synthesis of carbon nanotubes

7/23/2019 synthesis of carbon nanotubes

1/16

Arc discharge synthesis of carbon nanotubes: Comprehensive review

Neha Arora , N.N. SharmaDepartment of Mechanical Engineering, Birla Institute of Technology and Science, Pilani, India

a b s t r a c ta r t i c l e i n f o

Article history:

Received 2 June 2014Received in revised form 24 September 2014Accepted 3 October 2014

Available online 13 October 2014

Keywords:

Arc dischargeCarbon nanotube synthesis

In quest to synthesize high quality carbon nanotubes in bulk, different routes have been proposed andestablished over the last two decades. Arc discharge is the oldest and among the best techniques to producehigh quality carbon nanotubes. Though this synthesis technique has been exploredfor a long time, the nanotubegrowth mechanism is still unclear and the growth conditions lack strong correlation with the synthesized prod-uct. In this review, we attempt to present themechanism ofnanotube growth in arc dischargeand the factors af-fecting itsformation.In order to understandthe physicsof thismechanism,the effect of experimental parameterssuch assetupmodication,power supply, arc current, catalyst, pressure, grain size, electrode geometryand tem-peratureon sizeand yield of the nanotubes has beendetailed. The variation in synthesis parameters employed inliterature has been presented along with challenges and gaps that persist in the technique.

2014 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1362. Nanotube growth in arc discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

2.1. Arc discharge setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1362.2 . Growth mechanism of carbon nanotubes in arc discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

2.2.1. Vapour phase growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1372.2.2. Liquid phase growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1372.2.3. Solid phase growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

3. Physics of nanotube growth in arc discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1373.1. Effect of power supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

3.1.1. Type of power supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1373.1.2. Effect of voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1393.1.3. Effect of arc current . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1393.1.4. Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

3.2. Carbon precursors in arc discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1393.3. Effect of grain size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1393.4. Role of catalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1403.5. Role of atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1413.6. Effect of pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1413.7. Role of temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

3.8. Effect of setup modication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1423.9. Effect of cathode shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1423.10. Size and yield of CNT in arc discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143Appendix A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143Appendix B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Diamond & Related Materials 50 (2014) 135150

Corresponding author.E-mail address: [email protected](N. Arora).

http://dx.doi.org/10.1016/j.diamond.2014.10.001

0925-9635/ 2014 Elsevier B.V. All rights reserved.

Contents lists available atScienceDirect

Diamond & Related Materials

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / d i a m o n d
http://dx.doi.org/10.1016/j.diamond.2014.10.001http://dx.doi.org/10.1016/j.diamond.2014.10.001http://dx.doi.org/10.1016/j.diamond.2014.10.001mailto:[email protected]://dx.doi.org/10.1016/j.diamond.2014.10.001http://www.sciencedirect.com/science/journal/09259635http://www.elsevier.com/locate/diamondhttp://www.elsevier.com/locate/diamondhttp://www.sciencedirect.com/science/journal/09259635http://localhost/var/www/apps/conversion/tmp/scratch_5/Unlabelled%20imagehttp://dx.doi.org/10.1016/j.diamond.2014.10.001http://localhost/var/www/apps/conversion/tmp/scratch_5/Unlabelled%20imagemailto:[email protected]://dx.doi.org/10.1016/j.diamond.2014.10.001http://crossmark.crossref.org/dialog/?doi=10.1016/j.diamond.2014.10.001&domain=pdf


2/16

1. Introduction

Carbon nanotubes (CNTs) possess excellent mechanical, electronic,thermal, optical and chemical properties which have revolutionized thestate-of-the-art in nanotechnology. CNTs have been broadly classiedas Single-Walled Nanotubes (SWNTs), Double-Walled Nanotubes(DWNTs) and Multi-walled Nanotubes (MWNTs). Researchers have de-vised different routes to synthesizeCNTs fromvariouscarbon precursors.

The most popular and widely used nanotube synthesis techniques are:Arc Discharge, Chemical Vapour Deposition and Laser Ablation[116].Apart from these methods, Hydro-thermal synthesis[17], Electrolysis[18,19], and Ball milling[20,21]methods have also been used.

CNTs were rst synthesized by Iijima[22]in 1991 using arc dis-charge method. Over the last two decades, researchers have demon-strated a successful use of this technique in the production of highquality CNTs. A bar graph shown inFig. A.1(a) (Appendix A) depictsthe number of papers published year wise on CNT synthesis using arcdischarge. We gained some insights of nanotube growth process andunderstood the fundamental role of few arc discharge parameters.However, the literature lacks comprehensive study on the mechanismof nanotube formation and needs strong correlation between growthconditions and synthesized nanotubes. The requiem to understand therole of growth conditions lays in selective growth of nanotubes usingarc discharge, which remains largely unexplored.

For researchers taking up the nanotube synthesis using arc dis-charge, the availability of comprehensive reviewon the process is quin-tessential. Previously, Ando and Zhao[7]discussed the synthesis ofSWNTs andMWNTs using arc discharge. In 2007, Harris [8] investigatedthe models of nanotube growth in arc discharge and laser ablation pro-cesses. In 2010, Ando[10]presented the chronological aspect of arcgrown nanotubes in hydrogen atmosphere. In 2011, Tessonnier andSu [12] briey discussed thenucleation andgrowthprocess of nanotubein arc grown CNTs. In 2011, Prasek et al. [13]reviewed the differentroutes of nanotube synthesis. In 2012, Journet et al.[16]discussed lowor medium temperature techniques to synthesize CNTs.

The present review updates and details on experimental attempts tomanufacture carbon nanotubes using this technique. The growth mecha-

nism of nanotubes, as explored in literature has been briefed. It furtherdiscusses the role of synthesis parameters like setup modication,power supplies,arc current, catalyst, pressure,grain size, electrodegeom-etry and temperature on the nanotube production. Few parameters likepressure and catalyst have been investigated quantitatively in literaturebut the exact role of the growth conditions still requires extensive inves-tigation. The review has been divided into four sections. Section 2

discusses the nanotube growth in arc discharge.Section 3outlines thephysics of CNT growth in arc discharge process and the role of growthconditions on nanotube formation, which is followed by the conclusion.

2. Nanotube growth in arc discharge

2.1. Arc discharge setup

Arc discharge is the electrical breakdown of a gas to generate plasma.This old technique of generating arc using electric current was rst usedto produce CNTs by Iijima[22]in 1991. A schematic of an arc dischargechamber is shown inFig. 1. The chamber consists of two electrodeswhich are mounted horizontally or vertically; one of which (anode) islled with powdered carbon precursor along with the catalyst and theother electrode (cathode) is usually a pure graphite rod. The chamber islled with a gas or submerged inside a liquid environment. Afterswitching on the power supply (AC or DC), the electrodes are broughtin contact to generate an arc and are kept at an intermittent gap of12mm to attain a steady discharge. A constant current is maintainedthrough the electrodes to obtain a non-uctuating arc for which closedloop automation is employed to adjust the gap automatically. A uctuat-ingarcresultsin unstable plasma andthequalityof thesynthesized prod-uct is affected. Thearc current generatesplasmaof very high temperature~40006000 K, which sublimes the carbon precursorlled inside theanode. The carbon vapours aggregate in the gas phase and drift towardsthe cathode where it cools down due to the temperature gradient. Afteran arc application time of few minutes the discharge is stopped and ca-thodic deposit which contains CNTs along with the soot is collectedfrom the walls of chamber. The deposit is further puried and observedunder an electron microscope to investigate their morphology.

2.2. Growth mechanism of carbon nanotubes in arc discharge

Over the last two decades, researchers have investigated and sug-gested growth conditions for nanotube formation based on their exper-imental observations. Despite seminal studies, no clear understandingof the growth mechanism has been developed and a critical study lies

in understanding the physics of this mechanism which certainly helpsin predicting the optimum growth conditions of nanotubes. In this sec-tion,we present an outline of thesynthesis mechanism in thegrowth ofCNTs in arc discharge.

The two electrodes are brought in contact and upon application ofvoltage, constant currentows through them. The electric current re-sults in resistive heating and raises the temperature of the electrodes.

Glass window

Cathode Anode

Deposition

Precursor +

Catalyst

Closed Chamber

Base

Gas Inlet Gas Outlet

Power Supply

Fig. 1.Schematic of an arc discharge setup.

136 N. Arora, N.N. Sharma / Diamond & Related Materials 50 (2014) 135150
http://-/?-http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B1%80http://-/?-


3/16

The anode is moved back to maintain a desired gap (~1 mm) betweentheelectrodes forcontinuous deposition of carbon vapours. Meanwhile,the high temperature ~40006000 K facilitates the breakdown of thesurrounding gaslled insidethechamber.Thegas ionizes into electronsand ions and results in hot plasma formation between the electrodes.Stable plasma grows uniformly over the electrode surface correspond-ing to stable anode and cathode voltage. The collisions of ions and elec-trons in the plasma emit photons responsible for the glow in plasma.

The electrons are ejected from cathode hit the anode at high velocityand sputter the carbon precursorlled at the centre of the anode. Thehightemperatureresultingfrom resistive heating results in sublimationof carbon precursor and converts them into carbon vapours. Thecarbonvapours are decomposed in carbon ions. The decomposition occurs dueto high heatux or thermal energy of the plasma. The carbon vapoursaggregateto form viscous carbonclusters anddrift towards thecathode,which is cooler as compared to anode. The carbon vapours undergo aphase change and get converted into liquid carbon. The temperaturegradient at cathode and quenching effect of atmosphere solidies andcrystallizes the liquid carbon to form cylindrical deposits that growsteadily on cathode. The cathodic deposit is composed of a grey outershell and a dark inner core. The grey deposit consists of the rolls ofgraphene sheetsknownas carbonnanotubes. Theaddition of hexagonalcarbonatom clusters lead to thegrowthof nanotube.However, instabil-ity in plasma leads to capping of nanotubes. The diameter of the nano-tubes is governed by the density of carbon vapours in the plasma.Variation in temperature gradients strongly affects the diameter distri-bution of CNTs produced and nanotubes are formed in bundles due tovan der Waals interaction.

In reported literature, mechanism of nanotube synthesis has beenexplained by few researchers[2342] but the exact growth mechanismis still debatable due to different theories of nanotube growth in vapourphase[34], liquid phase[33], solid phase[35]and crystal phase[43].

2.2.1. Vapour phase growth

Gamaly and Ebbesen[34]detailed the vapour phase growth ofMWNTs and suggested that the carbon vapours condense and nucleate

to form nanotubes. They proposed a model for the velocity distributionof carbon vapours and divided the carbon vapours in two groups de-pending upon their velocity distributions. One group of carbon vapourswill have the isotropic (Maxwellian) velocity distribution. The othergroup of carbon particles has higher velocities than the rst groupwhich is due to the acceleration of carbon vapours between the elec-trodes. According to Gamaly and Ebbesen, the nanotube growth occursin three steps seed formation, tube growth and termination. Theseeds are formed as a result of the two velocity distributions. In the be-ginning, the carbon vapours possess Maxwellian distributions whichre-sult in nanoparticle formation. Upon increasing the current, the othergroup of directed carbon vapours results in open structures or seeds.Once thecurrent reaches a stablevalue, thecarbonionsow perpendic-ular to thecathode surface resultingin the nanotube growth. Finally the

nanotube growthis terminated dueto instabilities in theplasma.A sim-ilar theory has also been validated by other researchers[4447].

2.2.2. Liquid phase growth

The liquid phase growth model was suggested by Heer [33] in 2005.He proposed that the liquid carbon solidies to result in nanotubes.They found beads of amorphouscarbonwithinthe nanotubes which re-sult during solidication of liquid drops. According to them, when theelectric arc is applied to heat the carbon precursor, the electronsbombarding the surface result in localized heating, which causes thesurface to liquefy. The liquid droplets are ejected from the anode andare drifted towards cathode. These globules of liquid carbon tend tocool at the cathode surface. The cooling propagates from outer layer to-

wards the centre, resulting in multilayered tubular structures.

2.2.3. Solid phase growth

Solid phase growth of nanotubes was proposed by Harris et al.[48]in 1994. They found that nanotubes are synthesized by high tempera-ture heat treatment of fullerene soot. Based on their observations,they proposed that initially carbon vapours in the gas phase condenseonto the cathode to form fullerene soot. Since the temperature of cath-ode is high due to continuous arc, the fullerene soot is converted toMWNTs via the seed-growth-termination process. The requiem for

this process is rapid heating of fullerene soot since slow heating of ful-lerene results in nanoparticle formation[35]. In order to improveupon the existing theories, experimental investigations are necessaryto develop a strong correlation between the synthesized product andinput parameters. In literature, large number of experimental reportshas been published on arc discharge synthesis of nanotubes which hasbeen discussed in the next section.

3. Physics of nanotube growth in arc discharge

The arc discharge process is dependent on several parameters liketypeof power supply, environment, pressure, electrode geometry, cata-lyst and temperature that inuence the quality and quantity of the syn-thesized product. Researchers have studied the variation of theseparameters and have attempted to establish an optimal range forimproving quality and quantity of CNTs using arc discharge. In orderto develop the physics of nanotube growth in arc discharge, we rstneed to understand the role of various synthesis parameters on theproduct. The implication of these parameters and their correspondingphysics is detailed in the next subsections respectively.

3.1. Effect of power supply

The power supply controls the arc current and voltage which governtheenergydistribution of electron discharge. This affects theplasmatem-perature that plays decisive role in the product of arc discharge. In thissection we have detailed the affect of type of power supply, voltage, arccurrent and frequency on quality and yield of synthesized nanotubes.

3.1.1. Type of power supplyThe type of power supply plays a vital role in nanotube formation.

When DC arc current is applied across the electrodes, as shown inFig. 2(a), continuous emission of electrons from cathode bombards theanode at high velocity. The cathode diameter is usually larger than theanode diameter, which results in lower current density of cathode thananode. Thus, a high temperature gradient across the electrodes is ob-served which sublimes the carbon precursorlled inside the anode andresults intheformation of CNTs. In ACarc discharge, the current alters pe-riodically across the electrodes and no deposition is observed at eitherside of the electrodes. The carbon vapours y out of the plasma due tothermal effects which results in CNT formation at the walls of the cham-ber as shown inFig. 2(b). In pulsed arc method as shown inFig. 2(c),CNT formation takes place with some short range and long range pulses,

normally of the millisecond width. Pulse strikes the anode surface and itbegins to vaporize and deposition takes place on the cathode.

In case of DC supply, theionized gasdrifts towards cathode and hin-ders thecontinuous deposition of carbonions on cathode. In order to re-move the gas ions, the cathode should not be negatively charged. Thisproblem may be overcome by using an AC or pulsed power supply.The disadvantage of using AC supply is that the formation of nanotubesoccurs only in positive cycle and thus the yield is reduced. Thus, pulsedarc discharge is the most favourable for nanotube formation.

Ashraf et al. [49] compared thecontinuous andpulsedarc dischargesand their effect on nanotube formation.They observed variation in typeand quantity of atomic carbon species on using different power supply.According to them, the continuous arc discharge results in higheramount of C2 which is theprecursor fornanotube formation. Thepulsed

discharge contained additional amount of C1 due to disintegrationof C2,

137N. Arora, N.N. Sharma / Diamond & Related Materials 50 (2014) 135150


4/16

which does not form closed carbon structures. However, pulsed arc isadvantageous as it provides more control over the process.

The pulsed arc is found to be more energetic than a continuous arc.The energy of an electron is higher at the time of ejection and it takesmicroseconds for the electron to attain the Maxwellian energy distribu-tion. Thus, continuous bombardment of anode with pulses of widthshorter than this timeresults in higherelectron energy which increasesthe yield[50]. On contrary to this, continuous power supply has lesser

electron energy distribution. We have categorized the existing litera-ture based on the type of power supply used in the synthesis of CNTs.Pie charts showing the percentage of papers published in these catego-ries are shown inFig. A.2(Appendix A). It can be clearly seen fromFig. A.2(a) that the DC power supply is most widely used to synthesizeCNT using arc discharge method.

The rst attempts to synthesize MWNTs[22]and SWNTs[51,52]using DC power supply were reported in 1991 and 1993 respectively.Some of the early insightful works include large scale synthesis ofCNTs in helium atmosphere by Ebbesen and Ajayan[53]in 1992. In1995, Wang et al. [54]produced MWNTs in hydrogen atmosphereusing arc discharge. In 1997, Journet et al.[55]demonstrated largescale synthesis of SWNTs in helium atmosphere. Shi et al.[56]in 1999demonstrated bulk synthesis of SWNTs in high pressure helium atmo-

sphere. High yields of 6.5 g/h CNTs have been demonstrated by Zhao

[57]. In 2004, Itkis et al.[58]produced 515 g CNTs using DC arc dis-charge. Chronological attempts using DC power supply have been listedinTable B.1(Appendix B).

Researchers have also used AC power supply of constant frequencyto generate arc between the electrodes. The advantage of using AC arcdischarge is the formation of deposits on the wall of chamber. Veryfew experimental attempts have been made to synthesize CNTs usingAC arc discharge. In 1992, Ebbesen and Ajayan[53]reported CNTs

using AC arc discharge. In 1997, Zeng et al. [59]reported MWNTs at700 A arc current. In 1999, Ohkohchi[60] rst reported SWNTs usingAC arc discharge. In 2007, Matsuura et al.[61]reported poor quality ofCNT synthesized by AC arc discharge. CNT yield in AC is found to besmaller than that of DC as it is produced during one part of cycle[53].Chronological attempts using AC power supply have been listed inTable B.2(Appendix B).

Few researchers have synthesized carbon nanotubes using pulsedDC power supply. Parkansky et al. [62]suggested that the pulsed arc isadvantageous over other power supply as no pressure chamber is re-quired and can be performed in open air. Yield is found to be increasedin case of pulsed as compared to DC supply[63,64]. Sugai etal. [65] havefound yield of SWNT increases as pulse width increases. Cathode is alsofound to be consumed by pulsed arc method[66]. Oneobservation from

pulsed discharge is that the CNT formation is reported at low current

Fig. 2.Schematics showing the formation of CNTs using different power supplies.

http://-/?-http://-/?-http://-/?-http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B2%80http://-/?-http://-/?-http://-/?-


5/16

levels. Chronological attempts using pulsed power supply have beenlisted inTable B.3(Appendix B).

3.1.2. Effect of voltage

In arc discharge, voltage is applied across the electrodes to generateelectrical breakdown of the dielectric gas. The voltage across the elec-trodes for CNT synthesis ranges between 15 and 30 V. This voltagemust be kept constant for stable plasma. The sudden change in arc volt-

age results in the formation of bamboo like structures as observed by[67]. Scalese et al.[68]suggested that the major effect of voltage varia-tion wasfoundto be thespatial distribution andsharpness of CNTpillarsin the deposit. Jahanshahi et al.[69]observed fullerenes at 10 V andobserved CNTs upon increasing the voltage to 30 V. Antisari et al. [70]observed deposits of amorphous carbon with few nanotubes for volt-ages less than 22 V and found no cathodic deposits for voltages higherthan 27 V. They have also reported an increase in quality of nanotubeswhen the voltage is increased from 15 V to 20 V. However synthesis athigh voltages (~700 V) has also been reported[25,61,71-74].

Besides voltage, the synthesis time has also been investigated in lit-erature forCNT formation. Generally, thevoltage acrosstheelectrodes ismaintained for 30120 s. However CNTs have been synthesized at timeranging from milliseconds[75,76,77]to high synthesis time of 15 min[78,79], 20 min[80,81,82], 40 min[83], 30150 min[84], 60 min[85],6090 min[58], 12 h[86], and 330 h arc run time[65]. This is an in-teresting parameter and raises a question of impact of arc applicationtime in the formation of carbon nanotubes.

3.1.3. Effect of arc current

One of the signicant parameters in arc discharge method is the arccurrent. It affects the quality, yield and size of synthesized nanotubes. Inan arc discharge apparatus, arc current results in emission of electronsfrom cathode which travel at high velocity towards anode. These elec-trons hit theanode surface which results in sputteringof carbonprecur-sors. Upon increasing the arc current, the number of electrons strikingthe anode surface increases, thereby sputtering more carbon precursorfrom the anode. The applied current produces resistive heating whichresults in high temperature. Due to high temperature, the carbon pre-

cursor lled in anode sublimes to form carbon vapours and nucleatesat cathode resulting in nanotube formation. There also exists a mini-mum discharge current called as chopping current at which the plasmaproduction is insufcient and leads to arc extinction.

During the synthesis, non-uctuating current of an order of 50100A is maintainedto ablatetheanode. Ebbesen andGamaly [34] suggestedthat the current density for synthesis of CNTs should be around 165195 A/cm2. Cadek et al.[87]studied the dependence of arc current onthe yield of nanotubesand concluded that theyield increasesas currentdensity is varied from 165 A/cm2 to 195 A/cm2. Nishizaka et al.[88]found the optimal current density for SWNT formation as 250270 A/cm2. Matsuura et al. [89]observed better yieldsof SWNT at a current den-sity of ~450 A/cm2. In literature, vast variations have been reported in arccurrent,from 2.5A to 900A, which creates an ambiguityof theoptimum

arc current rating. A comparative view of thecurrent ratings used in lit-erature has beenshown above in Fig. A.1(b) (Appendix A). Wehave notincluded the experiments where ranges of current values were used. Itcan be observed from the graph that majority of the experiments havebeen conducted between the range of 50 and 100 A.

Effect of variation of current has been investigated in the literature[87,88,89]. It has been observed that at currents below 30 A, the arc isunstable and lower yields are obtained. The anomalies in the currentlevels corresponding to CNT formation are noteworthy. Okada et al.[90]have reported a low current synthesis of 120 A whereas DelongHe et al.[91]suggested that the nanotube formation takes place at cur-rent levels beyond 100 A. Takizawa et al. [92]found the highest nano-tube yield at 100 A. Researchers have produced CNTs at low arccurrent of 210 A[77,78,93,94,95]. However, high arc currents of

700 A[59]and 900 A[96]have also been employed to produce CNTs.

It has been observed that as current increases, the yield increases[74,89,97]however increasing the arc current does not improve the struc-ture of CNT[68]. He et al.[91]observed that the increase in arc currentreduces the amount of SWNTs and favours the formation of amorphouscarbon particles. Zhao et al.[98]studied the variation of pressure alongwith the variation in arc current. Tang et al. [99]investigated reductionin thearc current by changing theshape of cathode. Lange et al. [75] ob-servedthat a carbonprecursorwith smaller grain size requires lesserarc

current fornanotube formation. However, it raisesan interesting debateon the optimum current requirements for nanotube formation. Thus,there is a critical need to develop a strong correlation betweenoptimum current levels and formation of nanotubes.

3.1.4. Frequency

Thefrequency of thepower supplyhas been found to affectthequal-ity of thedeposits produced in arc discharge. First reported by Ohkohchiet al.[60]in 1999, AC arc discharge has been found to produce highquality SWNTs. In literature, the effect of variation in frequency hasnot been explored. Kia and Bonabi[78]found that at low frequenciesof 50 Hz, better growth is observed at anode, however at high frequen-cies of 400 Hz, the cathode soot and production rate increase. This canbe an interesting area for investigation and may lead to value additionto existing growth mechanism of CNTs.

3.2. Carbon precursors in arc discharge

The synthesis of CNTs is carried out by sublimation of a carbonprecursor using an external power source; in case of arc discharge, theapplied current ablates the carbon precursor to form carbon vapours,which nucleates to form a nanotube. A pie chart in Fig. A.2(b) showsthe percentage of papers published on CNT synthesis usingvarious car-bon precursors. It can be observed from the gure that most of the arcdischarge synthesis of CNTs has been carried out using graphite as pre-cursor. It is an excellent conductor of heat and electricity and is com-mercially available in high purity. When graphite is subjected to highcurrent, the lattice structure breaks and releases C1 or C2 carbon speciesthat vaporize due to high temperature.

Some groups have used carbon black as a raw material to produceCNTs[31,88,100,101,102,103]. Carbon black is amorphous in nature,easily available material on earthand can be a potential precursor to syn-thesizeCNT due to techno-economicreasons. The synthesis of CNTs fromcarbon black will requirelarger current value thangraphite. Carbon blackupon application of current gets rst converted to graphite and thenforms tubular structures of CNTs. Lange et al. [75]have used carbonblack as starting material to synthesize SWNT. Chen et al.[104]haveutilized carbon black as a dot carbon source for the production of DWNT.

Some researchers have used coal as a starting material to synthesizeCNTs. Coal is a mixture of aromatic and aliphatic hydrocarbon molecules,which arehighly reactive in nature. Qiuet al. [105] have discussed coal asan ideal starting material for large synthesis of DWNTs. When high arccurrent passes through coal, the weak linkages between the polymeric

aryl structures getbrokendown intoalkyneand aromatic species that fur-ther form DWNTs. Coal contains sulphur, which favours the growth ofDWNTs and affects the diameter distribution of the nanotubes produced.

Other carbon sources like fullerenewaste soot [106], toluene [90], tirepowder[107], poly-vinyl-alcohol[108],MWNT/carbon nanobers[109]and other hydrocarbons[32,86]have been utilized to produce CNTs byarc discharge method. Whereas graphite is the mostly investigated pre-cursor, the other precursor materials need further understanding andinvestigations.

3.3. Effect of grain size

One of thelessexplored aspects of arc discharge is theeffect of grainsize on current requirement and nanotube production. It seems logical

that a smaller particle requires lesser energy to vaporize. Solid state

http://-/?-http://-/?-


6/16

properties namely cohesive energy of cathode and anode play vital rolein the quality of the nanotubes formed. The cohesive energy[110]isfound to depend upon the size of nanoparticle[111].

Lange et al.[75]used carbon precursors with different grain sizesand found that the current requirement reduces drastically by changingthe grain size. Labedz et al.[103]investigated the effect of particle sizeand the carbon density of the electrode. They found that the anodewith smallest particle size and more electrode density yields lesser

CNTs and more carbonaceous particles. Also, the particle size mayaffectthe diameter of the synthesized nanotubes. It was observed that largergrain size results in smaller diameter CNT [103]. This may be a potentialarea for investigationand validations are required to develop an associ-ation with formation of nanotubes.

3.4. Role of catalyst

The catalyst used in arc discharge synthesis of CNTs is usually ametal, which is powdered and lled in either side of the electrodealong with a carbon precursor. The metal should have low boiling tem-perature and high evaporation rate to act as a good catalyst in nanotubeformation[23]. On application of arc current, the metal atoms vaporizealong with the carbon precursor. The metal particles agglomerate withcarbonvapours andnucleate at arc reactor walls.There is no establishedreason for deposition of SWNTs on walls of chamber instead of cathodeand is an open area for investigation. A probable reason for this

behaviour may be that the carbon vapours along with the liquid metalmove towards cathode in the plasma. Since metal particles do notstick to the surface of the cathode, theyy away due to their momen-tumand form SWNTsat thewalls.A graphicalrepresentationof thesug-gested mechanism is shown inFig. 3.

The catalyst favours the growth of SWNTs than MWNTs. Lin et al.[112]experimentally conrmed that the presence of metal catalyst ingas phase and on the surface of the cathode alters the temperature dis-

tribution and prevents the growth of MWNTs on cathode. The mostcommonly used catalysts are Ni, Fe, and Co along with Y, S, and Cradded as a promoter. Mostly, SWNTs are observed in the presence ofcatalyst, whereas MWNTs are produced in its absence. However somereports on MWNT formation in the presence of catalyst have been re-ported in literature [23,69,76,108,113,114,115,116]. The interesting ob-servation among these reports is that MWNT synthesis occurs upon theuse a binary catalyst. Zhao et al.[117]reported that no CNT formationoccurs without the use of catalyst.

Few researchers have investigated the role of various catalysts on thequality of the CNTs formed. Among themetals, Ni and Fe have been mostwidely used catalyst for the growth of high quality CNTs. Both Ni and Feincrease the yield and quality of CNTs. Nickel is found to produce morecrystalline nanotubes. Ni is mixed with elements like Y, Mo, Fe, Co, Cr toimprove the synthesized product. Fe is generally added with sulphur orW to facilitate growth of SWNTs. The drawback of using Fe as catalyst isthe formation of its oxides (Fe2O3) which doesn't act as catalyst and

Fig. 3.Role of catalyst in growth of CNTs.

http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E0%B3%80


7/16

retards the growth of SWNTs. Li et al. [118]suggested that the Fe pro-motes the length of the nanotubes. Zhao et al. [81] suggestedthat the ad-dition of sulphur to Fe catalyst results in the development of a core/shellat the cathode due to different melting points. This core/shell promotesthe growthof DWNTs.Zhaoetal. [117] observed a more uniform distribu-tion of ACNT diameterupon using FeS catalyst as comparedto Ni/Coalloy.Wang et al.[119] suggested that sulphur plays a major role in the forma-tion of branched CNTs. It forms an active site on the surface of thecatalyst

where nucleation of carbon vapours takes place. Ando et al. [120]ob-served that sulphur promotes the growth of CNTs by removal of the ter-minating species at the nanotube ends. The catalyst may also controlthe wall number of synthesized CNTs. Yang et al.[121] observed a controlon wall number of CNTs on addition of KCl to NiCo mixture. Montoroet al. [80] observed that VO groups also promote nucleation of carbon va-pours in the synthesis of CNTs. Qiu et al. [122] observed that KClalso pro-motes the formation of DWNTs.

The size of nanotubescan be readily controlled by using appropriatesize of metal catalyst particles. Thecarbonvapours move along with themetal particles and nucleate on these to form nanotubes. Thus, the sizeof metal particles plays a vital role in nanotube diameter distribution.The concentration of metal particles is found to affect the yield of nano-tubes produced. The composition of catalyst also plays a major role inthe existence of CNTs. Wang et al. [108]reported that the maximumconcentration of iron particles in the anode composition must not ex-ceed 10% for CNT formation. One reason for this maybe that larger con-centration of catalyst particles restricts the motion of carbon vapourstowards cathode. Keidar [27] suggested that SWNT synthesis isgoverned by catalyst carbon phasediagram. The effect of variouscat-alysts on CNT formation as reported in literature is tabulated in Table 1.

3.5. Role of atmosphere

The atmosphere plays an important role in thermo-ionic effects, plas-ma formation,andprovides thethermal growth of CNTandthenecessaryannealing effects. When the arc current ows through the electrodes, thegas gets ionized due to high temperatures and plasma is formed. The ion-ized gas acts as a highly conducting medium which provides transfer of

mass on either side. The ions in the plasma are thermally agitated andprovide thenecessaryenergyfor movement of carbonvapours. Theatmo-sphere also plays a signicant role in the thermal growth of CNTs. The

thermal conductivity of the atmosphere provides annealing effects atthe cathode which is essential for nanotube formation. The atmospherealso regulates the temperature of plasma depending on its ionization po-tential. Thegas with higher ionizationpotential will requirelarge arc cur-rent for breakdown. Thus, the choice of atmosphere depends on itsionization potential and thermal and electrical conductivity.

The arc chamber is pressurized with a gas like nitrogen, hydrogen,helium or argon or immersed in a liquid environment. Hydrogen has

the highest thermal conductivity and is regarded as the most ef

cientquencher in nanotube growth. Zhao and Ando[131] suggested that hy-drogen promotes the growth of CNTs and reduces the carbonaceousmaterials by forming hydrocarbons with them. Li et al.[132]foundthat hydrogen results in a cleaner CNT surface as it selectively etchesthe amorphous carbon impurities. However, Zhao[133]suggestedthat pure hydrogen is unfavourable for mass production of SWNT dueto the instability of arc discharge plasma. Tang et al. [134]observedthat rapid introduction of hydrogen prevents the ends of nanotubefrom closing. Due to this problem, hydrogen is generally mixed with anoble gas like argon or helium to stabilize the plasma. Liu et al.[135]found that changing of gas from hydrogen to helium promotes thegrowth of SWNT. Shi et al. [129]observed that helium atmospherestrongly affects the yield of SWNTs. Farhat et al. [28]controlled the di-ameter of nanotubes by changing the composition of argonheliumgas ratio. Su et al.[136]investigated the effect of CO concentration onthe amount of impurities present in SWNTs. They suggested thatlower concentration of CO should be preferred for better SWNT yield.Ando et al. [79] observed theincrease in yield uponaddition of nitrogento hydrogen when compared to H2Ar mixture. TheH2N2hasmoreen-thalpy than H2Ar which promotes the evaporation of CFe mixture.

Apart from the gaseous atmosphere, liquid environments have alsobeen used to produce CNTs. In case of liquid environments the CNTsare found to be oating on the surface of the liquid. The most popularchoice of the liquid environment in the literature is deionised water.Few works[116,126]have synthesized CNTs in NaCl solution owing toits good electrical conductivity, better cooling ability than deionisedwater, cheaper than liquid nitrogen and helium. NaCl dissociates intoNa+ andCl ions which provide better ionic conductivity to theplasma.

The concentration of Na+ ions is a critical parameter for better yields.The excess of Na+ ions hinders the locomotion of carbon vapoursfrom anode and reduces the yield of the SWNTs[126]. Other solutionslike LiCl[69], H3PO4aqueous solution[80], and beer froth[137]havebeen used as atmosphere in synthesizing CNTs using arc discharge.Kim et al.[137]suggested that the agglomerating effect of both beerfroth and carbon nanotubes helps in obtaining a cleaner product. TheCNTs stick to beer froth and the other carbonaceous materials get sepa-rated out. The effects of different atmosphere in arc discharge chamberon yield, quality and size of nanotubes are listed inTable 2.

3.6. Effect of pressure

In arc discharge method, the role of pressure is to provide energy to

the gas molecules and to act as a wall for a steady ow of ions betweenthe electrodes. The optimum range of pressure for CNT synthesis hasbeen found ranging from 300 to 700 Torr. At pressure below 300 Torr,the yield is found to be very low as the density of ions is low resultingin unstable plasma. Whereas at high pressure, more number of ionsparticipate intheplasmathereby restricting themotion of carbonvapoursfrom anodeto cathodeand decreases the yield. That is why high pressureis unfavourable for the synthesis of CNTs and very less number of re-searchers has used pressure beyond 700 Torr as shown inFig. A.1(c).

In literature the effects of pressure on yield and quality of synthesizednanotubes have been vastlyexplored. Cuiet al. [142] found thatthenano-tubeyield decreasesfor lowandhigh pressures of nitrogen. Grebenyukovet al.[141]found that the efcient pressure for nitrogen is 350 Torr. Incase of air, 4590 Torr has been found to be the optimum pressure

[143]. However, yield is reduced upon increasing the air pressure from

Table 1

Effect of catalyst on CNT formation.

Basecatalyst

Mixtureused

Comments

Ni Ni Promotes the growth of SWNT[23]NiY Y alone cannot synthesize CNT formation[58]

Addition of Y to Ni reduces the yield[123]Y promotes the growth of SWNT[55]

NiMo Mo does not affect CNT formation[90]NiMoFe MoFe favours CNT formation[90]

Ni

Co Results in SWNT formation[121]NiCo + KCl Results in DWNT formation[121]NiCr Results i n MWNT formation[113]NiHo Increases yield and purity[60]NiCoFeS Diameter of SWNT increases with sulphur[124,125]

Fe Fe Affects yield and diameter of MWNT[108]FeFeS Addition of sulphur promotes growth of DWNT and

increases the yield[81]FeFe2O3 Retards the growth of CNT[126]FeFe3O4 Increases the yield of SWNT[127]FeW Tungsten reduces the diameter and increases

yield of SWNT[128]FeFe(C5H5)2 Improves yield of SWNT[115]

Ca Ca Produces smaller diameter CNT than Ni[129,56]RhPt RhPt Increases the yield[130]Co Co Superior catalyst for DWNT production[109]

Increases the mean diameter of DWNT[66]CoCoS Addition of sulphur promotes CNT production[104]



8/16

40to 300 Torr[82]. Kim and Kim [97] observed MWNTsin airand heliumat 300 Torr and 500 Torr respectively. The pressure versus yield trend forhelium has been studied by few groups[28,73,129,144]and pressureranging from500 to 700 Torr hasbeen found to beoptimum. Howeverre-versal in trend has been shown by Park et al. in2002 [124] who observeda decrease in yield on increasing thepressure from 100 to 500 Torr. Also,Shi et al. [56]reported high yield of 2.5 g/h for helium pressure of1520 Torr. Optimum hydrogen pressure is found to be 500 Torr [57].The yield increases from 100 to 700 Torr but decreases beyond 700 Torr[122]. Farhat et al.[28]have reported that the optimum argon pressureis100 Torr. It may beconcluded from the reports inliteraturethatthe op-timum pressure condition for high yield of CNTs has been found to be~500 Torr. However, some variations that have been reported need tobe investigated for better understanding of the growth mechanism.

3.7. Role of temperature

The temperature ionizes the gas and forms the plasma. It simulta-neously sublimes the carbon precursor and provides thermal ux fordecomposition of carbon vapours into ions. It is responsible forthermo-ionic emission of carbon ions andnally helps in nucleation ofcarbon vapours at cathodeto form CNTs. Temperature variation in plas-masignicantlyaffects thequality andsize of theCNTs.Temperaturere-quired for growth and nucleation in arc discharge synthesis is achievedby the electric arc. The temperature increases as the current density isincreased, thereby subliming the anode at a faster rate. Higher temper-ature results in more crystalline CNTs. The temperature of the plasma isregulated by the thermal properties of atmosphere. Hydrogen plasmagenerates higher temperature of 36003800 K, whereas Argon plasma

restricts the plasma temperature to 22002400 K. Thetemperature gra-dient across the two electrodes is dependent on the diameter of theelectrodes. A smaller diameter anode has higher current density andhashigh temperature which is favourable forsublimationof carbonpre-cursor, whereas cathode with larger diameter has less current densityand is cooler than anode and facilitates nucleation of CNTs. This is whycathode diameter is selected to be greater than anode diameter.

In literature, the effect of temperature on nanotube formationhas notbeen vastly explored andis a potential area of investigation forbetter un-derstanding of the growth of CNTs. Among few published reports, Kimet al.[42]have discussed that nanotube growth occurs below 2000 Ceven though inter-electrode temperature approaches more than4000 C. Matsuura et al. [61] have suggested optimum range forCNT for-mation of 10001250 C. Keidar [27] observed nanotube formation in re-

gion of low plasma temperature of 1300

1800 K, where carbon reacts to

form large molecules or clusters. Lange et al. [145] have found an averageplasma temperature ranging from 4000 to 6500 C in water. Joshi et al.[29]have found that temperature requirement for MWNT is the leastfollowed by DWNT and SWNT for nucleation and growth. Sugai et al.[65] suggested that SWNT formation takes place in strong annealing con-dition. DohertyandChang [100] observeda decreasein yield of MWNTonan increase in temperature. Nishio et al. [146] observed high yield at highplasmatemperature. Liuet al.[135] found that an increase in temperatureincreases the yield of CNT. However Zhao and Liu [147]suggested thatSWNT diameter reduces with an increase in temperature but yield in-creases. Zhao et al.[117]suggested that optimum temperature for CNTformation is 600 C beyond which diameter decreases.

3.8. Effect of setup modication

Thebasic arcdischarge apparatus describedabove andshown in Fig. 1has beenmodied bytheresearchers overthepassage of time to improvethequality, sizeand yield ofCNTs. Zhao and Liu [147] reported an increasein yield of SWNTs by using six anodes. Researchers[29,4042] have useda rotating carbon cathode for a homogeneous micro-discharge resultingin good quality nanotubes. Joshi et al. [29]found a 5055% increase inyield at lower electrode disk rotation speed compared to higher diskspeeds. Lee et al.[40]suggested that rotating electrodes result in contin-uous growth of nanotubes as the carbon vapours moveoutdue to centrif-ugal force resulting in uniformly distributed plasma.

Kanai et al.[148]have proposed a gravity free discharge for highyields. Ando et al.[149]inclined the cathode and anode at an angle of30 to improve the yield of the process. Horvath et al. [86]also investi-gated theeffect of angle between twoelectrodes immersed in water andfound that highest yield is obtained at 90 inclination. Upon inclinationof the electrodes, a majority of the product does not get deposited onthe cathode and the deposition mostly ies away to the walls, therebyincreasing the yield of SWNTs.

3.9. Effect of cathode shape

The geometry of theelectrodes is also a potential area of research andgreatly affects thequality of theproduct. Usuallythe anodeis chosento beof smaller diameterthancathodewhich results in increased current den-sity at anode, thereby subliming the precursor at lower current levels.Cathode being larger in diameter will be at low temperature due to lesscurrent densityand allows theliquidcarbonto nucleate.Thus,owofcar-bonions is achieved from anode to cathode. Therelative size of electrodes

affects the plasma temperature distribution which directly affects the

Table 2

Effects of different environments on CNT formation.

Type of atmosphere Effect on yield and quality of nanotubes Effect on size of nanotubes Comments

Hydrogen Formation of MWNT is highly graphitizedand has crystalline perfection[120]

H2discharge is better compared to He as itproduces twice aspect ratio nanotubes[120]

Hydrogen is more effective forMWNT formation[97]

Hydrogenhelium H2He produces in growth of CNT with smalldiameter[138]as compared to H2Ar.

Hydrogenargon H2added to argon increases the yield[139,83]Helium Yield of SWNT increases in He[130] Diameter of synthesized CNT in air atmosphere

is smaller than in He[97]

He results in uniform cathode deposit[67]

Argon Argon produces smaller diameter CNT comparedto He[27,28]

Nitrogen N2at low pressure yield more SWNT[140]and MWNTs[70]

Diameter of MWNTs decreases with the increaseof the N2[141]

Nitrogen atom incorporated forclosure of CNT[142]

Hydrogennitrogen Diameter distribution can be controlled by varyingthe mixture ratio of H2and N2gas[79].

Open air Better yield in open air than He[29]CO CO plays requisite role for selective diameter

growth of SWNT[136]Krypton Yield of SWNT is more in Kr compared to Ar[65]NaCl solution Observed short CNTs in NaCl solution[126]



9/16

synthesized product. However, upon using similar electrodes the deposi-tion is not found on cathode but on the chamber walls[59].

The effectof changein electrodeshapeand size has beenstudied inlit-erature by few groups. Researchers[150,151]have observed depositionon anode for cathode diameter less than anode diameter. Joshi et al.[29]suggested that the cathode surface topography may play a decisiverole in nanotube formation. Raniszewski et al.[152]have suggested thatanode tip diameter should be lower than 10 mm. Fetterman et al.[153]

have found an increase in yield for smaller diameter anode. Wang et al.[154] suggest a reduction in cathode area in order to increase SWNT pro-duction which is formed at the walls of chamber. Tang et al.[99]havefound a reduction in current requirement by using a cupped cathodeand outer diameter of synthesized MWNT increases. Nishio et al.[146]and Shi et al.[56]found the yield decreases upon using a sharp cathode.Kia and Bonabi[78]concluded that the shape of electrodes affects theyield of CNTs.

3.10. Size and yield of CNT in arc discharge

The major concern of arc discharge CNT synthesis is to improve theyield of the product. However, arc discharge technique has not been suc-cessful in producing pure CNTs at large scale due to difculty in control-ling the experimental parameters. The yields of nanotubes depend onthe arc current, catalyst composition and its particle size, atmospheric ef-fects and electrode shape and composition. Nishio et al.[146]suggestedthat the yield of nanotube depends on the cooling rate. However, coolingrate is dependent on arc current, cathode size, and thermal conductivityof gas and temperaturegradient. Typicalyieldsare of theorderof few mil-ligrams for a synthesis that runs several minutes and thus makes thistechnique less preferred over CVD. However researchers have improvedthe yield through this technique by controlling the synthesis parameters.Few studies related to modied setups have been alreadybeen discussedabove. Some reports on variations in current and pressure values, elec-trode diameters, catalyst composition and atmosphere have been sug-gested to optimize the arc discharge. The typical yield of nanotubesobtained in arc discharge is ~2050 mg/min per synthesis. However

few attempts to produce CNT through semi-continuous and continuousarc discharge have been made and yields of 2 g/h[155]and 6.5 g/h[57]have been achieved. The control of size and diameter in nanotubes isstill unanswerable. The size of SWNTs may be controlled by using the de-sired size of catalyst particles[28].SWNT diameter increases by 0.2

upon increasing the argon helium ratio by 10%. The effect of variousparameters on size and yield of nanotubes hasbeen tabulated in Table 3.

4. Conclusion

We have reviewed the arc discharge method for synthesis of carbonnanotubes and have tabulated the synthesis conditions reported inliterature for CNT formation. We have discussed the growth mecha-

nism of nanotube formation and have presented discussions onphysics of the individual effect of the critical parameters on thequality and yield of the product as explored in literature. Some ofthe observations are:

The most signicant parameter of arc discharge synthesis is the arccurrent. Largevariations in current ratings corresponding to nanotubeformation have been reported creating an ambiguity. Thus, there is acritical need to further investigate and study its effect to develop anoptimum arc discharge.

The role of atmosphere has been studied vastly and similar observa-tions between pressure and nanotube formation have been reported.

Temperature effects in nucleation and growth of nanotubesin arc dis-charge technique are largely unexplored and considerable studies arerequired to understand the vital role of temperature in arc discharge

synthesis of CNT. Electrode geometry is a decisive parameter in the production of CNT.

Further explorations are necessary to understand the exact role of ge-ometry and shape of electrodes in carbon nanotube formation.

DC power supply has been commonly employed for arc discharge.However, effect of use of different types of power supply (AC andpulsed) has not been reported in large numbers.

Most of the synthesis has been done using graphite as a precursor,whereas other carbon materials like carbon black, and coal have notbeen utilized and are a potential area of investigation.

Mass production of carbon nanotubes is also a concern. Few attemptsto manufacture bulk synthesis have been reported; there is lot ofscope for improvisation by modifying the arc discharge parametersfor better yield.

There is a strong need for theoretical investigations andmathematicalmodels to approximate arc discharge synthesis and develop thegrowth mechanism of CNT in arc discharge.

Thougha large numberof experimental reportshavebeen publishedin the last two decades, this eld still needs experimental and

Table 3

Effect of different growth parameters on size and yield of CNT.

Parameter Effect on size of nanotube Effect on yield

Catalyst Diameter of SWNT increases with sulphur[124,125] Diameter increases by using Co[66] Tungsten reduces the diameter of SWNT[128]

Observed smaller diameter SWNT when Ca

Ni catalyst[56]

Yield of 4 g/h obtained by optimizing the compositionof catalyst[109]

Increase in yield with catalyst[127]

Temperature Increase in temperature reduces the diameter[147,117]. Diameter of MWNT increases with an increase in temperature[114]

Electrodes Electrode has signicant effect on diameter and purity of SWNT[152] Increasing the cathode diameter from 0.8 to 2 cm; length of SWNT

increases from 1.2 to 1.8 m[156]

Smaller anode diameters increase the yield [153] Adjusted the angle between two electrodes and found

that 90 angle produces the best yield [149,86].Atmosphere Observed short CNT in NaCl solution[126]

When atmosphere changes from H2Ar to H2He, it results in smaller diameter CNT[138], Ar affects the diameter of SWNT in HeAr mixture[28,27] Diameter distribution of SWNT changes by varying the kind of inert gas[157] Diameter of synthesized CNT in air atmosphere is smaller than in He[97].

Krypton increases the yield[65]

Grain size Smaller diameter CNTs were formed with larger grain size[103]Current Increasing the current increases the diameter[158] Increase in current increases the yield[74]

Maximum yield at 8 A[32]Setup modication Diameter under 0G condition is 3.7 times larger than 1G[148] Increase in yield by rotating electrodes[29,82]Pressure Growth of SWNT with small diameter can be suppressed when

pressure of CO greater than 4 kPa[136] SWNT diameter was independent of pressure in pure He environment[119]

Yield increases as pressure increases from 250 to400 Torr[159]

Low yield for pressure below 100 Torr[144]



10/16

Fig. A.1. (a):Bargraph showingpaperspublishedon arcdischargesynthesis of CNT. (b):Variation in current values usedin literatureto synthesizeCNTs using arcdischarge. (c): Bargraphshowing number of papers published at different range of pressure.

(a) (b) (c)

Fig. A.2. Pie chart showing percentage of papers published on arc discharge synthesisof CNT (a) using different power supplies, (b)various carbon precursors (c) under differentatmosphere.

Table B.1

Arc discharge synthesis of CNT using DC power supply.

Yearauthor Arc current, synthesistime

Precursor Catalyst Environment, pressure Type of CNTs,diameter, yield

1991Iijima[22] Graphite Argon, 100 Torr MWNT, 520 nm1992Ebbesen and Ajayan[53] 100 A Graphite Helium, 500 Torr CNT, 220 nm1993Iijima et al.[51]1993Bethune et al.[52] 95105 A Graphite Co Helium, 100500 Torr SWCNT, 20 nm1994Lin et al.[112] Graphite

Appendix A

Appendix B

http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E2%81%AE2http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E2%81%AE2http://localhost/var/www/apps/conversion/tmp/scratch_5/image%20of%20Fig.%E2%81%AE1


11/16

Table B.1 (continued)



1994Guerret-Plecourt et al.[160] 100110 A, 3060 min Graphite Helium, 0.6 bar CNT1995Wang et al.[54] 90 A Graphite Hydrogen, 50700 Torr MWNT1995Zhou and Seraphin[161] 220 A/cm2 Branched CNT1996Loiseau and Pascard[162] 100110 A, 3045 min Graphite Helium, 0.6 bar CNT1997Lange et al.[72] 6070 A, 4 min Graphite Helium, 500 Torr1997Journet et al. [55] 100 A, 2 min Graphite NiCoY Helium, 500 Torr SWNT, 520 nm

1997Zhao et al.[98] 20100 A Graphite Hydrogen, 10200 Torr MWNT, 1 nm1998Saito et al.[130] 70 A & 100 A, 13 min Graphite Ru/Pd/Rh/Pt Helium, 1520 Torr SWNT, 1.28 nm1998Chang et al.[163] 5565 A Graphite Helium, 500 Torr CNTs, 1040 nm1998Zhao and Ando[131] 50 A, 3060 s Graphite Hydrogen, 60 Torr MWNT, 120 nm1999Takizawa et al.[92] 100 A, 10 s Graphite NiY2O3 Helium, 500 Torr SWNT1999Yudasaka et al.[123] 100 A, 10 s Graphite Helium, 500 Torr SWNT1999Shi et al.[164] 40100 A Graphite NiY Helium, 100700 Torr SWNT, 1.3 nm, 1.5 g1999Zhao et al.[165] 50 A, 3060 s Hydrogen, 60 Torr MWNT1999Shi et al.[56] 40 A, ~120 min Graphite NiY Helium, 1520 Torr SWNT, 5 g1999Kiselev et al.[166] 80 A, 60 s Graphite Helium, 100 Torr MWNT, 115 mg/min1999Liu et al.[41] 150 A, 3 min Graphite Hydrogen, 200 Torr SWNT, 20 nm, 100 mg2000Ishigami et al.[167] 60 A Graphite Liquid nitrogen MWNT, 44 mg/min/cm2

2000Tang et al.[134] 58 A Graphite Helium, 400 Torr CNT2000Shi et al.[129] 4060 A, ~120 min Graphite NiY Helium, 500700 Torr SWNT, 5 g2000Ando et al.[168] 50100 A, ~3 min Graphite NiY Helium, 400700 Torr SWNT, 1.281.52 nm,

1.24 g/min2000Ando et al.[169] 50 A, 3060 s Graphite Hydrogen, 30120 Torr MWNT

2000

Cheng et al.[155] 150 A, 3 min Graphite Ni

Fe

Co

FeS Hydrogen, 200 Torr SWNT, 100 mg2001Ando et al.[149] 50100 A, 3 min Graphite NiY Helium, 400700 Torr SWNT, 1.341.53 nm2001Srivastav et al.[170] 200 A, 3 min Graphite Helium, 500 Torr CNT2001Farhat et al.[28] 100 A Graphite NiY Helium & argon, 495 Torr SWNT2001Hutchison et al.[83] 7580 A, 40 min Graphite Hydrogenargon, 350 Torr DWNT, 2.74.7 nm2001Kanai et al.[148] 2040 A, 30 min Graphite NiY Helium, 600 Torr SWNT2001Shimotani et al.[171] 100 A, 2 min Graphite Heliumhydrocarbons, 150500 Torr MWNT2001Li et al.[172] 30 A, 5 min Graphite Helium, 15 kPa MWNT/SWNT, 0.34 nm2001Ando et al.[120] 50 A, 3060 s Graphite Hydrogen, 8 kPa MWNT2001Lai et al.[173] 6090 A Hydrocarbon

(xylene, pyrene)Helium, 300600 Torr CNT, 1050 m

2002Osvath et al.[174] 100 A, 2 min Helium, 660 mbar Branched CNT, 10 & 20 nm2002Jain et al. [96] 900 A, 1520 min Graphite Helium, 800 Torr MWNT2002Tang et al.[99] 7585 A Graphite Helium, 400 Torr MWNT2002Li et al.[132] 50 A, 1 min Graphite HydrogenCO MWNT, 1020 nm2002Jong Lee et al.[40] 80120 A Graphite Helium, 500 Torr CNT2002Soo et al.[124] 6080 A, 5 min Graphite NiCoFeS Helium, 100500 Torr SWNT, 1020 nm2002Ando et al.[67] 60 A, 10 min Hydrogen, 13 kPa MWNT

2002Sheng et al.[175] 8595 A Graphite Helium, 375 Torr CNT2002Cadek et al.[87] ~57 A Graphite Helium, 500 Torr MWNT2002Zhen-Hua et al.[144] 5080A,10 min Graphite NiY Helium, 100200 Torr SWNT, 1020 nm2002Doherty and Chang[100] 100 A, 1 min Carbon black Helium, 100 Torr MWNT2003Antisari et al.[70] 3070 A, 1 min Graphite Liquid nitrogen, deionised water MWNT2003Doherty et al.[101] 100 A, 1 min Carbon black Helium, 100 Torr MWNT2003Saito et al.[125] 50 A Graphite FeSNiSCoS Hydrogenhelium, 300 Torr DWNT, 25 nm2003Qiu et al.[176] 5070 A, 20 min Coal Fe Helium, 490 Torr SWNT, 1020 nm2003Jung et al.[177] 80 A Graphite Liquid nitrogen MWNT, 2050 nm2003Zhao et al.[133] 3070 A, 3 min Graphite Fe Hydrogenargon, 60500 Torr SWNT2003Lange et al.[145] 3040 A, 10 min Deionised water Nanocarbon/CNT2003Cui et al.[142] 3544 A, 3 min Graphite Nitrogen, 0.02900 Torr MWNT2003Osvath et al.[178] 100 A, 2 min Graphite NiY Helium, 500 Torr Branched CNT, 10 nm2003Tarasov et al.[179] 50130 A Graphite NiCo + YNi2 Helium, 400800 Torr SWNT2004Liu et al.[135] 100 A, 10 min Graphite Hydrogen, 50 kPa Amorphous CNT2004Waldorff et al.[73] 78.5 A, 180 s Graphite NiY Helium, 500700 Torr SWNT, 2 nm2004Nishio et al.[146] 60 A, 28 s Graphite Helium, 760 Torr CNT

2004

Wang et al.[108] 60

450 A Polyvinyl alcohol Fe Helium, 260 Torr MWNT2004Zhao et al.[57] 80 A, 5 min Graphite NiCo Hydrogen, 500 Torr Amorphous CNT,

720 nm, 6.5 g/h2004Bera et al.[150] 35 A Graphite PdCl2 CNT2004Cui et al.[180] 3544 A, 50120 s Graphite Nitrogen, 0900 Torr MWNT/SWNT2004Zhao and Liu[147] 60 A, 5 min Graphite Helium, 400 Torr SWNT2004Jinno et al.[181] 50 A, 3060s Graphite Hydrogennitrogen, 70 Torr MWNT, 1 nm2004Keidar and Waas[182] CNT2004Itkis et al.[58] 90 A, 6090 min Graphite NiY Helium, 680 Torr SWNT, 515 g2004Qiu et al.[183] 5070 A Coal Helium, 250570 Torr MWNT2004Sano et al.[184] 60 A, 45 s Graphite Ni Liquid nitrogen SWNT2005Murr et al.[107] 50100 A Tire Powder Helium, 250 Torr MWNT, 550 nm2005Hahn et al.[185] 60 A Graphite Hydrogen, 13 kPa MWNT2005Shang et al.[186] 50 A Graphite Helium, 30120 Torr MWNT, 1.1 nm2005de Heer et al.[33] 100 A Graphite Helium, 375 Torr MWNT, 320 nm2005Ando et al.[79] 4070 A, 15 min Graphite Fe Hydrogennitrogen, 200 Torr SWNT

(continued on next page)



12/16




2005Yang et al.[121] 100120 A Graphite Hydrogenargon, 70 Torr H2, 50 Torr argon DWNT2005Guo et al.[139] 3070 A Graphite Fe Hydrogenargon, 200 Torr SWNT2005Li et al.[109] 120 A, 3 min Graphite NiCoFeS Hydrogen, 240 Torr DWNT, 4 g/h2005Wang et al.[126] 50 A, 60 s Graphite NaCl SWNT, 0.89 nm

MWNT, 8.9 nm2005Xin Lv et al.[84] 90120 A, 30150 min Graphite Helium, 530550 Torr SWNT

2005Makita et al.[140] 5060 A Graphite NiCo Helium, argon, nitrogen 501000 Torr SWNT2005Yao et al.[187] 90 A, 510 min Graphite NiHO2O3 Helium, 600 Torr SWNT, 1030 nm, 0.61 g2005Montoro et al.[80] 65 A, 20 min H3VO4aqueous solution SWNT/MWNT, 2030 nm2005Zhao et al.[117] 80 A Graphite Hydrogen, 500 Torr Amorphous CNT, 720 nm2005Tang et al.[188] 8595 A Graphite Helium, 400 Torr CNT2005Wang et al.[189] 50 A, 60 s Graphite Fe Deionised water, open air CNT2006Doherty et al.[102] 100 A, 1 min Carbon Black Helium, 100 Torr CNT2006Wang et al.[154] 70 A Graphite NiY Helium, 120 Torr SWNT, 10 g2006Chen et al.[104] 120 A Carbon black CoCoS Hydrogenargon DWNT2006Qiu et al.[122] 70 A, 15 min Graphite FeSKCl Hydrogen, 350 Torr DWNT2006Wang et al.[190] 70 A, 15 min Coal Argon, 8090 kPa CNT, 3080 nm2006Lange et al.[75] 25 A & 55 A Graphite/carbon

blackFe Hydrogenargon, 200 Torr SWNT

2006Zhao et al.[157] 50 A, 320 min Graphite Fe Hydrogen-inert gases, 50500 Torr SWNT2006Suzuki et al.[191] 60A, 10 min Graphite Hydrogenhelium, 100 Torr MWNT, 900 mg2006Yusoff et al.[25] 16 A Graphite Nitrogen, 760 Torr CNT2006HH Kim and HJ Kim[97] 4080 A Graphite Air, 100760 Torr MWNT

2006

Wang et al.[119] 70

80 A, 10 min Helium, 50

60 kPa Branched CNT2006HH Kim and HJ Kim[23] 40100 A Graphite NiCoTi Helium, 500 Torr MWNT/SWNT2007Cazzanelli et al.[113] 60 A, 100 s Graphite NiCr Helium, 375 Torr MWNT2007Okada et al.[90] 120 A, 30s Toluene NiMoFe Toluene CNT2007Mathur et al.[192] 100120 A Graphite + coke

powderHelium, 600 Torr MWNT/SWNT

2007Song et al.[114] 100 A, 58 min Graphite CoSPt Hydrogen, 300 Torr MWNT2007HH Kim and HJ Kim[115] 1070 A Xylene Ferrocene Xylene ferrocene, 30500 Torr MWNT/SWNT2007Guo et al.[193] 50 A Graphite Deionised water MWNT, 520 nm2007Duncan et al.[194] 80 A & 100 A, 5 min Heliumargon, 500700 mbar MWNT2007Qiu et al.[105] 5060 A, 10 min Coal Fe Helium, 375525 Torr DWNT, 15 nm2007Sun et al.[127] 70100 A, 10 min Graphite Fe3O4 Hydrogenargon, 200 Torr SWNT, 0.8 g2007Delong He et al.[91] 80280 A Graphite NiCo Helium, 500 Torr SWNT2007Xing et al.[195] 3075 A Graphite Deionised water MWNT, 1020 nm2008Grebenyukov et al.[141] 65 A Graphite NiY2O3 Nitrogen, 50760 Torr SWNT2008Li et al.[196] 90 A Graphite NiOY2O3 Helium, 300 Torr SWNT2008Fetterman et al.[153] 4070 A, 12480 s Graphite Helium, 600 Torr CNT2008Joshi et al.[29] 150 A/cm2 Graphite Open air, helium, 500 Torr MWNT

2008Keidar et al.[197] 7080 A Graphite NiY Helium, 500700 Torr SWNT2008Keidar et al.[198] 180 s Graphite NiY Helium, 500700 Torr SWNT2008Kim et al.[137] 10 A, 2 min Graphite Beer foam MWNT2009Kim et al.[199] 15 A Graphite Beer foam MWNT2009Charinpanitkul et al.[158] 50250 A Graphite Liquid nitrogen MWNT, 825 nm2009Ha et al.[200] 3070 A, 20 min Graphite Fe Hydrogen, 50450 Torr SWNT2009Labedz et al.[103] 4060 A Graphite/carbon

blackHydrogenargon, 30 kPa SWNT

2009Hoa et al.[201] 85 A/cm2, 1 min Graphite FeNiMo Hydrogen, 400 Torr SWNT2009Jahanshahi et al.[69] Graphite NiMo LiCl MWNT/SWNT, 7.7 mg/min2010Jahanshahi et al.[116] 100 A, 60 s Graphite NiMo NaCl MWNT2010Scalese et al.[30] 80 A, 60 s Graphite Liquid nitrogen MWNT2010Scalese et al.[68] 80 A, 30 s Graphite Liquid nitrogen CNT2010J Qiu et al. [106] 85110 A Fullerene

waste sootHydrogenargon, 300 Torr DWNT, 1.081.44 nm

2011Nishizaka et al.[88] 10 min Graphite +coal + carbon

black

Helium, 100 Torr SWNT

2011Ding et al.[202] 120 A Graphite Hydrogenhelium, 400 Torr CNTs2011Hou et al.[203] 150 A, 3 min Graphite NiFeCo Hydrogen, 200 Torr SWNT, 1 g2011Zhang et al.[204] 120 A Graphite Hydrogen, 240 Torr MWNT2011Su et al.[136] 100 A Graphite NiY HeliumCO, 225 Torr SWNT2012Tripathi et al.[205] 105 A, 58 min Graphite CNT2012Kim et al.[42] 100 A Graphite Argon MWNT2012Zhao et al.[206] 80 A Graphite Air, 60 Torr MWNT2012Wu et al.[26] 120 A Graphite NiY2O3 Hydrogenhelium, 530 Torr SWNT2012Su et al.[156] 90 A Graphite NiY Helium, 375 Torr SWNT2012Liang et al.[24] 80 A, 100 A, 150 A, 8 min Graphite Helium, 760 Torr MWNT2012Zhang[138] 50 A Graphite FeRh Hydrogenargon, 200 Torr Few-walled CNT,110 nm2012Zhao et al.[82] 80 A, 20 min Graphite Air, 40300 Torr MWNT2012Cai et al.[151] 210 A, 4 min Graphite Hydrogen, 50 Torr CNT, 4060 nm2013Zhao et al.[81] 80 A, 20 min Graphite Air, 50 Torr DWNT2013Mohammad et al.[159] 7595 A, 5 min Graphite NiY2O3 Hydrogenargon, 150 Torr CNTs, 3444 mg2013Fang et al.[128] 80 A, 10 min Graphite FeW Hydrogenargon, 200 Torr SWNT, 1030 nm,

100200 mg



13/16

theoretical investigations to establish a correlationbetween the synthe-sis parameters and nucleation of carbon nanotubes and help us betterunderstand the growth mechanism of carbon nanotubes.

References

[1] H. Dai, Carbon nanotubes: synthesis, integration, and properties, Acc. Chem. Res.35 (12) (Dec. 2002) 10351044.

[2] J. Liu, S. Fan,H. Dai,Recentadvances in methodsof forming carbon nanotubes, MRSBull. 29 (04) (2004) 244250.

[3] S. Arepalli , Laser ablation process for single-walled carbon nanotube production, J.Nanosci. Nanotechnol. 4 (4) (Apr. 2004) 317325.

[4] V. Shanov, Y. Yun, M. Schulz, Synthesis and characterization of carbon nanotubematerials, J. Univ. Chem. 41 (4) (2006) 377390.

[5] . ncel, Y. Yrm, Carbon nanotube synthesis via the catalytic CVD method: a re-view on the effect of reaction parameters, Fullerenes Nanotubes CarbonNanostruct. 14 (1) (Jun. 2006) 1737.

[6] C.T. Kingston, B. Simard,Recentadvances in laser synthesis of single-walled carbonnanotubes, J. Nanosci. Nanotechnol. 6 (5) (May 2006) 12251232.

[7] Y. Ando, X. Zhao, Synthesis of carbon nanotubes by arc-discharge method, NewDiam. Front. Carbon Technol. 16 (3) (2006) 123137.

[8] P.J.F. Harris, Solid state growth mechanisms for carbon nanotubes, Carbon N. Y. 45(2) (Feb. 2007) 229239.

[9] F. Danafar, a. Fakhru'l-Razi,M.A.M.Salleh, D.R.A. Biak, Fluidized bedcatalyticchem-ical vapor deposition synthesis of carbon nanotubesa review, Chem. Eng. J. 155(12) (Dec. 2009) 3748.

[10] Y. Ando, Carbon nanotube: story in Ando laboratory discovery of CNTs, 3 (1)

(2010) 3

22.

[11] M. Kumar, Y. Ando, Chemical vapor deposition of carbon nanotubes: a review ongrowth mechanism and mass production, J. Nanosci. Nanotechnol. 10 (6) (Jun.2010) 37393758.

[12] J.-P. Tessonnier, D.S. Su, Recent progress on the growth mechanism of carbonnanotubes: a review, ChemSusChem 4 (7) (Jul. 2011) 824847.

[13] J. Prasek, J. Drbohlavova, J. Chomoucka, J. Hubalek, O. Jasek, V. Adam, R. Kizek,Methods for carbon nanotubes synthesisreview, J. Mater. Chem. 21 (40) (2011)15872.

[14] D.B. Geohegan, A.A.Puretzky,X. Fan, M.A. Guillorn, M.L. Simpson, V.I.Merkulov,S.J.Pennycook, Mechanisms of single-wall carbon nanotube growth by the laser va-porization technique: in situ imaging and spectroscopy, MRS Proc. 593 (Feb.2011) 3.

[15] M. Lubej,I. Plazl, Theoreticaldescriptionsof carbonnanotubes synthesis in a chem-ical vapor deposition reactor: a review, Chem. Biochem. Eng. Q. 26 (3) (2012)277284.

[16] C. Journet, M. Picher, V. Jourdain, Carbon nanotube synthesis: from large-scaleproduction to atom-by-atom growth, Nanotechnology 23 (14) (Apr. 2012)142001.

[17] Y. Gogotsi, J.A. Libera, M. Yoshimura, Hydrothermal synthesis of multiwall carbonnanotubes, J. Mater. Res. 15 (12) (Jan. 2011) 25912594.

[18] W.K. Hsu, J.P. Hare, M. Terrones, H.W. Kroto, D.R.M. Walton, P.J.F. Harris,Condensed-phase nanotubes, Nature 377 (6551) (Oct. 1995) 687.

[19] J.B. Bai, Synthesis of SWNTs and MWNTs by a molten salt (NaCl) method, 365(2002) 184188.

[20] N. Pierard, A. Fonseca, Z. Konya, I. Willems, G. Van Tendeloo, Production ofshort carbon nanotubes with open tips by ball milling, 335 (February)(2001) 18.

[21] F. Liu, X. Zhang,J. Cheng,J. Tu, F. Kong, W.Huang, C.Chen,Preparation of short car-bon nanotubes by mechanicalball millingand their hydrogenadsorption behavior,

Carbon N. Y. 41 (13) (Jan. 2003) 2527

2532.




2013Su et al.[143] 90 A Graphite FeS Air, 0.75135 Torr Few-walled CNT, 1.66 nm2014Chaudhary et al.[207] 90 A Graphite Methane, 100, 300 and 500 Torr MWCNT, 1020 nm

Table B.2Arc discharge synthesis of CNT using AC power supply.

Yearauthor Arc current, synthesis time Precursor Environment, pressure Type of CNTs, diameter, yield

1992Ebbesen and Ajayan[53] 100 A Graphite Helium, 500 Torr CNT, 220 nm1997Zeng et al.[59] 700 A Graphite Helium, 140 Torr MWNT1999Ohkohchi M[60] 85 A, 12 min Graphite Helium, 400 Torr SWNT2003Biro et al.[74] 4585 A Graphite Deionised water MWNT, 510 mg/min2006Horvath et al.[86] 40 A, 12 h Graphite Deionised water MWNT, 1035 nm2007Matsuura et al.[61] 70100 A Graphite Helium, 600 Torr MWNT2010Matsuura et al.[89] 110120 A Graphite CNT2011Zhao et al.[32] 620 A Benzene Argon, 760 Torr MWNT2012Jia-Shiang Su[93] 2.5 A, 1.4 ms Graphite Air CNT2013Yousef et al.[208] 75 A Graphite Deionised water MWNT, 0.6g/h2013Kia and Bonabi[71] 50 A, 510 min Hydrocarbon Argon, 525 Torr MWNT, 50100 nm

Table B.3

Arc discharge synthesis of CNT using pulsed DC power supply.

Yearauthor Arc current, synthesis time Precursor Catalyst Environment, pressure Type of CNTs, diameter, yield

2000Sugai et al.[65] 22 A, 330 h Graphite NiCo Argon, 500 Torr SWNT, 210 mg2001Murooka et al.[94] 12 A, 13 s Graphite Argon, 250500 Torr Nanoparticles2003Sugai et al.[209] 4060 A Graphite NiY Argon DWNT2004Parkansky et al.[62] 7100 A Graphite Open air MWNT2006Imasaka et al.[63] 30A, 4 h Graphite De-ionized water MWNT2007Roch et al.[64] 100 A Graphite NiCo Argon, 75 Torr SWNT2008Tsai et al.[95] 22.5 A, 8001300 s Graphite Fe Open air MWNT2008Yoshida et al.[66] 50 A Graphite Metal catalysts Argon, 760 Torr DWNT2009Tsai et al.[76] 2.5 A, 1 ms Graphite Open air, 760 Torr MWNT2010Jia-Shiang Su[77] 35 A, 1.2 ms Graphite Open air MWNT

2012

Kia and Bonabi[78] 10 A, 15 min Acetone Argon, 525 Torr MWNT2012Takekosh et al.[85] 20 A, 60 min Graphite Metal catalysts De-ionized water CNT, 5 nm

http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0005http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0005http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0005http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0005http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0010http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0010http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0010http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0010http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0015http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0015http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0015http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0015http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0020http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0020http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0020http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0020http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0030http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0030http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0030http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0030http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0035http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0035http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0035http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0035http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0040http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0040http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0040http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0040http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0865http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0865http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0865http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0865http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0055http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0055http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0055http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0055http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0875http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0875http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0875http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0070http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0070http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0070http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0070http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0075http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0075http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0880http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0880http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0880http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0880http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0080http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0885http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0880http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0880http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0075http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0075http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0070http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0070http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0875http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0875http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0875http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0065http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0060http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0870http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0055http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0055http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0050http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0865http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0865http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0860http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0040http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0040http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0035http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0035http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0030http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0030http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0025http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0020http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0020http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0015http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0015http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0010http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0010http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0005http://refhub.elsevier.com/S0925-9635(14)00195-2/rf0005


14/16

[22] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (6348) (Nov. 1991)5658.

[23] H.H. Kim,H.J. Kim,The preparationof carbon nanotubes by dc arcdischargeusing acarbon cathode coated with catalyst, Mater. Sci. Eng. B 130 (13) (Jun. 2006)7380.

[24] F. Liang, T. Shimizu, M. Tanaka, S. Choi, T. Watanabe, Selective preparation of poly-hedral graphite particles and multi-wall carbon nanotubes by a transferred arcunder atmospheric pressure, Diamond Relat. Mater. 30 (Nov. 2012) 7076.

[25] H.M. Yusoff, R. Shastry,T. Querrioux, J. Abrahamson, Nanotube deposition in a con-tinuousarc reactor for varying arcgap and substrate temperature, Curr. Appl. Phys.6 (3) (Jun. 2006) 422426.

[26] Y.Wu, T.Zhang, F. Zhang,Y. Wang,Y. Ma, Y. Huang, Y.Liu,Y. Chen,In situ synthesisof graphene / single-walled carbon nanotube hybrid material by arc-discharge andits application in supercapacitors, Nano Energy 1 (6) (2012) 820827.

[27] M. Keidar, Factors affecting synthesis of single wall carbon nanotubes in arc dis-charge, J. Phys. D. Appl. Phys. 40 (8) (Apr. 2007) 23882393.

[28] S. Farhat, M. Lamy de La Chapelle, A. Loiseau, C.D. Scott, S. Lefrant, C. Journet, P.Bernier, Diameter control of single-walled carbon nanotubes using argonheliummixture gases, J. Chem. Phys. 115 (14) (2001) 6752.

[29] R. Joshi, J. Engstler,P.K. Nair, P. Haridoss, J.J. Schneider,High yield formation of car-bon nanotubes using a rotating cathode in open air, Diamond Relat. Mater. 17 (6)(Jun

synthesis of carbon nanotubes

Documents