1-s2.0-S0261306906000914-main.pdf

8/9/2019 1-s2.0-S0261306906000914-main.pdf

1/13

Carbon nanotubes Production and industrial applications

Melissa Paradise 1, Tarun Goswami *

Department of Mechanical Engineering, The T.J. Small College of Engineering, Ohio Northern University, Ada, OH 45810, United States

Received 8 August 2005; accepted 10 March 2006Available online 5 May 2006

Abstract

Carbon nanotubes are discussed in this paper from the time of their discovery to present day applications. Specifically the productionmethods, properties and industrial applications of carbon nanotubes are reviewed. Production methods include classical approaches suchas the arc method, chemical vapor deposition, laser ablation, and electric arc discharge along with new methods which are being testedsuch as through solar energy, plasma and microgravity environments. The electrical and mechanical properties and actual structure ofcarbon nanotubes are discussed in detail. Both current applications of carbon nanotubes along with potential uses are also elucidated inthis review. The data has been compiled from open literature to comment on trends in behavior of the carbon nanotubes.2006 Elsevier Ltd. All rights reserved.

Keywords: Single wall nanotubes; Multi-wall nanotubes; Nanometer; Chemical vapor deposition; Arc discharge; Carbon

1. Introduction

Carbon is known to be the most versatile element thatexists on the earth. It has many different properties whichcan be used in different ways depending on how the carbonatoms are arranged. For more than 6000 years carbon hasbeen used for the reduction of metal oxides. Carbon in theform of graphite was discovered in 1779, and 10 years laterin the form of a diamond. It was then determined that bothof these forms belong to a family of chemical elements. Itwas not until about 200 years later that the next advance-ments in carbon took place. In 1985 Kroto, Smalley andCurl2 discovered fullerenes[1]. A few years later the carbonnanotube was discovered.

Carbon nanotubes (CNT) were first discovered in 1991,by Sumio Iijima,3 in fullerene soot [2,3]. It was a productof the carbon-arc discharge method, which is similar to

the method used for fullerenes preparation. In this form,carbon is arranged in tubular formations on a nanoscopic

level. To observe such materials a high resolution transmis-sion electron microscopy was used[3,4]. Carbon nanotubesare a completely new type of carbon fibre which comprisescoaxial cylinders of graphite sheets, which range from 2 to50 sheets [5]. The first observations Sumio made [3] wereof multi-walled nanotubes, and it was not until two yearslater when single wall nanotubes were observed. Ijima alongwith Ichihasi [6] used carbon electrodes with a small amountof iron and filled the chamber around the carbon arc withmethane and argon gas which yielded the single wall carbonnanotube. Single wall nanotubes are basically a single ful-lerene molecule that has been stretched out so their length

is a million times its diameter [7]. Around this same timeDonald Bethune and colleagues also observed the singlewall carbon nanotube[4]. In 1996 Smalley synthesized bun-dles of single wall carbon nanotubes for the first time [5].

The name carbon nanotube is derived from their sizewhich is only a few nanometers wide. By definition carbonnanotubes are cylindrical carbon molecules with propertiesthat make them potentially useful in extremely small scaleelectronic and mechanical applications. These tubes consistof rolled up hexagons, 10,000 times thinner than a humanhair. Ideal nanotubes can be described as a seamless

0261-3069/$ - see front matter 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.matdes.2006.03.008

* Corresponding author. Tel.: +1 419 772 2385; fax: +1 419 772 2404.E-mail address:[email protected](T. Goswami).

1 Junior student in the Department of Mechanical Engineering OhioNorthern University, 45810, United States.2 Recipients of 1996 Nobel Prize in Chemistry for the discovery of

fullerenes.3 Recipient of 2002 Benjamin Franklin medal in Physics for his work on

carbon nanotubes.

www.elsevier.com/locate/matdes

Materials and Design 28 (2007) 14771489

Materials& Design

8/9/2019 1-s2.0-S0261306906000914-main.pdf

2/13

cylinder of rolled up hexagonal networks of carbon atoms,which is capped with half a fullerene molecule at the end[2]. Their strength is one to two orders of magnitude andweight six times lighter than steels. Possible applicationsrange from semiconductors, electronic memory, driveproducts, and medical delivery systems to uses in plastics

such as automobile body panels, paint, tires and as flameretardants in polyethylene and polypropylene[9].Carbon nanotubes have been the focus of considerable

study because of their unusual strength along with excellentmechanical, electrical, thermal and magnetic properties [1100]. Nanotechnology has been recently supported withNanotechnology Research and Development Act allowing$3.7 billion over the next four years to be administered bythe National Nanotechnology Initiative with plans to cre-ate a National Nanotechnology Program (NNP) [10] inthe United States.

2. Production of carbon nanotubes

Various methods since arc growth have been explored toproduce carbon nanotubes. Essentially nanotube structuresare all formed in the same way but the process which causesthe formation differs,Fig. 1. The first method for the pro-duction of multi-wall carbon nanotubes was through arcgrowth[14]Fig. 1(a), but most attractive method commer-cially used is condensationvaporization densation (CVD)method. Under this method there are different ways toinduce the carbon vaporization such as the electric arc dis-charge, continuous or pulsed laser ablation, or solar energy[11]Fig. 1(b). Chemical methods have also been found to

synthesize carbon materials such as the catalytic decompo-sition of hydrocarbons, the production by electrolysis(Fig. 1(c)), heat treatment of a polymer, the low tempera-ture solid pyrolysis, or the in situ catalysis [15]. Recentlya catalytic chemical vapor deposition (CCVD) has alsobeen experimented which may prove to be better than theregular CVD method[12]. Some other methods which alsohave been found to work in the production of carbonnanotubes is the plasma torch method[13]the underwateralternating current (AC) electric arc method [14]and pro-duction in a microgravity environment[8].

3. CVD process

In the CVD process growth involves heating a catalystmaterial to high temperatures (5001000C) in a tube fur-nace using a hydrocarbon gas through the tube reactorover a period of time[16]. The basic mechanism in this pro-cess is the dissociation of hydrocarbon molecules catalyzedby the transition metal and saturation of carbon atoms inthe metal nanoparticle [16]. Precipitation of carbon fromthe metal particle leads to the formation of tubular carbonsolids in a sp2 structure[16].

The characteristics of the carbon nanotubes produced byCVD method depend on the working conditions such as the

temperature and the pressure of operation, the volume and

concentration of methane, the size and the pretreatment ofmetallic catalyst, and the time of reaction. Many times acatalyst is added to speed up the process, to lower high pro-duction costs, and improve the quality of the final product[17]. The type of carbon nanotube produced depends on themetal catalyst used during the gas phase delivery[18]. In theCVD process single wall nanotubes are found to be pro-duced at higher temperatures with a well-dispersed and sup-ported metal catalyst while multi wall nanotubes are formedat lower temperatures and even with the absence of a metalcatalyst[19],Fig. 2. To eliminate impurities formed duringthe process such as graphite compounds, amorphous car-

bon, fullerenes, coal and metal nanoparticles a purification

Fig. 1. Schematic representation of various processes used to produceCNTs: (a) Electric-arc method used at the University of Montpelier(France). (b) Schematic representation of oven laser-vaporization appa-ratus used at Rice University (Houston, Texas, USA). (c) Electrolysisexperimental system (Brighton, UK). (d) Arc discharge and CNTformation and transport in the sheath. (e) Arc-discharge technique. (f)Laser ablation process. (g) Solar furnace from Odeillo (France). (h) Solarexperimental chamber used in Odeillo (France).

1478 M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489

8/9/2019 1-s2.0-S0261306906000914-main.pdf

3/13

is needed. This is achieved by oxidative treatments in the

gaseous phase, liquid phase, acid treatment, micro filtra-tion, thermal treatment and ultrasound methods. Afterthe process is complete the samples need to be characterizedfurther. Techniques such as Raman scattering (RS), thermalgravimetric analysis (TGA), scanning electronic micros-copy (SEM) and atomic force microscopy (AFM) have beenused for such characterization[15].

4. Arc method

The arc method[2], in which carbon nanotubes were dis-covered, is carried out in low pressure He or other neutral

atmosphere (Fig. 1(a)). Seales reaction chambers and vac-

uum equipment are needed to provide the atmosphere.

The products are known to be well graphitized but thereare some problems with this method. The growth needsto be interrupted to remove the product from the chamber[2]. The most widely used process in producing carbonnanotubes is the electric arc discharge method, Fig. 1(de). This same process is also used in producing fullerenes.In this method an electric arc discharge is generatedbetween two graphite electrodes under inert atmosphereof helium or argon. A very high temperature is obtainedwhich allows the sublimation of the carbon. Two kindsof synthesis can be performed in the arc: evaporation ofpure graphite or co-evaporation of graphite and metal

[11]. For the carbon nanotubes to be obtained, purification

Fig. 1 (continued)

M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489 1479

8/9/2019 1-s2.0-S0261306906000914-main.pdf

4/13

by gasification with oxygen or carbon dioxide is needed[20]. The first successful production of multi wall nano-tubes at the gram level was developed in 1992 by Ebbesenand Ajayan[21]. For single wall nanotubes to be obtaineda metal catalyst is needed and this first success of achievingsubstantial amounts came in 1993 by Bethune coworkers[22]. Process parameters involve small gaps between elec-trodes (>1 mm), high current (100 A), plasma betweenthe electrode at about 4000 K, voltage range (3035 V)under specified electrode dimensions.

5. Laser ablation method

The laser ablation method is the second technique forproducing carbon nanotubes which is very useful and pow-erful (Fig. 1(f)). This process is known to produce carbonnanotubes with the highest quality and high purity of singlewalls [23]. Laser ablation was the first technique used togenerate fullerenes in clusters. In this process, a piece ofgraphite is vaporized by laser irradiation under an inertatmosphere. This results in soot containing nanotubeswhich are cooled at the walls of a quartz tube. Two kindsof products are possible: multi walled carbon nanotubes orsingle walled carbon nanotubes [11]. For this process apurification step by gasification is also needed to eliminatecarbonaceous material. The effect of the gasificationdepends on the type of reactant used[24]. The first growthof high quality single wall nanotubes was achieved bySmalley and coworkers[25].

6. Other methods

Another method which is still being explored is throughsolar energy (Fig. 1(gh)). It was used only for fullereneproduction until 1996. In this method nanotubes are now

produced using highly concentrated sunlight from a solar

furnace. The sunlight is focused on a graphite sample andvaporizes the carbon. Soot containing the nanotubes is thencondensed in a dark zone of a reactor, which is collected in afilter and water cooled[11]. Carbon nanotubes can also beproduced under chemical methods. The catalytic decompo-sition of hydrocarbons is performed in a flow furnace at

high temperatures. It results in four structural forms: amor-phous carbon layers on the surface of the catalyst, filamentsof amorphous carbon, graphite layers covering metal parti-cles, and multi wall carbon nanotubes. Electrolysis pro-duces carbon nanotubes by passing an electric current in amolten ionic salt between graphite electrodes[11].

Other methods which have been recently developed suchas the plasma torch method, was designed on the basis thatcarbon nanotubes would naturally grow in any environ-ment in which both appropriate metal atoms and carbonatoms are present. The underwater AC electric arc methodactually combines the underwater growth with the use ofan AC controlled power supply. Using environments such

as microgravity can also help lead to better nanotubes andproduction by eliminating the effects of uncontrolled buoy-ancy[7].

Some of the methods are more effective than others buta problem that all methods face is the ability for the carbonnanotubes to self align. Many applications of carbon nano-tubes require controlled growth of aligned carbon nano-tubes with surface modification. Controlled synthesis ofwell aligned nanotubes in predetermined patterns is partic-ularly important in terms of fundamental studies and appli-cations[26](Fig. 3). Depending on which substrate is beingused in the CVD process two-dimensional (2D) or three-

dimensional (3D) micropatterns can be produced [26].Self-alignment is a key technology in silicon device manu-facturing and could benefit nanomechanical fabricationprocesses because patterned layers can be produced with-out additional lithography steps and could provide moreaccurate alignment than lithography. One successfulmethod has been performed through the synthesis of car-bon nanotubes in an enhanced CVD process on Si wafersand patterned Si wafers with parallel line arrays and holesand using Fe and CoSi

xas a catalyst[27]. This process suc-

cessfully produced carbon nanotubes and carbon nanorodswhich were aligned and parallel to the substrate whichfavors applications towards microelectronic devices[27].

7. Structure

Carbon nanotubes are built from sp2 carbon units andconsist of honeycomb lattices and are a seamless structure.They are tubular having a diameter of a few nanometersbut lengths of many microns. MWNTs are closed graphitetubules rolled like a graphite sheet, Fig. 2. Diameters usu-ally range between 2 and 25 nm, and the distance betweensheets is about 0.34 nm [28], Fig. 3. single-walled carbonnanotubes (SWNT) are made of a single seamlessly rolledgraphite sheet with a typical diameter of about 1.4 nm

which is similar to a buckyball (C60) [16] (Fig. 4). They

Fig. 2. TEM micrograph of a multi-walled carbon nanotube.

1480 M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489

8/9/2019 1-s2.0-S0261306906000914-main.pdf

5/13

have a tendency to form in bundles which are parallel incontact and consist of tens to hundreds of nanotubes [29].

Depending on how the grapheme walls of the nanotubeare rolled together they can result in an armchair, zigzag orchiral shapes (Fig. 5). These groups are distinguished bytheir unit cells which are determined by the chiral vectorgiven by the equation: Ch na1 ma2 where a1 and a2are unit vectors in the two-dimensional hexagonal lattice,and n and m are integers. Another important parameteris the chiral angle, which is the angle between Ch and a1(Fig. 6). When n=m and the chiral angle is 30 degrees it

is known as an armchair type. When m or n are zero and

the chiral angle is equal to zero the nanotube is known aszigzag. Chiral nanotubes are therefore when the chiralangles are between 0 and 30. The diameter is found bythe equation dt O3=paccm

2 mn n21=2, where acc

is the distance between neighboring carbon atoms in theflat sheet. The phase difference is known to be 2P, where,for example, 10 hexagons are around the circumferenceof a zigzag type, the 11th would collide with the first whenit comes around the circumference once [30].

The chiral angles along with diameter determine theproperties of the nanotube. Studies of optical properties

of nanotubes show that in most cases they act as semi

Fig. 3. Micrographs showing carbon nanotubes (a) macrograph of carbon nanotubes, (b) scanning electron micrographs of CNTs at 10,000 and 20,000magnification, (c) aligned carbon nanotubes.

M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489 1481

8/9/2019 1-s2.0-S0261306906000914-main.pdf

6/13

conductors but in a few rare cases they act as metallic. Thismetallic behavior only happens when n m= 3L andL= 0, resulting in the (HOMOLUMO) fundamentalgap being 0.0 eV. The electronic properties are a result ofthe electrons being normal to the nanotube axis. While act-ing as a semi conductor the fundamental gap was found tobe 0.5 eV, which was a function of the diameter whichcauses them to exist as ropes in their native state [30].The energy gap is found by Egap= 2y0acc/d, where y0 isthe CC tight bonding overlap energy (2.7 0.1 eV), accis the nearest neighbor CC distance (0.142 nm), and d isthe diameter. Studies also showed that a small gap wouldexist because ofP/r bonding orbital and P*/r* anti-bond-

ing orbital at the Fermi level. The Fermi energy is the high-

est occupied orbital, has finite density neighboring carbonatoms in the flat sheet. The phase difference is known to

be 2P, levels for a metallic tube and zero for a semiconduc-tor. The density state occurs at sharp peaks as the energylevel is increased[4].

8. Properties

Carbon nanotubes are unique nanostructures which areknown to have remarkable electronic and mechanicalproperties. These characteristics have sparked great interestin their possible uses for nano-electronic and nano-mechanical devices. Properties of carbon nanotubes canalso be expanded to thermal and optical properties as well.

Carbon nanotubes are predicted to have high stiffness andaxial strength as a result of the carboncarbon sp2 bonding[31]. Studies exploring the elastic response, inelastic behav-ior and buckling yield strength and fracture need to be con-ducted to find practical uses of the nanotubes.

The mechanical properties of a solid must ultimatelydepend on the strength of its interatomic bonds. Withknowledge of known properties of crystal graphite themechanical properties of carbon nanotubes can be pre-dicted with some confidence [32]. Experimental and theo-retical results have shown an elastic modulus of greaterthan 1 TPa (that of a diamond is 1.2 TPa) and havereported strengths 10100 times higher than the strongeststeel at a fraction of the weight[33]. It has been predictedthat carbon nanotubes have the highest Youngs modulusof all different types of composite tubes such as BN, BC3,BC2N, C3N4, CN, etc. [34] (Table 1). The definition ofYoungs modulus involves the second derivative of theenergy with respect to the applied stress/strain. In general,the strength of the chemical bonds determines the actualvalue of Youngs modulus and smaller diameters result ina smaller Youngs modulus. However, in tests conductedon carbon nanotubes show that little dependence existson the diameter of the tube with Youngs modulus, whichdoes help to hypothesize that carbon nanotubes do possess

the highest Youngs modulus which is expected around

Fig. 4. Structures of (a) diamond, graphite, and fullerene (from R.E.

Smalley), (b) a single-wall helical carbon nanotube [3].

Fig. 5. Illustrations of the atomic structure of (a) an armchair and (b) aziz-zag nanotube[33].

Fig. 6. Schematic diagram showing how a hexagonal sheet of graphite isrolled to form a carbon nanotube[33].

1482 M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489

8/9/2019 1-s2.0-S0261306906000914-main.pdf

7/13

1 TPa[35]. Experiments conducted have resulted in tensilestrengths in the range from 11 to 63 GPa, with dependenceon the outer shell diameter, which is not far from the the-oretical yield strength of 100 GPa.

Due to high in-plane tensile strength of graphite, bothsingle and multi wall carbon nanotubes, are expected tohave large bending constants since they mostly depend onYoungs modulus. The nanotube has been found to be veryflexible. It can be elongated, twisted, flattened, or bent into

circles before fracturing. Simulations conducted by Bern-holc and colleagues indicate it can regain their originalshape. Their kink-like ridges allow the structure to relaxelastically while under compression, unlike carbon fiberswhich fracture easily[4].

The unique elastic and inelastic properties have broughtabout more studies on the durability of carbon nanotubes.For single wall nanotubes simulations of deformationsshowed that each shape change corresponded directly toan abrupt release in energy and a singularity in thestress/strain curve. The nanotubes were found to have anextremely large breaking strain which decreased with tem-

perature. However, it was concluded single wall nanotubeswere subject to buckling under high pressure, which isresponsible for the pressure induced abnormalities of vibra-tion modes and electrical resistivity (Fig. 7). The elasticmodulus, Poissons ratio and bulk modulus were all foundto be directly affected by the tubes radius. A max bulkmodulus was found to be 38 GPa with samples having aradius of 0.6 nm. For multi-wall nanotubes the propertieswere a little more complicated to calculate. An empiricallattice dynamics model was used, which showed thatmulti-wall nanotubes were insensitive to parameters suchas the chirality, tube radius, and the number of layers.

Thermal properties including specific heat and thermal

conductivity of carbon nanotubes are determined primarilyby the phonons [31]. A phonon is a quantum acousticenergy similar to the photon. Phonons are a result of latticevibrations observed in the Raman spectra[4]. Especially atlow temperatures the phonon contribution to these quanti-ties dominates and is due to the acoustic phonons. Themeasurements of thermoelectric power of nanotube sys-tems give direct information for the type of carriers andconductivity mechanisms.

Theoretical and experimental results show superior elec-trical properties of carbon nanotubes. They can produceelectric current carrying capacity 1000 times higher than cop-

per wires[36]. For 1D systems cylindrical surface, transla-

tional symmetry with a screw axis could affect theelectronic structures and related properties. The electroniccapabilities possessed by carbon nanotubes are seen to arisepredominately from interlayer interactions,rather than frominterlayer interactions between multilayers within a singlecarbon nanotube or between different nanotubes[37].

These optical properties have proved to be especiallyunique with capabilities of acting as either a metallic orsemiconductor, which depends on tubule diameter and chi-ral angle. Studies have shown that metallic conduction canbe achieved without introduction of doping effects. Forsemiconducting nanotubes the band gaps have been foundto be proportional to a fraction of the diameter and with-out relation to the tubule chirality[37]. The I-tight-bindingmodel within the zone folding scheme shows, one third ofcarbon nanotubes are found to be metallic while two thirdsare semiconducting, depending on their indices[31]. Calcu-lations based on the use ofr and P bands, due to curva-ture induced mixing of these bands, are used to predict

that some metallic nanotubes are very-small-band-gapsemiconducting nanotubes [38] (Fig. 8). The symmetry ofthe structures basically relates all the calculations in bothsingle and multi-wall carbon nanotubes. Electronic proper-ties of bundles of single wall nanotubes can be derived,assuming the intertube interactions are not strong enoughto change the band structure. Broken symmetry causedby interactions between tubes in a bundle create a pseudo-gap of about 0.2 and 0.1 eV[39]. This pseudogap, which iscreated can modify electronic properties such as semimetal-like temperature dependence of the electrical conductivityand a finite gap in the infrared absorption spectrum is pre-

dicted[31].

Table 1Mechanical properties of carbon nanotubes[5]

Material Youngs modulus(GPa)

Tensile strength(GPa)

Density(g/cm3)

Single wall nanotube 1054 150Multi wall nanotube 1200 150 2.6Steel 208 0.4 7.8

Epoxy 3.5 0.005 1.25Wood 16 0.008 0.6

Fig. 7. TEM micrograph and computer simulation of nanotube buckling[33].

M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489 1483

8/9/2019 1-s2.0-S0261306906000914-main.pdf

8/13

9. Applications

Carbon nanotubes have attracted a great deal of atten-tion world wide with their unique properties which areleading to many promising applications. Potential practicalapplications have been reported such as chemical sensors[40], field emission materials [41], catalyst support [42],electronic devices [43], high sensitivity nanobalance fornanoscopic particles [43], nanotweezers [44], reinforce-ments in high performance composites, and as nanoprobesin meteorology and biomedical and chemical investiga-

tions, anode for lithium ion in batteries[45], nanoelectronicdevices [46], supercapacitors [47] and hydrogen storage[48]. New applications are likely in the diamond industrysince experiments have shown the conversion of carbonnanotubes to diamond under high pressure and high tem-peratures with the presence of a certain catalyst[49]. Theseare just a few possibilities that are currently being explored.As research continues, new applications will also develop.

10. Composites

Given the mechanical properties that have been reportedon carbon nanotubes, an entire new class of composite mate-rials may be possible with the use of carbon nanotubes. Thefirst commercially recognized use for multi wall nanotubeswas electrically conducting components in polymer compos-ites [50]. The matrices used in carbon nanotubes incorpo-rated into composites can improve the electrical propertieswhich can act as a polymer, metal, or metal oxide [14]. Car-bon nanotube metal or metal oxide composites have beenmade to improve electrical conductivity. For applicationsin polymer nanocomposites the elastic and fracture proper-ties of carbon nanotubes must be understood along withinteractions at the nanotube matrix interface. The perfor-mance of carbon nanotubes in a polymer or ceramic matrix

is well above traditional fillers such as carbon black or ultra

fine metal powders[51]. The major difference from conven-tional fiber-reinforced composites in that the scale is nar-rowed down to nanometers instead of micrometers[33].

Large similarities between mechanical properties of apolymer film and a SWNT matrix exist in that both havehigh viscoelasticity that can be evaluated using a nanoin-

dentor[52]. It would be difficult to replace all carbon fibersin their uses since there has been so much work done withthem. It is better for carbon nanotube research to look to anew market rather replace the old. The great novelty withcarbon nanotubes is that they can achieve high stiffnessalong with high strength[34]. Also studies have shown thatcarbon nanotubes do perform as reinforcing elements withpolymer [53], ceramic [54] and metallic matrices [55], butwithout alignment their performance in terms of strengthand stiffness fall short of traditional carbon fibers.

For industrial applications as composites large quanti-ties of nanotubes will be needed. It has been found thatthe best method for high quantity and low cost production

of nanotubes is provided through the CVD method. Costfactors also lead more to the use of multi wall nanotubesrather than single wall nanotubes[50]. Incorporating nano-tubes into plastics can lead to a dramatically increasedmodulus of elasticity and strength in structural materials.The main problem still lies in producing the nanotubes sothey are uniformly dispersed, achieving nanotube-matrixadhesion providing stress transfer and intra bundle slidingin single wall nanotubes [50]. Promising results have beenobserved by Biercuk and others to overcome these prob-lems by increasing Vickers hardness with single wall nano-tubes and increasing the modulus of elasticity and breaking

stress in polystyrene using multiwall nanotubes[56].Nanotube reinforced composites have already been suc-

cessfully created. Experiments on a fully integrated nano-tube composite using single wall nanotubes demonstrateddramatic enhancement of mechanical properties. To pro-duce these composites a reaction of terminal diamines withalkycarboxl groups attached to single wall nanotubes in thecourse of dicarboxxlic acid acyl peroxide treatment wasneeded. The ultimate strength and shear modulus increasedfrom 30% to 70% with only the addition of 14 wt% of sin-gle wall nanotubes. The strain to failure also increasedshowing an increase in toughness[57](Fig. 9).

Rubber compounds reinforced by nanotubes are poten-tial applications in tire industry. By replacing the carbonblack with carbon nanotubes improved skid resistanceand reduced abrasion of the tire have been found in exper-imental results[58]. Carbon nanotubes may provide a safer,faster, and eventually cheaper transportation [59] in thefuture. Although expectations of carbon nanotubes are veryhigh for their use in composites there has been some specu-lation against the results they produce when mixed withsome polymers and plastics. Carbon nanotubes themselvesare superior conductors by themselves but they may notexhibit the same level of conductivity when integrated intoother materials[60]. Experiments have shown the conduc-

tivity to increase thermal conductivity by two or threefold

Fig. 8. Band-gap values vs. nanotube diameters define nanotubes asmetallic or semiconducting[5].

1484 M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489

8/9/2019 1-s2.0-S0261306906000914-main.pdf

9/13

when it should have been close to 50 fold[60]. The problemis that carbon nanotubes vibrate at much higher frequencies

than the atoms in surrounding material which causes theresistance to be so high the thermal conductivity is limited[60]. Inducing stronger bonds between the nanotube andother material might help in solving the problem [60]. Theuse of carbon nanotubes to improve materials will be inves-tigated in the future as production increases and applicabil-ity in industrial settings become possible.

11. Sensors and probes

Carbon nanotubes have proved to have some advanta-ges for sensing applications. Their small size with larger

surface; high sensitivity, fast response and good reversibil-ity at room temperature enable them as a gas molecule sen-sor[61]; enhanced electron transfer when used as electrodesin electrochemical reactions[62]; and easy protein immobi-lization with retention of activity as potential biosensors[62]are among some of the desirable applications. Studieshave shown that surface modification performed onaligned carbon nanotubes even furthers the sensitivity ofnanotube sensors[25]. The main advantage of these sensorsare the nanscopic size of the nanotube sensing element andthe corresponding nanoscopic size of the material requiredfor a response[50]. The mechanical robustness of the nano-tubes and the low buckling force increase the probe lifealong and minimizes damage during repeated hard crashesinto substrates [49]. The cylindrical shape and small tubediameter also allow for imaging in narrow deep crevicesand improve resolution in comparison to conventionalnanoprobes, especially for high sample feature heights[63].

Electronic properties suggest carbon nanotubes will beable to mediate electron transfer reactions with electro activespecies in a solution when used as electrode material [64].This leads to the idea that carbon nanotube based electrodescan be used in miniature chemical sensing [65]. Electrodematerials with carbon nanotubes resulted in better behaviorthan traditional carbon electrodes including good conduct-

ing ability and high chemical stability [29]. The electrical

resistivity of single wall nanotubes have been found tochange sensitively on exposure to gaseous ambients contain-ing NO2 , NH3 ,andO2. By monitoring this change the pres-ence of gases could be detected. Results showed are at leastan order of magnitude faster than those currently availableand that they could be operated at room temperature or at

higher temperatures for sensing applications[66]. This sens-ing application is now being researched for its use on auto-motive tires. A tiny sensor would be able to monitor andreport tire pressure to the driver while being able to with-stand extreme temperature and vibrations[58].

Since multi wall nanotubes are conducting they can beused as scanning probes on microscope tips in instrumentssuch as a scanning tunneling microscope (STM), atomicforce microscope (AFM) and electrostatic force micro-scopes (Fig. 10). With their ultra high sensitivity, high res-olution electron microscopes which have sub-nanoscaleaccuracy have the ability to obtain information on theatomic arrangement, element identification and electronic

structure of nanocarbon materials [67]. Nanotubes tipscan also be used for high resolution imaging or as activetools for surface manipulation. On an AFM tip they canbe controlled like tweezers to pick up and release nanoscalestructures[68]. Nanoscopic tweezers have been made thatare driven by the electrostatic interaction between twonanotubes on a probe tip[69].

Studies have shown the reversible bending of nanotubescan be used to alter their conduction. Optimal designs suchas the zigzag and armchair nanotubes were observed tohave a difference in mechanical response at large bendingand the current passing through metallic structures

decreasing at larger bending angles as the semiconductorincreases [70]. The correspondence between mechanicalresponse and electronic transport has been proven poten-tial applications of nanotubes in such applications asnano-electro-mechanical sensors and even switches[71].

Fig. 10. Use of a MWNT as an AFM tip At the center of the vapor growncarbon fiber (VGCF) is a MWNT which forms the tip. The VGCFprovides a convenient and robust technique for mounting the MWNT

probe for use in a scanning probe instrument[66].

Fig. 9. Presence of 5 wt% multi wall carbon nanotubes results in a steeperslope in the stressstrain curve [66].

M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489 1485

8/9/2019 1-s2.0-S0261306906000914-main.pdf

10/13

Aligned multi-wall carbon nanotubes are now beingused for the development of an amperometric biosensor[70]. Electrodes modified with carbon nanotubes are usedfor the immobilization of enzymes and other redox pro-teins on the ends of aligned nanotube arrays [72], on thewalls of carbon nanotubes[73] and inside nanotubes[74].

It has been shown that small proteins can be entrappedinto the inner channel of opened carbon nanotubes by sim-ple absorption[75]. Azamaian et al.[73]demonstrated theprincipal where glucose oxidase was absorbed along thelength of carbon nanotubes and randomly distributed ona glassy carbon electrode. The key in this design is theestablishment of electron transfer between enzyme activesite and electrochemical conducer [71]. Small surface arealeads to constraints on enzyme loading[76]. Carbon nano-tubes posses the high surface area needed along with thestructure dependant metallic character to promote electrontransfer reactions at low potentials [77]. Based on results,chemical etching was proven to be most efficient when

opening carbon nanotubes and allowing the entrance ofthe enzyme at the inner shell [71].

Basic electronic properties of semiconducting carbonnanotubes change when placed in a magnetic field [78].The band gap shrank which is unique among knownmaterials [78]. Nanotubes band gaps are comparable withsilicon and gallium arsenide which are currently the main-stays of the computer industry because their narrow bandgaps correspond with how much electricity it takes to flipa transistor from on to off[78]. With the possibility of car-bon nanotubes band gap disappearing all together in thepresence of stronger magnetic fields, they could take over

the roles of silicon and gallium arsenide potentially revolu-tionizing the computer industry[78].

12. Field emission devices

Field emission is a quantum effect when compared tothermionic emission. For technological applications, elec-

tron emissive materials should have low threshold emis-sion fields and should be stable at high current density [66](Table 2). Carbon nanotubes posses the right combinationof properties: nanometer size diameter, structural integrity,high electrical conductivity, and chemical stability thatmake good electron emitters [79]. The first field emissionfrom carbon nanotubes was performed in 1995 by Rinzlerfrom single isolated multi wall nanotubes[80]and by multiwall nanotube film by de Heer [81]. Research on electronicdevices has since focused primarily on the use of singleand multi wall carbon nanotubes as field emission electronsources[82]for flat panel displays[83], lamps[84], gas dis-charge tubes providing surge protection [85], and X-ray[86] and microwave generators [87]. A potential appliedbetween a nanotube coated surface and an anode createshigh electric fields which is a result of a small radius of the

nanofiber tip and the length of the nanofiber [50]. The localfields cause electrons to tunnel from the nanotube tip to thetunnel. This process of nanotube tip electron emission dif-fers from that of bulk metals because it arises from discreteenergy states instead of continuous electronic bands and itsbehavior depends on the nanotube tip structure, single wallnanotubes[88]or multi wall nanotubes[84](Fig. 11).

Table 2Threshold electrical field values for different materials for a 10 mA/cm 2

current density[66]

Material Threshold electrical field (V/m)

Mo tips 50100Si tips 50100p-type semiconducting diamond 130

Undoped, defective CVD diamond 30120Amorphous diamond 2040Cs-coated diamond 2030Graphite powder (30 mA/cm2)Carbon nanotubesb 13 (stable at 1 A/cm2)

a Heat-treated in H plasma.b Random SWNT film.

Fig. 11. Left: Schematic of a prototype field emission display using carbon nanotubes. Right: A prototype 4.5_ field emission display fabricated by

Samsung using carbon nanotubes (image provided by Dr. W. Choi of Samsung Advanced Institute of Technologies).

1486 M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489

8/9/2019 1-s2.0-S0261306906000914-main.pdf

11/13

13. Flat panel displays

Flat panel displays are one of the more lucrative appli-cations of carbon nanotubes but are also the most techni-cally complex. Nanotubes are at an advantage over liquidcrystal displays since they have low power consumption,

high brightness, a wide viewing angle, a fast response rateand a wide operating system[50]. In the actual process elec-tric fields direct the field-emitted electrons toward theanode where phosphorus produces light for the flat paneldisplay[50]. Prototype matrix-addressable diode flat paneldisplays have been constructed at Northwestern University[66]. One demonstration consists of nanotube-epoxy stripeson the cathode glass plate and phosphor coated indium-tin-oxide (ITO) stripes on the anode plate [89]. Pixels are thenformed at the intersection of the cathode and anode stripes.Pulses of 150 V are switched among anode and cathodestripes to produce an image [66].

14. Nanotube-based lamps

Nanotube-based lamps are similar to displays compris-ing of a nanotube-coated surface opposing a phosphor-coated substrate, but they are less technically challengingand require less investment [50]. With lifetimes expectedin excess of 8000 h they can look to replace environmen-tally problematic mercury-based fluorescent lamps usedin stadium style displays [84]. Nanotube-based gas dis-charge tubes might also find commercial use in protectingtelecommunications networks from power surges [85].Another application arises if a metal target is used to

replace the phosphorescent screen at the anode. This causesthe accelerating voltage to increase producing X-raysinstead of light [50]. The compact geometry of the nano-tube based X-ray lead to potential uses for X-ray endo-scopes and medical exploration[50].

15. Energy storage

Graphite, carbonaceous materials and carbon fiber elec-trodes have been used for decades in fuel cells, batteriesand several other electrochemical applications[90]. Carbonnanotubes are now being considered for energy storageand production because of their small dimensions, a smoothsurface topology, and perfect surface specificity since onlythe graphite planes are exposed in their structure [66]. Theefficiency of the fuel cellsis determined by the rate of electrontransfer at carbon electrodes, which has been shown by sev-eral experiments to be fastest on carbon nanotubes[91].

The area of hydrogen storage is one of the most activestudies involving energy storage yet also the most contro-versial. Extremely high and reversible hydrogen storagehas been reported in materials containing single wall nano-tubes[92]along with graphite nanofibers fibers [93]whichhas attracted interest both in industry along with the aca-demic world (Table 3). The problem remains, however, in

a lack of understanding of the basic mechanisms of hydro-

gen storage in these materials. The main ways to storehydrogen is by metal hybrids, cryo-absorption, and bythe gas phase in metal hybrids [66]. Due to carbon nano-

tubes cylindrical shape and geometry, and nanometer scale diameters, it has been predicted that they will be ableto store liquid as gas in the inner cores through capillaryeffect improving metal hybrid batteries[94].

16. Electrochemical devices

Carbon nanotubes have been studied for their potentialuses as electrodes for devices that use electrochemical dou-ble layer charge injection because of their high electro-chemically accessible surface area of porous nanotubearrays combined with high electric conductivity [50].

Examples of such applications include Supercapacitorswhich have capacitances much larger than ordinary dielec-tric based capacitor and electrochemical actuators whichmay potentially be used in robots [50]. The capacitancefor an electrochemical device depends on the separationbetween the charge on the electrode and countercharge inthe electrolyte. Since this distance is about a nanometerfor nanotubes in electrodes compared to a micrometer inordinary dielectric capacitors, extremely large capacitancesresult from the high nanotube surface are accessible to theelectrolyte[50]. The use of nanotubes as electrodes in lith-ium batteries is a possibility because of the high reversiblecomponent of storage capacity at high discharge rates[50].The reversible capacity reported with single wall nanotubesis 1000 mA h/g compared to 372 mA h/g for graphite [95]and 708 mA h/g for ball milled graphite[79].

17. Nanometer-sized electronic devices

Recent advances have led to the idea that nanotubes willbe useful for downsizing circuit dimensions. Presently, cur-rent-induced electromigration causes conventional metalwires interconnects to fail when the diameter becomes toosmall[50]. The covalently bonded structure of carbon nano-tubes militates against similar breakdown of nanotube wires

and because of ballistic transport the intrinsic resistance of

Table 3Hydrogen storage of carbon nanotubes to other carbon materials[66]

Material Max. wt% H2 T(K) P(MPa)

SWNTs (low purity) 510 133 0.040SWNTs (high purity) 4 300 0.040GNFs (tubular) 11.26 298 11.35GNFs (herringbone) 67.55 298 11.35

GNS (platelet) 53.68 298 11.35Graphite 4.52 298 11.35GNFs 0.4 298773 0.101Li-GNFs 20 473673 0.101Li-graphites 14 473674 0.101K-GNFs 14

8/9/2019 1-s2.0-S0261306906000914-main.pdf

12/13

the nanotube should essentially vanish [50]. Experimentalresults have shown that metallic single wall nanotubes cancarry up to 109 A/cm2 compared to current densities fornormal metals being only 105 A/cm2 [96].

The research of field effect transistors (NT-FETs) aimsto replace source drain channel structure with a nanotube.

Transistors assembled with carbon nanotubes may or maynot work however depending on whether the chosen nano-tube is semiconducting or metallic, which the operator hasno control over[50]. It might be possible to peel back layersfrom multi-wall nanotubes to achieve desired propertiesbut advances in microlithography are still needed to perfectthis reduction method. Recent developments have focusedthe media attention to nanotube nanoelectronic applica-tions [50]. Crossed single wall nanotubes have been usedin producing three and four-terminal electronic devices[97] along with nonvolatile memory that functions like aelectromechanical relay [98]. Nanotube transistors [99]have also been reported using integrated nanotubes which

may lead to large scale integration. Patterned growth ofcarbon nanotubes on silicon wafers [100] may prove tobe the step needed to integrate nanotubes into electronics.

18. Conclusions

Carbon nanotubes may have only recently caught theattention of the world but many advances have been madesince their discovery about a decade ago. They are uniquenanostructures that display the desirable properties of anyother known material. The techniques of production havealso come a long way but still have some room to be more

efficient and cost effective. They have amazing electronicand mechanical properties which lead to incredible formsof strength, and conductivity. Due to these qualities thefield of applications is almost endless. From reinforcementsin composites, sensors and probes, energy storage, electro-chemical devices and nanometer sized electronics carbonnanotubes could revolutionize the world.

Acknowledgements

One of the authors (T.G.) acknowledge Mr. TomHughes of Applied Science Inc. for providing insights onthis subject and data. Mr. David Bennett assisted withthe illustrations. This research was a part of summer re-search experience for undergraduates funded by OhioNorthern University.

References

[1] Scharff P. New carbon materials for research and technology.Carbon 1998;36(56):4816.

[2] Biro LP, Horvath ZE, Szlamas L, Kertesz K, Weber F, Juhasz G,et al. Continuous carbon nanotube production in underwater ACelectric arc. Chem Phys Lett 2003:399402.

[3] Burstein E. A major milestone in nanoscale material science: the2002 Benjamin Franklin Medal in Physics presented to Sumio Iijima,

2003;340(34):22142.

[4] Dresselhaus MS, Dresselhaus G, Eklund PC. Science of fullerenesand carbon nanotubes. New York: Academic Press; 1996.

[5] Yamabe T. Recent development of carbon nanotubes. Synthetic Met1995:15118.

[6] Iijima S, Ichihashi T. Single shell carbon nanotubes of one diameter.Nature 1993;636:603.

[7] Alford JM, Mason GR, Feikema DA. Formation of carbonnanotubes in a microgravity environment, Sixth internationalmicrogravity combustion workshop, NASA Glenn Research Center,Cleveland, OH, CP-2001-210826, 2001. p. 2936, May 2224.

[8] Massachusetts Technology Collaborative and The Nano Science andTechnology Institute, Nanotechnology and its impact on industry.Industry Summaries, 1997.

[9] Dunn T, Morse BB, Pendleton PC. Nanotechnology: The bigpicture, The newsletter of the MIT enterprise forum of Cambridge,2004;22(8).

[10] Laplaze D, Alvarez L, Badie JM, Flamant G. Carbon nanotubes:dynamics of synthesis processes. Carbon 2002:162134.

[11] Journet C, Bernier P. Production of carbon nanotubes. Appl. Phys.A 1998:19.

[12] Endo M. Mass production, selective formation, and applications ofcarbon nanotubes; Shinsu University; 1995.

[13] Chen C, Perry W, Xu H, Jiang Y, Phillips J. Plasma torchproduction of macroscopic carbon nanotube structures. Carbon2003:255560.

[14] Liu Y, Gao L. A study of electrical properties of nanotubeNiFe2O4composites: effect of surface treatment of carbon nanotubes. Carbon2005:4752.

[15] Izaskun B, Ingio F, Ainara G, Belsue M, Roberto M. Synthesispurification and characterization of carbon nanotubes. Poster2004:137.

[16] Dai H. Carbon nanotubes: opportunities and challenges. Surface Sci2002;500(13):21841.

[17] Resasco DE, Alvarez WE, Pompeo F, Balzano L, Herrera JE,Kitiyanan B, et al. J Nanopart Res 2002;4:131.

[18] Johnson R. CVD process tames carbon nanotube growth. EE Times,2002.

[19] Qingwen L, Jin Y, Zhogfan L. Dependence of the formation ofcarbon nanotubes on the chemical structures of hydrocarbons. AdvNanomater Nanodev 2002:5971.

[20] Morishita K, Takarada T. Scanning electron microscope observa-tion of the purification behavior of carbon nanotubes. J Mater Sci1999;34(6):116974.

[21] Ebbesen TW, Ajayan PM. Large-scale synthesis of carbon nano-tubes. Nature 1992;358:2202.

[22] Bethune DS, Kiang CH, DeVries M, Gorman G, Savoy R, VazquezJ, et al. Cobalt-catalysed growth of carbon nanotubes with single-atomic-layer walls. Nature 1993;363:6057.

[23] Chiang M, Liu K, Lai T, Tsai C, Cheng H, Lin I. Electron fieldemission properties of pulsed laser deposited carbon films containingcarbon nanotubes. J Vac Sci Technol B 2001;19(3):10349.

[24] Srivastava D, Brenner DW, Schall JD, Ausman KD, Yu M, Ruoff

RS. J. Phys. Chem. B [in press].[25] Soundarrajan P, Patil A, Liming D. Surface modification of alignedcarbon nanotube arrays for electrochemical sensing applications.American Vacuum Society; 2002. p. 1198201.

[26] Wang B, Liu X, Wu D, Wang H, Jiang J, Wang X, et al.Controllable preparation of patterns of aligned carbon nanotubes onmetals and metal-coated silicon substrates. J Mater Chem2003:11246.

[27] Chang H, Lin C, Kuo C. Iron and cobalt silicide catalysts-assistedcarbon nanostructures on the patterned Si substrates. Thin SolidFilms 2002;420421:21924.

[28] Ajayan PM. Chem Rev 1999;99:1787.[29] Zhao Q, Gan Z, Zhuang Q. Electrochemical sensors based on

carbon nanotubes. Electroanalysis 2002(23):160913.[30] Saito Y, Uemura S. Field emission from carbon nanotubes and its

application to electron sources. Carbon 1999;38(2):16982.

1488 M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489

8/9/2019 1-s2.0-S0261306906000914-main.pdf

13/13

[31] Popov V. Carbon nanotubes: properties and application. Mater SciEng R. Rep 2004;43(3):61102.

[32] Ruoff R, Lorents D. Mechanical and thermal properties of carbonnanotubes. Carbon 1995;33(7):92530.

[33] Thoteson E, Ren Z, Chou T. Advances in the science and technologyof carbon nanotubes and their composites: a review. Compos SciTechnol 2001;61(13):1899912.

[34] Delmotte JP, Rubio A. Mechanical properties of carbon nanotubes:a fiber digest for beginners. Carbon 2002;40(10):172934.

[35] Whittaker, Brink J, Husseni M, Linford G, Davis M. Self-alignedmechanical attachment of carbon nanotubes to silicon dioxidestructures by selective silicon dioxide chemical-vapor deposition.Appl Sci Lett 2003;83(25).

[36] Collins PG, Avouris P. Nanotubes for electronics. Sci Am2000;283:629.

[37] Dresselhaus M, Dresselhaus G, Saito R. Physics of carbon nano-tubes. Carbon 1995;33(7):88391. Pergamon.

[38] Hamada N, Sawada S, Oshiyama A. New one-dimensional conduc-tors: graphitic microtubules. Phys Rev Lett 1992;68:1579.

[39] Kwon YK, Tomanek D. Electronic and structural properties ofmultiwall carbon nanotubes. Phys Rev B 1998;58:R16001.

[40] Ong JK, Franklin NR, Zhou C, Chapline MG, Peng S, Cho K,et al. Science 2000;287:622.

[41] Tans SJ, Verschueren RM, Dekker C. Nature 1998:393.[42] Planeix JM, Coustel N, Coq B, Brotons V, Kumbhar PS, Dutartre

R, et al. J Am Chem Soc 1994;116:7935.[43] Saito S. Science 1997;278:77.[44] Kim P, Lieber CM. Science 1999;286:2148.[45] Che G, Lakshmi BB, Fisher ER, Martin CR. Nature 1996;384:147.[46] Collins PG, Zettle A, Bando H, Thess A, Smally RE. Science

1997;278:100.[47] Che G, Lakshmi BB, Fisher ER, Martin CR. Nature 1998;393:346.[48] Journet C, Maser WK, Bernier P, Loiseau A, Lamy de la M,

Chapelle S, et al. Nature 1997;308:756.[49] Wang WK, Cao LM. Transformation of carbon nanotubes to

diamond at high pressure and high temperature. Russ Phys J2001;44(2):17882.

[50] Baugham R, Zakhidov A, Heer W. Carbon nanotubes the routetoward applications. Science 2002;297:78792.

[51] Rutkofsky M, Banash M, Rajagopal R, Chen J. Zyvex Application2005.

[52] Salvetat JP, Loubet JL [unpublished results].[53] Schadler LS, Giannaris SC, Ajayan PM. Load transfer in carbon

nanotube epoxy composites. Appl Phys Lett 1998;73(26):38424.[54] Peigney A, Laurent C, Rousset A. Synthesis and characterization of

alumina matrix nanocomposites containing carbon nanotubes. KeyEng Mat 1997;132136:7436.

[55] Kuzumaki T, Miyazawa K, Ichinose H, Ito K. Processing of carbonnanotube reinforced aluminum composite. J Mater Res1998;13(9):24459.

[56] Biercuk MJ et al. Appl Phys Lett 2002;80:2767.[57] Zhu J, Peng H, Rodriguez F, Margrave J, Khabashesku V, Imam A,

et al. Reinforcing epoxy polymer composites through covalentintegration of functionalized nanotubes. Adv Functional Mater2004:6438.

[58] Smith J. Slicing it extra thin, Tireview, 2005. Available from:http://www.tireview.com.

[59] Bernard A. Applied sciences aims its nanotubes at the near-termmarket for polymers, SMALLTIMES, 2002. Available from:http://www.smalltimes.com.

[60] Rensselaer Polytechnic Institute, Thermal superconductivity incarbon nanotubes not so Super when added to certain materials,Nano Tsunami, 2003. Available from:http://www.sciencedaily.com/releases/2003/11/031112072719.htm.

[61] Tang ZK et al. Science 2001;292:2462.[62] Tennent HG. US Patent 4,663,230 (5 May 1987).[63] Hafner JH, Cheung CL, Leiber CM. Nature 1999;398:761.[64] Britto PJ, Santhanam KSV, Ajayan PM. Bioelectrochem Bioenerg

1996;41:121.[65] Kong NR, Franklin C, Zhou MC, Chapline S, Peng K, Cho HD.

Science 2000;287(622):40615.[66] Ajayan P, Zhou O. Applications of carbon nanotubes, carbon

nanotubes. University of North Carolina; 2001. p. 391425.[67] Yumura M. Carbon nanotube industrial applications, Research

Center for Advanced Carbon Materials; 2004, No. 10.[68] Kim P, Lieber CM. Science 1999;286(2148):414.[69] Kim P, Lieber CM. Science 1999;286(2148).[70] Lee C, Liu X, Zhou C. Carbon nanotube field-effect inverters. Appl

Phys Lett 2001;79(20).[71] Farajian A, Yakobson B, Mizeseki H, Kawazoe Y. Electronic

transport through bent carbon nanotubes: nanoelectromechanicalsensors and switches. Phys Rev 2003;67:16.

[72] Gao M, Dai LM, Wallace GG. Electroanalysis 2003;15:1089.[73] Azamian BR, Davis JJ, Coleman KS, Bagshaw CB, Green MLH. J

Am Chem Soc 2002;124:12664.[74] Gooding J. Nanostructuring electrodes with carbon nanotubes: a

review on electrochemistry and applications for sensing. Electrochi-mica 2004;50(15):304960.

[75] Iijima S, Barbec C, Maiti A, Bernholc J. J Chem Phys1996;104:2089.

[76] Lammert PE, Zhang P, Crespi VH. Phys Rev Lett 2000;84:2453.[77] Tekleab D, Carroll DL, Samsonidze GG, Yakobson BI. Phys Rev B

2001;64:035419.[78] Rice University, Magnetic forces may turn some nanotubes into

metals, Nano Tsunami, 2004. Available from:http://www.science-daily.com/releases/2004/05/040521073051.htm.

[79] Bonnard JM, Salvetat JP, Stockli T, de Herr WA, Forro L,Chatelain A. Appl Phys Lett 1998;73(918):396.

[80] Rinzler AG, Hafner JH, Nikolaev P, Lou L, Kim SG, Tomanek D,et al. Science 1995;269:15503.

[81] De Heer WA, Chatelain A, Ugarte D. Science 1995;270:117980.[82] de Heer WA, Chatelain A, Ugarte D. Science 1995;270.[83] Lee NS et al. Diam Relat Mater 2001;10:265.[84] Saito Y, Uemura S. Carbon 2000;38:169.[85] Rosen R et al. Appl Phys Lett 2000;76:1668.[86] Sugie H et al. Appl Phys Lett 2001;78:2578.[87] Zhou O. personal communication.[88] Rinzler AG et al. Science 1995;269:1550.[89] Wang QH, Setlur AA, Lauerhaas JM, Dai JY, Seelig EW, Chang

RH. Appl Phys Lett 1998;72(2912):399400.[90] McCreery RL. Bard AJ, editor. Electroanalytical Chemistry, vol. 17.

New York: Marcel Dekker; 1991. p. 401.[91] Nugent J, Santhanam KSV, Rubio A, Ajayan PM. J Phys Chem 401

[in press].[92] Dillon AC, Jones KM, Bekkedahl TA, Kiang CH, Bethune DS,

Heben MJ. Nature 1997;386(377):4045.

[93] Chambers A, Park C, Baker RTK, Rodriguez NM. J Phys Chem B1998;102(4253):404.[94] Pederson, Broughton M. J Phys Rev Lett 1992;69(2689):405.[95] Gao B et al. Chem Phys Lett 1999;307:153.[96] Iijima S. Nature 1991;354(56):391.[97] Fuhrer MS et al. Science 2000;288:494.[98] Rueckes T et al. Science 2000;289:94.[99] Bachtold A, Hadley P, Nakanishi T, Dekker C. Science

2001;294:1317. published online 4 October 2001 (10.1126/science.1065824).

[100] Franklin NR, Li Y, Chen RJ, Javey A, Dai H. Appl Phys Lett2001;79:4571.

M. Paradise, T. Goswami / Materials and Design 28 (2007) 14771489 1489