Home
Search
Collections
Journals
About
Contact us
My IOPscience
Carbon nanotube network film directly grown on carbon cloth for high-performance solid-state flexible supercapacitors
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 035402 (http://iopscience.iop.org/0957-4484/25/3/035402) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 131.91.169.193 This content was downloaded on 11/11/2014 at 23:14
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 035402 (8pp)
doi:10.1088/0957-4484/25/3/035402
Carbon nanotube network film directly grown on carbon cloth for high-performance solid-state flexible supercapacitors Cheng Zhou and Jinping Liu Institute of Nanoscience and Nanotechnology, Department of Physics, Central China Normal University, Wuhan, Hubei 430079, People’s Republic of China E-mail: [email protected] Received 9 October 2013, revised 14 November 2013 Accepted for publication 25 November 2013 Published 20 December 2013 Abstract
Carbon nanotubes (CNTs) have received increasing attention as electrode materials for high-performance supercapacitors. We herein present a straightforward method to synthesize CNT films directly on carbon cloths as electrodes for all-solid-state flexible supercapacitors (AFSCs). The as-made highly conductive electrodes possess a three-dimensional (3D) network architecture for fast ion diffusion and good flexibility, leading to an AFSC with a specific capacitance of 106.1 F g−1 , an areal capacitance of 38.75 mF cm−2 , an ultralong cycle life of 100 000 times (capacitance retention: 99%), a good rate capability (can scan at 1000 mV s−1 , at which the capacitance is still ∼37.8% of that at 5 mV s−1 ), a high energy density (2.4 µW h cm−2 ) and a high power density (19 mW cm−2 ). Moreover, our AFSC maintains excellent electrochemical attributes even with serious shape deformation (bending, folding, etc), high mechanical pressure (63 kPa) and a wide temperature window (up to 100 ◦ C). After charging for only 5 s, three such AFSC devices connected in series can efficiently power a red round LED for 60 s. Our work could pave the way for the design of practical AFSCs, which are expected to be used for various flexible portable/wearable electronic devices in the future. Keywords: carbon nanotube, solid-state supercapacitors, flexible electronics, energy storage, charge storage (Some figures may appear in colour only in the online journal)
1. Introduction
lightweight, flexible, and of high power and energy densities are also promising to meet the various requirements of modern electronic products [3]. As one kind of SCs, electric double-layer capacitors (EDLCs) store energy in the form of charge separation between the double layer formed at the interface of the solid electrode surface and the liquid or solid electrolyte. Carbon materials which have high specific surface areas and exhibit a high diversity in crystallinity, morphology, porosity, and texture, are currently the most widely used materials for EDLCs. For this, several carbon materials such as activated carbons [4–6], carbon nanotubes [7–11] (multi- and single-walled CNTs),
Climate change and the decreasing availability of fossil fuels require society to move towards sustainable and renewable resources. At the same time, emerging energy storage devices are urgently expected to maintain the stability of the power supply from these energy sources. Electrochemical capacitors, also known as supercapacitors (SCs), are one kind of such energy devices. They have received increasing attention for applications in (hybrid) electric vehicles due to their fast charge–discharge characteristic, high power density and long cycle life [1–3]. SCs which are cheap, 0957-4484/14/035402+08$33.00
1
c 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
2. Experimental section
spherical carbon nanoparticles [12], and one- to few-layered graphene [13] have been explored. Commercial EDLCs using activated carbon as electrode materials have a much higher theoretical capacitance (100–300 F g−1 ) compared to conventional electrolytic capacitors [4] (on the order of micro- or picofarads). However, activated carbon electrodes are not ideal for multifunctional energy storage devices such as flexible EDLCs due to the fragile nature of particulate film. Carbon nanotubes (CNT) are an important family of carbonaceous materials, which has been attracting great attention in recent years [14–16]. Owing to the highly accessible surface area, superior electrical conductivity, electrochemical stability and mechanical flexibility, CNT is promising as an advanced electrode material for EDLCs. Preparation methods of CNT electrodes in the literature were, however, generally indirect, involving several procedures. The as-grown nanotubes were firstly harvested in the form of power or from the substrate and re-dispersed in liquids. These suspended nanotubes were then re-assembled on current collectors to form the porous electrodes via slurry casting [6], ink-jet printing [17], vacuum filtering [18], or electrophoretic deposition [19], etc. Since the CNTs were densely packaged in these electrodes, these approaches would decrease the available surface area of CNTs to some extent, which is very harmful to the SC device’s properties. Herein, we have constructed our CNT electrodes using a more facile approach, which involves the direct growth of CNT films on the current collector through a chemical vapor deposition (CVD) process using nickel as the catalyst. The as-grown CNTs are slightly entangled, forming a well-developed network with tremendous open pores. This porous three-dimensional (3D) architecture allows the rapid transport of ions from the electrolyte to the entire surface of the CNT electrode and makes it quickly available for electric double-layer formation. Flexible and lightweight SCs are promising nextgeneration energy storage devices. Recently, some studies have been carried out to integrate nanoscale materials with current collectors to prepare flexible, thin, lightweight, and wearable power conversion and storage devices [3, 9, 18, 20–27]. An easy-to-assemble integrated nanocomposite energy storage system could provide design ingenuity for a variety of devices operating under a hard environment condition, for example, high mechanical stress and a wide range of temperature [7]. In this work, we propose combining directly grown CNTs and flexible commercial carbon cloth (CC) in an integrated composite structure as electrodes and using H3 PO4 /poly(vinyl alcohol) (PVA) polymer gel as the electrolyte/separator to assemble an all-solid-state flexible supercapacitor (AFSC). As a result of the unique 3D CNT electrode architecture, the AFSC exhibits excellent properties such as ultralong lifetime and high rate capability. These attributes can still be maintained when high mechanical pressure (63 kPa) and high temperature (up to 100 ◦ C) are applied, demonstrating great promise for future flexible electronics. This work could open up new opportunities in the design and fabrication of high-performance AFSCs.
2.1. Preparation of CNTs/CC hybrid electrode
The CNT network film was synthesized directly on CC by a nickel-catalyzed CVD process. In detail, nickel nitrate hexahydrate was firstly dissolved in 50 ml absolute alcohol and ethylene glycol (1:1) under stirring to form a uniform mixture. Prior to the fabrication of CNTs, a piece of CC (CeTech, through-plane electrical conductivity: ∼2 × 102 S cm−2 , 2.0 × 4.0 cm2 ) was immersed into the above solution for 1 h to adsorb the nickel salt. The treated CC was then put at the center of a tube furnace and heated at a rate of 30 ◦ C min−1 to 800 ◦ C for 20 min with a mixed solution (ethanol and ethylene glycol with the volume ratio of 1:5) placed at the tube entrance as the carbon source under flowing argon atmosphere. The adsorbed nickel salt will be decomposed at high temperature and will subsequently be reduced to nickel catalyst by the vapor of ethanol and ethylene glycol, initiating the CNT growth [28]. 2.2. Characterization
Samples were characterized by scanning electron microscopy (SEM, JSM-6700F, 5 kV), and transmission electron microscopy (TEM, JEM-2010FEF, 200 kV), and Raman spectroscopy was carried out on a Witech CRM200 (532 nm). The Brunauer–Emmett–Teller (BET) specific surface area was measured on a Bel Sorp-mini (S/N-00230) analyzer. The mass of electrode materials was measured on an AX/MX/UMX balance (METTLER TOLEDO, maximum = 5.1 g; d = 0.001 mg). 2.3. Fabrication of AFSC
Firstly, the gel electrolyte was prepared: 12 g H3 PO4 was added into 60 ml deionized water and then 6 g PVA power was added. The whole mixture was heated to 85 ◦ C under stirring until the solution became clear. The AFSC device was assembled as follows: the polymer gelled electrolyte was slowly poured onto the CNTs/CC electrodes (100 µL electrolyte/1 cm2 of the electrode). This assembly was left under ambient conditions for 5 h to ensure that the electrolyte completely wet the electrode and to allow for evaporation of most excess water. Two electrodes were then assembled face-to-face and left overnight until the electrolyte solidified. This results in mechanically robust devices with the polymer electrolyte acting as both the electrolyte and the ion-porous separator. The solidified polymer electrolyte simplifies the device architecture and helps maintain the electrodes in closer proximity. 2.4. Electrochemical performance measurement
All the electrochemical performance was tested on a CS310 Electrochemical Workstation. The CNT electrode’s property was investigated in a three-electrode electrochemical cell, using a 1 M Li2 SO4 or H3 PO4 aqueous solution as the 2
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 1. (a) SEM image of the pure CC. The inset shows the optical image. (b) Optical image of CNTs/CC electrode. (c) and (d) SEM
images of CNTs/CC electrode at different magnifications.
figure 1(a). Figure 1(b) shows an optical image of a piece of CNTs/CC electrode. The electrode is still mechanically flexible even under folding conditions after the CNT growth. The representative SEM images of CNTs/CC electrode are shown in figures 1(c) and (d). It can be observed that entangled CNTs are uniformly grown onto the surface of CC microfibers over almost the entire porous CC, exhibiting a 3D network morphology. BET measurement reveals that the specific surface area of CNTs/CC electrode is ∼298 m2 g−1 , much higher than that of pure CC (∼27 m2 g−1 ). The TEM image in figure 2(a) clearly demonstrates the CNTs are multi-walled with ca. 0.34 nm interplanar spacing. Nickel nanoparticles are also detected in the CNT structure, confirming the vapor–liquid–solid (VLS) process [16]. The Raman spectrum of CNTs is presented in figure 2(b). The 1340 cm−1 peak (D band) corresponds to the disorder-induced feature due to the lattice distortion, while the 1570 cm−1 peak (G band) corresponds to the in-plane stretching vibration mode E2g of single crystal graphite. To investigate the charge storage behavior, CV curves of both the pristine CC and CNTs/CC electrodes were compared and the results are shown in figures 3(a)–(c). The CC exhibits a stable electrochemical performance at the potential window ranging from 0 to 0.8 V versus Ag/AgCl in 1 M Li2 SO4 aqueous solution. However, the current density is small, revealing the small capacitance of pristine CC. After the CNT growth, the current in the CVs increases dramatically (figures 3(b) and (c)) for the CNTs/CC electrode tested in both 1 M Li2 SO4 and H3 PO4 electrolytes. The rectangular shape of the CV curves at different scan rates (even up to 200 mV s−1 )
electrolyte to demonstrate the EDL capacitance. A Pt plate and Ag/AgCl were used as the counter-electrode and the reference electrode, respectively. The AFSC device’s performance was tested in a two-electrode mode. The specific capacitance (F g−1 ) was calculated based on the total mass of the CNT materials. Areal and specific capacitances were calculated based on cyclic voltammetry (CV) using equations R R I dV I dV Ca = 21 × 1V·υ·S and Csp = 12 × 1V·υ·m , respectively, where I is the current, 1V is the potential window, υ is the scan rate, m is the mass of the electrode material, and S is the geometrical area of the electrode. Electrochemical impedance spectroscopy (EIS) measurements were performed by applying an AC voltage with 5 mV amplitude in a frequency range from 0.01 Hz to 100 kHz at open circuit potential. To investigate the temperature effect, the device was placed in a small oven with precise temperature control and display. Two Pt wires were used to connect the device with the outer Electrochemical Station. To investigate the pressure effect, a 1 cm2 AFSC device was placed on a flat glass plate, and then regular iron blocks with different definite masses were added directly to the upper surface of the device one by one to generate the mechanical pressure. 3. Results and discussion
Figure 1 shows the structural characterization of the CNTs/CC hybrid electrode. The SEM image in figure 1(a) reveals that pristine CC consists of interwoven carbon fibers with diameters ranging from 10 to 12 µm. A photograph of the commercial CC is further displayed in the inset of 3
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 2. (a) TEM image of CNTs, showing the presence of nickel. (b) Raman spectrum of CNT film.
Figure 3. (a) CV curves of the pure CC electrode at different scan rates. (b) and (c) CV curves of the CNTs/CC hybrid electrode with 1 M
Li2 SO4 and 1 M H3 PO4 as aqueous electrolyte respectively at different scan rates. (d) Rate capability of the CC and CNTs/CC electrodes.
indicates the ideal electric double-layer capacitive behavior from the 3D CNT network film. The areal capacitances calculated from the CVs at different scan rates for pristine CC and CNTs/CC electrodes (in H3 PO4 electrolyte) are also shown in figure 3(d). It is noted that while both the electrodes demonstrate good rate capability, there is almost 100 times capacitance increase for the CNTs/CC electrode as compared to pure CC. Since the CNT network film on CC displayed excellent electrochemical properties, a two-electrode AFSC device was further constructed. Figure 4(a) shows the structure of the device, in which two pieces of CNTs/CC are
used respectively as the positive and negative electrodes with H3 PO4 /PVA gel as the electrolyte/separator. The EIS spectrum in figure 4(b) exhibits a small semicircle at high frequency and a long straight line at low frequency. The straight line with small angle to the imaginary axis reveals the good capacitive behavior of our device. From the extended spectrum in the inset, the equivalent series resistance (the high frequency intercept of the semicircle on the real axis) and charge-transfer resistance (diameter of the semicircle) can be determined as ∼9 and 3 , respectively. The small values of these resistances indicate the facile electron and ion transport/diffusion in the device. 4
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 4. (a) Schematic illustration of the structure of CNTs/CC-based AFSC. (b) The EIS of the entire device. (c) CV and (d) rate
capability and the related charge–discharge curves of the device. (e) Ragone plot of the device. (f) Cycle stability of the device up to 100 000 times.
Both CV and galvanostatic charge/discharge tests were used to evaluate the capacitive performance of the entire device. Figure 4(c) shows the CV curves of the AFSC measured at different scan rates of 20, 50, 100 and 200 mV s−1 with a potential window ranging from 0 to 0.8 V. These CV curves show ideal capacitive behavior with nearly rectangular shape and retain the rectangular shape without apparent distortion with increasing scan rate up to 200 mV s−1 , indicating stable and high electrochemical performance of the AFSC. At 50 mV s−1 , the device’s areal capacitance is determined to be 38.75 mF cm−2 (corresponding specific capacitance: 106.1 F g−1 ), which is much higher than that of conventional carbon-based SCs (µF cm−2 ), and even larger than those of state-of-the-art graphene SCs (∼4 mF cm−2 ) [3] and some SCs based on pseudocapacitive materials such as NiO–TiO2 nanotube arrays [29], Fe3 O4 –SnO2 nanorod films [30] and Ni–NiO core–shell inverse opals [31]. Further
investigation reveals that the device still achieves 35.5% capacitance retention when the scan rate increases 1000 times from 5 to 5000 mV s−1 , indicating a good rate capability. Figure 4(d) illustrates the rate capability of capacitance versus current density of the AFSC and the related galvanostatic charge/discharge curves at the current density of 0.3, 0.5, 1.0, 2.0 and 3.0 mA cm−2 , respectively. All the curves exhibit a straight-line shape and have a very symmetric nature, indicating a rapid current–voltage response and good electrochemical reversibility. Energy density and power density are two key factors for evaluating the practical applications of SCs. A good SC is expected to provide both high energy and power densities [1]. Figure 4(e) shows the Ragone plot for our CNTs/CC-based AFSC device, obtained from the CV data. The energy density decreases from 2.4 to 0.84 µW h cm−2 , while the power density increases from 0.05 up to 19 mW cm−2 as the 5
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 5. (a) CVs of the device at different temperatures at 50 mV s−1 . (b) The capacitance value variation of the device upon the gradual increase and then decrease of the temperature.
scan rate increases from 20 to 5000 mV s−1 . The overall performance of our device is comparable to that of traditional CNT-based SCs established with liquid electrolyte [16, 32]. Another important requirement for SC application is cycling capability. To evaluate this, the charge–discharge cycling test over 100 000 times for the device at a current density of 1 mA cm−2 was carried out. Figure 4(f) shows the capacitance retention of the AFSC device as a function of cycling number. It is interesting that there is only 1% loss in the capacitance after 100 000 continuous cycles, meeting the standard for practical application. The electrochemical performance of our AFSC device was also systematically studied under different operating temperatures, ranging from 25 to 100 ◦ C. As shown in figure 5(a), after heating to 100 ◦ C, the CV curve of the AFSC device shows a slight change in both the shape and integrated area. The rectangular shape is obviously deformed, which should be due to the interfacial variation between the electrodes and electrolyte upon heating. The slight increase of the capacitance should be due to the fact that the ion diffusion in the gel electrolyte and thus the ion accumulation on the surface of CNTs can be promoted with the temperature increase. Nevertheless, with the temperature reduced back to normal, the CV can almost recover. Figure 5(b) shows the capacitance value variation of the device upon the gradual increase and then decrease of the temperature. It is obvious that with even great changes in the working temperature, the capacitance of our AFSC device has no significant fluctuation. This result implies that the device can be operated at a wide range of temperatures. In order to evaluate the potential of our AFSC for flexible energy storage under real conditions, the device was placed under different shape deformations and its performance was analyzed. The CV performance of this device when tested under different bending conditions (90◦ and 180◦ ) is shown in figure 6(a). From the curves, it can be seen that the bending has almost no effect on the capacitive behavior. Moreover, the stability of the device was tested for more than 500 cycles in the bending state, with about 10% increase in the device capacitance, as shown in figure 6(b). The increase
of the capacitance upon serious bending may be attributed to the enhanced interfacial contact between the 3D CNT film and the solid-state electrolyte, ensuring more charge accumulation. We also measured the performance of our AFSC under different mechanical pressures (figure 6(c)). As can be seen, the capacitance of the device does not show obvious differences when the pressure is higher than 10 kPa. In addition, the device can be operated even when a pressure of 63 kPa was applied. This performance durability can be attributed to the high mechanical flexibility of the electrodes along with the interpenetrating 3D network structure between the CNTs/CC films and the gelled electrolyte. The electrolyte solidifies during the device assembly and acts like a glue that holds all the device components together, improving the mechanical integrity and increasing its cycle life even when tested under extreme bending and pressure conditions. As this remarkable performance has yet to be realized in commercial devices, our AFSC may be ideal for next-generation flexible, portable electronics. To demonstrate the potential use of our AFSC device, we connected SC units in series to drive light-emitting-diodes (LED). Each AFSC unit has the same electrode area of 1 cm2 and an output voltage of ∼0.8 V. We assembled three AFSCs in series, and after charging for only 5 s to ∼2.4 V, the integrated device could power a 3 mm-diameter red round LED indicator (1.8 V, 20 mA) for more than 60 s efficiently, as shown in figure 6(d). It is expected that the performance of the AFSC can be further improved by using an ionic-liquid-based gel electrolyte with a wider operation potential [3], increasing CNT mass loading, and combination with other pseudocapacitive materials. 4. Conclusions
In summary, we have successfully developed a straightforward method to prepare a CNT network film directly on the CC current collector and constructed an AFSC device. The AFSC device delivers high energy and power densities and exhibits a good rate capability and an outstanding cycling life exceeding 100 000 times. Furthermore, the electrochemical 6
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 6. (a) CV curves of the device under different bending conditions. Inset is the optical images showing the bending states. (b) Cycle
stability of device recorded under normal and bending conditions. (c) Plot of capacitance versus mechanical pressure. (d) Optical image showing that the red LED is powered by three tandem AFSC devices.
attributes of our device can be maintained even with serious shape deformation (bending, folding, etc), high mechanical pressure and wide temperature window. Our work presents an opportunity to establish an AFSC device that can be operated under harsh environmental conditions.
[3] El-Kady M F, Strong V, Dubin S and Kaner R B 2012 Science 335 1326 Meng C Z, Liu C H, Chen L Z, Hu C H and Fan S S 2010 Nano Lett. 10 4025 Horng Y Y, Lu Y C, Hsu Y K, Chen C C, Chen L C and Chen K H 2010 J. Power Sources 195 4418 [4] Zhang L L and Zhao X S 2009 Chem. Soc. Rev. 38 2520 [5] Kaempgen M, Chan C K, Ma J, Cui Y and Gruner G 2009 Nano Lett. 9 1872 [6] Gao Q, Demarconnay L, Raymundo-Pi˜nero E and B´eguin F 2012 Energy Environ. Sci. 5 9611 [7] Hu L, Choi J W, Yang Y, Jeong S, Mantia F La, Cui L F and Cui Y 2009 Proc. Natl Acad. Sci. 106 21490 Kim Y S, Kumar K, Fisher F T and Yang E H 2012 Nanotechnology 23 015301 [8] Li X, Rong J and Wei B 2010 ACS Nano 4 6039 [9] Li G R, Xu H, Lu X F, Feng J X, Tong Y X and Su C Y 2013 Nanoscale 5 4056 Do Q H, Fielitz T R, Zeng C C, Vanli O A, Zhang C and Zheng J P 2013 Nanotechnology 24 315401 [10] Fan Z J, Yan J, Zhi L J, Zhang Q, Wei T, Feng J, Zhang M, Qian W and Wei F 2010 Adv. Mater. 22 3723 [11] Niu Z Q et al 2011 Energy Environ. Sci. 4 1440 [12] Pech D, Brunet M, Durou H, Huang P H, Mochalin V, Gogotsi Y, Taberna P L and Simon P 2010 Nature Nanotechnol. 5 651 [13] Zhu Y W et al 2011 Science 332 1537 [14] NoKed M, Okashy S, Zimrin T and Aurbach D 2012 Angew. Chem. Int. Edn 51 1568 Basirico L and Lanzara G 2012 Nanotechnology 23 305401
Acknowledgments
Liu Jinping thanks the National Natural Science Foundation of China (Nos 51102105, 11104088), Science Fund for Distinguished Young Scholars of Hubei Province (No. 2013CFA023), Self-determined Research Funds of CCNU from the Colleges’ Basic Research and Operation of MOE (CCNU12A01009) and National Innovation Experiment Program for University Students (J1210068). References [1] Miller J R and Simon P 2008 Science 321 651 Mai L Q, Yang F, Zhao Y L, Xu X, Xu L and Luo Y Z 2011 Nature Commun. 2 381 [2] Simon P and Gogotsi Y 2008 Nature Mater. 7 845 Zhou C, Zhang Y W, Li Y Y and Liu J P 2013 Nano Lett. 13 2078 Liu J P, Jiang J, Cheng C W, Li H X, Zhang J X, Gong H and Fan H J 2011 Adv. Mater. 23 2076 7
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
[25] Lee S Y, Choi K H, Choi W S, Kwon Y H, Jung H R, Shin H C and Kim J Y 2013 Energy Environ. Sci. 6 2414 [26] Kang Y J, Chung H, Han C H and Kim W 2012 Nanotechnology 23 065401 [27] Hu Y, Zhao Y, Lu G W, Chen N, Zhang Z P, Li H, Shao H B and Qu L T 2013 Nanotechnology 24 195401 [28] Jiang J, Liu J P, Zhou W W, Zhu J H, Huang X T, Qi X Y, Zhang H and Yu T 2011 Energy Environ. Sci. 4 5000 [29] Kim J H, Zhu K, Yan Y F, Perkins C L and Frank A J 2010 Nano Lett. 10 4099 [30] Li R Z, Ren X, Zhang F, Du C and Liu J P 2012 Chem. Commun. 48 5010 [31] Kim J H, Kang S H, Zhu K, Kim J Y, Neale N R and Frank A J 2011 Chem. Commun. 47 5214 [32] Huang C, Wu Y, Hu C and Li Y 2007 J. Power Sources 172 460
[15] Hsu Y K, Hen Y C, Lin Y G, Chen L C and Chen K H 2012 J. Mater. Chem. 22 3383 [16] Du C, Yeh J and Pan N 2005 Nanotechnology 16 350 [17] Hu L, Wu H and Cui Y 2010 Appl. Phys. Lett. 96 183502 [18] Kang Y J, Chun S J, Lee S S, Kim B Y, Kim J H, Chung H, Lee S Y and Kim W 2012 ACS Nano 6 6400 [19] Du C and Pan N 2006 J. Power Sources 160 1487 [20] Lu X H et al 2012 Adv. Mater. 24 938 [21] Liu B, Zhang J, Wang X F, Chen G, Chen D, Zhou C W and Shen G Z 2012 Nano Lett. 12 3005 [22] Lu X H, Yu M H, Wang G M, Zhai T, Xie S L, Ling Y C, Tong Y X and Li Y 2013 Adv. Mater. 25 267 [23] Jiang J, Li Y Y, Liu J P, Huang X T, Yuan C Z and Lou X W 2012 Adv. Mater. 24 5166 [24] Si W P, Yan C L, Chen Y, Oswald S, Han L Y and Schmidt O G 2013 Energy Environ. Sci. 6 3218–23
8
Search
Collections
Journals
About
Contact us
My IOPscience
Carbon nanotube network film directly grown on carbon cloth for high-performance solid-state flexible supercapacitors
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 035402 (http://iopscience.iop.org/0957-4484/25/3/035402) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 131.91.169.193 This content was downloaded on 11/11/2014 at 23:14
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 035402 (8pp)
doi:10.1088/0957-4484/25/3/035402
Carbon nanotube network film directly grown on carbon cloth for high-performance solid-state flexible supercapacitors Cheng Zhou and Jinping Liu Institute of Nanoscience and Nanotechnology, Department of Physics, Central China Normal University, Wuhan, Hubei 430079, People’s Republic of China E-mail: [email protected] Received 9 October 2013, revised 14 November 2013 Accepted for publication 25 November 2013 Published 20 December 2013 Abstract
Carbon nanotubes (CNTs) have received increasing attention as electrode materials for high-performance supercapacitors. We herein present a straightforward method to synthesize CNT films directly on carbon cloths as electrodes for all-solid-state flexible supercapacitors (AFSCs). The as-made highly conductive electrodes possess a three-dimensional (3D) network architecture for fast ion diffusion and good flexibility, leading to an AFSC with a specific capacitance of 106.1 F g−1 , an areal capacitance of 38.75 mF cm−2 , an ultralong cycle life of 100 000 times (capacitance retention: 99%), a good rate capability (can scan at 1000 mV s−1 , at which the capacitance is still ∼37.8% of that at 5 mV s−1 ), a high energy density (2.4 µW h cm−2 ) and a high power density (19 mW cm−2 ). Moreover, our AFSC maintains excellent electrochemical attributes even with serious shape deformation (bending, folding, etc), high mechanical pressure (63 kPa) and a wide temperature window (up to 100 ◦ C). After charging for only 5 s, three such AFSC devices connected in series can efficiently power a red round LED for 60 s. Our work could pave the way for the design of practical AFSCs, which are expected to be used for various flexible portable/wearable electronic devices in the future. Keywords: carbon nanotube, solid-state supercapacitors, flexible electronics, energy storage, charge storage (Some figures may appear in colour only in the online journal)
1. Introduction
lightweight, flexible, and of high power and energy densities are also promising to meet the various requirements of modern electronic products [3]. As one kind of SCs, electric double-layer capacitors (EDLCs) store energy in the form of charge separation between the double layer formed at the interface of the solid electrode surface and the liquid or solid electrolyte. Carbon materials which have high specific surface areas and exhibit a high diversity in crystallinity, morphology, porosity, and texture, are currently the most widely used materials for EDLCs. For this, several carbon materials such as activated carbons [4–6], carbon nanotubes [7–11] (multi- and single-walled CNTs),
Climate change and the decreasing availability of fossil fuels require society to move towards sustainable and renewable resources. At the same time, emerging energy storage devices are urgently expected to maintain the stability of the power supply from these energy sources. Electrochemical capacitors, also known as supercapacitors (SCs), are one kind of such energy devices. They have received increasing attention for applications in (hybrid) electric vehicles due to their fast charge–discharge characteristic, high power density and long cycle life [1–3]. SCs which are cheap, 0957-4484/14/035402+08$33.00
1
c 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
2. Experimental section
spherical carbon nanoparticles [12], and one- to few-layered graphene [13] have been explored. Commercial EDLCs using activated carbon as electrode materials have a much higher theoretical capacitance (100–300 F g−1 ) compared to conventional electrolytic capacitors [4] (on the order of micro- or picofarads). However, activated carbon electrodes are not ideal for multifunctional energy storage devices such as flexible EDLCs due to the fragile nature of particulate film. Carbon nanotubes (CNT) are an important family of carbonaceous materials, which has been attracting great attention in recent years [14–16]. Owing to the highly accessible surface area, superior electrical conductivity, electrochemical stability and mechanical flexibility, CNT is promising as an advanced electrode material for EDLCs. Preparation methods of CNT electrodes in the literature were, however, generally indirect, involving several procedures. The as-grown nanotubes were firstly harvested in the form of power or from the substrate and re-dispersed in liquids. These suspended nanotubes were then re-assembled on current collectors to form the porous electrodes via slurry casting [6], ink-jet printing [17], vacuum filtering [18], or electrophoretic deposition [19], etc. Since the CNTs were densely packaged in these electrodes, these approaches would decrease the available surface area of CNTs to some extent, which is very harmful to the SC device’s properties. Herein, we have constructed our CNT electrodes using a more facile approach, which involves the direct growth of CNT films on the current collector through a chemical vapor deposition (CVD) process using nickel as the catalyst. The as-grown CNTs are slightly entangled, forming a well-developed network with tremendous open pores. This porous three-dimensional (3D) architecture allows the rapid transport of ions from the electrolyte to the entire surface of the CNT electrode and makes it quickly available for electric double-layer formation. Flexible and lightweight SCs are promising nextgeneration energy storage devices. Recently, some studies have been carried out to integrate nanoscale materials with current collectors to prepare flexible, thin, lightweight, and wearable power conversion and storage devices [3, 9, 18, 20–27]. An easy-to-assemble integrated nanocomposite energy storage system could provide design ingenuity for a variety of devices operating under a hard environment condition, for example, high mechanical stress and a wide range of temperature [7]. In this work, we propose combining directly grown CNTs and flexible commercial carbon cloth (CC) in an integrated composite structure as electrodes and using H3 PO4 /poly(vinyl alcohol) (PVA) polymer gel as the electrolyte/separator to assemble an all-solid-state flexible supercapacitor (AFSC). As a result of the unique 3D CNT electrode architecture, the AFSC exhibits excellent properties such as ultralong lifetime and high rate capability. These attributes can still be maintained when high mechanical pressure (63 kPa) and high temperature (up to 100 ◦ C) are applied, demonstrating great promise for future flexible electronics. This work could open up new opportunities in the design and fabrication of high-performance AFSCs.
2.1. Preparation of CNTs/CC hybrid electrode
The CNT network film was synthesized directly on CC by a nickel-catalyzed CVD process. In detail, nickel nitrate hexahydrate was firstly dissolved in 50 ml absolute alcohol and ethylene glycol (1:1) under stirring to form a uniform mixture. Prior to the fabrication of CNTs, a piece of CC (CeTech, through-plane electrical conductivity: ∼2 × 102 S cm−2 , 2.0 × 4.0 cm2 ) was immersed into the above solution for 1 h to adsorb the nickel salt. The treated CC was then put at the center of a tube furnace and heated at a rate of 30 ◦ C min−1 to 800 ◦ C for 20 min with a mixed solution (ethanol and ethylene glycol with the volume ratio of 1:5) placed at the tube entrance as the carbon source under flowing argon atmosphere. The adsorbed nickel salt will be decomposed at high temperature and will subsequently be reduced to nickel catalyst by the vapor of ethanol and ethylene glycol, initiating the CNT growth [28]. 2.2. Characterization
Samples were characterized by scanning electron microscopy (SEM, JSM-6700F, 5 kV), and transmission electron microscopy (TEM, JEM-2010FEF, 200 kV), and Raman spectroscopy was carried out on a Witech CRM200 (532 nm). The Brunauer–Emmett–Teller (BET) specific surface area was measured on a Bel Sorp-mini (S/N-00230) analyzer. The mass of electrode materials was measured on an AX/MX/UMX balance (METTLER TOLEDO, maximum = 5.1 g; d = 0.001 mg). 2.3. Fabrication of AFSC
Firstly, the gel electrolyte was prepared: 12 g H3 PO4 was added into 60 ml deionized water and then 6 g PVA power was added. The whole mixture was heated to 85 ◦ C under stirring until the solution became clear. The AFSC device was assembled as follows: the polymer gelled electrolyte was slowly poured onto the CNTs/CC electrodes (100 µL electrolyte/1 cm2 of the electrode). This assembly was left under ambient conditions for 5 h to ensure that the electrolyte completely wet the electrode and to allow for evaporation of most excess water. Two electrodes were then assembled face-to-face and left overnight until the electrolyte solidified. This results in mechanically robust devices with the polymer electrolyte acting as both the electrolyte and the ion-porous separator. The solidified polymer electrolyte simplifies the device architecture and helps maintain the electrodes in closer proximity. 2.4. Electrochemical performance measurement
All the electrochemical performance was tested on a CS310 Electrochemical Workstation. The CNT electrode’s property was investigated in a three-electrode electrochemical cell, using a 1 M Li2 SO4 or H3 PO4 aqueous solution as the 2
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 1. (a) SEM image of the pure CC. The inset shows the optical image. (b) Optical image of CNTs/CC electrode. (c) and (d) SEM
images of CNTs/CC electrode at different magnifications.
figure 1(a). Figure 1(b) shows an optical image of a piece of CNTs/CC electrode. The electrode is still mechanically flexible even under folding conditions after the CNT growth. The representative SEM images of CNTs/CC electrode are shown in figures 1(c) and (d). It can be observed that entangled CNTs are uniformly grown onto the surface of CC microfibers over almost the entire porous CC, exhibiting a 3D network morphology. BET measurement reveals that the specific surface area of CNTs/CC electrode is ∼298 m2 g−1 , much higher than that of pure CC (∼27 m2 g−1 ). The TEM image in figure 2(a) clearly demonstrates the CNTs are multi-walled with ca. 0.34 nm interplanar spacing. Nickel nanoparticles are also detected in the CNT structure, confirming the vapor–liquid–solid (VLS) process [16]. The Raman spectrum of CNTs is presented in figure 2(b). The 1340 cm−1 peak (D band) corresponds to the disorder-induced feature due to the lattice distortion, while the 1570 cm−1 peak (G band) corresponds to the in-plane stretching vibration mode E2g of single crystal graphite. To investigate the charge storage behavior, CV curves of both the pristine CC and CNTs/CC electrodes were compared and the results are shown in figures 3(a)–(c). The CC exhibits a stable electrochemical performance at the potential window ranging from 0 to 0.8 V versus Ag/AgCl in 1 M Li2 SO4 aqueous solution. However, the current density is small, revealing the small capacitance of pristine CC. After the CNT growth, the current in the CVs increases dramatically (figures 3(b) and (c)) for the CNTs/CC electrode tested in both 1 M Li2 SO4 and H3 PO4 electrolytes. The rectangular shape of the CV curves at different scan rates (even up to 200 mV s−1 )
electrolyte to demonstrate the EDL capacitance. A Pt plate and Ag/AgCl were used as the counter-electrode and the reference electrode, respectively. The AFSC device’s performance was tested in a two-electrode mode. The specific capacitance (F g−1 ) was calculated based on the total mass of the CNT materials. Areal and specific capacitances were calculated based on cyclic voltammetry (CV) using equations R R I dV I dV Ca = 21 × 1V·υ·S and Csp = 12 × 1V·υ·m , respectively, where I is the current, 1V is the potential window, υ is the scan rate, m is the mass of the electrode material, and S is the geometrical area of the electrode. Electrochemical impedance spectroscopy (EIS) measurements were performed by applying an AC voltage with 5 mV amplitude in a frequency range from 0.01 Hz to 100 kHz at open circuit potential. To investigate the temperature effect, the device was placed in a small oven with precise temperature control and display. Two Pt wires were used to connect the device with the outer Electrochemical Station. To investigate the pressure effect, a 1 cm2 AFSC device was placed on a flat glass plate, and then regular iron blocks with different definite masses were added directly to the upper surface of the device one by one to generate the mechanical pressure. 3. Results and discussion
Figure 1 shows the structural characterization of the CNTs/CC hybrid electrode. The SEM image in figure 1(a) reveals that pristine CC consists of interwoven carbon fibers with diameters ranging from 10 to 12 µm. A photograph of the commercial CC is further displayed in the inset of 3
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 2. (a) TEM image of CNTs, showing the presence of nickel. (b) Raman spectrum of CNT film.
Figure 3. (a) CV curves of the pure CC electrode at different scan rates. (b) and (c) CV curves of the CNTs/CC hybrid electrode with 1 M
Li2 SO4 and 1 M H3 PO4 as aqueous electrolyte respectively at different scan rates. (d) Rate capability of the CC and CNTs/CC electrodes.
indicates the ideal electric double-layer capacitive behavior from the 3D CNT network film. The areal capacitances calculated from the CVs at different scan rates for pristine CC and CNTs/CC electrodes (in H3 PO4 electrolyte) are also shown in figure 3(d). It is noted that while both the electrodes demonstrate good rate capability, there is almost 100 times capacitance increase for the CNTs/CC electrode as compared to pure CC. Since the CNT network film on CC displayed excellent electrochemical properties, a two-electrode AFSC device was further constructed. Figure 4(a) shows the structure of the device, in which two pieces of CNTs/CC are
used respectively as the positive and negative electrodes with H3 PO4 /PVA gel as the electrolyte/separator. The EIS spectrum in figure 4(b) exhibits a small semicircle at high frequency and a long straight line at low frequency. The straight line with small angle to the imaginary axis reveals the good capacitive behavior of our device. From the extended spectrum in the inset, the equivalent series resistance (the high frequency intercept of the semicircle on the real axis) and charge-transfer resistance (diameter of the semicircle) can be determined as ∼9 and 3 , respectively. The small values of these resistances indicate the facile electron and ion transport/diffusion in the device. 4
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 4. (a) Schematic illustration of the structure of CNTs/CC-based AFSC. (b) The EIS of the entire device. (c) CV and (d) rate
capability and the related charge–discharge curves of the device. (e) Ragone plot of the device. (f) Cycle stability of the device up to 100 000 times.
Both CV and galvanostatic charge/discharge tests were used to evaluate the capacitive performance of the entire device. Figure 4(c) shows the CV curves of the AFSC measured at different scan rates of 20, 50, 100 and 200 mV s−1 with a potential window ranging from 0 to 0.8 V. These CV curves show ideal capacitive behavior with nearly rectangular shape and retain the rectangular shape without apparent distortion with increasing scan rate up to 200 mV s−1 , indicating stable and high electrochemical performance of the AFSC. At 50 mV s−1 , the device’s areal capacitance is determined to be 38.75 mF cm−2 (corresponding specific capacitance: 106.1 F g−1 ), which is much higher than that of conventional carbon-based SCs (µF cm−2 ), and even larger than those of state-of-the-art graphene SCs (∼4 mF cm−2 ) [3] and some SCs based on pseudocapacitive materials such as NiO–TiO2 nanotube arrays [29], Fe3 O4 –SnO2 nanorod films [30] and Ni–NiO core–shell inverse opals [31]. Further
investigation reveals that the device still achieves 35.5% capacitance retention when the scan rate increases 1000 times from 5 to 5000 mV s−1 , indicating a good rate capability. Figure 4(d) illustrates the rate capability of capacitance versus current density of the AFSC and the related galvanostatic charge/discharge curves at the current density of 0.3, 0.5, 1.0, 2.0 and 3.0 mA cm−2 , respectively. All the curves exhibit a straight-line shape and have a very symmetric nature, indicating a rapid current–voltage response and good electrochemical reversibility. Energy density and power density are two key factors for evaluating the practical applications of SCs. A good SC is expected to provide both high energy and power densities [1]. Figure 4(e) shows the Ragone plot for our CNTs/CC-based AFSC device, obtained from the CV data. The energy density decreases from 2.4 to 0.84 µW h cm−2 , while the power density increases from 0.05 up to 19 mW cm−2 as the 5
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 5. (a) CVs of the device at different temperatures at 50 mV s−1 . (b) The capacitance value variation of the device upon the gradual increase and then decrease of the temperature.
scan rate increases from 20 to 5000 mV s−1 . The overall performance of our device is comparable to that of traditional CNT-based SCs established with liquid electrolyte [16, 32]. Another important requirement for SC application is cycling capability. To evaluate this, the charge–discharge cycling test over 100 000 times for the device at a current density of 1 mA cm−2 was carried out. Figure 4(f) shows the capacitance retention of the AFSC device as a function of cycling number. It is interesting that there is only 1% loss in the capacitance after 100 000 continuous cycles, meeting the standard for practical application. The electrochemical performance of our AFSC device was also systematically studied under different operating temperatures, ranging from 25 to 100 ◦ C. As shown in figure 5(a), after heating to 100 ◦ C, the CV curve of the AFSC device shows a slight change in both the shape and integrated area. The rectangular shape is obviously deformed, which should be due to the interfacial variation between the electrodes and electrolyte upon heating. The slight increase of the capacitance should be due to the fact that the ion diffusion in the gel electrolyte and thus the ion accumulation on the surface of CNTs can be promoted with the temperature increase. Nevertheless, with the temperature reduced back to normal, the CV can almost recover. Figure 5(b) shows the capacitance value variation of the device upon the gradual increase and then decrease of the temperature. It is obvious that with even great changes in the working temperature, the capacitance of our AFSC device has no significant fluctuation. This result implies that the device can be operated at a wide range of temperatures. In order to evaluate the potential of our AFSC for flexible energy storage under real conditions, the device was placed under different shape deformations and its performance was analyzed. The CV performance of this device when tested under different bending conditions (90◦ and 180◦ ) is shown in figure 6(a). From the curves, it can be seen that the bending has almost no effect on the capacitive behavior. Moreover, the stability of the device was tested for more than 500 cycles in the bending state, with about 10% increase in the device capacitance, as shown in figure 6(b). The increase
of the capacitance upon serious bending may be attributed to the enhanced interfacial contact between the 3D CNT film and the solid-state electrolyte, ensuring more charge accumulation. We also measured the performance of our AFSC under different mechanical pressures (figure 6(c)). As can be seen, the capacitance of the device does not show obvious differences when the pressure is higher than 10 kPa. In addition, the device can be operated even when a pressure of 63 kPa was applied. This performance durability can be attributed to the high mechanical flexibility of the electrodes along with the interpenetrating 3D network structure between the CNTs/CC films and the gelled electrolyte. The electrolyte solidifies during the device assembly and acts like a glue that holds all the device components together, improving the mechanical integrity and increasing its cycle life even when tested under extreme bending and pressure conditions. As this remarkable performance has yet to be realized in commercial devices, our AFSC may be ideal for next-generation flexible, portable electronics. To demonstrate the potential use of our AFSC device, we connected SC units in series to drive light-emitting-diodes (LED). Each AFSC unit has the same electrode area of 1 cm2 and an output voltage of ∼0.8 V. We assembled three AFSCs in series, and after charging for only 5 s to ∼2.4 V, the integrated device could power a 3 mm-diameter red round LED indicator (1.8 V, 20 mA) for more than 60 s efficiently, as shown in figure 6(d). It is expected that the performance of the AFSC can be further improved by using an ionic-liquid-based gel electrolyte with a wider operation potential [3], increasing CNT mass loading, and combination with other pseudocapacitive materials. 4. Conclusions
In summary, we have successfully developed a straightforward method to prepare a CNT network film directly on the CC current collector and constructed an AFSC device. The AFSC device delivers high energy and power densities and exhibits a good rate capability and an outstanding cycling life exceeding 100 000 times. Furthermore, the electrochemical 6
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
Figure 6. (a) CV curves of the device under different bending conditions. Inset is the optical images showing the bending states. (b) Cycle
stability of device recorded under normal and bending conditions. (c) Plot of capacitance versus mechanical pressure. (d) Optical image showing that the red LED is powered by three tandem AFSC devices.
attributes of our device can be maintained even with serious shape deformation (bending, folding, etc), high mechanical pressure and wide temperature window. Our work presents an opportunity to establish an AFSC device that can be operated under harsh environmental conditions.
[3] El-Kady M F, Strong V, Dubin S and Kaner R B 2012 Science 335 1326 Meng C Z, Liu C H, Chen L Z, Hu C H and Fan S S 2010 Nano Lett. 10 4025 Horng Y Y, Lu Y C, Hsu Y K, Chen C C, Chen L C and Chen K H 2010 J. Power Sources 195 4418 [4] Zhang L L and Zhao X S 2009 Chem. Soc. Rev. 38 2520 [5] Kaempgen M, Chan C K, Ma J, Cui Y and Gruner G 2009 Nano Lett. 9 1872 [6] Gao Q, Demarconnay L, Raymundo-Pi˜nero E and B´eguin F 2012 Energy Environ. Sci. 5 9611 [7] Hu L, Choi J W, Yang Y, Jeong S, Mantia F La, Cui L F and Cui Y 2009 Proc. Natl Acad. Sci. 106 21490 Kim Y S, Kumar K, Fisher F T and Yang E H 2012 Nanotechnology 23 015301 [8] Li X, Rong J and Wei B 2010 ACS Nano 4 6039 [9] Li G R, Xu H, Lu X F, Feng J X, Tong Y X and Su C Y 2013 Nanoscale 5 4056 Do Q H, Fielitz T R, Zeng C C, Vanli O A, Zhang C and Zheng J P 2013 Nanotechnology 24 315401 [10] Fan Z J, Yan J, Zhi L J, Zhang Q, Wei T, Feng J, Zhang M, Qian W and Wei F 2010 Adv. Mater. 22 3723 [11] Niu Z Q et al 2011 Energy Environ. Sci. 4 1440 [12] Pech D, Brunet M, Durou H, Huang P H, Mochalin V, Gogotsi Y, Taberna P L and Simon P 2010 Nature Nanotechnol. 5 651 [13] Zhu Y W et al 2011 Science 332 1537 [14] NoKed M, Okashy S, Zimrin T and Aurbach D 2012 Angew. Chem. Int. Edn 51 1568 Basirico L and Lanzara G 2012 Nanotechnology 23 305401
Acknowledgments
Liu Jinping thanks the National Natural Science Foundation of China (Nos 51102105, 11104088), Science Fund for Distinguished Young Scholars of Hubei Province (No. 2013CFA023), Self-determined Research Funds of CCNU from the Colleges’ Basic Research and Operation of MOE (CCNU12A01009) and National Innovation Experiment Program for University Students (J1210068). References [1] Miller J R and Simon P 2008 Science 321 651 Mai L Q, Yang F, Zhao Y L, Xu X, Xu L and Luo Y Z 2011 Nature Commun. 2 381 [2] Simon P and Gogotsi Y 2008 Nature Mater. 7 845 Zhou C, Zhang Y W, Li Y Y and Liu J P 2013 Nano Lett. 13 2078 Liu J P, Jiang J, Cheng C W, Li H X, Zhang J X, Gong H and Fan H J 2011 Adv. Mater. 23 2076 7
Nanotechnology 25 (2014) 035402
C Zhou and J Liu
[25] Lee S Y, Choi K H, Choi W S, Kwon Y H, Jung H R, Shin H C and Kim J Y 2013 Energy Environ. Sci. 6 2414 [26] Kang Y J, Chung H, Han C H and Kim W 2012 Nanotechnology 23 065401 [27] Hu Y, Zhao Y, Lu G W, Chen N, Zhang Z P, Li H, Shao H B and Qu L T 2013 Nanotechnology 24 195401 [28] Jiang J, Liu J P, Zhou W W, Zhu J H, Huang X T, Qi X Y, Zhang H and Yu T 2011 Energy Environ. Sci. 4 5000 [29] Kim J H, Zhu K, Yan Y F, Perkins C L and Frank A J 2010 Nano Lett. 10 4099 [30] Li R Z, Ren X, Zhang F, Du C and Liu J P 2012 Chem. Commun. 48 5010 [31] Kim J H, Kang S H, Zhu K, Kim J Y, Neale N R and Frank A J 2011 Chem. Commun. 47 5214 [32] Huang C, Wu Y, Hu C and Li Y 2007 J. Power Sources 172 460
[15] Hsu Y K, Hen Y C, Lin Y G, Chen L C and Chen K H 2012 J. Mater. Chem. 22 3383 [16] Du C, Yeh J and Pan N 2005 Nanotechnology 16 350 [17] Hu L, Wu H and Cui Y 2010 Appl. Phys. Lett. 96 183502 [18] Kang Y J, Chun S J, Lee S S, Kim B Y, Kim J H, Chung H, Lee S Y and Kim W 2012 ACS Nano 6 6400 [19] Du C and Pan N 2006 J. Power Sources 160 1487 [20] Lu X H et al 2012 Adv. Mater. 24 938 [21] Liu B, Zhang J, Wang X F, Chen G, Chen D, Zhou C W and Shen G Z 2012 Nano Lett. 12 3005 [22] Lu X H, Yu M H, Wang G M, Zhai T, Xie S L, Ling Y C, Tong Y X and Li Y 2013 Adv. Mater. 25 267 [23] Jiang J, Li Y Y, Liu J P, Huang X T, Yuan C Z and Lou X W 2012 Adv. Mater. 24 5166 [24] Si W P, Yan C L, Chen Y, Oswald S, Han L Y and Schmidt O G 2013 Energy Environ. Sci. 6 3218–23
8
