Carbon-Based Materials for Lithium-Ion Batteries, Electrochemical Capacitors, and Their Hybrid Devices.

DOI: 10.1002/cssc.201403490

Reviews

Carbon-Based Materials for Lithium-Ion Batteries, Electrochemical Capacitors, and Their Hybrid Devices Fei Yao,[a] Duy Tho Pham,[a, b] and Young Hee Lee*[a, b]

ChemSusChem 2015, 8, 2284 – 2311

2284

Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews A rapidly developing market for portable electronic devices and hybrid electrical vehicles requires an urgent supply of mature energy-storage systems. As a result, lithium-ion batteries and electrochemical capacitors have lately attracted broad attention. Nevertheless, it is well known that both devices have their own drawbacks. With the fast development of nanoscience and nanotechnology, various structures and materials have been proposed to overcome the deficiencies of both devices to improve their electrochemical performance further. In this Review, electrochemical storage mechanisms based on carbon materials for both lithium-ion batteries and electrochemical capacitors are introduced. Non-faradic processes (electric double-layer capacitance) and faradic reactions (pseu-

docapacitance and intercalation) are generally explained. Electrochemical performance based on different types of electrolytes is briefly reviewed. Furthermore, impedance behavior based on Nyquist plots is discussed. We demonstrate the influence of cell conductivity, electrode/electrolyte interface, and ion diffusion on impedance performance. We illustrate that relaxation time, which is closely related to ion diffusion, can be extracted from Nyquist plots and compared between lithiumion batteries and electrochemical capacitors. Finally, recent progress in the design of anodes for lithium-ion batteries, electrochemical capacitors, and their hybrid devices based on carbonaceous materials are reviewed. Challenges and future perspectives are further discussed.

1. Introduction With a fast-growing economy and human population, our global energy consumption has been dramatically increased.[1] Thus, the issue of the sustainability of energy supplies has attracted worldwide concern owing to a crisis in the rapid depletion of fossil energy resources along with serious environmental pollution issues.[2, 3] It is well known that solar cells and windmills not only exhibit limited application scopes but also require energy-storage systems because of their remote locations.[4] In addition, a rapidly developing market for portable electronic devices and hybrid electrical vehicles has also led to an urgent demand for mature energy-storage systems in modern society. Systems for electrochemical energy storage convert chemical energy into electrical energy. To value the energy content of a system, terms of “energy density” (or “specific energy”) and “power density” (or “specific power”) are used. “Energy density” is expressed in watt-hours per liter (W h L¢1) or in watt-hours per kilogram (W h kg¢1) and “power density” is expressed in watts per liter (W L¢1) or in watts per kilogram (W kg¢1).[5] To compare the performance of various energy-storage devices, a reprehensive chart known as the Ragone plot is shown in Figure 1. In such a plot, the specific energy is plotted versus specific power.[2] Among various energy-storage devices, it is clear that electrochemical capacitors (ECs) and supercapacitors can be considered as high power density systems with relatively low energy density, whereas lithium-ion batteries

[a] Dr. F. Yao,+ D. T. Pham,+ Prof. Y. H. Lee Center for Integrated Nanostructure Physics, Institute for Basic Science Sungkyunkwan University Suwon 440-746 (Republic of Korea) E-mail: [email protected]

[b] D. T. Pham,+ Prof. Y. H. Lee Department of Energy Science, Department of Physics Sungkyunkwan University Suwon 440-746, Republic of Korea) [+] These authors contributed equally to this work. This publication is part of a Special Issue on “Sustainable Chemistry at Sungkyunkwan University”. To view the complete issue, visit: http://onlinelibrary.wiley.com/doi/10.1002/cssc.v8.14/issuetoc.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Figure 1. Specific power against specific energy, also called a Ragone plot, for various electrical energy-storage devices. Reprinted with permission from Ref. [2] Copyright 2008, Macmillan Publishers, Ltd.

(LIBs) alone with high energy density show relatively low power characteristics. LIBs and ECs share some common features. They both consist of two electrodes that are in contact with the electrolyte. Requirements for electron and ion conduction in electrodes and electrolyte are valid for both systems. Furthermore, electron and ion transport are separated during the charge/discharge processes.[5] On the other hand, historically, differences between LIBs and ECs do exist. For instance, compared to traditional ECs, the electrodes of which are composed of the same materials (activated carbons) and therefore exhibit the same electrochemical potential, an inherent potential difference exists between the electrode materials in LIBs. The potential difference between the two electrodes in LIBs, for example, a graphite anode and a lithium cobalt oxide cathode, is ideally

2285


Reviews constant through discharge or charge, but the working voltage of ECs linearly declines with the extent of charge.[6] Furthermore, charge storage takes place by a faradic reaction (redox reactions) at the electrode in the case of LIBs. This is a diffusion-controlled slow-rate process and is also the origin of the low power density of LIBs. However, in the case of ECs, charges mainly form at the interface of the electrode and electrolyte by means of a non-faradic reaction through the formation of electrical double layers. Although this process exhibits a fast rate of ion diffusion to give rise to higher power density, the confinement of charges on the surface of the electrode leads to a low energy density for ECs. Nevertheless, it should be noted that these differences between LIBs and ECs are becomYoung Hee Lee is a director of the Center for Integrated Nanostructure Physics, Institute for Basic Science. He is also a professor in the Department of Energy Science and Department of Physics at Sungkyunkwan University, Korea. He received his BSc degree in physics from Chonbuk National University, Korea, and his PhD degree in physics from Kent State University, USA. His research interests include exploration of unprecedented physical and chemical properties of 2 D layered materials and carbon-based materials and their applications to electronic devices and energy storage. Fei Yao received her PhD degree in energy science from Sungkyunkwan University, Korea, and her second PhD degree in physics from Ecole Polytechnique, France, in 2013. She continues her postdoctoral research at the Center for Integrated Nanostructure Physics, Sungkyunkwan University. Her current research projects include the design of new electrode materials for Li-ion batteries, carbon-based materials as catalysts for the oxygen reduction reaction, and novel materials for electronic devices. Duy Tho Pham is a PhD candidate in the Department of Energy Science at Sungkyunkwan University, Korea. He received his BSc degree in material science from Hanoi University of Science and Technology, Vietnam, in 2010. His research involves the synthesis of nanomaterials and the fabrication of energy-storage devices mainly focused on carbon-based materials.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

ing less clear because of the intervention of innovative concepts such as asymmetric supercapacitors and Li-ion hybrid supercapacitors. Recently, LIBs and ECs have attracted attention from both industry and academia owing to the emergence of nanomaterials and nanotechnologies. Carbon-based materials including carbon nanotubes (CNTs), carbon nanofibers (CNFs), graphene, and their composites have been widely studied as electrode materials for both devices. Compared to traditional electrode materials, for example, graphite or activated carbons (ACs), these new types of carbon materials exhibit differences not only in dimensionality and morphology but also in the distribution of chemical bonding, which allows mixtures of local electronic structures between sp2 and sp3.[7] Therefore, the carrier-transport properties are different from classic carbon materials if they are in contact with the reactants. Novel carbon materials with high accessible surface areas and short diffusion lengths for ions open new perspectives for high energy and high power density devices. In this Review, on the basis of work performed in our group on batteries and supercapacitors with carbon-based materials, we will first provide a brief description of the fundamentals of electrochemical energy-storage devices. Storage mechanisms, influence of electrolyte, impedance behavior, and effect of the active mass loading/thickness are discussed. Furthermore, electrode performance based on different carbon materials will also be reviewed. Finally, general conclusions and perspectives are given.

2. Fundamentals 2.1. Storage mechanism In storing charges through an electrochemical reaction, there are basically two charge-storage mechanisms that exist: a nonfaradic process and a faradic reaction. The non-faradic process involves no charge transfer but only electrostatic force between charged ions and the electrode. This is mostly observed in electric double-layer capacitance (EDLC) in ECs, for which the cations and anions in the electrolyte are attached on both electrodes with an applied voltage to form two ideal double charged layers. No electrochemical reaction is involved during ion adsorption (charging). Electrons are released through an outer circuit to generate electricity during ion desorption (discharging) back to the electrolyte. For the faradic process, the major mechanism for carbon materials can be classified into pseudocapacitance, which occurs at the surface of the materials, and intercalation, which normally refers to a bulk reaction. Some other mechanisms still exist, for instance, alloying with Si-based materials and conversion for some transition-metal compounds.[5, 7] However, these mechanisms have been wildly reviewed and are beyond the scope of this work. 2.1.1. Non-faradic reaction (electric double-layer capacitance) Conventional dielectric capacitors store energy by the accumulation of charges on two parallel metal plate electrodes through a dielectric layer under an applied voltage (see Fig-

2286


Reviews of the electrode are accessible to the electrolyte ions.[4] Optimized pore-size distribution and surface wettability are necessary to enhance the accessible surface area, which increases the EDLC.

2.1.2. Faradic reaction 2.1.2.1. Pseudocapacitance Figure 2. Schematic illustration of a) a conventional dielectric capacitor and b) an electrochemical capacitor.

ure 2 a). The capacitance (C) of such a capacitor is expressed by [Eq. (1)]: C¼

Ae0 er d

Pseudocapacitance was first investigated by Conway and coauthors in the 1960s.[6] Different from EDLC, pseudocapacitance stores charges by means of a fast and reversible faradic redox reaction at the electrode surfaces, as shown in Figure 3 a.[6] If a potential is applied across the electrode/electrolyte interface, charge transfer takes place, similar to the electro-

ð1Þ

in which er is the dielectric constant, e0 is the permittivity of vacuum, A is the area of one metal plate, and d is the separation distance between the two electrodes. This capacitance is limited because of the small charge storage area (A) and the large separation distance (d on the micrometer scale). In contrast, ECs based on EDLC can store much more energy because of the large surface area of porous materials, for which a doubley charged layer is established on a porous surface of each electrode, and this gives rise to a charge separation distance of the order of 1 nm or less.[3] The concept of EDLC was first reported by von Helmholtz in the 19th century and was then further modified by Gouy, Chapman, and Stern.[3, 6] EDLC is established by storing charges in the double layer formed at an electrode/electrolyte interface upon application of an electric field.[8] The accumulation of charges is a non-faradic process; no electron transfer takes place through the electrode interface. During charging, electrons move from the positive electrode to the negative electrode through an external circuit. Whereas in the electrolyte, anions move toward the positive electrode and the cations move toward the negative electrode. The reverse process happens during discharging. This charge-storage process is completely reversible, which leads to excellent cycling stability and power density of the EDL capacitors. The most common electrode materials utilized for EDLC are based on carbons, such as ACs, CNTs, and graphene.[3, 9, 10] The capacitance of an EDL capacitor is also estimated by Equation (1),[6] but in this case, er is the dielectric constant of the electrolyte, A is the accessible surface area of the electrode, and d is the effective thickness of the EDL (or Debye length), which is the separation distance between the electrode surface and the electrolyte ion layer (see Figure 2 b). There is clearly a relationship between capacitance (C) and accessible surface area (A). Notably, the accessible surface area is different from the specific surface area (SSA) measured by the Brunauer– Emmet–Teller (BET) method. There is no linear relationship between capacitance and BET surface area, because not all pores ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Figure 3. Schematic illustration of different storage mechanisms. Faradic reactions include a) pseudocapacitance and c) intercalation. b) The non-faradic process is represented by EDLC. Combing either a/b) or b/c) forms d) an asymmetric supercapacitor or e) a Li-ion supercapacitor.

chemical process in batteries. The pseudocapacitance is derived from [Eq. (2)]: C¼

dðDqÞ dðDV Þ

ð2Þ

in which Dq is the charge acceptance and DV is the change in potential. Commonly active materials for such redox reactions include several transition-metal oxides (e.g., RuO2, MnO2, TiO2, NiO, V2O5, ZnO, WO3, Co3O4, and Fe2O3), conducting polymers [e.g., polyaniline (PANi), polypyrrole (PPy), and polythiophene (PTh)], and surface functional groups in carbons (e.g., oxygen and nitrogen).[9, 11, 12] Although the pseudocapacitance can be 10–100 times higher than the EDLC,[13] it suffers from low power density and poor cycling stability owing to the slow faradic process and the naturally poor electrical conductivity of the electrode materials.

2287


Reviews 2.1.2.2. Intercalation

2.2. Electrolyte

Intercalation is a simple, solid-state redox reaction in which ions are inserted into host materials. During intercalation, mobile guest ions such as Li + , Na + , and K + in the electrolyte are inserted into a solid layered material but the primary structural features are maintained with only a minimal amount of volume expansion.[14] Apart from protons, the smallest ion, Li + , is considered to be one of the best ion candidates for the intercalation reaction. In electrochemical storage, the concept of “intercalation” was first applied to cathode materials for lithium-ion batteries by Whittingham in the 1970s.[15] Transitionmetal oxides and chalcogenides with stable layered or tunnel structures were adopted as host materials. The redox reaction takes place only on the host lattice, whereas no faradaic changes occur in the guest ions.[16] Later on, carbonaceous materials were widely used as anodes in lithium-ion batteries by forming graphitic intercalation compounds (GICs) with a wellknown configuration of LixCn. During the charging process, positively charged lithium ions intercalate into negatively charged graphite layers; the ions maintain their ionic states rather than adopting their metallic states, as shown in Figure 3 c. The electrostatic force between the Li ions and graphite can provide the energy to overcome van der Waals forces between the graphene layers and thus the layer distance in graphite expands. In an extreme case of intercalation, the complete separation of graphene layers, so-called exfoliation, can occur. This kind of technique has been used in the exfoliation of graphite and other layered structures.

The electrolyte, as one of the most important components of the cell, participates in all reactions inside the electrochemical device. The role of the electrolyte is to serve as the medium for ion transfer between the electrodes. Also, the interfaces between the electrolyte and the two electrodes closely influence the performance of the cells. In particular, in LIBs, both the anode and the cathode materials are not stable with respect to the electrolyte solution. Therefore, formation of a protective layer at the solid (electrodes) electrolyte interface (SEI) through electrolyte decomposition at the onset of cell operation is considered one of the most effective ways to protect the electrode materials and to prevent further severe electrolyte decomposition.[26, 27] Nevertheless, the formation of a SEI consumes Li ions from the electrolyte, and therefore, it is detrimental to cell performance, especially in the first charge process. Thus, to reduce this irreversible process, SEI formation needs to be minimized. As a result, the choice of the electrolyte components directly affects the formation of a Li + -conducting SEI layer and, thus, the irreversible capacity and cycle life of a battery. The stability of the electrolyte can be described by the voltage range beyond which the electrolyte will be either oxidized or reduced, and this is known as the electrochemical window. The energy density of a device can be described according to the following equations [Eq. (3)]:[2]

2.1.3. Hybrid structure To achieve high energy and high power density for energystorage devices, hybrid devices that combine the fast charge/ discharge and long cycle life of EDLC with the high chargestorage capacity of either pseudocapacitance or intercalation (a few reports also outline the use of alloy/conversion materials) have attracted much attention over the past few years.[9, 17–20] There are two different types of hybrid devices: asymmetric supercapacitors and Li-ion supercapacitors (Figure 3). An asymmetric supercapacitor combines a pseudocapacitive electrode with an EDLC one (see Figure 3 d). Note that the device with both pseudocapacitive electrodes but different materials for each is also defined as an asymmetric supercapacitor.[21–25] Li-ion supercapacitors comprise a Li insertion anode and an EDLC cathode, as indicated in Figure 3 e. Because they operate through different electrochemical-storage mechanisms and in different electrode-potential ranges, the hybrid devices can perform over a wide potential window, which leads to an increase in the energy density while maintaining a high power density. Increasing the potential window is critical to increase the energy density, as the energy density is proportional to the square of the voltage. The design and optimization of appropriate electrode materials are key issues to obtain high electrochemical performance of the hybrid energy-storage devices. ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

E ¼ QV

or

1 E ¼ CV 2 2

ð3Þ

in which E is the energy density, V is the operation voltage, Q is the capacity (mAh g¢1 or mAh cm¢3) of the battery, and C is the capacitance (F g¢1 or F cm¢3) of the EC. Therefore, the electrochemical window of the electrolyte is closely related to the energy density of the storage device. In addition, the electrolyte not only affects the energy density of the cell but also the rate of mass flow or ion conductivity, which is related to the power density of the device.[16] Thus, choosing the electrolyte wisely is one of the most important steps for constructing a successful electrochemical energy-storage cell. An electrolyte must exhibit high electrochemical stability and high ionic conductivity; be composed of small solvated ions; have low viscosity, low toxicity, and high purity; and be low costing. In general, three types of electrolytes are widely used: aqueous electrolytes, organic electrolytes, and ionic liquids (ILs). 2.2.1. Aqueous electrolytes Aqueous electrolytes present several advantages, including low viscosity, small solvated ion size, low price, and an ionic conductivity that is two orders of magnitude higher than that of non-aqueous electrolytes.[28, 29] More importantly, aqueous electrolytes are environmentally friendly and nonexplosive, which is practically very important. Electrochemical cells containing aqueous electrolytes usually do not require the strict assembly conditions that are needed for cells containing non-

2288


Reviews aqueous electrolytes, as it is not necessary to prevent the invasion of water vapor; this reduces costs and makes electrochemical cells containing aqueous electrolytes adaptable for many practical applications.[30] The main disadvantage of aqueous electrolytes is their narrow electrochemical potential window, which is only approximately 1.23 V as a result of the decomposition of water. This fact limits the cell voltage and, therefore, the energy density of the cell. The operational voltage can be increased by using special structural designs, which require more careful investigation.[31] Aqueous electrolytes have been widely used in ECs and batteries. In the case of ECs, KOH, H2SO4, and Na2SO4 are commonly used in aqueous-based ECs with normally higher capacitance and power than ECs with organic electrolytes and ILs. In the case of rechargeable batteries, it is well known that KOH and H2SO4 electrolytes are also used in commercialized rechargeable aqueous batteries, such as nickel metal hydride (NiMH), nickel–cadmium (Ni-Cd), and lead-acid (Pb-acid) batteries.[31] These early-stage aqueous batteries suffer from a low cell working voltage, low energy density, and also poor cycling stability because of pulverization of their structure or an irreversible reaction of the electrode materials (dissolution/deposition of Pb or Cd).[28, 29] Recently, an aqueous solution containing a Li-containing solvent, such as LiNO3 and Li2SO4, has been adopted to develop aqueous rechargeable lithium-ion batteries after the initial report from Sony in 1994.[32–38] A mixture of LiOH/LiCl in water is considered a typical aqueous electrolyte for Li-O2 batteries.[39, 40] In addition to lithium-ion-based aqueous electrolytes, sodium-ion- and potassium-ion-based aqueous electrolytes have also been used in aqueous rechargeable sodium and potassium batteries.[41, 42] Many up-to-date Reviews on these new types of aqueous electrolyte batteries can be found; this topic is also beyond the scope of our Review and, therefore, will not be elaborated in detail herein.[28, 30, 43, 44] Although significant progress has recently been achieved with regard to these aqueous rechargeable devices, they are still in their infancy and will require more comprehensive examination in the future. 2.2.2. Organic electrolytes In accordance with requests to increase the operation potential window of aqueous electrolytes, organic electrolytes that normally provide an electrochemical window of 3 V are considered as alternatives to aqueous electrolytes.[5] Organic electrolytes usually consist of certain salts and non-aqueous compounds (solvents) that allow the salts to dissolve to sufficient concentrations. Thus, solvents with strong polar groups, such as C=O and C N, are preferred because of their high dielectric constants. An eligible organic electrolyte needs to maintain good electrochemical stability, especially to the surfaces of the electrodes. Furthermore, it should display low viscosity to allow facile ion transport. In ECs, tetraethylammonium tetrafluoroborate (TEABF4), tetraethylphosphonium tetrafluoroborate (TEPBF4), and ortriethylmethylammonium tetrafluoroborate (TEMABF4) are often used as organic salts.[9] Acetonitrile (AN) and propylene carbonate ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

(PC) are the most commonly used solvents. In LIBs, LiPF6/cyclic carbonates [e.g., ethylene carbonate (EC)]/linear carbonates [e.g., diethyl carbonate (DEC) and dimethyl carbonate (DMC)] is the most popular composition of the battery electrolyte.[31] LiPF6 exhibits reasonable conductivity, ionic mobility, and thermal stability relative to other lithium salts, and furthermore, its dissociation constant is also acceptable. As for ethylene carbonate, it shows a high dielectric constant and low viscosity. The most desirable property of ethylene carbonate compared to other solvents (i.e., PC) is that it forms an effective SEI layer on the graphite anode at the beginning of the charging process, and this prevents further vigorous degradation between the electrode and electrolyte. As a result, a stable electrochemical environment can be assured in the following reactions. Spectroscopic investigations have shown that the SEI is composed of two layers: an inorganic layer on the anode surface mainly containing Li2CO3 and LiF, which comes from decomposition of the cyclic carbonate solvent and the LiPF6 salt, and a second porous organic layer on top of the inorganic one. The total thickness of both layers ranges from 2 nm to several tens of nanometers.[27] Using linear carbonates as a cosolvent with ethylene carbonate can suppress the melting temperature, decrease the viscosity, and expand the electrochemical window of the electrolyte.[16] Although the application of an organic electrolyte allows the energy density of the device to be increased by widening the cell voltage, several drawbacks need to be considered. The solvated ion size is relatively large; therefore, ion transport is slow and, thus, the ionic conductivity in organic electrolytes is lower than that in aqueous electrolytes. Also, organic electrolytes are usually toxic and unstable in air, which leads to complicated cell fabrication and high production capital. In addition, it is of note that organic liquid carbonate electrolytes used in LIBs usually decompose at voltages less 5 V, and therefore, applications of high-voltage cathodes, for instance, LiNiPO4, with a voltage higher than 5 V versus Li/Li + are limited.[45] 2.2.3. Ionic liquids ILs are solvent-free, liquid forms of organic salts that have low melting temperatures (< 100 8C).[10] ILs have attracted great attention over the last decade as promising electrolytes for energy-storage devices. ILs display a particularly wide electrochemical potential window (up to 5.7 V).[11, 12] Such a wide voltage range not only benefits the energy density of the device but also provides better cathodic stability relative to that provided by organic electrolytes, especially for high-voltage cathode materials, as already mentioned. Furthermore, ILs are nonvolatile and nonflammable, which are very attractive properties from a battery safety point of view. In addition, their low vapor pressure and high thermal stability are also favorable features for electrolytes, especially for batteries that operate at high temperatures. Nevertheless, even though ILs are entirely composed of ions, they are not perfectly dissociated molten salts and still suffer from certain degrees of ion associations.[46] Therefore, ion mobility is severely hindered by both the size and charge of the

2289


Reviews cluster.[47] Furthermore, only a limited family of anions can produce a liquid at room temperature, which highly limits the commercial development of IL-based electrolytes.[48] Representative compositions of commonly used ILs in electrochemical storage systems consist of cations such as imidazolium and pyrrolidinium and anions such as tetrafluoroborate [BF4]¢ , bis(trifluoromethanesulfonyl)amide [NTf2]¢ , bis(fluorosulfonyl)imide [FSI]¢ , and hexafluorophosphate [PF6]¢ .[10, 49] In the case of LIBs, the issue is more complicated owing to dissolution of Li + into the bulk liquid and its influence on ion transport. A major concern of IL electrolytes in LIBs is whether they can form a protective SEI layer on the electrodes to prevent electrolyte decomposition. Recent research suggests that IL consisting of cyano group (C N) incorporated cations can enhance the interfacial properties of a Li + /IL system, therefore, to form a SEI layer.[48, 50–52] Similarly, small imide ion containing anions can increase the rate of charge transport and prevent irreversible intercalation of IL cations at graphitic carbon electrodes by providing a protective layer.[49, 51–54] To reduce viscosity and to increase ionic conductivity, organic solvents have been used as additives to dilute ILs. In the case of ECs, AN and PC are widely used, and an improvement in capacitance has been realized.[13, 14] In the case of LIBs, ethylene carbonate and vinylene carbonate have been tested, and a SEI can be formed from these organic additives in a way similar to that of classical organic electrolytes, as mentioned in Section 2.2.2.[51, 55] However, this approach reduces the operating voltage of the device because of the presence of organic solvents. Moreover, other problems such as toxicity and flammability may be encountered.[10]

tance, or intercalation) and usually involves diffusion-controlled reaction of the electrolyte ions. Electrochemical impedance spectroscopy (EIS) is used as a powerful technique to evaluate impedance. Briefly, EIS determines the cell impedance by applying a small ( 5 mV amplitude) alternating current signal at any constant direct current potential (preferably at the open-circuit voltage to minimize the direct currents), over a frequency range of approximately 10¢2 to 105 Hz. This is a rapid, easy, and nondestructive technique for cell examination. The most representative result obtained from EIS is the so-called Nyquist plot, which plots ¢Z’’ versus Z’. It normally consists of a depressed semicircle in the high frequency range and a tail in the low frequency range for both ECs and LIBs, as shown in Figure 4 a. The mathematical background of the shape of the spectra has been well explained in previous work.[4] Several special points that can be used to understand the dynamic behavior of the cell are

2.3. Impedance behavior One of the key factors to achieve a high power density and long cycling life for an energy-storage device is cell resistance. This resistance consists of electrode resistance, contact resistance between the electrode and electrolyte, ion-diffusion resistance in the electrolyte and through the porous structure, and any charge-transfer resistance. The lower the resistance, the better the performance. Given that a capacitive component always presents in an electrochemical system under an alternating current, alternating current impedance analysis is more appropriate to determine the power capability of the device.[4] The impedance characteristics define the voltage drop over the device if a current is applied and can be expressed as follows [Eq. (4)]: Z ðf Þ ¼

V ðf Þ ¼ Z 0 þ jZ 00 I ðf Þ

ð4Þ

in which j is the imaginary number, Z’ is the real part of the impedance that indicates the overall resistance of the cell, and Z’’ is the imaginary part that refers to capacitive behavior of the device.[6] Z’ includes resistance of the electrode and electrolyte, contact resistance, and any faradic resistance. Z’’ is related to the mechanism of charge storage (i.e., EDLC, pseudocapaciChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Figure 4. a) Nyquist plots of carbon-based materials for supercapacitors and lithium-ion batteries. The equivalent series resistance (ESR, only for the supercapacitor), series resistance (Rs), and charge-transfer resistance (Rct) are indicated. b) A typical electrical equivalent circuit often used in data fitting for which Cdl and Zw represent the double-layer capacitance and Warburg impedance related to the diffusion of ions, respectively. c) Relaxation time extraction from imaginary capacitance versus frequency plot.

2290


Reviews shown in Figure 4 a. The first point for which Z’’ = 0 comes from the series resistance (Rs) of the cell, which includes the resistance of the current collector, electrode, electrolyte, separator, and any contact resistance. The diameter of the semicircle indicates the emergence of a faradic reaction, which comes from the charge-transfer resistance (Rct) at the interface between the electrode and electrolyte or the porous structure of the electrode materials.[6] The tail that shows almost linear behavior implies a diffusion-controlled process of ions into the bulk material. In general, a smaller Rct and a steeper slope of the tail can be observed in aqueous electrolyte based ECs relative to that observed for organic electrolyte based LIBs, as can be seen in Figure 4 a. In the case of CNT-based symmetric ECs with a KOH electrolyte (Figure 4 a, &), the smaller Rct and the almost vertical line indicate a small charge-transfer resistance at the electrode/electrolyte interface and EDLC behavior with fast ion diffusion in the electrolyte, respectively. On the contrary, a CNFbased half-cell LIB with a LiPF6/EC/DMC electrolyte (Figure 4 a, *) displays a larger Rct and a smaller slope, which represent faradic behavior with a slow rate of Li + ion diffusion. Notably, the value of Z’ at the second point of Z’’ = 0 is also known as equivalent series resistance (ESR) in the case of ECs if the kinetic-controlled process is dominant (i.e., if the slope of the tail is near 908). The interpretation of the Nyquist plot normally relies on the construction of a suitable electrical equivalent circuit (EEC) to represent the electrochemical system and quantitative determination of the relevant electrical parameters. The most commonly used EEC model is based on Randles circuit, as shown in Figure 4 b. Nevertheless, in most cases, the Nyquist plot cannot be simulated unambiguously by a single model, which complicates the data interpretation process. Other useful information that can be extracted from EIS is the imaginary capacitance (C’’) versus frequency plot. This imaginary part of capacitance is calculated from Equation (5): C 00 ¼

Z0 wjZ j2

ð5Þ

Here, C = C’¢jC’’; C is the total effective capacitance, C’ is the real part component of C, and w is the angular frequency. C’’ reaches a maximum at a frequency f0, which yields a relaxation time of t = 1/f0 = 2p/w. The relaxation time, also called the RC time constant, is a measure of the time that is required to discharge 50 % of the total energy stored in the device.[56] This value is closely related to ion diffusion, as the time for electron transport in the electrode is rapid and negligible relative to that for ion transport in the device. A small value of t implies that the device can show high power (or rate capability). Figure 4 c displays typical examples of a C’’ versus frequency plot. The CNT-based symmetric EC exhibits a relaxation time of 1 s (Figure 4 c, &), which corresponds to a typically high response for EDLC. In contrast, the CNF-based half-cell LIB exhibits an extremely long relaxation time (Figure 4 c, *), which is beyond the frequency range of approximately 10¢2 to 105 Hz. In generChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

al, smaller values of Rs and Rct, a steeper slope of the tail, and a shorter relaxation time are preferred for fast ion diffusion and, therefore, high power capability of the device. Below, some important factors that affect these parameters will be introduced. 2.3.1. Cell conductivity The value of Rs is very sensitive to the contact resistance of all the components in the cell, such as the electrolyte and separator, and also to the conductivity of the active materials. For the electrolyte, an aqueous electrolyte containing relatively small ions and good ion conductivity is widely used in ECs and aqueous batteries to reduce the resistance and to improve the power density. A simple comparison was done to illustrate the effect of the cell components on Rs by using the same CNF anodes and the identical LiPF6/EC/DMC electrolyte. The only difference between these two half cells was that one of the cells consisted of a two-layer separator. The cell with the onelayer separator (Figure 4 a, *) shows a smaller Rs than the one with the bilayer separators (Figure 4 a, *). Among the characteristics of the different materials, the electrical conductivity of the specific materials is one of the most important factors in connection with the impedance and, therefore, the rate performance of the device. Improving the conductivity of the active material can reduce Rs. In the case of carbon materials, the electrical conductivity is significantly affected by two primary factors: intrinsic material structure and functionality.[57] Crystalline sp2 carbon atoms (e.g., CNTs and graphene) generally have higher electrical conductivity than amorphous carbons (e.g., ACs and CNFs). The higher surface area of carbon materials leads to poor conductivity.[56, 58] Oxygen functional groups often decrease electrical conductivity because they create sp3 hybridization and inhibit electron transport between the intrinsic crystal structures of the carbon materials.[57] Heat treatment at high temperature is necessary to reduce the functional groups or to increase the graphitic portion of the carbons, which thus improves electrical conductivity.[58, 59] Most of the active materials for pseudocapacitance (metal oxides/hydroxides, conducting polymers) display low intrinsic electrical conductivity. Therefore, adding highly conductive carbons into these materials provides a way to reduce Rs of the electrodes. This will be further discussed in Section 2.3.2. 2.3.2. Electrode/electrolyte interface As previously mentioned, Rct mostly occurs at the interface of the electrode and electrolyte. In the case of ECs, surface functionalities can enhance the wettability of carbonaceous electrodes, but they reduce electrical conductivity. This results in an improvement in electrolyte accessibility, which leads to a smaller value of Rct and efficient charge storage through the interface.[9, 60–63] In the case of LIBs, the situation of the electrode/electrolyte interface is more complicated owing to the formation of the SEI layer. Actually, the formation of the SEI layer can sometimes be observed through EIS (Figure 5). The common understand-

2291


Reviews

Figure 5. Schematic representation of a Nyquist plot containing SEI-related semicircles.

ing is that the small semicircle at higher frequencies is attributed to the presence of the SEI layer and the larger semicircle at lower frequencies is associated with Rct. However, the first semicircle is usually neglected if it is too small to be recognized, especially if the cell is newly fabricated.[64] Theoretically, the SEI layer in LIBs is supposed to be ionically conductive and should remain stable after the first few cycles. However, an SEI layer can continuously form at the freshly exposed surface of electrode materials that exhibit large volume expansion, for example, the Si electrode. As a result, Li-ion consumption from the electrolyte is more severe than that of carbon-based materials. Moreover, for an aged LIB, there is a possibility that the SEI layer will be weakened by a secondary reaction with HF and the formation of transition-metal clusters in the long run, which not only deteriorate the stability of the electrode/electrolyte interface but also trigger Li plating and severe safety problems.[65] Thus, a stable SEI layer that can prevent further electrolyte decomposition and protect the anode material from structural degradation induced by co-intercalation of the solvent is one of the key factors to reach satisfactory performance in LIBs.

contribute to the BET surface area the most. Micropores with diameters of 0.7–1 nm provide the best capacitive performance.[66, 67] However, structures containing only micropores are not desirable, because they restrict ion diffusion deep into the bulk active materials. Moreover, mesopores also contribute to the surface area and provide a pathway for the ions to access the internal micropores. Macropores have a negligible contribution to the surface area, but they can act as transport avenues to the interior of the carbon network.[68] An appropriate pore-size distribution is necessary to ensure a high accessible surface area and rapid ion diffusion thus to enhance both the energy and power density of carbon material based ECs. Ion diffusion is significantly improved if mesopores are dominant in the porous electrodes.[56] Microporous carbons exhibit high capacitance at low scan rates, whereas mesoporous carbons possess good capacitive performance at high rate measurement. Recently, our group demonstrated that vertically aligned MWCNTs (v-MWCNTs) outperform randomly entangled SWCNTs (re-SWCNT) in terms of rate capability.[56] The v-MWCNTs are mainly mesoporous in structure, whereas a major contribution of the surface area of re-SWCNTs comes from micropores. The micro-supercapacitor-based on vMWCNTs yields a relaxation time of 0.76 ms in aqueous Li2SO4 electrolyte and 2.23 ms in organic LiPF6 electrolyte. This is much smaller than the relaxation times of 17.7 and 35.96 ms for the re-SWCNT-based device (Figure 6).

2.3.3. Ion diffusion As mentioned above, the slope of the tail in the Nyquist plot can be different depending on the diffuFigure 6. Comparative complex capacitance plots for 10 mm long v-MWCNTs and sion-controlled reaction, which is mainly affected by 1.2 mm thick re-SWCNTs in a) Li SO and b) LiPF . c) Schematic of ion diffusion through 2 4 6 the size of the electrolyte ions and the porosity of v-MWCNT- and re-SWCNT-based microdevices. Reprinted with permission from Ref. [56]. the electrode. During the charge/discharge process, Copyright 2013, Macmillan Publishers, Ltd. electrolyte ions diffuse into or away from the electrode surface. This process induces a limitation on 2.4. Effect of mass loading/thickness ion transport in the energy-storage device. Therefore, designing electrode materials with reasonable surface areas and poNotably, the mass loading and thickness of an electrode materosities is very important to improve the ion-diffusion process, rial usually affect the performance of energy-storage cells. For especially in the case of ECs. Owing to the small size of the Li example, a small mass loading and a small thickness of the ion, porosity engineering in LIBs is not as important as in ECs. active material are preferred if investigating intrinsic charge Thus, the following content will focus on the influence of the storage. Nevertheless, the overall gravimetric energy and porosity for ECs. Porous carbon electrodes generally comprise power density may be overestimated.[69] A large mass loading three types of pores: micropores (diameter < 2 nm), mesopores of the active materials in the electrodes is necessary to obtain (diameter = 2–50 nm), and macropores (diameter > 50 nm).[4] high energy density in practice, but this increases the total difMicropores display a high surface area to volume ratio and fusion length, which sometimes leads to poor energy and ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2292


Reviews power density.[56] Therefore, it is necessary to indicate the loading amount or film thickness with mass density to compare the storage gravimetric energy and power performance precisely. Figure 7 indicates the effect of mass loading on supercapacitor performance. The data are reproduced from Li and co-

dried CCG films exhibit higher capacitances at a low current density of 0.1 A g¢1 than at a high current density of 1 A g¢1.

3. Carbon-Based Materials as Anodes in LIBs 3.1. Graphite as an anode in LIBs

Figure 7. Capacitance dependence on loading mass of active electrode materials. Reproduced with permission from Ref. [70]. Copyright 2013, American Association for the Advancement of Science.

Graphite, as a highly crystalline graphitic carbon, is a well-defined layered structure and is the most common anode in LIBs. The stacking order of graphene layers in graphite consists of AB (hexagonal graphite) and ABC (rhombohedra graphite). Two major changes in graphite occur from a structural point of view upon intercalation of Li ions into graphite layers: one, the stacking order of the graphene layers shifts to AA stacking; two, the interlayer distance between the graphene layers increases slightly ( 10.3 %) as a result of lithium intercalation, as shown in Figure 8 a, b. The maximum lithium content in graphite is one Li guest atom per six carbon host atoms (i.e., LiC6) at ambient pressure according to the following equation [Eq. (6)]:[72] 6 C þ x Liþ þ x e¢ ! Lix C6

ð5Þ

workers.[70] Two different electrodes based on chemically conin which x = 1 in LixC6. In LiC6, the Li ions avoid occupation of verted graphene (CCG) were employed. The H2SO4-mediated the nearest neighbor sites owing to Columbic repulsive forces, CCG (CCG/H2SO4) film with a mass density of 1.33 g cm¢3 was which yields a maximum Li-storage capacity of 372 mAh g¢1, as prepared by capillary pressure of the CCG hydrogel film indicated in the bottom panel of Figure 8 b. It is known that through controlled removal of a H2SO4 solution trapped in the the Li intercalation reaction occurs only at the edge plane of gel. The dried CCG film (mass density of 1.49 g cm¢3) was fabrigraphite. Through the basal plane, intercalation is possible cated directly by vaporizing water under vacuum inside the only at defect sites. The diffusion pathway of Li ions will be furCCG hydrogel film without exchanging any miscible solutions. ther discussed below.[73–77] The supercapacitors were assembled by using 1.0 m H2SO4 electrolyte and were tested at current densities of 0.1 and 1 A g¢1. It is clear that the capacitance decreases as the mass loading of the active materials increases with identical electrode materials and electrolytes used in the cell. This is evidence of ion-diffusion limitations at large material loadings. A similar phenomenon was also noticed in the case of LIBs with thicker films.[71] Besides, it is necessary to specify the current density or scan rate applied during the measurements to provide reasonable comparison between different devices. This is because larger amount of charges can be stored with smaller current densities owing Figure 8. a) Schematic illustration of graphite with AB stacking order; the layer distance ( 0.34 nm) is indicated. b) Stacking order of graphene layers shifts to AA stacking after Li intercalation with an increased interlayer disto a longer time for more reactance ( 0.37 nm). Li ions occupy the nearest neighbor sites, as shown in the bottom panel. c) Charge/discharge tions to occur.[69] As seen in profile of graphite carbon. Reproduced with permission from Ref. [14]. Copyright 1998, Wiley-VCH. d) Stage formaFigure 7, both CCG/H2SO4 and tion phenomenon corresponding to c); the stage indices are indicated. ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2293


Reviews An important feature of Li intercalation into graphite is “stage formation” (Figure 8 c). Stage formation implies stepwise formation of a periodic pattern of unoccupied graphitic layer gaps at a low concentration of Li.[14, 78–83] This stepwise process can be described by the stage index, s (s = I, II, III, IV), which is equal to the number of graphene layers between two nearest guest layers (Figure 8 d). Note that stage IV is not indicated in the figure, because the Li concentration is too low in the graphene layers. It is also known as a dilute stage if s > IV.[79, 81, 82] Stage formation can be easily observed in a charge/discharge profile in the form of a plateau by constant current measurements for a graphite anode. The plateaus indicate the coexistence of two phases. The formation of stages II, IIL (a transition stage between stage II and stage III), III, and IV were identified from experimental electrochemical curves[80, 81] and were confirmed by X-ray diffraction and Raman spectroscopy.[79, 83–85] Two factors determine the formation of stages during Li intercalation into graphite: one, the energy required to expand the van der Waals gap between two graphene layers; two, the repulsive interactions between the guest species. Therefore, compared to a random distribution of Li in the graphitic layers during the charge process, Li ions prefer first to occupy van der Waals gaps with a high Li ion density to reach an energetically stable state.[14] Ideally, Li + intercalation into graphite should be fully reversible, and the Li-storage capacity should not exceed 372 mAh g¢1 according to the LiC6 configuration. However, the charge accumulated in the first cycle is usually larger than the maximum theoretical specific capacity. Relative to the first charge, the first discharge capacity is much smaller. The excess amount of charge generated in the first cycle, which cannot be recovered, can be ascribed to the formation of a SEI layer. Decomposition of the electrolyte usually takes place at less than 1 V versus Li/Li + and appears as the first plateau in the charge curve (Figure 8 c).[14, 75, 80, 83] Because of the irreversible consumption of lithium and the electrolyte, a corresponding charge loss exists, so-called “irreversible specific charge” (Figure 8 c). Reversible lithium intercalation is called “reversible specific charge”.

3.2. Nanocarbons as anodes in LIBs The limited capacity of graphite has hindered the further development of battery technology. Researchers have been struggling for a long time to develop new materials and new structures to meet the ever-growing demands of the market. The emergence of nanoscience and nanotechnology, which led to a revolution in basic materials science and engineering, provided new opportunities to improve the performance of anode materials. Nanocarbon materials enable electrode reactions to occur that cannot take place for materials composed of micrometer-sized particles. The diffusion time constant (t) for Li ions is given by Equation (7): t¼

L2 D

ChemSusChem 2015, 8, 2284 – 2311

ð7Þ

www.chemsuschem.org

in which L is the diffusion length and D is the diffusion constant.[76] The reduced dimensions significantly increase the rate of lithium insertion/removal and electron transport because of the short distances for Li-ion transport within the particles.[86] High surface area permits a high contact area with the electrolyte and, hence, high Li-ion flux across the interface. The strain associated with intercalation is expected to be better accommodated in nanosized carbons. Owing to the advantages mentioned above, nanocarbon materials have been extensively investigated as anodes for LIBs. The discovery of nanoscale carbon materials includes CNTs, CNFs, and graphene, which had a profound impact on the development of clean energy storage and conversion systems. Low-dimensional carbons exhibit novel properties that are often superior to those of their bulk carbon counterparts, and this is associated with decreased size, unique shape, and defects. Therefore, the mechanism of Li storage and anodic behavior could be very different from bulk graphite. Compared to the charge/discharge curve of graphite, the phenomenon of stage formation cannot be observed. The special feature of nanocarbon anode materials is that they exhibit a larger voltage hysteresis between the charge and discharge processes than graphite, which resembles hard carbon materials.[14, 86, 87] Table 1 summarizes the anode performance with respect to the highlighted electrode materials published in the literature. 3.2.1. Carbon nanotubes as anode materials As an allotrope of graphite, CNTs are good candidates for lithium batteries because of their unique structure (1 D cylindrical tubule of a graphite sheet), high conductivity [106 S m¢1 at 300 K for single-walled CNTs (SWCNTs) and > 105 S m¢1 for MWCNTs], low density, high rigidity (Young’s modulus of the order of 1 TPa), and high tensile strength (up to 60 GPa).[88] All of these unique properties make them attractive candidates for the anodes of LIBs. Many theoretical studies related to the storage mechanisms of lithium ions in CNTs have been done in the past few years by using different calculation methods.[89–96] The main conclusions can be briefly summarized as follows: One, Li ions can be stored both outside the tube and inside the tube. The outer surface is more energetically favorable. Upon increasing the diameter of the tube, the adsorption energies of both the external and internal sites change.[89, 92] Two, Li diffusion through the sidewall of CNTs is forbidden, but Li ions can enter tubes through topological defects containing at least nine-membered rings or through the ends of open-ended nanotubes.[90, 92, 94, 97] Three, Li ions can intercalate at the interstitial spaces between nanotubes.[92, 95, 96] Experimentally, the capacity of raw CNTs varies significantly depending on their structures and morphologies. In general, SWCNTs display a capacity range from 300 to 600 mAh g¢1 and MWCNTs exhibit a capacity range of approximately 450 to 600 mAh g¢1. Kawasaki et al. have reported that the reversible Li-ion storage capacity of metallic SWCNTs is five times higher than that of semiconducting SWCNTs. They attribute the origin of the capacity difference to the difference in the Li-ion ad-

2294


Reviews Table 1. Performance summary of highlighted nanocarbon-based anode materials. Material[a]

Loading mass/thickness

d[b] [S cm¢1]

SSA [m2 g¢1]

Current density [mA g¢1]

Capacity [mAh g¢1]@cycles number

Columbic efficiency [%]

Ref.

CNTs VA-CNTs drilled CNTs CNx-NTs porous CNFs mesop-CNFs activated CNFs hollow CNFs PAN-CNF-550 8C AHTGCNFs NPCNFs Gr-300 8C Gr-1050 8C holey Gr N-doped Gr N-C sponges

3–5 mg[c] – 0.3 mg 0.45–0.55 mg cm¢2 – – 3 mg cm¢2 1.02 mg 150 mm – – 0.3 mm – ¢0.2–0.3 mg cm¢2 0.6 mg –

– – –

– – 314 – 91.8 872 167.9 20.35 – 538 2381 184 492.5 15 70 613

25 253 25 37 50 100 100 100 500 100 2000 50 100 200 5000 500

447@50 750@34 625@20 312@30 454@10 1132@100 512@10 390@10 333@500 1346@40 943@600 854@15 848@40 261@400 500@3000 870@300

50 54 40 71 52 51 44 53 65 58 36 40 62 76 52 38

[102] [98] [108] [117] [128] [134] [132] [133] [137] [138] [140] [143] [151] [154] [145] [112]

– – – – 7.35 Õ 10¢8 – 4.9 – – 1.0 – –

[a] VA-CNTs = vertically aligned carbon nanotubes, CNx-NTs = nitrogen-doped carbon nanotubes, mesoP-CNFs = mesoporous CNFs, AHTGCNFs = activated N-doped hollow-tunneled graphitic carbon nanofibers, NPCNFs = nitrogen-doped porous CNFs. [b] d = Electrical conductivity. [c] The mass is the total mass of the active material and the metal substrate.

sorption potential.[93] Jaber-Ansari et al. have observed a similar phenomenon. Furthermore, they have noticed that metallic SWCNTs display higher capacity only if they are not bundled. On the contrary, for bundled SWCNTs, the capacity of semiconducting SWCNTs can reach a level comparable to that of metallic SWCNTs.[91] Studies suggest that aligned CNTs can allow better contact with the current collector and increase ion diffusivity to significantly improve the bulk electron-transport properties, which thereby allows improved rate capabilities.[98–100] Furthermore, CNTs with shorter lengths, which can be produced under proper growth conditions, by ball milling or by solid-state cutting [NiO or FeS etching during chemical vapor deposition (CVD) growth], have been reported to display better performance.[101–106] This is because Li diffusion and intercalation/de-intercalation are limited through longer CNTs. Opening the cap of CNTs or creating defective sites on the side walls through gas/liquid phase chemical etching or ball milling can further improve the capacity.[104–110] A comprehensive Review regarding the influence of the morphology of CNTs on anode performance can be found in Ref. [106] In addition to the morphology engineering of CNTs, doping CNTs with heteroatoms has been demonstrated to be an effective way to improve the electrochemical performance of CNT anodes. Among different heteroatoms, N and B, the atomic sizes of which are close to that of carbon, are the preferred candidates.[111, 112] Most reports regarding doping of CNTs are based on N doping. It is well known that N contains five valence electrons for bonding with carbon atoms. Owing to the higher electronegativity of N, it will withdraw electrons from C atoms and, therefore, change the electronic properties of CNTs. Both theoretical studies and experimental work demonstrate that N doping can increase the conductivity and reactivity of CNTs.[111–115] Thus, the diffusion and storage mechanism of Li in CNTs may vary.[111, 112] It has been reported that graphitic N and pyridinic N can play a role in improving capacity.[115, 116] Recent ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

work has shown that N-doped CNT electrodes normally display higher reversible capacity and rate capability than raw CNT anodes.[111, 115, 117–119] The improved performance can be attributed to structural defects that enhance Li-ion absorption and ion diffusion. To be more specific, it is suggested that the Ndoping process generates a disordered carbon structure with extrinsic defects, and this enhances the Li-intercalation properties. Moreover, N doping can improve reactivity and electrical conductivity, which leads to increased active sites to absorb Li ions, and this can enhance capacity. N-Doped CNTs exhibiting better electrolyte wettability and a larger interlayer distance between the tube layers can also be effective factors to improve electrochemical storage performance.[118, 120] Storage mechanisms in CNTs as described above are now well accepted. Nevertheless, there are still several issues that need to be addressed before CNTs can be considered as an electrode material for industrial use. Precise control of their structural morphologies such as the number of walls, diameter, length, and metallicity is required to improve reproducibility of the data. Furthermore, electrode fabrication processes such as purification, stable dispersion, and post-treatment should also be refined. Furthermore, the high irreversible capacity, limited lithium-storage capacity, and large hysteresis of the voltage window of CNTs still hinder their use as replacements for graphite-based anodes. It might be desirable to combine CNTs with other non-carbon materials, and this will be described in Section 3.2.4. 3.2.2. Carbon nanofibers as anode materials As a very similar carbon family member to CNTs, CNFs can be prepared mainly through two methods. One is catalytic CVD, and the other is electrospinning followed by annealing.[121] Compared to CNFs prepared by the CVD method, CNFs produced by electrospinning display noticeable advantages in

2295


Reviews that the films are free-standing in nature, they are easy to fabricate, and are low costing. Compared to CNTs and graphene, CNFs as anodes in LIBs have been less reviewed. Therefore, we will review CNFs in more detail in this part. A polyacrylonitrile [PAN, (C3H3N)n]-derived CNF web (PAN-CNF) has been used as an anode material.[122] The PAN-CNF web shows a reversible discharge capacity (450 mAh g¢1) after annealing at 1000 8C that is slightly higher than that shown by graphite at a current density of 30 mA g¢1. However, the coulombic efficiency is rather low owing to the high irreversible capacity (500 mAh g¢1) induced by the formation of a SEI. To increase the surface area and to facilitate Li-ion diffusion, porous CNFs were brought into focus.[123–126] Ji and Zhang have fabricated a two-component nanofiber PAN/poly-l-acetic acid (PLLA) through electrospinning.[127] After postannealing, PLLA is eliminated during the carbonization of PAN, and therefore, porous CNFs are generated. According to EIS measurements, this type of porous CNF exhibits much faster charge transfer at the electrode–electrolyte interface and also more efficient lithium-ion diffusion than nonporous CNFs. The discharge capacity is approximately 435 mAh g¢1 after 50 cycles at a current density of 50 mA g¢1, which is better than that of nonporous CNFs. Nevertheless, the columbic efficiency is still low in the first cycle. The authors have also fabricated porous CNFs by using SiO2 as a pore generator and in situ CNF activation with ZnCl2.[128, 129] Porosity can also be created by KOH treatment on electrospun CNFs.[130–132] Hollow CNFs fabricated by the coaxial electrospinning technique have also been used as anode materials, as shown in Figure 9 a.[133] Furthermore, Xing et al. have made vertically aligned porous CNFs by using silica NPs as a pore generator in combination with AAO as a template during polymer carbonization (Figure 9 b).[134] After removal of the silica/AAO template, the as-sensitized porous CNF powder is mixed with a binder and then pressed onto a copper mesh. The porous CNFs show very high capacity of 1132 mAh g¢1 after 100 charge/discharge cycles at a current density of 100 mA g¢1.[134] So far, the reported capacity of pure CNF anodes normally lies within the range of approximately 300 to 500 mAh g¢1 depending on the structures and morphologies, which is not much better than that exhibited by graphite.[109, 122–129, 135] In accordance with doping CNTs as previously mentioned, N doping was introduced into CNF electrodes (N-CNF) to further improve the capacity of raw CNFs. The N content in the Ndoped carbon structure plays an important role in Li + -storage performance. CNFs with high N-doping levels exhibit high reversible capacity. The reason for the capacity improvement is similar to that in CNTs, as previously mentioned. Pyridinic N is preferable for Li storage, whereas graphitic N is not suitable.[136] Recently, Zhang et al. have found that the capacities of naturally N-doped PAN-CNF films (PAN-N-CNF) can be divided into two sources: Li + storage between the graphene layers (0.3 V) and reaction between the Li ions and the N-containing functional groups (1.5 V). They have demonstrated that high temperature treatment promotes graphitization and increases the electrical conductivity. However, the nitrogen content is also ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Figure 9. a) SEM image of hollow CNFs fabricated by coaxial electrospinning. Reproduced with permission rom Ref. [133]. Copyright 2012, Elsevier. b) SEM image of porous CNFs by using silica NPs as the pore generator. Reproduced with permission from Ref. [134]. Copyright 2014, American Chemical Society. TEM image of c) activated N-doped hollow-tunneled graphitic carbon nanofibers and d) the wall of the hollow structure. Reproduced with permission from Ref. [138]. Copyright 2014, Royal Society of Chemistry.

reduced. As a consequence, the PAN-N-CNFs obtained at high carbonization temperatures fail to deliver high capacities, although they may have a high Li-ion storage capability between the graphene layers. A better capacity value is obtained if both the graphitization and N content reach a reasonable value. The PAN-N-CNF films annealed at 550 8C with a N content of approximately 15 % show a capacity of 555 mAh g¢1 at a current density of 0.1 A g¢1 and 333 mAh g¢1 after 500 cycles at 0.5 A g¢1.[137] Chen et al. have successfully synthesized activated N-doped hollow-tunneled PAN-derived CNFs. This kind of unique structure shows a high degree of graphitization as a result of the catalytic effect of the Ni nanoparticles. Furthermore, Ni diffuses through the cracks on the wall of the CNFs, which induces the formation of hollow tunnels inside the fibers (see Figure 9 c, d). The activated N-doped hollow-tunneled PAN-derived CNFs exhibit a capacity of 1346 mAh g¢1 at a current density of 0.1 A g¢1 after 40 cycles and also very good rate ability.[138] Wang et al. have fabricated N-CNF webs by direct pyrolysis of polypyrrole (PPy) nanofibers (PPy-NCNF).[136] The PPy precursor is synthesized by a modified oxidative template assembly route.[139] The PPy-N-CNF shows a N content of 14 % and a capacity of approximately 605 (238) mAh g¢1 after 10 cycles at a rate of 0.1 (5) A g¢1. The initial columbic efficiency is 64 %. The same group further modified the N-CNF web anode through KOH treatment to increase the porosity and, thus, the Li + absorption sites.[140] The reversible capacity of this material is 633 mAh g¢1 after 1 cycle but it then gradually increases to 943 mAh g¢1 after 600 cycles at a current density of 0.1 A g¢1. The capacity still reaches 505 mAh g¢1 at 5 A g¢1 with an initial coulombic efficiency of 48.4 %. The lower coulombic efficiency relative to the previous one is due to the higher surface area, which induces more severe electrolyte decomposition.[140] Compared to CNTs, CNFs can be fabricated through electrospinning, which is suitable for mass production.

2296


Reviews In addition, CNFs also benefit from their cheap price. However, drawbacks such as high irreversible capacity, short cycle life, and lack of improvement in capacity are still challenging issues. Therefore, further extensive studies need to be done so that CNFs can outperform the graphite anode. 3.2.3. Graphene as an anode material

electrode material. Nevertheless, it is an attractive candidate for fundamental study of the mechanism of Li diffusion. In the case of graphite, one ambiguity in understanding the diffusion pathway of lithium ions is the coexistence of both an edge plane and a basal plane in the sample. Therefore, lithium-ion diffusion through the basal plane cannot exclusively be observed.[73–77] To obtain a comprehensive picture of the mechanism of lithium diffusion in LIBs, our group fabricated highquality single-layer graphene with a well-defined basal plane (Figure 10 a) and few-layer graphene with an enriched edge plane (Figure 10 b) through CVD, as indicated in Figure 10 c.[87] We found that Li-ion diffusion perpendicular to the basal plane of graphene is facilitated by defects, whereas diffusion parallel to the plane is limited by steric hindrance originating from aggregated Li ions adsorbed on the abundant defect sites (Figure 10 d). Furthermore, we noticed that the capacity of the graphene layers decreases continuously as the number of graphene layers increases (Figure 10 d). This indicates corrosion of the metal current collector and the protective nature of graphene. We predicted that the critical layer thickness (lc) to suffi-

Graphene composed of sp2 carbon atoms bonded into a 2 D sheet with the thickness of a single atom emerged and attracted broad attention in the field of energy storage. The good conductivity, high mechanical strength, high carrier mobility, and high surface area of graphene have made it an attractive electrode material for the anode of LIBs.[86, 141] The reported capacity of graphene anodes can easily reach over 1000 mAh g¢1, which is almost three times higher than that of graphite anodes.[141–146] The superb storage capacity is attributed to the unique interaction between the Li ions and the graphene layers besides the mechanisms of intercalation and Li adsorption in cavities or interstitial spaces, as mentioned above for CNTs.[86, 92, 95, 96, 141, 144] It is assumed that Li ions can be absorbed on both sides of graphene by forming a Li2C6 stoichiometry[147] or that they can be trapped at the benzene ring to form a covalent bond between two Li atoms to yield a LiC2 configuration.[148] To achieve high capacity and good rate capability, carefully engineered layer spacings between the reassembled graphene nanosheets, the defective sites, or nanopores on the graphene plane for Li diffusion need to be considered. Experimentally, pure graphene sheets reduced by different methods have been reported,[143, 149–151] and a capacity in the range of approximately 500 to 200 mAh g¢1 can be achieved.[143, 149–151] CNTs and C60 have been introduced to expand the layer distance in graphene.[152] The performance is improved in both cases. The anode capacity of C60-incorporated graphene increases up to 784 mAh g¢1 compared to 540 mAh g¢1 for pure gra- Figure 10. Optical micrographs of a) single-layer and b) multilayer graphene on a SiO2/Si phene sheets.[149] Furthermore, porous graphene or substrate. White dashed lines indicate wrinkles. Some portions of thicker graphene are holey graphene paper has also been demonstrated indicated by arrows. c) Schematic of i) single layer graphene with a well-defined basal plane and ii) edge plane enriched multilayer graphene. d) Schematics of the proposed to achieve high capacity and high rate capability for mechanism of Li diffusion through defects on the basal plane with different defect popuLIBs.[153–155] It has been suggested that graphene lations. Broad down arrows indicate Li-ion diffusion through defect sites of the basal edge defects and vacancies are desirable for improv- plane. Red glows represent steric hindrance for Li-ion diffusion formed by the accumulating the reversible capacity. Moreover, oxygen-con- ed Li ions or functional groups. The inset in the right indicates the relative magnitude of the diffusion coefficient. e) Related layer-dependent capacities. Two regimes of corrosiontaining functional groups that lead to the formation dominant and lithiation-dominant are indicated. Reproduced with permission from of a SEI layer are responsible for the loss of irreversi- Ref. [87]. Copyright 2012, American Chemical Society. ble capacity.[144, 146, 152, 156–160] In addition, similar to CNTs and CNFs, heteroatom doping has been widely studied to improve the performance of graphene ciently prohibit reaction of the substrate by using CVD-grown anodes.[112, 120, 161–170] Besides typical N doping, B, S, and P graphene layers is approximately six layers, independent of the doping, individually or as co-dopants, has also been reported defect population on the graphene layers (Figure 10 d). In addition, our DFT calculations show that divacancies and higher in the case of graphene electrodes.[160, 167–169] The improved performance is due to multiple factors, as explained in Secorder defects have reasonable diffusion barrier heights that tion 3.2.1 for N-doped CNTs. Related reports have been extenmay allow lithium diffusion through the basal plane but not sively reviewed.[118, 120, 159] through monovacancies or Stone–Wales defects.[87] Notably, Clearly, for lithium-ion storage, single-layer graphene sheets most of the work on graphene related to LIBs has been limited produced by CVD with the LiC6 configuration is not a promising to thin layers. Given that diffusion through the basal plane is ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2297


Reviews not possible, special care should be taken to prevent restacking of the graphene layers and to maintain facile ion diffusion without reducing the capacity and rate capability. 3.2.4. Carbon-based composites as anode materials Although anode performance can be improved relative to that of graphite by employing different nanocarbon materials or by engineering their structures, an improvement in the electrochemical storage of Li ions is still not satisfactory for applications in energy technologies. Fortunately, alloy/dealloy mechanism based materials (e.g., Si, Ge, Sn, and SnO2) and conversion mechanism based transition-metal compounds such as oxides, phosphides, sulfides, and nitrides (e.g., MxNy : M = Fe, Cu, Mn, Ni, Co, Ro, Mo; N = O, P, N, S) can display higher reversible capacities than intercalation-based carbonaceous materials.[26, 141, 171] Nevertheless, large capacity fading, poor cycling life induced by their relatively low conductivity, large volume change, and unstable SEI formation hinder real applications of these kinds of materials.[141, 172] An efficient way to improve the performance of anodes is to combine high-capacity materials with carbon-based materials by forming a composite. In such a hybrid system, the carbon-based material can provide good conductivity and also function as a strain buffer to confine volume changes of the host materials. This kind of composite anode has been widely reviewed.[26, 141, 171–173] Herein, we will take the carbon–silicon hybrid structure as a simple example to demonstrate the idea. Silicon as a high lithium storage capacity material (specific capacity of 3572 mAh g¢1 at room temperature, corresponding to Li15Si4) was recently proposed. Yet, the large volume expansion up to 400 % during charge/discharge causes severe structural pulverization, which makes this material impractical.[174, 175] For example, a Si thin film deposited on a metal substrate by chemical vapor deposition experiences the formation of cracks during cycling, and therefore, contact loss between the active Si material and the current collector takes place, which leads to poor cycling stability.[175] Owing to these difficulties, various carbon–silicon-based composite structures have been proposed.[71, 176–193] For instance, Fu et al. have fabricated a freestanding aligned Si-CNT sheet. The CNT sheet is drawn from CNT forests and rolled on a cylinder, and then Si is deposited afterward through CVD. An extra carbon layer is then coated on top of the Si layer to confine volume changes and also to create a stable SEI layer. The final structure is shown in Figure 11 a. The sample shows a reversible capacity of 1494 mAh g¢1 after 45 cycles with a capacity retention over 94 %.[183] Recently, our group fabricated a free-standing Sicoated CNF mat through combining an electrospun CNF mat with an electrodeposited Si layer.[71] The morphology and thickness of the Si layer can be tuned according to different electrochemical conditions. Figure 11 b shows a spaghetti-like Si layer with a reversible capacity of 730 mAh g¢1 after 50 cycles of charge/discharge.[71] Choi et al. have used a co-spinning method by adopting a dual nozzle and produced a core–shell Si–CNF structure (Figure 11 c). The as-synthesized fiber-type composite shows a capacity of 1384 mAh g¢1 with a good cyChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Figure 11. a) SEM image of CNT-Si-C sheet. Reproduced with permission rom Ref. [183]. Copyright 2013, Wiley-VCH. b) SEM image of electrochemically deposited Si on CNF. The cross-sectional images are shown in the insets.[71] Copyright 2014, Elsevier. c) A cross-sectional SEM image of Si nanoparticles wrapped by a carbon shell. Reproduced with permission from Ref. [189]. Copyright 2012, American Chemical Society. d) TEM image of Si nanoparticles with a graphene composite. Reproduced with permission from Ref. [193]. Copyright 2012, Wiley-VCH. e) SEM image of Si–carbon nanocables sandwiched between rGO sheets. Reproduced with permission from Ref. [192]. Copyright 2013, American Chemical Society. f) SEM image of a Al2O3–Si–carbon nanospheres composite. Reproduced with permission from Ref. [194]. Copyright 2014, Macmillan Publishers, Ltd.

cling life.[189] Si–graphene-based composites have also been widely studied.[190–193] Xiang et al. have prepared Si–graphene composites by using thermally reduced graphene oxide (rGO) and thermally expanded graphite. They found that the thermally expanded graphite with Si shows better performance owing to less structural defects. Zhou et al. have developed an electrostatic self-assembly method to produce a Si nanoparticle encapsulated graphene composite. The Si nanoparticles are uniformally dispersed between two layers of graphene (Figure 11 d).[193] Wang et al. have fabricated a Si nanowire encapsulated in overlapped graphene sheaths and reduced graphene oxide overcoats. The sample displays a capacity of 1600 mAh g¢1 with 80 % capacity retention after 100 cycles and superior rate capability (Figure 11 e).[192] Recently, our group fabricated a carbon nanosphere (CNS)/Si/Al2O3 core–shell structure and reported a reversible capacity of 1560 mAh g¢1 after 100 cycles at a current density of 1 A g¢1 (Figure 11 f).[194]

4. Carbon-Based Materials for ECs A variety of carbon materials, such as ACs, CNTs, carbon fibers, graphene, and carbon-based composites, have been employed as electrode materials in ECs. The development of next-generation supercapacitors with high energy and high power density as well as long cycling stability requires advanced carbon materials with a high accessible surface area, high electrical conductivity, appropriate pore size and pore volume, scalable production, safety, and low cost. Table 2 summarizes highlighted materials on improving the electrochemical performance of ECs recently published.

2298


ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2299

– – – – 0.75 0.49 – 0.5 – – – 0.6 0.94 0.048 0.4 0.75 – 1.25 1.06 1.12 – – – –

tea leaf based AC pomelo-based AC lignin-derived AC PPy-based AC SWCNT NA-CNT SWCNT film purest SWCNT PAN-based CNF N-doped porous CNF MWCNT-CMF//CNF CNT/graphene based CMF rGO LSG a-MEGO compressed a-MEGO PSS-GO EM-CCG ac-Gr/SWCNT nc-PDDA-Gr V2O5-CNF PANi-GO graphene/PANi PANi-rGO/CF

SSA [m2 g¢1] 2245–2841 2105 1148 3432 357 988 – 1300 705 562.5 – 396 630 1520 3100 707 – 167 652 663 700 – – –

Mass loading/ thickness

8 mg cm¢2 – 5–8 mg cm¢2 300 mm 150 mm 5–6 mg cm¢2 – 100 mm – 3.4 mg cm¢2 40 mm – 4.5 mg cm¢2 7.6 mm 4 mg cm¢2 4.2 mg cm¢2 45 mm 1–10 mg cm¢2 1.2 mg cm¢2 11.8 mm – – – – – – – – – – – 21 – – – 102 45 17.38 5 2.1 – 25 394 – 0.1 – – –

330 F g¢1@1 A g¢1 43.5 F g¢[email protected] A g¢1 102.3 F g¢1@1 mV s¢1 300 F g¢[email protected] A g¢1 180 F g¢1@1 mA cm¢2 98 F g¢1@1 A g¢1 140 F g¢1@20 mV s¢1 160 F g¢1@1 A g¢1 100 F g¢1@2 mV s¢1 202 F g¢1@1 A g¢1 6.3 mF cm¢1@2 mV s¢1 300 F cm¢[email protected] mA cm¢3 255 F g¢[email protected] A g¢1 276 F g¢1@5 A g¢1 166 F g¢[email protected] A g¢1 147 F g¢[email protected] A g¢1 190 F g¢[email protected] A g¢1 260 F cm¢[email protected] A g¢1 211 F cm¢[email protected] A g¢1 348 F cm¢[email protected] A cm¢3 214 F g¢1@2 mV s¢1 555 F g¢[email protected] A g¢1 640 F g¢[email protected] A g¢1 464 F g¢[email protected] A g¢1

Cs@scan rate 2 m KOH; 1 V 1 m NaNO3 ; 1,7 V 6 m KOH; 0.8 V EMIMBF4 ; 4.5 V 7.5 n KOH; 0.9 V BMIMBF4/AN; 3.5 V 1 m LiClO4/EC/DEC/DMC ; 3 V 1 m Et4NBF4/AN; 4 V 1 m Et4NBF4/AN; 2.3 V 6 m KOH; 1 V gel H3PO4/PVA; 1 V gel H3PO4/PVA; 0.9 V 6 m KOH; 1 V EMIMBF4 ; 4 V BMIMBF4/AN; 3.5 V BMIMBF4/AN; 3.5 V TEABF4/PC; 2.7 V EMIMBF4/AN; 3.5 V EMIMBF4 ; 4 V gel H2SO4/PVA; 1 V 2 m KCl; 1 V 1 m H2SO4 ; 1 V 1 m H2SO4 ; 1 V 1 m H2SO4 ; 1 V

Electrolyte[c] ; potential window

– 17.1 W h kg¢[email protected] kW kg¢1 8.34 W h kg¢1@– – 7 W h kg¢1@20 kW kg¢1 59 W h kg¢[email protected] kW kg¢1 43.7 W h kg¢[email protected] kW kg¢1 94 W h kg¢1@210 kW kg¢1 8 W h kg¢1@ 20 kW kg¢1 7.1 W h kg¢1@ 0.15 kW kg¢1 0.7 mW h cm¢[email protected] mW cm¢1 10.4 mW h cm¢3@ 1.8 W cm¢3 – 1.36 mW h cm¢3@ 1 W cm¢3 70 W h kg¢1@250 kW kg¢1 48 W h L¢1@ 17 kW L¢1 38 W h kg¢1@61 W kg¢1 88 W h L¢1@ 75 kW L¢1 117 W h L¢1@424 kW L¢1 6.7 mW h cm¢[email protected] W cm¢3 45 W h kg¢1@156 W kg¢1 – – –

Energy density@power density

2000 1100 – 10 000 – 10 000 – 1000 – 3000 1000 10 000 1200 10 000 10 000 – 14 860 5000 10 000 50 000 1000 2000 1000 1000

Cycles

92 90 – 108 – 91 – > 96 – – 90 93 93 97 97 – 88 95 98.2 92 80 92 90 89

CR[d] [%]

[208] [214] [215] [201] [234] [246] [247] [251] [255] [258] [263] [259] [282] [290] [294] [296] [310] [70] [311] [312] [328] [343] [316] [341]

Ref.

[a] NA-CNT = Nitrogen-doped and activated CNT, SWCNH = single-walled carbon nanohorn, LSG = laser-scribed graphene, a-MEGO = activated microwave exfoliated GO, PSS = polysodium 4-styrensulfonate, EMCCG = electrolyte-mediated chemically converted graphene, ac-Gr/SWCNT = activated graphene/SWCNT, nc-PDDA-Gr = nanochanneled poly(diallyldimethylammonium chloride)-mediated rGO. [b] d = Electrical conductivity; [c] EMIMBF4 = 1-Ethyl-3-methylimidazolium tetrafluoroborate, BMIMBF4 = 1-butyl-3-methylimidazolium tetrafluoroborate, Et4NBF4 = tetraethylammonium tetrafluoroborate, PVA = polyvinyl alcohol. [d] CR = Charge retention.

Mass density [g cm¢3]

4.1. Activated carbons for ECs

Electrode material[a]

d[b] [S cm¢1]

Activated carbons have been used as commercialized electrode materials for ECs because of their large SSA, relatively good electrical conductivity, high chemical/thermal stability, and low cost. ACs are usually synthesized from various types of initial carbon precursors (e.g., coal, pitch, coconut shells, bamboo, cellulose, and polymers) by physical and/or chemical activation.[195] In the case of physical activation, carbon precursors are carbonized under an inert atmosphere followed by activation in the temperature range of 600 to 1200 8C in the presence of oxidizing gases (e.g., steam, O2, and CO2).[196] Chemical activation is usually done by mixing carbonaceous materials with activating agents (e.g., KOH, NaOH, H3PO4, and ZnCl2) followed by heat treatment at 400–900 8C under an inert atmosphere.[196, 197] ACs produced by chemical activation usually have higher yields of the products and higher SSAs than ACs produced by physical activation.[196] Activating conditions affect the porosity and electrochemical performance of ACs significantly.[195, 198–200] A large content of microporosity (50–78 %) can be generated by CO2 physical activation. In contrast, a relatively low content (19–48 %) is achieved by steam activation.[9] During chemical activation, a high activation temperature, a long activation time, and a large mass ratio of the activation agent/precursor usually lead to an increase in the SSA, pore volume, and average pore size.[9, 63, 201–204] Carbon precursors also play an important role. Natural precursors (e.g., coal, pitch, petroleum coke, and biomass materials) are commonly used for the commercial production of ACs because of their abundance and low cost.[195, 203, 205–215] These ACs exhibit different SSAs in the range of 1000 to 3500 m2 g¢1, which results in a specific capacitance (Cs) of 150–350 F g¢1 in aqueous electrolytes and 100–200 F g¢1 in non-aqueous electrolytes. For instance, Peng et al., have reported a method for the preparation of ACs by high-temperature carbonization and KOH activation of waste tea leaves (Figure 12 a).[208] The obtained ACs exhibit high SSAs ranging from 2245 to 2841 m2 g¢1. ECs made with these ACs perform the maximum Cs of 330 F g¢1 at a current density of 1 A g¢1 in aqueous KOH electrolyte. However, these naturally

Table 2. Summary of the highlighted carbon-based materials for ECs.

Reviews


Reviews to the ionic liquid N-butyl-Nmethylpyrrolidiniumbis(trifluoromethanesulfonyl)imide (PYR14TFSI). Another important factor of ACs is surface functionalities (e.g., oxygen-, nitrogen-, phosphorus-, and sulfur-containing groups).[63, 115, 228–231] Surface functionalities not only enhance the wettability of the electrodes but also give additional pseudocapacitance. However, the presence of some active surface oxides and moisture in organic electrolytes leads to instability and a large resistance in ACs. They can also decompose organic electrolytes and ionic liquid.[68] Figure 12. a) Schematic illustration of the production of ACs by using waste tea leaves. Reproduced with permisTherefore, surface functionalities sion from Ref. [208]. Copyright 2012, Elsevier. b) SEM image at 700 8C and c) nonlocal density functional theory of ACs should be optimized to PSD of the PPy-derived ACs activated at three different temperatures. Reproduced with permission from improve the cycling life of ACRef. [201]. Copyright 2012, Wiley-VCH. based ECs. Although ACs are state-of-theart electrode materials for comderived ACs usually suffer from several drawbacks, such as low mercial ECs, their applications are still limited owing to their carbon purity and variations in their properties. Recently, varilow energy density and relatively low rate capability. The high ous polymers, such as polystyrene-based resins, polyfurfuryl alSSA of ACs cannot assure their high electrochemical percohols, phenol formaldehyde resins, polyaniline, and polypyrformance, as control of the PSD and pore structure is still chalrole (PPy), have attracted much attention as carbon precursors lenging. Therefore, producing ACs to achieve both a high SSA as a result of their commercial availability, uniform structure, and an optimized PSD with an interconnected pore structure, and high carbon purity.[201, 202, 216–219] Wei and co-workers have short pore length, and controlled surface chemistry is required presented a novel method to synthesize ACs based on the to improve the energy density while maintaining a high power direct KOH activation of PPy.[201] The as-produced ACs have density and a high cycling stability of AC-based ECs. a highly disordered porous carbon structure without graphite ribbons and crystalline impurities. An ultrahigh SSA up to 4.2. Carbon nanotubes for ECs 3400 m2 g¢1 with pore sizes in the range of 0.5 to 4 nm are obtained, which result in an outstanding Cs of approximately Over the last decades, CNTs have been intensively investigated for EC applications because of their unique internal structure, 300 F g¢1 in 1-ethyl-3-methylimidazolium tetrafluoroborate (EMIMBF4) ionic liquid at 60 8C (Figure 12 b, c). This Cs value is high electrical conductivity, mechanical strength, and chemical and thermal stability.[3, 232, 233] CNTs are generally used for high100–300 % higher than that of commercial carbons and other reported carbon-based materials for ECs. power electrode materials, as their energy density is still low The porosity of ACs is also an important factor influencing because of the limitation of surface area (usually the capacitive performance. Numerous studies have been > 500 m2 g¢1).[59, 60, 232–234] [58, 201, 220–224] done to increase the SSA. It is expected that with With the presence of mesopores coming from the central a higher SSA, more electrolyte ions can be stored at the solid canal and/or tube entanglement, CNTs have excellent access to electrolyte interface. However, there is no proportionality bethe electrolyte.[235] Our group demonstrated that most of the [4] tween the SSA (measured by BET) and Cs. The pore-size distriBET surface area of SWCNT electrodes synthesized by directbution (PSD) significantly affects Cs. A broad PSD with a wide current arc discharge contributes to the theoretically estimated range of average pore sizes (0.7–15 nm) may lead to a linear specific capacitance.[234] Although the SWCNT electrode has [225] constant of Cs. A narrow PSD with monodispersed pores a reasonable surface area (357 m2 g¢1), its intrinsic capacitance [226] can increase Cs and the energy density. The optimum pore can reach 50.4 mF cm¢2, which is much higher than that size depends on the size of the electrolyte ions. Recently, Pohlreached by ACs (< 10 mF cm¢2). This increased capacitance is mann et al. have demonstrated that the existence of a PSD due to the abundant PSD at lower pore sizes of 30–50 æ, and/or constrictions at the entrance of the pores leads to which is estimated from BET measurements. Heat treatment at significant changes in the Cs of ACs.[227] A total surface high temperature is necessary to increase Cs and to reduce the area with average micropore sizes above 1 nm is not accessible resistance of the SWCNT electrode. The electrode performs ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2300


Reviews with a maximum Cs of 180 F g¢1 and a measured power density of 20 kW kg¢1 at an energy density of 7 W h kg¢1 in 7.5 n KOH solution. Structural properties of CNTs such as diameter, length, electrical resistance, surface area, and doping play important roles in electrochemical performance.[236, 237] The capacitance of smaller diameter CNTs is usually higher than that of larger ones, which is due to the higher specific surface area in the network of smaller CNTs.[238] The low resistance of CNT electrodes ensures the high power of ECs. To reduce the contact resistance, CNTs can be grown directly on current collectors such as graphite[239] and Inconel alloy.[240] For example, Shaijumon et al. have synthesized arrays of CNTs and gold nanowire (AuNW) multisegmented hybrid structures through a combination of CVD and electrodeposition method.[241] Owing to the well-adhered interface between the CNTs and the AuNW segment, a very low ESR of 0.48 W is achieved for the CNT/AuNW electrode. This is much smaller than the large value of 3.4 W for the CNT-only electrode. This results in excellent electrochemical performance with a maximum power density of approximately 48 kW kg¢1. Increasing the SSA is a promising alternative way to improve the Cs and the energy density of CNTs. Several efforts directed at resolving this issue by using activation treatments (chemical and/or physical) have been reported.[242–245] Activation significantly enhances the SSA of the CNTs by opening their end tips and by introducing defects, and it also creates functional groups that contribute to the additional pseudocapacitance.[9] Recently, Yun et al. have shown that CNTs prepared by KOH activation and subsequent nitrogen doping possess a SSA of 988 m2 g¢1 with a hierarchical pore structure and rough surface, which are advantageous features for fast ion diffusion.[246] N doping changes the band structure of the CNTs, which results in improved electrical properties. ECs made with these N-doped activated CNTs exhibit a high energy density of 59 W h kg¢1 at a power density of 1750 W kg¢1 in 1-butyl-3-methylimidazolium tetrafluoroborate (BMIMBF4)/AN ionic liquid. CNTs have found potential applications in flexible ECs.[247–249] In these devices, CNT films act as highly conductive and flexible active materials, and they also increase the effective surface area in the electrodes, which maximizes the efficiency of thinfilm CNT based ECs. In addition, a direct comparison between aligned and entangled CNTs has been reported, and aligned electrodes are better than entangled CNTs in terms of rate capability.[56, 242, 250–252] Aligned CNTs possess relatively regular pore structures and conductive channels. This leads to a higher effective SSA, facilitates fast electron-ion transportation, and provides improved charge-storage and power properties, which is highly desirable for high-rate applications. As mentioned in Section 2.3.3, our group demonstrated that micro-supercapacitors based on v-MWCNTs (aligned CNTs) show better ion diffusion than devices based on re-SWCNTs (entangled CNTs, see Figure 6).[56] Despite excellent electrical conductivity and flexibility, CNTs still suffer from limited SSAs, difficult purification, and high production costs, and these are major obstacles precluding their practical application in high-energy performance ECs. ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Hence, much effort is still required to develop cost-effective and high-quality CNTs. 4.3. Carbon fibers for ECs Carbon fibers (CFs) are of great interest for EC applications, as they are inexpensive and massively synthesized by various methods (e.g., electrospinning and vapor growth) with variable fiber diameters, porosities, and surface chemistries.[11, 253–257] Similar to CNTs, CFs possess a high surface area to volume ratio, good electrical conductivity, and good flexibility. The pore size and SSA of CFs can be controlled through physical or chemical activation.[126] Our group reported the preparation of PAN-based CNF paper through electrospinning, followed by one-step carbonization/activation under a CO2 atmosphere at temperatures from 700 to 1000 8C.[255] ECs based on this CNF paper exhibit high power densities (up to 20 kW kg¢1) with insignificant degradation in energy densities ( 5–8 W h kg¢1) and outperform commercial AC-based ECs. The better performance can be attributed both to the high intrinsic conductivity of the CNFs and to the high diffusion rate of ions in the opened mesopores produced by CO2 activation. Recently, it has been demonstrated that N doping can improve electrical conductivity and wettability and, thus, the capacitance performance of CFs.[257–260] For instance, Yu’s group have presented a highly capacitive material based on N-doped porous CNFs synthesized by carbonization of macroscopic-scale CNFs coated with PPy at a temperature of 900 8C.[258] The resulting CNFs display a Cs of 202 F g¢1 at a current density of 1 A g¢1 in 6 m aqueous KOH electrolyte. This kind of N-doped CNF represents a promising candidate for an efficient EC electrode material. Owing to good electrical conductivity and mechanical properties, CFs have attracted increasing attention for wearable fiber-shaped ECs.[259, 261–265] Loading highly capacitive materials on CFs is an efficient approach to improve the electrochemical performance of devices. Recently, our group reported a coaxial fiber supercapacitor that was fabricated by using carbon microfiber (CMF) bundles coated with MWCNTs as the core electrode and CNF paper as the outer electrode (Figure 13).[263] The device exhibits a capacitance up to 6.3 mF cm¢1 (86.8 mF cm¢2) at a core electrode diameter of 230 mm and a measured energy density of 0.7 mW h cm¢1 (9.8 mW h cm¢2) at a power density of 13.7 mW cm¢1 (189.4 mW cm¢2) in H3PO4/polyvinyl alcohol (PVA) gel electrolyte. These results are much higher than those reported for previous studies on wearable fiber-shaped ECs. The device shows excellent cycling stability and negligible change in capacitance after 1808 of bending. The high electrochemical performance of the device has been attributed to a high effective surface area as a result of its coaxial configuration and also to the high electrical conductivity of the allcarbon electrode materials. Several challenges still remain to industrialize CFs as electrode materials for ECs. Free-standing webs of CFs are often brittle, which restricts their application to flexible and wearable devices. The large-scale production of CFs with precise control of fiber diameter, porosity, and distribution of the active com-

2301


Reviews exhibit a gravimetric capacitance of 255 F g¢1 and a volumetric capacitance of 196 F cm¢3 at a current density of 0.5 A g¢1.Capacitance retention of 93 % after 1200 cycles is obtained. Thus, these electrodes outperform previous rGO electrodes prepared by using hydrazine and NaBH4 as the reductants. Other than chemical reduction, rGO can be produced from GO by thermal annealing,[283–286] microwave irradiation,[287] focused solar irradiation,[288] and electrochemical reduction.[289] Our group presented a physical route to fabricate highly crystalline graphene sheets through high-temperature annealing (1900 8C), under vacuum (1.3 Õ 10¢3 Pa), of functional graphene sheets obtained from graphite oxide by low temperature thermal exfoliation.[285] The obtained Figure 13. Schematic illustration of the fabrication of a coaxial fiber supercapacitor. graphene sheets possess no appreciable oxygen cona) MWCNTs are dispersed in sodium dodecylbenzenesulfonate (NaDDBs) solution. tent, which results in a high electrical conductivity of b) MWCNTs are deposited onto planar CMFs by spray coating. c) The MWCNTs/CMFs are assembled into bundles after removing the surfactant. d, f) SEM images of a single unapproximately 56 500 S m¢1; this is comparable to coated CMF and a single CMF coated with MWCNTs. e) SEM image of a MWCNTs/CMF 100 900 S m¢1 of the precursor graphite. Kaner and bundle. g) SEM image of a CNF film and its enlargement in upper right (the inset is a digico-workers have reported a simple and rapid way to tal photo of the bendable CNF film). h) After soaking with polymer electrolyte, the core reduce GO films by using a laser from a standard MWCNTs/CMF bundle was wrapped with the separator and CNF film. i) Schematic and digital photo of a coaxial fiber supercapacitor. Reprinted with permission from Ref. [263]. LightScribe DVD optical drive.[290] The produced films, Copyright 2013, American Chemical Society. called laser-scribed graphene (LSG), are mechanically robust, highly electrically conductive, and possess a SSA of 1520 m2 g¢1. The LSG-based ECs exhibit ulponents to optimize their energy-storage capabilities is anothtrahigh gravimetric Cs of 276 F g¢1 in EMIMBF4 ionic liquid and er issue. excellent cycling stability. However, these LSG films have very small mass density of approximately 0.048 g cm¢3, which results in a low volumetric Cs of the device. 4.4. Graphene for ECs To increase the SSA of graphene-based materials, preparing Graphene has a high specific surface area, high flexibility, high porous structures has recently attracted much attention. It has chemical stability, and extremely high electronic and thermal been demonstrated that curved or crumpled graphene sheets conductivity, all of which make it attractive for potential applican enhance both the surface area accessible to the electrolyte cations in ECs.[266] Up to now, numerous methods have been and ion transport, which thus enhances the energy density employed for the large-scale synthesis of high-quality graand rate performance.[18, 291–293] Chemical activation is another [267] effective way to produce porous graphene.[294–297] Zhu et al. phene, including mechanical cleavage from graphite, epi[268, 269] [270, 271] taxial growth, CVD growth, solvothermal synthehave reported the KOH activation of microwave-exfoliated GO sis,[272] and exfoliation and reduction from chemically oxidized powders.[294] The activated microwave exfoliated GO (a-MEGO) [273, 274] Among them all, exfoliation and reduction from graphite. possesses a high SSA up to 3100 m2 g¢1, which results in chemically oxidized graphite are the most explored and apa high energy density of 70 W h kg¢1 at a maximum power plied methods for EC applications because of their simple density of 250 kW kg¢1 in ionic liquid. Restacking of graphene layers during the reduction process processing facilities, large-scale production possibilities, and low cost.[275] However, these methods suffer from disadvantagis a big challenge for graphene-based EC applications. This rees of possessing highly oxygen-containing functional groups stacking makes the graphene structure dense without retainand defects, which lead to low electrical conductivity of the asing the effective surface area for ion accessibility, and this derived graphene. leads to poor energy and power performance. Therefore, several efforts have been done to prevent graphene restacking by Many reduction methods have been reported to achieve high-quality rGO. At the very beginning, hydrazine was used as using incorporated spacers, such as nanocarbons (e.g., carbon a chemical agent to reduce GO.[276] Cs values of 135 and spheres, particles, and CNTs),[298–302] metal oxide/hydroxide ¢1 99 F g were obtained in aqueous and organic electrolytes, renanoparticles,[303–305] polymers,[306–308] and water molecules.[309] spectively. Although hydrazine rGO implies exciting potential For instance, our group presented a new GO material intercafor high-performance ECs, the high toxicity of hydrazine is lated by polysodium 4-styrensulfonate (PSS) for high-pera major drawback. Therefore, a variety of environmentally formance ECs.[306, 310] The interlayer distance of the PSS–GO is friendly chemical agents (e.g., NaBH4, hydrobromic acid, l-asincreased by approximately 1 æ relative to that of the original corbic acid, iron, and aluminum) have been developed.[277–281] GO (Figure 14 a, b).[310] The PSS–GO electrodes exhibit a Cs of Recently, Lei et al. have reported a facile and green route to 190 F g¢1 in organic electrolyte (i.e., TEABF4). A high energy [282] reduce GO by using urea. The as-fabricated rGO electrodes density of 38 W h kg¢1 at a power density of 61 W kg¢1 can be ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2302


Reviews obtained. More interestingly, Li’s group recently reported a simple soft approach enabling the subnanometer-scale integration of graphene sheets with an electrolyte (e.g., H2SO4 or EMIMBF4).[70] The films of electrolyte-mediated chemically converted graphene (EM-CCG) possess a continuous ion-transport network and controllable high mass densities ( 0.13– 1.33 g cm¢3, Figure 14 c, d). As a result, ECs based on the EM-CCG film can obtain a volumetric energy density of 60 W h L¢1 in ionic liquid. Lately, our group reported the synthesis of SWCNTbridged graphene 3 D building blocks through the Coulombic interaction of positively charged SWCNTs grafted by cetyltrimethylammonium bromide with negatively charged graphene oxide sheets, followed by KOH activation.[311] The as-obtained activated graphene/SWCNT (ac-Gr/SWCNT) films possess pillared SWCNTs intercalated into nanoporous graphene layers, which enhances the accessible surface area Figure 14. a) SEM image of PSS-GO. b) XRD patterns of pristine graphite, GO, and PSSand allows fast ion diffusion (Figure 15). These ac-Gr/ GO. Reproduced with permission from Ref. [310]. Copyright 2009, American Chemical SoSWCNT films are free standing and flexible and reveal ciety. c, d) SEM images of cross sections of the obtained EM-CCG films containing 78.9 a high electrical conductivity of 39 400 S m¢1 and and 27.2 vol % H2SO4, corresponding to packing densities of 0.42 and 1.33 g cm¢3, respeca reasonable mass density of 1.06 g cm¢3. As a conse- tively. e) XRD patterns of the EM-CCG films and dried CCG film with varied packing densities. Reproduced with permission from Ref. [70]. Copyright 2013, American Association quence, supercapacitors based on these films exhibit for the Advancement of Science. superior performance in neat EMIMBF4 electrolyte with a maximum energy density of 117.2 W h L¢1 or 110.6 W h kg¢1 at a maximum power density of 424 kW L¢1 or 400 kW kg¢1, which is based on the thickness or mass of the total active material. In other work, we present the facile and scalable fabrication of a nanochanneled poly(diallyldimethylammonium chloride)-mediated rGO (nc-PDDA-Gr) film for micro-supercapacitors.[312] The as-synthesized film possesses high packing density (1.12 g cm¢3) and efficient 2 D ion-transport pathways as a result of veinlike-textured nanochannels. The micro-supercapacitor based on this nc-PDDA-Gr film gives a high volumetric energy density of 6.7 mW h cm¢3 at a volumetric power density of 0.1 W cm¢3, a high volumetric capacitance of 348 F cm¢3, long cycling stability (92 % charge retention after 50 000 cycles), and fast frequency response (33 ms). The use of pseudocapacitive spacers (e.g., metal oxides/hydroxides and conducting polymers) is another interesting approach to prevent restacking of graphene.[303–308] These spacers not only efficiently avoid agglomeration of graphene but also contribute additional pseudocapacitance to the total Cs of the EC devices. Several N-rich polymers (e.g., melamine, cyanamide, and PANi)[313–316] can also be used as N-doping precursors that enhance the electrical conductivity, wettability, and capacitance of graphene-based electrodes. These kinds of grapheneFigure 15. Schematic illustration of the fabrication of the ac-Gr/SWCNT based composites will be discussed in Section 4.5. hybrid nanostructure. a) The cetyltrimethylammonium bromide (CTAB) graftIn spite of tremendous achievements, graphene still faces ed SWCNTs are positively charged, and the GO layers are negatively charged because of their functional groups. b) Schematic of the 3 D SWCNT-bridged some serious challenges for commercial EC applications such graphene block. Reprinted with permission from Ref. [311]. Copyright 2015, as the production of high-quality graphene on a large scale, American Chemical Society. with low cost, and by environmentally friendly procedures. Therefore, scalable, cost-effective, and environmentally friendly approaches to produce graphene without irreversible restacking are in urgent demand. In addition, comprehensive studies erties of graphene and its electrochemical behavior are neceson the relationship between the structural and physical propsary. ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2303


Reviews 4.5. Carbon-based composites for ECs Although carbon structures (e.g., ACs, CNTs, CFs, and graphene) possess good properties for EC applications such as high SSA, high electrical conductivity, and high mechanical strength, their capacitance is still low, as most of them exhibit EDLC behavior. Moreover, pseudocapacitive materials (e.g., metal oxides/hydroxides and conducting polymers) that exhibit high charge-storage capacity, environmental friendliness, and low cost have attracted great interest for ECs. However, these kinds of materials suffer from poor electrical conductivity and mechanical degradation during faradic charge/discharge.[3] Therefore, making composites requires a strategy to utilize the synergistic effects of all the active materials, which will lead to high electrochemical performance of the composite-based ECs. Much work has been done to prepare carbons–metal oxide/ hydroxide composites as electrodes for ECs, including AC– metal oxide/hydroxide,[317–319] CNT–metal oxide/hydroxide,[320–324] CF–metal oxide/hydroxide,[325–329] graphene–metal oxide/hydroxide,[22, 330–336] and 3 D porous carbon–metal oxide/ hydroxide.[18, 337, 338] For example, our group reported an ultrathin V2O5 layer electrodeposited on electrospun PAN-based CNFs.[328] The 3 nm thick V2O5-deposited CNF electrode exhibits a very high capacitance of 1308 F g¢1 in 2 m KCl electrolyte. If only the thin oxide layer is considered, it contributes over 90 % of the total capacitance (214 F g¢1). This high capacitance performance is attributed to the large external surface area of the CNFs and the numerous active sites of the ultrathin V2O5 layer for the redox reaction. Conducting polymers (e.g., PANi, PPy, and PTh) are another attractive pseudocapacitive material for EC applications. It has been demonstrated that the electrochemical performance of conducting polymers can be significantly improved if combined with carbon-based materials (e.g., ACs, CNTs, CFs, and graphene).[9, 315, 316, 339–343] For instance, Wei et al. have demonstrated that the combination of 1 D PANi nanowires with 2 D GO nanosheets avoids restacking of graphene, which results in a high Cs of up to 555 F g¢1 and good cycling stability.[343] Feng et al. have presented a new one-step large-scale electrochemical synthesis of graphene/PANi composite films with a high level of N doping.[316] EC electrodes made with these films possess high SSAs, high conductivity, fast redox properties, and perfect layered/encapsulated structures, all of which lead to a high Cs of 640 F g¢1 with a retention life of 90 % after 1000 charge/discharge cycles. More recently, Liu and co-workers have reported a PANi–rGO/CF composite paper that is highly conductive, light, and flexible.[341] The PANi–rGO/CF paper with a PANi deposition time of 72 h exhibits a Cs of 464 F g¢1 based on the masses of PANi and rGO, which is much larger than that of the rGO/CF paper (212 F g¢1). Carbon-based composites have significantly improved energy performance of storage devices. However, several issues still remain, including low loadings of the active materials, reasonable Cs, poor rate capability, short cycling life, and complex processing route. Hence, the optimization of procedures to obtain low-costing composites on large scale with high electrochemical performance still needs to be addressed. ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

5. Carbon-Based Materials for Hybrid EnergyStorage Devices Asymmetric supercapacitors and Li-ion supercapacitors are two types of well-known hybrid devices (Figure 3). Two electrodes (anode and cathode) within the cell exhibit two different potentials in the same electrolyte owing to differences in their inherent electrochemical potentials and charge-storage mechanisms (e.g., EDLC, pseudocapacitance, or intercalation). Generally, carbon materials incorporated with active pseudocapacitance or Li-insertion materials can be used as anodes and high SSA carbon materials are used as EDLC-type cathodes. Representative hybrid devices based on carbonaceous materials are summarized in Table 3. 5.1. Asymmetric supercapacitors Owing to their high SSAs (up to 3400 m2 g¢1) and low cost, ACs are common EDLC-type electrode materials.[344–346] For instance, Qu et al. have reported an asymmetric supercapacitor composed of AC and MnO2 in a K2SO4 electrolyte.[344] The fabricated device recycles the reversibly between 0 and 1.8 V with an energy density of 17 W h kg¢1 at a power density of 2 kW kg¢1. As a result of the nonfaradic reaction on the AC anode and the fast absorption/desorption reaction on the active surface of MnO2, excellent cycling behavior with no more than 6 % capacitance loss after 23 000 cycles is obtained. To achieve a higher operating voltage, Wang et al. have prepared a non-aqueous activated mesocarbon microbead//MnO2 nanowire-sphere asymmetric supercapacitor by using 1 m Et4NBF4 in acetonitrile as the electrolyte.[347] This device performs over a wide voltage range (0.0–3.0 V) with a high Cs of 228 F g¢1 at a scan rate of 10 mV s¢1, which leads to a high energy density of 128 W h kg¢1 on the basis of the total mass of the active materials and maintains desirable cycling stability and rate capability. Flexible and wearable asymmetric supercapacitors have been extensively investigated as emerging energy-storage devices. In such devices, the carbon network (e.g., CNTs and CNFs) provides a flexible support that can be decorated with high pseudocapacitive materials, and it also significantly enhances the electrical conductivity of the composite electrodes. Zhou and co-workers have reported the fabrication of a flexible asymmetric supercapacitor based on SWNT/In2O3 nanowire and SWNT/MnO2 nanowire.[21] In this design, charges can be stored not only through electrochemical double-layer capacitance from the SWCNT films but also through a reversible faradic process from the transition-metal oxide nanowires. The optimized device displays a high Cs of 184 F g¢1, an energy density of 25.5 W h kg¢1, and a power density of 50.3 kW kg¢1 with a 2 V potential window. In other work, superior electrochemical performance with a high energy density of 50 W h kg¢1 at a power density of 1000 W kg¢1 over the potential range of 0 to 2.8 V has been reported for a a-Fe2O3/MWCNT//MWCNT asymmetric supercapacitor.[348] These interesting results are attributed to the incorporation of MWCNTs into the a-Fe2O3 anode, which leads to a decrease in the internal resistance and

2304


Reviews Table 3. The electrochemical performance of hybrid energy-storage devices based on various carbon-based materials. Electrode material[a]

Electrolyte[b]

Potential window [V]

E [W h kg¢1] @P [kW kg¢1]

Cycles

CR[c] [%]

Ref.

MnO2//AC MnO2//AMCMB SWNT/In2O3//SWNT/MnO2 a-Fe2O3/MWCNT//MWCNT carbon-MnO2//AC rGO-RuO2//rGO-PANi rGO/MnO2//rGO/MoO3 Li4Ti5O12/AC//AC carbon-coated LiTi2(PO4)3//AC B-Si/SiO2/C//AC graphite//a-MEGO Fe3O4/graphene//3DGraphene

1m 1m 1m 1m 1m 2m 1m 1m 1m 1m 1m 1m

1.8 3 2 2.8 2 1.4 2 3 1.2 2.5 4 3

17@2 128@– [email protected] 50@ 1 63@ 0.23 26.3@ 0.15 42.6@ 0.28 32@6 27@ 0.1 128@ 0.2 147.8@– [email protected]

23 000 1200 – 600 5000 2500 1000 4000 1000 6000 – 1000

> 94 > 96 – – 92 70 – 70 > 85 70 – 70

[344] [347] [21] [348] [231] [350] [22] [354] [353] [20] [17] [19]

K2SO4 (aq) Et4NBF4/AN (org) Na2SO4 (aq) LiClO4/EM/DCM (org) Na2SO4 (aq) H2SO4 (aq) Na2SO4 (aq) LiPF6/EC/DEC (org) Li2SO4 (aq) LiPF6/DEC/DMC (org) LiPF6/EC/DEC (org) LiPF6/DEC/DMC (org)

[a] AMCMB = activated mesocarbon microbead, a-MEGO = activated microwave exfoliated GO, 3DGraphene = graphene-based 3 D porous carbon. [b] aq = Aqueous electrolyte, org = organic electrolyte. [c] CR = Charge retention.

an improvement in both ion diffusion and the integrity of the a-Fe2O3-containing films. More recently, Long et al. have reported the synthesis of nitrogen-doped carbon networks through the carbonization of PANi-coated bacterial cellulose.[231] The as-obtained carbon network serves as a substrate to obtain high capacitance electrode materials such as AC and carbon/MnO2 hybrid materials and also serves as a conductive network to integrate the active electrode materials. As a consequence, the as-assembled AC//carbon–MnO2 asymmetric supercapacitor performs a high energy density of 63 W h kg¢1 in 1.0 m Na2SO4 aqueous electrolyte with an operating voltage of 2 V. In addition, this asymmetric supercapacitor exhibits an excellent cycling performance with 92 % specific capacitance retention after 5000 cycles. Recently, graphene with a particular 2 D structure, high SSA, and high electrical conductivity has been introduced to enhance the electrochemical performance of asymmetric supercapacitor electrodes.[22, 25, 340, 349–352] For example, Zhang et al. have reported an asymmetric supercapacitor fabricated by using RuO2-modified rGO and PANi-modified rGO as the anode and cathode, respectively.[350] Owing to the broadened potential window in an aqueous electrolyte (1.4 V), the fabricated device exhibits a maximum energy density of 26.3 W h kg¢1 and a power density of 49.8 kW kg¢1, which is about two times higher than that of symmetric supercapacitors based on rGO– RuO2 (12.4 W h kg¢1) and rGO–PANi (13.9 W h kg¢1) electrodes. In other work, our group presented a self-assembled reduced graphene oxide (rGO)/MnO2 (GrMnO2) composite as a positive electrode and a rGO/MoO3 (GrMoO3) composite as a negative electrode for asymmetric supercapacitors based on an aqueous Na2SO4 electrolyte (Figure 16).[22] Owing to synergistic effects between the highly conductive graphene and the highly pseudocapacitive metal oxides, these two hybrid nanostructure electrodes exhibit high charge transport and expand the operational voltage window of the asymmetric device to 2.0 V in spite of using an aqueous electrolyte. The device shows a maximum specific capacitance of 307 F g¢1 and a high energy density of 42.6 W h kg¢1 at a power density of 276 W kg¢1. ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

Figure 16. a) Schematic illustration of the fabricated asymmetric supercapacitor device based on the GrMnO2 composite as the positive electrode and GrMoO3 as the negative electrode. High-magnification TEM images of b) GrMnO2 composite showing graphene coating the particles of flowerlike MnO2 and c) GrMoO3 composite showing graphene wrapping MoO3 nanosheets. Reproduced with permission from Ref. [22]. Copyright 2013, WileyVCH.

5.2. Li-ion supercapacitors The combination of the fast charging rate and high power density of supercapacitors with the high energy density of LIBs makes Li-ion supercapacitors the most promising energy-storage devices that can satisfy various demands for future practical applications. In this hybrid structure, the capacitive cathode requires high capacitance and fast ion transport, whereas the Li-insertion anode needs to have high Li-storage capacity. Therefore, carbon-based materials with high SSAs and appropriate micro/mesoporous structures are usually employed as cathodes, whereas various Li-based compounds such as LiMn2O4, Li4Ti5O12, and LiTi2(PO4)3 are widely used as anodes.[17, 18, 353, 354] This part introduces some novel carbon-

2305


Reviews based materials as electrodes in such Li-ion supercapacitors that exhibit outstanding electrochemical performance. Using AC as a cathode and Li4Ti5O12/AC as an anode, Choi et al. have prepared a hybrid Li-ion supercapacitor that displays a high energy density of 32 W h kg¢1 at a high power density of 6 kW kg¢1 in 1 m LiPF6/EC/DEC electrolyte.[354] In other work, Luo and Xia report a hybrid device based on an AC cathode and a carbon-coated LiTi2(PO4)3 anode in 1 m Li2SO4 aqueous electrolyte.[353] This Li-ion supercapacitor shows a slooping voltage profile from 0.3 to 1.5 V, delivers a capacity of 30 mAh g¢1 and an energy density of 27 W h kg¢1, and maintains over 85 % of its initial energy density after 1000 charge/ discharge cycles. More recently, Wang and co-workers have fabricated a hybrid Li-ion supercapacitor constructed of a Sibased anode and a porous carbon cathode in 1 m LiPF6/DEC/ DMC electrolyte.[20] The as-fabricated device shows both high power and high energy density; a high energy density of 128 W h kg¢1 at 1229 W kg¢1 can be achieved. Even if the power density increases to the level of a conventional supercapacitor (9704 W kg¢1), an energy density of 89 W h kg¢1 can be obtained. This device also exhibits excellent cycling stability (capacity retention of 70 % after 6000 cycles) and a low self-discharge rate (voltage retention of 82 % after 50 h). To improve the capacitance of Li-ion supercapacitor cathodes, other carbon materials with higher SSAs and electrical conductivities relative to that of ACs have been employed. Stoller et al. report a cathode material based on a-MEGO.[17] As a result of the high SSA of a-MEGO (3100 m2 g¢1), the as-fabricated Li-ion supercapacitor with graphite as the anode in 1 m LiPF6/EC/DEC electrolyte displays a high Cs of 266 F g¢1 at an operating potential of 4 V and yields a high energy density for a packaged cell of 53.2 W h kg¢1. In other work, Chen and coworkers prepare a graphene-based 3 D porous carbon material (3DGraphene) with a high surface area ( 3355 m2 g¢1) as a positive electrode material.[19] By using Fe3O4/graphene nanocomposite as the negative electrode material and 1 m LiPF6/EC/ DEC/DMC as the electrolyte, the hybrid Li-ion supercapacitor Fe3O4/graphene//3DGraphene shows an ultrahigh energy density of 147 W h kg¢1 at a power density of 150 W kg¢1, which remains at an energy density of 86 W h kg¢1 even at a high power density of 2587 W kg¢1. Although hybrid energy-storage devices have achieved notable progress in developing high electrochemical performance, challenges such as safety, cost effectiveness, ease of fabrication, and environmental issues still remain. In addition, such devices usually suffer from poor rate capability and limited cycling life owing to faradic redox reaction at the anodes. Strategies to enhance the electrical conductivity and porosity and to prevent the pulverization of anode materials urgently need to be addressed.

6. Summary and Outlook In this Review, we covered fundamental electrochemical storage principles of lithium-ion batteries (LIBs), electrochemical capacitors (ECs), and their hybrid devices. Furthermore, research efforts towards the development of carbonaceous maChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

terial based electrodes have been reviewed by highlighting research progress from our group. Novel materials and techniques have paved a way to the design and fabrication of appropriate nanostructured electrodes for high energy density and high power density devices. Nevertheless, the performance of these devices needs to be improved substantially to meet the requirements of both high energy and power density for future systems. Optimizing the electrical conductivity, electrode/electrolyte interface, and ion diffusion of electrode materials not only improves power density but also leads to high charge-storage capacity of LIBs and ECs, as well as hybrid devices. If graphene or carbon nanotubes (CNTs) are used as a powder, solution, or paste, an approach with a binder is a prerequisite. Although charge-storage performance with such samples is outstanding if the loading is small, this performance is not guaranteed if the loading is increased owing to limited ion diffusion. From an industry point of view, this is a serious limitation for real-life applications. Another issue is the trade-off between volumetric and gravimetric quantities. For example, highly porous carbons have low mass density, which is certainly advantageous for gravimetric energy density and power density but yields poor volumetric values. To maintain reasonable volumetric and gravimetric values, a medium mass density without sacrificing ion diffusion time is required. To compare electrode performance among the various data, mass density with sample thickness or loading should be provided. It is clear that nanocarbon materials provide opportunities to deliver better electrochemical performance than commercially available state-of-the-art materials for LIB anodes (e.g., graphite) and ECs [e.g., activated carbons (ACs)] as a result of their unique structures. Besides, careful structural engineering can further increase the capacity and rate capability of nanocarbon materials. For instance, nitrogen doping improves the conductivity and wettability of the material and also generates more active sites for ion absorption or reaction. However, the nitrogen content and composition of nitrogen bonds need to be carefully considered. The creation of pores or defective sites on nanocarbon materials by using chemical/physical activation or template-assisted methods gives rise to higher surface areas, which leads to a fast rate of ion diffusion and more absorption/reaction sites. Nevertheless, the generation of an excessive amount of pores or defects decreases the graphitization of the nanocarbon materials and, therefore, degrades the electrical conductivity of the electrode. This is eventually detrimental for capacity and rate capability. In addition, the low columbic efficiency at the first cycle in LIBs comes from the large irreversible capacity induced by vigorous electrolyte decomposition at large surface areas. This cannot be avoided in highly porous or defective electrodes. For ECs, a high specific surface area and an appropriate pore-size distribution of the electrode materials are required to facilitate high capacitance and fast power delivery. Therefore, optimizing the structural parameters of electrode materials is needed to compensate several trade offs, including mass density versus porosity and electrical conductivity versus porosity, which further improves energy and power density of the energy-storage devices.

2306


Reviews In general, carbon materials with good conductivity, high structure stability, and efficient pathways for ion diffusion are desirable for electrodes of both LIBs and ECs, as well as for their hybrid devices. However, each device requires specific electrode materials with more specific properties because of the different charge-storage mechanisms. Layered carbons (e.g., graphite and graphite oxides) are usually employed as host materials for intercalation of Li ions in LIB anodes. Emerging graphene-based composites with controllable interlayer distances could be promising candidates for LIB anodes. The combination of carbon materials possessing high conductivity and structure stability with Li-insertion or pseudocapacitance materials possessing high storage capacity are necessary to improve the energy and power performance of LIBs and pseudocapacitors. In the (EDL) capacitors, porous carbons with a high electrolyte-accessible surface area and a reasonable mass density are required for the electrodes. Heteroatom doping of carbon-based materials could be another solution to improve storage capacity but should be carefully designed to avoid conductivity degradation and structural deterioration. The underlying mechanism of dopant-induced performance improvement should be further explored. Noble hybrid device structures such as asymmetric supercapacitors or hybrid Li-ion supercapacitors could provide opportunities to further improve device performance. In such hybrid systems, the selection of materials for the positive and negative electrodes should be optimized. The high rate capability and long cycle life of (EDLC) cathodes as well as the high charge-storage capacity of redox faradic anodes should not be sacrificed. In addition, the mass ratio of both electrodes should be optimized to balance the specific capacity to obtain an optimal energy density.

Acknowledgements This work was supported by the Institute for Basic Science (IBS, EM1304) and in part by BK-Plus through the Ministry of Education, Korea. Keywords: carbon · electrochemical capacitors · impedance · lithium-ion batteries · nanostructures [1] [2] [3] [4] [5] [6]

[7] [8] [9] [10] [11] [12] [13] [14] [15]

S. Chu, A. Majumdar, Nature 2012, 488, 294. P. Simon, Y. Gogotsi, Nat. Mater. 2008, 7, 845. L. L. Zhang, X. S. Zhao, Chem. Soc. Rev. 2009, 38, 2520. A. Ghosh, Y. H. Lee, ChemSusChem 2012, 5, 480. M. Winter, R. J. Brodd, Chem. Rev. 2004, 104, 4245. B. E. Conway, Electrochemical Supercapacitors: Scientific Fundamentals and Technological Applications, Kluwer Academic Press/Plenum Publishers, New York, 1999 and citations therein. D. S. Su, R. Schlçgl, ChemSusChem 2010, 3, 117. J. R. Miller, R. A. Outlaw, B. C. Holloway, Science 2010, 329, 1637. J. Yan, Q. Wang, T. Wei, Z. Fan, Adv. Energy Mater. 2014, 4, 1300816. G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 2012, 41, 797. M. Zhi, C. Xiang, J. Li, M. Li, N. Wu, Nanoscale 2013, 5, 72. M. Seredych, D. Hulicova-Jurcakova, G. Q. Lu, T. J. Bandosz, Carbon 2008, 46, 1475. B. E. Conway, V. Birss, J. Wojtowicz, J. Power Sources 1997, 66, 1. M. Winter, J. O. Besenhard, M. E. Spahr, P. Novk, Adv. Mater. 1998, 10, 725. M. S. Whittingham, Science 1976, 192, 1126.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

2307

[16] K. Xu, Chem. Rev. 2004, 104, 4303. [17] M. D. Stoller, S. Murali, N. Quarles, Y. Zhu, J. R. Potts, X. Zhu, H.-W. Ha, R. S. Ruoff, Phys. Chem. Chem. Phys. 2012, 14, 3388. [18] S. B. Ma, K. W. Nam, W. S. Yoon, X. Q. Yang, K. Y. Ahn, K. H. Oh, K. B. Kim, Electrochem. Commun. 2007, 9, 2807. [19] F. Zhang, T. Zhang, X. Yang, L. Zhang, K. Leng, Y. Huang, Y. Chen, Energy Environ. Sci. 2013, 6, 1623. [20] R. Yi, S. Chen, J. Song, M. L. Gordin, A. Manivannan, D. Wang, Adv. Funct. Mater. 2014, 7433. [21] P. C. Chen, G. Shen, Y. Shi, H. Chen, C. Zhou, ACS Nano 2010, 4, 4403. [22] J. Chang, M. Jin, F. Yao, T. H. Kim, V. T. Le, H. Yue, F. Gunes, B. Li, A. Ghosh, S. Xie, Y. H. Lee, Adv. Funct. Mater. 2013, 23, 5074. [23] J. Xu, Q. Wang, X. Wang, Q. Xiang, B. Liang, D. Chen, G. Shen, ACS Nano 2013, 7, 5453. [24] P. Yang, Y. Ding, Z. Lin, Z. Chen, Y. Li, P. Qiang, M. Ebrahimi, W. Mai, C. P. Wong, Z. L. Wang, Nano Lett. 2014, 14, 731. [25] C. Long, T. Wei, J. Yan, L. Jiang, Z. Fan, ACS Nano 2013, 7, 11325. [26] X.-M. Liu, Z. d. Huang, S. w. Oh, B. Zhang, P.-C. Ma, M. M. F. Yuen, J.-K. Kim, Compos. Sci. Technol. 2012, 72, 121. [27] A. Lewandowski, A. S´widerska-Mocek, J. Power Sources 2009, 194, 601. [28] Z. Chang, Y. Yang, M. Li, X. Wang, Y. Wu, J. Mater. Chem. A 2014, 2, 10739. [29] J.-Y. Luo, W.-J. Cui, P. He, Y.-Y. Xia, Nat. Chem. 2010, 2, 760. [30] Y. Wang, J. Yi, Y. Xia, Adv. Energy Mater. 2012, 2, 830. [31] F. Beck, P. Rìetschi, Electrochim. Acta 2000, 45, 2467. [32] W. Li, J. R. Dahn, D. S. Wainwright, Science 1994, 264, 1115. [33] G. Wang, L. Fu, N. Zhao, L. Yang, Y. Wu, H. Wu, Angew. Chem. Int. Ed. 2007, 46, 295; Angew. Chem. 2007, 119, 299. [34] C. Wu, Z. Hu, W. Wang, M. Zhang, J. Yang, Y. Xie, Chem. Commun. 2008, 3891. [35] M. Minakshi, P. Singh, D. R. G. Mitchell, T. B. Issa, K. Prince, Electrochim. Acta 2007, 52, 7007. [36] L. Liu, F. Tian, M. Zhou, H. Guo, X. Wang, Electrochim. Acta 2012, 70, 360. [37] X. Wang, Q. Qu, Y. Hou, F. Wang, Y. Wu, Chem. Commun. 2013, 49, 6179. [38] X. Wang, Y. Hou, Y. Zhu, Y. Wu, R. Holze, Sci. Rep. 2013, 3, 1401. [39] P. He, Y. Wang, H. Zhou, Electrochem. Commun. 2010, 12, 1686. [40] J.-G. Zhang, D. Wang, W. Xu, J. Xiao, R. E. Williford, J. Power Sources 2010, 195, 4332. [41] B. Zhang, Y. Liu, X. Wu, Y. Yang, Z. Chang, Z. Wen, Y. Wu, Chem. Commun. 2014, 50, 1209. [42] J. F. Whitacre, A. Tevar, S. Sharma, Electrochem. Commun. 2010, 12, 463. [43] J. Lu, L. Li, J.-B. Park, Y.-K. Sun, F. Wu, K. Amine, Chem. Rev. 2014, 114, 5611. [44] H.-X. Yang, J.-F. Qian, J. Inorg. Mater. 2013, 28, 1165. [45] J. B. Goodenough, K.-S. Park, J. Am. Chem. Soc. 2013, 135, 1167. [46] K. J. Fraser, E. I. Izgorodina, M. Forsyth, J. L. Scott, D. R. MacFarlane, Chem. Commun. 2007, 3817. [47] D. R. MacFarlane, M. Forsyth, E. I. Izgorodina, A. P. Abbott, G. Annat, K. Fraser, Phys. Chem. Chem. Phys. 2009, 11, 4962. [48] M. Egashira, H. Todo, N. Yoshimoto, M. Morita, J.-I. Yamaki, J. Power Sources 2007, 174, 560. [49] D. R. MacFarlane, N. Tachikawa, M. Forsyth, J. M. Pringle, P. C. Howlett, G. D. Elliott, J. H. Davis, M. Watanabe, P. Simon, C. A. Angell, Energy Environ. Sci. 2014, 7, 232. [50] L. Zhao, J.-i. Yamaki, M. Egashira, J. Power Sources 2007, 174, 352. [51] M. Ishikawa, T. Sugimoto, M. Kikuta, E. Ishiko, M. Kono, J. Power Sources 2006, 162, 658. [52] M. Galin´ski, A. Lewandowski, I. Ste˛pniak, Electrochim. Acta 2006, 51, 5567. [53] T. Sato, T. Maruo, S. Marukane, K. Takagi, J. Power Sources 2004, 138, 253. [54] M. Holzapfel, C. Jost, P. Novak, Chem. Commun. 2004, 2098. [55] J. Xu, J. Yang, Y. NuLi, J. Wang, Z. Zhang, J. Power Sources 2006, 160, 621. [56] A. Ghosh, V. T. Le, J. J. Bae, Y. H. Lee, Sci. Rep. 2013, 3, 2939. [57] A. Davies, A. Yu, Can. J. Chem. Eng. 2011, 89, 1342. [58] L. Zhang, F. Zhang, X. Yang, G. Long, Y. Wu, T. Zhang, K. Leng, Y. Huang, Y. Ma, A. Yu, Y. Chen, Sci. Rep. 2013, 3, 1408.


Reviews [59] K. H. An, W. S. Kim, Y. S. Park, Y. C. Choi, S. M. Lee, D. C. Chung, D. J. Bae, S. C. Lim, Y. H. Lee, Adv. Mater. 2001, 13, 497. [60] J. Y. Lee, K. H. An, J. K. Heo, Y. H. Lee, J. Phys. Chem. B 2003, 107, 8812. [61] W. Li, D. Chen, Z. Li, Y. Shi, Y. Wan, G. Wang, Z. Jiang, D. Zhao, Carbon 2007, 45, 1757. [62] E. Iyyamperumal, S. Wang, L. Dai, ACS Nano 2012, 6, 5259. [63] L. Zhao, L.-Z. Fan, M.-Q. Zhou, H. Guan, S. Qiao, M. Antonietti, M.-M. Titirici, Adv. Mater. 2010, 22, 5202. [64] W. Waag, S. Kbitz, D. U. Sauer, Appl. Energy 2013, 102, 885. [65] B. Scrosati, J. Hassoun, Y.-K. Sun, Energy Environ. Sci. 2011, 4, 3287. [66] E. Raymundo-PiÇero, K. Kierzek, J. Machnikowski, F. B¦guin, Carbon 2006, 44, 2498. [67] J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon, P. L. Taberna, Science 2006, 313, 1760. [68] A. G. Pandolfo, A. F. Hollenkamp, J. Power Sources 2006, 157, 11. [69] J. Zhang, X. S. Zhao, ChemSusChem 2012, 5, 818. [70] X. Yang, C. Cheng, Y. Wang, L. Qiu, D. Li, Science 2013, 341, 534. [71] F. Yao, B. Li, K. So, J. Chang, T. H. Ly, A. Q. Vu, H. Mun, C. S. Cojocaru, H. Yue, S. Xie, Y. H. Lee, Carbon 2014, 79, 563. [72] T. Enoki, M. Suzuki, M. Endo, Graphite Intercalation Compounds and Applications, Oxford University Press, New York, 2003. [73] B. Jungblut, E. Hoinkis, Phys. Rev. B 1989, 40, 10810. [74] K. Persson, V. A. Sethuraman, L. J. Hardwick, Y. Hinuma, Y. S. Meng, A. van der Ven, V. Srinivasan, R. Kostecki, G. Ceder, J. Phys. Chem. Lett. 2010, 1, 1176. [75] T. Placke, V. Siozios, R. Schmitz, S. F. Lux, P. Bieker, C. Colle, H. W. Meyer, S. Passerini, M. Winter, J. Power Sources 2012, 200, 83. [76] T. Tran, K. Kinoshita, J. Electroanal. Chem. 1995, 386, 221. [77] Y. Yamada, K. Miyazaki, T. Abe, Langmuir 2010, 26, 14990. [78] J. O. Besenhard, H. P. Fritz, Angew. Chem. Int. Ed. Engl. 1983, 22, 950; Angew. Chem. 1983, 95, 954. [79] D. Billaud, F. X. Henry, M. Lelaurain, P. Willmann, J. Phys. Chem. Solids 1996, 57, 775. [80] C. Wang, I. Kakwan, A. J. Appleby, F. E. Little, J. Electroanal. Chem. 2000, 489, 55. [81] A. Funabiki, M. Inaba, T. Abe, Z. Ogumi, J. Electrochem. Soc. 1999, 146, 2443. [82] E. Hazrati, G. A. de Wijs, G. Brocks, Phys. Rev. B 2014, 90, 155448. [83] V. A. Sethuraman, L. J. Hardwick, V. Srinivasan, R. Kostecki, J. Power Sources 2010, 195, 3655. [84] M. Inaba, H. Yoshida, Z. Ogumi, T. Abe, Y. Mizutani, M. Asano, J. Electrochem. Soc. 1995, 142, 20. [85] M. D. Levi, E. Levi, Y. Gofer, D. Aurbach, E. Vieil, J. Serose, J. Phys. Chem. B 1999, 103, 1499. [86] D. S. Su, R. Schlçgl, ChemSusChem 2010, 3, 136. [87] F. Yao, F. Gìnes¸, H. Q. Ta, S. M. Lee, S. J. Chae, K. Y. Sheem, C. S. Cojocaru, S. S. Xie, Y. H. Lee, J. Am. Chem. Soc. 2012, 134, 8646. [88] C. de Las Casas, W. Li, J. Power Sources 2012, 208, 74. [89] Y. Liu, H. Yukawa, M. Morinaga, Comput. Mater. Sci. 2004, 30, 50. [90] C. Garau, A. Frontera, D. QuiÇonero, A. Costa, P. Ballester, P. M. Dey, Chem. Phys. Lett. 2003, 374, 548. [91] L. Jaber-Ansari, H. Iddir, L. A. Curtiss, M. C. Hersam, ACS Nano 2014, 8, 2399. [92] T. Kar, J. Pattanayak, S. Scheiner, J. Phys. Chem. A 2001, 105, 10397. [93] S. Kawasaki, T. Hara, Y. Iwai, Y. Suzuki, Mater. Lett. 2008, 62, 2917. [94] V. Meunier, J. Kephart, C. Roland, J. Bernholc, Phys. Rev. Lett. 2002, 88, 075506. [95] B. Song, J. Yang, J. Zhao, H. Fang, Energy Environ. Sci. 2011, 4, 1379. [96] J. Zhao, A. Buldum, J. Han, J. Ping Lu, Phys. Rev. Lett. 2000, 85, 1706. [97] K. Nishidate, M. Hasegawa, Phys. Rev. B 2005, 71, 245418. [98] D. T. Welna, L. Qu, B. E. Taylor, L. Dai, M. F. Durstock, J. Power Sources 2011, 196, 1455. [99] H. Zhang, G. Cao, Y. Yang, Z. Gu, J. Electrochem. Soc. 2008, 155, K19. [100] H. Zhang, G. Cao, Z. Wang, Y. Yang, Z. Shi, Z. Gu, Electrochim. Acta 2010, 55, 2873. [101] S. Yang, J. Huo, H. Song, X. Chen, Electrochim. Acta 2008, 53, 2238. [102] X. X. Wang, J. N. Wang, H. Chang, Y. F. Zhang, Adv. Funct. Mater. 2007, 17, 3613. [103] X. X. Wang, J. N. Wang, L. F. Su, J. Power Sources 2009, 186, 194. [104] B. Gao, C. Bower, J. D. Lorentzen, L. Fleming, A. Kleinhammes, X. P. Tang, L. E. McNeil, Y. Wu, O. Zhou, Chem. Phys. Lett. 2000, 327, 69.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

[105] N. Pierard, A. Fonseca, Z. Konya, I. Willems, G. Van Tendeloo, J. B. Nagy, Chem. Phys. Lett. 2001, 335, 1. [106] Z. Xiong, Y. Yun, H.-J. Jin, Materials 2013, 6, 1138. [107] H. Shimoda, B. Gao, X. Tang, A. Kleinhammes, L. Fleming, Y. Wu, O. Zhou, Phys. Rev. Lett. 2001, 88, 015502. [108] H. S. Oktaviano, K. Yamada, K. Waki, J. Mater. Chem. 2012, 22, 25167. [109] J. Y. Eom, H. S. Kwon, J. Liu, O. Zhou, Carbon 2004, 42, 2589. [110] S. Klink, E. Ventosa, W. Xia, F. La Mantia, M. Muhler, W. Schuhmann, Electrochem. Commun. 2012, 15, 10. [111] L. G. Bulusheva, A. V. Okotrub, A. G. Kurenya, H. Zhang, H. Zhang, X. Chen, H. Song, Carbon 2011, 49, 4013. [112] X. Li, X. Zhu, Y. Zhu, Z. Yuan, L. Si, Y. Qian, Carbon 2014, 69, 515. [113] C.-C. Kaun, B. Larade, H. Mehrez, J. Taylor, H. Guo, Phys. Rev. B 2002, 65, 205416. [114] S. Lim, R. Li, W. Ji, J. Lin, Phys. Rev. B 2007, 76, 195406. [115] N. Gavrilov, I. A. Pasˇti, M. Vujkovic´, J. Travas-Sejdic, G. C´iric´-Marjanovic´, S. V. Mentus, Carbon 2012, 50, 3915. [116] Y. Wu, S. Fang, Y. Jiang, Solid State Ionics 1999, 120, 117. [117] W. Ren, D. Li, H. Liu, R. Mi, Y. Zhang, L. Dong, L. Dong, Electrochim. Acta 2013, 105, 75. [118] W. J. Lee, U. N. Maiti, J. M. Lee, J. Lim, T. H. Han, S. O. Kim, Chem. Commun. 2014, 50, 6818. [119] X. Huang, R. Zhang, X. Zhang, G. Wen, H. Yu, Y. Zhou, Scr. Mater. 2012, 67, 987. [120] J. P. Paraknowitsch, A. Thomas, Energy Environ. Sci. 2013, 6, 2839. [121] M. Inagaki, Y. Yang, F. Kang, Adv. Mater. 2012, 24, 2547. [122] C. Kim, K. S. Yang, M. Kojima, K. Yoshida, Y. J. Kim, Y. A. Kim, M. Endo, Adv. Funct. Mater. 2006, 16, 2393. [123] G. Larsen, R. Spretz, R. Velarde-Ortiz, J. Mater. Chem. 2004, 14, 1533. [124] A. L. Yarin, E. Zussman, J. H. Wendorff, A. Greiner, J. Mater. Chem. 2007, 17, 2585. [125] C. Kim, Y. I. Jeong, B. T. N. Ngoc, K. S. Yang, M. Kojima, Y. A. Kim, M. Endo, J.-W. Lee, Small 2007, 3, 91. [126] C. Kim, B. T. N. Ngoc, K. S. Yang, M. Kojima, Y. A. Kim, Y. J. Kim, M. Endo, S. C. Yang, Adv. Mater. 2007, 19, 2341. [127] L. Ji, X. Zhang, Nanotechnology 2009, 20, 155705. [128] L. Ji, Z. Lin, A. J. Medford, X. Zhang, Carbon 2009, 47, 3346. [129] L. Ji, X. Zhang, Electrochem. Commun. 2009, 11, 684. [130] S.-H. Yoon, S. Lim, Y. Song, Y. Ota, W. Qiao, A. Tanaka, I. Mochida, Carbon 2004, 42, 1723. [131] M. Sevilla, A. B. Fuertes, Energy Environ. Sci. 2011, 4, 1765. [132] C. Chen, R. Agrawal, Y. Hao, C. Wang, ECS J. Solid State Sci. Technol. 2013, 2, M3074. [133] B.-S. Lee, S.-B. Son, K.-M. Park, W.-R. Yu, K.-H. Oh, S.-H. Lee, J. Power Sources 2012, 199, 53. [134] Y. Xing, Y. Wang, C. Zhou, S. Zhang, B. Fang, ACS Appl. Mater. Interfaces 2014, 6, 2561. [135] W. E. Tenhaeff, O. Rios, K. More, M. A. McGuire, Adv. Funct. Mater. 2014, 24, 86. [136] Z. Wang, X. Xiong, L. Qie, Y. Huang, Electrochim. Acta 2013, 106, 320. [137] B. Zhang, Y. Yu, Z.-L. Xu, S. Abouali, M. Akbari, Y.-B. He, F. Kang, J.-K. Kim, Adv. Energy Mater. 2014, 4, 1301448. [138] Y. Chen, X. Li, X. Zhou, H. Yao, H. Huang, Y.-W. Mai, L. Zhou, Energy Environ. Sci. 2014, 7, 2689. [139] Z. Liu, X. Zhang, S. Poyraz, S. P. Surwade, S. K. Manohar, J. Am. Chem. Soc. 2010, 132, 13158. [140] L. Qie, W.-M. Chen, Z.-H. Wang, Q.-G. Shao, X. Li, L.-X. Yuan, X.-L. Hu, W.-X. Zhang, Y.-H. Huang, Adv. Mater. 2012, 24, 2047. [141] S. Goriparti, E. Miele, F. De Angelis, E. Di Fabrizio, R. Proietti Zaccaria, C. Capiglia, J. Power Sources 2014, 257, 421. [142] H. J. Hwang, J. Koo, M. Park, N. Park, Y. Kwon, H. Lee, J. Phys. Chem. C 2013, 117, 6919. [143] D. Pan, S. Wang, B. Zhao, M. Wu, H. Zhang, Y. Wang, Z. Jiao, Chem. Mater. 2009, 21, 3136. [144] J. Hou, Y. Shao, M. W. Ellis, R. B. Moore, B. Yi, Phys. Chem. Chem. Phys. 2011, 13, 15384. [145] Z.-L. Wang, D. Xu, H.-G. Wang, Z. Wu, X.-B. Zhang, ACS Nano 2013, 7, 2422. [146] T. Bhardwaj, A. Antic, B. Pavan, V. Barone, B. D. Fahlman, J. Am. Chem. Soc. 2010, 132, 12556. [147] T. Zheng, W. Xing, J. R. Dahn, Carbon 1996, 34, 1501.

2308


Reviews [148] K. Sato, M. Noguchi, A. Demachi, N. Oki, M. Endo, Science 1994, 264, 556. [149] E. Yoo, J. Kim, E. Hosono, H.-s. Zhou, T. Kudo, I. Honma, Nano Lett. 2008, 8, 2277. [150] P. Guo, H. Song, X. Chen, Electrochem. Commun. 2009, 11, 1320. [151] P. Lian, X. Zhu, S. Liang, Z. Li, W. Yang, H. Wang, Electrochim. Acta 2010, 55, 3909. [152] C. Tang, Q. Zhang, M.-Q. Zhao, J.-Q. Huang, X.-B. Cheng, G.-L. Tian, H.J. Peng, F. Wei, Adv. Mater. 2014, 26, 6100. [153] S. Yin, Y. Zhang, J. Kong, C. Zou, C. M. Li, X. Lu, J. Ma, F. Y. C. Boey, X. Chen, ACS Nano 2011, 5, 3831. [154] X. Zhao, C. M. Hayner, M. C. Kung, H. H. Kung, ACS Nano 2011, 5, 8739. [155] Y. Fang, Y. Lv, R. Che, H. Wu, X. Zhang, D. Gu, G. Zheng, D. Zhao, J. Am. Chem. Soc. 2013, 135, 1524. [156] A. Abouimrane, O. C. Compton, K. Amine, S. T. Nguyen, J. Phys. Chem. C 2010, 114, 12800. [157] H. Wang, C. Zhang, Z. Liu, L. Wang, P. Han, H. Xu, K. Zhang, S. Dong, J. Yao, G. Cui, J. Mater. Chem. 2011, 21, 5430. [158] L. Wang, Z. Sofer, J. Luxa, M. Pumera, J. Mater. Chem. C 2014, 2, 2887. [159] K. N. Wood, R. O’Hayre, S. Pylypenko, Energy Environ. Sci. 2014, 7, 1212. [160] Z.-S. Wu, W. Ren, L. Xu, F. Li, H.-M. Cheng, ACS Nano 2011, 5, 5463. [161] D. Cai, S. Wang, P. Lian, X. Zhu, D. Li, W. Yang, H. Wang, Electrochim. Acta 2013, 90, 492. [162] T. Hu, X. Sun, H. Sun, G. Xin, D. Shao, C. Liu, J. Lian, Phys. Chem. Chem. Phys. 2014, 16, 1060. [163] Z.-J. Jiang, Z. Jiang, ACS Appl. Mater. Interfaces 2014, 6, 19082. [164] X. Li, D. Geng, Y. Zhang, X. Meng, R. Li, X. Sun, Electrochem. Commun. 2011, 13, 822. [165] Y. Liu, X. Wang, Y. Dong, Z. Wang, Z. Zhao, J. Qiu, J. Mater. Chem. A 2014, 2, 16832. [166] C. Ma, X. Shao, D. Cao, J. Mater. Chem. 2012, 22, 8911. [167] X. Ma, G. Ning, C. Qi, C. Xu, J. Gao, ACS Appl. Mater. Interfaces 2014, 6, 14415. [168] X. Ma, G. Ning, Y. Sun, Y. Pu, J. Gao, Carbon 2014, 79, 310. [169] C. Zhang, N. Mahmood, H. Yin, F. Liu, Y. Hou, Adv. Mater. 2013, 25, 4932. [170] A. L. M. Reddy, A. Srivastava, S. R. Gowda, H. Gullapalli, M. Dubey, P. M. Ajayan, ACS Nano 2010, 4, 6337. [171] C.-H. Lai, M.-Y. Lu, L.-J. Chen, J. Mater. Chem. 2012, 22, 19. [172] S. L. Candelaria, Y. Shao, W. Zhou, X. Li, J. Xiao, J.-G. Zhang, Y. Wang, J. Liu, J. Li, G. Cao, Nano Energy 2012, 1, 195. [173] L. Ji, Z. Lin, M. Alcoutlabi, X. Zhang, Energy Environ. Sci. 2011, 4, 2682. [174] T. D. Hatchard, J. R. Dahn, J. Electrochem. Soc. 2004, 151, A838. [175] J. P. Maranchi, A. F. Hepp, A. G. Evans, N. T. Nuhfer, P. N. Kumta, J. Electrochem. Soc. 2006, 153, A1246. [176] R. Zhang, Y. Du, D. Li, D. Shen, J. Yang, Z. Guo, H. K. Liu, A. A. Elzatahry, D. Zhao, Adv. Mater. 2014, 26, 6749. [177] G. Jeong, J.-G. Kim, M.-S. Park, M. Seo, S. M. Hwang, Y.-U. Kim, Y.-J. Kim, J. H. Kim, S. X. Dou, ACS Nano 2014, 8, 2977. [178] W. J. Lee, T. H. Hwang, J. O. Hwang, H. W. Kim, J. Lim, H. Y. Jeong, J. Shim, T. H. Han, J. Y. Kim, J. W. Choi, S. O. Kim, Energy Environ. Sci. 2014, 7, 621. [179] T. D. Bogart, D. Oka, X. Lu, M. Gu, C. Wang, B. A. Korgel, ACS Nano 2014, 8, 915. [180] F. M. Hassan, V. Chabot, A. R. Elsayed, X. Xiao, Z. Chen, Nano Lett. 2014, 14, 277. [181] R. Yi, F. Dai, M. L. Gordin, H. Sohn, D. Wang, Adv. Energy Mater. 2013, 3, 1507. [182] B. Wang, X. Li, T. Qiu, B. Luo, J. Ning, J. Li, X. Zhang, M. Liang, L. Zhi, Nano Lett. 2013, 13, 5578. [183] K. Fu, O. Yildiz, H. Bhanushali, Y. Wang, K. Stano, L. Xue, X. Zhang, P. D. Bradford, Adv. Mater. 2013, 25, 5109. [184] Y. Park, N.-S. Choi, S. Park, S. H. Woo, S. Sim, B. Y. Jang, S. M. Oh, S. Park, J. Cho, K. T. Lee, Adv. Energy Mater. 2013, 3, 206. [185] X. Zhou, L.-J. Wan, Y.-G. Guo, Small 2013, 9, 2684. [186] N. T. Hieu, J. Suk, D. W. Kim, J. S. Park, Y. Kang, J. Mater. Chem. A 2014, 2, 15094. [187] H. Wu, G. Zheng, N. Liu, T. J. Carney, Y. Yang, Y. Cui, Nano Lett. 2012, 12, 904. [188] L. Ji, X. Zhang, Energy Environ. Sci. 2010, 3, 124.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

[189] T. H. Hwang, Y. M. Lee, B.-S. Kong, J.-S. Seo, J. W. Choi, Nano Lett. 2012, 12, 802. [190] X. Zhou, Y.-X. Yin, L.-J. Wan, Y.-G. Guo, Chem. Commun. 2012, 48, 2198. [191] H. Xiang, K. Zhang, G. Ji, J. Y. Lee, C. Zou, X. Chen, J. Wu, Carbon 2011, 49, 1787. [192] B. Wang, X. Li, X. Zhang, B. Luo, M. Jin, M. Liang, S. A. Dayeh, S. T. Picraux, L. Zhi, ACS Nano 2013, 7, 1437. [193] X. Zhou, Y.-X. Yin, L.-J. Wan, Y.-G. Guo, Adv. Energy Mater. 2012, 2, 1086. [194] B. Li, F. Yao, J. J. Bae, J. Chang, M. R. Zamfir, D. T. Le, D. T. Pham, H. Yue, Y. H. Lee, Sci. Rep. 2015, 5, 7659. [195] L. Wei, G. Yushin, Carbon 2011, 49, 4830. [196] J. Wang, S. Kaskel, J. Mater. Chem. 2012, 22, 23710. [197] M. Sevilla, R. Mokaya, Energy Environ. Sci. 2014, 7, 1250. [198] G. Hasegawa, M. Aoki, K. Kanamori, K. Nakanishi, T. Hanada, K. Tadanaga, J. Mater. Chem. 2011, 21, 2060. [199] D. Hulicova-Jurcakova, A. M. Puziy, O. I. Poddubnaya, F. Surez-Garca, J. M. D. Tascûn, G. Q. Lu, J. Am. Chem. Soc. 2009, 131, 5026. [200] T. E. Rufford, D. Hulicova-Jurcakova, K. Khosla, Z. Zhu, G. Q. Lu, J. Power Sources 2010, 195, 912. [201] L. Wei, M. Sevilla, A. B. Fuertes, R. Mokaya, G. Yushin, Adv. Funct. Mater. 2012, 22, 827. [202] Y. Lv, F. Zhang, Y. Dou, Y. Zhai, J. Wang, H. Liu, Y. Xia, B. Tu, D. Zhao, J. Mater. Chem. 2012, 22, 93. [203] J. Wang, M. Chen, C. Wang, J. Wang, J. Zheng, J. Power Sources 2011, 196, 550. [204] I.-J. Kim, S. Yang, M.-J. Jeon, S.-I. Moon, H.-S. Kim, Y.-P. Lee, K.-H. An, Y.H. Lee, J. Power Sources 2007, 173, 621. [205] W. Chen, H. Zhang, Y. Huang, W. Wang, J. Mater. Chem. 2010, 20, 4773. [206] W.-S. Choi, W.-G. Shim, D.-W. Ryu, M.-J. Hwang, H. Moon, Microporous Mesoporous Mater. 2012, 155, 274. [207] X. He, R. Li, J. Qiu, K. Xie, P. Ling, M. Yu, X. Zhang, M. Zheng, Carbon 2012, 50, 4911. [208] C. Peng, X.-b. Yan, R.-t. Wang, J.-w. Lang, Y.-j. Ou, Q.-j. Xue, Electrochim. Acta 2013, 87, 401. [209] C.-W. Huang, C.-T. Hsieh, P.-L. Kuo, H. Teng, J. Mater. Chem. 2012, 22, 7314. [210] G. Dobele, T. Dizhbite, M. V. Gil, A. Volperts, T. A. Centeno, Biomass Bioenergy 2012, 46, 145. [211] X. Li, W. Xing, S. Zhuo, J. Zhou, F. Li, S.-Z. Qiao, G.-Q. Lu, Bioresour. Technol. 2011, 102, 1118. [212] W. Huang, H. Zhang, Y. Huang, W. Wang, S. Wei, Carbon 2011, 49, 838. [213] K.-Y. Lee, H. Qian, F. Tay, J. Blaker, S. Kazarian, A. Bismarck, J. Mater. Sci. 2013, 48, 367. [214] C. Peng, J. Lang, S. Xu, X. Wang, RSC Adv. 2014, 4, 54662. [215] D. Saha, Y. Li, Z. Bi, J. Chen, J. K. Keum, D. K. Hensley, H. A. Grappe, H. M. Meyer, S. Dai, M. P. Paranthaman, A. K. Naskar, Langmuir 2014, 30, 900. [216] G. Sun, K. Li, L. Xie, J. Wang, Y. Li, Microporous Mesoporous Mater. 2012, 151, 282. [217] Z. Zheng, Q. Gao, J. Power Sources 2011, 196, 1615. [218] S.-J. Han, Y.-H. Kim, K.-S. Kim, S.-J. Park, Curr. Appl. Phys. 2012, 12, 1039. [219] V. Ruiz, A. G. Pandolfo, J. Power Sources 2011, 196, 7816. [220] Y. J. Kim, C.-M. Yang, K. C. Park, K. Kaneko, Y. A. Kim, M. Noguchi, T. Fujino, S. Oyama, M. Endo, ChemSusChem 2012, 5, 535. [221] M.-J. Jung, E. Jeong, S. I. Lee, Y.-S. Lee, J. Ind. Eng. Chem. 2012, 18, 642. [222] T. Ohta, I.-T. Kim, M. Egashira, N. Yoshimoto, M. Morita, J. Power Sources 2012, 198, 408. [223] D. Tashima, H. Yoshitama, T. Sakoda, A. Okazaki, T. Kawaji, Electrochim. Acta 2012, 77, 198. [224] S. L. Candelaria, B. B. Garcia, D. Liu, G. Cao, J. Mater. Chem. 2012, 22, 9884. [225] T. A. Centeno, O. Sereda, F. Stoeckli, Phys. Chem. Chem. Phys. 2011, 13, 12403. [226] S. Kondrat, C. R. Perez, V. Presser, Y. Gogotsi, A. A. Kornyshev, Energy Environ. Sci. 2012, 5, 6474. [227] S. Pohlmann, B. Lobato, T. A. Centeno, A. Balducci, Phys. Chem. Chem. Phys. 2013, 15, 17287. [228] M. P. Bichat, E. Raymundo-PiÇero, F. B¦guin, Carbon 2010, 48, 4351. [229] D.-W. Wang, F. Li, L.-C. Yin, X. Lu, Z.-G. Chen, I. R. Gentle, G. Q. Lu, H.-M. Cheng, Chem. Eur. J. 2012, 18, 5345.

2309


Reviews [230] C. Wang, Y. Zhou, L. Sun, P. Wan, X. Zhang, J. Qiu, J. Power Sources 2013, 239, 81. [231] C. Long, D. Qi, T. Wei, J. Yan, L. Jiang, Z. Fan, Adv. Funct. Mater. 2014, 24, 3953. [232] D. N. Futaba, K. Hata, T. Yamada, T. Hiraoka, Y. Hayamizu, Y. Kakudate, O. Tanaike, H. Hatori, M. Yumura, S. Iijima, Nat. Mater. 2006, 5, 987. [233] S. Iijima, Nature 1991, 354, 56. [234] K. H. An, W. S. Kim, Y. S. Park, J. M. Moon, D. J. Bae, S. C. Lim, Y. S. Lee, Y. H. Lee, Adv. Funct. Mater. 2001, 11, 387. [235] E. Frackowiak, K. Metenier, V. Bertagna, F. B¦guin, Appl. Phys. Lett. 2000, 77, 2421. [236] G. Che, B. B. Lakshmi, E. R. Fisher, C. R. Martin, Nature 1998, 393, 346. [237] C. Niu, E. K. Sichel, R. Hoch, D. Moy, H. Tennent, Appl. Phys. Lett. 1997, 70, 1480. [238] H.-J. Ahn, J. I. Sohn, Y.-S. Kim, H.-S. Shim, W. B. Kim, T.-Y. Seong, Electrochem. Commun. 2006, 8, 513. [239] J. H. Chen, W. Z. Li, D. Z. Wang, S. X. Yang, J. G. Wen, Z. F. Ren, Carbon 2002, 40, 1193. [240] S. Talapatra, S. Kar, S. K. Pal, R. Vajtai, L. CiL, P. Victor, M. M. Shaijumon, S. Kaur, O. Nalamasu, P. M. Ajayan, Nat. Nanotechnol. 2006, 1, 112. [241] M. M. Shaijumon, F. S. Ou, L. Ci, P. M. Ajayan, Chem. Commun. 2008, 2373. [242] W. Lu, L. Qu, K. Henry, L. Dai, J. Power Sources 2009, 189, 1270. [243] E. Frackowiak, S. Delpeux, K. Jurewicz, K. Szostak, D. Cazorla-Amoros, F. B¦guin, Chem. Phys. Lett. 2002, 361, 35. [244] Q. Jiang, M. Z. Qu, G. M. Zhou, B. L. Zhang, Z. L. Yu, Mater. Lett. 2002, 57, 988. [245] B.-J. Yoon, S.-H. Jeong, K.-H. Lee, H. Seok Kim, C. Gyung Park, J. Hun Han, Chem. Phys. Lett. 2004, 388, 170. [246] Y. S. Yun, G. Yoon, K. Kang, H.-J. Jin, Carbon 2014, 80, 246. [247] 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. 2011, 4, 1440. [248] L. Hu, M. Pasta, F. L. Mantia, L. Cui, S. Jeong, H. D. Deshazer, J. W. Choi, S. M. Han, Y. Cui, Nano Lett. 2010, 10, 708. [249] M. Kaempgen, C. K. Chan, J. Ma, Y. Cui, G. Gruner, Nano Lett. 2009, 9, 1872. [250] B. Zhang, J. Liang, C. L. Xu, B. Q. Wei, D. B. Ruan, D. H. Wu, Mater. Lett. 2001, 51, 539. [251] A. Izadi-Najafabadi, S. Yasuda, K. Kobashi, T. Yamada, D. N. Futaba, H. Hatori, M. Yumura, S. Iijima, K. Hata, Adv. Mater. 2010, 22, E235. [252] A. Izadi-Najafabadi, T. Yamada, D. N. Futaba, M. Yudasaka, H. Takagi, H. Hatori, S. Iijima, K. Hata, ACS Nano 2011, 5, 811. [253] Y.-E. Miao, W. Fan, D. Chen, T. Liu, ACS Appl. Mater. Interfaces 2013, 5, 4423. [254] S. Hu, S. Zhang, N. Pan, Y.-L. Hsieh, J. Power Sources 2014, 270, 106. [255] E. J. Ra, E. Raymundo-PiÇero, Y. H. Lee, F. B¦guin, Carbon 2009, 47, 2984. [256] Z. Jin, X. Yan, Y. Yu, G. Zhao, J. Mater. Chem. A 2014, 2, 11706. [257] Y. Fan, Z. Zhao, Q. Zhou, G. Li, X. Wang, J. Qiu, Y. Gogotsi, Carbon 2013, 58, 128. [258] L.-F. Chen, X.-D. Zhang, H.-W. Liang, M. Kong, Q.-F. Guan, P. Chen, Z.-Y. Wu, S.-H. Yu, ACS Nano 2012, 6, 7092. [259] D. Yu, K. Goh, H. Wang, L. Wei, W. Jiang, Q. Zhang, L. Dai, Y. Chen, Nat. Nanotechnol. 2014, 9, 555. [260] D. Yu, K. Goh, Q. Zhang, L. Wei, H. Wang, W. Jiang, Y. Chen, Adv. Mater. 2014, 26, 6790. [261] T. Chen, L. Dai, J. Mater. Chem. A 2014, 2, 10756. [262] Y. Meng, Y. Zhao, C. Hu, H. Cheng, Y. Hu, Z. Zhang, G. Shi, L. Qu, Adv. Mater. 2013, 25, 2326. [263] V. T. Le, H. Kim, A. Ghosh, J. Kim, J. Chang, Q. A. Vu, D. T. Pham, J.-H. Lee, S.-W. Kim, Y. H. Lee, ACS Nano 2013, 7, 5940. [264] J. Ren, L. Li, C. Chen, X. Chen, Z. Cai, L. Qiu, Y. Wang, X. Zhu, H. Peng, Adv. Mater. 2013, 25, 1155. [265] X. Xiao, T. Li, P. Yang, Y. Gao, H. Jin, W. Ni, W. Zhan, X. Zhang, Y. Cao, J. Zhong, L. Gong, W.-C. Yen, W. Mai, J. Chen, K. Huo, Y.-L. Chueh, Z. L. Wang, J. Zhou, ACS Nano 2012, 6, 9200. [266] J. Zhu, D. Yang, Z. Yin, Q. Yan, H. Zhang, Small 2014, 10, 3480. [267] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Science 2004, 306, 666.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

[268] K. V. Emtsev, A. Bostwick, K. Horn, J. Jobst, G. L. Kellogg, L. Ley, J. L. McChesney, T. Ohta, S. A. Reshanov, J. Rohrl, E. Rotenberg, A. K. Schmid, D. Waldmann, H. B. Weber, T. Seyller, Nat. Mater. 2009, 8, 203. [269] P. W. Sutter, J.-I. Flege, E. A. Sutter, Nat. Mater. 2008, 7, 406. [270] K. S. Kim, Y. Zhao, H. Jang, S. Y. Lee, J. M. Kim, K. S. Kim, J.-H. Ahn, P. Kim, J.-Y. Choi, B. H. Hong, Nature 2009, 457, 706. [271] X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo, R. S. Ruoff, Science 2009, 324, 1312. [272] M. Choucair, P. Thordarson, J. A. Stride, Nat. Nanotechnol. 2009, 4, 30. [273] S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen, R. S. Ruoff, Nature 2006, 442, 282. [274] D. Li, M. B. Muller, S. Gilje, R. B. Kaner, G. G. Wallace, Nat. Nanotechnol. 2008, 3, 101. [275] X. Huang, Z. Zeng, Z. Fan, J. Liu, H. Zhang, Adv. Mater. 2012, 24, 5979. [276] M. D. Stoller, S. Park, Y. Zhu, J. An, R. S. Ruoff, Nano Lett. 2008, 8, 3498. [277] L. Lai, H. Yang, L. Wang, B. K. Teh, J. Zhong, H. Chou, L. Chen, W. Chen, Z. Shen, R. S. Ruoff, J. Lin, ACS Nano 2012, 6, 5941. [278] Y. Chen, X. Zhang, D. Zhang, P. Yu, Y. Ma, Carbon 2011, 49, 573. [279] X. Zhang, Z. Sui, B. Xu, S. Yue, Y. Luo, W. Zhan, B. Liu, J. Mater. Chem. 2011, 21, 6494. [280] Z. Fan, K. Wang, T. Wei, J. Yan, L. Song, B. Shao, Carbon 2010, 48, 1686. [281] Z.-J. Fan, W. Kai, J. Yan, T. Wei, L.-J. Zhi, J. Feng, Y.-m. Ren, L.-P. Song, F. Wei, ACS Nano 2010, 5, 191. [282] Z. Lei, L. Lu, X. S. Zhao, Energy Environ. Sci. 2012, 5, 6391. [283] S. R. C. Vivekchand, C. Rout, K. S. Subrahmanyam, A. Govindaraj, C. N. R. Rao, J. Chem. Sci. 2008, 120, 9. [284] W. Lv, D.-M. Tang, Y.-B. He, C.-H. You, Z.-Q. Shi, X.-C. Chen, C.-M. Chen, P.-X. Hou, C. Liu, Q.-H. Yang, ACS Nano 2009, 3, 3730. [285] M. Jin, T. H. Kim, S. C. Lim, D. L. Duong, H. J. Shin, Y. W. Jo, H. K. Jeong, J. Chang, S. Xie, Y. H. Lee, Adv. Funct. Mater. 2011, 21, 3496. [286] M. Jin, H.-K. Jeong, T.-H. Kim, K. P. So, Y. Cui, W. J. Yu, E. J. Ra, Y. H. Lee, J. Phys. D 2010, 43, 275402. [287] Y. Zhu, S. Murali, M. D. Stoller, A. Velamakanni, R. D. Piner, R. S. Ruoff, Carbon 2010, 48, 2118. [288] V. Eswaraiah, S. S. Jyothirmayee Aravind, S. Ramaprabhu, J. Mater. Chem. 2011, 21, 6800. [289] Y. Shao, J. Wang, M. Engelhard, C. Wang, Y. Lin, J. Mater. Chem. 2010, 20, 743. [290] M. F. El-Kady, V. Strong, S. Dubin, R. B. Kaner, Science 2012, 335, 1326. [291] C. Liu, Z. Yu, D. Neff, A. Zhamu, B. Z. Jang, Nano Lett. 2010, 10, 4863. [292] Z. Fan, Q. Zhao, T. Li, J. Yan, Y. Ren, J. Feng, T. Wei, Carbon 2012, 50, 1699. [293] J. Yan, Y. Xiao, G. Ning, T. Wei, Z. Fan, RSC Adv. 2013, 3, 2566. [294] Y. Zhu, S. Murali, M. D. Stoller, K. J. Ganesh, W. Cai, P. J. Ferreira, A. Pirkle, R. M. Wallace, K. A. Cychosz, M. Thommes, D. Su, E. A. Stach, R. S. Ruoff, Science 2011, 332, 1537. [295] L. L. Zhang, X. Zhao, M. D. Stoller, Y. Zhu, H. Ji, S. Murali, Y. Wu, S. Perales, B. Clevenger, R. S. Ruoff, Nano Lett. 2012, 12, 1806. [296] S. Murali, N. Quarles, L. L. Zhang, J. R. Potts, Z. Tan, Y. Lu, Y. Zhu, R. S. Ruoff, Nano Energy 2013, 2, 764. [297] T. Kim, G. Jung, S. Yoo, K. S. Suh, R. S. Ruoff, ACS Nano 2013, 7, 6899. [298] Z. Lei, N. Christov, X. S. Zhao, Energy Environ. Sci. 2011, 4, 1866. [299] Z. Fan, J. Yan, L. Zhi, Q. Zhang, T. Wei, J. Feng, M. Zhang, W. Qian, F. Wei, Adv. Mater. 2010, 22, 3723. [300] M.-Q. Zhao, Q. Zhang, J.-Q. Huang, G.-L. Tian, T.-C. Chen, W.-Z. Qian, F. Wei, Carbon 2013, 54, 403. [301] J. Yan, T. Wei, B. Shao, F. Ma, Z. Fan, M. Zhang, C. Zheng, Y. Shang, W. Qian, F. Wei, Carbon 2010, 48, 1731. [302] N. Jung, S. Kwon, D. Lee, D.-M. Yoon, Y. M. Park, A. Benayad, J.-Y. Choi, J. S. Park, Adv. Mater. 2013, 25, 6854. [303] W. Shi, J. Zhu, D. H. Sim, Y. Y. Tay, Z. Lu, X. Zhang, Y. Sharma, M. Srinivasan, H. Zhang, H. H. Hng, Q. Yan, J. Mater. Chem. 2011, 21, 3422. [304] Q. Qu, S. Yang, X. Feng, Adv. Mater. 2011, 23, 5574. [305] H. Wang, H. S. Casalongue, Y. Liang, H. Dai, J. Am. Chem. Soc. 2010, 132, 7472. [306] H.-K. Jeong, M. Jin, E. J. Ra, K. Y. Sheem, G. H. Han, S. Arepalli, Y. H. Lee, ACS Nano 2010, 4, 1162. [307] D.-W. Wang, F. Li, J. Zhao, W. Ren, Z.-G. Chen, J. Tan, Z.-S. Wu, I. Gentle, G. Q. Lu, H.-M. Cheng, ACS Nano 2009, 3, 1745.

2310


Reviews [308] P. A. Mini, A. Balakrishnan, S. V. Nair, K. R. V. Subramanian, Chem. Commun. 2011, 47, 5753. [309] X. Yang, J. Zhu, L. Qiu, D. Li, Adv. Mater. 2011, 23, 2833. [310] H.-K. Jeong, M. H. Jin, K. H. An, Y. H. Lee, J. Phys. Chem. C 2009, 113, 13060. [311] D. T. Pham, T. H. Lee, D. H. Luong, F. Yao, A. Ghosh, V. T. Le, T. H. Kim, B. Li, J. Chang, Y. H. Lee, ACS Nano 2015, 9, 2018. [312] J. Chang, S. Adhikari, T. H. Lee, B. Li, F. Yao, D. T. Pham, V. T. Le, Y. H. Lee, Adv. Energy Mater. 2015, 1500003. [313] F. Ma, H. Zhao, L. Sun, Q. Li, L. Huo, T. Xia, S. Gao, G. Pang, Z. Shi, S. Feng, J. Mater. Chem. 2012, 22, 13464. [314] Z. Wen, X. Wang, S. Mao, Z. Bo, H. Kim, S. Cui, G. Lu, X. Feng, J. Chen, Adv. Mater. 2012, 24, 5610. [315] M. Xue, F. Li, J. Zhu, H. Song, M. Zhang, T. Cao, Adv. Funct. Mater. 2012, 22, 1284. [316] X.-M. Feng, R.-M. Li, Y.-W. Ma, R.-F. Chen, N.-E. Shi, Q.-L. Fan, W. Huang, Adv. Funct. Mater. 2011, 21, 2989. [317] Y. J. Lee, H. W. Park, S. Park, I. K. Song, Curr. Appl. Phys. 2012, 12, 233. [318] A. Yuan, Q. Zhang, Electrochem. Commun. 2006, 8, 1173. [319] C. Lin, J. A. Ritter, B. N. Popov, J. Electrochem. Soc. 1999, 146, 3155. [320] Z. Chen, Y. Qin, D. Weng, Q. Xiao, Y. Peng, X. Wang, H. Li, F. Wei, Y. Lu, Adv. Funct. Mater. 2009, 19, 3420. [321] W. Chen, Z. Fan, L. Gu, X. Bao, C. Wang, Chem. Commun. 2010, 46, 3905. [322] X. Zhang, W. Shi, J. Zhu, D. J. Kharistal, W. Zhao, B. S. Lalia, H. H. Hng, Q. Yan, ACS Nano 2011, 5, 2013. [323] L. Basiricý, G. Lanzara, Nanotechnology 2012, 23, 305401. [324] X. Wang, G. Li, Z. Chen, V. Augustyn, X. Ma, G. Wang, B. Dunn, Y. Lu, Adv. Energy Mater. 2011, 1, 1089. [325] C.-M. Chuang, C.-W. Huang, H. Teng, J.-M. Ting, Compos. Sci. Technol. 2012, 72, 1524. [326] M. Zhi, A. Manivannan, F. Meng, N. Wu, J. Power Sources 2012, 208, 345. [327] L. Yang, S. Cheng, Y. Ding, X. Zhu, Z. L. Wang, M. Liu, Nano Lett. 2012, 12, 321. [328] A. Ghosh, E. J. Ra, M. Jin, H.-K. Jeong, T. H. Kim, C. Biswas, Y. H. Lee, Adv. Funct. Mater. 2011, 21, 2541. [329] X. Lu, T. Zhai, X. Zhang, Y. Shen, L. Yuan, B. Hu, L. Gong, J. Chen, Y. Gao, J. Zhou, Y. Tong, Z. L. Wang, Adv. Mater. 2012, 24, 938. [330] M.-S. Wu, Y.-P. Lin, C.-H. Lin, J.-T. Lee, J. Mater. Chem. 2012, 22, 2442. [331] Y.-Q. Zhao, D.-D. Zhao, P.-Y. Tang, Y.-M. Wang, C.-L. Xu, H.-L. Li, Mater. Lett. 2012, 76, 127. [332] Z.-S. Wu, G. Zhou, L.-C. Yin, W. Ren, F. Li, H.-M. Cheng, Nano Energy 2012, 1, 107.

ChemSusChem 2015, 8, 2284 – 2311

www.chemsuschem.org

[333] X. Zhao, L. Zhang, S. Murali, M. D. Stoller, Q. Zhang, Y. Zhu, R. S. Ruoff, ACS Nano 2012, 6, 5404. [334] C. Xiang, M. Li, M. Zhi, A. Manivannan, N. Wu, J. Mater. Chem. 2012, 22, 19161. [335] H. Kim, M.-Y. Cho, M.-H. Kim, K.-Y. Park, H. Gwon, Y. Lee, K. C. Roh, K. Kang, Adv. Energy Mater. 2013, 3, 1500. [336] Y. Hou, L. Chen, P. Liu, J. Kang, T. Fujita, M. Chen, J. Mater. Chem. A 2014, 2, 10910. [337] L. Hu, W. Chen, X. Xie, N. Liu, Y. Yang, H. Wu, Y. Yao, M. Pasta, H. N. Alshareef, Y. Cui, ACS Nano 2011, 5, 8904. [338] X.-C. Dong, H. Xu, X.-W. Wang, Y.-X. Huang, M. B. Chan-Park, H. Zhang, L.-H. Wang, W. Huang, P. Chen, ACS Nano 2012, 6, 3206. [339] H. Jiang, J. Ma, C. Li, J. Mater. Chem. 2012, 22, 16939. [340] J. Yan, Q. Wang, C. Lin, T. Wei, Z. Fan, Adv. Energy Mater. 2014, 4, 1400500. [341] L. Liu, Z. Niu, L. Zhang, W. Zhou, X. Chen, S. Xie, Adv. Mater. 2014, 26, 4855. [342] H. Fan, H. Wang, N. Zhao, X. Zhang, J. Xu, J. Mater. Chem. 2012, 22, 2774. [343] J. Xu, K. Wang, S.-Z. Zu, B.-H. Han, Z. Wei, ACS Nano 2010, 4, 5019. [344] Q. T. Qu, P. Zhang, B. Wang, Y. H. Chen, S. Tian, Y. P. Wu, R. Holze, J. Phys. Chem. C 2009, 113, 14020. [345] T. Tomko, R. Rajagopalan, M. Lanagan, H. C. Foley, J. Power Sources 2011, 196, 2380. [346] H. A. Mosqueda, O. Crosnier, L. Athouel, Y. Dandeville, Y. Scudeller, P. Guillemet, D. M. Schleich, T. Brousse, Electrochim. Acta 2010, 55, 7479. [347] H. Q. Wang, Z. S. Li, Y. G. Huang, Q. Y. Li, X. Y. Wang, J. Mater. Chem. 2010, 20, 3883. [348] X. Zhao, C. Johnston, P. S. Grant, J. Mater. Chem. 2009, 19, 8755. [349] B. G. Choi, M. Yang, W. H. Hong, J. W. Choi, Y. S. Huh, ACS Nano 2012, 6, 4020. [350] J. Zhang, J. Jiang, H. Li, X. S. Zhao, Energy Environ. Sci. 2011, 4, 4009. [351] J. Yan, Z. Fan, W. Sun, G. Ning, T. Wei, Q. Zhang, R. Zhang, L. Zhi, F. Wei, Adv. Funct. Mater. 2012, 22, 2632. [352] Z. Zhang, F. Xiao, L. Qian, J. Xiao, S. Wang, Y. Liu, Adv. Energy Mater. 2014, 4, 1400064. [353] J. Y. Luo, Y. Y. Xia, J. Power Sources 2009, 186, 224. [354] H. S. Choi, J. H. Im, T. Kim, J. H. Park, C. R. Park, J. Mater. Chem. 2012, 22, 16986. Received: December 31, 2014 Revised: April 20, 2015 Published online on July 3, 2015

2311


Ultrasonic modification of carbon materials for electrochemical capacitors.

carbon hybrid nanoparticle electrodes for high-capacity electrochemical capacitors.

Graphene Sheets Composites as Electrode Materials for Electrochemical Capacitors.

Hybrid supercapacitor-battery materials for fast electrochemical charge storage.

Facile synthesis of porous Mn2O3 nanoplates and their electrochemical behavior as anode materials for lithium ion batteries.

"Electroanalysis with New Electrode Materials and Advanced Electrochemical Devices".

The use of engineered protein materials in electrochemical devices.

Controllable synthesis, morphology evolution and electrochemical properties of LiFePO4 cathode materials for Li-ion batteries.

C nanosheets and microspheres as anode materials for lithium-ion batteries.

MOF-derived crumpled-sheet-assembled perforated carbon cuboids as highly effective cathode active materials for ultra-high energy density Li-ion hybrid electrochemical capacitors (Li-HECs).

Improved Electrochemical Performance of Fe-Substituted NaNi0.5Mn0.5O2 Cathode Materials for Sodium-Ion Batteries.

WS₂@graphene nanocomposites as anode materials for Na-ion batteries with enhanced electrochemical performances.

Development of hybrid materials based on sponge supported reduced graphene oxide and transition metal hydroxides for hybrid energy storage devices.

Synthesis and Electrochemical Properties of LiNi0.5Mn1.5O4 Cathode Materials with Cr3+ and F- Composite Doping for Lithium-Ion Batteries.

Hybrid nanostructured C-dot decorated Fe3O4 electrode materials for superior electrochemical energy storage performance.

One-pot rapid synthesis of core-shell structured NiO@TiO2 nanopowders and their excellent electrochemical properties as anode materials for lithium ion batteries.

Neurosurgical materials and devices.

Flexible patterned micro-electrochemical capacitors based on PEDOT.

Nanowire electrodes for electrochemical energy storage devices.

Certification and classification programs for dental materials and devices. List of certified dental materials and devices revised to Jan 1, 1977. Council on Dental Materials and Devices.

Hybrid materials.

Nanostructured Materials for Li-Ion Batteries and Beyond.

carbon composite powders as anode materials for lithium-ion batteries.

Nanostructured Electrode Materials for Electrochemical Capacitor Applications.