
Phil. Trans. R. Soc. Lond. A (2004)

2226
unit: nanometre

M. Endo and others

intensity (arb. units)

1.222 1.389 arc-discharge 1.223 0.989 0.845

1.667 1.389

CVD

100

200 Raman frequency (cm1)

300 1200

1400 Raman frequency

1600 (cm1)

1800

Figure 4. Raman spectra of arc-discharge and CVD derived SWNTs. The values inserted in this gure are calculated through the correlation between diameter and RBM Raman frequency (Bandow et al . 1998).

Figure 5. A graphical description of the tip-growth and root-growth nanotube models, in which the catalytic particles are located at the tip and at the root, respectively. Carbon species are illustrated as benzene molecules, and the base is the simplied model of the zeolite (Hayashi et al . 2003).

Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

(a) (b)

2227

3 nm

2 nm

Figure 6. TEM images of the tubes with the catalytic particles (dark contrast) at (a) the root of the tubes, and (b) the tip of the tube. This demonstrates that the tube growth in the zeolite oating-substrate method occurs both at the root and at the tip of the tube (Hayashi et al . 2003).

have enabled the production of large amounts of high-purity SWNTs (Nikolaev et al . 1999). More recently, we reported an alternative route for the large-scale synthesis of SWNTs by combining the use of catalytic substrates and oating methods (Hayashi et al . 2003). The template (substrate) prevents metal particle aggregation, thus resulting in the formation of high-purity SWNTs. The method is able to produce either individual nanotubes or nanotube bundles, exhibiting a wide range of diameters (0.44 nm). In particular, nanosized zeolites were impregnated with Fecontaining compounds (seeding method) and placed inside a furnace (ca. 1000 C) together with benzene vapour, and H2 as the carrier gas. The resulting material was puried via immersion in hydrouoric acid (HF). Scanning electron microscope (SEM) images (gure 3a) show web-like structures attached to the surface of the zeolite template. Transmission electron microscopy (TEM) studies conrm the presence of isolated SWNTs and bundles of nanotubes of various diameters (gure 3b). We could attribute the diameter variations of the tubes to the size of the catalytic particles embedded in the zeolite. For the case of bundle formation, the diameter distribution of SWNTs within a bundle is small (gure 3c), which may be due to the presence of homogeneous surfaces containing catalyst particles of very similar size; growth conditions, such as carbon density, reaction temperature and ow rate, should also be similar. Raman spectroscopy is a useful technique for characterizing SWNTs because it reveals information related to tube diameter and the degree of structural perfection of the graphene sheets (Dresselhaus & Eklund 2000; Saito et al . 2000). Figure 4 displays Raman spectra of arc-discharge and CVD-produced SWNTs. The radial breathing modes (RBMs) are observed between 100 and 300 cm1 , whereas the G-bands, with a zone-folding spread, are observed at ca. 1590 cm1 for both spectra. From the spectra it is clear that CVD SWNTs exhibit a wide diameter distribution (0.845
Phil. Trans. R. Soc. Lond. A (2004)

2228

M. Endo and others

1.667 nm) compared with those produced by the arc-discharge method. In addition, the intensity of the D-band (ca. 1350 cm1 ), in the case of CVD-derived SWNTs, is higher than that of arc-discharge SWNTs (see Dresselhaus & Eklund (2000) and Saito et al . (2000) for details on Raman spectroscopy). This result indicates a larger defect concentration, including point defects due to a lower tube-formation temperature. It is important to emphasize that each Raman spectrum exhibits dierent RBMs, thus conrming the diversity of SWNTs produced, due to the dierent growth conditions of the metal particles embedded in the zeolite template. There are two major mechanisms for the catalytic growth of carbon nanotubes depending on the position of the catalyst with respect to the substrate. Figure 5 shows both the tip-growth mechanism, whereby the catalyst particle is always located at the tip of a growing tube, and the root-growth mechanism, in which the catalytic particle is placed at the root (bottom) of the tube (Oberlin et al . 1976; Tibbetts 1984; Tibbetts et al . 1987; Baker 1989). For the SWNTs, catalytic particles can be found at both: the roots and the tips of the tubes (gure 6). This indicates that SWNTs are able to grow within these two frameworks. Moreover, we believe that both growth mechanisms occur simultaneously and depend on the position of the catalytic particle, which acts as a carbon extruder. Regarding the bulk production of nanotubes (concentric graphene tubules) and nanobres (irregular carbon nanolaments) for industrial applications, it is important to mention that in the early 1990s, Hyperion Catalysis International, Inc. (Cambridge, MA) started the large-scale production of MWNTs. This company has a wide range of patents on synthesis and applications of nanotubes and nanobres of carbon, which expire in 2004. The Carbon Nanotech Research Institute (CNRI), a subsidiary of Mitsui & Co., Ltd., plans to engage in developing technologies for the large-scale production of 120 t of MWNTs annually. Others companies such as Applied Sciences, Inc. (API) and Showa Denko (SDK) already have a large-scale production capacity for MWNTs, which exhibit a wide distribution of diameters in the range from 70 to 200 nm. The most interesting point is that all companies selected a catalytic CVD method, and furthermore three companies (not Hyperion) adopted the oating-reactant technique for the large-scale production of MWNTs. The development of the oating-reactant/catalyst techniques allows a three-dimensional dispersion of the hydrocarbon together with the catalytic particles derived from the pyrolysis of organometallic compounds. This results in a semi-continuous production of MWNTs (Endo 1988; Tibbetts et al . 1994; Lake 2001; Collins et al . 2002). Therefore, it is clear that a great deal of progress has been achieved related to the bulk synthesis of carbon nanotubes. However, at present the most important challenge is to tailor the chirality of SWNTs and MWNTs.

3. Purication of carbon nanotubes

Even though synthetic techniques have been improved to obtain high-purity carbon nanotubes, the formation of byproducts containing impurities such as metal-encapsulated nanoparticles, metal particles in the tip of a carbon nanotube, and amorphous carbon has been an unavoidable phenomenon, because the metal nanoparticles are essential for the nanotube growth. Thus, extensive research has been dedicated to the purication of carbon nanotubes in order to remove foreign nanoparticles that modify the physico-chemical properties of carbon nanotubes.
Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

(a) (b) (c)

2229

0.34 nm

10 nm

10 nm

Figure 7. (a) HR-TEM image of highly disordered carbon nanotubes. (b) HR-TEM image of an annealed carbon nanotube at 2800 C showing linear, sti graphene layers along the tube axis. (c) Enlarged HR-TEM image of (b). Note that the distance between adjacent graphene layers is 0.34 nm.

Chemical methods have been applied for purifying SWNTs. A generalized method for SWNT purication developed by Smalley and co-workers (Liu et al . 1998) consists of reuxing as-grown SWNTs in nitric acid solutions. Subsequently, more-eective purication techniques have been developed with minor physical damage of the tubes (Bandow et al . 1997; Rinzler et al . 1998; Dillon et al . 1999). Chattopadhyay et al . (2002) reported the complete removal of metallic impurities through a sonicationmediated treatment of as-grown SWNTs in a mixture of hydrouoric and nitric acids. Also, Martinez et al . (2002) reported a combined technique of high-temperature air oxidation in conjunction with microwave acid treatments, thus removing a high portion of metal particles in relatively short periods of time. For MWNTs, high-temperature treatments in an inert atmosphere (graphitization or annealing) are the most eective methods for removing structural defects (heptagons and heptagonpentagon pairs) or impurities such as metallic compounds (Andrews et al . 2001; Grimes et al . 2001; Endo et al . 2001a). In this context, gure 7 shows the structural transformation of a disordered MWNT into a highly ordered multi-wall carbon nanotube. This crystallization is a stepwise process: (i) straightening and rearrangement of distorted graphene layers, (ii) fusion between graphene layers, and (iii) growth to larger graphene layers along the tube axis and removal of stacking faults between graphene layers (Kim et al . 2003).

4. Electrochemical system
(a) Lithium-ion battery The outstanding mechanical properties and the high surface-to-volume ratio (due to their small diameter) make carbon nanotubes potentially useful as anode materials (Endo et al . 1995; Tatsumi et al . 1999; Frackowiak et al . 1999; Wu et al . 1999; Gao et al . 1999; Frackowiak & Beguin 2002) or as additives (Endo et al . 2001b) in lithium-ion battery systems. An SEM image of an electrode containing 10 wt % of
Phil. Trans. R. Soc. Lond. A (2004)

2230
(a)

M. Endo and others

(b) 100

cyclic efficiency (%)

95

90

85 added weight percents of nanotube 0 1 5 10 80 0 10 20 30 40 50 cycle number 60 70

3 m

Figure 8. (a) SEM micrographs of the anode sheet containing carbon nanotubes in a commercial lithium-ion battery. (b) The cyclic eciency of synthetic graphite, heat treated at 2900 C, as a function of weight percent, in the range 01.5 V, with a current density of 0.2 mA cm2 .

carbon nanotubes as the additive displays a homogeneous distribution of nanotubes in synthetic graphite (gure 8a). Figure 8b reveals the cyclic eciency of a synthetic graphite anode as a function of the weight percent of carbon nanotube. With increasing weight percent of carbon nanotubes, the cyclic eciency of the synthetic graphite battery anode increases continuously, and, in particular, when 10 wt % of the nanotubes was added, the cyclic eciency was maintained at almost 100% up to 50 cycles. At higher concentrations, the nanotubes interconnect graphite powder particles together to form a continuous conductive network. The characteristics of a carbon nanotube when used as a ller in the electrodes of lithium-ion batteries can be summarized as follows. (i) The small diameter of the nanotube makes it possible to distribute the nanotubes homogeneously in the thin electrode material and to introduce a larger surface area to react with the electrolyte. (ii) The improved electrical conductivity of the electrode is related to the high electrical conductivity of the tubes, and the function of the electrical bridge between graphite particles. (iii) The relatively high intercalation ability of nanotubes did not in itself lower the capacity of anode materials upon cycling. (iv) A high exibility of the electrode is also achieved due to the network formation of the nanotube in a tube-mat structure. (v) The electrode has high endurance due to the presence of nanotubes, which absorb the stress caused by intercalation of lithium ions. (vi) Improved penetration of the electrolyte due to the homogeneous distribution of the tubes surrounding the anode material.
Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

100 discharge capacity (%) addition of CNTs to negative plate 80

2231

60 no addition 40 0 200 400 600 cycle number 800 1000

Figure 9. The eects of carbon nanotubes embedded in a conductive ller in polyvinylidenederived carbon electrode and compared with a carbon black/polyvinylidene composite.

(vii) The cyclic eciency of the lithium-ion battery was improved for a relatively long cycle when compared with that of carbon black. (b) Additives to the electrodes of lead-acid batteries In order to increase the conductivity of electrodes in lead-acid batteries, dierent weight percents of carbon nanotubes are added to the active anode material (with average diameters of ca. 25 mm) of the positive electrode. The resistivity of the electrode is lowered for the case of 1.5% nanotube addition. When this sample (0.51 wt %) is incorporated in the negative electrode, the cycle characteristics are greatly improved compared with those of an electrode without additive (gure 9) (Endo et al . 2001b). This is probably due to the ability of carbon nanotubes to act as a physical binder, resulting in electrodes that undergo less mechanical disintegration and shedding of their active material. Therefore, it is expected that the use of carbon nanotubes as an electrodes ller should produce an enhanced cyclic behaviour for electrodes in lead-acid batteries compared with electrodes using conventional graphite powder, because the unusual morphology of the carbon nanotube, such as the concentric orientation of their graphite crystallites along the bre cross-section, induces a high resistance towards oxidation, and furthermore the nanotube network embedded in the polymer would enhance the reactivity of the electrode. (c) The electric double-layer capacitor The merit of the electric double-layer capacitor (EDLC) is considered to be a high discharge rate (Conway 1999), which makes them applicable as a hybrid energy source for electric vehicles and portable electric devices (Miyadera 2002). EDLCs containing carbon nanotubes in the electrode exhibit relatively high capacitances resulting from the high surface area accessible to the electrolyte (Niu et al . 1997; An et al . 2001; Frackowiak & Beguin 2002). On the other hand, the most important factor in commercial EDLCs is considered to be the overall resistance. In this context, carbon nanotubes with strong electrical and mechanical properties can be used as an electrical conductive additive in the electrode of EDLC. It has been demonstrated
Phil. Trans. R. Soc. Lond. A (2004)

2232

M. Endo and others

10 nm
Figure 10. An HR-TEM image of highly dispersed Pt nanoparticles deposited on both outer and inner surfaces of shortened stacked-cone nanotubes/nanotubes.

that the addition of carbon nanotubes results in an enhanced capacity at higher current densities, when compared with electrodes containing carbon black (Takeda et al . 2001). (d ) Fuel cells Fuel cells have been considered as the next generation of energy devices because this type of system transforms the chemical reaction energy from hydrogen and oxygen into electric energy (Williams 2001). Carbon nanotubes decorated with metal nanoparticles as the electrode have doubled the performance of the fuel cell, owing to the increased catalytic activity of nanotube-based electrodes (Britto et al . 1999; Che et al . 1999; Yoshitake et al . 2002). In this context, we have reported the ecient impregnation of Pt nanoparticles (outer diameter less than 3 nm) on cup-stackedtype carbon nanotubes (Endo et al . 2003). The method involves dispersion of the bres in H2 PtCl6 , followed by low-temperature annealing. The Pt particle deposition is always homogeneous, and can be controlled selectively on the outer or inner core using the hydrophobic nature of the material (see gure 10). Since the Pt particle activity on the bres is high, this material could nd application as an ecient catalyst or biological device. It may be that carbon nanotube technology will contribute to the development of fuel cells as a catalyst support and also as a main component of bipolar systems. However, additional basic and applied research is necessary.

5. Multifunctional llers in polymer composite

It has been shown that carbon nanotubes could behave as the ultimate one-dimensional material with remarkable mechanical properties (Lu 1997; Li et al . 2000). The density-based modulus and strength of highly crystalline SWNTs are 19 and 56 times that of steel. Based on a continuum shell model, the armchair tube exhibits larger stressstrain response than the zigzag tube under tensile loading (gure 11).
Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

150 D = 1.37 nm tensile stress (GPa) 100 armchair (10,10)

2233

zigzag (17,0)

50

E = 0.94 TPa 0 5 10 strain, (%) 15

CNT b = 1163 GPa b ~ 12% D = 1.35 nm 20

Figure 11. Stressstrain response of a carbon nanotube under tensile loading.

nanogear

human hair

0.2 mm

0.1 mm 2 m

Figure 12. SEM image of the smallest working gear (carbon nanotube/nylon composite); the inset shows the fractured surface.

Strong mechanical properties of carbon nanotubes due to a strong carboncarbon covalent bond are highly dependent upon the atomic structure of nanotubes and the number of shells. Moreover, carbon nanotubes exhibit strong electrical and thermal conducting properties. Therefore, carbon nanotubes (single and multi-walled) have been studied intensively as llers in various matrices, especially polymers (Ajayan et al . 1994; Lau & Hui 2002; Calvert 1999; Mamedov et al . 2002; Zhan et al . 2003).
Phil. Trans. R. Soc. Lond. A (2004)

2234

M. Endo and others

The best use of the intrinsic properties of these brous nanocarbons in polymers can be achieved by optimizing the interface interaction of the nanotube surface and the polymer. Therefore, surface treatments via oxidation could be used in order to improve adhesion properties between the ller and the matrix. This results in a good stress transfer from the polymer to the nanotube. There are various surface oxidative processes, such as electrochemical, chemical and plasma techniques. From the industrial point of view, ozone treatment is a very attractive technique. In addition, the dispersion of the nanotubes/nanobres in the polymer should be uniform within the matrix. The smallest working composite gear has been prepared by mixing nanotubes into molten nylon and then injecting into the tiny mould. As shown in gure 12, this gear is as small as the diameter of a human hair. This piece exhibits a high mechanical strength, high abrasion resistance and also good electrical and thermal conductivity. When cup-stacked-type carbon nanotubes are incorporated in polypropylene, the improvement of the tensile strength is very noticeable (up to 40%). This remarkable result can be explained by the particular morphology of cup-stacked-type carbon nanotubes. In other words, a large portion of edge sites on the outer surface of the nanotubes might act as nucleation sites, resulting in good adhesion between nanotubes and polymers (good stress transfer). Recently, various studies on the nucleation eect of nanotubes on the crystallization of semi-crystalline polymers have been reported (Valentini et al . 2003; Bhattacharyya et al . 2003; Cadek et al . 2003).

6. Conclusion
This account has mainly described the possible routes to large-scale synthesis of carbon nanotubes (single and multi-walled), with emphasis on their applications in electrochemical systems and polymer nanocomposites. The unique electronic properties of nanotubes also make them good candidates in the electronics industry for replacing silicon-based technologies with those based on nanocarbons (Collins & Avouris 2002). At present, researchers are working on the following areas in order to apply carbon nanotubes in emerging technologies. (i) Production of defect-free and high-purity carbon nanotubes. (ii) Establishment of useful techniques for quantifying the number of defects (their types, where they are, etc.) in the nanotube structure. (iii) Development of eective purication techniques (below the parts per million level) for the metal particles within carbon nanotubes, especially for biological and electronic applications. (iv) Achieving homogeneous carbon nanotube dispersions in polymer composites. The possible applications of carbon nanotubes range from electronics to eldemission displays, energy-storage devices and functional llers in composites. These have attracted both industrial and academic interest. Therefore, it is important that some of the basic knowledge is transferred to industry very shortly so that real and novel technologies appear commercially.
We are thankful for nancial support from the CLUSTER of Ministry of Education, Culture, Sports, Science and Technology. We also thank CONACYT-Mxico, grants W-8001-Millennium e Initiative (H.T., M.T.), 37589-U (M.T.) and 41464-Inter American Collaboration (M.T.).
Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

2235

References
Ajayan, P. M. & Zhou, O. Z. 2001 Application of carbon nanotubes. In Carbon nanotubes. Topics in applied physics (ed. M. S. Dresselhaus, G. Dresselhaus & Ph. Avouris), vol. 80, pp. 381425. Springer. Ajayan, P. M., Stephan, O., Colliex, C. & Trauth, D. 1994 Aligned carbon nanotube arrays formed by cutting a polymer resinnanotube composite. Science 265, 12121214. An, K. H., Kim, W. S., Park, Y. S., Moon, J. M., Lim, S. C., Lee, Y. S. & Lee, Y. H. 2001 Electrochemical properties of high power supercapacitors using singlewalled carbon nanotubes electrodes. Adv. Mater. 11, 387392. Andrews, R., Jacques, D., Qian, D. & Dickey, E. C. 2001 Purication and structural annealing of multiwalled carbon nanotubes at graphitization temperatures. Carbon 39, 16811687. Baker, R. T. K. 1989 Catalytic growth of carbon laments. Carbon 27, 315323. Bandow, S., Rao, A. M., Thess, A. & Smalley, R. E. 1997 Purication of single wall carbon nanotubes by microltration. J. Phys. Chem. B 101, 88398842. Bandow, S., Asaka, S., Saito, Y., Rao, A. M., Grigorian, L., Richter, E. & Eklund, P. C. 1998 Eect of the growth temperature on the diameter distribution and chirality of single-wall carbon nanotubes. Phys. Rev. Lett. 80, 37793782. Baughman, R. H., Zakhidov, A. A. & De Heer, W. A. 2002 Carbon nanotubesthe route toward applications. Science 297, 787792. Bhattacharyya, A. R., Sreekumar, T. V., Liu, T., Kumar, S., Ericson, L. M., Hauge, R. H. & Smalley, R. E. 2003 Crystallization and orientation studies in polypropylene/single wall carbon nanotube composite. Polymer 44, 23732377. Britto, P. J., Santhanam, K. S. V., Rubio, A., Alonso, A. & Ajayan, P. M. 1999 Improved charge transfer at carbon nanotube electrodes. Adv. Mater. 11, 154157. Cadek, M., Coleman, J. N., Barron, V., Hedicke, K. & Blau, W. J. 2003 Morphological and mechanical properties of carbon-nanotube-reinforced semicrystalline and amorphous polymer composites. Appl. Phys. Lett. 81, 51235125. Calvert, P. 1999 Nanotube composites: a recipe for strength. Nature 399, 210211. Chattopadhyay, D., Galeska, I. & Papadimitrakopoulos, F. 2002 Complete elimination of metal catalysts from single wall carbon nanotubes. Carbon 40, 985988. Che, G., Lakshmi, B. B., Fisher, E. R. & Martin, C. R. 1999 Carbon nanotubule membranes for electrochemical energy storage and production. Nature 393, 346349. Cheng, H., Li, F., Su, G., Pan, H. & Dresselhaus, M. S. 1999 Large-scale and low-cost synthesis of single-walled carbon nanotubes by the catalytic pyrolysis of hydrocarbons. Appl. Phys. Lett. 72, 32823284. Collins, P. G. & Avouris, P. 2002 Nanotubes for electronics. Scientic American, December, pp. 6269. Collins, S., Brydson, R. & Rand, B. 2002 Structural analysis of carbon nanobres grown by the oating catalyst method. Carbon 40, 10891100. Colomer, J. F., Stephan, C., Lefrant, S., Van Tendeloo, G., Willems, I., Konya, Z., Fonseca, A., Laurent, Ch. & Nagy, J. B. 2000 Large-scale synthesis of single-wall carbon nanotubes by catalytic chemical vapor deposition (CCVD) method. Chem. Phys. Lett. 317, 8389. Conway, B. E. 1999 Electrochemical supercapacitorsscientic fundamentals and technological applications. Dordrecht: Kluwer Academic. Dillon, A. C., Gennett, T., Jones, K. M., Alleman, J. L., Parilla, P. A. & Heben, M. J. 1999 A simple and complete purication of single-walled carbon nanotube materials. Adv. Mater. 11, 13541358. Dresselhaus, M. S. & Eklund, P. C. 2000 Phonons in carbon nanotubes. Adv. Phys. 49, 705814. Dresselhaus, M. S., Dresselhaus, G. & Eklund, P. C. 1996 Science and fullerenes and carbon nanotubes, 1st edn, pp. 1505. Academic.
Phil. Trans. R. Soc. Lond. A (2004)

2236

M. Endo and others

Ebbesen, T. W. & Ajayan, P. M. 1992 Large-scale synthesis of carbon nanotubes. Nature 358, 220222. Endo, M. 1988 Grow carbon bers in the vapor phase. CHEMTECH 18, 568576. Endo, M., Nakamura, J., Sasabe, Y., Takahashi, T. & Inagaki, M. 1995 Lithium secondary battery using vapor grown carbon bers as a negative electrode and analysis of the electrode mechanism by TEM observation. Trans. IEE Jpn A 115, 349356. Endo, M., Kim, Y. A., Fukai, Y., Hayashi, T., Terrones, M., Terrones, H. & Dresselhaus, M. S. 2001a Comparison study of semi-crystalline and highly crystalline multiwalled carbon nanotubes. Appl. Phys. Lett. 79, 15311533. Endo, M., Kim, Y. A., Hayashi, T., Nishimura, K., Matushita, T., Miyashita, K. & Dresselhaus, M. S. 2001b Vapor-grown carbon bers (VGCFs): basic properties and their battery applications. Carbon 39, 12871297. Endo, M., Kim, Y. A., Ezaka, M., Osada, K., Yanagisawa, T., Hayashi, T., Terrones, M. & Dresselhaus, M. S. 2003 Selective and ecient impregnation of metal nanoparticles on cupstacked-type nanobers. Nano Lett. 3, 723726. Frackowiak, E. & Beguin, F. 2002 Electrochemical storage of energy in carbon nanotubes and nanostructured carbons. Carbon 40, 17751787. Frackowiak, E., Gautier, S., Gaucher, H., Bonnamy, S. & Beguin, F. 1999 Electrochemical storage of lithium in multiwalled carbon nanotubes. Carbon 37, 6169. Gao, B., Kelinhammes, A., Tang, X. P., Bower, C., Wu, Y. & Zhou, O. 1999 Electrochemical intercalation of single-walled carbon nanotubes with lithium. Chem. Phys. Lett. 307, 153157. Ge, M. & Sattler, K. 1993 Vapor-condensation generation and STM analysis of fullerene tubes. Science 260, 515518. Grimes, C. A., Dickey, E. C., Mungle, C., Ong, K. G. & Qian, D. 2001 Eect of purication of the electrical conductivity and complex permittivity of multiwall carbon nanotubes. J. Appl. Phys. 90, 41344137. Harris, P. J. F. 1999 Carbon nanotubes and related structures: new materials for the twenty-rst century, 1st edn, pp. 1293. Cambridge University Press. Hayashi, T., Kim, Y. A., Matoba, T., Ezaka, M., Nishimura, K., Tsukada, T., Endo, M. & Dresselhaus, M. S. 2003 The smallest freestanding single-wall carbon nanotube. Nano Lett. 3, 887889. Iijima, S. 1991 Helical microtubules of graphic carbon. Nature 354, 5658. Journet, C., Maser, W. K., Bernier, P., Loiseau, A., Delachapelle, M. L., Lefrant, S., Deniard, P., Lee, R. & Fisher, J. E. 1997 Large-scale production of single-walled carbon nanotubes by the electric-arc technique. Nature 388, 756758. Kim, Y. A., Osada, K., Hayashi, T., Endo, M. & Dresselhaus, M. S. 2003 Annealing eect on disordered multi-wall carbon nanotubes. Chem. Phys. Lett. 380, 319324. Kroto, H. W., Heath, J. R., OBrien, S. C., Curl, R. F. & Smalley, R. E. 1985 C60: Buckminsterfullerene. Nature 318, 162163. Lake, M. L. 2001 Large scale production of VGCF. In Carbon laments and nanotubes: common origins, diering applications? Applied Science, Series E, vol. 372, pp. 187196. Dordrecht: Kluwer Academic. Lau, K. T. & Hui, D. 2002 The revolutionary creation of new advanced materialscarbon nanotube composites. Composites B 33, 263277. Li, F., Cheng, B. S., Su, G. & Dresselhaus, M. S. 2000 Tensile strength of single-walled carbon nanotubes directly measured from their macroscopic ropes. Appl. Phys. Lett. 77, 31613163. Liu, J. (and 14 others) 1998 Fullerene pipes. Science 280, 12531256. Lu, J. P. 1997 Elastic properties of single and multilayered nanotubes. J. Phys. Chem. Solids 58, 16491652.
Phil. Trans. R. Soc. Lond. A (2004)

Applications of carbon nanotubes in the twenty-rst century

2237

Mamedov, A. A., Kotov, N. A., Prato, M., Guldi, D. M., Wicksted, J. P. & Hirsch, A. 2002 Molecular design of strong single-wall carbon nanotube/polyelectrolyte multilayer composites. Nature Mater. 1, 190194. Martinez, M. T., Callejas, M. A., Benito, A. M., Maser, W. K., Cochet, M., Andres, J. M., Schreiber, J., Chauvet, O. & Fierro, J. L. 2002 Microwave single walled carbon nanotubes purication. Chem. Commun., pp. 10001001. Mintmire, J. W. & White, C. T. 1995 Electronic and structural properties of carbon nanotubes. Carbon 33, 893902. Miyadera, K. 2002 Development of ES3 small sized low-fuel consumption vehicle. Toyota Tech. Rev. 52, 2227. Nikolaev, P., Bronikowski, M. J., Bradley, R. K., Rohmund, F., Colbert, D. T., Smith, K. A. & Smalley, R. E. 1999 Gas-phase catalytic growth of single-walled carbon nanotubes from carbon monoxide. Chem. Phys. Lett. 313, 9197. Niu, C., Sickel, E. K., Hoch, R., Moy, D. & Tennent, H. 1997 High power electrochemical capacitors based on carbon nanotube electrodes. Appl. Phys. Lett. 70, 14801482. Oberlin, A., Endo, M. & Koyama, T. 1976 Filamentous growth of carbon through benzene decomposition. J. Cryst. Growth 32, 335349. Rinzler, A. G., Liu, J., Dai, H., Nikolaev, P., Human, C. B. & Rodriguez-Marcias, F. J. 1998 Large-scale purication of single-wall carbon nanotubes: process, product, and characterization. Appl. Phys. A 67, 2937. Saito, R., Dresselhaus, G. & Dresselhaus, M. S. 1998 Physical properties of carbon nanotubes, 1st edn, pp. 1259. London: Imperial College Press. Saito, R., Dresselhaus, G. & Dresselhaus, M. S. 2000 Trigonal warping eect of carbon nanotubes. Phys. Rev. B 61, 29812990. Takeda, T., Takahashi, R., Kim, Y. J., Koshiba, K., Ishi, K., Kasahi, T. & Endo, M. 2001 Eect of VGCF addition into the carbon electrode for electric double layer capacitor (EDLC). TANSO 196, 1418. Tanaka, K., Yamabe, T. & Fukui, K. 1999 The science and technology of carbon nanotubes, 1st edn, pp. 1190. Elsevier. Tatsumi, K., Zaghib, K., Abe, H., Higuchi, S., Ohsaki, T. & Sawada, Y. 1999 A modication in the preparation process of a carbon whisker for the anode performance of lithium rechargeable batteries. J. Power Sources 54, 425427. Thess, A. (and 14 others) 1996 Crystalline ropes of metallic carbon nanotubes. Science 273, 483487. Tibbetts, G. G. 1984 Why are carbon laments tubular? J. Cryst. Growth 66, 632638. Tibbetts, G. G., Devour, M. G. & Rodda, E. J. 1987 An adsorptiondiusion isotherm and its application to the growth of carbon laments on iron catalyst particles. Carbon 25, 367375. Tibbetts, G. G., Bernnardo, C. A., Gorkiewicz, D. W. & Alig, R. L. 1994 Role of sulfur in the production of carbon bers in the vapor phase. Carbon 32, 569576. Valentini, L., Biagiotti, J., Kenny, J. M. & Santucci, S. 2003 Eects of single-walled carbon nanotubes on the crystallization behavior of polypropylene. J. Appl. Polym. Sci. 87, 708 713. Wilder, J. W. G., Venema, L. C., Rinzler, A. G., Smalley, R. E. & Dekker, C. 1998 Electronic structure of atomically resolved carbon nanotubes. Nature 391, 5962. Williams, M. C. 2001 Status and promise of fuel cell technology. Fuel Cells 2, 8791. Wu, G. T., Wang, C. S., Zhang, X. B., Yang, H. S., Qi, Z. F., He, P. M. & Li, W. Z. 1999 Structure and lithium insertion properties of carbon nanotubes. J. Electrochem. Soc. 146, 16961701. Yasuda, E., Inagaki, M., Kaneko, K., Endo, M., Oya, A. & Tanabe, Y. 2003 Carbon alloys: novel concept to develop carbon science and technology, 1st edn, pp. 1569. Elsevier.
Phil. Trans. R. Soc. Lond. A (2004)

2238

M. Endo and others

Yoshitake, T., Shimakawa, Y., Kuroshima, S., Kimura, H., Ichihashi, T., Kubo, Y., Kasuya, D., Takahashi, K., Kohai, F., Yudasaka, M. & Iijima, S. 2002 Preparation of ne platinum catalyst supported on single-wall carbon nanohorns for fuel cell application. Physica B 323, 124126. Zhan, G. D., Kuntz, J. D., Wan, J. & Mukherjee, A. K. 2003 Single-wall carbon nanotubes as attractive toughening agents in alumina-based nanocomposites. Nature Mater. 2, 3842.

Phil. Trans. R. Soc. Lond. A (2004)