composites: part b - icce-nano.orgsite.icce-nano.org/clients/iccenanoorg/2013 controllable...

Composites: Part B xxx (2013) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Composites: Part B

journal homepage: www.elsevier .com/locate /composi tesb

Controllable synthesis of carbon coils and growth mechanismfor twinning double-helix catalyzed by Ni nanoparticle

1359-8368/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.compositesb.2013.06.010

⇑ Corresponding author. Tel./fax: +86 028 87600454.E-mail address: [email protected] (Z. Zhou).

Please cite this article in press as: Jian X et al. Controllable synthesis of carbon coils and growth mechanism for twinning double-helix catalyzenanoparticle. Composites: Part B (2013), http://dx.doi.org/10.1016/j.compositesb.2013.06.010

Xian Jian a, Dingchuan Wang a, Hongyang Liu b, Man Jiang a, Zuowan Zhou a,⇑, Jun Lu a, Xiaoling Xu a,Yong Wang a, Li Wang c, Zizheng Gong c, Mingli Yang d, Jihua Gou e, David Hui f

a Key Laboratory of Advanced Technologies of Materials (Ministry of Education), School of Materials Science and Engineering, Southwest Jiaotong University, Chengdu 610031, Chinab Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, Chinac China Academy of Space Technology, Beijing 100094, Chinad Institute of Atomic and Molecular Physics, Sichuan University, Chengdu 610065, Chinae Department of Mechanical, Materials and Aerospace Engineering, University of Central Florida, Orlando, FL 32816, USAf Department of Mechanical Engineering, University of New Orleans, New Orleans, LA 70148, USA

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

Article history:Received 25 March 2013Accepted 10 June 2013Available online xxxx

Keywords:A. Carbon fibreA. Nano-structuresE. Chemical vapour deposition (CVD)E. Heat treatment

A simple approach using �90 nm Ni nanoparticles to tune the growth of carbon fibers is developed basingon the bottom-up regulation in this study. By adjusting the preparing temperature and atmospheric com-position, the nickel-containing catalysts with specific sizes and shapes are alternately prepared from90 nm to 500 nm. The straight carbon nanofibers and three kinds of carbon coils (single-helix carbonnanocoils, single-helix carbon microcoils and twinning double-helix carbon microcoils) with the coildiameter ranging from 150 nm to 3 lm are achieved by using the as-prepared catalyst. The relationshipbetween the carbon coil and catalyst particle, and the growth mechanism for different kinds of helices arediscussed in details. The results suggest that catalytic anisotropy come from growing tendency and rateof carbon deposition on the facets, edges and vertices of catalyst grain. The twinning structure exists ineach fiber of the double-helix carbon microcoils regardless of circular or flat fiber, which is separated bythe tip of the catalyst particle due to the different rate of carbon deposition at edge and vertex.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction So far, two main strategies have been developed to synthesize

Since the discovery of carbon helix, the synthesis of carbon coilsinvolving fibers or tubes with helical diameter sized from 16 nm to30 lm [1,2] has attracted significant attention, not only for funda-mental scientific [3,4] interests, but also for their potential applica-tions [5–7]. Though carbon nanofiber or tubes can strength somematrix [8–12] such as multiscale syntactic foams [9], epoxy[13,14], plastic laminates [15], polyurethane [16] and phenoliccomposites [17], helical carbon fibers/tubes are also a goodstrengthening fibers in composite materials [18–20]. Besides, car-bon coils exhibit peculiar morphology-dependent electrical,mechanical, magnetic and chemical properties, compared withthose of their bulk materials. However, for applications in elec-tro-mechanical sensors, the morphological uniformity is criticalfor achieving a stable piezoresistive behavior of carbon nano-springs. Besides, morphology control of the helical structure hassignificant influence on chirality, which is essential for the electro-magnetic wave absorbers [21]. Thus, the rational synthesis of spe-cial coil with high purity is of key importance for technologicalapplications.

helical structures of carbon coils. one is the ‘‘catalyst-component’’adjustment depending on the catalytic activity, and the other is the‘‘growth-condition’’ method being governed by the chemical kinet-ics. Several helical carbon structures (double/single-helix, micro/nano-coils, helical fibers/tubes, etc.) were obtained by adjustingthe component of metallic catalysts (Fe, Co, Ni, Cu, In or their al-loys) [22–24]. The advantage of the catalytic activity methods isthe production with high pertinent characteristic, but the proce-dure of preparing the special catalyst seems quite complicated.The chemical kinetic methods can be used to synthesize shape-controlled carbon coils by choosing different carbon sources [24],adding the substrate [25,26] or introducing magnetic/micro-wavefield [27,28], but the equipment is usually fussy and the procedureis multi-step. In order to simplify the process to obtain special car-bon coils, it is necessary to explore an effective method to combinethe catalytic activity with chemical kinetics.

Nickel is one of the typical catalysts for growing the helical car-bon fibers (CFs). For example, pure Ni powder sized of �5 lm [29]or 500 nm [30], activated carbon nanotubes supported Ni �0.8 lm[31], nickel foam [32] and Ni plate [33] were used to preparehelical CFs with an approximate fiber diameter of 500 nm. It canbe noticed that some carbon coils with similar sizes and diameterswere prepared using Ni catalyst with different particle size.

d by Ni

http://dx.doi.org/10.1016/j.compositesb.2013.06.010

mailto:[email protected]


http://www.sciencedirect.com/science/journal/13598368

http://www.elsevier.com/locate/compositesb


2 X. Jian et al. / Composites: Part B xxx (2013) xxx–xxx

Nevertheless, few literatures provided reasonable explanations tothe relationship between the initial Ni catalyst size and the diam-eter of the CFs. In addition, it has been proposed that the drivingforce for the coiling of vapor grown CFs is the presence of catalyticanisotropy among the crystal facets of the catalyst grain. However,the direct evidence of this presumption has not been reported yet.

In this work, we reported the size control of Ni catalyst bychemical reduction and the rational synthesis of several differentcarbon coils by catalytic chemical vapor deposition (CCVD) methodusing acetylene as carbon source. The as-synthesized CFs could becontrolled into linear carbon nanofibers (CNFs) and carbon coilsincluding single-helix carbon nanocoils (SH-CNCs), single-helixcarbon microcoils (SH-CMCs) and twinning double-helix carbonmicrocoils (TDH-CMCs). In addition, the existence of catalyticanisotropy was experimentally approved and the expoundedgrowth mechanism for these carbon species was preliminarilysuggested.

2. Experimental section

For the synthesis of Ni catalyst, the mixed solution containing20 mL of 0.8 mol/L (M) sodium hydroxide solution, 20 mL of 6 Mhydrazine hydrate aqueous solution and 1 g polyvinylpyrrolidone(PVP-K30) was firstly prepared. Then 20 mL of 1 M nickel sulfateaqueous solution was slowly added to the mixed solution. Finally,Ni nanoparticles were prepared by chemical reduction with hydra-zine hydrate in distilled water at 70 �C with magnetic stirring for1.5 h. Ni nanoparticles, namely the black precipitates, were sepa-rated from the mother liquor by magnetic separation, and washedseveral times with distilled water and acetone, respectively, untilthe pH was 7. The obtained Ni nanoparticles, grey–black powder,were dried in vacuum at 25 �C.

The Ni catalyst (30 mg/cm2) supported by graphite plate wasplaced in the central part of a horizontal reaction tube (quartz,24 mm i.d.), which was heated from the outside by an electricallyheated horizontal furnace. Then linear and helical CFs were pre-pared on the graphite plate using a gas mixture of C2H2, H2 andN2 containing a small amount of thiophene as catalytic auxiliary.The furnace was firstly flushed until 600 �C at the rate of 5 �C/min under nitrogen, and it was unceasingly flushed from 600 to750 �C at 3 �C/min under hydrogen. Acetylene along with thio-phene vapor was allowed to slowly enter the furnace at a rate of20 mL/min. Thiophene was introduced at a rate of about 80 mg/min by bubbling acetylene through a vessel containing thiopheneat 40 �C. Meanwhile, the flow rate of H2 was fixed at 60 mL/min.Nitrogen was used as the carrier gas at a given rate from 100 to140 mL/min. The fluffy products of various CFs including linearCNFs, SH-CNCs, SH-CMCs and TDH-CMCs were synthesized with-out further treatment.

The crystal structure of the catalyst particles and the helical CFswas investigated using X-ray diffraction (XRD, Panalytical X’PertPRO diffractometer with Ni-filtered, the Netherlands). The sizeand morphology analyses of the nickel particles and the CFs wereperformed using scanning electron microscope (SEM, Fei, Quanta200 and Inspect-F) with an accelerating voltage of 20.0 kV andtransmission electron microscopy (TEM, H-700H) at an accelerat-ing voltage of 160 kV.

3. Results and discussion

3.1. Preparation and characterization of Ni catalyst

Generally, catalyst plays an important role in the control ofhelical carbon materials with desired category and morphology.For example, three-dimensional spring-like carbon nanocoils are

Please cite this article in press as: Jian X et al. Controllable synthesis of carbonanoparticle. Composites: Part B (2013), http://dx.doi.org/10.1016/j.composite

obtained with high purity by the catalytic pyrolysis of acetyleneat 750–790 �C using a Fe-based catalyst, and the nanocoils have ananotube of 10–20 nm [34]. Besides, the carbon nanocoils with coildiameters of 50–450 nm can be obtained using sputtered thin filmsof Au and Au/Ni as catalysts [27]. Considering that Ni particle is oneof the typical catalysts for the mass production of carbon micro-coils (CMCs), we prepared high-purity and monodispersed Ninanoparticles by the reduction of a nickel salt with hydrazine hy-drate employing the surfactant of PVP to prevent agglomerationof the nanoparticles. The effects of reaction temperature (Fig. S1in Supporting information), NaOH concentration (Fig. S2) and reac-tion time for the preparation of nickel powders were systemati-cally investigated. Spherical and relatively uniform Ni particles(Fig. 1a) were obtained under the optimal condition that the molarconcentration of NaOH solution was 0.8 M at 70 �C with magneticstirring for 1.5 h. Particle-size distribution (Fig. 1c) of the obtainedmonodispersed Ni nanoparticles was determined by software ofNano Measurer [35] program based on the corresponding SEMimages. The average particle size was 90.12 nm with a relativelynarrow distribution. The TEM image of the as-prepared Ni nano-particles placed in ethyl alcohol is shown in Fig. 1b. It could be seenthat sometimes Ni nanoparticles adhered to each other forminginto some larger Ni grains of about 150 nm, which might be dueto the ferromagnetism of them. The XRD pattern of the nanoparti-cles, shown in Fig. 1d, revealed an fcc Ni crystalline structure. In theXRD profile, the three characteristic diffraction peaks correspondingto the (111), (200) and (220) diffraction planes of only cubic Niphase over 40� were clearly observed, which was well in agreementwith the standard nickel diffraction pattern (ICDD, PDF file No. 01-070-1849). The crystallite size of Ni for the most intense peak(111) plane was determined from the X-ray diffraction data usingthe Debye–Scherrer formula, and the average value was about5.07 nm when the concentration of NaOH was 0.8 M. In addition, itcould be found that the particle size calculated from the XRD patternwas rather smaller than that determined from the SEM image. Theresults suggested that the spherical nickel particles contained anumber of ultra-fine crystals of about 5 nm, which agreed with theTEM observation of Ni particles’ morphology.

3.2. Rational synthesis of CFs

The CMCs are usually prepared using typical nickel catalystsuch as Ni plate [33] and Ni powder with particle sizes from 0.5to 5 lm [29–31]. Besides, Ni particles with a particle size below50 nm [36] or about 500 nm [37] have also been chosen for the cat-alyst to prepare CMCs. It is can been seen that the Ni catalyst are ofvarious kinds and their original size does not determine the fiberdiameter of the corresponding carbon coil. The cause of this casemight be that it was difficult to obtain carbon nanocoils on Ninanoparticles below 100 nm. Similarly, we obtained linear CNFswithout helical shape using the as-prepared Ni nanoparticles asthe catalyst under the reaction conditions that the bath tempera-ture of thiophene was 40 �C and the flow rate of C2H2, H2 and N2

was 20, 60 and 100 mL/min, respectively, at the reaction tempera-ture of 660 �C for 3 min. A typical SEM image of the linear CNFs(Fig. 2) displayed an almost uniform fiber diameter of about100 nm. This was due to the presence of an almost uniform sizeof catalyst particles and confirmed that the linear CNFs diametersdepended on the size of the Ni particles in chief. The linear CNFswere straight or complexly curved, and interlaced with each other.This might be associated with the real-time shape of crystallinemetallic Ni particles with symmetric or asymmetric shape duringthe growth of the CFs [38]. Since spherical catalyst particle hassymmetric planes on condition that they pass through any diame-ter of the Ni nanoparticles, and the anisotropic behaviors of thecatalyst do not show. Then straight CFs could be obtained using

n coils and growth mechanism for twinning double-helix catalyzed by Nisb.2013.06.010


Fig. 1. (a) SEM and (b) TEM images, (c) size distribution and (d) XRD pattern of nickel particles obtained under the optimal condition that the concentration of NaOH solutionwas 0.8 M at 70 �C with magnetic stirring for 1.5 h.

Fig. 2. SEM image of linear CNFs synthesized by CCVD route at 660 �C for 3 minwhen the gas flow rate of C2H2, H2 and N2 was 20, 60 and 140 mL/min, respectively.

X. Jian et al. / Composites: Part B xxx (2013) xxx–xxx 3

fine Ni particles with spherical shape, and curved CFs formed onthe asymmetric spheroidal Ni particles. From the early study[39], linear CFs were synthesized using the catalyst containing Coor Fe alloy. This is a supplementary method for the synthesis of lin-ear CFs, differing from the noodle-like twin fibers obtained at700 �C by Chen and Motojima [40], which might be potential andeffectual to prepare CFs networks.

As the basic unit of bulk materials, Ni nanoparticle has the advan-tage of forming the required Ni catalyst by the gas-adjusting ap-proach. Three kinds of carbon coils, namely, SH-CNCs, SH-CMCsand TDH-CMCs, were prepared using the as-prepared Ni nanoparti-cles by adjusting the gas flow rate of N2 at 140, 120 and 100 mL/minrespectively. The other experimental parameters were unchanged.


The bath temperature of thiophene was 40 �C, the flow rate ofC2H2 and H2 was 20 and 60 mL/min respectively, and the reactionwas carried out at 750 �C for 120 min. SEM images of the three kindsof carbon coils are shown in Fig. 3. The SH-CNCs (Fig. 3a) mostly havethe fiber diameter of about 150 nm and the coil diameter of about200 nm, and the coil pitch ranging from 200 to 400 nm. The mor-phology of SH-CNCs was analogous to the CFs obtained from theinteraction between the copper–nickel (7:3) and the ethylene/hydrogen (4:1) at 600 �C by Kim et al. [41] Compared with the linearfibers, the fiber diameters of SH-CNCs were generally larger. Theprobable reason for this phenomenon was that the Ni particlesunderwent a growing-up experience by agglomerating and/orrecrystallizing as the reaction temperature increased from 660 to750 �C. The fiber diameters of CFs depend on the size of Ni particles,thus the SH-CNCs have a little thicker fibers than linear CFs atrelatively higher temperature of 750 �C.

When the flow rate of N2 reduced to 120 mL/min, the SH-CMCs(Fig. 3b) could be observed and they had a fiber diameter of500–600 nm and a coil pitch of 2–3 lm. Different from conven-tional SH-CNCs, the coil pitch was generally quite large while thecoils still kept the regular helical forms. Besides, the SH-CMCs con-tain an abnormal interlayer of a wavy morphology in the middlepart of the fibers, which is similar to the spring-like carbon coilswith wave-form laces obtained by changing the composition andthe solid state conditions of the catalysts (71Fe–18Cr–8Ni–3Mo)supported on ceramic nano-powders [34]. Then, it is suggestedthat the SH-CMCs can be obtained just by controlling the flow rateof the carrier gas at 120 mL/min in this case.

Conventional DH-CMCs are usually prepared by the catalyticpyrolysis of acetylene using Ni powder of several micrometers indiameter. For example, Kawaguchi prepared double helix regularcarbon microcoils with the catalyst of Ni powder having an averagediameter of 5 lm [42]. Here we designed the process of getting thelarger Ni particles by the bottom-up method, which is useful for



Fig. 3. SEM images of three kinds of carbon coils: (a) SH-CNCs, (b) SH-CMCs, and (c)typical DH-CMCs.


precise control the helical structure of carbon coils. Before thegrowth of carbon coils, the as-prepared Ni nanoparticles wereheated under H2 at 700 �C for 1 h to make the Ni nanoparticleswith melting point less than 750 �C be melt and annexed by lagerNi particles until the state of thermodynamic equilibrium. The uni-form Ni particles with diameter of several micrometers formedafter the above heat treatment. Typical DH-CMCs (Fig. 3c) were ob-tained when the flow rate of N2 was 100 mL/min, and they had afiber diameter of 500–600 nm and a helical pitch of 1–3 lm. Fromthese observations, it is necessary to explain why the three kinds ofhelical carbon species differ in filament diameter and helical pitch.


The growth mechanism for the three kinds of helical carbon mate-rials will be discussed in the next section.

3.3. Analysis and mechanism for carbon coils

Amelinckx et al. proposed a formation mechanism for catalyti-cally growing of helix-shaped graphite nanotube [43]. They re-ported the geometrical parameters of the resulting helix could beexpressed in terms of the extreme values VM (maximum velocity),Vm (minimum velocity) of the extrusion velocity and of the inclina-tion angle. The growth mechanism could also be sufficiently gen-eral to account for the variability in pitch of helices. About theCFs, the size of its diameter depended on the particle size of thecatalyst, and the pitch and diameter of the coil were decided bythe extrusion speed of VM, Vm and the inclination angle. Then itcan be speculated that the direct cause for the discrepancies ofthe three kinds of helical fibers is the differences of the size andthe anisotropy catalysis of the catalysts. Three kinds of carbon coilscan be rationally synthesized by adjusting the flow rate of N2 dem-onstrating that the flow rate makes huge effects on the catalyticperformance of Ni particles.

Nitrogen flow rate seems to be the only changed parameter dur-ing synthesizing, which results in different carbon coils under thecorresponding condition. However, the change of N2 flow rate isnot the essential cause, because N2 is non-active and does not takepart in the chemical reaction. Moreover, the N2 flow rate doesinfluence the quantity of carbon source and H2 for a certain periodof time. From the difference of the diameters of helical CFs, it canbe reasonably speculated that the fine Ni particles happen to formlarger agglomerates. For example, as the fine Ni particles of�100 nm have the volume of V0, they can grow up into larger par-ticles sized of 150 nm (�3.375V0) or 500 nm (�125V0). Though itcan be noticed that hydrogen plays an essential role in the forma-tion of helical CFs, it is difficult to in situ observe the changing pro-cess of the catalyst particles and the forming course of the carboncoils. To elucidate the mystery, the effects of hydrogen on catalystparticles were investigated and the results displayed that the fineNi particles changed not only their particles sizes, but also the ex-posed crystal facets and the topography after heat treatment underH2 at 750 �C for 1 h as shown in Fig. 4. The fine Ni particles grewinto larger particles by recrystallization. Meanwhile, some concav-ity formed on the surface of the aggregative Ni particles high-lighted by the small ellipse in Fig. 4a, and some of them hadpolyhedral shape as shown in Fig. 4b. These changes might be as-cribed to two reasons. One is the changes of surface energy underH2. From the energy’s point of view, the Ni facets with lower sur-face energy tend to expose and wrap around the catalyst particle.The other reason might be the existence of the reaction betweenH2 and Ni, which causes forming the new species of NiHx. Thenthe role of H2 during the growth of carbon coils can be summedup in terms of three major points. Firstly, the flowing H2 adjuststhe rate of the reaction of C2H2 = 2C + H2, by diluting the concen-tration of the carbon source gas, and accordingly affects the growthrate of the CFs. Secondly, hydrogen prevents the catalyst from poi-soning, and reactivate the poisoned catalyst particles according tothis work. The main reaction are shown in Fig. 5, where NixCy issome poisoned catalyst particles and the NimCn is the activated cat-alyst such as Ni3C [44]. Thirdly, hydrogen induces reconstruction ofthe catalyst particles and control the exposed metal faces [45],which influences the size, the exposed crystal facet and topographyof the catalyst particles [46] and the presence of H2 during synthe-sis was found to have a positive effect on CCF formation at givencondition [47].

As the reaction temperature increases from 600 to 750 �C underH2, the Ni particles grow up to larger ones until they reach a stabi-lized state. Accordingly, the variable size and shape of the catalyst



Fig. 4. SEM images of Ni particles after heat treatment under H2 at 750 �C for 1 h. (a) Some concavity formed on the surface of the aggregative Ni particles and (b) the districtof the Ni particle with polyhedral shape.

Fig. 5. The related reactions during the poisoned catalyst particles changing intoactivated ones under the effects of hydrogen. where NixCy is some poisoned catalystparticles, the NimCn is the activated catalyst such as Ni3C.


particles from spherical particle (�100 nm) or special particle(�150 nm) to polyhedral micro-particle (�500 nm) could be ad-justed by N2 flow. Thus the three kinds of helical CFs can be ratio-nally prepared with the same Ni nanoparticles just by controllingthe reaction temperature and the N2 flow. Fig. 6 depicts the forma-tion of the linear CFs, SH-CNCs, SH-CMCs with lace and TDH-CMCs,and the corresponding Ni particles. To lower their surface energy,the small-sized nanoparticles tend to adopt spherical structuresand exhibit catalytic isotropy, resulting in the formation of linearCFs. As the Ni particles growing up, their sizes become larger andtheir shapes turn to be more regular. Then various carbon coilsformed on the corresponding nanoparticles of anisotropic catalysis.Apart from the familiar morphologies of SH-CNCs and TDH-CMCs,sometimes, SH-CMCs with lace grow on the asymmetric catalystgrain such as the one composed of hemisphere and a hexahedron.The suggestion will be verified in the future study.

Fig. 6. Diagrams of the growth of four kinds of CFs on Ni particles with different size and sregular polyhedron (�500 nm), and (d) regular polyhedron (�500 nm).


It was reported that the driving force of curling and spiral coil-ing of the carbon coils was the catalytic anisotropies between cat-alyst crystal faces [30]. However, the catalytic anisotropy of thecatalyst grain is not well defined, and the origin of the catalyticanisotropy is not well described yet. Boskovic et al. reported thatcarbon flowed to the nanofiber on the surface of the catalyst parti-cle rather than diffuse through the particle [48]. Then the catalyticanisotropy mainly depends on the catalyst surface, leading to for-mation of various carbon coils with different cross section. In addi-tion, Boskovic et al. observed that graphene layers were orientedparallel to the facets of the catalyst particle, while the middle partof the CNFs was composed of amorphous carbon corresponding tothe edge-zone of the catalyst particle. Shen et al. proposed a mech-anism for the regular coiled carbon nanotubes [49], basing on theirmechanism analysis, a suggestion is put forward that the catalyticanisotropy is the discrepancy of forming carbon atoms on the sur-face of the catalyst, and the catalytic anisotropy origins from theatomic density of the catalyst surface and reflects in the form offacets, edges and vertices of the catalyst grain. In general, the for-mation of carbon coils is often related to the existence of polyhe-dral catalyst grain no matter microcoil [50] or nanocoil [51]. Inorder to analyze the polyhedral Ni compound particles, three kindsof typical regular polyhedrons, namely octahedron, hexahedronand dodecahedron, are sketched in Fig. 7a, b, and c, respectively.The cuboid has the sectional plane of symmetry more or less. Forinstance, the regular hexahedron divides into two comparativeparts by the diagonal plane as presented in the shadow tetragonof Fig. 7b. Two main CFs (X and Y in Fig. 7e) grow bi-directionally

hape: (a) small sphere (�80 nm), (b) small regular polyhedron (�150 nm), (c) quasi-



Fig. 7. Diagram of Ni particles with shapes of (a) octahedron, (b) hexahedron and (c) dodecahedron, and the DH-CMCs having cross sections of circle, rectangle and pentagon.

Fig. 8. SEM images of two kinds of TDH-CMCs with (a and b) circular and (c and d) flat cross sections.


from the symmetric catalyst grain and then entwine with eachother to form helical pattern of the conventional DH-CMCs. Simi-larly, two twinning fibers developed with the catalysis of a quarterof the catalyst particles. The cross section of the DH-CMCs formsaccording to the topography of the corresponding Ni particles. Sothe circular (Fig. 7d) and rectangular (Fig. 7e) cross sections ofthe carbon coils usually exist and are reported with adjusting somereaction condition [52]. Besides, the carbon coils having pentago-nal fiber are presented in Fig. 7f. This is in good agreement with


the experimental observation of the ruptured tip and of variousgraphite coils [53].

Kawaguchi et al. suggested that each crystal plane of a Ni seedhad different catalytic ability for the growth of coiled fibers. TwoCFs began to grow in opposite directions from each pair of planes(X and Y) on the Ni catalytic surface [42]. In the low magnificationimage (Fig. 8a) of typical DH-CMCs, they seemed to be made up oftwo circular fibers. The as-prepared regular TDH-CMCs could ex-tend to 3–8 times of its original length as shown in Figs. S4 and




8b. The extension characteristic was better than conventional DH-CMCs reported by Motojima et al. [33], but smaller than the super-elastic carbon microcoils [54]. Interestingly, we found the TDH-CMCs containing four small CFs as shown in Fig. 8b. The thick fiberY consists of two fibers (Y1 and Y2) as shown in the high magnifica-tion image. Besides, some obscure scoop channel was also found inflat carbon filaments (Figs. 8c, d and S6). Thus, the growth mecha-nism proposed by Kawaguchi et al. [42] could not be applied to thegrowth of twinning double helical fibers. We then proposed ananisotropically carbon deposition on the facets, edges and verticesof catalytic polyhedron as the growth mechanism for carbon coils.The number of the branching fiber is mostly two [45] and three[50], sometimes is four and six [53], which determined by theanisotropical level of the catalyst grain. The catalyst particles hadthe sectional plane of symmetry in terms of catalytic ability forthe growth of the fibers. Take the helical flat CFs as an example,the fiber M and fiber N grew on each comparative part of the cat-alyst particle. Due to the different forming rate of carbon atomson the facet, edge and vertex of catalyst grain, two pieces of CFsgrew more quickly on the crystal plane than at the edge or vertex,and the fibers twisted with each other leading to forming two coils.The diameter of the coil depended on the diversity of each compar-ative part of the catalyst grain. Similarly, fibers M1 and M2 generateon quarter part of catalyst grain and they lightly separate fromeach other due to the smaller different rate of carbon depositionat edge and vertex.

4. Conclusion

Linear CFs and three kinds of carbon coils were rationally pre-pared by gas-adjusting method using regular spherical Ni nanopar-ticles (�90 nm) as catalyst obtained by liquid phase reduction withhydrazine hydrate. The nano-scaled linear CFs often grow onspherical Ni particle in the form of straight and curved shape. Threekinds of carbon coils, namely, SH-CNCs, SH-CMCs and TDH-CMCsare rationally prepared by controlling the flow rate of N2 rangingfrom 100 to 140 mL/min. The morphological differences of the car-bon species are caused by the characteristic of the real-time Ni cat-alyst, involving size, shape and external planar face. The catalyticanisotropy reflects in the form of facets, edges and vertices of thecatalyst grain. The conventional DH-CMCs consisting of circularand flat DH-CMCs in this work had the twinning structure, fromwhich a mechanism is proposed based on different growth rateof carbon nanoparticles on the facet, edge and vertex of catalystgrain. The mechanism is also useful for predicting the existenceof some novel carbon coils, such as DH-CMCs with pentagonalcross section.

Acknowledgments

This work was financially supported by the National NaturalScience Foundation of China (Nos. 51173148 and 90305003) theSpecial Research Fund for Doctoral Program of Higher Education(No. 20060613004), the NBRP (Grant No. 2011CB606200), the2011 Doctoral Innovation Funds of Southwest Jiaotong Universityand the Fundamental Research Funds for the Central Universities(Nos. 2010XS31 and SWJTU11ZT10).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.compositesb.2013.06.010.


References

[1] Zhang X, Zhang X, Bernaerts D, Tendeloo GV, Amelinckx S, Landuyt JV, et al.The texture of catalytically grown coil-shaped carbon nanotubules. EurophysLett 1994;27(2):141–6.

[2] Chen X, Yang S, Sawada N, Motojima S. The design and performance of tactile/proximity sensors made of carbon microcoils. In: Mukhopadhyay SC, Gupta GS,editors. Smart sensors and sensing technology. Berlin, Heidelberg: Springer;2008. p. 251–61.

[3] Su DS. Inorganic materials with double-helix structures. Angew Chem Int Ed2011;50(21):4747–50.

[4] Qi X, Zhong W, Yao X, Zhang H, Ding Q, Wu Q, et al. Controllable and large-scale synthesis of metal-free carbon nanofibers and carbon nanocoils overwater-soluble NaxKy catalysts. Carbon 2012;50(2):646–58.

[5] Nitze F, Mazurkiewicz M, Malolepszy A, Mikolajczuk A, Kedzierzawski P, Tai C-W, et al. Synthesis of palladium nanoparticles decorated helical carbonnanofiber as highly active anodic catalyst for direct formic acid fuel cells.Electrochim Acta 2012;63:323–8.

[6] Raghubanshi H, Hudson MSL, Srivastava ON. Synthesis of helical carbonnanofibres and its application in hydrogen desorption. Int J Hydrogen Energy2011;36(7):4482–90.

[7] Qin Y, Vogelgesang R, Eßlinger M, Sigle W, van Aken P, Moutanabbir O, et al.Bottom-up tailoring of plasmonic nanopeapods making use of the periodicaltopography of carbon nanocoil templates. Adv Funct Mater 2012;22(24):5157–65.

[8] Yang J-P, Chen Z-K, Feng Q-P, Deng Y-H, Liu Y, Ni Q-Q, et al. Cryogenicmechanical behaviors of carbon nanotube reinforced composites based onmodified epoxy by poly(ethersulfone). Compos B: Eng 2012;43(1):22–6.

[9] Colloca M, Gupta N, Porfiri M. Tensile properties of carbon nanofiber reinforcedmultiscale syntactic foams. Compos B: Eng 2013;44(1):584–91.

[10] Yang Z, McElrath K, Bahr J, D’Souza NA. Effect of matrix glass transition onreinforcement efficiency of epoxy-matrix composites with single walledcarbon nanotubes, multi-walled carbon nanotubes, carbon nanofibers andgraphite. Compos B: Eng 2012;43(4):2079–86.

[11] Chen Q, Zhao Y, Zhou Z, Rahman A, Wu X-F, Wu W, et al. Fabrication andmechanical properties of hybrid multi-scale epoxy composites reinforced withconventional carbon fiber fabrics surface-attached with electrospun carbonnanofiber mats. Compos B: Eng 2013;44(1):1–7.

[12] Belhaj M, Deleglise M, Comas-Cardona S, Demouveau H, Binetruy C, Duval C,et al. Dry fiber automated placement of carbon fibrous preforms. Compos B:Eng 2013;50:107–11.

[13] Sánchez M, Rams J, Campo M, Jiménez-Suárez A, Ureña A. Characterization ofcarbon nanofiber/epoxy nanocomposites by the nanoindentation technique.Compos B: Eng 2011;42(4):638–44.

[14] Jia X, Li G, Liu B, Luo Y, Yang G, Yang X. Multiscale reinforcement andinterfacial strengthening on epoxy-based composites by silica nanoparticle-multiwalled carbon nanotube complex. Compos A: Appl Sci Manuf 2013;48:101–9.

[15] Krishnaraj V, Prabukarthi A, Ramanathan A, Elanghovan N, Senthil Kumar M,Zitoune R, et al. Optimization of machining parameters at high speed drillingof carbon fiber reinforced plastic (CFRP) laminates. Compos B: Eng2012;43(4):1791–9.

[16] Tijing LD, Park C-H, Choi WL, Ruelo MTG, Amarjargal A, Pant HR, et al.Characterization and mechanical performance comparison of multiwalledcarbon nanotube/polyurethane composites fabricated by electrospinning andsolution casting. Compos B: Eng 2013;44(1):613–9.

[17] Tai N-H, Yeh M-K, Peng T-H. Experimental study and theoretical analysis onthe mechanical properties of SWNTs/phenolic composites. Compos B: Eng2008;39(6):926–32.

[18] Lau K, Lu M, Hui D. Coiled carbon nanotubes: synthesis and their potentialapplications in advanced composite structures. Compos B: Eng 2006;37(6):437–48.

[19] Li X-F, Lau K-T, Yin Y-S. Mechanical properties of epoxy-based compositesusing coiled carbon nanotubes. Compos Sci Technol 2008;68(14):2876–81.

[20] Lau K-t, Lu M, Liao K. Improved mechanical properties of coiled carbonnanotubes reinforced epoxy nanocomposites. Compos A: Appl Sci Manuf2006;37(10):1837–40.

[21] Wang G, Gao Z, Tang S, Chen C, Duan F, Zhao S, et al. Microwave absorptionproperties of carbon nanocoils coated with highly controlled magneticmaterials by atomic layer deposition. ACS Nano 2012;6(12):11009–17.

[22] Pan L, Zhang M, Nakayama Y. Growth mechanism of carbon nanocoils. J ApplPhys 2002;91(12):10058–61.

[23] Dong L, Yu L, Cui Z, Dong H, Ercius P, Song C, et al. Direct imaging of coppercatalyst migration inside helical carbon nanofibers. Nanotechnology2012;23(3):035702.

[24] Li D, Pan L, Wu Y, Peng W. The effect of changes in synthesis temperature andacetylene supply on the morphology of carbon nanocoils. Carbon2012;50(7):2571–80.

[25] Shaikjee A, Franklyn PJ, Coville NJ. The use of transmission electronmicroscopy tomography to correlate copper catalyst particle morphologywith carbon fiber morphology. Carbon 2011;49(9):2950–9.

[26] Shaikjee A, Coville NJ. The effect of copper catalyst reducibility on lowtemperature carbon fiber synthesis. Mater Chem Phys 2011;125(3):899–907.

[27] Kuzuya C, In-Hwang W, Hirako S, Hishikawa Y, Motojima S. Preparation,morphology, and growth mechanism of carbon nanocoils. Chem Vapor Depos2002;8(2):57–62.




http://refhub.elsevier.com/S1359-8368(13)00321-1/h0005
























































































[28] Xie J, Sharma PK, Varadan V, Varadan V, Pradhan BK, Eser S. Thermal, Ramanand surface area studies of microcoiled carbon fiber synthesized by CVDmicrowave system. Mater Chem Phys 2002;76(3):217–23.

[29] Motojima S, Itoh Y, Asakura S, Iwanaga H. Preparation of micro-coiled carbonfibres by metal powder-activated pyrolysis of acetylene containing a smallamount of sulphur compounds. J Mater Sci 1995;30(20):5049–55.

[30] Motojima S, Chen Q. Three-dimensional growth mechanism of cosmo-mimeticcarbon microcoils obtained by chemical vapor deposition. J Appl Phys1999;85(7):3919–21.

[31] Liu Y, Shen Z. Preparation of carbon microcoils and nanocoils using activatedcarbon nanotubes as catalyst support. Carbon 2005;43(7):1574–7.

[32] Chen Y, Liu C, Du J-H, Cheng H-M. Preparation of carbon microcoils by catalyticdecomposition of acetylene using nickel foam as both catalyst and substrate.Carbon 2005;43(9):1874–8.

[33] Motojima S, Kawaguchi M, Nozaki K, Iwanaga H. Growth of regularly coiledcarbon filaments by Ni catalyzed pyrolysis of acetylene, and their morphologyand extension characteristics. Appl Phys Lett 1990;56(4):321–3.

[34] Yang S, Chen X, Motojima S, Ichihara M. Morphology and microstructure ofspring-like carbon micro-coils/nano-coils prepared by catalytic pyrolysis ofacetylene using Fe-containing alloy catalysts. Carbon 2005;43(4):827–34.

[35] Shi Y, Wang Y, Wang D, Liu B, Li Y, Wei L. Synthesis of hexagonal prism (La, Ce,Tb)PO4 phosphors by precipitation method. Cryst Growth Des2012;12(4):1785–91.

[36] Du J-H, Sun C, Bai S, Su G, Ying Z, Chenga H-M. Microwave electromagneticcharacteristics of a microcoiled carbon fibers/paraffin wax composite in Kuband. J Mater Res 2002;17(5):1233.

[37] Chen X, Saito T, Kusunoki M, Motojima S. Three-dimensional vapor growthmechanism of carbon microcoils. J Mater Res 1999;14(11):4329–36.

[38] Lin M, Tan JPY, Boothroyd C, Loh KP, Tok ES, Foo YL. Dynamical observation ofbamboo-like carbon nanotube growth. Nano Lett 2007;7(8):2234–8.

[39] Mukhopadhyay K, Porwal D, Lal D, Ram K, Narayanmathur G. Synthesis ofcoiled/straight carbon nanofibers by catalytic chemical vapor deposition.Carbon 2004;42(15):3254–6.

[40] Chen X, Motojima S. The growth patterns and morphologies of carbon micro-coils produced by chemical vapor deposition. Carbon 1999;37(11):1817–23.

[41] Kim M, Rodriguez N, Baker R. The role of interfacial phenomena in thestructure of carbon deposits. J Catal 1992;134(1):253–68.


[42] Kawaguchi M, Nozaki K, Motojima S, Iwanaga H. A growth mechanism ofregularly coiled carbon fibers through acetylene pyrolysis. J Cryst Growth1992;118(3–4):309–13.

[43] Amelinckx S, Zhang X, Bernaerts D, Zhang X, Ivanov V, Nagy J. A formationmechanism for catalytically grown helix-shaped graphite nanotubes. Science1994;265(5172):635–9.

[44] Motojima S, Asakura S, Hirata M, Iwanaga H. Effect of metal impurities on thegrowth of micro-coiled carbon fibres by pyrolysis of acetylene. Mater Sci Eng B1995;34(1):L9–L11.

[45] Chen X, Yang S, Motojima S. Carbon nanocoils with changed coiling–chirality formed over Ni/molecular Sieves catalyst. J Mater Sci 2004;39(9):3227–33.

[46] Shaikjee A, Coville NJ. The role of the hydrocarbon source on the growth ofcarbon materials. Carbon 2012;50(10):3376–98.

[47] Hanus MJ, MacKenzie KJ, King AAK, Dunens OM, Harris AT. Parametric study ofcoiled carbon fibre synthesis on an in situ generated H2S-modified Ni/Al2O3

catalyst. Carbon 2011;49(13):4159–69.[48] Boskovic BO, Stolojan V, Khan RUA, Haq S, Silva SRP. Large-area synthesis of

carbon nanofibres at room temperature. Nat Mater 2002;1(3):165–8.[49] Wen Y, Shen Z. Synthesis of regular coiled carbon nanotubes by Ni-catalyzed

pyrolysis of acetylene and a growth mechanism analysis. Carbon 2001;39(15):2369–74.

[50] Chen X, Yang S, Motojima S. Morphology and growth models of circular andflat carbon coils obtained by the catalytic pyrolysis of acetylene. Mater Lett2002;57(1):48–54.

[51] Jian X, Jiang M, Zhou Z, Yang M, Lu J, Hu S, et al. Preparation of high purityhelical carbon nanofibers by the catalytic decomposition of acetylene and theirgrowth mechanism. Carbon 2010;48(15):4535–41.

[52] In-Hwang W, Chen X, Kuzuya T, Kawabe K, Motojima S. Effect of externalelectromagnetic field and bias voltage on the vapor growth, morphology andproperties of carbon micro coils. Carbon 2000;38(4):565–71.

[53] In-Hwang W, Kuzuya T, Iwanaga H, Motojima S. Oxidation characteristics ofthe graphite micro-coils, and growth mechanism of the carbon coils. J MaterSci 2001;36(4):971–8.

[54] Chen X, Motojima S, Iwanga H. Vapor phase preparation of super-elasticcarbon micro-coils. J Cryst Growth 2002;237–239(3):1931–6.















































































composites: part b - icce-nano.orgsite.icce-nano.org/clients/iccenanoorg/2013 controllable...

Documents