NANO LETTERS
Enhanced Cyclic Performance and Lithium Storage Capacity of SnO2/ Graphene Nanoporous Electrodes with Three-Dimensionally Delaminated Flexible Structure
2009 Vol. 9, No. 1 72-75
Seung-Min Paek, EunJoo Yoo, and Itaru Honma* Energy Technology Research Institute, National Institute of AdVanced Industrial Science and Technology (AIST), Umezono 1-1-1, Central 2, Tsukuba, Ibaraki 305-8568, Japan Received August 14, 2008; Revised Manuscript Received November 6, 2008
ABSTRACT To fabricate nanoporous electrode materials with delaminated structure, the graphene nanosheets (GNS) in the ethylene glycol solution were reassembled in the presence of rutile SnO2 nanoparticles. According to the TEM analysis, the graphene nanosheets are homogeneously distributed between the loosely packed SnO2 nanoparticles in such a way that the nanoporous structure with a large amount of void spaces could be prepared. The obtained SnO2/GNS exhibits a reversible capacity of 810 mAh/g; furthermore, its cycling performance is drastically enhanced in comparison with that of the bare SnO2 nanoparticle. After 30 cycles, the charge capacity of SnO2/GNS still remained 570 mAh/g, that is, about 70% retention of the reversible capacity, while the specific capacity of the bare SnO2 nanoparticle on the first charge was 550 mAh/g, dropping rapidly to 60 mAh/g only after 15 cycles. The dimensional confinement of tin oxide nanoparticles by the surrounding GNS limits the volume expansion upon lithium insertion, and the developed pores between SnO2 and GNS could be used as buffered spaces during charge/discharge, resulting in the superior cyclic performances.
Lithium-ion batteries (LIB), as power sources for mobile communication devices, portable electronic devices, and electrical/hybrid vehicles, have attracted special attention in the scientific and industrial fields due to their high electromotive force and high energy density. For an anode material in LIB, graphite is usually employed as a standard electrode because it can be reversibly charged and discharged under intercalation potentials with reasonable specific capacity.1,2 However, to meet the increasing demand for batteries with higher energy density, much research attempt has been made to explore new electrode materials or design novel nanostructures of electrode materials.3-7 Especially, among the carbonaceous materials, graphene-based materials could be one of the promising alternatives as an anode in LIB because these materials have superior electrical conductivities than graphitic carbon, high surface area of over 2600 m2/g, chemical tolerance, and a broad electrochemical window that would be very advantageous for application in energy technologies.8-12 Recently, we are quite successful in the controlled reassembling of exfoliated graphene nanosheets * Towhomcorrespondenceshouldbeaddressed.E-mail:
[email protected]. Tel/Fax: +81-29-861-5648. 10.1021/nl802484w CCC: $40.75 Published on Web 12/17/2008
2009 American Chemical Society
(GNS) with carbon nanotubes and fullerenes.13 According to our previous experiments, the obtained GNS families show very large reversible capacity, which can be attributed to the increased basal spacing of GNS in comparison with that of graphite. Such an exfoliation-reassembling method can be extended to fabricate a new hybrid electrode material that is composed of GNS and transition metal oxide nanoparticles. For example, tin oxide (SnO2) could be a good substitute for the carbon anode in LIB because its high theoretical Li+ storage capacity of 782 mAh/g is much larger than that (372 mAh/g) of graphite. However, similar to the other lithium reactive electrode materials, tin oxide shows very large volume change of about 300% during charge/discharge process, which causes crumbling and cracking of electrode, leading to electrical disconnection from current collectors.14-16 Therefore, most of tin oxide electrode materials suffer from the rapid fading of capacity. Although several attempts have been made to prepare new nanostructures based on the tin oxide,17,18 most efforts so far proved not to be very successful in enhancing the cyclability of tin oxide based electrodes. To circumvent these problems, we have attempted to reassemble graphene nanosheets (GNS) in the presence of
Figure 1. SEM and TEM observation of GNS and SnO2/GNS: SEM images for (a) GNS, (b) SnO2/GNS. Cross-sectional TEM images for (c) GNS, (d) GNS (high magnification), (e) as-prepared SnO2/GNS, (f) heat-treated SnO2/GNS. The white arrows denote the graphene nanosheets.
Figure 2. (A) Charge/discharge profile for SnO2/GNS. (B) Cyclic performances for (a) bare SnO2 nanoparticle, (b) graphite, (c) GNS, and (d) SnO2/GNS.
tin oxide nanoparticles without any deterioration of fundamental electrochemical properties of both components. In the obtained SnO2/GNS, not only GNS but also tin oxide nanoparticles could play a role as electrode materials to get a synergetic effect. As seen in Scheme 1, our key strategy in this work is to develop nanostructured SnO2/GNS electrodes in which the dimensional confinement of tin oxide nanoparticles by the surrounding GNS limits the volume Nano Lett., Vol. 9, No. 1, 2009
expansion upon lithium insertion, and the developed nanopores between SnO2 and GNS could be used as buffered spaces during charge/discharge, resulting in the superior cyclic performances and eventually higher reversible capacities in comparison with the ordinary SnO2. Graphene nanosheets (GNS) were prepared via the chemical reduction of exfoliated graphite oxide materials, while tin oxide (SnO2) nanoparticles were obtained by the con73
Scheme 1. Schematic Illustration for the Synthesis and the Structure of SnO2/GNS
trolled hydrolysis of SnCl4 with NaOH. The reduced graphene nanosheets were dispersed in the ethylene glycol, and then, reassembled in the presence of SnO2 nanoparticles as shown in the Scheme 1. The molar ratio of tin to carbon ([Sn]tin oxide/[C]GNS) was fixed to 1.5. As to the experimental details, please see Supporting Information. To investigate the morphology of the products, field emission scanning electron microscopic (FE-SEM) images were taken for the GNS and the SnO2/GNS. Figure 1a presents the representative SEM image of GNS from the top view, showing the layered platelets composed of curled nanosheets. However, as shown in the inset of Figure 1a that was taken from the edge-side of GNS, the restacked GNS were made up of the finely divided nanoplates in which a thickness of an individual stack is estimated to be less than 10 nm, suggesting that the self-restacked GNS consist of several layers. As shown in Figure 1b, the SEM image of the SnO2/ GNS also exhibits a similar morphology to that of the GNS, revealing that a fine structural manipulation of the GNS is successfully achieved even after the reassembling process with tin oxide nanoparticles. The cross-sectional transmission electron microscope (TEM) analyses for products were also used to elucidate the structural features of GNS and SnO2/ GNS materials. The TEM image of typical GNS (Figure 1c) reveals a crumpled and rippled structure where the black stripes can be attributed to the graphene nanosheets. Some parts of GNS are bent and wavy as a result of elastic deformation upon the exfoliation and reassembling process. A closer inspection of TEM image of GNS was made at higher magnification (Figure 1d). The d-spacing between two graphene nanosheets is estimated to be about 0.39 nm that is much larger than that (0.335 nm) of graphite, indicating that graphene interlayer distances are increased by more than 9% after reassembling reaction. It is highly plausible that such a significant expansion of interlayer distance might provide additional intercalation sites for accommodation of lithium ions, resulting in the enhancement of specific capacity.13 On the other hand, these graphene layers dispersed in the ethylene glycol solution were reassembled in the 74
presence of SnO2 nanoparticles to prepare the SnO2/GNS nanocomposite. As seen in Figure 1e, the TEM images of the as-prepared SnO2/GNS exhibit two distinct images, lines, and spherical shapes. The former are due to the exfoliated nanosheets of graphene whereas the latter are from the SnO2 nanoparticles. Therefore, it is quite clear that the graphene nanosheets are homogeneously distributed between the loosely packed SnO2 nanoparticles in such a way that the nanoporous composite with large amount of void spaces could be prepared. It is worthwhile to note here that the thin graphene layers are randomly hybridized with SnO2 nanoparticles to preserve the 3-dimensionally delaminated structure of the graphene nanosheets. In the TEM image of SnO2/ GNS heat-treated at 400 °C for 2 h (Figure 1f), the black stripes due to the graphene layers are still discernible between SnO2 nanoparticles, supporting that the delaminated structure of graphene layers in this nanocomposite still maintained even after the heat-treatment. An average particle size of SnO2 particles in the heat-treated SnO2/GNS nanocomposite are about 5.4 ( 2.1 nm (100 counts). It seems that the randomly distributed graphene layers surrounding SnO2 nanoparticles inhibit the particle growth of SnO2 upon calcination. In the inset of Figure 1f, about 0.33 nm size of the lattice fringe of SnO2 is clearly observed, which can be indexed as the (110) plane of the rutile structure.19 Furthermore, no significant reduction of interparticle distances could be seen even after calcination, supporting that the present nanocomposite sustains its nanoporous structure. In the calcined sample, its crystallinity is slightly increased as a result of surface curing during the heat-treatment, which could be observed in the round shape of the particles. The highly nanoporous and delaminated structure of SnO2/ GNS might be applied as potential anode materials in the LIB. The lithium insertion/extraction profiles of SnO2/GNS at a current density of 50 mA/g are demonstrated in Figure 2A, in which the voltage range is fixed from 0.05 to 2 V. Normally, SnO2-based anode showed classical plateaus around 0.8 V similar to that of a bulk SnO2 system, which is well known as the reaction of SnO2 with lithium to form the solid electrolyte interface (SEI) layers.20,21 However, in case of the SnO2/GNS, this plateau around 0.8 V nearly disappeared after the first cycle, indicating that Li2O is formed only in small quantities as a result of small particle size of SnO2 in this nanocomposite. Such a result is related to the enhanced surface electrochemical reactivity due to the large surface-to-volume ratio.22,23 The second charge capacity of the SnO2/GNS is 860 mAh/g at the current density of 50 mA/g, suggesting that the extraordinary high capacity is due to the nanocrystalline nature of the SnO2/GNS composite. A comparison of the charge/discharge cyclic performances for graphite, GNS, bare SnO2, and SnO2/GNS composite is shown in Figure 2B. After 30 cycles, graphite shows about 78% retention (240 mAh/g) of the initial capacity, whereas GNS exhibits about 57% retention (300 mAh/g) of the initial capacity. This shows that the specific capacity of GNS fades faster than that of graphite; however, the reversible capacity of GNS is still larger than that of graphite even after 30 cycles. On the other hand, SnO2/GNS exhibits a reversible Nano Lett., Vol. 9, No. 1, 2009
capacity of 810 mAh/g. Furthermore, its cycling performance is drastically enhanced, as seen in the figure. After 30 cycles, the charge capacity still remained 570 mAh/g that is about 70% retention of the reversible capacity, while the specific capacity of SnO2 nanoparticle on the second charge was 550 mAh/g, dropping rapidly to 60 mAh/g only after 15 cycles. The Coulombic efficiency of SnO2/GNS at the first cycle is around 43%; however, it is above 95% after 10 cycles, which is slightly smaller than that of graphite (see the Supporting Information). In order to clarify the influence of the hybridization on the electrochemical performance of SnO2/ GNS, we have calculated theoretically the capacity of physical mixture of the pristine materials (SnO2 and graphite) from the theoretical capacities of the SnO2 (782 mAhg-1) and the graphite (372 mAhg-1), for comparison. The elemental ratio of both components in this hypothetical mixture was adjusted to the chemical composition of SnO2/ GNS with 40% of carbon and 60% of SnO2. On the basis of the equation described below, we could calculate a theoretical capacity (C) of the hypothetical mixture, as follows: Ctheoretical ) CSnO2 * %mass of SnO2 + CGraphite * %mass of Graphite ) 782 * 0.6 + 372 * 0.4 ) 618 mA hg-1. The observed capacity (570 mA hg-1) of the SnO2/GNS is comparable to the theoretical capacity of the physical mixture even after 30 cycles, highlighting the synergetic effect for the enhanced cyclic performance. As pointed out before, the main reason for rapid fading of SnO2 electrode is that a large volume expansion of the SnO2 material occurs during the cycling, leading to the pulverization of the electrode.14-16 In the case of the SnO2/GNS, graphene nanosheets are homogeneously distributed between tin oxide nanoparticles. Such a dimensional confinement of the SnO2 nanoparticles by the surrounding graphene limits the volume expansion upon lithium insertion, therefore the stress formed during the process of lithium insertion was avoided. Even though the volume expansion has happened, the electrode was not pulverized because SnO2/GNS has enough void spaces to buffer volume change. Furthermore, graphene nanosheets in hybrid material could act as not only lithium storage electrodes but also electronic conductive channels to improve the electrochemical performances. The present SnO2/GNS materials are
Nano Lett., Vol. 9, No. 1, 2009
expected to be used as anode materials by tuning the charge/ discharge profiles. Acknowledgment. Dr. Eiji Hosono in AIST Tsukuba is acknowledged for the SEM measurements and the fruitful discussion on results. Supporting Information Available: Materials and methods, XRD patterns, FT-IR spectra, N2 adsorption-desorption isotherms, and the charge/discharge profiles of the products. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Wu, Y. P.; Rahm, E.; Holze, R. J. Power Sources 2003, 114, 228. (2) Buqa, H.; Goers, D.; Holzapfel, M.; Spahr, M. E.; Novak, P. J. Electrochem. Soc. 2005, 152, A474. (3) Idota, Y.; Kubota, T.; Matsufuji, A.; Maekawa, Y.; Miyasaka, T. Science 1997, 276, 1395. (4) Maier, J. Nat. Mater. 2005, 4, 805. (5) Lou, X. W.; Wang, Y.; Yuan, C.; Lee, J. Y.; Archer, L. A. AdV. Mater. 2006, 18, 2325. (6) Chan, C. K.; Zhang, X. F.; Cui, Y. Nano Lett. 2008, 8, 307. (7) Kim, D. W.; Hwang, I. S.; Kwon, S. J.; Kang, H. Y.; Park, K. S.; Choi, Y. J.; Choi, K. J.; Park, J. G. Nano Lett. 2007, 7, 3041. (8) Li, X.; Wang, X.; Zhang, L.; Lee, S.; Dai, H. Science 2008, 319, 1229. (9) Meyer, J. C.; Geim, A. K.; Katsnelson, M. I.; Novoselov, K. S.; Booth, T. J.; Roth, S. Nature 2007, 446, 60. (10) Watcharotone, S.; Dikin, D. A.; Stankovich, S.; Piner, R.; Jung, I.; Dommett, G. H. B.; Evmenenko, G.; Wu, S. E.; Chen, S. F.; Liu, C. P.; Nguyen, S. T.; Ruoff, R. S. Nano Lett. 2007, 7, 1888. (11) Wang, X.; Zhi, L.; Mu¨llen, K. Nano Lett. 2008, 8, 323. (12) Gilje, S.; Han, S.; Wang, M.; Wang, K. L.; Kaner, R. B. Nano Lett. 2007, 7, 3394. (13) Yoo, E.; Kim, J.; Hosono, E.; Zhou, H. S.; Kudo, T.; Honma, I. Nano Lett. 2008, 8, 2277. (14) Wang, Y.; Lee, J. Y.; Zeng, H. C. Chem. Mater. 2005, 17, 3899. (15) Wei, X. M.; Zeng, H. C. Chem. Mater. 2003, 15, 433. (16) Fan, J.; Wang, T.; Yu, C.; Tu, B.; Jiang, Z.; Zhao, D. AdV. Mater. 2004, 16, 1432. (17) Deng, D.; Lee, J. Y. Chem. Mater. 2008, 20, 1841. (18) Zhao, N.; Wang, G.; Huang, Y.; Wang, B.; Yao, B.; Wu, Y. Chem. Mater. 2008, 20, 2612. (19) Wang, W.; Xu, C.; Wang, X.; Liu, Y.; Zhan, Y.; Zheng, C.; Song, F.; Wang, G. J. Mater. Chem. 2002, 12, 1922. (20) Kim, C.; Noh, M.; Choi, M.; Cho, J.; Park, B. Chem. Mater. 2005, 17, 3297. (21) Sun, X.; Liu, J.; Li, Y. Chem. Mater. 2006, 18, 3486. (22) Ahn, H. J.; Choi, H. C.; Park, K. W.; Kim, S. B.; Sung, Y. E. J. Phys. Chem. B 2004, 108, 9815. (23) Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J. M. Nature 2000, 407, 495.
NL802484W
75
