- Email: [email protected]
Pd nanoparticles with tunable diameter deposited on carbon nanotubes with enhanced hydrogen storage capacity
Energy 75 (2014) 549e554
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Pd nanoparticles with tunable diameter deposited on carbon nanotubes with enhanced hydrogen storage capacity Karolina Wenelska, Beata Michalkiewicz, Xuecheng Chen, Ewa Mijowska* Institute of Chemical and Environment Engineering, West Pomeranian University of Technology, ul. Pulaskiego 10, 70-322, Szczecin, Poland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 December 2013 Received in revised form 22 May 2014 Accepted 4 August 2014 Available online 29 August 2014
In situ synthesis of Pd nanoparticles supported on multiwalled CNT (carbon nanotubes) is reported. The size of the nanoparticles can be easily tuned via application of different experimental conditions. The hydrogen storage properties of Pd supported on CNT at room temperature were examined in the pressure range of 0e50 bar. Carbon material with palladium particles of 3 nm diameter exhibits the highest hydrogen capacity at low moderate pressure than the raw materials and CNT with larger particles size of 17 and 9 nm, respectively. We also propose the mechanism of hydrogen storage in the studied samples. © 2014 Elsevier Ltd. All rights reserved.
Keywords:: Hydrogen storage Energy Carbon nanotubes Palladium nanoparticle
1. Introduction The discovery of carbon nanotubes and carbon nanotubes based materials has inspired scientists for a range of potential applications. Nanotubes have attracted extensive attention for their intriguing and potentially useful structural, electrical, and mechanical properties, have unique atomic structure, very high aspect ratio [1e6]. Composites reinforced by carbon nanotubes have improved electrical conductivity and electrostatic charging behaviour, optical emitting devices, and in lightweight, high strength composites [7]. Hydrogen storage is one of the key issues for the realization of fuel-cell powered vehicles using hydrogen as the energy carrier. The advantages of hydrogen as energy sources lie in the fact that its byproduct is water, and it can be easily regenerated. Owing to the lack of a suitable storage system satisfying a combination of both volume and weight limitations, the use of hydrogen energy technology has been restricted from automobile application [8]. Hydrogen storage remains one of the main challenges in the implementation of a hydrogen-based energy economy. Although several different approaches are being pursued, sorption onto a porous high-surface-area material is a one contender. There is a great interest in finding porous solid materials that can store hydrogen for use in fuel cell vehicles [9]. Carbon-based materials,
* Corresponding author. E-mail address: [email protected] (E. Mijowska). http://dx.doi.org/10.1016/j.energy.2014.08.016 0360-5442/© 2014 Elsevier Ltd. All rights reserved.
due to their low cost and weight, have been considered as suitable adsorption substrates for the reversible storage of hydrogen [10e14]. Hydrogen storage in these materials is still at a research level and is not yet mature enough for industrial application [15e20]. Ideally, these materials would adsorb large amounts of hydrogen gas reproducibly at room temperature and moderate pressure. Among these hydrogen storage materials, nanostructured and porous carbon materials, including carbon nanotubes, graphite nanofibers, activated carbon, and graphite, have received considerable research interest due to their lightweight, high surface areas, and relative chemical stabilities [21]. Many studies have revealed that hydrogen storage capacity is enhanced by added metals e.g. (Ag, B, Ca, Fe, K, Li, Ni, Pd, Pt, Ru, Ti, TiO2) [22] to carbon structures. Recently, it was reported that hydrogen storage at room temperature can be improved by a phenomenon known as spillover effect initiated by metal deposited on the adsorbate surface [23]. A mechanism involves the dissociation of H2 on the metal particles, atomic H diffusion from the particles to the sorbent, and chemisorption of atomic hydrogen on the material's winding sites. A very important attribute of this mechanism is that it is operative at ambient temperature as opposed to storage of H2 by physisorption, which requires very low temperature [24]. In our study we present in situ deposition of palladium particles on the surface of carbon nanotubes. This method allows loading of well dispersed Pd nanoparticles with controlled diameter on CNT surface. The carbon nanotubes with Pd nanoparticles are shown to have much higher hydrogen adsorption than the pure carbon nanotubes. It is observed that the amount and the size of Pd
550
K. Wenelska et al. / Energy 75 (2014) 549e554
nanoparticles supported on CNT are the crucial parameters determining the hydrogen storage capacity. Different stirring time resulted in the synthesis of different diameter of palladium nanoparticle and different diameter distribution. We believe that this method will pave the new way for the future storage and transport of H2 and application of H2 energy.
reflux for 24 h in 110 C. In the typical synthesis of carbon nanotubes functionalized by Pd nanoparticles. 50 mg of CNT and 50 mg palladium acetate were dispersed in 150 ml of acetone and sonicated for 2 h. After sonication process, the mixture was stirred for 24, 48, 72 h respectively. Finally, each sample was dried in air at 100 C for 24 h.
1.1. Experimental section
1.1.2. Characterization Transmission electron microscopy (TEM) characterization of the samples were performed on a Tecnai F30 electron microscope operated at an acceleration voltage of 200 kV. Hydrogen adsorption capacity was measured using a Sievert-type volumetric apparatus (IMI Hiden Analytical Ltd). The TGA (thermogravimetric analysis) measurements were carried out in a DTA-Q600 SDT instrument TA
1.1.1. Synthesis of CNT witch palladium particle In a typical experiment, the surface treatment process is as follows. Multiwalled CNT (carbon nanotubes) (50.0 mg, from SHEENZEN NANOPORT) were weighed and placed in a 50 mL glass flask with 10 mL of HNO3 (69%) and 30 mL of H2SO4 (96.2%) and
Fig. 1. TEM images of pristine carbon nanotubes (a, a’, a”) CNT with Pd nanoparticles Pd@CNT_24 h (b, b’, b”), Pd@CNT_48 h (c, c’, c”), Pd@CNT_72 h (d, d’, d”).
K. Wenelska et al. / Energy 75 (2014) 549e554
551
Fig. 2. Diameter distributions of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d).Pd@CNT_72 h.
device in which the samples were heat-treated in flowing air at 900 C. XRD (X-ray powder diffraction) patterns were measured on a Philips diffractometer using Cu Ka radiation. The BrunauereEmmetteTeller surface area (SBET) of the CNT with palladium was measured with Micromeritics ASAP 2010 M instrument. Raman scattering was conducted on a Renishaw micro Raman spectrometer (l ¼ 785 nm). FTIR (Fourier transform infrared spectroscopy) Measurement ware made with a Nicolet model 380 Fourier transform-infrared spectrometer. 2. Results and discussion The final samples dried in air were investigated using transmission electron microscopy. The pristine CNT are very uniform in a diameter of about 20e50 nm. Fig. 1 shows typical TEM images of
the Pd nanoparticles supported on the surface CNT, where palladium acetate (Pd (ac)2) was a source of Pd nanoparticles. The Pd nanoparticles are highly dispersed on the carbon nanotubes surface and they are deposited homogeneously. The detailed TEM analysis of the samples allowed to reveal the diameter distribution of the samples. The histograms presenting them are shown in Fig. 2(aec). With the increase of the mixing time the mean diameter and diameter distribution of palladium nanoparticle increased. It is noted that most of Pd nanoparticles are centred at 3 nm for Pd@CNT_24 h (Fig. 2a), 7 nm for Pd@CNT_48 h (Fig. 2b) and 9 nm for Pd@CNT_72 h (Fig. 2c). To confirm the crystal phase composition of the samples, XRD (X-ray diffraction) analysis was used (Fig. 3A). The XRD patterns of the obtained products showed the characteristic diffraction peaks of the palladium cubic phase, indicating the formation of metallic
Fig. 3. A Powder X-ray diffraction patterns of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h, B TGA profiles of the (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h.
552
K. Wenelska et al. / Energy 75 (2014) 549e554
Pd. Four reflections in the XRD pattern at 40 , 46 ,68 , are assigned to centred cubic crystal structure of palladium [25]. The appearance of the broad strong diffraction peaks at 25 and 43 are assigned to graphitic carbon from carbon nanotubes composition. This broadening of the diffraction peak suggests the lack of long-range order and it is a signature of graphene-based carbon [26]. TGA measurements (Fig. 3B) provide information about the carbon content as well as the quantity of palladium particles. Wellordered graphite starts to oxidize at above 600 C, whereas the carbon nanotubes are oxidized at above 500 C. The carbon nanotubes begin to decompose at 517 C in air. As the temperature is further increased, the weight loss increases rapidly until all of the carbon nanotubes are exhausted at about 650 C. The ash content of the carbon nanotubes after combustion at 900 C is 4.6 wt%, thus implying that about 4.6 wt% this is the residue of the oxidized catalysts. TGA results of the Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h, have the ash content of 6.3 wt%, 8.4 wt%, 38 wt% respectively. In comparison to pure carbon nanotubes, the stability of the carbon nanotubes with palladium particles is decreased. The weight loss of Pd@CNT_24 h and Pd@CNT_48 h started at 177 C and for Pd@CNT_72 h at 437 C. This may be due to the interaction of Pd and carbon atoms inducing defects in the crystal structure of CNT. The bonding and crystallinity of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h, specimens were studied by Raman spectroscopy (Fig. 4A). Generally, the Raman response of all the samples is nearly the same. Raman spectra show G band at ~1614 cm1 arising from the stretching of the CeC bond in graphitic materials. It indicates the in-plane vibration of sp2 carbon atoms. D band is detected at ~1316 cm1 which corresponds to the defect band in graphite, which arises from the disorder in the sp2 hybridization. The intensity of D-band exceeds the G-band’s, suggesting that the structure comprise a lot of defects. It is well-known that the position of the G-band is very sensitive to the strain present in the specimen in all sp2 systems. The ratio of the intensity of the G-to D-band expressed as IG/ID is taken as a measure of the degree of defects presented in the sample [27,28]. Upon deposition of the Pd nanoparticles, relation of G-to-D band intensities decreased what confirmed that additional defects in CNT structure are formed upon deposition of the metal particles (see inset of Fig. 4A). For most of the samples the G/D ratio decreases with
Table 1 Specific surface area and total pore volume for CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h.
CNT Pd@CNT_24 h Pd@CNT_48 h Pd@CNT_72 h
SBET [m2/g]
Total pore volume [cm3/g]
173.95 89.06 93.89 102.06
1.248 1.088 0.615 0.991
increasing amount of the palladium particles. To support Raman data, IR (infrared) analysis of the samples was performed (Fig. 4B). In the spectra of palladium containing samples the additional peak near 1590 cme1 attributed to PdeC bonding is detected [29]. This mode is the most prenounced in Pd@CNT_72 h. This indicates that the highest concentration of PdeC bonds is in this material. The samples were also measured by N2 adsorption/desorption method. The adsorption isotherms of CNT and CNT decorated by palladium nanoparticles showed Table 1. Surface areas are commonly reported as BET (BrunauereEmmetteTeller) surface areas obtained by applying the theory of Brunauer, Emmett, and Teller to nitrogen adsorption isotherms measured at 77 K. This is a standard procedure that allows for comparison of different materials. The total specific surface area of CNT was 174 m2/g. However, it is significantly dropped to 102 m2/g for Pd@CNT_72 h, 93 m2/g for Pd@CNT_48 h and 89 m2/g for Pd@CNT_24 h when Pd nanoparticles are deposited on CNT. The smaller pore volume and surface area were observed in the samples with the Pd particles. This suggests that the Pd nanoparticles block the pores of CNT. The total pore volume was calculated and it was 1248 cm3/g, 1088 cm3/g, 0.615 cm3/g, 0.991 cm3/g for CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h, respectively. 3. High-pressure hydrogen adsorption studies Hydrogen adsorption studies were carried out for the CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h in the pressure range of 0e50 bar and in the temperature of 25 C. The instrument has been calibrated with high-purity hydrogen (99.99%) at various initial pressures. Fig. 5 shows the pressurecomposition isotherms of the samples acquired in the pressure range of 0e50 bar H2
Fig. 4. A Raman spectra of the (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h B. FTIR spectra of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d).Pd@CNT_72 h.
K. Wenelska et al. / Energy 75 (2014) 549e554
Hydrogen adsorption [wt.%]
1,2
CNT Pd@CNT_24h Pd@CNT_48h Pd@CNT_72h
0,9
0,6
0,3
0,0
0
10
20
30
40
50
Pressure [bar] Fig. 5. Hydrogen adsorption capacity at 25 C CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h.
553
for by considering the contribution of palladium particles on the hydrogen storage. However, all the proposed samples of Pd@CNT showed enhanced gravimetric capacity with respect to the CNT. We have shown that a Pd atom prefers to adsorb on the CeC bridge site as compared to any other available site on the carbon nanotubes. Results show that the CeC bonds near the Pd adsorption site are weakened after Pd decoration process due to the formation of PdeC bonds. On adsorption of hydrogen new PdeH bonds were found to form at the expense of weakening of PdeC and CeC bonds. Most of bonds PdeC shows a sample CNT@Pd_24 h and hydrogen capacity is the biggest [24]. Additionally, it is well-known that useful materials for hydrogen storage should possess excellent regenerability and stability with many adsorptiondesorption cycles [31]. Therefore, to test the stability of CNT and Pd@CNT 10 cycles of adsorptiondesorption were carried out. The first and last adsorption curves of each sample are shown in Fig. 6. All the samples were stable even after 10 adsorptiondesorption cycles. 4. Conclusions
respectively. The maximum capacity of 0.87 wt % has been achieved at 25 C and 50 bar of hydrogen. In general, all the isotherms exhibit the typical feature that the hydrogen adsorption amount increases monotonically with rising pressure. However, measured values show that the carbon nanotubes and Pd@CNT uptake capacity increases from 0.52% for CNT to 0.87% for Pd@CNT_24 h, to 0.82% for Pd@CNT_48 h and to 0.71% for Pd@CNT_72 h at 50 bars. Palladiumloaded carbon materials have higher hydrogen capacity than the raw materials in the pressure above 15 bars. However, at 50 bar, the difference in the hydrogen capacity between the loaded and the raw materials is the most prenounced. We observed a clear Pd nanoparticles size dependence on their hydrogen adsorption. To the best of our knowledge, these are the first quantitative data concerning size Pd nanoparticles deposited on CNT. These results imply that hydrogen-storage properties of CNT can be controlled by changing the Pd size [30]. These results imply that hydrogenstorage properties of metals can be controlled by changing their size. The hydrogen-storage capacity of Pd decreases with increase of Pd nanoparticles. Therefore, carbon nanotubes with deposited Pd nanoparticles are more effective in the adsorption of hydrogen than the pristine carbon nanotubes. The enhanced hydrogen storage capacity observed for Pd@CNT_24 h can be qualitatively accounted
In summary, in this contribution palladium nanoparticles with different size and diameter distribution were deposited on the surface of carbon nanotubes were generated using a simple in situ technique. We tested the influence of the obtained samples on their hydrogen storage capacity. It was found that hydrogen storage properties are strongly enhanced in respect to the pristine sample. In addition, hydrogen storage capacity was the most enhanced for carbon nanotubes with palladium particle centred at diameter of 3 nm and it was decreasing with the increase of the particles size. The experiments revealed that, the system composed of CNT and Pd has a potential as hydrogen storage medium due to the enhanced H2 adsorption capacity and stability after multiple adsorptiondesorption cycles. This knowledge can be important in the development of not only hydrogen-storage media, but also of novel catalysts in hydrogen handling. Acknowledgements This research was funded by National Science Centre SONATA BIS e UMO-2012/07/E/ST8/01702. References
Hydrogen adsorption [wt.%]
1,2
CNT Pd@CNT_24h Pd@CNT_48h Pd@CNT_72h 0,9 After 10 cycles ads/des CNT Pd@CNT_24h Pd@CNT_48h 0,6 Pd@CNT_72h
0,3
0,0 0
10
20
30
40
50
Pressure [bar] Fig. 6. Hydrogen adsorption before and after 10 adsorptionedesorption cycles at 25 C for CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h.
[1] Iijima S. Helical microtubules of graphitic carbon. Nature 1991;354:56e8. [2] Rinzler AG, Hafner JH, Nikolaev P, Lou L, Kim SG, Tomanek D, et al. Unraveling nanotubes: field emission from an atomic wire. Science 1995;269:1550e3. [3] de Heer WA, Chatelain A, Ugarte D. A carbon nanotube field-emission electron source. Science 1995;270:1179e80. [4] Collins PG, Zettl A, Bando H, Thess A, Smalley RE. Nanotube nanodevice. Science 1997;278:100e2. [5] Wagner HD, Lourie O, Feldman Y, Tenne R. Stress-induced fragmentation of multiwall carbon nanotubes in a polymer matrix. Appl Phys Lett 1998;72(2): 188e90. [6] Dagani R. Putting the 'nano' into composites. Chem Eng News 1999;77(23): 25e37. [7] Gong X, Liu J, Baskaran S, Voise RD, Young JS. Surfactant-assisted processing of carbon nanotube/polymer composites. Chem Mater 2000;12(4):1049e52. [8] Odom TW, Huang J-L, Kim P, Ouyang M, Lieber CM. Structure and electronic properties of carbon nanotubes. J Phys Chem B 2000;104:2794e809. [9] Satyapal S, Petrovic J, Thomas G. Gassing up with hydrogen. Sci Am 2007: 81e7. [10] Zhong ZY, Xiong ZT, Sun LF, Luo JZ, Chen P, Wu X, et al. Nanosized nickel (or cobalt)/graphite composites for hydrogen storage. J Phys Chem B 2002;106: 9507e13. [11] Han SS, Goddard WA. Lithium-doped metal-organic frameworks for reversible H2 storage at ambient temperature. J Am Chem Soc 2007;129:8422e3. [12] Belof JL, Stern AC, Eddaoudi M, Space B. On the mechanism of hydrogen storage in a metal-organic framework material. J Am Chem Soc 2007;129: 15202e10.
554
K. Wenelska et al. / Energy 75 (2014) 549e554
[13] El-Kaderi HM, Hunt JR, Mendoza-Cortes JL, Cote AP, Taylor RE, O'Keeffe M, et al. Designed synthesis of 3D covalent organic frameworks. Science 2007;316:268e72. [14] Furukawa H, Miller MA, Yaghi OM. Independent verification of the saturation hydrogen uptake in MOF-177 and establishment of a benchmark for hydrogen adsorption in metal-organic framework. J Mater Chem 2007;17: 3197e204. [15] Reyhani A, Mortazavi SZ, Mirershadi S, Moshfegh AZ, Parvin P, Nozad Golikand A. Hydrogen storage in decorated multiwalled carbon nanotubes by Ca, Co, Fe, Ni, and Pd nanoparticles under ambient conditions. J Phys Chem C 2011;115(14):6994e7001. [16] Li B, Wei L, Yang H, Wang X, Chen H. The enhancing mechanism of calcium oxide on water gas shift reaction for hydrogen production. Energy 2014;68: 248e54. [17] Chaia YJ, Donga YM, Menga HX, Jiaa YY, Shena J, Huangb YM, et al. Hydrogen generation by aluminum corrosion in cobalt (II) chloride and nickel (II) chloride aqueous solution. Energy 2014;68:204e9. [18] Singh V, Soni SL, Sharma D. Engine performance of optimized hydrogenfueled direct injection engine. Energy 2014;65:pp.116e122. [19] de Santolia L, Bassoa GL, Bruschib D. Energy characterization of CHP (combined heat and power) fuelled with hydrogen enriched natural gas blends. Energy 2013;60:13e22. [20] Guo WQ, Ding J, Cao GL, Chen C, Zhou XJ, Ren NQ. Accelerated startup of hydrogen production expanded granular sludge bed with l-cysteine supplementation. Energy 2013;60:94e8. [21] Ganz E, Dornfeld MJ. Storage capacity of metal-organic and covalentorganic frameworks by hydrogen spillover. Phys Chem C 2012;116: 3661e6.
[22] Singh AK, Ribas MA, Yakobson BI. H-spillover through the catalyst saturation: an Ab initio thermodynamics study. J Am Chem Soc 2009;3:1657e62. [23] Psofogiannakis GM, Froudakis GE. DFT study of the hydrogen spillover mechanism on Pt-doped graphite. J Phys Chem C 2011;115:4047e53. [24] Parambhath VB, Nagar R, Ramaprabhu S. Effect of nitrogen doping on hydrogen storage capacity of palladium decorated graphene. Langmuir 2012;28:7826e33. [25] Lengke MF, Fleet ME, Southam G. Synthesis of palladium nanoparticles by reaction of filamentous cyanobacterial biomass with a palladium (II) chloride complex. Langmuir 2007;23:8982e7. [26] Wenelska K, Michalkiewicz B, Gong J, Tang T, Kalenczuk R, Chen X, et al. In situ deposition of Pd nanoparticles with controllable diameters in hollow carbon spheres for hydrogen storage. Int J Hydrogen Energy 2013;38:16179e84. [27] Dresselhaus MS, Dresselhaus G, Jorio A. Raman spectroscopy of carbon nanotubes in 1997 and 2007. J Phys Chem C 2007;111:17887e93. [28] Fujimori T, Urita K, Aoki Y, Kanoh H, Ohba T, Yudasaka M, et al. Fine nanostructure analysis of single-wall carbon nanohorns by surface-enhanced. Raman scattering. J Phys Chem C 2008;112(20):pp.7552e7556. [29] C. C. Chusuei, M. Wayu, Electronic properties of carbon nanotubes. ISBN 978953-307-499-3. [30] Yamauchi M, Ikeda R, Kitagawa H, Takata M. Nanosize effects on hydrogen storage in palladium. J Phys Chem C 2008;112:pp.3294e3299. [31] Bose PK, Banerjee MR, Majumder A. Multi objective optimization of performance parameters of a single cylinder diesel engine running with hydrogen using a Taguchi-fuzzy based approach. Energy 2013;63:375e86.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Pd nanoparticles with tunable diameter deposited on carbon nanotubes with enhanced hydrogen storage capacity Karolina Wenelska, Beata Michalkiewicz, Xuecheng Chen, Ewa Mijowska* Institute of Chemical and Environment Engineering, West Pomeranian University of Technology, ul. Pulaskiego 10, 70-322, Szczecin, Poland
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 December 2013 Received in revised form 22 May 2014 Accepted 4 August 2014 Available online 29 August 2014
In situ synthesis of Pd nanoparticles supported on multiwalled CNT (carbon nanotubes) is reported. The size of the nanoparticles can be easily tuned via application of different experimental conditions. The hydrogen storage properties of Pd supported on CNT at room temperature were examined in the pressure range of 0e50 bar. Carbon material with palladium particles of 3 nm diameter exhibits the highest hydrogen capacity at low moderate pressure than the raw materials and CNT with larger particles size of 17 and 9 nm, respectively. We also propose the mechanism of hydrogen storage in the studied samples. © 2014 Elsevier Ltd. All rights reserved.
Keywords:: Hydrogen storage Energy Carbon nanotubes Palladium nanoparticle
1. Introduction The discovery of carbon nanotubes and carbon nanotubes based materials has inspired scientists for a range of potential applications. Nanotubes have attracted extensive attention for their intriguing and potentially useful structural, electrical, and mechanical properties, have unique atomic structure, very high aspect ratio [1e6]. Composites reinforced by carbon nanotubes have improved electrical conductivity and electrostatic charging behaviour, optical emitting devices, and in lightweight, high strength composites [7]. Hydrogen storage is one of the key issues for the realization of fuel-cell powered vehicles using hydrogen as the energy carrier. The advantages of hydrogen as energy sources lie in the fact that its byproduct is water, and it can be easily regenerated. Owing to the lack of a suitable storage system satisfying a combination of both volume and weight limitations, the use of hydrogen energy technology has been restricted from automobile application [8]. Hydrogen storage remains one of the main challenges in the implementation of a hydrogen-based energy economy. Although several different approaches are being pursued, sorption onto a porous high-surface-area material is a one contender. There is a great interest in finding porous solid materials that can store hydrogen for use in fuel cell vehicles [9]. Carbon-based materials,
* Corresponding author. E-mail address: [email protected] (E. Mijowska). http://dx.doi.org/10.1016/j.energy.2014.08.016 0360-5442/© 2014 Elsevier Ltd. All rights reserved.
due to their low cost and weight, have been considered as suitable adsorption substrates for the reversible storage of hydrogen [10e14]. Hydrogen storage in these materials is still at a research level and is not yet mature enough for industrial application [15e20]. Ideally, these materials would adsorb large amounts of hydrogen gas reproducibly at room temperature and moderate pressure. Among these hydrogen storage materials, nanostructured and porous carbon materials, including carbon nanotubes, graphite nanofibers, activated carbon, and graphite, have received considerable research interest due to their lightweight, high surface areas, and relative chemical stabilities [21]. Many studies have revealed that hydrogen storage capacity is enhanced by added metals e.g. (Ag, B, Ca, Fe, K, Li, Ni, Pd, Pt, Ru, Ti, TiO2) [22] to carbon structures. Recently, it was reported that hydrogen storage at room temperature can be improved by a phenomenon known as spillover effect initiated by metal deposited on the adsorbate surface [23]. A mechanism involves the dissociation of H2 on the metal particles, atomic H diffusion from the particles to the sorbent, and chemisorption of atomic hydrogen on the material's winding sites. A very important attribute of this mechanism is that it is operative at ambient temperature as opposed to storage of H2 by physisorption, which requires very low temperature [24]. In our study we present in situ deposition of palladium particles on the surface of carbon nanotubes. This method allows loading of well dispersed Pd nanoparticles with controlled diameter on CNT surface. The carbon nanotubes with Pd nanoparticles are shown to have much higher hydrogen adsorption than the pure carbon nanotubes. It is observed that the amount and the size of Pd
550
K. Wenelska et al. / Energy 75 (2014) 549e554
nanoparticles supported on CNT are the crucial parameters determining the hydrogen storage capacity. Different stirring time resulted in the synthesis of different diameter of palladium nanoparticle and different diameter distribution. We believe that this method will pave the new way for the future storage and transport of H2 and application of H2 energy.
reflux for 24 h in 110 C. In the typical synthesis of carbon nanotubes functionalized by Pd nanoparticles. 50 mg of CNT and 50 mg palladium acetate were dispersed in 150 ml of acetone and sonicated for 2 h. After sonication process, the mixture was stirred for 24, 48, 72 h respectively. Finally, each sample was dried in air at 100 C for 24 h.
1.1. Experimental section
1.1.2. Characterization Transmission electron microscopy (TEM) characterization of the samples were performed on a Tecnai F30 electron microscope operated at an acceleration voltage of 200 kV. Hydrogen adsorption capacity was measured using a Sievert-type volumetric apparatus (IMI Hiden Analytical Ltd). The TGA (thermogravimetric analysis) measurements were carried out in a DTA-Q600 SDT instrument TA
1.1.1. Synthesis of CNT witch palladium particle In a typical experiment, the surface treatment process is as follows. Multiwalled CNT (carbon nanotubes) (50.0 mg, from SHEENZEN NANOPORT) were weighed and placed in a 50 mL glass flask with 10 mL of HNO3 (69%) and 30 mL of H2SO4 (96.2%) and
Fig. 1. TEM images of pristine carbon nanotubes (a, a’, a”) CNT with Pd nanoparticles Pd@CNT_24 h (b, b’, b”), Pd@CNT_48 h (c, c’, c”), Pd@CNT_72 h (d, d’, d”).
K. Wenelska et al. / Energy 75 (2014) 549e554
551
Fig. 2. Diameter distributions of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d).Pd@CNT_72 h.
device in which the samples were heat-treated in flowing air at 900 C. XRD (X-ray powder diffraction) patterns were measured on a Philips diffractometer using Cu Ka radiation. The BrunauereEmmetteTeller surface area (SBET) of the CNT with palladium was measured with Micromeritics ASAP 2010 M instrument. Raman scattering was conducted on a Renishaw micro Raman spectrometer (l ¼ 785 nm). FTIR (Fourier transform infrared spectroscopy) Measurement ware made with a Nicolet model 380 Fourier transform-infrared spectrometer. 2. Results and discussion The final samples dried in air were investigated using transmission electron microscopy. The pristine CNT are very uniform in a diameter of about 20e50 nm. Fig. 1 shows typical TEM images of
the Pd nanoparticles supported on the surface CNT, where palladium acetate (Pd (ac)2) was a source of Pd nanoparticles. The Pd nanoparticles are highly dispersed on the carbon nanotubes surface and they are deposited homogeneously. The detailed TEM analysis of the samples allowed to reveal the diameter distribution of the samples. The histograms presenting them are shown in Fig. 2(aec). With the increase of the mixing time the mean diameter and diameter distribution of palladium nanoparticle increased. It is noted that most of Pd nanoparticles are centred at 3 nm for Pd@CNT_24 h (Fig. 2a), 7 nm for Pd@CNT_48 h (Fig. 2b) and 9 nm for Pd@CNT_72 h (Fig. 2c). To confirm the crystal phase composition of the samples, XRD (X-ray diffraction) analysis was used (Fig. 3A). The XRD patterns of the obtained products showed the characteristic diffraction peaks of the palladium cubic phase, indicating the formation of metallic
Fig. 3. A Powder X-ray diffraction patterns of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h, B TGA profiles of the (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h.
552
K. Wenelska et al. / Energy 75 (2014) 549e554
Pd. Four reflections in the XRD pattern at 40 , 46 ,68 , are assigned to centred cubic crystal structure of palladium [25]. The appearance of the broad strong diffraction peaks at 25 and 43 are assigned to graphitic carbon from carbon nanotubes composition. This broadening of the diffraction peak suggests the lack of long-range order and it is a signature of graphene-based carbon [26]. TGA measurements (Fig. 3B) provide information about the carbon content as well as the quantity of palladium particles. Wellordered graphite starts to oxidize at above 600 C, whereas the carbon nanotubes are oxidized at above 500 C. The carbon nanotubes begin to decompose at 517 C in air. As the temperature is further increased, the weight loss increases rapidly until all of the carbon nanotubes are exhausted at about 650 C. The ash content of the carbon nanotubes after combustion at 900 C is 4.6 wt%, thus implying that about 4.6 wt% this is the residue of the oxidized catalysts. TGA results of the Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h, have the ash content of 6.3 wt%, 8.4 wt%, 38 wt% respectively. In comparison to pure carbon nanotubes, the stability of the carbon nanotubes with palladium particles is decreased. The weight loss of Pd@CNT_24 h and Pd@CNT_48 h started at 177 C and for Pd@CNT_72 h at 437 C. This may be due to the interaction of Pd and carbon atoms inducing defects in the crystal structure of CNT. The bonding and crystallinity of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h, specimens were studied by Raman spectroscopy (Fig. 4A). Generally, the Raman response of all the samples is nearly the same. Raman spectra show G band at ~1614 cm1 arising from the stretching of the CeC bond in graphitic materials. It indicates the in-plane vibration of sp2 carbon atoms. D band is detected at ~1316 cm1 which corresponds to the defect band in graphite, which arises from the disorder in the sp2 hybridization. The intensity of D-band exceeds the G-band’s, suggesting that the structure comprise a lot of defects. It is well-known that the position of the G-band is very sensitive to the strain present in the specimen in all sp2 systems. The ratio of the intensity of the G-to D-band expressed as IG/ID is taken as a measure of the degree of defects presented in the sample [27,28]. Upon deposition of the Pd nanoparticles, relation of G-to-D band intensities decreased what confirmed that additional defects in CNT structure are formed upon deposition of the metal particles (see inset of Fig. 4A). For most of the samples the G/D ratio decreases with
Table 1 Specific surface area and total pore volume for CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h.
CNT Pd@CNT_24 h Pd@CNT_48 h Pd@CNT_72 h
SBET [m2/g]
Total pore volume [cm3/g]
173.95 89.06 93.89 102.06
1.248 1.088 0.615 0.991
increasing amount of the palladium particles. To support Raman data, IR (infrared) analysis of the samples was performed (Fig. 4B). In the spectra of palladium containing samples the additional peak near 1590 cme1 attributed to PdeC bonding is detected [29]. This mode is the most prenounced in Pd@CNT_72 h. This indicates that the highest concentration of PdeC bonds is in this material. The samples were also measured by N2 adsorption/desorption method. The adsorption isotherms of CNT and CNT decorated by palladium nanoparticles showed Table 1. Surface areas are commonly reported as BET (BrunauereEmmetteTeller) surface areas obtained by applying the theory of Brunauer, Emmett, and Teller to nitrogen adsorption isotherms measured at 77 K. This is a standard procedure that allows for comparison of different materials. The total specific surface area of CNT was 174 m2/g. However, it is significantly dropped to 102 m2/g for Pd@CNT_72 h, 93 m2/g for Pd@CNT_48 h and 89 m2/g for Pd@CNT_24 h when Pd nanoparticles are deposited on CNT. The smaller pore volume and surface area were observed in the samples with the Pd particles. This suggests that the Pd nanoparticles block the pores of CNT. The total pore volume was calculated and it was 1248 cm3/g, 1088 cm3/g, 0.615 cm3/g, 0.991 cm3/g for CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h, respectively. 3. High-pressure hydrogen adsorption studies Hydrogen adsorption studies were carried out for the CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h in the pressure range of 0e50 bar and in the temperature of 25 C. The instrument has been calibrated with high-purity hydrogen (99.99%) at various initial pressures. Fig. 5 shows the pressurecomposition isotherms of the samples acquired in the pressure range of 0e50 bar H2
Fig. 4. A Raman spectra of the (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d) Pd@CNT_72 h B. FTIR spectra of (a) CNT, (b) Pd@CNT_24 h, (c) Pd@CNT_48 h, (d).Pd@CNT_72 h.
K. Wenelska et al. / Energy 75 (2014) 549e554
Hydrogen adsorption [wt.%]
1,2
CNT Pd@CNT_24h Pd@CNT_48h Pd@CNT_72h
0,9
0,6
0,3
0,0
0
10
20
30
40
50
Pressure [bar] Fig. 5. Hydrogen adsorption capacity at 25 C CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h.
553
for by considering the contribution of palladium particles on the hydrogen storage. However, all the proposed samples of Pd@CNT showed enhanced gravimetric capacity with respect to the CNT. We have shown that a Pd atom prefers to adsorb on the CeC bridge site as compared to any other available site on the carbon nanotubes. Results show that the CeC bonds near the Pd adsorption site are weakened after Pd decoration process due to the formation of PdeC bonds. On adsorption of hydrogen new PdeH bonds were found to form at the expense of weakening of PdeC and CeC bonds. Most of bonds PdeC shows a sample CNT@Pd_24 h and hydrogen capacity is the biggest [24]. Additionally, it is well-known that useful materials for hydrogen storage should possess excellent regenerability and stability with many adsorptiondesorption cycles [31]. Therefore, to test the stability of CNT and Pd@CNT 10 cycles of adsorptiondesorption were carried out. The first and last adsorption curves of each sample are shown in Fig. 6. All the samples were stable even after 10 adsorptiondesorption cycles. 4. Conclusions
respectively. The maximum capacity of 0.87 wt % has been achieved at 25 C and 50 bar of hydrogen. In general, all the isotherms exhibit the typical feature that the hydrogen adsorption amount increases monotonically with rising pressure. However, measured values show that the carbon nanotubes and Pd@CNT uptake capacity increases from 0.52% for CNT to 0.87% for Pd@CNT_24 h, to 0.82% for Pd@CNT_48 h and to 0.71% for Pd@CNT_72 h at 50 bars. Palladiumloaded carbon materials have higher hydrogen capacity than the raw materials in the pressure above 15 bars. However, at 50 bar, the difference in the hydrogen capacity between the loaded and the raw materials is the most prenounced. We observed a clear Pd nanoparticles size dependence on their hydrogen adsorption. To the best of our knowledge, these are the first quantitative data concerning size Pd nanoparticles deposited on CNT. These results imply that hydrogen-storage properties of CNT can be controlled by changing the Pd size [30]. These results imply that hydrogenstorage properties of metals can be controlled by changing their size. The hydrogen-storage capacity of Pd decreases with increase of Pd nanoparticles. Therefore, carbon nanotubes with deposited Pd nanoparticles are more effective in the adsorption of hydrogen than the pristine carbon nanotubes. The enhanced hydrogen storage capacity observed for Pd@CNT_24 h can be qualitatively accounted
In summary, in this contribution palladium nanoparticles with different size and diameter distribution were deposited on the surface of carbon nanotubes were generated using a simple in situ technique. We tested the influence of the obtained samples on their hydrogen storage capacity. It was found that hydrogen storage properties are strongly enhanced in respect to the pristine sample. In addition, hydrogen storage capacity was the most enhanced for carbon nanotubes with palladium particle centred at diameter of 3 nm and it was decreasing with the increase of the particles size. The experiments revealed that, the system composed of CNT and Pd has a potential as hydrogen storage medium due to the enhanced H2 adsorption capacity and stability after multiple adsorptiondesorption cycles. This knowledge can be important in the development of not only hydrogen-storage media, but also of novel catalysts in hydrogen handling. Acknowledgements This research was funded by National Science Centre SONATA BIS e UMO-2012/07/E/ST8/01702. References
Hydrogen adsorption [wt.%]
1,2
CNT Pd@CNT_24h Pd@CNT_48h Pd@CNT_72h 0,9 After 10 cycles ads/des CNT Pd@CNT_24h Pd@CNT_48h 0,6 Pd@CNT_72h
0,3
0,0 0
10
20
30
40
50
Pressure [bar] Fig. 6. Hydrogen adsorption before and after 10 adsorptionedesorption cycles at 25 C for CNT, Pd@CNT_24 h, Pd@CNT_48 h, Pd@CNT_72 h.
[1] Iijima S. Helical microtubules of graphitic carbon. Nature 1991;354:56e8. [2] Rinzler AG, Hafner JH, Nikolaev P, Lou L, Kim SG, Tomanek D, et al. Unraveling nanotubes: field emission from an atomic wire. Science 1995;269:1550e3. [3] de Heer WA, Chatelain A, Ugarte D. A carbon nanotube field-emission electron source. Science 1995;270:1179e80. [4] Collins PG, Zettl A, Bando H, Thess A, Smalley RE. Nanotube nanodevice. Science 1997;278:100e2. [5] Wagner HD, Lourie O, Feldman Y, Tenne R. Stress-induced fragmentation of multiwall carbon nanotubes in a polymer matrix. Appl Phys Lett 1998;72(2): 188e90. [6] Dagani R. Putting the 'nano' into composites. Chem Eng News 1999;77(23): 25e37. [7] Gong X, Liu J, Baskaran S, Voise RD, Young JS. Surfactant-assisted processing of carbon nanotube/polymer composites. Chem Mater 2000;12(4):1049e52. [8] Odom TW, Huang J-L, Kim P, Ouyang M, Lieber CM. Structure and electronic properties of carbon nanotubes. J Phys Chem B 2000;104:2794e809. [9] Satyapal S, Petrovic J, Thomas G. Gassing up with hydrogen. Sci Am 2007: 81e7. [10] Zhong ZY, Xiong ZT, Sun LF, Luo JZ, Chen P, Wu X, et al. Nanosized nickel (or cobalt)/graphite composites for hydrogen storage. J Phys Chem B 2002;106: 9507e13. [11] Han SS, Goddard WA. Lithium-doped metal-organic frameworks for reversible H2 storage at ambient temperature. J Am Chem Soc 2007;129:8422e3. [12] Belof JL, Stern AC, Eddaoudi M, Space B. On the mechanism of hydrogen storage in a metal-organic framework material. J Am Chem Soc 2007;129: 15202e10.
554
K. Wenelska et al. / Energy 75 (2014) 549e554
[13] El-Kaderi HM, Hunt JR, Mendoza-Cortes JL, Cote AP, Taylor RE, O'Keeffe M, et al. Designed synthesis of 3D covalent organic frameworks. Science 2007;316:268e72. [14] Furukawa H, Miller MA, Yaghi OM. Independent verification of the saturation hydrogen uptake in MOF-177 and establishment of a benchmark for hydrogen adsorption in metal-organic framework. J Mater Chem 2007;17: 3197e204. [15] Reyhani A, Mortazavi SZ, Mirershadi S, Moshfegh AZ, Parvin P, Nozad Golikand A. Hydrogen storage in decorated multiwalled carbon nanotubes by Ca, Co, Fe, Ni, and Pd nanoparticles under ambient conditions. J Phys Chem C 2011;115(14):6994e7001. [16] Li B, Wei L, Yang H, Wang X, Chen H. The enhancing mechanism of calcium oxide on water gas shift reaction for hydrogen production. Energy 2014;68: 248e54. [17] Chaia YJ, Donga YM, Menga HX, Jiaa YY, Shena J, Huangb YM, et al. Hydrogen generation by aluminum corrosion in cobalt (II) chloride and nickel (II) chloride aqueous solution. Energy 2014;68:204e9. [18] Singh V, Soni SL, Sharma D. Engine performance of optimized hydrogenfueled direct injection engine. Energy 2014;65:pp.116e122. [19] de Santolia L, Bassoa GL, Bruschib D. Energy characterization of CHP (combined heat and power) fuelled with hydrogen enriched natural gas blends. Energy 2013;60:13e22. [20] Guo WQ, Ding J, Cao GL, Chen C, Zhou XJ, Ren NQ. Accelerated startup of hydrogen production expanded granular sludge bed with l-cysteine supplementation. Energy 2013;60:94e8. [21] Ganz E, Dornfeld MJ. Storage capacity of metal-organic and covalentorganic frameworks by hydrogen spillover. Phys Chem C 2012;116: 3661e6.
[22] Singh AK, Ribas MA, Yakobson BI. H-spillover through the catalyst saturation: an Ab initio thermodynamics study. J Am Chem Soc 2009;3:1657e62. [23] Psofogiannakis GM, Froudakis GE. DFT study of the hydrogen spillover mechanism on Pt-doped graphite. J Phys Chem C 2011;115:4047e53. [24] Parambhath VB, Nagar R, Ramaprabhu S. Effect of nitrogen doping on hydrogen storage capacity of palladium decorated graphene. Langmuir 2012;28:7826e33. [25] Lengke MF, Fleet ME, Southam G. Synthesis of palladium nanoparticles by reaction of filamentous cyanobacterial biomass with a palladium (II) chloride complex. Langmuir 2007;23:8982e7. [26] Wenelska K, Michalkiewicz B, Gong J, Tang T, Kalenczuk R, Chen X, et al. In situ deposition of Pd nanoparticles with controllable diameters in hollow carbon spheres for hydrogen storage. Int J Hydrogen Energy 2013;38:16179e84. [27] Dresselhaus MS, Dresselhaus G, Jorio A. Raman spectroscopy of carbon nanotubes in 1997 and 2007. J Phys Chem C 2007;111:17887e93. [28] Fujimori T, Urita K, Aoki Y, Kanoh H, Ohba T, Yudasaka M, et al. Fine nanostructure analysis of single-wall carbon nanohorns by surface-enhanced. Raman scattering. J Phys Chem C 2008;112(20):pp.7552e7556. [29] C. C. Chusuei, M. Wayu, Electronic properties of carbon nanotubes. ISBN 978953-307-499-3. [30] Yamauchi M, Ikeda R, Kitagawa H, Takata M. Nanosize effects on hydrogen storage in palladium. J Phys Chem C 2008;112:pp.3294e3299. [31] Bose PK, Banerjee MR, Majumder A. Multi objective optimization of performance parameters of a single cylinder diesel engine running with hydrogen using a Taguchi-fuzzy based approach. Energy 2013;63:375e86.