Hydrogen storage in coiled carbon nanotubes

Hydrogen storage in coiled carbon nanotubes

international journal of hydrogen energy 35 (2010) 1313–1320 Available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/he Hydrog...

1MB Sizes 3 Downloads 95 Views

international journal of hydrogen energy 35 (2010) 1313–1320

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Hydrogen storage in coiled carbon nanotubes V. Gayathri*, N.R. Devi, R. Geetha Department of Physics, Thiagarajar College of Engineering, Madurai – 625 015, India

article info

abstract

Article history:

We report a density functional calculation of the adsorption of molecular hydrogen on the

Received 13 October 2009

external surface of coiled carbon nanotube (CCNT). Binding energies of single molecule

Received in revised form

have been studied as a function of three different orientations and at three different sites

21 November 2009

like hexagon, pentagon and heptagon. The binding energy values are larger than linear (5,5)

Accepted 22 November 2009

armchair nanotube, which has approximately same diameter as that of coiled carbon

Available online 17 December 2009

nanotube. The curvature and topology of CCNT are responsible for this considerable enhancement. The system with full coverage is also studied. When the nanotube surface is

Keywords:

fully covered with one molecule per graphitic hexagon, pentagon and heptagon gives the

Carbon nanotubes

6.8 wt% storage capacity. The binding energy per molecule decreases due to repulsive

Coiled carbon nanotubes

interactions between neighbor molecules. It gives good storage medium for hydrogen.

Adsorption

Almost it meets the DOE target.

Hydrogen storage

1.

ª 2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.

Introduction

Shrinking resources of fossil fuels and global warming leads to the undesired climate changes. This necessitates alternate sources of energy carrier such as hydrogen (H2), one of the profuse element present in the universe. It is wellknown that, hydrogen is predicted to be an efficient fuel due to lot of interesting properties like high-energy content, lightweight and pollution free [1,2]. However, the explosive nature and large volume are the two main hindrances for the transportation and storage purposes, which constraints the development of hydrogen economy. Currently, hydrogen fuel is stored as gas on solid adsorption, compressed or liquefied form in metal hydrides. So far no method could achieve hydrogen storage capacity of 6.5 wt% recommended by U.S Department of Energy (DOE) for automobile applications. Hence the search is still going on to identify a suitable material and storage technique to achieve this target level. Materials at nanoscale prove to be promising contenders for this race as these materials exhibit novel physical and

chemical properties and possess nanosize. Carbon nanotubes (CNTs) one of the versatile material [3] with high surface area, porous nature, low density, high strength and nanosize is predicted to have tremendous potential application in energy storage apart from its considerable role in molecular electronics and nanosensors etc., CNTs are rolled up sheets of graphite into seamless cylinder discovered by Sumio Iijima in 1991 [4]. CNTs are classified as multi walled nanotubes (MWNT), single walled nanotubes (SWNT), double walled nanotubes (DWNT) and coiled carbon nanotubes (CCNT). Depending on the rolling, CNTs can be further classified as armchair, zigzag and chiral types. Of these only armchair nanotubes are truly metallic [5], while other tubes zigzag as well as chiral 1/3 are narrow-gap and 2/3 wide-gap semiconductors [6]. Several authors analyzed the storage of hydrogen through gas adsorption in carbon nanotubes. Ab initio and density functional theory (DFT) studies of molecular hydrogen interaction in achiral single-wall carbon nanotubes are carried by Ferre-Vilaplana et al. and Arellano et al. [7,8]. Ju li et al. [9] and Yuchen Ma et al. [10] performed the theoretical calculations

* Corresponding author. Tel.: þ91 452 2482240; fax: þ91 452 2483427. E-mail address: [email protected] (V. Gayathri). 0360-3199/$ – see front matter ª 2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2009.11.083

1314

international journal of hydrogen energy 35 (2010) 1313–1320

Table 1 – Parameters of coiled carbon nanotube. Parameters

Coiled carbon nanotube One pitch (nm)

Tube diameter Length of the tube No of carbon atoms No of pentagons No of heptagons Height of one pitch

Fig. 1 – Three different configurations of Hydrogen molecule.

on storage capacity of hydrogen in single walled carbon nanotubes. Experimental investigation of hydrogen adsorption through the monolayer surface of carbon nanotubes was studied by Zuttel et al. [11]. It has been reported that the storage capacity of carbon nanofibers at 10.5 MPa is 0.7 wt% [12] and in CNT film is 8 wt% [13] at room temperature and ambient pressure. Recently, Reddy et al. [14] reported the platinum dispersion on SWNT enhances the storage capacity. Gayathri et al. [15] have studied the H2 adsorption in defected CNTs. It has been observed both theoretically and experimentally that hydrogen can be adsorbed on carbon nanotubes in two different ways namely physisorption and chemisorption, where the former one is suitable for hydrogen storage. We have three different molecular configurations such X, Y and Z, where H2 molecular axis positioned i) parallel and above C–C bond (X) ii) perpendicular and above carbon ring and (Y) iii) parallel and above carbon ring respectively (Z) shown in Fig. 1. Active sites, molecular configurations, defects and dopant are some of the important factors that strongly affect the hydrogen adsorption in CNTs. In this paper, we have investigated the effect of all the above factors on hydrogen adsorption in the un-doped coiled carbon nanotubes (CCNTs) an innovative material predicted to have tremendous potential applications [16]. Dunlap [17] discussed the dramatic tubule transformation from the linear tube by the joining together crenelated (L, L) and a sawtooth tubule (2L, 0) of appropriate diameters using a single pentagon and single heptagon. The same mechanism followed by Fonseca et al. [18] which is between the chiral and achiral nanotubes. CCNTs are first observed experimentally by Amelinkx et al. [19]. CCNTs are one of the classifications of carbon

a

0.64 (0.65 ) 1.6 112 4 4 1.6 (1.11a)

Three pitches (nm) 0.64 (0.65a) 5.319 336 12 12 1.6 (1.11a)

a Tube diameter and pitch values obtained by the L.P. Biro et al. [25] studied the STM observation of tightly wound Single-Wall coiled carbon nanotube.

nanotubes with spiral like structure predicted to exhibit unique physical and chemical properties as that of its linear tubes. The occurrence of heptagons and pentagons along with the hexagonal network gives raise to coiled (helical) structure of the CNTs. In recent times there is a lot of synthesis and characterization studies exist on CCNTs [20,21]. This structure is mechanically stable under axial, shearing, torsional and bending forces. CCNT exhibits two kinds of chirality’s one are geometrical and other structural type. Kazuto et al. [22,23] have discussed about the electronic band structure of CCNT and observed the semimetallic characteristics, in addition to metallic and semiconducting behaviors. This is due to the fact that according to the simple tight-binding models, the electronic structures of coiled nanotubes vary with the position of pentagon–heptagon pairs of the constituent carbon nanotubes. In coiled tubes pitch and coil diameter are the two important parameters that decides the physical and chemical properties. Defects are invariably present in all coiled tubes unlike the production of pure linear tubes with all hexagonal rings of carbon. Coiled carbon nanotubes show excellent mechanical, electrical and magnetic properties. It has remarkable applications in nanocomposites, nanoelectronics and nanoelectromechnical devices. Here, it is interesting to note that coiled form of CNTs is predicted to be a suitable material for hydrogen adsorption and storage. Since, CCNTs has very high surface area and contains large number of defects both important for enhanced hydrogen adsorption. In the current scenario there are lot of research papers devoted to the different synthesis methods and construction mechanism of CCNT. The details of CCNT construction for the present work are given in the next section.

Fig. 2 – Horizontal view of the constructed Coiled Carbon Nanotube (CCNT) with three pitches.

1315

international journal of hydrogen energy 35 (2010) 1313–1320

Table 2 – Adsorption binding energy (eV) in CCNT for three different configurations at three different sites. Hydrogen molecule configuration with respect to Molecular axis Site

Hexagon Pentagon Heptagon

Parallel to C–C bond (X)

Perpendicular and above carbon ring (Y)

Parallel and above carbon ring (Z)

0.080 (3) 0.086 (2.94) 0.098 (3.49)

0.125 (2.85) 0.104 (3.1) 0.114 (2.79)

0.122 (2.7) 0.074 (3.6) 0.127 (2.81)

˚ ) are quoted in brackets. Note: H2-CCNT wall separation values (A

Fig. 3 – H2 adsorption for configuration (Y) with respect to Hexagon in CCNT.

2.

Construction of geometry

Several methods exist in literature for the construction of an optimized coiled structure [17,18] consisting of seven, six and five fold rings of carbon atoms. We have constructed coiled carbon nanotube by taking single layer of (7,0) sawtooth tube and affixing polygons periodically that is in this structure one of hexagon is replaced with pentagon. It gives small amount of bending (convex surface) in the layer.

In order to get concave surface the heptagon is introduced just opposite to that of pentagon. Then the second row of hexagons is introduced in the optimal position along the circumference to get the symmetry of the structure. To get high curvature further heptagon, pentagon and hexagons are added in the appropriate place, which is straight to that of previously introduced pentagon and heptagon. The above mechanism followed reversibly to get the coiled structure. This process can be repeated to increase the number of pitches in CCNT. The coiled carbon nanotube constructed with the above mechanism has the diameter of 0.64 nm including three pitches and pitch length 1.6 nm. It composed of 336 carbon atoms and 12 numbers of pentagon and heptagon. Ihara et al. [24] made the simulation studies on helically coiled cage derived from toroidal structures and declared the helix C540 has 0.6 nm diameter. Biro et al. [25,26] examined the structure of tightly wound single-wall coiled carbon nanotube through scanning tunneling microscopy and reported the diameters in the range of 0.5–0.9 nm having the pitch 0.9–1.25 nm. These tubes were synthesized by catalytic thermal decomposition of acetylene by Co nanoparticles

Fig. 4 – Binding energy variation of H2 as a function of separation for configuration (Y) with respect to Hexagon in CCNT.

1316

international journal of hydrogen energy 35 (2010) 1313–1320

Fig. 5 – H2 adsorption for configuration (Y) with respect to Pentagon in CCNT.

Fig. 7 – H2 adsorption for configuration (Z) with respect to Heptagon in CCNT. supported on silica. The CCNT that we have constructed for the present study has almost the same diameter and pitch value as reported in the above literature [24,25]. As a first case the adsorption binding energy calculations are carried out for a single pitch of CCNT. The simulated tube has 112 carbon atoms and four numbers of pentagon and heptagon. The horizontal view of the constructed tube is given in Fig. 2. The computational details are discussed in the next section and the various parameters of the constructed tube are listed in Table 1.

3.

Computational details

The molecular dynamics study on the adsorption binding energy of constructed coiled carbon nanotube was performed using DFT calculation with Dmol3. The total energy of the system was estimated by solving Kohn–Shams equation. The electron exchange and correlation effect are

Fig. 6 – Binding energy variation of H2 as a function of separation for configuration (X and Y) with respect to Pentagon in CCNT.

international journal of hydrogen energy 35 (2010) 1313–1320

1317

Fig. 8 – Binding energy variation of H2 as a function of separation for configuration (Y) with respect to Heptagon in CCNT.

introduced through Generalized Gradient Approximation (GGA). We have chosen the Perdew, Burke and Enzerhof (PBE) potential for the present study, as it is more suitable for adsorption energy calculation. We have used the Double Numerical Polarization (DNP) basis set and fixed Global ˚ for the hydrogen molecular adsorption Orbital Cutoff as 3 A on CCNT. The binding energy values are estimated for three different molecular configurations from the well-known equation, Eb ¼ EðCCNTÞ þ EðH2 Þ  EðCCNT þ H2 Þ

(1)

Where, E (CCNT þ H2), E (CCNT) and E (H2) are total energy of CCNT with H2 molecule, free coiled carbon nanotube, and single hydrogen molecule respectively. The energy values are minimized by adjusting the separation between the molecule and adsorption site from the wall of the CCNT. The binding energy of the system was calculated for single Hydrogen molecule for three different configurations namely X, Y and Z at three different sites like hexagon, pentagon and heptagon.

Fig. 9 – Horizontal view of full coverage of H2 adsorption for configuration (Y) in CCNT.

4.

Results and discussion

As given in the introduction, three different molecular configurations can be explained where H2 molecular axis positioned i) parallel and above C–C bond (X) ii) perpendicular and above carbon ring (Y) and iii) parallel and above carbon ring (Z). The hydrogen adsorption binding energy results on CCNT and the effect of various adsorption sites like hexagon, pentagon and heptagon carbon rings are discussed in this section.

4.1.

Site effect on CCNT

4.1.1.

Hexagon site

In the case of regular hexagon site in coiled carbon nanotube, the adsorption binding energy comes out to be 0.080 eV for the

Fig. 10 – Side view of full coverage of H2 adsorption for configuration (Y) in CCNT.

1318

international journal of hydrogen energy 35 (2010) 1313–1320

Fig. 11 – Full coverage Binding energy variation of H2 as a function of separation for configuration (Y) in CCNT.

molecular configuration X. We could obtain molecule-nano˚ . For the next configuratube surface separation value as 3 A tion Y the binding energy increases to a value of 0.125 eV. Third configuration Z gives a smaller value of 0.122 eV. These results show that configuration Y is more stable and the molecule is able to fit into the electron density valley that exists on the center of hexagon. It is interesting to note that for a linear (5,5) armchair tube of diameter same as that of the present case Alonso et al. [27] have reported the H2 adsorption binding energy as 0.070 eV for the Y configuration. In our earlier work [15] with linear CNT, we could show a binding energy value as 0.072 eV for the same configuration. Compared to these values CCNTs exhibit 80% increase in hydrogen adsorption. Fig. 3 shows the hydrogen-adsorbed tube for configuration (Y) with respect to hexagon. Fig. 4 shows the variation of adsorption binding energy as a function of separation for configuration (Y) with respect to

hexagon in CCNT. The estimated value of binding energy is given in Table 2.

4.1.2.

4.1.3.

Fig. 12 – Full coverage of H2 adsorption for configuration (Y) in (5,5) metallic tube.

Pentagon site

Pentagons and heptagons are considered as structural defects in linear CNTs. Whereas in CCNT they are part of a repeating unit. It is observed that the defects enhance the adsorption in linear tubes that motivates as to make the adsorption studies at these sites. For X configuration we could observe almost the same value as that of hexagon. As a special case of Z configuration we have considered where the H2 molecule is placed with its molecular axis parallel to the carbon ring and tube axis. Binding energy value slightly decreases for Y and Z configuration compared to hexagon site. But again Y configuration becomes a stable one for the pentagon site also. Our results show 50% increase in adsorption binding energy compared to linear tube. The hydrogen-adsorbed tube for configuration (Y) with respect to pentagon is given in Fig. 5. The graph shown in Fig. 6 gives the binding energy changes as a function of separation for configuration (X and Y) with respect to pentagon. The binding energy values are given in Table 2.

Heptagon site

For the heptagon site, we could obtain a little higher value of 0.098 eV compare to hexagon and pentagon for the X configuration. While, the Y configuration gives out a value of 0.114 eV which is almost 0.01 eV less than that of hexagon and greater than that of pentagon. Unlike the case of other sites Z configuration becomes a stable one for the heptagon with the binding energy of 0.127 eV, which is 0.002 eV higher than that of hexagon. We could observe an increase of 80% in the binding energy than the linear tube. Fig. 7 shows the H2 adsorbed tube for configuration (Z) with respect to heptagon. The calculated value of binding energy is given in Table 2.

1319

international journal of hydrogen energy 35 (2010) 1313–1320

Fig. 13 – Full coverage Binding energy variation of H2 as a function of separation for configuration (Y) in (5,5) metallic tube.

Table 3 – The length of tube (nm), diameter (nm), number of carbon atoms, number of H2 molecules, Binding energy (eV) and storage capacity (wt%) values in (5,5) and CCNT. Tube (5,5) 2UC CCNT one pitch

Length (nm)

Diameter (nm)

Number of carbon atoms

Number of H2 molecules

Binding energy (eV)

Storage capacity (wt%)

Tube length 0.38 Coil length 1.6

0.68 0.64

40 112

10 49

0.094 0.056

4 6.8

Binding energy changes as a function of separation for configuration (Y) with respect to heptagon is given in Fig. 8.

4.2.

Storage capacity

We have considered the full coverage of single pitch coiled carbon nanotube with hydrogen molecules adsorbed on the entire hexagon, pentagon and heptagon site. Here we considered the most stable Y (perpendicular and above carbon ring) configuration for binding energy calculation. The binding energy per adsorbed molecule is calculated in a usual way. From the results we observed the binding energy per molecule for full coverage decreases than the binding energy for a single hydrogen molecule in the same configuration (Y) due to the repulsive interactions between some neighbor molecules. We obtain 6.8 wt% storage capacity, almost meets the DOE target. Fig. 9 shows horizontal view of the full coverage on coiled carbon nanotubes for configuration (Y). The side view of full coverage of H2 adsorption on coiled carbon nanotubes for configuration (Y) is given in Fig. 10. The corresponding binding energy changes as a function of separation for full coverage on CCNT is given in Fig. 11. In order to compare our present result with the linear tube under full coverage, we have considered metallic (5,5) armchair nanotubes for full coverage for the same Y configuration. We obtain the binding energy per molecule value as

0.094 eV, which is greater than CCNTs value. The Full coverage of H2 adsorption on (5,5) metallic tube is shown in Fig. 12. The graph shown in Fig. 13 gives the corresponding binding energy changes as a function of separation for full coverage on (5,5) metallic tube. Table 3 gives the length of tube (nm), diameter (nm), no of carbon atoms, number of H2 molecules, binding energy (eV) and storage capacity (wt%) values in (5,5) and CCNT. A part of this work was presented in WHTC-2009 conference held at New Delhi, India.

5.

Conclusions

In this paper, we have examined the hydrogen binding energy values in CCNTs at various active sites. The different molecular configurations are considered to find out the favorable one. We observed the configuration Y is more suitable for hexagon and pentagon and Z configuration proves to be a stable one for the heptagon. Our results show an increment of 80% in the binding energy compared to linear tube for case of hexagon site and heptagon site. For the pentagon site we could observe an increase of 50%. The physisorption energies per molecule decrease when the surface of CCNT is fully covered with Hydrogen molecule. This is due to the repulsive interactions between some neighbor molecules. It gives 6.8 wt% storage capacity. Adsorption through full coverage increases the

1320

international journal of hydrogen energy 35 (2010) 1313–1320

molecule storage capacity. But it leads to strong repulsive interactions. From these we conclude that the coiled types of CNTs are most suitable material for H2 adsorption. The tubes with more number of pitches will further may enhance the adsorption energy values and the work is under progress.

Acknowledgement This work is funded by the Ministry of New and Renewable energy (MNRE), Government of India, New Delhi.

references

[1] Jouanne AV, Husain I, Wallace A, Yokochi A. Innovative hydrogen/fuel electric vehicle infrastructure based on renewable energy sources. IEEE 2003;0-7803-7883-0/03. [2] Jain IP. Hydrogen the fuel for 21st century. Int J Hydrogen Energy 2009;XXX:I-II. [3] Dillon AC, Gennett T, Alleman JL, Jones KM, Parilla PA, Heben MJ. Carbon nanotube materials for hydrogen storage. In: Proceedings of the 1999 U.S. DOE/NREL Hydrogen program review 1999;NREL/CP-570-26938. [4] Ijima S. Helical microtubules of graphitic carbon. Nature 1991;354:56–8. [5] Mintmire JW, Dunlap BI, White CT. All fullerene tubules metallic? Phys Rev Lett 1992;68:631–4. [6] Hamada N, Sawada S, Oshiyama A. New one-dimensional conductors: graphitic microtubules. Phys Rev Lett 1992;68: 1579–81. [7] Vilaplana AF. Ab initio computational investigation of physisorption of molecular hydrogen on achiral singlewalled carbon nanotubes. J Chem Phys 2005;122:214724–6. [8] Arellano JS, Molina LM, Rubio A, Lopez MJ, Alonso JA. Interaction of molecular and atomic hydrogen with (5,5) and (6,6) single-wall carbon nanotubes. J Chem Phys 2002;117: 2281–8. [9] Li Ju, Furuta T, Goto H, Ohashi T, Fujiwara Y, Yip S. Theoretical evaluation of hydrogen storage capacity in pure carbon nanostructures. J Chem Phys 2003;119:2376–85. [10] Ma Y, Xia Y, Zhao M, Ying M. Hydrogen storage capacity in single-walled carbon nanotubes. Phys Rev B: Condens Matter 2002;65:155430–6. [11] Zuttel A, Sudan P, Mauron PH, Kiyobayashi T, Emmenegger CH, Schlapbach L. Hydrogen storage in carbon nanostructures. Int J Hydrogen Energy 2002;27:203–12.

[12] Poirier E, Chahine R, Bose TK. Hydrogen adsorption in carbon nanostrucutres. Int J Hydrogen Energy 2001;26:831–5. [13] Qikun W, Changchun Z, Weihua L, Ting W. Hydrogen storage by carbon nanotube and their films under ambient pressure. Int J Hydrogen Energy 2002;27:497–500. [14] Mohana Reddy AL, Ramaprabhu S. Hydrogen adsorption properties of single- walled carbon nanotube – nanocrystalline platinum compistes. Int J Hydrogen Energy 2008;33:1028–34. [15] Gayathri V, Geetha R. Hydrogen adsorption in defected carbon nanotubes. Adsorpt 2007;13:53–9. [16] Lau KT, Lu M, Hui D. Coiled carbon naotubes: synthesis and their potential applications in advanced composite structures. Composites: Part B 2006;37:437–48. [17] Dunlap BI. Connecting carbon tubules. Phys Rev B: Condens Matter 1992;46:1933–6. [18] Fonseca A, Hernadi K, Nagy JB, Lambin PH, Lucas AA. Model structure of perfectly graphitizable coiled carbon nanotubes. Carbon 1995;33:1759–75. [19] Amelinckx S, Zhang XB, Bernaerts D, Zhang XF, Ivanov V, Nagy JB. A formation mechanism for catalytically grown helix-shaped graphite nanotubes. Science 1994;265:635–9. [20] Yang S, Hasegawa M, Chen X, Motojima S. Synthesis and morphology of carbon microcoils produced using methane as a carbon source. Carbon 2007;45:1583–95. [21] Lu M, Liu WM, Guo XY, Li XL. Coiled carbon nanotubes growth via reduced-pressure catalytic technical vapor deposition. Carbon 2004;42:805–11. [22] Ryo KA, Tamura R, Tsukada M, Itoh S, Ihara S. Electronic structure of helically coiled cage of graphitic carbon. Phys Rev Lett 1995;74:2307–10. [23] Ryo KA, Tamura R, Tsukada M, Itoh S, Ihara S. Electronic structure of helically coiled carbon nanotubes: relation between phason lines and energy band features. Phys Rev B: Condens Matter 1996;53:2114–20. [24] Ihara S, Itoh S, Kitakami JI. Helically coiled cage forms of graphitic carbon. Phys Rev B: Condens Matter 1993;48: 5643–7. [25] Biro LP, Lazarescu SD, Thiry PA, Fonseca A, Nagy JB, Lucas AA, et al. Scanning tunneling microscopy observation of tightly wound single-wall coiled carbon nantoubes. Europhys Lett 2000;50:494–500. [26] Biro LP, Mark GI, Koos AA, Nagy JB, Lambin PH. Coiled carbon nanotube structures with supraunitary nonhexagonal to hexagonal ring ratio. Phys Rev B: Condens Matter 2002;66: 165401–6. [27] Alonso JA, Arellano JS, Molina LM, Rubio A, Lopez MJ. Interaction of molecular and atomic hydrogen with single-wall carbon nanotubes. IEEE Trans Nanotechnol 2004;3:304–10.