Hydrogen storage of dense-aligned carbon nanotubes

Hydrogen storage of dense-aligned carbon nanotubes

20 July 2001 Chemical Physics Letters 342 (2001) 510±514 www.elsevier.com/locate/cplett Hydrogen storage of dense-aligned carbon nanotubes Anyuan C...

228KB Sizes 0 Downloads 58 Views

20 July 2001

Chemical Physics Letters 342 (2001) 510±514


Hydrogen storage of dense-aligned carbon nanotubes Anyuan Cao *, Hongwei Zhu, Xianfeng Zhang, Xuesong Li, Dianbo Ruan, Cailu Xu, Binqing Wei, Ji Liang, Dehai Wu Department of Mechanical Engineering, State Key Laboratory of Automotive Safety and Energy, Tsinghua University, Beijing 100084, People's Republic of China Received 18 January 2001

Abstract It has been considered so far that the inner cavities of carbon nanotubes are the desired place for the storage of hydrogen molecules. Here we show that the inter-nanotube space between densely aligned carbon nanotubes, produced by catalytic pyrolysis of ferrocene, could also contribute to the e€ective uptake of molecular hydrogen. This provides a new way to improve the properties of carbon nanotubes in storing hydrogen or other elemental molecules. Ó 2001 Elsevier Science B.V. All rights reserved.

1. Introduction Since their discovery in 1991 [1], carbon nanotubes have shown unique and attractive properties, such as in mechanical [2], electrical [3,4] and thermal [5] aspects. Applications of nanotubes in ®eld emission ¯at panel displays [6,7], advanced scanning probes [8], chemical molecular sensors [9] and hydrogen storage [10±12] have been reported in recent years. In particular, single-walled carbon nanotubes (SWNTs) have shown a hydrogen storage density of 5±10 wt% at 293 K [10], higher than traditional materials. Therefore carbon nanotubes could be used as an e€ective energy carrier in fuel cell vehicles. Hydrogen molecules could be stored inside the tube-cavities [13], between the graphitic layers [14] or in the inter-tube spacing of SWNT bundles [15]. Large-scale production of multi-walled nanotubes (MWNTs) is available


Corresponding author. E-mail address: [email protected] (A. Cao).

now [16], which enables preparation of enough material for hydrogen storage. However, it is dif®cult for the MWNTs to meet the general requirement of 6.5 wt% in typical fuel cell systems. Although SWNTs have a predicted storage capacity of 14 wt% [13], it is still important to increase the hydrogen storage density of MWNTs, as they could be produced on a large scale in a cost-e€ective way. In this Letter, we ®nd that the aligned MWNTs have a higher hydrogen adsorption than non-aligned ones, which is attributed to the inter-nanotube space between adjacent parallel nanotubes. Controllable synthesis of aligned nanotubes with a desired nanotube density would improve the hydrogen storage capacity of MWNTs. 2. Experimental Well-aligned carbon nanotubes have been synthesized by catalytic pyrolysis of ferrocene [16,17]. We have also obtained vertical carbon nanotube

0009-2614/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 9 - 2 6 1 4 ( 0 1 ) 0 0 6 1 9 - 4

A. Cao et al. / Chemical Physics Letters 342 (2001) 510±514

arrays on quartz substrates by this method [18,19]. A detailed description of the experimental process has been reported in [19]. The density of aligned carbon nanotubes increases with reaction temperature from 700°C to 900°C. With a temperature higher than 900°C, both the alignment and growth rate of nanotubes decreases considerably. The aligned and random nanotube samples were obtained at 900°C, peeled o€ the plane quartz sheets and the internal wall of the reaction tube, respectively. SEM images reveal that the two samples have very di€erent degrees of alignment. Whereas nanotubes grown on the plane quartz sheets are well-aligned with a high density, those on the internal reaction tube wall consist of only randomly distributed nanotubes (Fig. 1). The aligned nanotube ®lms divided into numerous smaller blocks when they were peeled o€ the substrate, but the nanotubes in each block kept the original alignment (Fig. 1a). The block size ranges from several to tens of micrometers with an estimated nanotube


density of about 100 lm 2 . The average nanotube length, or the block thickness, is 250 lm for a reaction time of 80 min. The roof and bottom of the aligned nanotube blocks is very smooth, allowing for the entrance of hydrogen molecules from both sides during hydrogen storage. Diameters of the MWNTs in both samples range from 10 to 40 nm and they have relatively smaller inner cavities (Fig. 1c). Transmission electronic microscopy (TEM) examinations of the two samples reveal that carbon nanotubes have closed ends. Random growth of nanotubes on the tube wall is probably due to the unstable gas ¯ow near its surface, resulting in concentration ¯uctuation of the catalytic and carbon source. Hydrogen storage testing was performed on the two samples without any other pre-treatment except for drying at 100°C for 24 h, in order to keep the original orientation of nanotubes. Hydrogen adsorption of the aligned and random samples (100 mg for each) at various pressures was

Fig. 1. A block of densely aligned carbon nanotubes peeled o€ from the quartz sheet (a) and the random nanotubes grown on the internal reaction tube wall (b). The horizontal parallel lines on the block side are due to the presence of Fe particles in the nanotubes [19]. A MWNT with outer and inner tube diameters of 45 and 5 nm, respectively (c).


A. Cao et al. / Chemical Physics Letters 342 (2001) 510±514

Fig. 2. Hydrogen adsorption of the aligned and random samples. The random sample shows only a weak adsorption even at a high pressure of 10 MPa, while the H2 adsorption of the aligned sample increases monotonically with the hydrogen pressure.

measured at 290 K (Fig. 2). The random sample shows only a weak hydrogen adsorption of 0.68 wt% at 10 MPa. However, the aligned sample shows a higher adsorption than the random one, which increases with hydrogen pressure and reaches 2.4 wt% at 10 MPa. The apparatus used for the H2 adsorption experiments likes that described elsewhere [14]. 3. Results and discussions Speci®c surface area was measured for the aligned and random samples (by Sorptomatic 1990, ThermoQuest Italia S.p.A.), which was calculated to be 60 and 27 m2 =g, respectively. The larger surface area of the aligned sample indicates that it has more micro-pores than the random one. Size distribution of pores within 100 nm shows that the aligned sample has a higher volume percentage of pores with sizes less than 50 nm compared to the random one (Fig. 3). This is because aligned nanotubes are more closely arranged. After the adsorption and desorption of hydrogen at 10 MPa, the pore distribution of the aligned sample was measured again. The size of most of the pores in the sample was reduced to less than 10

Fig. 3. Pore width distribution within 100 nm in the two samples consisting of the random and aligned nanotubes (before and after the hydrogen storage testing), respectively. The curves of the random and aligned sample (before storage) are enlarged three times.

nm, and there is a strong peak at about 7 nm and a moderate peak at 3 nm (Fig. 3). The reduction of pore sizes may be due to the hydrogen pressure applied on the sample when H2 ®lls the vacuum chamber, squeezing aligned nanotubes into a more close arrangement and therefore compressing the pores. Pores in the aligned sample exist mainly as cylindrical micro-channels formed by adjacent parallel nanotubes, while in the random sample they are generally irregular. The length and diameters of such micro-channels are equal to the block thickness and the distance between adjacent nanotubes, respectively (Fig. 4). As-grown aligned nanotubes contain a large number of such channels with diameters ranging from 3 to 100 nm and they were further squeezed by the hydrogen pressure. The micro-long and narrow channels in the aligned sample could be compressed much more easily than the irregular pores in the random sample. The compression makes the aligned sample favorable to store hydrogen molecules, because the micro-channels with a larger size could not uptake H2 . The critical size for the micro-channels to uptake hydrogen molecules e€ectively has not

A. Cao et al. / Chemical Physics Letters 342 (2001) 510±514

Fig. 4. A schematic view of the cross-section of as-grown aligned nanotube ®lms on the quartz substrate, showing microchannels between adjacent parallel nanotubes (arrow c points). The channel length is equal to the ®lm thickness (L), or the nanotube length. Hydrogen molecules (arrow d) would enter the numerous micro-channels present in the densely aligned nanotubes (see the dashed arrows). Arrows a and b refer to the closed ends and the Fe particles on the ends of nanotubes, respectively, which hamper H2 entrance during the adsorption experiments.

been determined yet, but we suppose it should be less than 10 nm. The as-grown aligned nanotubes generally have closed ends when their growth is stopped by the chemical vapor deposition (CVD) method. The samples have not been subject to treatment such as oxidation or acid washing, which could open their ends e€ectively, so by our way carbon nanotubes still have closed ends during hydrogen adsorption. The random sample shows only a weak adsorption value because H2 could not enter nanotubes through their ends. Fe particles present inside the nanotube-cavities would hamper the entrance of H2 also. However, the aligned sample shows a higher H2 adsorption at the same pressures. As both of the two samples consist of carbon nanotubes with closed ends, the di€erence in hydrogen storage is due to the alignment degree of nanotubes in them. So we attribute the high H2 adsorption to the e€ective uptake of H2 by the inter-nanotube space, that is, the micro-channels between the aligned carbon nanotubes. Moreover, H2 adsorption of the aligned sample increases monotonically with the hydrogen pressure. The micro-channels should be compressed to a higher degree at a high pressure and their size decreases further, which could uptake H2 more e€ectively. However, the H2 storage capacity is 2.4%, lower than that reported for SWNT bundles [10,11]. Although there are also many micro-pores in


random nanotubes, hydrogen molecules could easily escape from these irregular and shallow pores. Hydrogen molecules could be physically or chemically adsorbed on the internal or external nanotube walls [13,20]. They may be adsorbed by the outer-walls of nanotubes and stay in the micro-channels between them. As these microchannels have a high aspect ratio, of the same magnitude as individual nanotubes, adsorbed H2 inside them could not escape easily due to the capillary e€ect. The number and diameter of these micro-channels are two critical factors determining the storage capacity of molecular hydrogen. Opening the closed nanotube ends by oxidation would improve the hydrogen storage of aligned nanotubes.

4. Conclusion In summary, dense alignment of nanotubes provides a potential way to improve their hydrogen storage property by adsorbing molecules on the outer-walls as well as inner-walls. This way involves controllable synthesis of aligned nanotubes with desired length and density. Nanotube length is determined by the reaction time in our way (the average nanotube growth rate is about 3±5 lm/min), and the density depends on the reaction temperature and ferrocene concentration. Optimum size range of micro-channels among parallel nanotubes, including length and diameter, should be de®ned and the inter-nanotube adsorption mechanism remains unclear, which needs further study.

Acknowledgements The authors wish to thank Mr. Y.J. Yan and Professor L.P. You for SEM and HRTEM study of the carbon nanotube samples. This work was supported under the State Key Program for Fundament Research of MOST, China (Grant No. G20000264-04) and the Doctor Dissertation Foundation of Tsinghua University.


A. Cao et al. / Chemical Physics Letters 342 (2001) 510±514

References [1] S. Iijima, Nature 354 (1991) 56. [2] P. Poncharal, Z.L. Wang, D. Ugarte, W.A. de Heer, Science 283 (1999) 1513. [3] H.J. Dai, E.W. Wong, C.M. Lieber, Science 272 (1996) 523. [4] B. Wei, R. Spolenak, P. Kohler-Redlich, M. R uhle, E. Arzt, Appl. Phys. Lett. 74 (1999) 3149. [5] J. Hone, M.C. Llaguno, N.M. Nemes, A.T. Johnson, J.E. Fisher, D.A. Walters, M.J. Casavant, J. Schmidt, R.E. Smalley, Appl. Phys. Lett. 77 (2000) 666. [6] H. Murakami, M. Hirakawa, C. Tanaka, H. Yamakawa, Appl. Phys. Lett. 76 (2000) 1776. [7] W.B. Choi, D.S. Chung, J.H. Kang, H.Y. Kim, Y.W. Jin, I.T. Han, Y.H. Lee, J.E. Jung, N.S. Lee, G.S. Park, J.M. Kim, Appl. Phys. Lett. 75 (1999) 3129. [8] H.J. Dai, J.H. Hafner, A.G. Rinzler, D.T. Colbert, R.E. Smally, Nature 384 (1996) 147. [9] J. Kong, N.R. Franklin, C.W. Zhou, M.G. Chapline, S. Peng, K. Cho, H.J. Dai, Science 287 (2000) 622.

[10] A.C. Dillon, K.M. Jones, T.A. Bekkedahl, C.H. Kiang, D.S. Bethune, M.J. Heben, Nature 386 (1997) 377. [11] C. Liu, Y.Y. Fan, M. Liu, H.T. Cong, H.M. Cheng, M.S. Dresselhaus, Science 286 (1999) 1127. [12] N. Rajalakshmi, K.S. Dhathathreyan, A. Govindaraj, B.C. Satishkumar, Electrochim. Acta 45 (2000) 4511. [13] S.M. Lee, Y.H. Lee, Appl. Phys. Lett. 76 (2000) 2877. [14] A. Chambers, C. Park, R.T.K. Baker, N.M. Rodriguez, J. Phys. Chem. B 102 (1998) 4253. [15] F. Darkrim, D. Levesque, J. Chem. Phys. 109 (1998) 4981. [16] B.C. Satishkumar, A. Govindaraj, C.N.R. Rao, Chem. Phys. Lett. 307 (1999) 158. [17] C.N.R. Rao, R. Sen, B.C. Satishkumar, A. Govindaraj, Chem. Commun. (1998) 1525. [18] A. Cao, L. Ci, D. Li, B. Wei, C. Xu, J. Liang, D. Wu, Chem. Phys. Lett. 335 (2001) 150. [19] A. Cao, L. Ci, G. Wu, B. Wei, C. Xu, J. Liang, D. Wu, Carbon 39 (2001) 152. [20] H. Zhu, A. Chen, Z. Mao, C. Xu, X. Xiao, B. Wei, J. Liang, D. Wu, J. Mater. Sci. Lett. 19 (2000) 1237.