Tin nanoparticle-loaded porous carbon nanofiber composite anodes for high current lithium-ion batteries

Tin nanoparticle-loaded porous carbon nanofiber composite anodes for high current lithium-ion batteries

Journal of Power Sources 278 (2015) 660e667 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/lo...

3MB Sizes 5 Downloads 38 Views

Journal of Power Sources 278 (2015) 660e667

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Short communication

Tin nanoparticle-loaded porous carbon nanofiber composite anodes for high current lithium-ion batteries Zhen Shen a, Yi Hu a, b, *, Yanli Chen a, Xiangwu Zhang c, Kehao Wang a, Renzhong Chen a a

Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education, Zhejiang Sci-Tech University, Hangzhou 310018, China Engineering Research Center for Eco-Dying & Finishing of Textiles, Ministry of Education, Zhejiang Sci-Tech University, Hangzhou 310018, China c Fiber and Polymer Science Program, Department of Textile Engineering, Chemistry and Science, North Carolina State University, Raleigh, NC 27695-8301, USA b

h i g h l i g h t s  Sn nanoparticle-loaded porous and nonporous CNF anodes are introduced.  A porous structure was formed by the volatilization of mineral oil.  The morphology exhibits nanosized Sn dispersed in porous CNF uniformly.  Ultra-disperse Sn and porous CNF led to high capacity in high current.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 October 2014 Received in revised form 18 December 2014 Accepted 22 December 2014 Available online 23 December 2014

Metallic Sn is a promising high-capacity anode material for use in lithium-ion batteries (LIBs), but its huge volume variation during lithium ion insertion/extraction typically results in poor cycling stability. To address this, we demonstrate the fabrication of Sn nanoparticle-loaded porous carbon nanofiber (SnPCNF) composites via the electrospinning of Sn(II) acetate/mineral oil/polyacrylonitrile precursors in N,N-dimethylformamide solvent and their subsequent carbonization at 700  C under an argon atmosphere. This is shown to result in an even distribution of pores on the surface of the nanofibers, allowing the Sn-PCNF composite to be used directly as an anode in lithium-ion batteries without the need to add non-active materials such as polymer binders or electrical conductors. With a discharge capacity of around 774 mA h g1 achieved at a high current of 0.8 A g1 over 200 cycles, this material clearly has a high rate capability and excellent cyclic stability, and thanks to its unique structure and properties, is an excellent candidate for use as an anode material in high-current rechargeable lithium-ion batteries. © 2014 Elsevier B.V. All rights reserved.

Keywords: Sn(II) acetate Electrospinning Porous carbon nanofibers Lithium-ion battery

1. Introduction Lithium-ion batteries (LIBs) have attracted increasing research interest over recent years due to their superior energy density, operating voltage and cycle life when compared with the majority of other rechargeable battery technologies [1]. However, although the graphite used as an anode material in commercial LIBs offers stable cycling performance and low cost [2], its low theoretical capacity (372 mA h g1) greatly limits its further application in future applications [3,4]. Accordingly, metallic Sn in the form of a

* Corresponding author. 928 Second Avenue, Xiasha Higher Education Zone, Hangzhou 310018, China. E-mail address: [email protected] (Y. Hu). http://dx.doi.org/10.1016/j.jpowsour.2014.12.106 0378-7753/© 2014 Elsevier B.V. All rights reserved.

Li4.4Sn alloy has been proposed as a more suitable anode material based on its much higher theoretical capacity of 992 mA h g1 [5,6]; however, unlike graphite, these anodes suffer from an enormous volume expansion and contraction during Li insertion/extraction [7,8]. The resulting cracking of the Sn particles invariably leads to mechanical failure, loss of electrical contact, and capacity fading [6e9]. This is further exacerbated by the coalescence of Sn particles during the discharge process [10,11], and is the reason why the practical realization of pure-Sn anodes has not been achieved. Various strategies have been developed to try and overcome the inherent weaknesses of Sn-based anodes, which include using combining a carbon composite with nanocrystallization [12,13]. Indeed, a great deal of research effort has already been directed toward reducing the size of Sn particles [14,15], as well as developing new carbon materials such as carbon nanofibers (CNF),

Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667


carbon nanotubes, graphene, graphene oxide, etc. Of these, CNF prepared by electrospinning has been demonstrated to offer significant advantages in terms of its high flexibility, large surface area and good conductivity [16,17], with the addition of nanostructured Sn reported to improve cycle performance by enhancing conductivity, while limiting the volume change in the Sn particles [18,19]. This electrochemical performance can be further improved by making the smooth fibrous structure of CNF more porous to create an ultrahigh specific surface area that can provide more charge transfer [20e22]. When compared to traditional electrodes, the use of binder-free electrodes in LIBs can provide a high energy density, good electrical conductivity and greater ease of preparation [11,23,24]. Building on these past studies, we fabricated a Sn nanoparticleloaded porous carbon nanofiber (Sn-PCNF) composite by means of electrospinning a precursor of Sn(II) acetate/mineral oil/polyacrylonitrile nanofibers, which was then stabilized in air and carbonized in an argon atmosphere. This process is shown schematically in Fig. 1, and relies on the volatilization of the mineral oil to create the porous structure needed. The structural morphology of this material is herein discussed, and its electrochemical performance when used as a binder-free anode in a LIB is assessed.

tube furnace at 700  C for 6 h in an argon atmosphere (heating rate: 2  C min1) to obtain the Sn-PCNF composites. For comparison, Sn nanoparticle-loaded carbon nanofiber composites were designed under the same conditions without mineral oil.

2. Experimental

The electrochemical performance of the Sn-PCNF anode material was evaluated by using it as a working electrode without any binder or conductive filler in CR2032 coin-type cells with lithium foil as the counter electrode and 1 M LiPF6 dissolved in a mixture of ethyl carbonate (EC), dimethyl carbonate (DMC) and diethyl carbonate (DEC) (1:1:1, v/v/v) as the electrolyte. These cells were assembled in a glovebox (Mbraun Labstar 1450/780) filled with highly pure argon gas, with direct contact between the working and counter electrodes prevented by using a Celgard 2400 separator. Once assembled, the cells were aged for 24 h prior to any measurements so as to ensure adequate percolation of the electrolyte to the electrodes. Charge and discharge measurements were carried out using a LAND-CT2001A battery-testing system at a high current density of 0.8 A g1 and cut-off potentials of 0.01 and 3.00 V. The rate performance was measured at current densities of 0.8, 1.6, 2.4, and 4 A g1. Cyclic voltammogram (CV) experiments were conducted at room temperature on an electrochemical workstation (Metrohm, PGSTAT 302N) at a scan rate of 0.5 mV s1, with the same workstation also used to measure the electrochemical impedance spectroscopy (EIS) at a voltage of 10 mV within a frequency range of 100 kHz to 0.1 Hz after first equilibrated the cells for 1 h.

2.1. Preparation of samples The precursor solution for electrospinning was prepared by dissolving 0.40 g polyacrylonitrile (PAN, Mw ¼ 150,000 g mol1, J&K Chemical Inc.) and 0.05 g mineral oil (Acros) in 4.43 g N,Ndimethylformamide (DMF, TianJin Yongda Chemical Reagent Co. Ltd) by magnetic stirring for 24 h at around 60  C. After that, 0.12 g Sn(II) acetate (J&K Chemical Inc.) was added into precursor solution, and then vigorously stirred for 6 h. As for a typical electrospinning process, the spinneret had an inner diameter of 0.43 mm and grounded aluminum foil was used as the collector. A distance of 15 cm and a direct current voltage of 15 kV were maintained between the tip of the spinneret and the collector. The as-collected electrospun nanofibers were stabilized in an air environment at 280  C for 2 h (heating rate: 5  C min1) and then carbonized in a

2.2. Characterization The structure of Sn-PCNF composites were investigated by powder X-ray diffraction (XRD, Thermo ARL XTRA) and Raman spectra were recorded using a confocal Renishaw Raman microscope with a 633 nm of HeeNe laser. X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific K-Alpha) was conducted to support the composition of Sn-PCNF composites. The morphology of Sn-PCNF composites were examined by scanning electron microscope (SEM, Zeiss vltra55) and transmission electron microscope (TEM, JEOL JEM-2100). EDX mapping has been acquired at Kedge of C and O and L-edge of Sn. Nitrogen adsorption/desorption isotherms were measured on a Micromeritics ASAP 2020. The BrunauereEmmetteTeller (BET) methods used to analyze the specific surface areas of the samples. 2.3. Electrochemical measurements

3. Results and discussion 3.1. Structure and composition of the Sn-PCNF anode material

Fig. 1. Schematic of Sn-PCNF composite.

In the X-ray diffraction (XRD) patterns of the Sn-PCNF composites and CNF given in Fig. 2a, the main reflections of metallic Sn are indexed according to Miller's notation, while the reflections from SnO2 are marked with an asterisk. From this, we can see that the strong peaks in the 2q regions of 30e33 and 44e46 correlate to the (200), (101), (220) and (211) peaks of metallic Sn. Reflections from SnO2 are also evident, which may well be because carbonization at 700  C in an argon atmosphere does not allow for the complete reduction of any SnO2 that may be formed through decomposition and oxidation of Sn(II) acetate during stabilization at 280  C in air [20]. Nevertheless, the SnO2 content is substantially lower than that of metallic Sn. Also seen in Fig. 2a is a broad and weak diffraction peak at around 2q ¼ 26 , which is attributed to the (002) planes of the amorphous CNF matrix [25,26]. This is supported by the Raman spectrum of the Sn-PCNF composite (Fig. 2b),


Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667

Fig. 2. (a) XRD patterns of Sn-PCNF composites and CNF. (b) Raman spectrum of as-prepared Sn-PCNF composites. (c) Low resolution XPS spectrum of Sn-PCNF composites. (d) High resolution XPS spectrum of O1s.

in which the two broad peaks at 1351 and 1588 cm1 can be assigned to the D and G bands of amorphous carbon, respectively [27]. The corresponding ID/IG intensity ratio was ~1.12, which infers the existence of abundant vacancies and defects in the amorphous CNF. These would not only accelerate the transfer of lithium ions, but also contribute to the reversible capacity by providing extra sites for Li storage [28]. The chemical composition of the surface of the Sn-PCNF composite was investigated by X-ray photoelectron spectroscopy (XPS), with the low-resolution XPS spectrum in Fig. 2c proving that this material is composed of tin, nitrogen, oxygen and carbon. It is known that some of the N species in polyacrylonitrile can be retained following carbonization at 700  C in Ar, with the resulting N-doped carbon providing high conductivity [29,30]. In addition, the existence of CeO, C]O and SneO bonding were confirmed by the high-resolution O1s XPS spectrum of Sn-PCNF composite (Fig. 2d), which indicated that the presence of oxygen is ascribed to carbon matrix, H2O and SnO2. The Sn content of the Sn-PCNF composite was determined through energy dispersive spectroscopy (EDS, Supplementary information Fig. S1) to be approximately 39.4 wt % and the embedded Sn content is much greater than that of the spilled Sn content, meaning that the latter has little impact on the overall performance of the electrodes (Fig. S2). The SEM images of the Sn-PCNF composite taken at different magnifications in Fig. 3a and b show it to be made up of

independent nanofibers with an average diameter of 200 nm. A small number of bright nanoparticles around 100e300 nm in size can also be seen to form on the surface of the nanofibers (Fig. 3b). Transmission electron microscopy (TEM) was used to get a clearer view of the surface texture, which as shown in Fig. 3c, revealed the presence of numerous pores created through the decomposition and volatilization of the mineral oil used in the electrospinning and carbonization process [31]. From the inset of Fig. 3c, it is clear that the Sn-PCNF composite exhibits a BrunauereEmmetteTeller (BET) surface area of 872.16 m2 g1 and a total pore volume of 0.990 cm3 g1, both values being much higher than those of Sn-CNF composites (708.22 m2 g1 and 0.814 cm3 g1, respectively). It is the existence of abundant pores that is considered responsible for this larger surface area, which should beneficial to Li storage [32]. It can also be observed in Fig. 3d that the nanoparticles load and halfembed on the surface of the nanofibers, while the high-resolution transmission electron microscopy (HRTEM) images in Fig. 3e show that these nanoparticles are also homogeneously dispersed throughout the nanofiber. Given the small size of these particles, this should improve the cycle performance when used in a LIB [14]. The small amount of crystals observable in Fig. 3f have a layer distance of 0.338 nm, which corresponds to the d110 value of SnO2 [23]. The energy dispersive X-ray (EDX) maps of C, O and Sn in Fig. 4a show that there is a uniform distribution of these elements

Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667


Fig. 3. (a, b) SEM, (c, d) TEM and (e, f) HRTEM images at different magnifications of the as-prepared Sn-PCNF composite. Inset: Nitrogen adsorptionedesorption isotherms of SnPCNF and Sn-CNF composites.

throughout the entirety of the Sn-PCNF composite, while the EDS spectrum in Fig. 4b indicates that the molar ratio of Sn to O is less than 1:2, thus further confirming that only a small portion of metallic Sn was oxidized during fabrication [33]. Finally, the EDX/ EDS results obtained for a nanoparticle on the surface of the SnPCNF composite (Fig. 4c and d) show that it has very little residual carbon surrounding it that could potentially restrain its volume change during lithiation/delithiation [22].

3.2. Electrochemical performance Fig. 5 shows typical cyclic voltammograms (CV) for the Sn-PCNF composite electrode for the first three cycles at a scan rate of 0.5 mV s1 and over a voltage range of 0e3.0 V. In the first cathodic scan, the broad reduction peak from 0.6 to 0.8 V reflects the reduction of SnO2 to Sn. This is accompanied by the formation of a solid-electrolyte interphase (SEI) film due to the decomposition of the electrolyte, which results in an irreversible capacity fade [34,35]. A second reduction peak occurs at around 0.08 V and


Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667

  Sn þ xLiþ þ xe 4Lix Sn 0  x  4:4


C þ xLiþ þ xe /Lix C

(3) 1

Fig. 4. (a, c) EDX elemental mapping of STEM images and (b, d) EDS spectra obtained from different areas of a Sn-PCNF composite.

corresponds to the alloying of LixSn [19,36,37]. Two oxidation peaks can also be observed at around 0.12 and 0.7 V in the anodic scan of later cycles, and these correspond to the extraction of lithium ions from the carbon matrix and the de-alloying of LixSn, respectively [38]. Meanwhile, the weak oxidation peak at around 1.22 V demonstrates the partial reversibility of the reduction [19]. The fact that these CV curves almost overlap from the second to third scan indicates that a stable SEI is formed during the first cycle. From the room temperature charge/discharge curves of the SnPCNF composite anode obtained at different cycles between 0.01 and 3.0 V with a constant high current density of 0.8 A g1 (Fig. 6), it is apparent the principal electrode reactions are [35,39]:

SnO2 þ 4Liþ þ 4e /Sn þ 2Li2 O


For the first cycle, the discharge capacity of 1880 mA h g and charge capacity of 1162 mA h g1 equates to an irreversible capacity of 718 mA h g1 and a Coulombic efficiency of approximately 61.8%. This can be attributed to the inevitable formation of a large quantity of SEI film (which has a potential plateau at 0.8 V [20]) due to the high surface area of the Sn-PCNF composite [40], as well as the irreversible reduction of SnO2 to Sn (Eq. (1)). The cycling performance of PCNF matrix, Sn-PCNF and Sn-CNF composites electrodes was also investigated, as shown in Fig. 7, revealing the Sn-CNF composite to have a lower initial discharge capacity (1332 mA h g1) and Coulombic efficiency (56.5%) than the porous-structured Sn-PCNF composite. More importantly, even after 200 deep charge/discharge cycles the Sn-PCNF composite continues to have a higher discharge capacity (774 mA h g1) and Coulombic efficiency (>93%) than the Sn-CNF composite (355 mA h g1 and an unstable Coulombic efficiency). This is considered to be mainly the result of the PCNF matrix acting as a buffer against the volume change of Sn during charge/discharge process [41]. The discharge capacity of PCNF is around 327 mA h g1 and its Coulombic efficiency under the same conditions is 94%, which indicates that PCNF enhances the capacity and stability of the electrode. Interestingly, the specific capacity fades rapidly in the first 25 cycles, but then increases during the subsequent 100 cycles, which would seem likely to be the result of severe pulverization caused by the dramatic volume change of large Sn particles exposed on the surface of nanofibers [22], followed later by an infiltration of the electrolyte into the active material [42]. In Fig. 8, it can be seen that the Sn-PCNF composite displays an excellent rate capability, and is capable of delivering discharge capacities of around 792.7, 586.6, 482.3 and 349.2 mA h g1 at high current densities of 0.8, 1.6, 2.4 and 4 A g1, respectively. In contrast, the discharge capacities of the Sn-CNF composite at equivalent current densities are 834.2, 224.5, 141.8, 55.7 mA h g1, indicating that it decays much more rapidly. Moreover, once the current rate returned to 0.8 A g1, the discharge capacities of the Sn-PCNF and Sn-CNF composites increased back up to 726.0 and 632.2 mA h g1, respectively. This clearly indicates that the Sn-PCNF composite is able to withstand a much greater variation and has a better capacity recovery performance, further confirming the advantage of using a porous rather than a smooth fibrous structure in high-current LIB applications. To further understand the reasons for the better rate performance of the Sn-PCNF composite, its electrochemical impedance spectroscopy (EIS) at different stages is compared to the Sn-CNF composite in Fig. 9. The depressed semicircles in the high to middle frequency range represent the initial interfacial resistance and charge-transfer impedance [43], and as the diameter of these is clearly smaller with the Sn-PCNF composite is smaller than that of the Sn-CNF composites in the high-medium frequency region and after the fifth lithiation to 0.3 V. This observation suggests that the Sn-PCNF composite anode has a lower initial interfacial resistance and charge-transfer impedance, and therefore a better rate performance, due to the greater number of charge transfer pathways afforded by the porous CNF [44]. TEM imaging was used to observe the morphological change in the Sn-PCNF composite anode after 200 cycles at a high current density of 0.8 A g1 (Fig. 10). When compared to Fig. 3, it is evident that all surfaces of the nanofibers become rough after cycling due to repeated Li insertion/extraction [44]. Fig. 10a and b also show that

Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667

Fig. 5. Cyclic voltammetry (CV) curves for a Sn-PCNF composite electrode at a 0.5 mV s1 scanning rate.

Fig. 6. Charge and discharge curves for Sn-PCNF at a current density of 0.8 A g1.


Fig. 8. Rate performance of Sn-PCNF and Sn-CNF composites cycled at different current densities.

Fig. 9. Electrochemical impedance spectra obtained from fresh coin cells (Sn-PCNF0 and Sn-CNF-0) and after the fifth lithiation to 0.3 V (Sn-PCNF-5 and Sn-CNF-5).

separated from the PCNF (see Fig. 1) as part of the capacity fade that occurs in the first 25 cycles (see Fig. 7). The outstanding electrochemical performances of the Sn-PCNF anode can be explained as follows. First, the uniform dispersion of Sn nanoparticles in the CNF matrix provides a large number of active sites for Li ion storage and shortens the distance for Li ion transfer, thus leading to a higher reversible capacity and rate capability. Second, the large surface area associated with the porous CNF helps maintain highly efficient electron and ion transport, thereby contributing to the notable improvement in reversible capacity and rate performance. Finally, the amorphous CNF matrix can readily accommodate the volume change in the Sn nanoparticles, resulting in excellent cycling stability at high current. 4. Conclusions Fig. 7. Cycling performance and Coulombic efficiency of the PCNF matrix, Sn-PCNF and Sn-CNF composites at a high current density of 0.8 A g1.

the Sn-PCNF composite anode retains its original fibrous geometry even after 200 cycles, and that its better cycling performance results in a surface that is more uniform than that of Sn-CNF (Fig. 10c and d). Furthermore, as there was no evidence of any particles, it can be safely assumed that any large particles are pulverized and

A nanostructured Sn-PCNF composite, in which Sn is uniformly distributed throughout a porous CNF matrix, has been successfully synthesized through a readily scalable electrospinning approach and subsequent carbonization. The Sn-PCNF composites obtained exhibited superior rate performance at high current densities and long-term cycle stability when used as a binder-free electrode material in a Li-ion battery. A high capacity of 774 mA h g1 was achieved at a high current density of 0.8 A g1 after 200 cycles, with this excellent electrochemical performance attributed to the high-


Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667

Fig. 10. TEM images of the Sn-PCNF and Sn-CNF composite anodes after 200 cycles.

dispersion of Sn and the unique porous structure of the amorphous CNF. This Sn-PCNF composite therefore represents a promising Snbased material for use as a high-capacity, high-current anode in next generation LIBs. Acknowledgments The Project Supported by Zhejiang Provincial Natural Science Foundation of China (LY12E03005) and Supported by Young Researchers Foundation of Zhejiang Provincial Top Key Academic Discipline of Chemical Engineering and Technology (ZYG2011004), and Supported by Zhejiang Provincial Science and Technology Innovation Team Project (1101318-E), 2013 Jia Xing High-tech Achievement Transformation Project (2013BY31001). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2014.12.106. References [1] W.W. Lee, J.-M. Lee, J. Mater. Chem. A 2 (2014) 1589e1626. [2] Y.P. Wu, E. Rahm, R. Holze, J. Power Sources 114 (2003) 228e236. [3] Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa, T. Miyasaka, Science 276 (1997) 1395e1397. [4] M. Endo, C. Kim, K. Nishimura, T. Fujino, K. Miyashita, Carbon 38 (2000) 183e197. [5] W.M. Zhang, J.S. Hu, Y.G. Guo, S.F. Zheng, L.S. Zhong, W.G. Song, L.J. Wan, Adv. Mater. 20 (2008) 1160e1165.

[6] L. Ji, Z. Tan, T. Kuykendall, E.J. An, Y. Fu, V. Battaglia, Y. Zhang, Energy Environ. Sci. 4 (2011) 3611e3616. [7] J. Zai, K. Wang, Y. Su, X. Qian, J. Chen, J. Power Sources 196 (2011) 3650e3654. [8] H. Liu, R. Hu, W. Sun, M. Zeng, J. Liu, L. Yang, M. Zhu, J. Power Sources 242 (2013) 114e121. [9] N. Li, H. Song, H. Cui, G. Yang, C. Wang, J. Mater. Chem. A 2 (2014) 2526e2537. [10] J. Qin, C. He, N. Zhao, Z. Wang, C. Shi, E.-Z. Liu, J. Li, ACS Nano 8 (2014) 1728e1738. [11] W. Song, X. Hou, X. Wang, B. Liu, D. Chen, G. Shen, Energy Technol. 2 (2014) 370e375. [12] N. Zhang, Q. Zhao, X. Han, J. Yang, J. Chen, Nanoscale 6 (2014) 2827e2832. [13] X.W. Lou, C.M. Li, L.A. Archer, Adv. Mater. 21 (2009) 2536e2539. [14] X. Zhou, Z. Dai, S. Liu, J. Bao, Y.G. Guo, Adv. Mater. 26 (2014) 3943e3949. [15] Z. Zhu, S. Wang, J. Du, Q. Jin, T. Zhang, F. Cheng, J. Chen, Nano Lett. 14 (2013) 153e157. [16] Z. Dong, S.J. Kennedy, Y. Wu, J. Power Sources 196 (2011) 4886e4904. [17] C. Lai, Z. Zhou, L. Zhang, X. Wang, Q. Zhou, Y. Zhao, Y. Wang, X.-F. Wu, Z. Zhu, H. Fong, J. Power Sources 247 (2014) 134e141. [18] Y. Yu, Q. Yang, D. Teng, X. Yang, S. Ryu, Electrochem. Commun. 12 (2010) 1187e1190. [19] J. Zhu, D. Lei, G. Zhang, Q. Li, B. Lu, T. Wang, Nanoscale 5 (2013) 5499e5505. [20] I. Meschini, F. Nobili, M. Mancini, R. Marassi, R. Tossici, A. Savoini, M.L. Focarete, F. Croce, J. Power Sources 226 (2013) 241e248. [21] G.-H. An, H.-J. Ahn, Carbon 65 (2013) 87e96. [22] Y. Yu, L. Gu, C. Wang, A. Dhanabalan, P.A. van Aken, J. Maier, Angew. Chem. 48 (2009) 6485e6489. [23] L. Zou, L. Gan, F. Kang, M. Wang, W. Shen, Z. Huang, J. Power Sources 195 (2010) 1216e1220. [24] Y. Liu, K. Huang, Y. Fan, Q. Zhang, F. Sun, T. Gao, Z. Wang, J. Zhong, Electrochim. Acta 102 (2013) 246e251. [25] D. Deng, J.Y. Lee, Angew. Chem. 48 (2009) 1660e1663. [26] K. Fu, L. Xue, O. Yildiz, S. Li, H. Lee, Y. Li, G. Xu, L. Zhou, P.D. Bradford, X. Zhang, Nano Energy 2 (2013) 976e986. [27] J. Hwang, S.H. Woo, J. Shim, C. Jo, K.T. Lee, J. Lee, ACS Nano 7 (2013) 1036e1044. [28] Z. Zhu, F. Cheng, J. Chen, J. Mater. Chem. A 1 (2013) 9484e9490. [29] Y. Wu, S. Fang, Y. Jiang, J. Power Sources 75 (1998) 201e206.

Z. Shen et al. / Journal of Power Sources 278 (2015) 660e667 [30] C. Hu, Y. Xiao, Y. Zhao, N. Chen, Z. Zhang, M. Cao, L. Qu, Nanoscale 5 (2013) 2726e2733. [31] D. Li, Y. Xia, Nano Lett. 4 (2004) 933e938. [32] T. Yang, B. Lu, Phys. Chem. Chem. Phys. 16 (2014) 4115e4121. [33] F.M. Hassan, Z. Chen, A. Yu, Z. Chen, X. Xiao, Electrochim. Acta 87 (2013) 844e852. [34] T. Chen, L. Pan, X. Liu, K. Yu, Z. Sun, RSC Adv. 2 (2012) 11719e11724. [35] R. Demir-Cakan, Y.-S. Hu, M. Antonietti, J. Maier, M.-M. Titirici, Chem. Mater. 20 (2008) 1227e1229. [36] C. Wang, A. John Appleby, F.E. Little, J. Power Sources 93 (2001) 174e185. [37] R.A. Guidotti, P.J. Masset, J. Power Sources 183 (2008) 388e398. [38] Y. Wang, J.Y. Lee, H.C. Zeng, Chem. Mater. 17 (2005) 3899e3903.


[39] M.-S. Park, S.A. Needham, G.-X. Wang, Y.-M. Kang, J.-S. Park, S.-X. Dou, H.K. Liu, Chem. Mater. 19 (2007) 2406e2410. [40] L. Li, X. Yin, S. Liu, Y. Wang, L. Chen, T. Wang, Electrochem. Commun. 12 (2010) 1383e1386. [41] Y. Li, J.P. Tu, X.H. Huang, H.M. Wu, Y.F. Yuan, Electrochem. Commun. 9 (2007) 49e53. [42] H. Wu, G. Yu, L. Pan, N. Liu, M.T. McDowell, Z. Bao, Y. Cui, Nat. Commun. 4 (2013) 1943. [43] H. Li, X. Huang, L. Chen, J. Power Sources 81e82 (1999) 340e345. [44] J.H. Kong, Z.L. Liu, Z.C. Yang, H.R. Tan, S.X. Xiong, S.Y. Wong, X. Li, X.H. Lu, Nanoscale 4 (2012) 525e530.