Controllable synthesis of carbon nanotubes by changing the Mo content in bimetallic Fe–Mo/MgO catalyst
Xiangju Xu,Shaoming Huang,Zhi Yang,Chao Zou,Junfan Jiang,Zhijie Shang
DOI: https://doi.org/10.1016/j.matchemphys.2011.02.028
IF: 4.778
2011-01-01
Materials Chemistry and Physics
Abstract:Research highlights ► Increasing the Mo content in the Fe–Mo/MgO catalysts resulted in an increase in wall number, diameter and growth yield of carbon nanotubes. ► The Fe interacts with MgO to form complex (MgO) x (FeO) 1− x (0 < x < 1) and is then dispersed into the MgO support uniformly, resulting in very small Fe nanoparticles after reduction. ► The Mo interacts with MgO to form stoichiometry compound MgMoO 4 and relative large metal Mo particles can be generated after reduction. ► The avalanche-like reduction of MgMoO 4 makes the catalyst particles to be small thus enhances the utilize efficiency of Fe nanoparticles. Abstract A series of Fe–Mo/MgO catalysts with different Mo content were prepared by combustion method and used as catalysts for carbon nanotube (CNT) growth. Transmission electron microscopy studies of the nanotubes show that the number of the CNT walls and the CNT diameters increase with the increasing of Mo content in the bimetallic catalyst. The growth yield determined by thermogravimetric analysis also follows the trend: the higher the Mo content, the higher the yield of the CNTs. However, the increase of Mo content leads to the lower degree of graphitization of CNTs. A comparative study on the morphology and catalytic functions of Fe/MgO, Mo/MgO and Fe–Mo/MgO catalysts was carried out by scanning electron microscopy and X-ray diffraction. It is found that the Fe interacts with MgO to form complexes and is then dispersed into the MgO support uniformly, resulting in very small Fe nanoparticles after reduction. The Mo interacts with MgO to form stoichiometry compound MgMoO 4 and relative large metal Mo particles can be generated after reduction. High yield CNTs with small diameter can be generated from Fe–Mo/MgO because the avalanche-like reduction of MgMoO 4 makes the catalyst particles to be small thus enhances the utilize efficiency of Fe nanoparticles. Keywords Carbon nanotubes Chemical vapor deposition Scanning electron microscopy 1 Introduction Carbon nanotubes (CNTs) have attracted a great deal of research attention since the first observation by Iijima [1] . Their unique properties [2–5] due to their quasi-1D structure and potential applications in many fields [6] have excited researchers all over the world to grope for the convenient and controllable method to synthesis this fancy material. Currently, arc-discharge [7,8] , laser ablation [9,10] , and chemical vapor deposition (CVD) are the main methods for CNT production. The first two methods have the advantages in the high qualities of the CNTs, however at the same time have disadvantages in low yields, high energy consumption, impurities of the products and high cost. So they are not considered as the promising means for an industrial production of the CNTs. CVD is the most promising way for industrial production because of its advantages in large scale production, easy controllability and low cost etc. A lot of efforts have been put on the CVD synthesis of CNTs [11–16] and understanding of the growth mechanism [17–21] . As far as the CVD process commonly employed by now times is concerned, CNTs can usually grow at 800–1200 °C using hydrocarbons, CO and small molecular weight alcohol as carbon sources [22,23] . The growth temperature can be decreased about 200 °C when plasma or micro-waves and other assistant means are introduced into the CVD system [24,25] . It is worth being mentioned that the exploring of series of transitional metal catalysts played an important role and possessed a crucial significance [26,27] . A lot of researches have proved that the catalyst was the one of the key steps within the whole CVD process [12,26–30] . To explore catalysts with high catalytic activity is particularly important for bulk production of CNTs [31] . Mo has been regarded as an effective doped additive to many other metal catalysts such as Fe–Mo cluster and supported Fe–Mo and Co–Mo etc., for producing both single-walled CNTs (SWCNTs) and multi-walled CNTs (MWCNTs) [32–42] . Several authors have reported that the yield of CNTs is increased by adding a certain quantity of Mo to Fe [33–37] , Co [38–40] , and Ni [41,42] . A phase diagram was used to explain the optimized conditions for the Fe–Mo system [33,34] . However, the function of Mo in bimetallic catalyst is still not very clear and some conflict results were reported in the literatures. In addition, it is still unclear why the catalytic nature of bimetallic catalyst is better than monometallic catalyst [32–42] . In this paper, we focus our effort on the controllable growth of CNTs by altering the Mo content in the bimetallic Fe–Mo/MgO catalyst and understanding of the different behaviors of the metals during catalyst preparation and CNT growth as well. The morphology and structures of the CNTs have been investigated by scanning electron microscopy (SEM), transmission electron microscopy (TEM), and Raman spectroscopy. The catalytic functions and efficiency of the catalysts have been studied by X-ray diffraction (XRD) and thermogravimetric analysis (TGA). The results indicate that the number of the CNT walls, the CNT diameters and the growth yield can be controlled by changing the Mo content in the Fe–Mo/MgO catalyst. It is also found that Fe and Mo have different behaviors during catalyst preparation, reduction and catalytic growth of the CNTs. 2 Experimental 2.1 Catalyst preparation The catalysts were prepared by a combustion method. Typically in our study, a mixture of iron nitrate (Fe(NO 3 ) 3 ·9H 2 O, A.R.), ammonium molybdate ((NH 4 ) 6 Mo 7 O 24 ·4H 2 O, A.R.), magnesium nitrate (Mg(NO 3 ) 2 ·6H 2 O, A.R.), citric acid (C 6 H 8 O 7 ·H 2 O, A.R.) and glycine (C 2 H 5 NO 2 , A.R.) at a designed ratio was dissolved into distilled water and heated up to evaporate water till the solution became a foamy solid. The foamy solid will be ignited to become ashes when it was heated further. The ashes were then collected and loaded into a 550 °C muffle furnace for 60 min. The powder material obtained was used for growth of the CNTs. The catalysts with different composition are summarized in Table 1 . 2.2 Growth of CNTs CNTs were synthesized by the catalytic decomposition of methane diluted with hydrogen (500 sccm methane and 500 sccm hydrogen) at 1000 °C for 20 min. The preparation was carried out in a fixed-bed flow reactor made of a quartz tube (i.d. 2.5 cm, length 80 cm) laid on a horizontal furnace. One quartz boat with 50 mg of catalyst was placed in the central part of the reactor. The catalyst was heated to 1000 °C in nitrogen (500 sccm). Subsequently, the nitrogen was switched to hydrogen and methane during reaction process. The CNTs with catalysts were collected from the quartz boat. 2.3 Characterization The morphology and structures of the CNTs were characterized by SEM (JEOL, JSM-5610, 10 kV), TEM (JEOL, JEM-2010, 200 kV), and Raman spectroscopy (Thermo Nicolet Almega, ALMEGA-Dispersive Raman, Nd: YAG, excited at 532 nm). TGA was used to analyze the growth yield of CNTs (SDT, Q600, air, 10 °C/min). The morphology and structures of the catalysts were determined by SEM and XRD (Bruker, D8, CuKα1, λ = 1.5604 Å, the patterns were recorded over 20° < 2 θ <90°). 3 Results and discussion In order to examine the effect of the catalyst composition on the activity of the catalysts, bimetallic catalyst of Fe 0.1 Mo 0.1 Mg 2 O 2.45 and monometallic catalysts of Fe 0.1 Mg 2 O 2.15 and Mo 0.1 Mg 2 O 2.3 were used to produce CNTs. Fig. 1 presents the typical SEM and TEM images of the as-grown nanotubes from different catalysts. From Fig. 1 a, it can be seen that the nanotubes from Fe 0.1 Mo 0.1 Mg 2 O 2.45 are long and straight. Their average diameters were about 5 nm ( Fig. 1 d). The numbers of the walls were mainly in the range of 2–5. From TEM observation ( Fig. 1 d), the surface of the CNTs was smooth, indicating fewer defects in the CNT surface. As clearly seen from Fig. 1 b, the CNTs grew from Fe 0.1 Mg 2 O 2.15 are bunchy and the diameters of the CNTs are smaller than 5 nm ( Fig. 1 e), which is slightly smaller than that of CNTs grew from Fe 0.1 Mo 0.1 Mg 2 O 2.45 . Comparatively, CNTs obtained from Mo 0.1 Mg 2 O 2.3 are flexural, and the diameters are bigger than that of the former two ( Fig. 1 c and f). Fig. 1 f exhibits a single CNT with a diameter of around 14 nm. Raman spectroscopy was also applied to monitor the quality of the as-grown CNTs from different catalysts. Three peaks in the Raman spectrum, G-band, D-band and radial breathing mode (RBM), are generally used to characterize the CNTs’ structures. CNTs can be approximately looked as a duct rolled with graphitic sheets, so the peak appears at about 1585 cm −1 which corresponds to graphitic usually stands for the graphitic walls, the stronger the peak appears, the better the walls graphitize. Carbon atoms on graphitic walls present sp 2 hybridized state while the sp 3 state corresponds to the spatial topology caused by curve and transmutation which is generally looked as the disadvantage factors to the perfect 1D structures of the CNTs. Sp 3 state of carbon atoms gives signals at about 1350 cm −1 in the Raman spectra and corresponds to the defect of CNTs or amorphous carbon. So, the stronger the peak here appears, the lower the quality is [43–47] . According to the research of Rao et al. [48,49] , CNTs, especially small diameter CNTs and SWCNTs have a characterized vibration mode which gives signals at low frequencies below 400 cm −1 , they called this mode as RBM. From Fig. 2 a , the Raman spectrum of CNTs obtained from Fe 0.1 Mo 0.1 Mg 2 O 2.45 shows a strong peak of G-band, low intensity of D-band and the RBM peaks (see inset), indicating the high quality and small diameter of the CNTs. Fig. 2 b is the Raman spectrum of the CNTs from Fe 0.1 Mg 2 O 2.15 catalyst and gives very strong peak of G-band, very feeble peak of D-band and the RBM peaks in the low frequencies (see inset), indicating high quality of the CNTs as well. Based on the Raman results, the samples from these two catalysts contain small diameter CNTs, which is also confirmed by TEM observations ( Fig. 1 d and e). Fig. 2 c is the Raman spectrum of the CNTs from Mo 0.1 Mg 2 O 2.3 , the high magnitude of D/G ratio indicates many defects on the CNTs. The absence of RBM signals in the low frequencies area indicates that almost no small diameter CNTs in the samples, which is in good agreement with the TEM observation ( Fig. 1 f). Wholly speaking, the CNTs grew from Fe 0.1 Mo 0.1 Mg 2 O 2.45 and Fe 0.1 Mg 2 O 2.15 catalysts are high quality with small diameters while the CNTs from Mo 0.1 Mg 2 O 2.3 catalyst are relative low quality with large diameters. According to the electron microscopy observations (both SEM and TEM) and Raman spectra, the wall-number and the diameter of the CNTs correlate well with the catalysts. In order to investigate the effect of the Mo amount on the structure of the CNTs, the content of Mo was varied from 0.5 to 2 (the molar ratio of Mo:Fe) in the bimetallic Fe–Mo/MgO catalysts. As shown in Fig. 3 a–c , it was found that small diameter CNTs can be produced for all the catalysts used. However, the diameter and the number of the CNT walls apparently increased with the increasing of Mo content. Fig. 3 d–i gives the distributions of the wall numbers and diameters of the CNTs for these three catalysts. Clearly, the CNTs produced by Fe 0.1 Mo 0.05 Mg 2 O 2.3 (the molar ratio of Mo:Fe is 0.5:1) have predominant triple-walled CNTs ( Fig. 3 d). When the atomic ratio of Fe:Mo is 1:1, mainly quadruple-walled CNTs were found ( Fig. 3 e). Increasing the Fe:Mo to be 1:2 leads to the predominant formation of CNTs with five walls ( Fig. 3 f). It is worth noting that the diameters of the CNTs also follow the similar trend: the higher the Mo content, the larger the diameter of the CNTs. The variation in the CNT diameter from 2 to 10 nm can be seen in Fig. 3 g–i. CNTs with about 5 nm and 10 nm were synthesized at low and high content of Mo, respectively ( Fig. 3 g–i). The content of Mo in the bimetallic catalyst affects not only the wall numbers and diameters of the CNTs but also the yield of the CNTs. We changed the Mo content while fixing the molar ratio of Fe:MgO in Fe–Mo/MgO catalysts. TGA has been used to study the growth yield of CNTs obtained from Fe 0.1 Mo 0.01 Mg 2 O 2.18 , Fe 0.1 Mo 0.05 Mg 2 O 2.3 , Fe 0.1 Mo 0.1 Mg 2 O 2.45 , Fe 0.1 Mo 0.15 Mg 2 O 2.6 , and Fe 0.1 Mo 0.2 Mg 2 O 2.75 . As shown in Fig. 4 , the samples underwent mass losses of 21.2%, 40.5%, 45.7%, 59.9% and 64.8%, respectively. From the figure, it is found that almost no mass losses appeared before 400 °C and this indicated that the absence of water and amorphous carbon in the samples. The burning of CNTs themselves begins above 400 °C. The residue remains mainly due to the presence of catalyst. The yield of CNTs was calculated by the formula: yield (%) = 100 × M CNTs / M catalyst , where M CNTs is the mass loss above 400 °C and M catalyst is the mass remained. According to the formula, the yields of samples are 26.9%, 68.1%, 84.2%, 149.4% and 184.1%, respectively. It can be seen that the yield of the CNTs increased with increasing the Mo content. That is, the addition of Mo promotes the formation of CNTs. In order to understand the above result, the morphology and structures of these three catalysts before and after reduction were monitored by SEM and XRD. The SEM observations as shown in Fig. 5 a–c clearly indicate that the three catalysts have similar particles size before reduction by hydrogen. It was found that the average sizes of catalyst (mainly MgO microcrystal) tended to decrease with the addition of Mo to the catalysts ( Fig. 5 d–f) because of the avalanche-like reduction of MgMoO 4 in the bimetallic catalysts [50] . The reduced size of MgO particles will make metal ions in the catalyst much easy to be reduced by hydrogen thoroughly, which benefits the growth yield of small diameter nanotubes and SWCNTs. This feature was shown more clearly in Fig. 6 by growing the CNTs only for 1 min. Few of CNTs appeared from Fig. 6 a when the monometallic Fe catalyst was used (Fe 0.1 Mg 2 O 2.15 ). No CNTs were observed on pure Mo catalyst (Mo 0.1 Mg 2 O 2.3 ) although the catalyst particles are small ( Fig. 6 b). However, the yield of CNTs with small diameter is greatly enhanced by the addition of Mo (Fe 0.1 Mo 0.1 Mg 2 O 2.15 ), as indicated in Fig. 6 c. XRD analysis of the catalysts before and after CNT growth was further applied to monitor the structural changes of the catalysts. Fig. 7 shows the XRD patterns of Fe 0.1 Mg 2 O 2.15 , Mo 0.1 Mg 2 O 2.3 and Fe 0.1 Mo 0.1 Mg 2 O 2.45 catalysts before reduction. MgO/Fe 3 O 4 , MgO/MgMoO 4 , and MgO/Fe 3 O 4 /MgMoO 4 phases were found in the above three catalysts, respectively. It was noticed from Fig. 7 that the intensity of MgO in Mo 0.1 Mg 2 O 2.3 and Fe 0.1 Mo 0.1 Mg 2 O 2.45 drastically decreased comparing with Fe 0.1 Mg 2 O 2.15 , and the half-width of the peak become broad due to the presence of Mo in the catalysts because of the formation of MgMoO 4 . After reduction at 1000 °C for 5 min no Fe and Mo diffraction except MgO were detected in all the three catalysts. The possible reason is that the particles size of the Fe and Mo are too small to be detected (under the detect limitation). In order to confirm this, a catalyst with high content of Fe and Mo (Fe 0.1 Mo 0.1 MgO 1.45 , the molar ratio of Fe:Mo:MgO is 1:1:10) was then prepared and further examined by XRD as shown in Fig. 8 . Metal Mo peaks appeared but no metal Fe was traced except small amount of Mg 1.55 Fe 1.6 O 4 complex after reduction as shown in the Fig. 8 b. The big different behavior of Fe and Mo during the reduction process is believed to be related to the different interaction between them and MgO. The ferroferric oxide created from the combustion soaked into the MgO crystal at high temperature, resulting in the dispersion of Fe good enough to be reduced to form nano-sized particles even smaller than the detection limitation of XRD. These small metal Fe nanoparticles are suitable for the growth of small diameter CNTs and even SWCNTs. Detailed XRD investigation in the reduction of Fe/MgO catalyst showed the existence of some complexes such as (MgO) 0.91 (FeO) 0.09 during reduction process, which further confirms the good dispersion of Fe species in the MgO support. However, on the other hand, the good dispersion makes them difficult to be reduced, particularly when the catalyst particle size is large. Thus the utilized efficiency of Fe is crucial for Fe/MgO catalyst. For Mo/MgO catalyst, stoichiometric compound MgMoO 4 can be formed during the combustion and calcination process. MgMoO 4 can be easily reduced to be larger Mo metal particles to give XRD peaks and relative large diameter CNTs finally. In bimetallic Fe–Mo/MgO catalyst, the addition of Mo and the formation of MgMoO 4 in the catalyst make the catalyst particle much smaller than Fe/MgO catalyst because of the avalanche-like reduction of MgMoO 4 during reduction process. Thus the ferroferric oxide can be more easily reduced and the utilized efficiency of Fe could be enhanced greatly. As the increase of Mo content in Fe–Mo/MgO catalyst, Mo can catalyze the growth of CNTs independently, result in the increase of the growth yield and the average diameter of the CNTs. After growth of CNTs, strong Mo 2 C peaks can be observed as shown in Fig. 8 c. As most researchers found that carbides such as Fe carbide, Co carbide, Ni carbide and Mo carbide [37,40,42,51] were formed during the growth of CNTs and they were believed to play an important role in the CNT growth. 4 Conclusions The Fe–Mo/MgO bimetallic catalysts with different Mo content were used as catalysts for CNT growth by catalytic decomposition of methane. The number of the CNT walls, CNT diameters and the growth yield can be controlled by changing the content of Mo in the Fe–Mo/MgO catalyst. Increasing the Mo content in the Fe–Mo/MgO bimetallic catalysts resulted in an increase in wall number, diameters and growth yield of CNTs. It was found that the catalytic behavior of the active metals during the CVD process strongly depended on the interaction between the active metals and the support, size and dispersion of active metals in the catalysts. Fe and Mo interact with MgO in different ways. The Fe interacts with MgO to form complex (MgO) x (FeO) 1− x (0 < x < 1) and is then dispersed into the MgO support uniformly, resulting in very small Fe nanoparticles after reduction by hydrogen. The Mo interacts with MgO to form stoichiometry compound MgMoO 4 and relative large metal Mo particles can be generated after reduction. As a result, long and small diameter CNTs with fewer defects can be obtained from Fe 0.1 Mg 2 O 2.15 (without Mo). Relative large diameter CNTs with high yield can be grown from Mo 0.1 Mg 2 O 2.3. However, high yield CNTs with small diameter can be generated from Fe–Mo/MgO because the avalanche-like reduction of MgMoO 4 makes the catalyst particles to be small thus enhances the utilize efficiency of Fe nanoparticles. The understanding of the different behaviors of Fe and Mo in the bimetallic catalyst could provide a way to design more active catalysts for controllable synthesis of desirable CNTs. Acknowledgments The author would like to thank STBWZ (G20070083), NSFC (50772076), NSFZJ (R4090137) and ZJED Innovative Team for Huang S. for their financial support. References [1] S. Iijima Nature 354 1991 56 58 [2] A. Vijayaraghavan K. Kanzaki S. Suzuki Y. Kobayashi H. Inokawa Y. Ono S. Kar P.M. Ajayan Nano Lett. 5 8 2005 1575 1579 [3] S.F. Wang L. Shen W.D. Zhang Y.J. Tong Biomacromolecules 6 2005 3067 3072 [4] X. Gong J. Liu S. Baskaran R.D. Voise J.S. Young Chem. Mater. 12 4 2000 1049 1052 [5] X. Gui J. Wei K. Wang A. Cao H. Zhu Y. Jia Q. Shu D. Wu Adv. Mater. 22 2010 617 621 [6] R.H. Baughman A.A. Zakhidov W.A. de Heer Science 297 2002 787 792 [7] T.W. Ebbesen P.M. Ajayan Nature 358 6383 1992 220 222 [8] M. Ishigami J. Cumings A. Zettl S. Chen Chem. Phys. Lett. 319 2000 457 459 [9] T. Guo P. Nikolaev A.G. Rinzler D. Tomanek D.T. Colbert R.E. Smalley J. Phys. Chem. B 99 27 1995 10694 10697 [10] A. Thess R. Lee P. Nikolaev H.J. Dai P. Petit J. Robert C. Xu Y.H. Lee S.G. Kim A.G. Rinzler D.T. Colbert G.E. Scuseria D. Tomanek J.E. Fischer R.E. Smalley Science 273 1996 483 487 [11] M.J. Yacaman M.M. Yoshida L. Rendon J.G. Santiesteban Appl. Phys. Lett. 62 1993 202 204 [12] H. Jaeger T. Behrsing Compos. Sci. Technol. 51 1994 231 242 [13] G.L. Hornyak L. Grigorian A.C. Dillon P.A. Parilla K.M. Jones M.J. Heben J. Phys. Chem. B 106 11 2002 2821 2825 [14] T. Hiraoka S. Bandow H. Shinohara S. Iijima Carbon 44 9 2006 1853 1859 [15] J.A. Kong A.M. Cassell H.J. Dai Chem. Phys. Lett. 292 1998 567 574 [16] P. Nikolaev M.J. Bronikowski R.K. Bradley F. Rohmund D.T. Colbert K.A. Smith R.E. Smalley Chem. Phys. Lett. 313 1999 91 97 [17] A. Gorbunov O. Jost W. Pompe A. Graff Carbon 40 1 2002 113 118 [18] H.M. Cheng F. Li X. Sun S.D.M. Brown M.A. Pimenta A. Marucci G. Dresselhaus M.S. Dresselhaus Chem. Phys. Lett. 289 1998 602 610 [19] S.B. Sinnott R. Andrews D. Qian A.M. Rao Z. Mao E.C. Dickey F. Derbyshire Chem. Phys. Lett. 315 1999 25 30 [20] S.M. Huang M. Woodson R.E. Smalley J. Liu Nano Lett. 4 6 2004 1025 1028 [21] G.X. Du S.A. Feng J.H. Zhao C. Song S.L. Bai Z.P. Zhu J. Am. Chem. Soc. 128 48 2006 15405 15414 [22] H.J. Jeong K.K. Kim S.Y. Jeong M.H. Park C.W. Yang Y.H. Lee J. Phys. Chem. B 108 46 2004 17695 17698 [23] H. Qi C. Qian J. Liu Chem. Mater. 18 24 2006 5691 5695 [24] W.L. Wang X.D. Bai Z. Xu S. Liu E.G. Wang Chem. Phys. Lett. 419 2006 81 85 [25] Y.C. Choi Y.M. Shin Y.H. Lee B.S. Lee G.S. Park W.B. Choi N.S. Lee J.M. Kim Appl. Phys. Lett. 76 2000 2367 2369 [26] E. Dervishi Z. Li A.R. Biris D. Lupu S. Trigwell A.S. Biris Chem. Mater. 19 2 2007 179 184 [27] V.Z. Mordkovich E.A. Dolgova A.R. Karaeva D.N. Kharitonov I.A. Maslov A.A. Kamenev V.F. Tretjakov Carbon 45 1 2007 62 69 [28] W. Zhou Z. Han J. Wang Y. Zhang Z. Jin X. Sun Y. Zhang C. Yan Y. Li Nano Lett. 6 12 2006 2987 2990 [29] K. Mizuno K. Hata T. Saito S. Ohshima M. Yumura S. Iijima J. Phys. Chem. B 109 7 2005 2632 2637 [30] D. Takagi Y. Homma H. Hibino S. Suzuki Y. Kobayashi Nano Lett. 6 12 2006 2642 2645 [31] P. Ball Nature 414 2001 142 144 [32] G. Ning F. Wei Q. Wen G. Luo Y. Wang Y. Jin J. Phys. Chem. B 110 3 2006 1201 1205 [33] H.M. Christen A.A. Puretzky H. Cui K. Belay P.H. Fleming D.B. Geohegan D.H. Lowndes Nano Lett. 4 10 2004 1939 1942 [34] T. Saito W.-C. Xu S. Ohshima H. Ago M. Yumura S. Iijima J. Phys. Chem. B 110 12 2006 5849 5853 [35] H. Ago N. Uehara N. Yoshihara M. Tsuji M. Yumura N. Tomonaga T. Setoguchi Carbon 44 14 2006 2912 2918 [36] E. Lamouroux P. Serp Y. Kihn P. Kalck Appl. Catal. A 323 2007 162 173 [37] H. Yoshida T. Shimizu T. Uchiyama H. Kohno Y. Homma S. Takeda Nano Lett. 9 11 2009 3810 3815 [38] M. Hu Y. Murakami M. Ogura S. Maruyama T. Okubo J. Catal. 225 2004 230 239 [39] H. Ago S. Ohshima K. Tsukuagoshi M. Tsuji M. Yumura Curr. Appl. Phys. 5 2 2005 128 132 [40] H. Sugime S. Noda S. Maruyama Y. Yamaguchi Carbon 47 1 2009 234 241 [41] Y. Li X.B. Zhang X.Y. Tao J.M. Xu W.Z. Huang J.H. Luo Z.Q. Luo T. Li F. Liu Y. Bao H.J. Geise Carbon 43 2 2005 295 301 [42] L.-P. Zhou K. Ohta K. Kuroda N. Lei K. Matsuishi L. Gao T. Matsumoto J. Nakamura J. Phys. Chem. B 109 10 2005 4439 4447 [43] H. Hiura T.W. Ebbesen K. Tanigaki H. Takahashi Chem. Phys. Lett. 202 6 1993 509 512 [44] C.Y. Jiang J.L. Zhao H.A. Therese M. Friedrich A. Mews J. Phys. Chem. B 107 34 2003 8742 8745 [45] S. Reich C. Thomsen Philos. Trans. R. Soc. Lond. A 362 2004 2271 2288 [46] S.K. Doorn L.X. Zheng M.J. O’Connell Y.T. Zhu S.M. Huang J. Liu J. Phys. Chem. B 109 9 2005 3751 3758 [47] Y.Y. Zhang H.B. Son J. Zhang M.S. Dresselhuas J. Kong Z.F. Liu J. Phys. Chem. B 111 5 2007 1983 1987 [48] A.M. Rao E. Richter S. Bandow B. Chase P.C. Eklund K.A. Williams S. Fang K.R. Subbaswamy M. Menon A. Thess R.E. Smalley G. Dresselhaus M.S. Dresselhaus Science 275 1997 187 191 [49] S. Bandow S. Asaka Phys. Rev. Lett. 80 17 1998 3779 3782 [50] Z. Shang S. Huang X. Xu J. Chen Mater. Chem. Phys. 114 1 2009 173 178 [51] M. Lin J.P.Y. Tan C. Boothroyd K.P. Loh E.S. Tok Y.-L. Foo Nano Lett. 6 3 2006 449 452