(2004)High-rate flame synthesis of vertically aligned carbon nanotubes using electric field control

High-rate flame synthesis of vertically aligned carbonnanotubes using electric field controlWilson Merchan-Merchan a ,Alexei V.Saveliev b ,Lawrence A.Kennedyb,*aDepartment of Materials Sciences and Engineering,University of Illinois at Chicago,Chicago,IL 60607,USAbDepartment of Mechanical and Industrial Engineering,University of Illinois at Chicago,851South Morgan Street,Chicago,IL 60607,USAReceived 31July 2003;accepted 17December 2003AbstractThe electric field controlled synthesis of carbon nanomaterials on a Ni-based catalytic support positioned at the fuel side of the opposed flow oxy-flame is studied experimentally.Carbon nanomaterials formed on the probe surface are comparatively analyzed for two characteristic operational modes:a grounded probe mode and a floating probe mode.In a grounded mode a number of various carbon nanostructures are formed depending on the probe location in flame.Observed nanoforms include multi-walled carbon nanotubes (MWNTs),MWNT bundles,helically coiled tubular nanofibers,and ribbon-like coiled nanofibers with rectan-gular cross-section.The presence of various carbon nanoforms is attributed to the space variation of flame parameters,namely flame temperature and concentration of chemical species.It is found that the presence of an electric potential (floating mode operation)provides the ability to control the nanostructure morphology and synthesis rate.A thick layer (35–40l m)of vertically aligned carbon nanotubes (VACNTs)is found to be formed on the probe surface in the floating potential mode.This layer is characterized by high uniformity and narrow distribution of nanotube diameters.Overall,the electric field control method demonstrates sta-bilization of the structure in a wide flame region while growth rate remains dependent on flame location.Ó2004Elsevier Ltd.All rights reserved.Keywords:A.Carbon nanotubes;B.Catalyst support;C.Scanning electron microscopy1.IntroductionIn recent years great effort has been devoted to the study of synthesis of carbon nanotubes in flames.Since the discovery of nanotubes in 1991by Iijima [1]con-ventional methods such as arc plasma discharges [2,3],pulsed laser vaporization [4],and chemical vapor deposition (CVD)[5]have dominated the field.Re-cently,it has been demonstrated in a number of publi-cations that flames can be applied as a relatively inexpensive,robust,pyrolysing carbon source for growing tubular nanomaterials [6–14].Mainly studies in co-flow diffusion flames with the introduction of a cat-alyst in the form of nano-aerosol or in the form of solid support have been reported.Yuan et al.[6]analyzed flame-based growth of nanotubes on a Ni/Cr support in laminar co-flow methane–air diffusion flames.Forma-tion of entangled MWNTs with a diameter range of 20–60nm was reported.In related work,Yuan and co-workers [7]used a catalytic support in the form of a stainless steel grid to produce MWNTs in an ethylene–air diffusion flame.The electroplating of the grid with cobalt resulted in synthesis of aligned MWNTs.The metal catalyst dispersed on TiO 2substrate was used by Vander Wal [8]to generate MWNTs in ethylene/air and acetylene/air diffusion co-flow flames.It was demon-strated that the structure of the obtained nanotubes and nanofibers strongly depends on the catalytic particle shape and chemical composition.Recently,Vander Wal and co-workers [9,10]also reported that single-walled carbon nanotubes (SWNTs)can be grown in premixed co-flow flames by seeding the fuel line with ferrocene and compositions of metal nitrates serving as catalyst precursors for the formation of nanotubes.It was demonstrated that ferrocene and Fe nanoparticles can yield bundles of self-assembled single-wall nanotubes with diameters as small as two nanometers.The rich premixed flame synthesis of carbon nanotubes using supported catalyst [11]was optimized by selection of optimal flame conditions including fuel composition and*Corresponding author.Tel.:+1-312-996-2400;fax:+1-312-996-8664.E-mail address:lkennedy@ (L.A.Kennedy).0008-6223/$-see front matter Ó2004Elsevier Ltd.All rights reserved.doi:10.1016/j.carbon.2003.12.086Carbon 42(2004)599–608fuel-to-air ratio.The studied fuels include methane, ethane,ethylene,acetylene,and propane.A recent work by Height and co-workers[12]used premixed co-flow flames to grow SWNTs.A detailed characterization of SWNT growth was given with emphasis onflame posi-tion andflame air-to-fuel ratio.The non-catalytic for-mation of carbon nanotubes was reported in opposed flow oxygen enrichedflame studies[13].The strong potential of theseflames for carbon nanomaterial syn-thesis was recently proven employing a catalytic probe [14].If compared with CVD and plasma methods,a typi-calflame is a reacting medium characterized by strong thermal and chemical non-uniformity.It is not surpris-ing,that a number offlame studies show a high mor-phological and growth rate sensitivity of formed carbon nanomaterials to theflame location.An efficient control method is required to improve uniformity and produc-tivity offlame based synthesis and utilization of elec-tromagneticfield control is one of the promising approaches.The electricfield control is successfully tested in chemical vapor deposition(CVD)and plasma synthesis studies[15–21]as reported recently by several authors:Kuzuya et al.[15]applied an electricfield for effective control of producing helically coiled carbon materials using CVD;Avigal et al.[16]grew short aligned nanotubes by positively biasing the substrate during the nanotube formation in CVD;Srivastava et al.[17]showed that nanotubes that are grown under the influence of an electricfield in plasma can permanently maintain their alignment when the electricfield force is removed;Lee et al.[18]using CVD and magneticfields showed that it is possible to grow aligned carbon na-notubes at any angle with respect to the substrate sur-face;Ural et al.[19]used CVD to grow carbon nanotubes with and without the influence of electric fields to show the ability of electricfields to grow aligned carbon nanotubes;Colbert et al.[20]used an arc plasma discharge to show that an electricfield is essential for nanotube growth by stabilizing the open tip and pre-venting it from closure;Srivastava et al.[21]used plas-ma varying voltages to study the size,length,and alignment of carbon nanotubes.To our knowledge,electricfields have not been ap-plied to control growth of carbon nanotubes inflames prior to this study.In this article the carbon nanoforms formed on a catalytic probe in opposedflow oxy-flame are comparatively analyzed for various probe potentials. It is shown that implementation of an electricfield control allows to increase their structural uniformity and growth rate.2.ExperimentalFig.1shows the schematic of the experimental setup employed in the present study.A counter-flow burner forms two opposite streams of gases;the fuel(methane seeded with4%of acetylene)is supplied from the top nozzle and the oxidizer(50%O2+50%N2)is introduced from the bottom nozzle.The fuel and oxidizerflows impinge against each other to form a stable stagnation plane,with a diffusionflame established from the oxi-dizer side.The introduction of co-flowing nitrogen through a cylindrical annular duct around the outer edge of the oxidizer nozzle has the function of extin-guishing theflame near the outer jacket and isolating it from the environment.A detailed description of the burner is given by Beltrame et al.[22].Technical purity methane(98%,AGA Gas Inc.)was seeded with atomic absorption purity acetylene(99.8%,AGA Gas Inc.).The experiments were conducted with constant fuel and oxidizer velocities and a strain rate equal to20sÀ1.A40mm long catalytic probe was introduced through theflame-protecting shield to the yellow soot-containing region of theflame.The central part oftheFig.1.Schematic of the experimental setup.600W.Merchan-Merchan et al./Carbon42(2004)599–608probe($25mm)was used to study the structure of deposited materials.The0.64mm diameter probe is an alloy with compositions of73%Ni+17%Cu+10%Fe. The axial position of the probeðZÞwas controlled by the positioning system.Reported experiments were con-ducted with residence time of10min.To generate radial electricfields on the probe surface, the probe was supported on Teflonâisolators while the burner nozzles were kept at ground potential(Fig.1).In general this configuration allows to generate a variety of electricfields controlling the probe potential with the external electric source.In the present work,experi-ments were conducted at two characteristic probe potentials:grounded probe and probe atfloating po-tential.When nodes one and two are connected no electric field is generated;this mode is further referred as the grounding probe mode(GPM).On the other hand, when node one and two are disconnected,electricfield is generated between the surface of the catalytic probe and the edges of the fuel and oxidizer nozzles;this configu-ration is further referred as thefloating potential mode (FPM).Due to the small size of the probe relative to size of the burner nozzles the electricfield around the probe can be well approximated as radial.The source of the floating potential is the transport of ions and electrons formed in high-temperature oxygenflame to the probe surface;a potential difference close to300mV was typically measured when the probe was in FPM.The initial surface scans of the catalyst probe were performed by a scanning electron microscope(JEOL Inc.,Model JSM-6320F)with a coldfield emission source.The specimens for TEM examinations were prepared by ultrasonic dispersion of carbon deposits collected from a specific probe location in acetone;a drop of the suspension was placed on the electron microscope grid.The detailed structural characteristics of the deposited material were obtained from high-res-olution electron microscopy studies performed in a JEOL JEM-3010electron microscope with the magnifi-cation range from50to1,500,000times.Images were collected on a Gatan digital imaging system and pro-cessed by Digital Micrograph software.3.Results and discussionWhen the catalytic probe was inserted in GPM a variety of carbon nanostructures were observed depending on position of catalytic probe in the fuel zone of the opposedflowflame.Those carbon structures in-clude MWNTs and MWNT bundles,nanofibers with varying degree of crystallinity,helical regularly coiled tubular carbon nanotubes,ribbon-like coiled nanofibers with rectangular cross section,and,finally,long($0.2 mm)uniform-diameter($100nm)tubular nanofibers with regular internal structure of carbon layers.Some of the observed structures,such as coiled tubular and rib-bon-like nanofibers,have been reported[23,24]to grow only under special conditions in CVD.Fig.2(a)shows a typical regularly coiled carbon nanofiber formed in ourflames.The diameters of these coils ranged from20to100nm with lengths of several microns.These structures were typically observed in the zone close to Z¼9mm,in the vicinity of theflame front.Qualitatively,the formation of helical nanotubes can be explained by variation of deposition rates and, hence,extrusion velocities along the contact curve be-tween the active catalytic particle and thealready Fig.2.SEM images of carbon nanomaterials grown in various loca-tion of the opposedflowflame on a single catalytic support:(a)heli-cally coiled spiral carbon nanotube,(b)multi-walled carbon nanotubes,(c)coiled carbon nanofiber with rectangular cross-section.W.Merchan-Merchan et al./Carbon42(2004)599–608601formed tube[23].The sharp gradients of temperature and chemical species in the vicinity of theflame zone induce sensible variations of carbon deposition rates providing the condition for the growth of helical struc-tures.Irregular carbon nanotubes are widely observed in theflame region close to Z¼8:5mm.They form a coating layer that typically contains entangled web of randomly directed tubes,carbon nanofibers and soot particulates,as shown in Fig.2(b).The tube diameters vary significantly,typically from10to45nm.The internal tube structure is regular,although carbon layers often form an angle with the tube axis.As a result bamboo-like structures are widely observed.Irregularly oriented nanotubes frequently change direction of growth;kinks and bends are common for these struc-tures.The other distinctive coiled carbon structure observed in ourflames possesses a nontubular form,Fig.2(c). These structures are typically found to be present in the low temperature zone Z¼8mm.The nanofiber has a distinctive ribbon-like appearance with unwound ribbon rolls.The rectangular cross section has a height of300 nm and a thickness of100nm.It was reported previously [25],that growth of these rarely observed ribbon-like filaments could be catalyzed by small iron-containing particles in CO containing atmosphere at700°C,which correspond to theflame environment at the vicinity of Z¼8mm.The variation of structure and morphology of formed carbon nanomaterials is directly attributed to the strong variation of temperature and chemical composition in the studiedflame region.The distributions of several major hydrocarbons(CH4,C2H2,CO,and C6H6)that can contribute to the growth of carbon deposits are shown in Fig.3along with the temperature profile[22]. It is easy to conclude that all above components vary significantly in theflame region of interest.Thus,con-centration of CH4and C2H2diminishes below100ppm for Z>10:5mm;the concentration of CO grows with Z reaching its maximum at this point;C6H6is present in essential quantities only from8to10mm,maximizing at 9.5mm.The essential variations of temperature and chemical composition are inevitable in a variety offlame config-urations.The effective control method is necessary to stabilize growth of carbon nanomaterials with specified structure and morphology.The electricfield control method was implemented in this study electrically insulating the probe from the grounded support and operating it at afloating potential.The catalytic probe of the same metallic composition was inserted at the axialflame position of Z¼8:5mm in FPM(approximate probe potential)300mV).The produced carbon deposits were analyzed with SEM (Figs.4and5).Fig.4shows that controlled electricfield growth generates a coat of VACNTs.Low magnifica-tion image shows that VACNT layer uniformly coat the catalytic substrate.The generated nanotubes are char-Fig.4.SEM image of orderly VACNT layer covering the probe sur-face;floating potential mode,Z¼8:5mm.The layer is partially re-moved revealing the bare catalytic surface.Fig.5.High-resolution SEM of the wall edge of the layer shown in Fig.4displays nanotube purity and alignment.602W.Merchan-Merchan et al./Carbon42(2004)599–608acterized by high purity and alignment,as shown in Fig.5.For this axial position,hundreds of nanotube diam-eters were measured.The results show a very narrow diameter distribution with a mean diameter close to 38nm.Fig.2(b)and Figs.4,5depict the carbon nanomate-rials grown on identical probes in the same flame loca-tion respectively in GPM and FPM conditions.Analysis of SEM images on the synthesized carbon materials on both catalytic substrates suggests that nanotube growth is greatly enhanced when the electric field is present (FPM)if compared with the case when the electric field is absent (GPM).Other interesting aspects,displayed in Figs.4and 5,include not only the presence of VACNTs but also the absence of contaminants such as soot or other nontubular carbon structures that are often pres-ent in the flame synthesis of carbon nanotubes.It should be noted that the samples analyzed in this study were never purified by any kind of chemical and/or physical treatment.With electric field stabilization,flame position re-mains the important factor in controlling the thickness of the coat layer of nanotubes formed under FPM.Figs.6and 7show the variation of nanotube formation along the burner axis.Fig.6is a low resolution SEM image of the catalytic substrate inserted in the FPM at the axial distance of Z ¼9:5mm.The well-defined layer formed here shows highly ordered carbon nanotube arrays,similar to those found in the previous position.The application of high-resolution imaging in Fig.6reveals a highly dense bundle of nanotubes attached to the tips of the undisturbed nanotube layer,as shown in Fig.7.High-resolution imaging shows that these nanotube bundles are free of contaminants as well.It is evident by comparing micrographs obtained at the axial position of Z ¼8:5mm to those obtained at Z ¼9:5mm that the thickness of the coating layer decreased when the sub-strate is inserted further down from the edge of the fuel nozzle.Several SEM images were collected at variouslocations of the probe surface;an average layer thick-ness of 9l m was measured for this flame location.Another interesting aspect of these nanotube layers is the strong van der Waals body attraction force existing between the tubes in the self-formed macro bundles.This aspect is unique since it greatly simplifies their harvesting.That is,micron size arrays of VACNTs can be easily removed from its roots.A catalytic substrate inserted at the axial flame po-sition of Z ¼10:5mm shows no apparent sign of arrays of VACNTs (Fig.8).But high-resolution SEM reveals that thinner and less uniform layers of VACNTs are still being grown as shown in Fig.9.The nanotube layer at this position is still aligned and defined but not as well as in the two previous positions.It can be speculated that at this flame position the conditions of CNT growth are far from optimal and the electric field control has only limited effect.A thickness of VACNT arrays of 3l m is characteristic for this flame position.Additionally,high and low resolution SEM analysis was performed on the surface of the support catalytic substrate at different locations in order to confirm the uniformity of the nanotube layer.In this particular experiment,the probe was inserted into the flame ataFig.6.Low resolution SEM image shows the catalyst substrate coated with a layer of carbon nanotubes (FPM),Z ¼9:5mm.Fig.7.Higher resolution SEM of a micro area in Fig.6.Fig.8.Low resolution SEM image on the catalyst substrate coated with a layer of carbon material (FPM),Z ¼10:5mm.W.Merchan-Merchan et al./Carbon 42(2004)599–608603position where the most thick layer of nanotubes were observed,Z¼8:5mm.After the exposure to theflame, the probe was placed on a stainless steel stud for microscope studies.Fig.10represents SEM images of arrays of orderly and high purity nanotubes scanned at top,center,and bottom of the probe surface,respectively.For all probe locations considered for microscopic analysis,arrays of nanotubes covered the probe with orientation perpen-dicular to the probe surface.By inspection of Fig.10it is evident that most of the formed material remains at-tached to the surface.Higher resolution SEM on the walls of this material showed that the bulk material is composed of arrays of nanotubes.The separated mate-rial always tends to remains packed like bundles of hay and always maintains its original length.In the upper right corner of Fig.10(b)a bundle of highly dense na-notubes is observed.This bundle has a cylindrical shape and a length of approximately40l m which coincide with the length of nanotubes in the attached VACNT layer.The material remains packed due to the strong van der Waals forces that exist between the tubes in the bundles.The same sample probe was then rotated180°for further SEM scanning and again,layers of nanotu-bes were present covering the catalytic substrate surface.High-resolution TEM was employed to study the inner morphology of the aligned nanotubes.For TEM studies,nanotubes formed on the catalytic support were carefully peeled offand transferred to acetone.After sonication,the drop of the suspension was placed ona Fig.9.Higher resolution SEM of a micro area in Fig.8.Fig.10.Low and high resolution(inserts)SEM images of scanned probe surface showing uniformity of VACNT coating layer.604W.Merchan-Merchan et al./Carbon42(2004)599–608copper-substrate/carbon-film of microscope specimen grid and dried.TEM images of nanotubes synthesized at Z ¼8:5mm are shown in Figs.11and 12.Very low resolution TEM imaging on the transferred material shows the presence of high-density closely packed nanotube bundles (Fig.11).It appears that even after sonication,closely packed bundles of nanotubes remain intact.Fig.11shows typical bundles of nanotubes ob-served by TEM in a sample grid,here the bundle pos-sesses a length and width of 31and 10l m,respectively.In a closer zoom on the nanotube bundle,no material other than tubular structures are present,Fig.12.From a number of micrographs the average diameter of the cylindrical multi-shells was measured,a monodisperse diameter distribution averaging 38nm was obtained.The application of high-resolution TEM imaging on these carbon nanotubes reveals a texture of well-aligned and highly graphitized concentric graphitic cylinders.The average interplanar distance of the concentric cylindrical graphene sheets was measured to be 0.34nm,as indicated in Fig.13.The layer planes appear to be perfectly parallel to the central tube axis.The experimental results show the strong influence of the electric field on alignment,size distribution,internal structure,and growth rate of carbon nanotubes.The mechanism of alignment is widely discussed in the lit-erature [26–28].As an example,employing a plasma discharge Merkulov and coworkers [27]successfully demonstrated,that the direction of the electric field lines determines the orientation of the carbon nanotubes and nanofibers;electric field alignment of single-walled na-notubes were considered by Zhang et al.[28].Consider the alignment mechanism of a single CNT growing on a metal catalytic support at the presence of the electric field E (Fig.14a).At the synthesis tempera-ture,both metallic and semiconducting CNT can be treated as good conductors.Consequently,the nanotube can be represented as a conductive cylinder of length L and radius R in contact with a conducting plane with a surface charge density r .For a cylinder of finite length,an electric field generated by the plate in the normal direction will induce a bound charge at the nanotube end to screen the external field over the nanotube vol-ume.The field-screening area of the single nanotube can be estimated as L 2.Accordingly,the total charge of the nanotube is q %r L 2¼e 0L 2E .The total electrostatic force acting on the nanotube is F %qE ¼e 0L 2E 2.The nanotube inclined at an angle h with respect to the normal direction will have axial F a and tangential F t components of the force F inducing respectively axial stress and alignment force.For a small deviation of the nanotube from the normal direction d ,the alignment force F t ¼F sin h ¼e 0LE 2d is proportional to the nano-tube length.The more rigorous solution of the electro-static problem [29]gives an expression for the total electrostatic force acting on the cylinder as F ¼pe 0E 2L 2=ðln ð2R =L ÞÀ3=2Þ,showing that the acting force is a weak function of R .The solution also shows the pres-ence of finite surface charges over the lower part of the nanotube with the charge density distribution close tolinear.Fig.11.Low resolution TEM image of material removed from the catalytic support.The material is produced using FPM at Z ¼8:5mm.Fig.12.TEM image of metal catalyzed carbon nanotubes from a micro area shown in Fig.11.Fig.13.High-resolution TEM image of the wall of a tubule produced in floating potential mode.W.Merchan-Merchan et al./Carbon 42(2004)599–608605For CNTs growing on a conductive substrate as a vertically aligned array with a tube-to-tube separation distance of D,the externalfield is reduced by mutual screening(Fig.14b).The depth of thefield penetration can be also estimated as D.Similar consideration for the charge induced on the nanotube tip provides estimation of the charge value q%e0D2E and the electrostatic force acting on the single tube F%qE¼e0D2E2.Conse-quently,the aligning force F t¼e0DE2d is linearly pro-portional to the nanotube separation D and the deviation from the vertical position d.The surface charges of the same polarity induced near the nanotube ends ensure their separation and alignment.These repulsive forces overcome van der Waals attraction forces.It is worth mentioning that the repulsive align-ment is quiet different from the alignment induced by van der Waals interactions.The latter mechanism leads to the formation of nanotube bundles which in turn reduces the thermal randomization of the CNT growth. However,a direction of the bundle growth remains arbitrary.Overall,the electricfields near the tips of growing nanotubes can be extremely high.Even applied poten-tials as small as few tens of millivolts can develop an electricfield exceeding1000V/cm at the characteristic nanotube diameter.The experimentally measuredfield enhancement factors are reaching800for multi-walled nanotubes[30]and3000for single-walled nanotubes [31].The enhancement of the electricfield at the tip of closed conducting nanotube was calculated by Maiti et al.[32];the resulting force estimated from the axial stress is in a good agreement with Taylor solution[29].The important aspect of the aligned growth is that the constant orientation of catalytic particles at the tips of the growing nanotubes is preserved by the electricfield. The non-symmetric catalytic particle is polarized in the electricfield,and the induced dipole moment tends to be aligned along the electricfield lines.The formation of helical and spiral nanotubes requires variation of the particle orientation.As a result the constant orientation of the catalytic particle stabilizes the linear CNT struc-tures.The high concentration of the electric charges near the nanotube end also induces axial stress between the growing nanotube body and conductive catalyst parti-cle.This can lead to the increase of the CNT growth rate.For clarity,we further consider a nanotube with a catalytic particle at the tip.The tip-growth proceeds via several important stages,including decomposition of flame hydrocarbons on catalytic surface,diffusion of carbon through and over metal particulate,formation of ordered CNT structure on the opposite side of the catalytic particle.The solution of electrostatic problem suggests strong concentration of charge and electricfield near the tip of the growing nanotube,namely on the catalytic particle attached to the nanotube end.The stress introduced between the particle and the CNT body may be estimated as r s¼F=p R2.While the force F has only weak dependence on R,the stress shows dra-matic increase with the reduction of contact area and can reach a critical value rÃscorresponding to the pos-sible particle separation.This can discriminate the growth of nanotubes with R<RÃ.The moderate value of stress,however,can increase the transformation of carbon to the CNT structure.As a result the concen-tration of free carbon on the CNT side will decrease leading to the faster carbon diffusion and the CNT growth will proceed at the increased rate.The sensitiv-ity of the stress to the nanotube diameter is a possi-ble explanation for the narrow diameter distribution which is characteristic for the synthesized VACNT arrays.Finally,the electricfield can influence the transport of charged particles inflames that include ions and charged soot particles.In this way,soot entrapment in the growing layer can be controlled by the electricfield.The effect of the electricfield control using various electricfield amplitudes and geometries is an interesting subject for the future studies.The selectedflame con-figuration allows ease of application of internal(probe generated)and external(generated by external elec-trodes)electricfields,as well as time-dependent elec-tromagneticfields.4.ConclusionsThe electricfield controlled synthesis of carbon nanomaterials on a Ni-based catalytic support posi-tioned at the fuel side of the opposedflow oxy-flame is studied experimentally.Theflame environment is formed by fuel and oxidizer with compositions of CH4+4%C2H2and50%O2+50%N2,respectively.Car-bon nanomaterials formed on the probe surface are comparatively analyzed for two characteristic opera-tional modes:a grounded probe mode and afloating probe mode.In a grounded mode a number of various carbon nanostructures are formed depending on the probe location inflame.Observed nanoforms include multi-walled carbon nanotubes(MWNTs),MWNT bundles, helically coiled tubular nanofibers,and ribbon-like coiled nanofibers with rectangular cross-section.The presence of various carbon nanoforms is attributed to the space variation offlame parameters,namelyflame temperature and concentration of chemical species.It is found that a presence of an electric potential (floating mode operation)provides an ability to control nanostructure morphology and synthesis rate.A thick layer(35–40l m)of vertically aligned carbon nanotubes (VACNTs)is found to be formed on the probe surface in thefloating potential mode.This layer is character-ized by high uniformity and narrow distribution of nanotube diameters.Electricfield control method demonstrates stabilization of the linear nanotube structure in a wideflame region.The growth rate re-mains dependent onflame location but the strong enhancement effect is observed with application of electricfield.AcknowledgementsThis work was supported by the National Science Foundation grant CTS-0304528.This work was par-tially supported by the Air Liquide Corp.under an Unrestricted Laboratory Development Grant.The au-thors extend special thank to Dr.Alan Nicholls,Mr. John Roth,Ms.Linda J.Juarez,and Ms.Kristina Jarosius from the UIC Research Resource Center for day-to-day assistance in SEM and TEM studies, encouragement and helpful discussions.The authors would also like to thank Mr.Attilio Milanese for his input and helpful discussions.References[1]Iijima S.Helical microtubules of graphitic carbon.Nature1991;354:56–8.[2]Kiang CH,Goddard WA,Beyers R,Salem JR,Bethune DS.Catalytic synthesis of single-layer carbon nanotubes with wide range of diameters.J Phys Chem1994;98:6612–8.[3]Shi Z,Lian Y,Zhou X,Gu Z,Zhang V,Iijima S,et al.Massproduction of single-wall carbon nanotubes by arc discharge method.Carbon1999;37:1449–53.[4]Guo T,Nikolaev P,Thess A,Colbert DT,Smalley RE.Catalyticgrowth of single-walled nanotubes by laser vaporization.Chem Phys Lett1995;243:49–54.[5]Andrews R,Jacques D,Rao AM,Derbyshire F,Qian D,Fan X,et al.Continuous production of aligned carbon nanotubes:a step closer to commercial realization.Chem Phys Lett1999;303:467–74.[6]Yuan L,Saito K,Pan C,Williams FA,Gordon AS.Nanotubesfrom methaneflames.Chem Phys Lett2001;340:237–41.[7]Yuan L,Saito K,Hu W,Chen Z.Ethyleneflame synthesis of well-aligned multi-walled carbon nanotubes.Chem Phys Lett2001;346: 23–5.[8]Vander Wal RL.Flame synthesis of substrate-supported metal-catalyzed carbon nanotubes.Chem Phys Lett2000;324:217–23.[9]Vander Wal RL,Lee JH.Ferrocene as a precursor reagent formetal-catalyzed carbon nanotubes:competing effbust Flame2002;130:27–36.[10]Vander Wal RL.Fe-catalyzed single-walled carbon nanotubesynthesis within aflame bust Flame2002;130: 37–47.[11]Vander Wal RL,Hall LJ,Berger GM.Optimization offlamesynthesis for carbon nanotubes using supported catalyst.J Phys Chem2002;B106:13122–32.[12]Height MJ,Howard JB,Tester JW.Synthesis of single-walledcarbon nanotubes in a premixed acetylene/oxygen/argonflame.Proceedings of the Third Joint Meeting of the U.S.Section of the Combustion Institute(2003).[13]Merchan-Merchan W,Saveliev A,Kennedy LA,Fridman A.Formation of carbon nanotubes in counter-flow,oxy-methane diffusionflames without catalyst.Chem Phys Lett2002;354:20–4.[14]Saveliev A,Merchan-Merchan W,Kennedy LA.Metal catalyzedsynthesis of carbon nanotubes in an opposedflow methane oxygenflbust Flame2003;135:27–33.[15]Kuzuya C,Kohda M,Hishikawa Y,Motojima S.Preparation ofcarbon micro-coils with the application of outer and inner electromagneticfields and bias voltage.Carbon2002;40:1991–2001.[16]Avigal Y,Kalish R.Growth of aligned carbon nanotubes bybiasing during growth.Appl Phys Lett2001;78(16):2291–3. [17]Srivastava AK,Srivastava ON.Curious aligned of carbon nano-tubes under applied electricfield.Carbon2001;39:201–6. [18]Lee KH,Cho JM,Sigmund W.Control of growth orientation forcarbon nanotubes.Appl Phys Lett2003;82(3):448–50.[19]Ant U,Yiming L,Hongjie D.Electric-field-aligned growth ofsingle-walled carbon nanotubes on surfaces.Appl Phys Lett 2002;81(18):3464–6.[20]Colbert DT,Smalley RE.Electric effects in nanotube growth.Carbon1995;33(7):921–4.[21]Srivastava A,Srivastava AK,Srivastava ON.Effect of externalelectricfield on the growth of nanotubules.Appl Phys Lett1998;72(14):1685–7.W.Merchan-Merchan et al./Carbon42(2004)599–608607。