MnOx nanosheets for improved electrochemical performances through bilayer nano-architecting

MnOx nanosheets for improved electrochemical performances through bilayer nano-architecting

Journal of Power Sources 286 (2015) 394e399 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/lo...

2MB Sizes 0 Downloads 33 Views

Journal of Power Sources 286 (2015) 394e399

Contents lists available at ScienceDirect

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

MnOx nanosheets for improved electrochemical performances through bilayer nano-architecting Yating Hu, John Wang* Department of Materials Science and Engineering, National University of Singapore, Singapore 117574, Singapore

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Bilayer MnOx nano-architecture are integrated onto Ni foam.  High mass loading of 2.4 mg cm1 is achieved.  Supercapacitor electrode shows specific capacitance of 559.5 F g1 and excellent cycle ability.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 February 2015 Received in revised form 24 March 2015 Accepted 30 March 2015 Available online 31 March 2015

A new bilayer nanostructure of manganese oxide is developed by two-step hydrothermal reactions. Much higher mass loading of manganese oxide is successfully maintained (~2.4 mg cm2) through this new type of nanoarchitecture. In the meanwhile, the desired high specific surface area, good conductivity and ion accessibility are obtained, therefore giving rise to significantly improved electrochemical performance for manganese oxide (e.g., specific capacitance of 559.5 F g1 and excellent cycle ability). By growing the second layer of MnOx on the Ni foam current collector that is pre-covered by a MnO2 substrate layer, uniformly thin and continuous nanosheets from the second layer MnOx with excellent surface coverage and open mesoporous structure are obtained, leading to large surface area and fast charge transfer. This novel architecting explores the new potential of manganese oxides as high performance supercapacitor electrodes. © 2015 Elsevier B.V. All rights reserved.

Keywords: Supercapacitor Bilayer MnOx High mass loading

1. Introduction Manganese oxide-based (MnOx) supercapacitor electrode materials are being extensively studied as they exhibit high theoretical capacitance, being environmental friendly and abundant [1,2]. However, most of the previously investigated systems were having difficulties to achieve a practically high specific capacitance, and

* Corresponding author. Department of Materials Science and Engineering, Block E3A, Engineering Drive 1, National University of Singapore, Singapore 117574, Singapore. E-mail address: [email protected] (J. Wang). http://dx.doi.org/10.1016/j.jpowsour.2015.03.177 0378-7753/© 2015 Elsevier B.V. All rights reserved.

the typical values reported in powder forms are from 100 to 300 F g1, which is only about 20% of the theoretical value [3,4]. The charge storage of MnOx is mainly by Faradic reactions occurring on an electrode surface with electrolytes, which involves the surface adsorption of electrolyte cations on MnOx. Due to the intrinsically low electric conductivity and the surface Faradic reaction, it is important to develop a structure with high surface area and to be highly electrolyte-accessible [5]. Various thin film based nanostructures such as hybrid core@shells, sandwiches and nanopillars have been synthesized recently, showing much improved specific capacitance [6e9]. In terms of synthesis methodology, approaches such as the template-based assembling, wet chemical

Y. Hu, J. Wang / Journal of Power Sources 286 (2015) 394e399

reaction, and electrochemical deposition have been employed in synthesizing MnOx nanostructures [10e12]. However, one of the limitations for 2D MnOx-based supercapacitor electrodes is the low mass loading, usually no more than 1 mg cm2. At increasing loading amount or film thickness, the gravimetric capacitance will decrease dramatically, while the increase in areal capacitance is also unsatisfactory [7,13]. The redox reaction between the III and IV oxidation states of Mn ions occurs spontaneously during the chargeedischarge process [14]. Thus both MnO2 and Mn3O4 are widely studied for electrochemical applications, as they are showing best electrochemical performances among the various stable manganese oxides (MnO, Mn2O3, Mn3O4 and MnO2) [15]. In the present work, we have developed a new bilayer structure for the MnOx growing on Ni foam, where high specific capacitance is obtained with much higher mass loading as compared to several previously reported structures [8,9,13]. Ni foam is an ideal current collector which has been used to grow MnO2 or other metal oxides based electrochemical materials from aqueous based solutions [16]. The second thin layer of continuous nanosheet could be obtained from a gel formation reaction followed by growth of MnOx nanocrystals during hydrothermal reaction. Due to the largely hydrophobic nature of the Ni foam, the MnOx gel is not properly attached to Ni foam surface during hydrothermal reaction. This is solved by growing the first layer MnO2 from KMnO4 solution prior to the second layer deposition, by which the wettability is much improved as compare to bare Ni foam (Table S1). The first layer therefore serves mainly as a substrate to grow the second layer and to provide the desired charge transfer to the current collector (Ni Foam). Thus, this layer needs to be uniform and possess good coverage and adhesion to the Ni foam. The second layer is responsible for most of the pesudocapacitance, as it provides a large surface area for the redox reactions to take place. The unique bilayer nanoarchitecture therefore delivers much improved conductivity and surface area, as well as high ion accessibility, which greatly improve the capacitance of MnOx nanosheet with high mass loading. 2. Experimental 2.1. Materials synthesis-step 1 Ni foam was immersed in a 1 M HCl solution for 1 h to clear the surface oxide layer. Cleaned and dried Ni foam was then immersed in 0.1 M KMnO4 solution and gone through 200  C hydrothermal reaction for 30 min, 2 h and 4 h respectively. The resulting samples were washed with DI water and dried in vacuum oven; the Ni form covered with the first layer of MnO2 is then obtained (Sample MnO2-1, MnO2-2, MnO2-3). Loading mass of the active material is calculated based on the mass gain of the dry Ni foam. 2.2. Materials synthesis-step 2 Pyrrole was added into 0.2 M KMnO4 drop wise. A gel was then formed. MnO2-covered Ni foam synthesized from step 1 was then immersed into this gel and undergone hydrothermal of 85  C for 12 h or 130  C for 6 h. The product grown on Ni foam was washed and dried for further characterizations. Powders were scratched off for transmission electron microscopy sample preparation. 2.3. Materials characterization X-ray diffraction (XRD) patterns were collected on a Bruker AXS X-ray powder diffractometer (D8 Advance, Cu Ka, l ¼ 0.154 nm). Xray photoelectron spectroscopy (XPS) studies were conducted with

395

a Kratos Axis Ultra DLD photoelectron spectrometer. Transmission electron microscopy (JOEL 2010F, 200 kV) and field-emission scanning electron microscopy (SUPRA 40 ZEISS, Germany) were used to examine the morphology of the samples. N2 adsorption at the temperature of 77 K with relative pressure (P/P0) ranging from 0.02 to 1 (Micromeritics ASAP2020) was employed to study the pore texture and specific surface area. The BrunauereEmmetteTeller (BET) method was utilized to calculate the specific surface area. Pore size distribution was calculated based on the BarretteJoynereHalenda (BJH) model for mesopores. Raman scattering spectra were recorded on a LABRAM-HR Raman spectrometer excited with 514.5 nm Arþ laser. 2.4. Electrochemical measurements The samples grown on Ni foam of 1.0 cm2 are directly tested as working electrode for electrochemical measurements in 0.5 M Na2SO4 electrolyte after pressing. A Pt plate and Ag/AgCl electrode were used as the counter electrode and the reference electrode, respectively. All potentials were referred to the reference electrode. Electrochemical performance was evaluated by cyclic voltammetry (CV), galvanostatic chargeedischarge and impedance spectroscopy by using Solartron System 1470E and 1400A, respectively. 3. Results and discussion The first layer of MnO2 nanonetwork is formed by hydrothermal reaction of directly immersing the cleaned Ni foam in KMnO4 solution. MnO-4 would be reduced and MnO2 could then be formed spontaneously [12]. MnO2 nanosheets with different thicknesses are obtained by controlling the hydrothermal reaction time. The second layer of MnOx is developed by a gel formation reaction of KMnO4 and Pyrrole, followed by the growth of MnOx nanocrystals on the previous MnO2 layer covering Ni foam through further hydrothermal reaction. As mentioned, this step of growth of the active material onto the current collector can only be successful when the Ni foam is already covered with a uniform thin layer of MnO2, thus the presence of the first layer is necessary. Through controlling the hydrothermal reaction temperature and time, either MnO2 or Mn3O4 could be harvested. The commonly used cationic surfactant hexadecyl trimethyl ammonium bromide (CTAB), which is known to assist in nanosheets growth, is also examined [6]. The mass loading ratio of the first to second layer is controlled at about 1:3. To obtain the optimal thickness of the first MnO2 layer, three samples with the first layer MnO2 grown only are compared, coded MnO2-1, MnO2-2 and MnO2-3, which are obtained by hydrothermal reaction for 30 min, 2 h and 4 h, respectively. The scanning electron microscopy (SEM) image in Fig. 1A shows good continuity and uniform coverage of the first layer, with Ni foam being exposed for contrast. When hydrothermal reaction time is too long (sample MnO2-3), the first layer becomes too thick, cracks are observed and the nanosheets' coverage is bad (Supplementary data, Fig. S1). The interface between the first layer MnO2 and Ni foam shows that the MnO2 networks bind closely to the Ni foam (Fig. 1B). Through X-ray photoelectron spectroscopy (XPS) study, it is confirmed that MnO2 are successfully grown on Ni foams (Fig. 1C). By conducting X-ray powder diffraction (XRD) and Raman spectra, it is found that this layer is largely amorphous (Fig. S2 and S3) [17]. The thicknesses of the first layer for the three samples are ~100 nm, 300 nm and 800 nm, respectively. The MnO2 mass loading in the three samples were measured to be 0.3 mg cm2, 0.6 mg cm2 and 1.3 mg cm2, respectively. It is found that a higher mass loading gives rise to a higher areal capacitance but a lower gravimetric capacitance (Fig. S4). In addition, the conductivity drops when this MnO2 layer became thicker (Fig. S5). By considering the combination of

396

Y. Hu, J. Wang / Journal of Power Sources 286 (2015) 394e399

Fig. 1. SEM images of: (A) MnO2 nanosheets grown on Ni Foam of sample MnO2-2 and (B) Side view of first layer MnO2 and Ni foam interface of sample MnO2-2; (C) XPS spectra of MnO2-2.

coverage, both areal and gravimetric capacitance and conductivity, sample MnO2-2 with specific capacitance of 267.9 F g1 and 0.19 F cm2 is chosen for developing the 2nd layer MnOx growth. The two layers are integrated onto Ni foam as illustrated by the scheme of Fig. 2A, which also shows that the growth directions of the two layers are quite different. In addition, the first layer is of network structure while the second layer is of continuous nanosheet structure. The sample derived from hydrothermal reaction at 85  C for 12 h is coded MnO2-MnOx-1. With CTAB being added during the gel formation reaction, the sample is coded MnO2MnOx-2. The sample derived from hydrothermal reaction at 130  C for 6 h, without CTAB, is coded MnO2-MnOx-3. All the samples were thermally annealed in N2 atmosphere at 350  C. This moderate annealing temperature is used, as when the crystallinity is too high in MnOx, the protonation or deprotonation reaction will be limited, as well as surface area loss will be experienced, though a high crystallinity can give rise to a better conductivity [1]. The preannealed sample is coded MnO2-MnOx-1-pre ANL. The XRD patterns of the three annealed samples (Fig. 2B and C) show that, either MnO2 or Mn3O4 can be obtained from different hydrothermal reaction conditions. For sample MnO2-MnOx-1, both a-MnO2 and Mn3O4 crystalline phases are observed. This could be due to that the MnO2 nanosheets formed in the first step is further reduced by pyrrole during the second step reaction, as the SEM image in Fig. 3C shows that the first layer is no longer in nanosheets but nanofiber morphology. Thus, Mn3O4 crystallites occurred in the first layer while a-MnO2 in the second layer (Fig. 2B). This also proves that the second layer of MnOx nanosheets or nanofiber binds to the first layer of MnO2 through chemical reactions during hydrothermal reactions. Point Raman spectroscopy done on the second layer further confirmed that it is a-MnO2, as there are two peaks at around 580 and 630 cm1 (Fig. S3) [18]. For sample MnO2MnOx-2 that with CTAB added in the gel formation step, only

amorphous MnO2 is formed based on the results of XRD (Fig. 2B) and Raman spectroscopy (Fig. S3). Sample MnO2-MnOx-3, which is obtained at higher hydrothermal temperature, shows only Mn3O4 peak. This could be explained as that the hydrated MnO2 loses the physisorbed water and protons and becomes Mn2O3 or Mn3O4 at higher temperature [17]. Raman spectra show consistent corresponding results (Fig. S3) [19]. As mentioned above, the main purpose to grow the second layer is to provide the desired surface area for the redox reaction. Fig. 2D shows the N2 adsorption isotherm of sample MnO2-MnOx-1, indicating a mesoporous structure [20]. The pore size distribution shows a well-defined peak at 8.19 nm. Table 1 summarizes the specific surface areas of various samples before and after thermal annealing at 350  C. For the single layer MnO2 grown on Ni foam (MnO2-2), the overall surface area is significantly low even before the thermal annealing. After the growth of the second layer, the surface area increased dramatically. A large portion is retained after thermal annealing for sample MnO2-MnOx-1. Sample MnO2-MnOx2 and MnO2-MnOx-3 show lower surface areas, which are related to their morphologies. For sample MnO2-MnOx-2, much smaller nanosheets are observed (Fig. 3B). The small flowers appear to stack together with a large degree of agglomeration, preventing the interior to be exposed. Thus the measured surface area is much lower compared to that of sample MnO2-MnOx-1, which is of open structure nanosheets, as shown in Fig. 3A and C. The average thickness of the second layer of sample MnO2-MnOx-1 is ~7 mm as shown in the Fig. 3A's insertion. The high magnification SEM image (Fig. S6A) shows that the nanosheet in the second layer is grown on top of the fibres from the first layer. In theory, this structure type should benefit the ion transfer as it is interconnected and highly electrolyte-accessible [21,22]. Fig. 3E shows the transmission electron microscopy (TEM) image of a single piece of the nanosheet

Y. Hu, J. Wang / Journal of Power Sources 286 (2015) 394e399

397

Fig. 2. (A) Illustration of the two layer Ni foam-MnO2-MnOx nanoarchitecture; XRD patterns of: (B) sample MnO2-MnOx-1 and MnO2-MnOx-2, and (C) sample MnO2-MnOx-3; (D) N2 adsorption isotherm of sample MnO2-MnOx-1; (E) Pore size distribution of sample MnO2-MnOx-1.

which is in good continuity. At higher hydrothermal reaction temperature used (sample MnO2-MnOx-3), most of the nanosheets had disappeared. Instead, entangled nanowires present (Fig. 3D and F), resulting in the surface area being much lower than that of sample MnO2-MnOx-1. As mentioned above, the surface area was improved significantly after the second layer growth. The electrical conductivity also plays an important role in affecting the capacitance of MnOx materials. Fig. 4A shows the nyquist plots for sample MnO2-2, MnO2-MnOx-1, MnO2-MnOx-2 and MnO2-MnOx-3, measured in 0.5 M Na2SO4 electrolyte. The internal resistance of each sample, which could be represented by the interception of the plot with the real axis (Z0 ) [23], appears to be similar for all samples, except for sample MnO2-2 that shows a slightly lower internal resistance than others. Due to the large void size of Ni foam (~200 mm), MnOx grows on the Ni foam scaffold without clogging the voids. Thus, the uniformly covered Ni foam retains its high electrolyte accessibility (Fig. S6B). As mentioned above, the first layer of MnO2 networks binds closely to the Ni foam, thus sample MnO2-2 shows similar internal resistance with bare Ni foam (Fig. S7). The shape of the nyquist curve is related to the ratio between internal resistance and charge transfer resistance, by adopting the continuum-mechanics-based analysis [24]. When the charge transfer resistance is much higher than the internal resistance, the

nyquist plot does not come back toward the real axis when frequency becomes lower, as in the plot of sample MnO2-2, and MnO2MnOx-3. This indicates that the electrode behaves as a purely capacitive system of significant charge transfer resistance [25]. When the charge transfer resistance value is close to the internal resistance, the curve has a higher tendency of coming back towards the real axis, indicating it is more conductive as a capacitor, as observed in the plot of sample MnO2-MnOx-1 and MnO2-MnOx-2 [24,26]. In addition, the slope of the nyquist plot at lower frequency for sample MnO2-MnOx-1 is steeper, indicating it has better accessibility and lower ion transfer resistance for the electrolyte [27]. In short, sample MnO2-2 shows a slightly lower internal resistance while sample MnO2-MnOx-1 shows a much lower charge transfer resistance, most probably owing to the structure of thin, uniform and continuous sheets as shown in Fig. 3A, C and E. Electrochemical tests are performed on these samples. The specific capacitance values of sample MnO2-2, MnO2-MnOx-1, MnO2-MnOx-2 and MnO2-MnOx-3 under scan rate of 1 mV s1 are: 267.9, 559.6, 307.1, and 386.6 F g1, respectively. The mass loading of the second layer is measured to be ~2 mg cm1 for all samples. Comparing the specific capacitance at various scan rates (Fig. 4B), it is obvious that by growing the second layer of MnO2, the specific capacitance improved significantly. Together with the high surface area and uniform pores, sample MnO2-MnOx-1 gives rise to a very

398

Y. Hu, J. Wang / Journal of Power Sources 286 (2015) 394e399

Fig. 3. SEM images of: (A) MnO2-MnOx-1 (insertion: side view and thickness measurement of second layer MnOx), (B) MnO2-MnOx-2, (C) the first and second layer interface of MnO2-MnOx-1, (D) MnO2-MnOx-3; TEM of: (E) MnO2-MnOx-1 powder, and (F) MnO2-MnOx-3 powder.

Table 1 Specific surface areas for samples before and after thermal annealing at 350  C. Sample

Single layer MnO2-2 Bilayer MnO2-MnOx-1 Bilayer MnO2-MnOx-2 Bilayer MnO2-MnOx-3

Surface area(m2 g1) Pre-annealing

Post-annealing

31.7 241.0 170.7 163.2

e 207.0 52.5 77.9

promising specific capacitance value of 559.6 F g1. The CV curve in Fig. 4C for sample MnO2-MnOx-1 retains symmetrical nearrectangular shape and indicates that there is no obvious resistive behaviour [7]. The redox peaks at around 0.4 V for sample MnO2MnOx-1 might be due to the side reaction of insertion and deinsertion of Naþ and Hþ from the electrolyte in the tunnel sites of MnO2 or Mn3O4, which only take place at low scan rates [28,29]. The retain rates of specific capacitance from scan rate of 1 mV s1 to 25 mV s1 are calculated and listed in Table S2, Supporting information. Development of the second layer of MnO2 does not bring down the retain rate (about 60% of the capacitance from 1 mV s1 scan rate retained when using 25 mV s1 scan rate), except for sample MnO2-MnOx-2. Capacitance is also calculated from chargeedischarge tests at 2A g1 current density (Table S2), where sample MnO2-MnOx-1 shows the highest specific capacitance of 473.9 F g1. Charge-discharge curve of sample MnO2-2 and MnO2MnOx-1 are shown in Fig. 4D. The very symmetric charge and

discharge curves indicate that the electrode is of good electrochemical capacitive characteristics and capable for superior reversible redox reactions [8]. Using the same reduction method of KMnO4 by pyrrole, the MnO2 powders obtained could not form nanosheets structure. Instead, nanorods are formed, with specific capacitance ~200 F g1 under 1 mV s1 scan rate (SEM image of such nanorods shown in Fig. S8). In addition, due to being in the powder form, the prepared electrode has much higher internal resistance (Fig. S7). As for sample MnO2-MnOx-2, both surface area and conductivity are not as good as those of sample MnO2-MnOx-1 (poorer conductivity may well be due to the low degree of crystallinity as mentioned earlier), thus the specific capacitance is not as high. Therefore, it is concluded that the smaller sizes of nanosheets does not improve electrochemical performances in this case. Sample MnO2-MnOx-3 is also showing a lower specific capacitance, due to the relatively low surface area posed by the agglomerated Mn3O4 nanofibers and poorer conductivity. For sample MnO2-MnOx-1-pre ANL, although it has a slightly higher surface area compare to the annealed sample (MnO2-MnOx-1), the charge transfer resistance is much higher, thus the specific capacitance is lower (Fig. S9A and S9B). The cycle ability was evaluated under 2 A g1 current density of chargeedischarge for samples MnO2-2 and MnO2-MnOx-1 (Fig. S10). Significant cyclic increment during first 2000 cycles for sample MnO2-MnOx-1 is observed, which is probably due to the slow dissolution and re-deposition of the MnO2 during the redox

Y. Hu, J. Wang / Journal of Power Sources 286 (2015) 394e399

399

Fig. 4. (A) Nyquist plot of various samples; (B) Comparison of specific capacitance of various samples at different scan rates; (C) CV curves at 5 mV s1 of sample MnO2-2 and MnO2MnOx-1; (D) Charge-discharge curve at 2A g1 current density of sample MnO2-2 and MnO2-MnOx-1.

reaction in the electrolyte [7]. Both samples could retain 98% of its initial capacitance after 5000 cycles which could be considered as great improvement [2,17], and starts to drop after that. MnO2-2 could retain 65% and sample MnO2-MnOx-1 could retain only 58% after 7500 cycles, which might be due to that the cycled MnO2 sheets have become thicker from the re-deposition in 0.5 M Na2SO4 electrolyte. 4. Conclusions In summary, we have devised and successfully developed a new bilayer structure for manganese oxides to explore their full electrochemical performances. By growing the second layer of MnO2 on the Ni foam that is pre-covered by MnO2 substrate layer, uniformly thin and continuous nanosheets with excellent coverage and open structure are successfully obtained. The bilayer nanosheets architecture, which exhibits a high mass loading of manganese oxides (~2.4 mg cm1), possesses high surface area for the redox reactions and allows for fast charge transfer, delivers much improved electrochemical performances. Such bilayer nanoarchitecture greatly extends the practical application potential of MnOx based materials as supercapacitor electrode. Acknowledgement This work is supported by MOE, Singapore Ministry of Education (Tier 2, MOE2012-T2-2-102), conducted at the National University of Singapore (Tier 2, MOE2012-T2-2-102). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2015.03.177. References [1] G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 41 (2012) 797e828.

[2] K. Zhang, X. Han, Z. Hu, X. Zhang, Z. Tao, J. Chen, Chem. Soc. Rev. 44 (2015) 699e728. [3] M.-J. Deng, P.-J. Ho, C.-Z. Song, S.-A. Chen, J.-F. Lee, J.-M. Chen, K.-T. Lu, Energy & Environ. Sci. 6 (2013) 2178e2185. [4] G. Yu, L. Hu, M. Vosgueritchian, H. Wang, X. Xie, J.R. McDonough, X. Cui, Y. Cui, Z. Bao, Nano Lett. 11 (2011) 2905e2911. langer, Chem. Mater. 16 (2004) 3184e3190. [5] M. Toupin, T. Brousse, D. Be [6] H. Chen, L. Hu, Y. Yan, R. Che, M. Chen, L. Wu, Adv. Energy Mater. 3 (2013) 1636e1646. [7] J. Kang, A. Hirata, L. Kang, X. Zhang, Y. Hou, L. Chen, C. Li, T. Fujita, K. Akagi, M. Chen, Angew. Chem. Int. Ed. 52 (2013) 1664e1667. [8] Q. Li, Z.-L. Wang, G.-R. Li, R. Guo, L.-X. Ding, Y.-X. Tong, Nano Lett. 12 (2012) 3803e3807. [9] Z. Yu, B. Duong, D. Abbitt, J. Thomas, Adv. Mater. 25 (2013) 3302e3306. [10] J. Liu, J. Jiang, C. Cheng, H. Li, J. Zhang, H. Gong, H.J. Fan, Adv. Mater. 23 (2011) 2076e2081. [11] W. Li, G. Li, J. Sun, R. Zou, K. Xu, Y. Sun, Z. Chen, J. Yang, J. Hu, Nanoscale 5 (2013) 2901e2908. [12] X. Xia, Y. Zhang, D. Chao, C. Guan, Y. Zhang, L. Li, X. Ge, I.M. Bacho, J. Tu, H.J. Fan, Nanoscale 6 (2014) 5008e5048. [13] J. Kang, L. Chen, Y. Hou, C. Li, T. Fujita, X. Lang, A. Hirata, M. Chen, Adv. Energy Mater. 3 (2013) 857e863. [14] Q. Jiangying, G. Feng, Z. Quan, W. Zhiyu, H. Han, L. Beibei, W. Wubo, W. Xuzhen, Q. Jieshan, Nanoscale 5 (2013) 2999e3005. [15] W. Wei, X. Cui, W. Chen, D.G. Ivey, Chem. Soc. Rev. 40 (2011) 1697e1721. [16] L. Yu, G. Zhang, C. Yuan, X.W. Lou, Chem. Commun. 49 (2013) 137e139. [17] C. Xu, S. Shi, Y. Sun, Y. Chen, F. Kang, Chem. Commun. 49 (2013) 7331e7333. [18] S. Cheng, L. Yang, D. Chen, X. Ji, Z.-J. Jiang, D. Ding, M. Liu, Nano Energy 9 (2014) 161e167. [19] L. Malavasi, P. Galinetto, M.C. Mozzati, C.B. Azzoni, G. Flor, Phys. Chem. Chem. Phys. 4 (2002) 3876e3880. [20] M. Thommes, Chem. Ing. Tech. 82 (2010) 1059e1073. [21] K.A. Stoerzinger, M. Risch, J. Suntivich, W.M. Lu, J. Zhou, M.D. Biegalski, H.M. Christen, Ariando T. Venkatesan, Y. Shao-Horn, Energy & Environ. Sci. 6 (2013) 1582e1588. langer, T. Brousse, J.W. Long, Electrochem. Soc. Interface 17 (2008). [22] D. Be [23] M.D. Stoller, S. Park, Y. Zhu, J. An, R.S. Ruoff, Nano Lett. 8 (2008) 3498e3502. [24] P.M. Gomadam, J.W. Weidner, Int. J. Energy Res. 29 (2005) 1133e1151. [25] Z. Niu, W. Zhou, J. Chen, G. Feng, H. Li, W. Ma, J. Li, H. Dong, Y. Ren, D. Zhao, S. Xie, Energy & Environ. Sci. 4 (2011) 1440e1446. [26] R.A. Huggins, Advanced Batteries Materials Science Aspects, Springer, Stanford, 2009. [27] H. Yoo, M. Min, S. Bak, Y. Yoon, H. Lee, J. Mater. Chem. A 2 (2014) 6663e6668. [28] G. Qiu, H. Huang, S. Dharmarathna, E. Benbow, L. Stafford, S.L. Suib, Chem. Mater. 23 (2011) 3892e3901. [29] Y.H. Lin, T.Y. Wei, H.C. Chien, S.Y. Lu, Adv. Energy Mater. 1 (2011) 901e907.