Effect of process conditions on the synthesis of carbon nanotubes by catalytic decomposition of meth

Available online at www.sciencedirect.com.●ScienceDirectChinaParticuologyEL SEVIERChina Particuology 5 (2007)213- -219www.elsevier.com/locate/cpartEffect of process conditions on the synthesis of carbon nanotubes bycatalytic decomposition of methaneShuanglin Zhan a,b, Yajun Tian a.*, Yanbin Cuia,b, Hao Wu°,Yonggang Wang°, Shufeng Yea, Yunfa Chenaa lstitute ofProcess Engineering, Chinese Academy of Sciences, Beijing 100080, Chinab Graduate University of Chinese Academry of Sciences, Beijing 1000 China“China University of Mining and Technology, Beijing 10083, ChinaReccived 25 December 2006; accepted 24 March 2007AbstractA new dual-compostion catalyst based on Ni- Mo/MgO with high eficiency of producing carbon nanotubes (CNTs) from methane was reportedrecently. In the prescnt article, with this type of catalyst, the impact of such experimental parameters as reaction temperature, reaction time,concentration of H2, flow rate raio of CH4 to H2 on yield and graphitization were investigated, leading to the fllowing optimal growth conditions:reaction time 60min, reaction temperature 900°C, CH4:H2 about 100:20 mL/min, under which high-yield muti-valled CNTs bundles weresythesized. Raman measrement indicated that the as synthesized product was well graphitized, and the purity was estimated over 95% byTG-DSC analyis. In terms of the above results, an explanation of hghgfficiency formation of CNTs bundles and the co catalysis mechanism ofNi-MoMgO were suggested.◎2007 Chinese Society of Particuology and Institute of Process Enginering, Chinese Academy of Sciences. Published by Elsevier B.V.All rights reserved.Keywords: Carbon nanotubes; Catalytic decomposition; Carbon yield; Graphitization degree1. IntroductionKomatsu, Ichihashi, & ljima, 1997), and catalytic decomposi-tion ofhydrocarbons (Liuetal, 1997; Panet al., 1998; Zhu et al,Since carbon nanotubes (CNTs) were first discovered by2002). Catalytic decomposition is the most promising methodIijima (1991), they have been recognized as one of the most for industrialization due to the advantages of low cost, mildimportant materials due to their superior surface property, highpreparation conditions and casy control.tennacity, high electron conductivity, excellent field emissionIn most cases, CNTs are random and their productivity isproperty, metal and semiconductor property (Collins, Zett,still low, thus greatly restricting their industrialization and appli-Bando, Thess & Smalley, 1997; Dresselhaus, 1998; Yu et al,cation. Recently, a new dual-composition catalyst based on200)and potential applications in the fields of emission display, Ni- Mo/MgO prepared by solution combustion synthesis (SCS)nanoprobe, gas storage, single-molecular transistors, quantum(Coquay, Peigney, De Grave, Vandenberghe, & Laurent, 2002;wires and nanoscale electronic devices (Dai, Hafner, Rinzler,Lima, Bonadiman, De Andrade, Toniolo, & Bergmann, 2006;Colbert, & Smalley, 1996; de Heer et al, 1995; Gadd et al, Pati, 1993) for eficient production of CNTs was reported (Li1997; Mintmire, Dunlap, & White, 1992; Tans, Verschueren, &et al, 2005). Generally SCS can rapidly produce homogencousDekker, 1998; Tans et al, 1997). At present, many synthesis oxide mixtures with high surface area and small crystal size viamethods have been developed including arc-discharge (ijjma,an exothermic combustion reaction. This paper reports the syn-1991; ljima & Ichihashi, 1993; Liu et al, 1999), laser vaporiza-thesis of this type of catalyst and large-scale production ofCNTstion (Guo, Nikolaev, Thess, Cobert, & Smalley, 1995; Yudasaka, bundles. More importantly, the process parameters, includingreaction temnerature. reaction time. concentration of H2 andflowr中国煤化工ied in order to study theireffect* Coresponding author. Tel: +86 10 62588029.optimiYHCNMHGionsoastodelerminetheE-mail adress: yjian@home ipeac .cn (Y Tian).1672-2515/5 - see front matter。2007 Chinese Scietey of Pricology and Intitute of Pocess Enineering. Chinse Academy of Sciences. Published by Esevier B.V.All nghts rerved.doi:10.1016/j. cpart2007.03.004214s. Zhan et al. /China Paricuology 5 (2007)213 -2192. ExperimentalThe catalyst was prepared by SCS， by dissolvingOnly MgMoO, can beNi(NO3)2 6H2O (0.291 g), (NH4)6Mo7O24.4H2O (2.120 g) andcleary ienified.Mg(NO3)2 6H2O (2.564 g) in 19 g of ethylene glycol to form ahomogeneous solution by sirring and refluxing at 90°C for 5h.The solution was then heated at 700°C for 15 min in a muf-fle furmace. Finally, a foamy material was obtained which was。ground to a fine powder.The experiments were carried out in a quartz tube fixed-bedreactor in a horizontal tube furmace. After first evacuating thequartz tube reactor to remove air, about 50 mg catalyst spreadout in a small porcelain boat was placed in the center of thequartz tube. Then the furmace was heated from room temperature30106C70to 900°C in nitrogen, which was then switched to a mixture of20()methane and hydrogen at a given flow rate for a given periodFig. 2. XRD spectra of catalyst (a) and typical product (b) synthesized underof time. After growth of CNTs, the furnace was cooled to roomoptimal reaction conditions.temperature under nitrogen protection.The as-prepared catalyst + product was characterized bywhere MTotal is the total mass of final catalyst + product andscanning electron microscopy (SEM, JEOL, JSM-6700),Mcat is the initial mass of catalyst.transmission eletron microscopy (TEM, HITACH, H-8100),X-ray diffraction (XRD, Panalytical X'Pert PRO), Raman3. Results and discussionspectroscopy (JY-HR800) and thermal gravimetry/differentialscanning calorimetry (TG- DSC, SETARAM Labsys). The cat-3.I. Characterization of catalyst and typical productalyst was further characterized by XRD and SEM.The carbon yield, average growth rate and methane con-Fig. 1a is the SEM image showing the morphology of theversion are calculated by using the following equations,catalyst prior to synthesizing CNTs. The particle shape is ireg-respectively:ular, and not uniform in size, ranging from several hundreds ofcarbonyield =Mroual - Mcau. x 100%(1)nanometers to several micrometers. There are always some pla-nar facets on the particles, suggesting they are well crystallized.McatFig. 1b exhibits the SEM image of the catalyst reduced in H2MTotal - MCatat 700°C for 60 min, showing many finer particles which coveraverage growthrate =(mg/min)(2)growth timethe surface of flakes with a size ranging from 10 to 20nm.Fig. 2a is an X-ray diffraction spectrum of the catalyst.Only one phase, MgMoO4, can be clearly identified. Duringmethane conversioncalcinations at high temperature, molybdenum may be first oxi-MToal - Maudized into MoO3, which then reacts with MgO to form theFlow rate (L/min) x Time (min) + 22.4(L/mol) x 12(g/mol)new MgMoO4 phase. Interestingly, no Ni-containing phase wasx 100%(3) traced in the diffraction pattern, implying that Ni ion has entered中国煤化工FHCNMHG300nmFig. 1. SEM images of catalyst (a) and catalyst reduced in H2 for 60 min (b).S. Zhan et al. /China Paricuology 5 (2007) 213. -2192152um5000md200nmFig. 3. Electron micrograms of typical product syothesized under the optimal reaction conditions: (a and b) SEM images; (c and d) TEM images.the MgMoO4 lttie, to form a solid solution. It is well known fibers shown in Fig. 3b. Fig. 3c and d are TEM images of thethat both NiO and MgO possess a rock salt type crystal struc- bundles dispersed by ultrasonic treatment. It is clear that theture, and that the ionic radius of Ni2+ (0.070 nm) is quite close fibers are hollow with outer diameters between 10 and 20 nmto that of Mg2+ (0.065 nm) (Chen, Zhang, Lin, Hong, & Tsai, which are consistent with the diameter range of the particles1997), so that it is quite likely that Ni4+ replaces Mg2+ to form shown in Fig. 1b, indicating that the diameter of CNTS dependsthe NixMg1- xO solid solution. Therefore, the final catalyst can on the size of the associated particles.be written as Mg1. xNi.MoO4.Fig. 2b is an X-ray diffraction spectrum of a typical prod-3.2. Eifect of reaction temperatureuct obtained under the optimal reaction conditions. The peakaround 26° was assigned to (002) peak of graphite, which isFig.4 shows the influence of reaction temperature on car-the most important characteristic peak of CNTs, and graphite bon中国煤化工e of 60min and CH4/H2(1 01). From Fig. 2b, the MgMoO4 phase of the initial cata- flow ICNMHs found that carbon yieldlyst disappeared, having changed to others (MoC, MoO3) after increa，and reached a maximumreaction.at 900°C, and then declined over 900 °C. Noticeably, no car-Fig. 3a d demonstrates the electron micrographs of a typical bon was produced ai 600°C, because the catalyst could not beproduct. Seen in Fig. 3a are bundles comprised of hundreds of activated at that temperature.216s. Zhan et al. / China Particwology 5 (2007)213- 21980016007001400500。1200500系10000|440033000o2200100-100600 700 800 900 1000 11004080.12016020024Reaction time (min)Temperature (°C)Fig. 4. Reaction temperature dependence of carbon yield underconstant reactionFig. 6. Reaction time dependence of cartbon yield and average growth rate underconstant reaction temperature of 900°C and CH4/H2 = 40/20.time of 60 min and CH4/H2 =40/20.Methane decomposition is a moderately endothermic crack-ing reaction. Consequently, increase in temperature results in anexponential increase of the equilibrium constant, so the yieldrises with increasing temperature, but at even higher tempera-ture, the formation rate of carbon over the catalyst surface mightexceed the growth rate of CNTs thus resulting in encapsulating墓the active catalytic particles (Piao, Li, Chen, Chang, & Lin, 2002;Snoeck, Froment, & Fowles, 1997).Fig. 5 shows that the intensity of the characteristic peakof graphite (002) depends on the temperature consistent withyield, namely graphitization of CNTs is the best at 900°C. The102030500intensity declines above 900°C, possibly because the productscontain more amorphous carbon at the higher temperature, thusfsetting the contribution of CNTs Additionally, Chen, Dai,Fig. 7. XRD results of as-prepared products :Huang, and Jehng (2006) also investigated the influence of reac-temperature of 900°C, CH4/H2 = 40/20 and at reaction time of (a) 10 min, (b)tion temperature on the graphitization degree of CNTs by the30 min, (c) 60 min, (d) 120 min and (c) 240 min.XRD technique, and he drew a similar conclusion. Based on theabove analyses, the optimum temperature to synthesize CNTs3.3. Effect of reaction timeis about 900°C.Fig. 6 shows the effect of reaction time on carbon yield andaverage growth rate under the constant reaction temperature of900°C and CH4/H2 fow rate ratio of 40/20 mL/min, indicatingthat carbon yield increases with extended reaction time. Evenafter 240 min, this increase did not stop, implying that the cata-lyst could remain active for a long time. The trend of the averagedgrowth rate indicates that productivity declines after 60 min,evidently because of loss of catalyst activity with reaction.In addition, Fig.7 shows that the intensity of the characteristicpeak of graphite (00 2) reaches a maximum at around 60 minand does not recover after that, implying that time is needed toremedy the defects of the CNTs structure. Integrating the aboveanalyses, the optimal reaction time is chosen to be 60 min.中国煤化工46C703.MYHCNMHG20(9)Fig. 8 displays the effect of hydrogen flow rate on carbonFig. s. XRDresults of as-prepared products synthesized under constant reactiontime of 60 min, CH4/H2 = 40/20 and at various reaction temperatures of (a)yield under constant reaction time of 60 min, at 900°C and CH4700°C, (b) 800°C, (c) 900 °C, (d) 1000°C and (e) 1100°C. .flow rate of 40 mL /min, showing that the carbon yield increasess. Zhan et al. /China Pricology s (2007)213-21917750，26700180065060022; 14000量g 550名12008810004080035406080100120140160180200220206CH4 flow rate (mUmin)H2 fow rate (mL/min)Fig. 10. Dependence of carbon yield and carbon converion on CH fow rateFig. 8. H2 flow rate dependence of caron yield unte deodene of Carbon yield under constant reaction tem-under constant reaction temperature of 900°C, reaction time of 60min andH2peralure of°C, reaction time of 60min and CH4 flow rate of 40 ml /min.flow rate of 20 mLU/min.3.5. Efect of methane flow ratewith increasing H2 flow rate, but flls off as the flow rate exceeds20 mL/min.With a reactin time of 60 min, a temperature of 900°C,In the presence of H2, the catbon containing species co-and a H2 flow rate of 20mL/min, Fig. 10 shows that carbonproduced during CNT growth over the catalyst surface are likelyyield increases and methane conversion decreases with increas-to be removed by the reversible reaction, which would favoring CH4 flow rate. Methane conversion is almost constant forself-cleaning of the catalyst surface by inhibiting the depositionCHa Aow rate between 70 and 100m/min, but sarts to dropforof the“encapsulating" carbon, and slow down deactivation of >100mU/min. In addition, Fig.11 clearly shows that the inten-the catalyst. However, a high content of hydrogen would cer-sity of the charateristic graphite peak (002) increases withtainly reverse the direction of the decomposition reaction andincreasing CH4 fow rate, and remains at a high level when thehence suppress CNT growth (Chen et al, 1997; Makris, Giorgi,flow rate is over 100 mL/min, showing that short resident timeGiorgi, Lisi, & Salemitano, 2005; Piao et al, 2002; Snoeck etof methane on the calyst surface is benficiall to sytesizingal., 1997). Fig. 9 shows that the variation of the intensity ofwell-graphitized CNTs (Chen et al, 2006). According to thethe graphite peak (002) fllows that of carbon yield, suggest-above analysis, more CH4 tends to drive the reaction towarding that the concentration of H2 afeets not only carbon yielddecomposition, leading to more CNTs.but also gaphization. Based on the above analysis, it ppearsReferring to Figs.8 and9 again, one can conclude that higherthat hydrogen in the reactor is necessary but should not beH2 concentration enhances not only cartbonyield but also graphi-excessive.tization, when the ratio of CH4 to H2 is around 5:1, producingthe CNT morphology displayed in Fig. 3. In our experiments,入Munn !MrC.ba10305060中国煤化工50720(9)Fig..MYHCNMHGFig. 9. XRD results olsynthesized under constantRD results of as-prepared products synthesized under constant reac-reaction tempraure of 900C, reaction time of 60min and H2 flow ratle .tion teprature of 900°C, reaction time of 60min and CH4 flow nace ofof 20mL/min, and at CH4 flow rates of (a) 50mU/min,, () 70mL/min,40mU/min, and at H2 fow ratcs of (a)0mL/min, (b) 10mL/min, (c) 20mL/min,(e) 90mL/min, (d) 100mL/min,120mL/min, (0) 150 mL/min and (g)(d) 40mU/min, (e) 60 mL/min and ( 80mL/min.200 mL/min.18s. Zhan et al / China Paricuology s (2007)213-219product measured in air. From the TG curve, it can be seen that2500the itial burning temperature is about 490°C, and there is onlyone sharp exothermic peak around 530°C in the DSC curve.These results reflet the high purity of the CNTs, for otherwisethere would be signals between 300 and 400°C associated with旁1500amorphous carbon, as is consitent with the observation inFig.3.Moreover, acording to the TG curve, the percent of CNTS inthe product is calculated to be over 95%.When CH4 and H2 were admited for CNTs growh, itis00understandable that spliting of the catalyst would occur underthe strongly reducing atnosphere, and then clusters of finer cat-alyst particles were formed and afixed over a faced plane,200 400 600 800 10000 1200 1400 1600 1800 2000as ilustrated in Fig.1b. Initiated by the catalyst particles andRaman shit (cm*)owing to van der Waals attraction between nanotubes, the CNTsFig. 12. Raman specrum of a tpical produec.bundles were formed (Ning et al, 2002; Xu et al, 2004).The high carbon yield under the optimal conditions may bewe chose 100 and 20mL/min as the optimal flow rate of CH4atributed to the co-catalysis mechanism of the Ni- Mo cata-and H2, respectively. For appropriate CNTs yield, methane con-lyst. Mo as the catalytic center promotes the aromatization ofversion and graphitization, the reaction temperature and time aremethane. The intermediate aromatic secies generated over Mochosen to be 900°C and 60 min, respectively.sites can feed the adjacent Ni sites to grow CNTs with high ffIn order to invstigate the crsalinity and purity of as-ciency and high yield Cassell, Raymakers, Kong, & Dai, 199;synthesized CNTs, bobh Raman and TG DSC caracerizationsNing et al, 2002; Tang ct al, 2001). Franklin and Dai (2000)were conducted for the product derived from the above opti-studied the efuent of the methane CVD system at 900C formized conditions. Two peaks around 1577 and 1347 cm-1 were growing SWNTs through mass spectroscopy, suggesting thatobserved in the Raman spectrum shown in Fig. 12. The for- the methane underwent negligible self-pyrolysis and the smallmer is atibuted to the tangential C C stretching mode (Gconcentration of benzene converted from methane on catalystband) of the carbon material with a sp2 orbital structure, andfavored the growth of the product because the highly reacivethe ltter, to the disorder induced phonon mode (D band). Nobenzene could provide an eficient catbon- fedstoc for nan-peaks near 150 -250cm-', which are charateristic of vibrationotubes, and a six membered-ring-based growh model has beenmodes of single- wall CNTs, were observed, indicating that fewpresented by Tian et al. (2004) with benzene as raw material.single-wall CNTs were obtained under our conditions. The peakThe highdispersion of Ni-species in the NizMg1 20 solid solu-intensity ratio Il, which refers to the ratio of graphitic struc-tion and the efet of valence sbilizatini by MgO cystal-fieldtures to the defective structures of multi-wall CNTSs, can be usedwould favor the inhibition of deep reduction of the Ni2+ to Ni0to ases the graphitization degree of CNTs (Thostenson, Ren,and aggregation of the Ni0 to fom large particles of metal-& Chou, 2001). Compared to the Raman results by Chenet allhic nickel at the surface, making CNTs grown on this catalyst(2006), the CNTsproduced under our optimal conditions canberelatively small and even (Chen et al, 1997). Further study isconsidered to possess good quality.needed in order to confirm the growth mechanism with directUsing the TG DSC technique, the puity of the typicalexperimental evidence.product, afer removing the catalyst by hydrochloric acid, was4. Conclusionsroughly estimnated, Fig. 13 is the TG- DSC curve of the typicalWell-graphitized and hig-ield multiwall CNTs bundles00-50were synthesized under optimal reaction conditions by thedecompositin of methane on Ni MoMgO catalyst, and the30infuence of relevant operating parameters (reaction time, reac-tion temperature, concentration of H2 and the ratio of CH4 toH2)onCNTsyield, methane conversion and graphitization werestudied stetilly Proposed optimal conditions are as fol-20lows: reaction time 60 min, reaction temperature 900°C and Aowrate ratio of CH4 to H2 about 100: 20 mL/min. The purity of theproduct is over 95%. A Dossible exolanation for the formationof hig中国煤化Iented.1002003004005006007008009010AcknoMHCNMHGTemperature (°C)This work was supported fnancially by the National Nat-Fig. 13. TC-DSC curves of a typical product.ural Science Foundation of China No.20506010), BejingS. Zhan et al. /China Paricuology 5 (2007)213- 219219Novel Program (2006A74) and Natural Science Fund of Shanxi Liu, J, Dai, H,. Hafner, J. H. Colbern, D. T, Smalley,R. E., Tans,S. J. ctal.Province No.20063004).(1997). Fullerene“crop circles". Nature, 385, 780-781.Makris, T. D. Giorgi, L, Giorgi, R.. Lisi, N, & Salemitano, E. (2005). CNTReferencesgrowth on alumina supported nickel catalyst by thermal CVD. Diamond andRelated Materials, 14, 815- -819.Mintmire,J. W, Dunlap, B. I, & White, C. T. (1992). Are fllerene tubulesCassell, A. M., Raymakers, J. A. Kong, J, & Dai, H. (999) Large scale CVDmaeallic? Physical Review Letters, 68, 631-634.synthesis of sngl-walled carbon nanotubes. Journal ofPhysical ChemistryNing, Y. H, Zhang, X. B.,. Wang, Y. W. Sun, Y.L, Shen,L. H, Yang, x. F, etal.B, 103, 6484 -6492.(2002). Bulk production of multi-wall carbon nanoube bundles on sol gelChen, C. M, Dai, Y. M, Huang, J. G.. & Jehng, J. M. (2006). Inermtalieprepared catalyst. Chemical Physics Letters, 366, 555- 560.calalyst for carbon nanoubes (CNTS) growth by thermal chemical vapor Pan,Z W. Xie, s. s.. Chang, B. H Wang,C Y, Lu, L. Liu, wW, et al. (998).deposition method. Carbon, 44, 1808 -1820.Very long carbon nanotubes. Nature, 394, 631-632.Chen, P, Zhang. H. B.. Lin, G. D.. Hong, Q. & Tsai, K. R. (1997). Growth ofPatil, K. C. (1993). Advanced ceramics: Combustion synthesis and properties.carbon nanotubes by catalytic decomposition of CH4 or CO on a Ni MgOBulletin of Materials Science,3- -541.catalyst. Carbon, 35, 1495 -1501.Piao, L, Li, Y, Chen, J, Chang, L, & Lin, J. Y. s. (2002). Methane decom-Collins, P. G., Zet, A.. Bando, H., Thess, A., & Smalley, R. E. (1997). Nanoubeposition to carbon nanotubes and hydrogen on an alumina supported nickelnanodevice. Science, 278, 100- -103.aerogel catalyst. Catalysis Today, 74, 145 -155.Coquay, P, Peigney, A., De Grave, E.. Vandenberghe, R. ., & Laurent, C.Snoeck, J. W., Froment, G. F, & Fowles, M. (1997). Filamentous carbon for-(2002). Carbon nanotubes by a CVD method. Part I: Formation of nan-mation and gasification: Thermodynamics, driving force, nucleation, andotubes from (Mg, Fe)O catalysts. Joumal of Physical Chemistry B, 106,steady-statec growth. Journal of Catalysis, 169, 240 -249.13199- -13210.Tang, s. Zhong, Z. Xiong, Z, Sun, L.. Liu, L, Lin, J., ct al. (2001). Contolled)ai, HI, Hafner, J. H, Rinzler, A. G., Colbert, D. T, & Smalley, R. E.growth of single-walled carbon nanotbes by catalytic decomposition of(1996). Nanotubes as nanoprobes in scanning probe microscopy. Nature,CH4 on Mo/Co/MgO catalysts. Chemical Physics Letters, 350, 19 26.384, 147-150.de Heer, W. A., Bacsa, w. S. Chatelain, A, Gerfin, T., Humphrey-Baker, R.,Tans, s. J, Devorel, M. H, Dai, H, Thess, A., Smalley, R. E,. Geerligs, L. J..et al. (1997). Individual single walled carbon nanotubes as quantum wires.Foro, L, et al. (1995). Aligned carbon nanotubes fiIms: Production andNature, 386, 474477.optical and electronic properies. Science, 268, 845 -847.Tans, S. J, Verschueren, A. R. M., & Dekker, C. (1998). Room: temperatureDresselhaus, M. S. (1998). Nanotechnology: New tricks with nanotubes. Nature,transistor based on a single carbon nanotube. Nature, 393, 49- 52.391, 19- 20.Thostenson, E. T, Ren, Z., & Chou, T. W. (2001). Advances in the science andFranklin, N., & Dai, H. (000.0 An enbanced CVD approach to extensive nan-technology of carbon nanotubes and their composites: A review. Compositesotube networks with directionality. Advanced Materials, 12, 890 -894.Science and Technology, 61, 1899- 1912.Gadd, G. E., Backford, M., Moricca, s, Webb, N, Evans, P. J, Smith, A. M,Tian, Y. J, Hu,Z, Yang, Y, Wang, X. z.. Chen, X, Xu, H., etal. (2004). In situet al. (1997). The world's smallest gas cylinders? Science, 277, 933- -936.TA-MS study of the six-membered-ring based growth of carbon nanotubesGuo, T, Nikolaev, P, Thess, A., Cober, D. T, & Smalley, R. E. (1995), Catalyticwith benzeneJourmal of the American Chemical Society, 126,growth of single-walled nanotubes by laser vaporization. Chemical Physics1180-1183.Iijima, s. (1991). Helical microtubules of graphitic carbon. Naure, 354, 56 -58.Letters, 243, 49 -54.Xu, J. M, Zhang, X. B., Li, Y, Tao, x. Y, Chen, E, Li,T, et al.(2004). Preparation of Mg1- -xFex.MoO4 catalyst and its application to growIijima, S, & Ichihashi, T. (1993). Single-shell carbon nanotubes of 1-nm diam-MWNTs with high eficiency. Diamond and Related Materials, 13, 1807-eter. Nature, 363, 603- 605.1811.Li, Y, Zhang, X. B, Tao,X. Y., Xu,J. M., Huang, W.Z, Luo,J. H,etal. (2005).Yu, M. F, Lourie, 0.. Dyer, M. J, Moloni, K, Kelly,T F, & Ruoff, R. s. (0000Mass production of high-quality multi-walled carbon nanotube bundles onStrength and breaking mechanism of mulialled carbon nanotubes undera Ni/MoMgO catalyst. Carbon, 43, 295- -301.tensile load. Science, 287, 637- 640.Lima, M. D.. Bonadiman, R., De Andrade, M. J.. Toniolo, J.. & Bergmann, C.Yudasaka, M, Komatsu, T, Ichihashi, T, & ljima, s. (1997). Nanoholes onP. (2006). Synthesis of multi- walled carbon nanotubes by catalytic chemi-silicon surface created by electron iradiation under ulrahigh vacuum envi-cal vapor deposition using Cr2- .Fe.O3 as catalyst. Diamond and Relatedronment. Chemical Physics Letters, 278, 102- 106.Materials, 15, 1708 -1713.Liu, C.. Cong, H. T, Li, F, Lu, K, Cheng, H. M. & Zhou, B. L. (999. Semi-Zhu, H. w., Xu, C. L, Wu, D. H, Wei, B. Q., Vajtai, R., & Ajayan, P. M. (2002).Direct synthesis of long singl-walled carbon nanotube strands. Science,continuous syothesis of single-wall carbon nanotubes by a hydrogen arc296, 884 886.discharge method. Carbon, 37, 1865-1868.中国煤化工MYHCNMHG