Fe Battery Based on Graphene

Nov 17, 2014 - The development of portable and wearable electronics has promoted increasing demand for high-performance power sources with high ...
8 downloads 0 Views 5MB Size
Download PDF
Letter pubs.acs.org/NanoLett

A Flexible Alkaline Rechargeable Ni/Fe Battery Based on Graphene Foam/Carbon Nanotubes Hybrid Film Jilei Liu,†,∥ Minghua Chen,† Lili Zhang,‡ Jian Jiang,† Jiaxu Yan,† Yizhong Huang,§ Jianyi Lin,∥ Hong Jin Fan,*,† and Ze Xiang Shen*,† †

Division of Physics and Applied Physics, School of Physical and Mathematical Sciences, Nanyang Technological University, 637371, Singapore ‡ Heterogeneous Catalysis, Institute of Chemical Engineering and Sciences, A*star, 1 Pesek Road, Jurong Island, 627833, Singapore § School of Materials Science and Engineering and ∥Energy Research Institute @NTU (ERI@N), Nanyang Technological University, 639798, Singapore S Supporting Information *

ABSTRACT: The development of portable and wearable electronics has promoted increasing demand for high-performance power sources with high energy/power density, low cost, lightweight, as well as ultrathin and flexible features. Here, a new type of flexible Ni/Fe cell is designed and fabricated by employing Ni(OH)2 nanosheets and porous Fe2O3 nanorods grown on lightweight graphene foam (GF)/carbon nanotubes (CNTs) hybrid films as electrodes. The assembled f-Ni/Fe cells are able to deliver high energy/power densities (100.7 Wh/kg at 287 W/kg and 70.9 Wh/kg at 1.4 kW/kg, based on the total mass of active materials) and outstanding cycling stabilities (retention 89.1% after 1000 charge/ discharge cycles). Benefiting from the use of ultralight and thin GF/CNTs hybrid films as current collectors, our f-Ni/Fe cell can exhibit a volumetric energy density of 16.6 Wh/l (based on the total volume of full cell), which is comparable to that of thin film battery and better than that of typical commercial supercapacitors. Moreover, the f-Ni/Fe cells can retain the electrochemical performance with repeated bendings. These features endow our f-Ni/Fe cells a highly promising candidate for next generation flexible energy storage systems. KEYWORDS: Ni/Fe cell, flexible battery, graphene foam, carbon nanotubes, hybrid electrodes

T

Zn batteries);15,16 (ii) both Ni and Fe are earth-abundant elements and exhibit low toxic or corrosive effect.13,15,16 Indeed, the nickel−iron battery, which was invented in 1899−1902, has been well developed.15,16 More recently, the century-old nickel− iron battery was revisited and rediscovered to explore its potential application for modern energy supply systems. Different from traditional Ni/Fe batteries, hybrid materials based on novel inorganic nanoparticles (or rods) and carbon allotropes (carbon fibers,12 carbon nanotubes, and graphene13) are now chosen as electrode materials, which indeed have improved the rate performance and energy density of Ni/Fe cells. Nevertheless, in the recent demonstrated high-performance Ni/Fe batteries13 the active materials are in powder form, so that carbon black and binders are still required. These additives, together with the heavy Ni foam current collector, will lower the gravimetric capacity of the full cells. To push the performance particularly in flexible electronic devices, it is highly desirable to design novel electrode architecture that is binder-free, bendable, and durable in long-term cycling.

he sharp proliferation of portable electronics and electrical vehicles has promoted increasing demand for highperformance power sources that have high energy density and power density and with lightweight, ultrathin, flexible, costeffective, and environmentally friendly characteristics.1−3 Among various energy storage devices, aqueous rechargeable batteries have attracted much attention due to their high ionic conductivity, environmental issues, good safety, and low cost.4−7 Many types of aqueous rechargeable batteries have been explored, including aqueous alkali-ion (Li+, Na+, K+) batteries,8,9 aqueous metal-ion batteries (Al3+, Zn2+)10,11 and aqueous nickel/metal (zinc, cadmium, cobalt, and iron) batteries.5,6,12,13 In contrast to reversible intercalation/deintercaltion chemistry for aqueous alkali-ion batteries and metalion batteries, the nickel/metal battery are mainly based on faradaic reactions that involve one or multielectron reactions on electrode materials.5 Numerous electrochemical redox couples have been explored for alkaline rechargeable nickel/metal batteries, such as nickel/cadmium,6,7 nickel/zinc,14 nickel/ cobalt,5 and nickel/iron.12,13 Among these, aqueous Ni/Fe batteries are particularly favorable because (i) both Ni and Fe active materials are insoluble in alkaline solution and has no requirement for the separator (in contrast with Ni/Zn and Ag/ © XXXX American Chemical Society

Received: October 7, 2014 Revised: November 7, 2014

A

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Information Figure S3a) indicates that the Ni(OH)2 nanosheets are polycrystalline. Well-resolved lattice fringes of 0.26 nm is observed (Supporting Information Figure S3b), corresponding to the (101) plane of α-Ni(OH)2 (JCPDS #38-0715). X-ray diffraction (XRD) analysis further confirms the formation of rhombohedral α-Ni(OH)2. As shown in Figure 2c, all the peaks could be indexed to rhombohedral α-Ni(OH)2 except the crystalline peaks at 26.4° and 44.3° that result from GF/CNTs substrate. X-ray photoelectron spectroscopy (XPS) analysis was conducted to shed more light on composition and oxidization stage of Ni(OH)2. Two typical peaks centered at 855.6 and 873.0 eV were observed (Figure 2d), corresponding to Ni 2p3/2 and Ni 2p1/2 of Ni(OH)2, respectively.19 Meanwhile, two satellite lines associated with Ni 2p also appeared. GF/CNTs/Fe2O3 hybrid films were synthesized by a facile hydrothermal reaction of FeOOH nanorods followed by thermal annealing (see details in Supporting Information Materials and Methods section). Representative SEM images in Figure 3a and Supporting Information Figure S4 reveal the well-defined interconnected network structure of GF/CNTs/Fe2O3, where the columnar Fe2O3 nanorods grown out of CNTs uniformly with an outer diameter of 200−300 nm. TEM image (Figure 3b) clearly identifies the mesoporous structure of the Fe2O3 nanorods with a diameter of 50−150 nm and a length of ranging 50−500 nm. This is further confirmed by magnified TEM image in Supporting Information Figure S5a, where the pore size is around 5−10 nm. Corresponding Brunauer−Emmett−Teller (BET) specific surface area of the GF/CNTs/Fe2O3 hybrid film is around 68 m2/g (Supporting Information Figure S5c), which is comparable with rGO/Fe2O3 composites20 reported previously and higher than that of mesoporous Fe2O3.21 In addition, the homogeneous distribution of both Fe and O corroborates the uniform coating of Fe2O3 (Figure 3b). The selected area electron diffraction (SAED, inset of Supporting Information Figure S5a) pattern indicates that Fe2O3 nanorodes are polycrystalline. The well-defined lattice fringes with distances of 0.19 and 0.27 nm (Supporting Information Figure S5b) correspond to the dspacings of (024) and (104) planes, respectively. The XRD pattern of the as-prepared GF/CNTs/Fe2O3 is shown in Figure 3c. The characteristic peaks in XRD can be well indexed as tetragonal α-Fe2O3 (JCPDS 33-0664).22 XPS result (Figure 3d) exhibits typical Fe 2p spectrum of Fe2O3 with two peaks centered at 711.0 and 724.6 eV, corresponding to Fe 2p3/2 and Fe 2p1/2, respectively.23 The presence of satellite line of the main Fe 2p3/2 located at 719.0 eV further verifies the Fe2O3 phase rather than Fe3O4.23 Electrochemical Properties of GF/CNTs/Ni(OH)2 and GF/CNTs/Fe2O3 Electrodes. The electrochemical properties of both GF/CNTs/Ni(OH)2 cathode and GF/CNTs/Fe2O3 anode were investigated in a three-electrode configuration containing 6 M KOH solution. Data for electrodes of other mass loadings are provided in Supporting Information (Figures S6 and S7). Hereby, we start our discussion based on the chargematching electrodes, viz., GF/CNTs/Ni(OH)2-0.9 positive electrode and GF/CNTs/Fe2O3-0.6 negative electrode (0.9 and 0.6 denote the areal mass densities of the corresponding active materials in unit of mg/cm2; the areal mass density of GF/ CNTs hybrid films is 0.65 mg/cm2). Figures 4a,b exhibit typical cyclic voltammetry (CV) curves of GF/CNTs/Ni(OH)2-0.9 and GF/CNTs/Fe2O3-0.6 in aqueous electrolyte. The oxidization and reduction potential peaks of Ni(OH)2 are observed at 0.36 and 0.21 V (vs SCE), respectively. This corresponds to the reversible reaction as follows7,12,24

Recently, we have developed a class of graphene foam/carbon nanotubes (GF/CNTs) hybrid films as an excellent current collector for electrodes (Supporting Information Figure S1).17 These hybrid films are lightweight, ultrathin, highly conductive and have large surface-to-volume ratios. They are ideal for depositing nanometer-sized active materials without binders or carbon additives toward flexible and high-performance electrochemical storage devices. In this work, rechargeable Ni/Fe batteries are constructed by direct growth of Ni(OH) 2 nanosheets (cathode material) and mesoporous Fe2O3 nanorods (anode material) on the GF/CNTs hybrid films. Figure 1 shows

Figure 1. Schematics of the flexible Ni/Fe (f-Ni/Fe) cells. (a) GF/ CNTs hybrid films are used as the current collector for both electrodes, and Ni(OH)2 nanosheets and porous Fe2O3 nanorods are grown around individual CNTs. (b) The f-Ni/Fe cell structure. (c) Working mechanism of the f-Ni/Fe cell and involved electrochemical reactions.

schematically the architecture of the electrodes and the full cell structure, as well as the working mechanism. These novel flexible Ni/Fe cells demonstrate superior energy and power densities stemming from the nanosized feature of active materials and the hierarchical structure of the electrodes. Stable capacities on repeated bending and long-term (up to 2000) cycling are demonstrated. Structural Characterization of GF/CNTs/Ni(OH)2 and GF/CNTs/Fe2O3 hybrid films. The in situ electrochemical deposition method was employed to prepare Ni(OH)2 on GF/ CNTs hybrid films because it not only produces thin nanosheet structure of Ni(OH)2 active material but also allows a fine-tuning of mass loading of Ni(OH)2 via modulating the concentrations of electrolyte and deposition time.18 FESEM images of the resulting GF/CNTs/Ni(OH)2 hybrid film in Figure 2a reveal that flowerlike Ni(OH)2 nanosheets surrounding the CNTs uniformly with outer diameters ranging from 150 to 300 nm (depending on the mass loading amount, Supporting Information Figure S2). The areal mass loading of Ni(OH)2 could be easily tuned from 0.5 to 1.2 mg/cm2 by adjusting the deposition time (Supporting Information Figure S2). Transmission electron microscopy (TEM) images (Figure 2b and Supporting Information Figure S3a) collected from Ni(OH)2 nanostructure also verify its nanosheet morphology, and the corresponding selected-area electron diffraction (SAED) pattern (inset in Supporting B

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 2. Characterization of the GF/CNTs/Ni(OH)2 hybrid film. (a) Typical FESEM image (inset is magnified SEM image) and (b) TEM image of the hybrid film and corresponding element mappings of Ni and O. (c) XRD pattern and (d) XPS spectra.

Figure 3. Characterization of the GF/CNTs/Fe2O3 hybrid film. (a) Typical FESEM image (inset is enlarged view) and (b) TEM image TEM and corresponding element mappings of Fe and O. (c) XRD pattern and (d) XPS spectra.

The GF/CNTs/Fe2O3 electrode displays two oxidization peaks around −0.89 and −0.71 V in the anodic scan, corresponding to the formation of Fe(OH)2 (with the oxidization from Fe0 to

Ni(OH)2 + OH− ↔ NiOOH + H 2O + e− 0

(E = 0.249 V vs SCE)

(1) C

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 4. Electrochemical characterizations of the hybrid film electrodes. Left column: GF/CNTs/Ni(OH)2-0.9 cathode. Right column: GF/CNTs/ Fe2O3-0.6 anode. (a,b) Cyclic voltammetry curves of at a scan rates of 5 mV/s. (c,d) Variation of anodic peak current (Ipa) with scan rate. (e,f) Galvanostatic discharge curves of at various current densities. (g,h) Cycling performance. Insets are the corresponding CV curves after 1000 cycles.

Fe2+) and FeOOH (with the oxidization from Fe2+ to Fe3+), respectively. One well-defined reduction peak at around −1.05 V was found in the cathodic curve, which is assigned to the reduction from Fe3+ to Fe2+. The missing of another cathodic peak (corresponding to the reduction of Fe2+ to Fe0) may be due to distortion by the H2 evolution. The current intensity and the shifts of both the anodic peak (to a more positive potential) and cathodic peak (to a more negative potential) positions, ΔEa,c, increase with increasing scan rate.25,26 These are attributed to an increase in the charge diffusion

polarization within the pseudoactive material when the scan rate increases.27 The peak currents (Ipa) are plotted as a function of sweep rate (υ) and υ1/2, respectively, in Figure 4c for the GF/ CNTs/Ni(OH)2-0.9 electrode and in Figure 4d for the GF/ CNTs/Fe2O3-0.6 electrode. It can be seen that the Ipa versus υ1/2 plots exhibit a linear relationship, whereas the Ipa vs υ plots display a nonlinear behavior. Generally, Ipa vs υ1/2 plot shows linear relationship regardless of the scan rate for a kinetically uncomplicated redox reaction in semi-infinite diffusion controlled CV, and the Ipa versus υ plots are expected to be linear for D

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 5. Performance of the flexible Ni/Fe full cells. Cyclic voltammetry curves (a) and corresponding gravimetric capacity (b) at different scan rates. (c) Galvanostatic discharge curves at various current densities and (d) the Ragone plot. Comparisons are made in two ways: based on the active material mass only (upper half) and based on the whole device mass (bottom shaded half). (e) Cycling performance of the f-Ni/Fe cell at a current density of 1.3 A/g for 1000 cycles. Inset shows the charge/discharge curves for the 1st and 1000th cycles. (f) Cycling performance of f-Ni/Fe cell at different bending angles. Inset shows the charge/discharge curves at different bending angles.

an adsorption process.28−33 Therefore, it is reasonable to deduce that the oxidization processes are diffusion limited based on the linear relationship between Ipa and υ1/2 for both GF/CNTs/ Ni(OH)2-0.9 positive electrode and GF/CNTs/Fe2O3-0.6 negative electrode, consisting well with previous reported results about LiFePO4.31 The typical discharge curves were plotted for Ni(OH)2 electrode (Figure 4e) and Fe2O3 (Figure 4f) with various discharge current densities. A well-defined plateau at around 0.25 V was observed for Ni(OH)2 electrode, consisting well with eq 1. The oxidization reactions of iron anode in alkaline electrolytes could be described as7,15,34−37 Fe + 2OH− ↔ Fe(OH)2 + 2e−

Fe(OH)2 + OH− ↔ FeOOH + H 2O + e−

(4)

The well-defined plateau around −0.77 V (vs SCE) can be assigned to the reaction 4. The plateau corresponding to reaction 2 is not evident in either CV or discharge curves, which suggests that the anodic process is dominated by reaction 4. Possibly reaction 2 takes place synchronously with the hydrogen evolution during the cathodic scan when the H2 evolution occurs at a more positive potential. The gravimetric capacity of GF/CNTs/Ni(OH)2 electrode and GF/CNTs/Fe2O3 electrode at different current densities (presented in Figures 4e and 4f) were calculated from the corresponding galvanostatic discharge curves following the equation C* = IΔt/m, where I is the discharging current, Δt is discharging time, and m is the mass of individual electrode. The gravimetric capacity (based on the mass of active material) of GF/CNTs/Ni(OH)2-0.9 and GF/CNTs/Fe2O3-0.6 are 195 and 278 mAh/g, respectively, at 1 A/g. At a high scan rate of 4 A/g, the gravimetric capacities retain at 93 and 109 mAh/g, respectively. This implies good rate capabilities for both hybrid

(E0 = − 1.076 V vs SCE) (2)

3Fe(OH)2 + 2OH− ↔ Fe3O4 + 4H 2O + 2e− (E0 = −0.859 V vs SCE)

(E0 = − 0.756 V vs SCE)

(3)

or/and E

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Ragone plots are important to evaluate the performance of a battery. The energy and power densities (E and P) were calculated using E = ∫ Δt 0 IV (t)dt and P = E/Δt, where I is the discharging current, V is discharging voltage, dt is time differential, and Δt is the discharge time. We constructed the plot in two ways, viz., one based on the active material mass only (open symbols); and the other based on the whole device mass (solid symbols) (see Figure 5d. Detailed calculation and comparisons are presented in Supporting Information Table S1−S3). By considering only the total mass of active materials, our f-Ni/Fe cell delivers an energy density of 100.7 Wh/kg at 287 W/kg and 70.9 Wh/kg at 1.4 kW/kg, respectively. Even at a high power density of 6.0 kW/kg, an energy density of 37.4 Wh/kg can still be achieved. When the total mass of the cell is considered, the use of lightweight GF/CNTs hybrid films current collector in our f-Ni/Fe cell results in a gravimetric energy density of 30.3 Wh/kg. This value is three folds that of typical commercial supercapacitors and much higher than that of Ni/Fe cells using nickel foams as the current collector (Supporting Information Table S3).13 The maximum power density is 1.8 kW/kg, which is comparable to that of commercial supercapacitor and four-folds higher than that of lithium thin film battery. In order to explore real applications of the f-Ni/Fe cell, the volumetric energy and power densities based on the whole cell configuration were also calculated (Supporting Information Table S1 and Figure S12). A maximum volumetric energy density of 16.6 Wh/l has been achieved, which is two-folds higher than the reported MnO2-based38 and Ni(OH)2-based asymmetric supercapacitors27 (Supporting Information Table S3), and both are two typical high-capacity supercapacitive materials. The maximum volumetric power density is 1.0 kW/l. Moreover, the f-Ni/Fe cell exhibits good cycling stability with capacity retention of 89.1% after 1000 charge/discharge cycles (Figure 5e). This result is better than the recently reported Ni/Fe cells which showed 80% capacity retention after 800 cycles.13 Moreover, the capacity could retain at 78% when the stability test extended further to 2000 cycles (Supporting Information Figure S11). Also, the CNTs-supported structures were well maintained although the surfaces of electrodes show some degree of aggregations during long charge/discharge process (Supporting Information Figure S11). The development of multifunctional flexible electronics requires power sources that are flexible and lightweight in addition to high performance. For practical application consideration, the f-Ni/Fe cells were mechanically bent during the measurement of charge/discharge curves (Figure 5f). It was noted that there was little changes in the charge/discharge curves at different bending angles (up to 60°). Therefore, our f-Ni/Fe cell can be bent to a large extent without sacrificing its performance.39 The practical application of our f-Ni/Fe cell was further demonstrated by powering light-emitting diodes (LEDs) or a fan using two f-Ni/Fe cells connected in series. Benefiting from the high power/energy density, the f-Ni/Fe cell could power one red LED (1.8 V, 20 mA, 5 mm diameter) and one yellow LED (1.8 V, 30 mA, 5 mm diameter) simultaneously after charging for only 30 s (Supporting Information Figure S11) or drive a small rotation motor (3 V, 0.45 W) after charging for only a few seconds. The Ni/Fe cells demonstrated in this work could potentially bridge the gap between conventional thin film Li-ion batteries and supercapacitors for application in flexible electronics. The nanostructured active materials are grown directly around the

electrodes. Additionally, these hybrid electrodes show good cycle stability with capacity retention at 97.2% for GF/CNTs/ Ni(OH)2-0.9 (Figure 4g) and 96% for GF/CNTs/Fe2O3-0.6 (Figure 4h) after 1000 cycles at a scan rate of 10 mV/s. The electrochemical performance of hybrid electrodes is strongly affected by mass loading of active materials. The gravimetric capacity and rate performance both degrade with the increase in mass loading of active materials (Supporting Information Figures S6c,d and S7d), although the corresponding current response increases. This can be attributed to the decrease in electric conductivity related to the increase in the thickness (Supporting Information Figures S6b and S7e). It is noted that the equivalent series resistance (ESR) increases from 2.9 Ω for GF/CNTs/Ni(OH)2-0.5 to 4.6 Ω for GF/CNTs/Ni(OH)2-1.2 (Supporting Information Figure S6b). Similar trend was noted for GF/CNTs/Fe2O3 electrodes with different mass loading (Supporting Information Figure S7e). This dependence of electrochemical property on mass loading of active materials demonstrates the importance of electrode design to achieve optimized performance of full cells. Performance of the f-Ni/Fe Cell. On the basis of the results above, Ni/Fe full cells were fabricated by using GF/CNTs/ Ni(OH)2-0.9 hybrid film as cathode and GF/CNTs/Fe2O3-0.6 as anode. These two electrodes were chosen because the mass loading ratio of Ni(OH)2 to Fe2O3 (0.9 mg: 0.6 mg) corresponds to a reasonably good charge balance between them (see Supporting Information Figure S8). The total active materials weight is 6 mg with cell size at 4 cm2. Figure 5a shows the CV curves of the f-Ni/Fe cell at various scan rates. Well-defined redox couples were observed, corresponding to the overall reaction in the Ni/Fe cell12 Ni(OH)2 +

1 1 1 Fe2O3 ↔ NiOOH + Fe + H 2O 6 3 2

(5)

Similar to the results of single electrodes measured in threeelectrode configuration, the difference in the anodic peak and cathodic peak positions, ΔEa,c, increases with increasing scan rate for f-Ni/Fe cell. The Ipa versus υ1/2 plot of f-Ni/Fe cell (Supporting Information Figure S9) exhibits the same trend as that of single electrodes and indicates that a diffusion limited oxidization reaction takes place during the charging process.28−31 The f-Ni/Fe cell delivers high specific capacity of 119 mAh/g at scan rate of 5 mV/s and 78 mAh/g at scan rate of 40 mV/s (Figure 5b), implying its good rate capability. Typical galvanostatic discharge curves at different current densities are shown in Figure 5c. Consistent to the CV curves, the discharge profiles of the f-Ni/Fe cell exhibit good reversibility with distinct discharge voltage plateau from 0.9 to 1.1 V, depending on current density, suggesting its application feasibility for energy storage.12,13,36 The specific capacity calculated based on discharge curves ranges from 118 to 50 mAh/g when the current density changes from 0.3 to 8 A/g. It is noteworthy that the discharge (Figure 5c) completes in short time ranging from 20 min to 30 s, depending on the current density. These discharge process is much faster than conventional Ni/Fe batteries that usually require hours. These values also outperform a recently reported Ni/Fe cell based on carbon fibers12 with a specific capacity of 80 mAh/g and charge/discharge time of hours (based on the mass of active materials). The series resistance of the devices can be estimated to be 5.9 Ω (Rs = 3.8 Ω; Rct = 2.1 Ω) from the EIS data (Supporting Information Figure S10). The relatively small serial resistances of our f-Ni/Fe cell imply a fast charge transfer between the electrolyte and electrodes. F

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

(3) Pushparaj, V. L.; Shaijumon, M. M.; Kumar, A.; Murugesan, S.; Ci, L.; Vajtai, R.; Linhardt, R. J.; Nalamasu, O.; Ajayan, P. M. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 13574−13577. (4) Gao, X. P.; Yang, H. X. Energy Environ. Sci. 2010, 3, 174−189. (5) Gao, X.-P.; Yao, S.-M.; Yan, T.-Y.; Zhou, Z. Energy Environ. Sci. 2009, 2, 502−505. (6) Kohler, U.; Antonius, C.; Bauerlein, P. J. Power Sources 2004, 127, 45−52. (7) Shukla, A. K.; Venugopalan, S.; Hariprakash, B. J. Power Sources 2001, 100, 125−148. (8) Pasta, M.; Wessells, C. D.; Huggins, R. A.; Cui, Y. Nat. Commun. 2012, 3, 1149. (9) Chen, L.; Gu, Q. W.; Zhou, X. F.; Lee, S. X.; Xia, Y. G.; Liu, Z. P. Sci. Rep. 2013, 3, 1074. (10) Xu, C.; Li, B.; Du, H.; Kang, F. Angew. Chem., Int. Ed. 2012, 51, 933−935. (11) Liu, S.; Hu, J. J.; Yan, N. F.; Pan, G. L.; Li, G. R.; Gao, X. P. Energy Environ. Sci. 2012, 5, 9743−9746. (12) Liu, Z.; Tay, S. W.; Li, X. Chem. Commun. 2011, 47, 12473− 12475. (13) Wang, H.; Liang, Y.; Gong, M.; Li, Y.; Chang, W.; Mefford, T.; Zhou, J.; Wang, J.; Regier, T.; Wei, F.; Dai, H. Nat. Commun. 2012, 3, 917. (14) Gong, M.; Li, Y.; Zhang, H.; Zhang, B.; Zhou, W.; Feng, J.; Wang, H.; Liang, Y.; Fan, Z.; Liu, J.; Dai, H. Energy Environ. Sci. 2014, 7, 2025− 2032. (15) Chakkaravarthy, C.; Periasamy, P.; Jegannathan, S.; Vasu, K. I. J. Power Sources 1991, 35, 21−35. (16) Halpert, G. J. Power Sources 1984, 12, 177−192. (17) Liu, J.; Zhang, L.; Wu, H. B.; Lin, J.; Shen, Z.; Lou, X. W. Energy Environ. Sci. 2014, 7, 3709−3719. (18) Wang, Y.-M.; Zhao, D.-D.; Zhao, Y.-Q.; Xu, C.-L.; Li, H.-L. RSC Adv. 2012, 2, 1074−1082. (19) McIntyre, N. S.; Cook, M. G. Anal. Chem. 1975, 47, 2208−2213. (20) Kim, I. T.; Magasinski, A.; Jacob, K.; Yushin, G.; Tannenbaum, R. Carbon 2013, 52, 56−64. (21) Xu, Y.; Jian, G.; Liu, Y.; Zhu, Y.; Zachariah, M. R.; Wang, C. Nano Energy 2014, 3, 26−35. (22) Liu, J. L.; Zhou, W. W.; Lai, L. F.; Yang, H. P.; Lim, S. H.; Zhen, Y. D.; Yu, T.; Shen, Z. X.; Lin, J. Y. Nano Energy 2013, 2, 726−732. (23) Fujii, T.; de Groot, F. M. F.; Sawatzky, G. A.; Voogt, F. C.; Hibma, T.; Okada, K. Phys. Rev. B 1999, 59, 3195−3202. (24) Shukla, A. K.; Ravikumar, M. K.; Balasubramanian, T. S. J. Power Sources 1994, 51, 29−36. (25) Kamath, P. V.; Ahmed, M. F. J. Appl. Electrochem. 1993, 23, 225− 230. (26) Corrigan, D. A.; Bendert, R. M. J. Electrochem. Soc. 1989, 136, 723−728. (27) Ji, J.; Zhang, L. L.; Ji, H.; Li, Y.; Zhao, X.; Bai, X.; Fan, X.; Zhang, F.; Ruoff, R. S. ACS Nano 2013, 7, 6237−6243. (28) Bing, L.; Huatang, Y.; Yunshi, Z.; Zuoxiang, Z.; Deying, S. J. Power Sources 1999, 79, 277−280. (29) MacArthur, D. M. J. Electrochem. Soc. 1970, 117, 729−733. (30) Schrebler Guzmán, R. S.; Vilche, J. R.; Arvía, A. J. J. Electrochem. Soc. 1978, 125, 1578−1587. (31) Come, J.; Taberna, P. L.; Hamelet, S.; Masquelier, C.; Simon, P. J. Electrochem. Soc. 2011, 158, A1090−A1093. (32) Augustyn, V.; Come, J.; Lowe, M. A.; Kim, J. W.; Taberna, P.-L.; Tolbert, S. H.; Abruna, H. D.; Simon, P.; Dunn, B. Nat. Mater. 2013, 12, 518−522. (33) Simon, P.; Gogotsi, Y.; Dunn, B. Science 2014, 343, 1210−1211. (34) Periasamy, P.; Babu, B. R.; Iyer, S. V. J. Power Sources 1996, 58, 35−40. (35) Periasamy, P.; Babu, B. R.; Iyer, S. V. J. Power Sources 1996, 63, 79−85. (36) Huo, G.; Lu, X.; Huang, Y.; Li, W.; Liang, G. J. Electrochem. Soc. 2014, 161, A1144−A1148. (37) Ujimine, K.; Tsutsumi, A. J. Power Sources 2006, 160, 1431−1435.

highly conductive CNTs. This feature may have contributed the following merits: (1) A favorable electric contact and electron transportation; (2) Good mechanical integrity, accounting for the high cycling stability and flexibility; (3) Relatively high mass loading of active materials compared to direct growth onto the GF or planar metallic current collectors. This also eliminates the additives (e.g., carbon black and binder). Benefiting from the unique hierarchical design, the gravimetric energy density and volumetric energy density of our f-Ni/Fe cells are comparable to thin film lithium ion battery and better than typical commercial supercapacitors. In conclusion, we report the design and successful fabrication of flexible Ni/Fe cells by growing nanostructured active materials, that is, Ni(OH)2 nanosheets and Fe2O3 nanorods, onto lightweight and flexible GF/CNTs current collectors without using binders or carbon additives. The high volumetric energy/power densities, good cycling stability, and flexibility of the f-Ni/Fe cell are clearly demonstrated. Such high-performance alkaline batteries could bridge the energy density gap between supercapacitor and thin film lithium ion battery and be a promising candidate for next generation flexible energy storage systems. The design we proposed is scalable for mass production and could be extended to the fabrication of other binder-free and flexible Ni/Metal alkaline batteries.

■

ASSOCIATED CONTENT

S Supporting Information *

Synthesis and calculation methods; SEM images of GF/CNTs hybrid film (Figure S1); more SEM images of the GF/CNTs/ Ni(OH)2 (Figure S2) with different loading amounts and corresponding HRTEM, SAED images (Figure S3); more SEM images of the GF/CNTs/Fe2O3 (Figure S4) at different loading amounts and HRTEM, SAED images, and BET result (Figure S5); CV, charge/discharge curves, and EIS spectra of the GF/ CNTs/Ni(OH)2 hybrid electrode (Figure S6) and GF/CNTs/ Fe2O3 hybrid electrode (Figure S7); charge balance (Figure S8), variation of anodic peak current (Ipa) with scan rate (Figure S9), EIS spectra (Figure S10), SEM images before and after cycling test (Figure S11), and Ragone plot based on the total volume of device (Figure S12) for f-Ni/Fe Cell; electrochemical parameters of f-Ni/Fe Cell (Table S1), commercial supercapacitors (Table S2), and some reported asymmetric supercapacitors (Table S3). This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (H.J.F.). *E-mail: [email protected] (Z.X.S.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge support from the Energy Research Institute @NTU (ERI@N). H.J.F. thanks the support by SERC Public Sector Research Funding (Grant 1121202012), Agency for Science, Technology, and Research (A*STAR).

■

REFERENCES

(1) Wang, Y.; Xia, Y. Adv. Mater. 2013, 25, 5336−5342. (2) Yang, X.; Cheng, C.; Wang, Y.; Qiu, L.; Li, D. Science 2013, 341, 534−537. G

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

(38) Huang, M.; Zhang, Y.; Li, F.; Zhang, L.; Ruoff, R. S.; Wen, Z.; Liu, Q. Sci. Rep. 2014, 4, 3878. (39) El-Kady, M. F.; Strong, V.; Dubin, S.; Kaner, R. B. Science 2012, 335, 1326−1330.

H

dx.doi.org/10.1021/nl503852m | Nano Lett. XXXX, XXX, XXX−XXX