Synthesis and application of iron-based

0 downloads 0 Views 4MB Size Report
May 24, 2018 - So Nanoscience Letters (SCIRP),. Research & Reviews in ...... D. Mecerreyes and M. Armand, ACS Macro Lett., 2016, 5,. 59–64. 206 Y. Zeng ...
Journal of

Materials Chemistry A View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

REVIEW

Cite this: J. Mater. Chem. A, 2018, 6, 9332

View Journal | View Issue

Synthesis and application of iron-based nanomaterials as anodes of lithium-ion batteries and supercapacitors Shijin Yu, a Vincent Ming Hong Ng,b Fajun Wang,c Zhuohao Xiao, Ling Bing Kong, *ab Wenxiu Que*e and Kun Zhou *f

d

Cuiyun Li,a

Lithium-ion batteries and supercapacitors have great potential as power supplies in portable electronic devices and electric vehicles. Their performance depends greatly on the properties of electrode materials. Many attempts have been devoted to the development of new electrode materials with advanced electrochemical performances. Due to their high theoretical specific capacitance, low cost and non-toxicity, iron-based materials are considered as very promising candidates for anode materials. However, low electrical conductivity and poor cycle stability are two major problems plaguing iron-based materials. Nanomaterial design has emerged as a promising solution to these fundamental issues in LIBs and SCs. Here, we review the synthesis of iron oxide (Fe2O3 and Fe3O4) nanomaterials with various structures, including 1D (nanorods, nanowires, and nanotubes), 2D (nanosheets) and 3D (nanospheres, hollow nanostructures, flower-like structures, and nanoarrays). Nanocomposites, consisting of iron oxides and different supports (such as carbonaceous materials, other metal oxides, and polymers), are also covered in this review. Furthermore, the synthesis and structural characteristics of iron hydroxides (FeOOH) and iron sulfides (FeS2) will also be elaborated. Finally, applications of iron-based nanomaterials in LIBs and SCs are summarized. Ultimately, we Received 20th February 2018 Accepted 9th April 2018

wish to provide an in-depth and reasonable understanding of how to effectively improve the electrochemical performance of iron-based anodes by selecting suitable nanostructures and optimizing their

DOI: 10.1039/c8ta01683f

chemical compositions. Hopefully, these concepts and strategies can be extended to other nanomaterials,

rsc.li/materials-a

as a reference for future development in the areas of energy conversion, storage and environmental protection.

a

School of Mechanical and Electronic Engineering, Jingdezhen Ceramic Institute, Jingdezhen 333403, P. R. China. E-mail: [email protected]; lbkongntunus@126. com

b

School of Materials Science and Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798

c Key Laboratory for Microstructural Control of Metallic Materials of Jiangxi Province, Nanchang Hangkong University, Nanchang 330063, P. R. China

Shijin Yu received his PhD in Materials Physics and Chemistry from Huazhong University of Science and Technology in 2009. He then served as a Senior Engineer at the Quality Metrology Supervision and Inspection Institute in Shantou, Guangdong Province (2009–2013). He is currently working at the School of Mechanical and Electronic Engineering at Jingdezhen Ceramic Institute. His research interests focus on the design and synthesis of functional nanostructured materials for energy storage and conversion applications.

9332 | J. Mater. Chem. A, 2018, 6, 9332–9367

d

School of Materials Science and Engineering, Jingdezhen Ceramic Institute, Jingdezhen 333403, P. R. China

e

Electronic Materials Research Laboratory, School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, Shaanxi, China. E-mail: [email protected]

f

School of Mechanical & Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798. E-mail: [email protected]

Ling Bing Kong has published 225 journal papers, 4 books, 16 book chapters and 80 conference presentations/abstracts. He serves as a referee for more than 50 international journals and a member of the Editorial Board and a guest editor for several journals. He is a member of the Editorial Board for So Nanoscience Letters (SCIRP), Research & Reviews in Electrochemistry, ISRN Ceramics, and Journal of Advanced Dielectrics, and an Associate/Academic Editor of American Chemical Science Journal, Indian Science and Engineering Journal and Open Physics.

This journal is © The Royal Society of Chemistry 2018

View Article Online

Review

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

1. Introduction Energy has been one of the most important and hot research topics in recent decades. With the rapid consumption of fossil fuels and the environmental crisis, renewable clean energy has attracted worldwide attention. Solar energy, wind, tidal energy and other clean energies can be used to generate electricity, so they have been extensively studied. However, there is a fatal aw of energy instability in these clean energies. A high-energydensity storage system needs to be added to achieve the effect of cutting peaks and valleys in the smart grid. In addition, the rapid development of electric vehicles for power batteries and portable consumer electronics also places high demand on the power density, energy density, transparency and exibility of energy storage systems. Electrochemical energy storage systems, which have the characteristics of high efficiency, high exibility and versatility, are currently the most successful devices for energy storage applications. Lithium-ion batteries (LIBs) and supercapacitors (SCs) are the most widely used electrochemical energy storage devices. LIBs have the highest energy density, while SCs can provide very high power density. Both devices usually consist of a cathode, an anode, an electrolyte and a separator. Electrochemical energy storage is achieved by the charging/discharging of electrons/ions. During charging and discharging, electrochemical reactions occur at the electrodes. Therefore, the performance of electrode materials plays a decisive role in determining the performance of LIBs and SCs. Graphitic carbons are the conventional anode materials for LIBs and SCs, which suffer from the drawbacks of limited specic capacity (372 mA h g1) and cycling life. Therefore, new anode materials with high capacity, long cycling lifetime and high safety are urgently demanded. Signicant achievements have been made in the search for alternative anode materials with high capacity, such as Si, intermetallic alloys and transition metal oxides. Although Si and alloy materials have high theoretical specic capacities, they suffer from large volume changes during lithium intercalation/–deintercalation.

Wenxiu Que has published more than 230 ISI journal papers, 2 books, and 7 book chapters and has been granted 22 patents. His research interests cover transparent laser ceramics, organic– inorganic hybrid materials based on organically modied silanes for photonic applications, oxide semiconductor nanostructured arrays/polymer organic–inorganic nanocomposite materials for clean energy and environment applications, semiconductor compounds and organic thin lm solar cells, as well as organic–inorganic perovskite solar cells, etc.

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

For example, the volume expansion ratio of Si electrodes aer lithium insertion can reach 400%.53,54 As one of the transition metals, iron-based materials are considered as one of the most promising anode materials for LIBs and SCs. Firstly, they have multiple valence states (Fe0, Fe2+ and Fe3+) to provide rich redox pairs, such as Fe0/Fe2+, Fe0/Fe3+ and Fe2+/Fe3+. Therefore, ironbased materials have a higher theoretical capacity than the graphite carbon electrode. Secondly, iron-based materials have stable and wide operating windows at negative potentials and thus they are well suited for the fabrication of high-performance asymmetric supercapacitors (ASCs). Moreover, iron is the most abundant transition metal in the Earth's crust (Fig. 1a). In addition, its cost is roughly equivalent to that of commercial activated carbon, thus making it economically viable for largescale applications. Lastly, it is less toxic and greener than other transition metal oxides/hydroxides. Therefore, many papers have been published regarding the synthesis and application of iron-based nanomaterials over the past ve years (Fig. 1b and c). However, as an electrode material, iron-based materials have poor conductivity and cycle stability. It has been acknowledged that nanomaterials could be a promising solution to these issues. Nanomaterials have been studied for decades and great achievements have been made, especially in the design and preparation of controlled nanostructures. At present, various iron-based nanostructures have been developed, such as onedimensional (1D) structures (nanowires, nanotubes, and nanorods), two-dimensional (2D) nanosheets and threedimensional (3D) structures (nanospheres, hollow nanostructures, ower-like nanostructures, and arrays). These nanostructures can be obtained by using various methods, such as sol–gel,56 chemical vapor deposition,50 electrospinning,29,57 hydro/solvothermal,6,20,58,59 electrochemical deposition,60 spray pyrolysis,61 detonation,62 corrosion32 and electron beam deposition.51,63 In the past few decades, although many high-quality reviews64–66 have been published, covering the preparation of iron-based oxides and their applications in LIBs or SCs, no attention has been paid to low-dimensional nanostructured (1D

Kun Zhou is an Associate Professor at the School of Mechanical & Aerospace Engineering, Nanyang Technological University, Singapore. His research interests focus on micro/nanomechanics of materials and structures, novel computational methods for modeling material behaviour, and sustainable energy and green technologies. He worked as a Postdoctoral Fellow at the Center for Surface Engineering & Tribology, Northwestern University, USA from 2007 to 2010.

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9333

View Article Online

Journal of Materials Chemistry A

Review

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

2.1. 1D Fe2O3 nanostructures

Fig. 1 (a) Elemental abundance in the Earth's crust, (b) line chart of the research trend of iron-based materials of LIBs and SCs in recent years, (c) pie chart of the ratio of iron materials applied in LIBs and SCs.

and 2D) iron oxides. More recently, a few reviews67,68 have referred to the preparations and applications of iron oxides, but only emphasizing hollow structures in energy storage and conversion. Moreover, there is a lack of systematic review on the preparation and properties of iron oxide nanocomposites. In addition, the research on iron-based hydroxides and FeS2 has made great progress, but so far there has been no relevant review regarding their application in LIBs and SCs. Here, we attempt to provide a comprehensive review on the preparation, characterization and application of various iron-based materials, particularly the synthetic strategies that have newly emerged in the past ve years. The application of iron-based nanomaterials as LIB and SC electrodes, the analysis of the relationship of different morphologies, the composition and electrochemical performance are the main highlights of this review article.

2. Synthesis of iron oxide (Fe2O3) nanostructures Among various iron-based materials, Fe2O3 has attracted a great deal of attention, as an anode of LIBs and SCs, due to its high theoretical specic capacitance, abundant resources and excellent environmental compatibility. Various methods have been used to synthesize Fe2O3 nanostructures, such as hydro/solvothermal, electrospinning, spray pyrolysis, sol–gel, vapor deposition and electrodeposition. Unfortunately, it suffers from poor conductivity, volumetric expansion and particle agglomeration. Nanostructures have proven to be an effective solution to address these issues. In this section, 1D, 2D and 3D Fe2O3 nanostructures are thoroughly reviewed.

9334 | J. Mater. Chem. A, 2018, 6, 9332–9367

Generally, 1D nanostructures are dened as those that have a high aspect ratio of length along the longitudinal direction to the width along the transverse axis. When they are densely packed into electrode materials, they have a large specic surface area and can form suitable internal voids. Various 1D iron oxide nanostructures, such as nanorods, nanowires, nanoribbons and nanotubes, have been synthesized and applied not only to LIBs and SCs but also to waste treatment, sensors and catalysts.69–71 2.1.1 Fe2O3 nanorods. Fe2O3 nanorods can be synthesized by using various approaches, such as solution methods, electrospinning methods and electrochemical deposition.57,60,72,73 For example, Kim et al.60 rst deposited and grew FeOOH nanorods by using electrochemical deposition processes on the surface of carbon nanobers, from which Fe2O3 nanorods were obtained aer calcination at 400  C. It was found that length of the nanorods increased with increasing current density during the deposition. Cherian et al.57 prepared Fe2O3 nanorods by using electrospinning and heat treatment. Iron acetylacetonate (Fe(acac)3) was added to an alcoholic solution of polyvinylpyrrolidone (PVP) to form a PVP/Fe(acac)3 composite precursor. The electrospun PVP/Fe(acac)3 nanorods were calcined in air at 500  C for 5 h to obtain Fe2O3 nanorods, which had an average diameter of 150 nm. Microstructural analysis indicated that these Fe2O3 nanorods are formed through the assembly of Fe2O3 nanoparticles. In recent years, hydrothermal methods have been widely used to prepare nanorods.74–77 For instance, Balogun et al.58 prepared a-Fe2O3 nanorods by using a hydrothermal process. Zhang et al.76 deposited a-Fe2O3 nanorods on carbon cloth from FeCl3 and Na2SO4 precursor solutions through hydrothermal treatment. The a-Fe2O3 nanorods are single-crystalline with diameters of 100–150 nm and are vertically grown on the carbon bers. Chen et al.16 obtained Fe2O3 nanorods with exposure of the (001) and (010) planes by using a hydrothermal method, with FeCl3 and NaOH precursor solutions. The resulting Fe2O3 nanorods have a very uniform size distribution with an average length of about 496 nm, while their widths and thicknesses are 50 and 15 nm, respectively, as shown in Fig. 2a and b. The sample has a specic surface area of 26.81 m2 g1, which is larger than those of Fe2O3 nanosheets. The crystal structure of nanorods has a signicant effect on their electrochemical properties.78–81 Controlling the morphology of the nanorods is an effective strategy to improve the electrochemical properties of Fe2O3 as electrodes for LIBs and SCs. A variety of shape controlling agents (SCAs) have been used to assist the hydrothermal synthesis of nanorods, in order to control their morphologies.82,83 With ferrous sulphate heptahydrate (FeSO4$7H2O) and sodium acetate (CH3COONa) as raw materials, Chaudhari et al.84 synthesized porous Fe2O3 nanorods by using a wet chemical approach. CH3COONa was used as the precipitating agent, which slowly released hydroxide ions during the reaction, so as to control the growth rate of the nanorods. Li et al.44 prepared uniform a-Fe2O3 nanorods with an aspect ratio of >10 by using a 1,2-propanediamine-assisted

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

Fig. 2 SEM (a) and TEM (b) images of the Fe2O3 nanorods [reprinted with permission from ref. 16, copyright 2016, Nature]. SEM images of the products obtained at different volume ratios of aqueous FeCl3 solution to 1,2-propanediamine, (c) without 1,2-propanediamine and (d) at a 1/1 ratio [reprinted with permission from ref. 44, copyright 2009, IOP].

hydrothermal method. The results showed that 1,2-propanediamine not only provided OH during the formation of the Fe2O3 nanorods, but also played an important role in determining the morphology of the nanorods, as illustrated in Fig. 2c and d. The SCA adjusted the substituents on the diamine unit to affect the electron density and the steric interaction, which facilitated control over the crystal morphology.83 Srinivasan's group85 studied the controlled formation of a-Fe2O3 nanorods by using different vicinal diamine derivatives, such as ethylenediamine, N-methylethylenediamine, 2,3-diaminobutane, and 1,2-diaminopropane. Due to the electron density enhancement (+M effect) and steric hindrance, the location and number of substituents (H and H3C) affect the directional structure formation of ferric iron species. 1,2-diaminopropane exhibits the strongest shape controlling capability. The prepared nanorods have an average length of about 390 nm, with an aspect ratio of about 5.0. In this case, the H3C-groups at the same C amine group enhance the chelation ability of the iron species by increasing the electron density (+M effect), while causing less steric hindrance. It is observed that the shape controllability decreased in the order of 1,2-diaminopropane > ethylenediamine $ 2,3-diaminobutane [ N-methylethylenediamine. Through the selection of SCAs and the adjustment of the Fe3+ concentration, the Fe2O3 nanorods could be controlled to have lengths in the range from 240–400 nm and aspect ratios of 2.6– 5.7. The nanorods, with lengths of 240–280 nm and aspect ratios of 2.6–3.0, exhibited optimized electrochemical properties. 2.1.2 Fe2O3 nanowires. Fe2O3 nanowires are mainly used for biocatalysis and sensors.69 In recent years, they have been explored as electrode materials for LIBs and SCs. Fe2O3 nanowires can be efficiently synthesized by using traditional methods, such as chemical precipitation. Wang's group2 mixed FeCl3 solution with NaBH4 aqueous solution to obtain

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

a precipitate, which was annealed at 500  C to form Fe/Fe2O3 nanowires with a core–shell structure. The Fe/Fe2O3 nanowires have lengths from several micrometers to tens of micrometers and diameters of 80–130 nm. The shell thickness of the nanowire is about 4.0 nm, as observed in Fig. 3a and b. Lu's group29 developed Pt-doped Fe2O3 nanowires by using electrospinning. The Fe2O3 nanowires contained some micropores, with a diameter of about 100 nm. Pt doping has a signicant effect on the morphology of the nanowires, as demonstrated in Fig. 3c and d. With increasing content of Pt, the nanoparticles become more and more compact, so that the porosity of the nanowires was gradually reduced, but the cracks on the walls became more obvious. With acetylacetone iron as the precursor and polyvinyl pyrrolidone as raw material, Sivakumar's team86 synthesized Fe2O3 porous nanobres by using electrospinning. The nanobers have diameters in the range of 700–750 nm, with a specic surface area of about 61 m2 g1, an average pore size of 5.7 nm and a pore volume of 0.087 cm3 g1. Bioinspired strategies are a new method to synthesize nanomaterials. A hierarchical FeOOH nanostructure array was synthesized by using a bioinspired gas–liquid diffusion method at the air–water interface.87 Similarly, Huang and co-workers15 fabricated a Fe2O3 nanowire network@graphene transparent lm (FNW@Gr-TF) by using a bioinspired gas–liquid diffusion method at the air-solution interface. During the reaction process, NH3 vaporized and diffused into the ferrous sulfate solution. At the gas–liquid junction, NH3 reacted with ferrous sulfate to form a yellow lm. Excess NH3 continued to diffuse

Fig. 3 SEM (a) and HRTEM (b) images of the Fe@Fe2O3 nanowires [reprinted with permission from ref. 2, copyright 2016, Wiley]. SEM images of the products obtained at different mole number ratios of Pt to Fe2O3: (c) 0 and (d) 3 mol% [reprinted with permission from ref. 29, copyright 2018, Elsevier]. SEM (e) and TEM (f) images of the graphene wrapped Fe2O3 nanowires [reprinted with permission from ref. 15, copyright 2017, IOP]. SEM (g) and TEM (h) images of the Ag NWs@Fe2O3 nanowires [reprinted with permission from ref. 1, copyright 2015, Wiley].

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9335

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

into the solution, reacting with the ferrous sulfate solution underneath the top layer, where nanowires developed from the nucleus and grew towards the solution. At the end of the reaction Fe2O3 nanowires were generated with a diameter of 10.5 nm. Then, ITO/PET substrate and graphene sheets were respectively introduced as the substrate and the conductive layer. The nanowire layers with a thickness of about 500 nm showed a porous network structure. Aer the thin and transparent graphene wrapped the Fe2O3 nanowire layers, there are still many transparent holes, as seen in Fig. 3e and f. Using the existing nanowires as a template, a metal oxide source solution is coated on the surface of the nanowires to prepare metal oxide nanowires. Zhang's team88 has synthesized Fe2O3 nanobers on the surface of carbon bers by electrospinning. Geng and co-workers1 have synthesized a layer of Fe2O3 on the surface of silver nanowires to form Ag@Fe2O3 nanocables similar to the coaxial cable structure. As the iron source concentration increases, the Fe2O3 nanosheets become uniform and dense, and the thickness increases accordingly. Representative SEM and TEM images of the Ag nanowires and Ag@Fe2O3 nanowires are depicted in Fig. 3g and h. 2.1.3 Fe2O3 nanotubes. Template methods are the most common for the preparation of nanotubes. Aer preparing the Fe2O3 coating on the surface of the nano-metal wire, the metal wire template is etched to form the nano-tube. Chen et al.89 have thermally decomposed Fe(NO3)3 precursors to prepare Fe2O3 nanotubes by using an anodic aluminum membrane as a template. Aer impregnation with Fe(NO3)3$9H2O solution, the templates were dried under vacuum and annealed. The solution impregnation and annealing process was repeated twice. NaOH solution was used as an etchant to remove the alumina template. The preparation process can well copy the template morphology. The Fe2O3 nanotubes arrange in an ordered manner in the alumina matrix. Similarly, the length of the Fe2O3 nanotubes is 60 microns, as high as the template channels. Geng and co-workers1 have synthesized a layer of Fe2O3 on the surface of silver nanowires to form Ag@Fe2O3 nanocables similar to the coaxial cable structure. When Ag was corroded by aqueous ammonia, the Ag@Fe2O3 nanocables were transformed into hollow Fe2O3 nanotubes. Representative SEM and TEM images of the Fe2O3 nanotubes are shown Fig. 4a and b. Ma and co-workers14 have synthesized FeOOH nanotubes with MoO3 nanorods as a template. By adjusting the molar ratio of Fe3+ to MoO3 and the hydrolysis temperature of Fe3+, nanotubes with different thicknesses and diameters can be obtained. Aer calcination, the FeOOH nanotubes are transformed into Fe2O3 nanotubes with wall thicknesses ranging from 50 to 128 nm, as illustrated in Fig. 4c. When the molar ratio of Fe3+ to MoO3 is 2 : 1 and the hydrolysis temperature of Fe3+ is 70  C, the resulting a-Fe2O3 nanotubes are built up of nanoparticles with a diameter of 10 nm, as seen in Fig. 4d. The Fe2O3 nanotubes possess a BET surface area of as high as 149 m2 g1. Sacricial template methods are a strategy for the preparation of nanotubes. For example, by using Cu nanowires as a sacricial template, Lou's group90 has prepared Fe2O3 nanotubes by the substitution reaction method. The prepared Fe2O3 nanotubes well replicated the Cu nanowire structure. The

9336 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

SEM (a) and TEM (b) images of the Ag nanowires@Fe2O3 nanotubes [reprinted with permission from ref. 1, copyright 2015, Wiley]. SEM (c) and HRTEM (d) images of the Fe2O3 nanotubes with a molar ratio of Fe3+ to MoO3 of 2 : 1 at a hydrolysis temperature of 70  C [reprinted with permission from ref. 14, copyright 2015, Springer]. SEM images of FeC2O4$2H2O (e), the FeOOH precursor (f) and Fe2O3 nanotubes (g) [reprinted with permission from ref. 33, copyright 2017, Elsevier]. SEM (h) and TEM (i) images of the as-synthesized Fe2O3 nanotube and corresponding high resolution TEM image showing the lattice fringes (inset) of the sample [reprinted with permission from ref. 48, copyright 2017, Elsevier]. Fig. 4

diameter of the Fe2O3 nanotubes is in the range of 50–100 nm, while the thickness of the nanotube wall is about 10 nm. Despite the very thin nanotubes, the wall thickness is very uniform and has good structural stability. Gu et al.33 chose FeC2O4$2H2O nanorod precursors as a self-sacricial template to obtain uniform Fe2O3 nanotubes. During the growth of FeOOH nanotubes, FeC2O4$2H2O nanorod precursors were used as raw materials and the initial framework. The FeOOH nanotubes were converted to hierarchical Fe2O3 nanotubes through heat treatment. The Fe2O3 nanotubes well inherited the morphology and size of the precursors. The Fe2O3 nanotubes retained a tubular morphology, without collapse aer calcination. Many nanosheets are randomly distributed on the surface of the Fe2O3 nanotubes, as observed in Fig. 4e–g. The tubular structure and high surface area allow the electrode material to provide more electroactive sites and greater scalability to alleviate volume changes during the charge–discharge cycles. In addition to the template method, there are other methods that can be used to prepare Fe2O3 nanotubes, such as hydrothermal and solution impregnation methods. With the aid of NH4H2PO4 solution, Lee et al.59 prepared Fe2O3 nanotubes by hydrothermally treating FeCl3 solution. The obtained nanotubes had an average length of about 370 nm, with the wall thickness in the range of 15–40 nm. Xiang et al.48 adopted a hydrothermal and bio-inspired strategy to synthesize Fe2O3 nanotubes. Using tannic acid as coating agent, a composite coating was prepared on the surface of the hydrothermally synthesized Fe2O3 nanotubes. The Fe2O3 nanotubes have lengths of 300–400 nm and an outer diameter of about 90 nm, as shown in Fig. 4h–i. Fig. 4h shows that Fe2O3 nanotubes contained pores, as indicated by the arrows.

This journal is © The Royal Society of Chemistry 2018

View Article Online

Review

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

2.2. 2D Fe2O3 nanostructure The lateral dimension of 2D nanosheets is at least two orders of magnitude greater than the thickness, which combine well the advantages of 0D and 1D nanostructures. 2D nanosheets have unique mechanical properties that can be wrinkled and twisted or even folded into 3D nanostructures. Therefore, 2D nanosheets have the ability to buffer the lithiation/delithiation volume change of lithium ions.91 In addition, nanosheets have a larger surface area and shorter Li+ ion diffusion paths. They also provide 2D transmission channels between layers to facilitate pseudocapacitive storage of the electrodes.92 These supercial special electronic structures make nanosheets well suited for energy devices, thus attracting much interest in recent years.93 Chen's group16 used a hydrothermal method to prepare Fe2O3 nanosheets with exposure of the (001) plane. The nanostructure exhibited a discharge capacity of 865 mA h g1 at a rate of 200 mA g1 over 80 cycles, which is higher than that (456 mA h g1) of Fe2O3 nanorods by 90%. Firstly, the (001) plane of haematite has a high packing density of Fe3+ and O2, which is a key factor for high electrochemical performance. In comparison, the (010) facet is the mainly exposed crystal plane in nanorods, with a ratio of (010) plane to (001) plane of 77 : 23. The proportion of the (001) plane in nanosheets is almost 100%. Secondly, the charge transfer resistances of Fe2O3 nanosheets is 53 U, which is smaller than that (179 U) for Fe2O3 nanorods. As a result, Fe2O3 nanosheets offer a more conductive pathway and lower Li+ migration resistance. Thirdly, according to potentiostatic intermittent titration data, the average DLi+ value of the nanosheets is 2.2  1010 cm2 s1, which is increased by about 15.7% as compared with those of the nanorods. The high kinetic parameters are responsible for the discharge cycle stability and rate ability. Due the anisotropic structure of Fe2O3 crystals, it is possible to enlarge the exposure of the (001) facet and control the growth of Fe2O3 nanostructures along the [001] direction.94 Fe2O3 nanosheets along the [001] direction with preferentially exposed (001) facets have been grown by using hydro/solvothermal, template-assisted oriented and acid etching methods.6,95,96 Lu et al.94 prepared Fe2O3 nanodiscs by using a hydrothermal method with NaAc as the control agent. It has been well known that the hydrolysis of Fe3+produces H+, as shown in eqn (1). As the concentration of H+ in FeCl3 solution is sufficiently high, the hydrolysis of Fe (H2O)63+ is inhibited. In the absence of NaAc, the number of Fe2O3 nuclei generated by the hydrothermal processes is so less that the crystals grow very fast. With the presence of NaAc, two equilibrium reactions occur in the solution, as shown in eqn (2) and (3). With increasing the amount of NaAc, the original balance is broken. The concentration of H+ is decreased, while that of OH is increased. Then, eqn (1) will shi in a forward direction, i.e., hematite precipitation takes place, thus producing a large number of nuclei. As a consequence, the crystals will have a relatively small size, which appear as hexagonal nanodisks with exposed (001) planes and thicknesses of only about 27 nm. The concentrations of OH and H+ have a signicant effect on the morphology of the nal products.97

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

2Fe(H2O)63+ 4 Fe2O3 + 6H+ + 12H2O

(1)

CH3COO + H+ / CH3COOH

(2)

CH3COO + H2O 4 CH3COOH + OH

(3)

Studies have indicated that it is possible to control the directional growth of Fe2O3 nanocrystals by introducing foreign matters or ions in order to better tailor the crystal morphology and explore the crystal growth mechanism. For instance, Al3+ and Ni2+ ions have been shown to be suitable crystal growth control agents.98 Jin et al.6 synthesized Fe2O3 nanosheets with a thickness of 1.3 nm by using a hydrothermal method, as shown in Fig. 5a and b. It was found that the addition of Al3+ is crucial for the formation of Fe2O3 nanosheets. The Fe2O3 nanoparticles synthesized without the presence of Al3+ were irregular in shape. When Al3+ ions were added, they were adsorbed onto the (001) surface of the Fe2O3 nanoparticles. Therefore, Fe2O3 nanoparticles grew preferentially along the [100] direction, while the growth of (001) is inhibited. Jin's group27 prepared Fe2O3 hexagonal nanoplates by using a molten salt method with the assistance of Ni2+ ions. The thickness and morphology of the Fe2O3 nanoplates could be tailored by controlling the content of Ni2+. The Ni2+ ions were adsorbed onto the (001) facet through coordination bonding with the O atoms, resulting in slower growth along the (001) planes. With increasing content of Ni2+, the nanosheets became more and more uniform, the edges of the items became more and more sharp, and the thickness of nanoplates decreased rapidly from 300 nm to 50 nm, as shown in Fig. 5c and d. In addition to direct synthesis in solution, template and substrate methods have also been employed to prepare 2D Fe2O3 nanostructures. Liu and co-worker51 fabricated Fe2O3 nanolms by using Al2O3 as the sacricial layer. The Al2O3 lms and the Fe nanolms were sequentially deposited on a glass substrate. Then, Fe nanolms were obtained by etching the Al2O3 sacricial layer with aqueous NaOH. Aer annealing in air

Fig. 5 SEM (a, b) images of the ultrathin Fe2O3 nanosheets [reprinted with permission from ref. 6, copyright 2017, Elsevier]. SEM images of the Fe2O3 nanosheets free of Ni2+ ions (c) and with 0.85 wt% NiCl2 (d) [reprinted with permission from ref. 27, copyright 2014, Royal Society of Chemistry]. SEM image (e) of the as-synthesized Fe2O3 nanosheets [reprinted with permission from ref. 46, copyright 2016, Royal Society of Chemistry].

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9337

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

at 450  C, 2D Fe2O3 nanostructures were obtained, with thicknesses of 103–165 nm. The Fe2O3 nanolms were free-standing, with a relatively rough surface due to the assembly of nanoparticles. Because of the presence of internal differential stress, the Fe2O3 nanolms were prone to wrinkling and bending. It is expected that such a structural feature should be very effective to buffer the lithium/delithiation induced strain. Wu's team46 synthesized a spinous Fe2O3 nanosheet hierarchical structure on a Ni foam substrate by using a hydrothermal process. The assynthesized spinous Fe2O3 sample consists of a large number of nanosheets, with thicknesses at the nanometer scale, as seen in Fig. 5e.

2.3. 3D Fe2O3 nanostructures The synthesis of 3D Fe2O3 nanostructures of Fe2O3 has been extensively reported in the open literature. Various 3D Fe2O3 nanostructures, such as spherical nanostructures,20,61,99,100 hollow nanostructures,11,38,101 nanoarrays,8,25,39 ower-like nanostructures4,10,102 and microbox structures,103 have been synthesized by using different synthetic methods. 2.3.1 Spherical structures. 3D spherical nanostructures have been widely studied for the applications not only in LIBs and SCs but also in water treatment and sensors. In this section, the synthesis of 3D Fe2O3 spherical nanostructures will be presented. Based on whether a template is used or not, there are two categories: (i) template synthesis20,61,104 and a (ii) templatefree method.70,100,105 2.3.1.1 Template-based approaches. Template synthesis is a simple and stable strategy to obtain spherical nanostructures. By selecting a specic template, one can readily control the shape, size and uniformity of the nal products. In general, template synthesis consists mainly of four steps: (1) synthesis of the template, (2) modication of the template surface, (3) deposition of the designed material and (4) deletion of the template. Each step is important, but their importance is slightly different case by case. For instance, the second step to introduce functional groups on the template surface can effectively solve the problem of incompatibility between the template surface and the shell material. The third step requires effective deposition of the designed material on the shell material at the micro–nanolevel, which is generally considered to be the most challenging step. The second step of the template surface modication and the third step of the surface deposition can be carried out simultaneously, with the formation of a dense coating on the template surface. Li's group20 reported the synthesis of Ag–Fe2O3 hollow spheres by using Ag@C nanosphere templates. Fig. 6a shows a schematic diagram describing the synthesis process of Ag– Fe2O3 nanospheres. The Ag@C template has diameters of 650– 700 nm and each sphere consists of a carbon shell with a thickness of 230 nm and an Ag core with a diameter of 150 nm. Ag–Fe2O3 hollow spheres were obtained aer the growth and annealing of Fe2O3. The diameter of the Ag–Fe2O3 nanospheres is about 600 nm, which is almost the same as that of the Ag@C template sphere. The surface of the Ag–Fe2O3 hollow nanospheres is covered with numerous Fe2O3 nanorods which are

9338 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

60–70 nm in length and 30–40 nm in width, as illustrated in Fig. 6b and c. In addition, the Ag nucleus decomposed into nano-sized particles of about 10 nm in size and uniformly distributed on the surface of the Fe2O3 spheres. Chen's group61 fabricated ultrane Fe2O3 nanoparticles embedded in nitrogendoped hollow carbon spheres by pyrolyzing a Fe-based zeolite imidazole framework (Fe–ZIF). The Fe–ZIF precursor is composed of Fe2+ ions and a 2-methylimidazolate linker. When the Fe–ZIF precursor was annealed in air at 620  C, the 2methylimidazolate linker was converted to a nitrogen-doped carbon network which served as carbon and nitrogen sources. At the same time, Fe2+ ions were oxidized to form Fe2O3 nanoparticles, which were uniformly dispersed in the nitrogendoped carbon network. This unique structure can effectively prevent agglomeration of the Fe2O3 nanoparticles. In addition, Fe atoms served as catalysts to promote rapid carbonation of the carbon-containing linker, leaving a hollow structure. When such Fe2O3 nanospheres are used as electrode materials for LIBs, the hollow structure can act as a reservoir for Li+ ions and electrolytes, effectively accommodate huge volumetric changes and shorten the path of Li+ ion diffusion and electron transfer. Similar to the traditional template, sacricial templates are also an effective approach in controlling the structural orientation and determining the shape and size of the sphere. However, the signicant difference is that the sacricial template is partially or completely consumed during the reaction. The sacricial template forms a shell structure by sacricing the template itself, so that additional surface functionalization steps can be omitted. Therefore, this method is generally considered to be more efficient, especially when the template is completely depleted during the shell formation process. These attractive features have been widely applied to the formation of a variety of hollow spherical structures. A large number of special formation mechanisms can be proposed to understand the process of sacricial template synthesis of hollow spheres, such as the Kirkendall effect106–110 and galvanic replacement.99

(a) Schematic illustration of a template-guided synthesis mechanism of the Ag–Fe2O3 nanospheres, (b) SEM images of the Ag@C templates and (c) Ag–Fe2O3 composites [reprinted with permission from ref. 20, copyright 2017, Nature]. Fig. 6

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

Yan's group99 synthesized a-Fe2O3 hierarchical hollow spheres through galvanic replacement reactions with CuO nanospheres. The CuO nanospheres and FeCl2 were mixed in solution, sealed in an autoclave and held at 170  C for 30 min. In this process, Cu atoms in the CuO nanospheres were replaced by Fe ions to form hollow a-Fe2O3 nanospheres. Cho et al.17 used a combination of spray pyrolysis and the nanoscale Kirkendall diffusion effect to synthesize new structures called ‘hollow nanoball agglomerates’. Scheme 1 shows the formation mechanism of the hollow nanosphere aggregates and the chemical conversion process of the FeOx–carbon nanosphere surface. Aer polymerization and carbonization in N2, droplets of the spray pyrolysis were converted into FeOx–carbon composite spherical powders (Scheme 1-a-②). The FeOx powders were then reduced to dozens of nanometers of Fe nanocrystals and uniformly dispersed in the carbon microspheres (Scheme 1-a-③ and Scheme 1-b-①). In an oxidizing atmosphere, the surface of Fe nanocrystals is partially oxidized to form a thin Fe2O3 layer (Scheme 1-a-④ and Scheme 1-b-②). Due to the Kirkendall effect, Kirkendall voids are generated in the vicinity of the Fe/Fe2O3 interface (Scheme 1-b-③). As the Fe nanocrystals are completely oxidized to iron oxide, hollow Fe2O3 nanosphere agglomerates are formed (Scheme 1-a-⑤ and Scheme 1-b-④). 2.3.1.2 Template-free approaches. Template synthesis of microspheres usually involves complicated synthesis procedures and thus has high costs. Moreover, the template residue can also affect the activity of the prepared materials. Therefore, various template-free methods have been developed to prepare complex microsphere structures.111,112 Hydrothermal methods to obtain hollow nanospheres are based on the Ostwald ripening process. Nanoparticles in the hydrothermal solutions aggregate into spheres to reduce the total surface energy. The growth process involves the continuous consumption of small crystals or particles, thus leading to the formation of larger nanospheres. Chai and co-workers113 synthesized Fe2O3@C nanospheres (50 nm) by using a one-step hydrothermal method. Small Fe2O3 nanocrystals (8 nm) were embedded in carbon nanospheres. Due to their small size and well-crystallized structure, these Fe2O3 nanocrystals would ensure a short diffusion distance for electrons and ions. In addition, the carbon shells not only provide high electrical conductivity, but also buffer the volumetric expansion/contraction during the process of lithiation/delithiation. A solvothermal method is developed on the basis of a hydrothermal method, which uses organic solvents rather than water. Pan's group100 prepared Fe2O3 nanocrystalline microspheres consisting of 52 nm nanocrystals by using a solvothermal reaction and thermal oxidation. Zhu and Xu114 prepared Fe2O3 mesoporous microspheres by using a surfactant-free solvothermal method combined with precursor thermal transformation. Fe(NO3)3 and tartaric acid were dissolved in dimethylformamide (DMF). Aer reaction at 160  C for 8 h, ferrous tartrate (C4H4O6Fe) precursor microspheres were obtained. The precursor microspheres were further calcined in air at 320  C, resulting in gFe2O3 mesoporous microspheres. The as-prepared g-Fe2O3 mesoporous microspheres have a high specic surface area of 138.6 m2 g1.

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

Scheme 1 (a) Formation mechanism of the hollow Fe2O3 nanosphere aggregates through Kirkendall-type diffusion. (b) Chemical conversion process in the surface region of a Fe sphere in the C matrix [reprinted with permission from ref. 17, copyright 2015, Royal Society of Chemistry].

According to previous studies, ammonia, L-arginine and sodium hydroxide have all been used as hydrolysis-controlling agents, leading to nanopolyhedra, nanoparticles and nanoakes.115,116 Zhang's group105 developed hierarchically porous Fe2O3 microspheres with lysine as a hydrolysis-controlling agent. It is well known that the number of amino and carboxylic acid groups in a molecular structure determines the pH of an amino acid. Lysine molecules include two amino acids and one carboxylic acid group. Because of its pH being above 7, lysine can act as a hydrolysis-controlling agent to control the formation of Fe2O3 during hydrothermal processes. With the hydrolysis of lysine, more and more OH ions are produced in the hydrothermal reaction solution, so that pH value of the solution increases greatly. The Fe(OH)3 phase is then dehydrated to obtain Fe2O3 particles. In addition to the hydro/solvothermal method, heat treatment has been used to prepare microspheres, such as spray pyrolysis and thermal annealing. Zhang et al.117 developed a ‘spray drying-carbonization–oxidation’ strategy to prepare a-

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9339

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

Fe2O3–graphitic carbon (a-Fe2O3@GC) composite microspheres. The a-Fe2O3 nanoparticles with sizes in the range of 30–50 nm are coated with onion-like GC shells with thickness ranging from 5 to 10 nm. In this particular structure, the aFe2O3 nanoparticles act as the main active material and contribute most of the capacity. Because of its simple feature, this preparation process has attracted widespread interest from the research community of LIBs. It can be widely employed to synthesize other MOx–GC nanocomposites, such as MnO2@GC, SnO2@GC, NiO@GC and Co3O4@GC. Kore et al.118 synthesized mesoporous Fe2O3 nanoparticles by a using solvent decient approach. Without the use of any solvent, ammonium bicarbonate (NH4HCO3) was ground with hydrated ferric nitrate nonahydrate (Fe(NO3)3$9H2O) to obtain a precursor. Then, mesoporous Fe2O3 nanoparticles were achieved aer annealing at 350  C for 2 h. 2.3.2 Hollow structures. Due to their low density, high surface area and shell permeability, hollow structures have attracted a great deal of attention in the elds of energy storage and conversion, gas sensors and drug delivery.119 Hollow structures are usually composed of a core, a middle gap and a shell. As expected, the interior voids in the Fe2O3 yolk–shell particles would provide extra space to alleviate structural strain/ stress during the charge/discharge process, when they are used as anode materials of LIBs. Yolk–shell and multi-shell are typical hollow structures. Hydrothermal synthesis is a common method to prepare hollow structures. Lu's group11 adopted a template-free hydrothermal synthesis to develop double-shell hollow Fe2O3 nanospheres, with K3[Fe(CN)6] and NH4H2PO4 as the raw materials. In the early stages of the hydrothermal reaction, single-shell hollow spheres were formed due to the Ostwald ripening process. When the reaction was carried on, H+ ions from NH4H2PO4 corroded the inner shell, forming a multishell hollow nanosphere structure. The diameter of the ball is about 400 nm, which consists of a large number of primary particles with diameters of tens of nanometers. Multilayered internal structures can be clearly observed from cracked nanospheres, as shown in Fig. 7a–c. The diameter of the inner sphere is about 250 nm. The thicknesses of the inner shell and the outer shell are 60 and 30 nm, respectively, while the distance between the two shells is about 60 nm. Yu's group101 developed a novel structure of iron oxide/carbon. The a-Fe2O3 nanoparticles were prepared by using a hydrothermal method. Then, tetraethoxysilane and dopamine hydrochloride were used for hydrolysis and carbonization. Finally, the SiO2 layer was etched away with NaOH to form a hollow multi-layer shell Fe2O3 sphere. The internal void space will allow the active material to freely expand without damaging the outer protective shell. However, the void space will affect the performance of the volume energy density. They optimized the void space by adjusting the thickness of the sacricial layer of silicon oxide, so that the electrode material accommodates volume expansion on the one hand and reduces the loss of volumetric energy density on the other hand. Because of its simple process, spray pyrolysis is one of the most common methods to develop hollow structures. Spray pyrolysis temperature is the most critical factor affecting the

9340 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

Fig. 7 FESEM (a and b) and TEM (c) images of the double-shell Fe2O3

hollow spheres [reprinted with permission from ref. 11, copyright 2017, Elsevier]. TEM images of the Fe2O3 particles prepared by spray pyrolysis at various temperatures, (d) 600  C, (e) 800  C, and (f) 1000  C [reprinted with permission from ref. 38, copyright 2015, Royal Society of Chemistry].

number of shells. Kang's group38 prepared yolk–shell structured Fe2O3 particles by using a continuous spray pyrolysis method. Compact (non-hollow) Fe2O3 spheroidal particles were prepared through spray pyrolysis of a sucrose-free solution. The sucrose concentration is one of the key factors for the formation of the yolk–shell. The optimized concentration of sucrose in the spray solution was 0.7 M. The temperature of the preparation process can affect the rate of combustion of the Fe salt and carbon components, thus controlling the formation and internal structure of the yolk–shell structured particles. The microsized carbon–Fe2O3 composite particles had yolk–shell structures aer the reactions of sucrose polymerization and carbonization, repeated combustion and shrinkage. When the temperature was 600  C, the combustion rate of the composite particles was slow, while the Fe2O3 particle had a yolk–shell structure with four shells. Due to rapid combustion and an Ostwald ripening process, the shell number of the carbon–Fe2O3 particles was reduced to two, as the temperature was increased to 1000  C. Representative SEM and TEM images are shown in Fig. 7d–f. The rate of temperature change is another important factor affecting the number of shells. Yu's group120 prepared a-Fe2O3 multi-shelled hollow spheres by using a spray drying method. From the broken part of the microspheres, it was observed that the microspheres have a multi-shell structure. According to the mechanism of non-uniform shrinkage, the process of the shell formation is closely related to the rate of temperature change. At lower rates (0.5  C min1), fewer shells are produced, corresponding to single-shell hollow spheres and yolk–shell structures. At higher rates (2 and 5  C min1), more shells were formed and multilayer hollow spheres were obtained. 2.3.3 Flower-like structures. Flower-like structures have high specic surface areas, which is conducive to the diffusion and adsorption of ions and electrons. Various ower-like structures have been prepared and used as electrode materials for LIBs and SCs.121–123 Shivakumara's team124 synthesized ower-like a-Fe2O3 nanostructures by using glycol-mediated self-assembly. The ower petals in the nanostructure have many

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

independent pores. These pores and the ower-shaped structure provide a larger surface area. Wang's group125 obtained ferric(III)-hydroxide precursor hydrolyzed from FeCl3 solution and a ower-like Fe2O3 structure was achieved through heat treatment. No templates or catalysts are required in this process. The individual ower-shaped structure was made up of dozens of self-assembled nanosheets with a thickness of about 20 nm. Cao and co-workers4 prepared ower-shaped 3D Fe2O3 nanosheets on copper foil. FeSO4$7H2O, urea and NH4F were dissolved in deionized (DI) water. The solution and the pretreated copper foil were hydrothermally treated at 80  C for 12 h. The product was nally annealed in Ar at 450  C. In the early stages of reaction, the nanoparticles rst aggregated into owerlike dodecahedrons. Then, petals grew out and started to divide. Large petals divided into layered nanosheets, thus leaving enough room between themselves and nally forming a 3D framework (Fig. 8a). The ower-like Fe2O3 structure has a high BET specic surface area of 16.1 m2 g1. Lu's group10 prepared a porous a-Fe2O3 microscopic ower-like structure, which was derived from FeSO4(OH) precursors synthesized by using a simple ethanol-mediated method. Fe2(SO4)3$xH2O was added to absolute ethanol and stirred for 3 h, followed by a hydrothermal reaction at 150  C for 24 h. The resulting yellow colored precursor was calcined at 700  C, yielding hybrid structure of Fe2O3 with an average diameter of about 2 mm. The ower-like structure is composed of a large number of nanoparticle nanoporous bodies, as illustrated in Fig. 8b and c. In order to better control the ower-like structures, various adjuvants are used during the synthesis of iron oxides. Wan et al.102 synthesized Fe2O3 with a 3D ower-like structure with urea as the control agent. Urea plays a decisive role in the synthesis of the precursor of Fe2O3 and the ower-like structure self-assembly process. When ethylene glycol forms an iron alkoxide with FeCl3, H+ ions are generated as a by-product. Hydrolysis of urea produces OH ions to neutralize H+ ions, which favors the precursor synthesis. Liang et al.42 prepared a 3D ower-like Fe2O3 nanostructure through a urea-assisted hydrothermal route. A representative SEM image of the nanostructures is shown in Fig. 8d. Urea and NaOH played key roles in the formation of Fe2O3 microowers. Similar aids have been reported in the literature.121,122 Liang et al.31 found that sodium citrate played a very important role in the formation of Fe2O3 ower-like microspheres obtained by using hydrothermal synthesis. On the one hand, the introduction of citric acid chelated with iron ions helped to reduce the concentration of free Fe3+ in the solution and the formation rate of Fe2O3 nanoparticles. On the other hand, citrate acted as a shape regulator and control agent. The diameter of Fe2O3 ower-like microspheres was about 6.8 mm. This ower structure had a rough surface, which consisted of numerous randomly arranged rice-like granular particles, as demonstrated in Fig. 8e. 2.3.4 Nanoarrays. With large surface areas and hierarchical structures, nanoarrays, which could provide a high surface area and short transmission paths for electron and ion conduction, are ideal candidates for electrode materials of SCs. There are various strategies to prepare nanoarrays, which can be roughly divided into top-down and bottom-up categories. The former

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

Fig. 8 (a) SEM image of the 3D flower-like Fe2O3 [reprinted with permission from ref. 4, copyright 2015, Wiley]. (b and c) SEM images of the as-prepared porous a-Fe2O3 microflowers [reprinted with permission from ref. 10, copyright 2017, Elsevier]. (d) SEM image of the Fe2O3 microflowers [reprinted with permission from ref. 31, copyright 2014, Elsevier]. (e) SEM image of the a-Fe2O3 microflowers [reprinted with permission from ref. 42, copyright 2013, Elsevier].

mainly involves etching, removing unnecessary parts and preserving the desired material array. The latter is a step-by-step growth of a given material, removing the intermediate process material, leading to a hierarchical array structure. Fe2O3 nanoarrays are mainly prepared by using a bottom-up method. In order to solve the problem of low conductivity of iron oxide, materials with high conductivity are usually used in either the core or the shell of the arrays. Zhai's group8 designed and fabricated hierarchical core–shell heterostructure composites which consisted of NiCo2S4 nanoneedle arrays (NNAs) and Fe2O3 nanorods (NRs). Free-standing NiCo2S4 NNAs with diameters of 50–100 nm and a length of about 800 nm were grown on a Ti substrate, on which Fe2O3 nanorods with a diameter of about 5 nm and lengths of 5–15 nm were grown by using a hydrothermal method, as illustrated in Fig. 9a and b. The Fe2O3 nanorods were directly graed onto the NiCo2S4 NNAs to provide a large active surface area for electrochemical reactions. The NiCo2S4@Fe2O3 sample had a Brunauer– Emmet–Teller (BET) specic surface area of 58.5 m2 g1, which is much larger than that of Fe2O3 (21.6 m2 g1). The high conductivity of the core (NiCo2S4 NNAs) and the high specic surface area of the shell (porous Fe2O3) would ensure a high charge storage capacity of the Fe2O3 composites. Several special treatment methods have been used in the preparation process to improve the stability of the nanoarray, such as a sacricial template or an adhesive lm layer. Hu's group25 synthesized tectorum-like Fe2O3/polypyrrole (PPy) nanoarrays. The preparation process includes the growth of Fe2O3 nanoarrays from sacricial ZnO templates and the in situ vapor-phase polymerization of PPy on the surface of the nanoarrays. The formation of the Fe2O3/PPy on the conductive carbon cloth is shown in Fig. 9c. The ZnO nanorods with diameters of 100–300 nm were transformed into tectorum-like Fe2O3 nanoarrays in Fe(NO3)3 aqueous solution. The surface of the Fe2O3 nanoarray was coated with a PPy layer with an optimized thickness of 3 nm by adjusting the amount of PPy. The

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9341

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

Review

surfaces grew up accordingly. The Prussian blue microcubes were converted into hollow Fe2O3 microboxes by thermally induced oxidative decomposition at 350  C. When the annealing temperature was 550  C, Fe2O3 nanoparticles grew up, forming porous Fe2O3 microboxes. Further increase in annealing temperature resulted in further growth of the Fe2O3 nanocrystals, thus leading to the formation of nanoplatelets. Fe2O3 hierarchical microboxes were produced at 650  C. Therefore, the morphology and structure of the Fe2O3 microboxes could be controlled simply through the annealing temperature.

3. Synthesis of ferriferous oxide (Fe3O4) nanomaterials 3.1

Fig. 9 (a) SEM images of the NiCo2S4 NNAs and (b) hierarchical NiCo2S4@Fe2O3 core–shell NNAs [reprinted with permission from ref. 8, copyright 2016, Elsevier]. (c) Schematic illustration of the fabrication of the Fe2O3/Ppy nanostructure and (d) TEM image of the Fe2O3/PPy [reprinted with permission from ref. 25, copyright 2017, Wiley]. (e) TEM image of the NiNTAs@Fe2O3 nanoneedles [reprinted with permission from ref. 39, copyright 2017, Wiley].

length and diameter of the Fe2O3/PPy arrays were about 1 mm and 250 nm, respectively. The arrays were made up of nanosheets that are tens of nanometers in thickness, as shown in Fig. 9d. The PPy shell was uniformly attached to surface of the Fe2O3 cores. Xia's team39 deposited iron oxide nano-needles on ultrane Ni nanotube arrays (NiNTAs) to form NiNTAs@Fe2O3 nanostructures with a hybrid core–shell structure. ZnO nanorod arrays were used as the template, while an Au lm was sputtered on their surface before an ultrathin Ni lm was coated through electrodeposition. Finally, Fe2O3 nano-needles were deposited. Interestingly, the thin layer of Au (thickness of about 5 nm) sputtered on the ZnO nanorods with an average diameter of 150 nm ensured the uniform growth of the Ni nanolm on the ZnO nanorods. A representative TEM image of the nanostructure is shown in Fig. 9e. Without the presence of the Au lm on the ZnO nanorods, a uniform Ni lm could not be formed, instead, the Ni lm was non-uniform and discontinuous, composed of large Ni particles. 2.3.5 Other structures. Zheng et al.126 prepared an iron oxide hollow nanoshuttle, which was used as an electrode of SCs. The hollow nanoshuttles had an average wall thickness and length of 30 nm and 100 nm, respectively. Zhang and coworkers103 synthesized a series of Fe2O3 microboxes with various shell structures by annealing Prussian blue (PB) microcubes. The microboxes were made up of numerous nanoparticles. As the annealing temperature was increased, the particles on the

9342 | J. Mater. Chem. A, 2018, 6, 9332–9367

Low-dimensional nanostructures

Due to their magnetic properties, low-dimensional Fe3O4 nanomaterials are widely used in data storage, magnetic resonance, gas sensors, spintronic devices and biomedical elds. Fe3O4 is also an attractive anode material for LIBs and SCs, owing to its high capacity, environmental friendliness and high electrical conductivity. However, the stress caused by the volume expansion during charging and discharging leads to poor cycle stability of the Fe3O4 electrode and low rate capability. In this case, low dimensional Fe3O4 nanostructures could be a solution to these problems. This is because nanostructures ensure a high contact area with the electrolyte and can buffer strain and volume changes to avoid structural changes or cracks. Gupta's team127 synthesized nearly monodisperse Fe3O4 nanocrystals by using a wet chemical method, with an average particle size of 8  2 nm. Zhang et al.128 prepared carbonencapsulated Fe3O4 nanoparticles by using a hydrothermal reaction combined with heat treatment. It was found that the crystal morphology and particle size of the Fe3O4 nanoparticles could be controlled by adjusting the concentration of iron nitrate and glucose solution during the reaction process. 1D nanowires can promote the transport of electrons and lithium ions, which is advantageous for achieving high rate performance. Manthiram's group129 prepared Fe3O4 single crystal nanowires and carbon-coated Fe3O4 nanowires by using a microwave assisted hydrothermal method. Xia and coworkers130 synthesized Fe3O4 nanotubes by using a two-step hydrothermal method. Firstly, Fe2O3 nanotubes were obtained by using a hydrothermal method. Then, a thin carbon layer was coated on the surface of the nanotubes by using another hydrothermal treatment step. The Fe2O3 nanotubes were completely reduced and converted to Fe3O4 nanotubes aer thermal annealing. O'Neill et al.131 prepared 1D Fe3O4 nanowires by using a spray deposition method. They were wrapped with carbon nanotubes to form a composite electrode for SCs. Doong's group132 deposited Fe3O4 nanoparticles on the surface of halloysite nanotubes to get Fe3O4 nanotubes by using a coprecipitation method. The unique structure of 2D nanosheets also ensures high conductivity, while the spacing in between the sheets is able to alleviate the problem of volume expansion. Gu et al.133 prepared Cu/Fe3O4 core–shell nanorod arrays by using a hydrolysis-

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

coupled redox reaction. Thin Fe3O4 nanosheets are uniformly assembled on the Cu nanorods. Ding et al.23 used sodium citrate as the surfactant to induce the growth of Fe–O into hexagonal ultrathin nanosheets. The addition of sodium citrate not only neutralizes H+ as a OH donor but also acts as a surfactant to direct the growth of the nanosheets and prevent their aggregation. In the process of growth, Fe–O layers and ethylene glycolate anion layers stack alternatively and crystallize in the c-axis direction. Therefore, the citrate groups coordinate with the Fe3+ ions exposed on the (001) plane through their carboxyl and hydroxyl ligands, which can inhibit the deposition and aggregation of Fe–glycolate along the c-axis direction. As the concentration of sodium citrate was increased from 15 mM to 20 mM, the average thickness of the Fe–ethylene glycolate nanosheets was decreased from 370 nm to 23 nm, because the growth in the c-axis direction was more severely inhibited than that in the a–b direction as the sodium citrate concentration was increased. Fig. 10a–c shows SEM images of the Fe–ethylene glycolate nanosheets obtained at different concentrations of sodium citrate. In addition, the Fe–ethylene glycolate nanosheets tended to be curved, preventing them from being closely packed in the thickness direction. Similar results have been reported by others in the literature.94,97 Xin et al.134 prepared a hierarchical structure of Fe3O4 nanosheets by annealing nanosheet ferric alkoxide precursors in N2. Huang et al.135 deposited porous Fe3O4 nanosheets by using a chemical bath. The nanosheets consisted of nanoribbons with a width of less than 10 nm. 3.2. 3D nanostructures Due to special physical reasons such as Ostwald ripening,136 spherical or ower-shaped 3D Fe3O4 nanostructures have been synthesized through stacking. Sun's group137 synthesized Fe3O4 microowers by using a solvothermal method combined with subsequent thermal annealing. When the Fe3+ solution was mixed with urea and ethylene glycol, burst nucleation occurred, resulting in the aggregation of supersaturated nuclei. Subsequently, nanoparticles began to form and grow. Due to Ostwald ripening, these nanoparticles with an average size of 10 nm were interconnected and assembled into nanosheets with an average thickness of 60 nm. In order to minimize the energy of the system, the nanosheets are crosslinked together to form a microower with diameters of 3–6 mm. Zhu and Xu114 prepared Fe3O4 mesoporous microspheres by using a solvothermal method without the use of surfactants. Fe(NO3)3 and tartaric

Fig. 10 SEM images of the Fe–ethylene glycolate nanosheets with different thicknesses, synthesized at different concentrations of sodium citrate: (a) 370 nm/10 mM, (b) 110 nm/15 mM and (c) 23 nm/20 mM [reprinted with permission from ref. 23, copyright 2016, American Chemical Society].

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

acid were dissolved in dimethylformamide solvent and reacted at 160  C for 8 h to prepare ferrous tartrate (C4H4O6Fe) precursor microspheres. The microspheres were further calcined in N2 at 400  C, leading to Fe3O4 mesoporous microspheres. The as-prepared Fe3O4 mesoporous microspheres had a high specic surface area of 122.3 m2 g1. He's group138 fabricated Fe3O4@C nanospheres with ferrocene as the Fe source by using a solvothermal method. Suh's group139 prepared Fe3O4@carbon microspheres through suspension polymerization combined with heat treatment. The surface morphology of microspheres varied with the content of Fe3O4 nanoparticles. With a lower content (72 wt%), the microsphere surface was relatively smooth. However, if the content was too high (98 wt%), the microspheres had a macroporous sponge-like structure and the surface was rather rough. Wan's group140 synthesized porous 3D nanohybrids. N-doped carbon (NC) was used to encapsulate the Fe3O4 nanoparticles. The Fe3O4@NC nanoparticles were connected by nitrogen-doped graphene to form a conductive network of 3D interconnects. The resulting 3D nanostructures exhibited an ultra-low density (8.47 mg cm3). Hollow porous structures have attracted much interest due to their specic morphology. The hollow and porous structure offers a high specic surface area and a large number of pores, providing a large contact area between the electrode and the electrolyte and thus producing more redox activation sites. More importantly, the hollow structure may provide sufficient space to mitigate the mechanical stresses caused by the volume changes during the repeated Li+ insertion/extraction. Therefore, the hollow structured electrode will have high cycling stability. Various methods have been employed to prepare hollow porous Fe3O4 nanostructures, such as self-assembly, ion exchange, solid state decomposition and chemical etching.141–145 Hollow nanostructures include hollow spheres, multi-shell hollow structures, hollow nanocapsules and yolk–shell structures. Lou's group19 synthesized a hollow microsphere composed of Fe3O4 nanoplates by using a solvothermal method. The uniform sphere with a diameter of about 5 mm was hollow inside, whereas the wall of the sphere is made up of nanoplates through stacking. These wedged plates had a certain curvature to construct the spherical structure, as demonstrated in Fig. 11a and b. Similar Fe3O4 hollow spheres were reported by Geng and co-workers45 by using a hydrothermal method combined with heat treatment. The Fe3O4 hollow nanospheres had an average diameter of about 200 nm, with a rather rough surface. The hollow spheres were made of nanocrystals with a diameter of about 10 nm, as seen in Fig. 11c and d. He's team7 prepared Fe3O4 double carbon-shelled hollow spheres, where a double C– C–SiO2 hollow sphere was used as a hard template. 2,4-Dihydroxybenzoic acid-formaldehyde was coated on the surface of the C–C–SiO2 particles through in situ polymerization. Fe3+ ions were locked through the electrical attraction between the negatively charged RF–COO– and the positively charged Fe3+ ions, which were converted to iron oxide during the subsequent pyrolysis. Then, double-layered hollow C–C–Fe3O4 microspheres were obtained by carbonizing the resin layer and removing the intermediate silica layer. The as-prepared C–C– Fe3O4 hollow spheres had a typical hierarchical porous

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9343

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

structure. The rst layer consisted of macropores with diameters of 50–100 nm, as clearly observed from the broken shell in which the holes were well connected to one another (white arrow in Fig. 11e). The second layer consisted of mesopores/ micropores in the carbon wall that was thinner than 3 nm. The C–C–Fe3O4 microspheres had a high specic surface area of 317.3 m2 g1 and a pore volume of 0.591 cm3 g1, indicating that most of the pores in the shell were connected to the opening outside the spheres. Prussian blue (PB) templates have also been used to synthesize hollow microcapsules. Xu's group145 prepared Fe3O4/ VOx hollow microcapsules through a template reaction between a PB cube and Na3VO4. The synthesized PB cube was a 3D mini cube with a uniform size. Addition of Na3VO4 resulted in the conversion of PB into multi-component composites near the surface by hydrolysis and ion exchange. An interlayer gap appeared between the shell and the remaining PB core. The reaction occurred from the outside to the inside. As the reaction time was extended to 20 min, the core disappeared and an open hollow micro-cassette was formed. Yolk–shell is a special type of hollow nanostructure consisting of an outer shell and an inner core, with a gap between them, which can be prepared by using a variety of methods. The rst step is to prepare a multi-layer core–shell structure, from which the middle layer is then removed, thus leaving a gap to achieve a yolk–shell nanostructure. There are two main ways to remove the middle layer. One is the corrosion method. Guan's group146 synthesized a yolk–shell Fe3O4@C composite with a high loading ratio of 90 wt% by using a modied St¨ ober method. Firstly, 3-aminopropyltriethoxy-silane was applied to modify the NH2 groups. Then, Fe3O4 spheres were coated with a SiO2 layer. Secondly, a carbonaceous layer was deposited through a hydrothermal reaction, followed by thermal

Review

annealing. Finally, NaOH solution was used to etch the SiO2 layer. In this way, middle-hollow Fe3O4@C yolk–shell spheres were obtained, in which Fe3O4 was the core, a carbon layer with 10 nm thickness was the shell and the void thickness between the Fe3O4 core and the carbon shell was about 80 nm. A similar yolk–shell Fe3O4 nanostructure was prepared by Yang's group.47 The main difference is that they used ionic liquids as the sources of carbon and nitrogen. Another way to remove the middle layer is to generate a gas through a chemical reaction, thus forming the voids. Chen's group12 prepared yolk–shell Fe3O4@nitrogen-doped carbon (Fe3O4@N–C) nanocapsules by using a hydrothermal method combined with thermal annealing. Spindle-like b-FeOOH nanoparticles were prepared through a hydrothermal reaction rst. Dopamine was coated on the surface of the b-FeOOH nanoparticles as a carbon source and a nitrogen source to form the b-FeOOH@PDA core–shell nanostructure. As water molecules ran away from the spindle-like b-FeOOH during annealing, some pores were le on the surface of the materials. At the same time, the core volume shrank at high calcination temperatures, creating void spaces inside the carbon shell. The b-FeOOH@PDA core–shell nanostructure was then transformed into a Fe3O4@N–C yolk shell nanostructure. The Fe3O4@N–C nanocapsules exhibited a BET surface area of 94.2 m2 g1. In addition, there are many reports on the synthesis of nanobeads, nanocages and hollow nanotubes. Based on directional assembly and Ostwald ripening, Xue's group147 fabricated a hollow Fe3O4 nanobead by using a solvothermal method. Using a kapok ber as a microtubule biotemplate, Du et al.148 synthesized Fe3O4 hollow nanotubes. Song's group149 prepared hollow nitrogen-doped Fe3O4/carbon nanocages with hierarchical porosities by carbonizing polydopamine coated Prussian blue.

4. Synthesis of iron oxide (Fe2O3, Fe3O4) nanocomposites

Fig. 11 (a and b) SEM images of the Fe3O4 hollow microspheres [reprinted with permission from ref. 19, copyright 2013, Wiley]. (c and d) SEM and TEM images of the hollow Fe3O4 nanostructure (inset: magnified image) [reprinted with permission from ref. 45, copyright 2014, Royal Society of Chemistry]. (e) SEM image revealing the hierarchical porous structure of the C–C–Fe3O4 hollow spheres [reprinted with permission from ref. 7, copyright 2015, Elsevier].

9344 | J. Mater. Chem. A, 2018, 6, 9332–9367

Volume changes and powder agglomeration are the major problems suffered by transition metal oxide (MOx) electrode materials. In order to solve these problems, the rst approach is to design and fabricate various types of nanostructures such as nanotubes, nanowires, nanospheres, and hierarchical structures. The nanostructure material can provide a high specic surface area and short electron-transmission distance, which improves the large contract area between the active material and the electrolyte, high rate capability and cycling performance during the lithium insertion/extraction process. The second strategy is by compositing MOx with other materials to reduce agglomeration and volume changes during the charge/ discharge cycles. The third effective method is the design and synthesis of a hybrid electrode.74,150–156 In this section, we will focus on the complexing of iron oxide with other materials, and review the synergistic effects of the various components and enhance the electrochemical behavior. Firstly, the composites of different carbonaceous supports such as iron oxide and amorphous carbon, carbon nanotubes,

This journal is © The Royal Society of Chemistry 2018

View Article Online

Review

graphene and carbon cloth will be reviewed. Then, the nanohybrid structures formed by iron oxide and other metal oxides will be introduced. Finally, the composite with other polymer conductors will be discussed.

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

4.1

Iron oxides and carbonaceous nanocomposites

The involvement of carbonaceous materials can not only reduce electrode resistance and increase lithium ion and electron transport rates but also provides a buffer layer to mitigate the volume changes and the agglomeration of the activated particles during the cycling process. Various carbonaceous materials have been adopted to form composites with iron oxide. Based on the carbon sources, they are classied into graphene, carbon nanotubes and other carbonaceous materials. Graphene is a sp2-bonded carbon atom with a 2D single-atom thickness, which has the characteristics of high specic surface area, excellent electrical conductivity and mechanical exibility. When graphene and iron-based composite materials are used as the electrode of LIBs, the graphene layer not only acts as a structural buffer layer to effectively accommodate the volume change, but also promotes electron conduction and shortens the lithiumion transport pathway. Moreover, the combination of graphene sheets and nanoparticles can also effectively inhibit the re-accumulation of graphene and the iron-based nanoparticles. Also, the composite maintains a high active surface area. In recent years, many iron oxide and graphene composites have been reported in the open literature.34,157–163 Yu's group164 prepared Fe2O3/graphene hybrids by using a liquid phase li off method, in which highly conductive few-layer graphene sheets and covalently bound Fe2O3 nanoparticles were employed to build up a sandwich structure. Tour's group165 fabricated a Fe2O3/Fe3C–graphene nanoporous composite by using chemical vapor deposition. In this composite, the graphene layer coated on the surface of the nanoporous material is mainly used for its conductive effect. The introduction of FeC increases the structural stability of the nanoporous composite, which provides more capacity due to the interfacial lithium storage effect. Zapien's group166 prepared Fe3O4 nanoparticles and graphene composites by using a thermal evaporation-induced anhydrous strategy. The Fe3O4 nanoparticles with sizes of 10–20 nm were uniformly anchored onto graphene sheets, which prevented the agglomeration of the graphene sheets. Choi and Kang3 prepared Fe3O4/graphene composite powder by using a spray pyrolysis process. During the drying process, the graphene sheets were bent due to shrinkage. A plurality of curved graphene sheets were agglomerated into hollow Fe3O4/graphene balls (Fig. 12a and b). Fig. 12j shows the formation process of the hollow Fe3O4/graphene balls and the mechanism of lithium intercalation/deintercalation. The BET surface area of the Fe3O4/graphene balls was up to 130 m2 g1. Hu et al.167 prepared a sandwich-structured graphene–Fe3O4–carbon composite. The composite was characterized with a high mass content of Fe3O4 (85%), ultrane Fe3O4 particles (5 nm) and a high carbon layer coverage of the Fe3O4 surface. Zhao et al.168 prepared a carbon-coated Fe3O4 quantum dots/graphene composite by a hydrothermal method. In that composite, the

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

graphene nanosheets showed a curly and wavy morphology, while the Fe3O4 nanoparticles with sizes of 7–10 nm were immobilized on the graphene nanoplatelets. Gao et al.169 synthesized a quasi-hexagonal Fe2O3 nanoplates/–graphene composite by using a solvothermal method. It was found that, as the concentration of FeCl3 was increased, the morphology of the Fe2O3 nanosheets became more and more regular. Fe2O3 nanosheets with sizes of 20–80 nm were uniformly dispersed on the surface of rGO, so that their agglomeration was prevented. Xu's group170 prepared Fe3O4@graphene composites on graphene by etching an FeAl alloy. The Al was selectively leached from the FeAl alloy with NaOH solution. During the etching process, the number of octahedron nanoparticles was increased and the shape became more and more regular. Longer corrosion time (48 h) was conducive for the formation of uniform Fe3O4 octahedrons. Fe atoms underwent natural oxidation and aggregation to produce Fe3O4

Fig. 12 (a and b) SEM and TEM images of the Fe3O4/graphene hollow balls [reprinted with permission from ref. 3, copyright 2014, Elsevier]. (c) SEM images of the 3D rGO NSs [reprinted with permission from ref. 28, copyright 2017, American Chemical Society]. (d and e) SEM and TEM images of the Fe3O4/CNF necklaces [reprinted with permission from ref. 41, copyright 2015, Royal Society of Chemistry]. (f and g) SEM and TEM images of the Fe2O3/SWCNT composites [reprinted with permission from ref. 50, copyright 2017, IOP]. (h and i) TEM images of the urchin-like Fe3O4@N–C composites [reprinted with permission from ref. 55, copyright 2016, Elsevier]. (j) Schematic illustration of the formation of the hollow Fe3O4/graphene balls and the lithium intercalation/deintercalation mechanism [reprinted with permission from ref. 3, copyright 2014, Elsevier].

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9345

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

octahedrons of about 500 nm in size. The positively charged Fe3O4 octahedron was xed on the graphene by using an electrostatic assembly method. Kumar and co-workers28 synthesized a 3D hybrid composite of Fe3O4 nanoparticles embedded in the network of reduced graphene oxide nanosheets (rGO NSs) by using an in situ microwave method. The rGO NSs had a cotton-like uffy structure, as observed in Fig. 12c .The surface of the rGO NSs presented a microscopic open structure and the rGO NSs interconnect with each other to form a network morphology. The rGO NSs had a high surface area and large reaction edge, which was favorable for the attachment of the nanoparticles. Fe3O4 nanoparticles about 50–200 nm in size were uniformly dispersed in the framework of rGO nanoparticles. The edges of the rGO NSs prevented the agglomeration of the Fe3O4 nanoparticles, which would increase the electrochemical capacity and stability of the composite. Fu's team41 synthesized Fe3O4 nanospheroids with necklace-like structures on carbon nanobers (CNFs) by using a solvothermal method. Ferric chloride and CNFs were added to the ethylene glycol solution and then hydrothermally treated in an autoclave at 200  C for 16 h. The resulting Fe3O4 nanospheres had a diameter of about 300 nm and were uniformly strung on the CNFs, as illustrated in Fig. 12d and e. The mass content of Fe3O4 was about 86%. The Fe3O4/CNF necklace could be directly sprayed onto a large area current collector to form a binderless electrode, with promising performances for LIBs and SCs. Due to their super high conductivity, carbon nanotubes (CNTs) can be used as the conductive substrate and skeleton of composite electrode materials.60 Their composites have been studied extensively as electrode materials for energy storage systems, including both LIBs and SCs. Cao et al.171 prepared a hybrid lm which is composed of a-Fe2O3 nanoparticles and single-walled carbon nanotubes (SWCNTs). Shang's group50 synthesized Fe/SWCNT lms by using a chemical vapor deposition method, which were converted into Fe2O3/SWCNT composite lms through thermal annealing. When the annealing temperature was increased from 450  C to 600  C, the mass fraction of Fe2O3 increased from 63 to 96%, while the Fe2O3/SWCNT composites still maintained their porous network structure, as demonstrated in Fig. 12f and g. The interconnected Fe2O3/– SWCNT networks, which contain highly loaded active materials, have a synergic effect and are promising anode materials for LIBs. Yu et al.172 prepared a quadrangular carbon nanotube (qCNT)–Fe3O4–C composite by using a hydrothermal method. Cheng's team173 reported nanoporous Fe2O3-5 wt% CNT composites as the electrode of SCs by using an atmospheric pressure plasma jet method. Wang et al.174 prepared Fe– Fe2O3@N-doped C nanoparticles uniformly anchored on Ndoped carbon nanotubes (NCNTs). The nano-Fe–Fe2O3 particles had a core–shell structure, consisting of a Fe–Fe2O3 core and an N-doped C shell with an average diameter of about 10 nm. The N-doped C shell effectively interconnected the Fe–Fe2O3 with the CNT scaffold, which can improve the mass/charge transfer properties and structural stability. Sun's group63 developed Fe2O3@CNT composites by depositing Fe2O3 nanoparticles on carbon nanotubes through atomic layer deposition. The Fe2O3 nanoparticles were uniformly distributed on the CNTs. The size

9346 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

of Fe2O3 nanoparticles was increased with the increasing number of atomic layer deposition cycles. In addition, explosive detonation is the process of generating very high temperatures and pressures in a very short period of time, where a large amount of heat is released due to the conversion of the chemical energy involved in the energy molecules into heat energy. In recent years, detonation has been used to prepare bamboo-like carbon nanotubes.175 Chen's team62 has prepared a Fe3O4–Fe@CNT composite by the detonation method. In the preparation process, bamboo carbon nanotubes wrapped with Fe3O4/Fe akes were prepared by onestep detonation using ferrocene as the carbon source and using hexane to generate high temperature and high pressure. Fe3O4 akes and iron nanoparticles were generated during the detonation assisted decomposition and were nally encapsulated in carbon nanotubes. The resulting carbon nanotubes have a bamboo-like structure, presenting a multi-section entangled morphology. Nanotubes (about 18–20 nodes) are more than 600 nm in length and 20–30 nm in diameter. In addition to graphene and carbon nanotubes, a variety of other carbonaceous organic and inorganic materials have been used as carbon sources. Graphite and porous carbon were used as inorganic carbon sources to prepare iron oxide and carbon composites by Yang et al.176 and Ye et al.177 Sucrose,178 glucose,179 melamine,180 ethylene glycol,181 petroleum pitch,182 petroleum alkali lignin,183 gelatin56 and other organic matter as a carbon source, have been widely studied. Polymers can also be used as a carbon source. For instance He et al.184 and Pan et al.185 synthesized iron oxide/carbon composites with polyacrylonitrile and polydimethyldiallylammonium as carbon sources, respectively. Chen et al.55 prepared an N-doped urchin-like Fe3O4@C composite by using a hydrothermal method. Urchin-like hydroxyferric oxide (a-FeOOH) was used as a template and polydopamine (PDA) was used as carbon and nitrogen sources. During high temperature carbonization, a-FeOOH was converted to oxide and the outer carbon shell also turned into hollow carbon tubes, as shown in Fig. 12h and i. On the one hand, water molecules evaporated from the a-FeOOH framework, leaving pores in the original spinous process surface. On the other hand, the appearance of the carbon shell remained unchanged, forming a hollow carbon tube. The Fe3O4@N–C urchin-like microspheres exhibited a BET surface area of 72.6 m2 g1, which was much larger than that (47.7 m2 g1) of the Fe3O4 microspheres. The Fe3O4@N–C urchin-like microspheres were made through the stacking of nanorods with sizes of 20–50 nm and lengths of 500–800 nm. The thickness of the carbon shell was 5–8 nm and the pore size was 5–25 nm. Therefore, urchin-like Fe3O4@N–C was a mesoporous material. It is expected that this porous structure of the carbon shell would provide an effective internal transport path and offer sufficient internal void space to relieve the volumetric expansion during lithiation/delithiation.

4.2

Iron oxides nanocomposites with other metal oxides

Due to their higher theoretical capacities (1494 mA h g1 for SnO2, 1232 mA h g1 for MnO2, and 890 mA h g1 for Co3O4),

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

SnO2, MnO2 and Co3O4 have been chosen to construct the heterostructures with iron oxide. Solvothermal and hydrothermal methods are the most commonly used ones to prepare Fe2O3/SnO2 heterostructures. Liu's team186 and Yang's team187 synthesized Fe3O4@SnO2 and Fe2O3@SnO2 core–shell nanorods by using a hydrothermal process. Yu's group188 grown 1D SnO2 nanorods on 2D Fe2O3 nanoakes to form a comb-shaped heterostructure. Jin et al.189 prepared SnO2 hollow spheres by using a solvothermal method rst, onto which a layer of Fe2O3 was coated. The hierarchical Fe2O3/SnO2 hollow spheres offer breathable aggregates which can buffer the volume expansion during the charge/discharge process. Moreover, the unique porous structure of the hollow sphere and the nanosized network can provide faster ion diffusion and better cycling stability. In addition, a ame-assisted spray method has been employed by Li et al.190 to decompose iron and tin precursors to prepare core–shell Fe2O3@SnO2 heterostructures. He's group191 prepared an N-doped amorphous carbon coated Fe3O4@SnO2 coaxial nanober by using electrospinning technology combined with chemical bath deposition. Yang's group9 prepared MnO2/Fe2O3 dendrite nanorods through a hydrothermal reaction. MnO2 nanorods with diameters of 30–120 nm were employed as the skeleton of the dendritic nanorods. During the hydrothermal process of FeOOH, a large number of tiny Fe2O3 nanorods were deposited on the surface of the nanorod framework. The Fe2O3 nanorods with an average diameter of 30 nm and a length of 140 nm were deposited perpendicularly onto the sides of the MnO2 nanorods, resulting in typical branched nanostructures, as demonstrated in Fig. 13a–d. The Fe2O3 branches were arranged along four-fold axis on the side of the MnO2 nanorods, forming high yields of branched nanostructures. The specic surface area of the dendritic nanorods reached 32.7 m2 g1. These branched nanostructures should be benecial to increase reaction sites and the interface area during the discharge/charge process, thus achieving high specic capacity. Wang's group192 prepared a-Fe2O3 nanotube@MnO2 nanosheet hierarchical networks as electrodes for SCs. Zhao et al.32 obtained a Mn3O4/Fe3O4 nanoower by etching Mn5Fe5Al90 alloys. In this case, the more active components of the alloys are removed, so that voids are formed inside the materials. NaOH solution was used as the etchant to selectively etch the Al atoms from the Mn5Fe5Al90 alloy. Then, the exposed Fe and Mn atoms were rapidly oxidized to form oxide nuclei. The oxide nuclei aggregated into nanoower structures in the alkaline environment. The individual nanoower-like structure consisted of an array of regularly-shaped hexagonal nanosheets. Hexagonal nanosheets had side lengths of 600–900 nm and thickness of about 150 nm, as revealed in Fig. 13e and f. Using ZnO as a sacricial template, Tu's group43 prepared hierarchical Fe2O3@Co3O4 nanowire arrays. The Co3O4 nanowires were rst fabricated by using a hydrothermal process. The nanowires were then immersed into a Zn source solution and heat treated to form the intermediate template ZnO@Co3O4 nanowires. Finally, the nanowires were transferred to a Fe source reaction solution. Aer the replacement reaction, Fe2O3@Co3O4 nanowires were formed, as shown in Fig. 13g and h.

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

Fig. 13 SEM images of (a) MnO2 nanorods, (b) branched nanorods of MnO2/FeOOH and (c) branched nanorods of MnO2/Fe2O3. (d) Schematic diagram of the formation of the MnO2/Fe2O3 branched nanostructures [reprinted with permission from ref. 9, copyright 2013, Wiley]. (e and f) SEM images of the Mn3O4/Fe3O4 sample after dealloying the Mn5Fe5Al90 alloy for 10 h in 2 M NaOH solution [reprinted with permission from ref. 32, copyright 2015, Elsevier]. (g and h) SEM and TEM images of the Fe2O3@Co3O4 nanowire array [reprinted with permission from ref. 43, copyright 2013, Elsevier]. (i, j) SEM and TEM images of the a-Fe2O3@Ni(OH)2 nanosheet hybrids [reprinted with permission from ref. 52, copyright 2016, Nature].

Wu's team193 synthesized Fe2O3@Co3O4 composites by using a two-step hydrothermal method. The a-Fe2O3 nanosheets were covered with a layer of well aligned Co3O4 particles. Similarly, Yang et al.194 fabricated Co3O4 nanowire@Fe2O3 nanorods by using a hydrothermal method. In addition to the above three oxides, there have also been reports on other oxides used to form complex heterostructures with iron oxide. Owing to high Li-insertion voltage and zerostrain insertion, spinel Li4Ti5O12 has attracted considerable attention.195–198 Diao's group199 synthesized a core–shell a-Fe2O3@Li4Ti5O12 composite through a facile hydrothermal reaction. The spinel Li4Ti5O12 wrapped a-Fe2O3 ellipsoids, which broke the vast majority of the contacts between the active material a-Fe2O3 and the electrolyte. The core–shell a-Fe2O3@Li4Ti5O12 composite can prevent the formation process of a solid electrolyte interface (SEI) layer, thus minimizing the initial capacity loss. Gao's group52 fabricated an a-Fe2O3 nanosheet@Ni(OH)2 nanosheet hybrid composite by using a two-step hydrothermal method. The a-Fe2O3 nanosheet was wrapped with a wrinkled Ni(OH)2 nanosheet, as observed in Fig. 13i and j. Xiang et al.200 embedded RuO2–Fe2O3 nanoparticles into ordered mesoporous carbon (OMC) by using impregnation and in situ heating methods. The RuO2–Fe2O3 nanoparticles were uniformly dispersed in the pore walls of the

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9347

View Article Online

Journal of Materials Chemistry A

Review

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

2D mesoporous carbon. The average particle size of RuO2–Fe2O3 was 1.96 nm. The RuO2–Fe2O3/OMC composites had a large pore volume of 1.01 cm3 g1, an appropriate pore size of 4.3 nm and a high specic surface area of 768 m2 g1. The nanocomposite particles would not block the mesoporous channels and they also enhance the electron and ion conduction. 4.3

Iron oxide nanocomposites with conductive polymers

Conductive polymers can be used to improve the electrochemical performance of energy storage devices, thus having attracted a great deal of interest in recent years.201,202 Polyaniline (PANI) has a small particle size and high electrical conductivity, which can provide high specic surface area and fast electron transport. PANI not only acts as a host material for insertion/ extraction of Li+ ions but also serves as an intermediate between the active material and the electrolyte, facilitating the Li+ ion insertion/extraction process. Wang et al.203 developed a Fe2O3/ PANI composite through the in situ polymerization of aniline, while the Fe2O3 was synthesized by using hydrothermal synthesis. Lee's group24 fabricated a hierarchical Fe2O3@PANI core–shell hollow structure, in which urchin-like Fe2O3 spheres were synthesized by using a template-free sonochemical method. PANI was then coated on the surface of the Fe2O3 spheres. Fig. 14a shows a schematic diagram of the synthesis process. A representative TME image of the Fe2O3@PANI nanocomposite is shown in Fig. 14b. As an ultra-stable conductive polymer, poly(3,4-ethylenedioxythiophene) (PEDOT) not only can effectively improve the conductivity of the material, but also can be used as a protective layer to prevent structural damage.204,205 Lu's group206 designed and prepared Ti-doped Fe2O3@PEDOT nanorod arrays on carbon cloth. In the composites, Ti doped Fe2O3 nanorods form the core, the conductive PEDOT layer is the shell, and the carbon cloth is the carrier. Ti4+ partially replaced Fe3+ and some Fe3+ ions were accordingly reduced to Fe2+. In this way, the donor concentration in Fe2O3 was significantly increased, thus making it possible to increase its capacitance. The unique core/shell composite should be a promising candidate for the electrode, with high conductivity, large interfacial area for the reaction and enhanced diffusion of electrolyte ions. Polypyrrole (PPy) and poly (ST-AN) (PSA) have also been used to combine with iron oxide to make high performance electrode materials. Shen's group207 directly heated iron oxide foil in air to obtain iron oxide nanoakes, which were then coated with a conductive PPy layer through chemical polymerization. Zhang et al.208 synthesized a PSA–Fe3O4@C mesoporous microsphere by using a hydrothermal method. 4.4

Multi-component nanocomposites based on iron oxides

Multi-component nanocomposites (ternary or quaternary) based on iron oxides have also been developed. Xie's group209 proposed a hybridization strategy for the layered assembly of TiO2 nanorods and Fe3O4 nanoparticles on pristine graphene (PG) sheets, so as to construct TiO2/Fe3O4–PG ternary heterostructures. Each of the three components contributed in

9348 | J. Mater. Chem. A, 2018, 6, 9332–9367

Fig. 14 (a) Schematic diagram illustrating the procedure to fabricate hierarchical Fe2O3@PANI through simultaneous in- and exterior construction, (b) TEM images of the Fe2O3@PANI samples [reprinted with permission from ref. 24, copyright 2013, Wiley].

a mutually supportive manner, demonstrating a signicant synergy in terms of composition and morphology. In the composition, TiO2 as the primary active material can fully maintain the high cycle stability of the composite, Fe3O4 as a secondary active material can provide high capacity, while graphene ensures sufficient conductivity. At the microscopic level, Fe3O4 nanoparticles and TiO2 nanorods acted as spacers to prevent the re-stacking of the graphene nanoplatelets. Graphene nanosheets served as a mounting substrate that also prevented the agglomeration of the TiO2 nanorods and Fe3O4 nanoparticles. Huang's group21 prepared a nanobrous Fe3O4– TiO2–carbon ternary composite by using natural cellulose as a structural support and carbon source. A layer of TiO2 was deposited on the cellulose surface by using a sol–gel process. Then, Fe3O4 nanoparticles were grown the TiO2-cellulose substrate through a hydrothermal reaction, followed by carbonization and thermal annealing. In this way, the Fe3O4 nanoparticles with a diameter of about 30 nm and a length of less than 100 nm were anchored to the surface of the TiO2 coated carbon nanobers. Due to the loss of water molecules during calcination, pores were formed inside the Fe3O4 particles, as depicted in Fig. 15a and b. The BET specic surface area of the Fe3O4–TiO2–carbon composite was 258 m2 g1, while the average pore size was about 3.9 nm. Using MnOx/Fe2O3 nanotubes as a template, Jin et al.37 synthesized MnOx/Fe2O3/polypyrrole(PPy) nanotubes through polymerization. The MnOx/Fe2O3/PPy nanotubes exhibited a rough surface. The thin PPy layer was grown uniformly on the MnOx/Fe2O3 nanotubes to form 1D nanostructures, as demonstrated in Fig. 15c and d. The combination of multiple components created more internal spaces that can effectively buffer the volume change and inhibit the aggregation of the materials during the charge and the discharge process. Duan's team210 prepared Fe3O4@C@Mn3O4 multilayer core–shell porous spheres by using bovine serum albumin as the carbon source. Guo et al.211 developed SnO2–Fe2O3@C ternary nanocomposites for lithium battery electrodes by using a sol–gel in situ polymerization process combined with carbonization. The in situ polymerization not only inhibited the agglomeration and growth of the metal oxide particles, but also provided excellent conductivity pathways aer carbonization. Sun's group212 prepared reduced graphene–Fe3O4–SnO2–C quaternary hybrid nanocomposites by using a homogeneous precipitation method and solvothermal treatment. Nanosized Fe3O4 and SnO2 were uniformly dispersed on the reduced graphene nanosheets and

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

Fig. 15 (a and b) SEM and TEM images of an individual Fe3O4–TiO2– carbon composite nanofiber, with the inset showing the TEM image of an Fe3O4 nanoparticle [reprinted with permission from ref. 21, copyright 2016, American Chemical Society]. (c and d) SEM and TEM images of the MnOx/Fe2O3/PPy nanotube [reprinted with permission from ref. 37, copyright 2017, Elsevier].

were entirely encapsulated by the outermost carbon layer to form a sandwich-like buffer structure. The multi-particle layers were stacked to form nanopores with an average size of 5.8 nm. These structural features would enhance the LIBs electrode performance of the composites.

5. Synthesis of iron hydroxide (FeOOH) nanomaterials Due to its attractive negative potential window and high theoretical specic capacitance, FeOOH is emerging as a new anode material in recent years. FeOOH particles with various shapes and compositions have been developed as electrodes for supercapacitors. With a large surface area and high ion mobility, 1D coaxial nanostructured materials have shown great potential to improve electrochemical performances. In order to develop electrode materials with high energy density and high power density, Wei et al.213 prepared FeOOH@PPy, a composite of 1D FeOOH nanoparticles wrapped in ultra-thin polypyrrole. Lou's group214 prepared a layered urchin-like a-FeOOH solid sphere or hollow sphere by adjusting the amount of glycerol in the reaction system. The specic surface area of the a-FeOOH hollow sphere can reach 96.9 m2 g1. Chen's group18 used electroplating to synthesize iron oxyhydroxide lepidocrocite (gFeOOH) nanosheets with different lamellar channels. These nanosheets had smooth surfaces, with an average length of about 1.4 mm and thicknesses of 30–50 nm. Different nanosheets intertwine to form multiple 2D channels, as revealed in Fig. 16a. As a result, Li+ ions can reversibly insert/leave the 2D channels in between the [FeO6] octahedral subunits, thus improving the pseudocapacitive effect. Yu's group215 prepared uorine-doped b-FeOOH nanorods on carbon cloth. The introduction of uoride anions enhanced the conductivity of the materials. Fluorine doped b-FeOOH as the electrode material can solve the low energy density problem.

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

FeOOH has also been incorporated with other materials, such as metal oxides and carbonaceous materials, to form composites. Zhang et al.216 synthesized a CoFe2O4/FeOOH hierarchical electrode nanocomposite for supercapacitors by using a hydrothermal method. A ower-like CoFe2O4/FeOOH nanostructure was obtained by adjusting the urea content. Lv et al.35 obtained urchin-like a-FeOOH@MnO2 core–shell hollow microspheres by using a two-step hydrothermal method. The rst step was to prepare urchin-like a-FeOOH hollow spheres composed of nanorods. The second step was to grow ribbonshaped MnO2 nanostructures on the surface of the hollow spheres to form urchin-like a-FeOOH@MnO core–shell hollow microspheres. As shown in Fig. 16b, the middle black part is FeOOH while the outer white bar is MnO2. Hao's group217 prepared a FeOOH nanoparticle modied nitrogen-doped graphene composite. In the synthesis process, urea acted as not only a reductant and dopant for N-doped graphene, but also a hydroxyl donor for the precipitation of the metal hydroxides. FeOOH nanorods were grown on graphene sheets through the combined action of N-doped graphene and urea. Zhang's group218 developed ultrane a-FeOOH nanorods/graphene oxide composites as the electrode material of supercapacitors by using a hydrothermal method. Graphene oxide and iron acetate were used as the raw materials through direct reaction without the use of any additives. The a-FeOOH nanorods had an average diameter of 6 nm and an average length of 75 nm. Although iron hydroxide is a promising supercapacitor electrode in terms of its crystal structure, the material cannot easily expand or contract which limits ion penetration and diffusion. In comparison, amorphous materials could exhibit excellent electrochemical performances due to the disordered structure. Wu's group36 synthesized amorphous FeOOH/MnO2 composites, which were made into electrodes on PET, paper and textile substrates by using screen printing. The supercapacitors on all three substrates were highly exible and could be bent without sacricing device performance. Xia's group30 prepared amorphous FeOOH quantum dots/graphene hybrid nanosheets. The amorphous FeOOH quantum dots with an average size of 2 nm were tightly anchored on the graphene sheets, forming continuous mesoporous nanolms. Wong's group219 prepared amorphous sh scale FeOOH nanostructures on Ni foam through electrodeposition. This nanostructure has a large number of surface active sites, while the amorphous

(a) SEM images of the g-FeOOH NS [reprinted with permission from ref. 18, copyright 2014, Wiley]. (b) TEM image of the as-prepared a-FeOOH@MnO2 [reprinted with permission from ref. 35, copyright 2017, Springer]. (c) SEM of FeOOH/PPy on carbon fibers [reprinted with permission from ref. 49, copyright 2017, Royal Society of Chemistry]. Fig. 16

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9349

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

characteristics of the FeOOH would promote the diffusion and reaction of electrolyte ions, thus improving the capacitive properties of the materials. With the rapid development of wearable electronics, ber SCs have been considered as promising energy storage devices for wearable applications, due to their advantages of small size, light weight, high exibility and good weaving properties.220,221 Lee's group49 developed nanostructured FeOOH/polypyrrole (PPy) on carbon bers (CF) through electrodeposition. The 1D FeOOH nanowires were grown vertically on the carbon bers, thus leading to the formation of open porous structures, as illustrated in Fig. 16c. PPy was subsequently applied to the FeOOH/CF, while the porous morphology was still retained. The diameter of the FeOOH nanowires in the composite was about 10 nm, while the thickness of the PPy coating was about 3 nm. Yuan's group222 prepared Ti-doped FeOOH quantum dot (QD)/ graphene (GN) composites. The composites were then uniformly dispersed on a bacterial cellulose (BC) substrate to form a exible capacitor electrode. The Ti-doped FeOOH QD/GN material acted as the active material, while the BC ensured the exibility and mechanical strength of the exible SCs. Electronic devices such as foldable displays and self-powered transparent LCD displays require electrode materials to have not only high power/energy density but also high transparency. In order to meet these requirements, asymmetric transparent exible supercapacitors have emerged. The application of carbon materials in transparent electrodes is limited by their low theoretical specic capacitance and low transmittance. Due to their high specic pseudocapacitive effect, transition metal hydroxide materials can be used as transparent microstructured electrodes. Zhang et al.223 fabricated transparent graphenecoated FeOOH nanowire arrays and Co(OH)2 nanosheet lms by using a bioinspired method. Both nanostructures were encapsulated in a graphene shell and exhibited a porous structure, which have been evaluated as a transparent asymmetric pseudocapacitor electrode. The unique structure of the composite materials provided a large effective contact area and high electrical conductivity, offering a 3D transmission path for ions and electrons. O'Neill and co-workers131 employed a spray deposition method to prepare mesoporous porous composite electrodes, in which 1D Fe3O4/FeOOH nanowires were entangled with carbon nanotubes. FeOx and carbon nanotubes, with a similar size, were closely ‘entangled’ together aer spraying. Self-supporting networks were formed by friction, whereas voids with diameters of 50–250 nm were formed in between the two 1D materials. Transparent electrode materials require some porosity to improve electrode transparency, but voids reduce material density and capacitance. Therefore, a trade-off should be established between the two aspects.

6. Synthesis of iron sulfide (FeS2) nanomaterials Iron sulde (FeS2) with the typical characteristics of metal suldes such as high theoretical capacity (890 mA h g1), low cost and environmental friendliness has been considered as

9350 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

a promising electrode material. FeS2 can be used as either a cathode or an anode because its voltage platform occurs at about 1.5 V.224–227 However, an FeS2 electrode has poor electrochemical reversibility. During the rst discharge, highly active Fe0 and high resistance Li2S are formed, which cannot be completely reversed. In addition, the produced lithium polysuldes (Li2Sx, 2 < x < 8) have high solubility in the electrolyte solution and diffuse to the lithium metal electrode to form insoluble Li2S on the surface. The presence of Li2S and sulfur results in an increase in resistance, thus leading to serious capacity deterioration and irreversible electrode destruction. Therefore, various strategies have been developed to solve these problems.228 One strategy is to use a liquid electrolyte that has a functional group to inhibit of dissolution of the polysulde. Another strategy is to coat the electrode material with a layer of polymer or carbon to effectively reduce the polysulde dissolution in the electrolyte. Covering a layer of carbon on the surface of FeS2 can effectively prevent the polysuldes from dissolving. Carbon coating was previously formed through solid-phase reactions, such as ball milling FeC2O4$2H2O powder, S powder and glucose mixture.229 Nanostructures of FeS2 electrodes have been proven to exhibit enhanced electrochemical performances. Wang's team230 synthesized 1D porous FeS2@C nanowires as anode materials for LIBs. The composite nanowires consist of an inner core of FeS2 and an outer thin amorphous carbon layer. The 1D FeS2@C nanostructures have short distances for electron transport and ion diffusion, a high surface area for the reaction, and more space to accommodate volumetric changes in lithium ion insertion/removal processes. The carbon layer outside the nanowires can inhibit the internal polysulde dissolution in the electrolyte. Yang's team231 utilized a hydrothermal method to polymerize glucose on the surface of FeS2 nanocrystals, followed by carbonization through thermal annealing, thus forming FeS2@C composites. Yu's group226 synthesized FeS2@C porous nano-octahedra by using self-sacricing template routes. During carbonization of the carbon-rich polysaccharide layer coated on FeS2 octahedra, the partially encapsulated FeS2 was decomposed into FeS to a certain degree. Hydrochloric acid was used to remove the acid-soluble FeS, leaving void space that can buffer the volumetric expansion of lithium intercalation. Besides synthetic carbon, graphene and carbon nanotubes have also been used to form composites with FeS2, in order to improve the conductivity of the electrode material and enhance the mechanical strength and structural stability. Lee's group13 used reduced graphene oxide to coat FeS2, leading to FeS2/rGO microspheres. The FeS2 nanoparticles were wrapped with rGO sheets on the surface, which can effectively protect polysulde electrolyte. This structure is also expected to reduce the charge transfer resistance and improve the structural stability of the composite. Shen's group5 prepared a FeS2/graphene oxide composite by using a hydrothermal method. The cubic FeS2 microparticles had diameters in the range of 0.5–1 mm. The FeS2 particles were assembled from cube-shaped blocks with sizes of 200–300 nm, resulting in a rough surface. The particles were uniformly anchored to the graphene oxide sheets, as shown in Fig. 17a. Han's group232 synthesized FeS2@hierarchical porous

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

carbon (HPC) composites by reacting FeS2 nanocrystals with HPC. The Fe ions adsorbed in the HPC were rst calcined to form Fe3O4 and then reacted with sulfur in a vacuum to obtain FeS2. The HPC would ensure a highly conductive network, while its porous structure would have a buffer effect on the volume change. Park and Sridhar233 prepared FeS2–CNF composites, in which FeS2 was crosslinked to carbon nanobers (CNFs). Through microwave pyrolysis, ferrocene was decomposed into iron, followed by sulfation with L-cysteine, so that the resulting FeS2 nanoakes were crosslinked with CNFs. Tao's team234 developed pyrite modied sulfur-doped carbon (FeS2@S–C) bers by using biological templates. Cotton was used as the biological template and served as the carbon source. The FeS2 nanoparticles were uniformly attached to the carbon bers. In order to increase the specic area of FeS2, porous structures and ower-shaped structures have been proposed. He et al.26 synthesized FeS2 anchored on 3D graphene foam by using a hydrothermal method. During the hydrothermal reaction, FeS2 particles self-assembled into cauliower-shaped nanostructures with sizes of 1–2 mm, as demonstrated in Fig. 17b. The cauliower-like structure would be able to reduce the volume expansion and prevent the FeS2 aggregation during cycling. Shen's group235 prepared a ower-like FeS2/graphene airgel composite by using a hydrothermal self-assembly method. Ma et al.40 fabricated porous micro/nanostructured FeS2 microspheres (PPMS) by using a solvothermal method. The PPMS was a stack of nanosheets that have many pores of about 100 nm in size, as seen in Fig. 17c. The electrical conductivity of marcasite (m-FeS2) is higher than that of pyrite p-FeS2, due to its lower semiconductor gap, stronger Fe–S bonds and less S–S interaction. Therefore, m-FeS2 should be more suitable as an electrode material. However, mFeS2 is rarely studied due to the difficulty in synthesis. Zhang's group22 prepared m-FeS2 and carbon nanober composites (mFeS2/CNFs) with a cluster structure by using a hydrothermal method. The m-FeS2 particles were uniformly distributed on the CNFs and intertwined with the CNFs to form a grape-like hybrid morphology, as illustrated in Fig. 17d. When Li+ ions are inserted, the m-FeS2/CNFs composite is transformed into a mimosa leaf-opening shape. In the same way, it turns into a closed mimosa when Li+ ions are detached.

7. Potential applications Iron-based materials have been extensively studied for a wide range of applications in photocatalysis, water treatment, gas sensors, magnetic storage, superconductors, lithium batteries, supercapacitors and other elds.54,64–66,236–247 As stated earlier, the applications of iron-based materials only focus on lithiumion batteries and supercapacitors. 7.1

Lithium-ion batteries (LIBs)

Because of their high energy density, lack of memory effect, lightweight nature, small self-discharge, and environmentally friendly features, lithium-ion batteries have the absolute dominance in the consumer electronics power market.54,65,89,248

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

Fig. 17 (a) SEM image of the FeS2/RGO composite [reprinted with permission from ref. 5, copyright 2015, Royal Society of Chemistry]. (b) SEM image of the 3D GF-FeS2 [reprinted with permission from ref. 26, copyright 2016, Elsevier]. (c) SEM image of the pyrite porous microspheres [reprinted with permission from ref. 40, copyright 2017, Elsevier]. (d) SEM image of the m-FeS2/CNFs composite [reprinted with permission from ref. 22, copyright 2017, American Chemical Society].

They are also considered as the preferred energy storage unit for electric vehicle powertrain and smart grid energy storage systems. These emerging applications put higher demands on the energy density of LIBs. Therefore, the performance of LIBs must be signicantly improved. It is well known that LIBs are composed of anodes and cathodes, which are separated by an ion-permeable membrane. Meanwhile, lithium ions in the electrolyte connect the two electrodes. The charging and discharging processes of LIBs are accompanied by charge and mass transfer between the electrodes. Therefore, the performance of the electrode greatly affects the performance of the battery. Commercial graphite anodes can provide the lithium ion storage space with little change in volume and thus have high structural stability, but their relatively low theoretical capacity (372 mA h g1) limits their application in power supply with high energy density. Due to their high theoretical capacity, ironbased materials have been considered to be the next generation electrode materials of LIBs. However, their poor conductivity and large volumetric changes during lithiation/delithiation cause rapid capacity decay and unsatisfactory rate performance. To solve the above problems, two strategies have been developed. One strategy is to increase the electrical conductivity and improve structural stability by using composites. The other strategy is to utilize intricate structures with a large area to make them come into contact with the electrolyte and porosity to buffer the volumetric change. Representative Fe-based materials and their electrochemical performances as the anode of LIBs are listed in Table 1. 7.1.1 Iron-based materials and their composite electrodes. In order to improve the electrochemical performance of ironbased materials, forming nanocomposites has been a widely used strategy. For instance, carbon materials are used as a shell to encapsulate iron-based materials to reduce their crushing

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9351

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

and agglomeration, so as to improve the cycle stability. Combination with a material having excellent conductivity can effectively solve the problem of poor electrical conductivity of iron-based materials. In addition, the composites of iron-based materials and various transition metal oxides have been developed to increase the reversible capacity and cycle stability, due to their synergistic effect. According to the components, ironbased composites will be discussed in ve categories, namely (i) iron–carbon composites, (ii) iron–graphene composites, (iii) iron–carbon nanotube composites, (iv) other metal oxide composites and (v) iron–polymer composites. The coating of carbon increases the conductivity of the electrode and prevents agglomeration of iron-based materials. A foam-like Fe3O4/C composite electrode fabricated by Mu et al.56 possessed a reversible capacity of 1008 mA h g1 at 200 mA g1 over 400 cycles. Compared with other Fe3O4@C composites, pomegranate-like Fe3O4@C nanoparticles128 exhibited an excellent initial specic capacity of 1215 mA h g1, together with a signicantly enhanced rate capability (573 mA h g1 at a current density of 1500 mA g1) and cycling performance (806 mA h g1 aer 100 cycles). Such a high performance was attributed to their unique pomegranate-like structure, because it could endure the mechanical deformation caused by Li+ insertion/extraction reactions and prevent Fe3O4@C nanoparticles from breaking. In addition, the small sizes of the Fe3O4 nanocrystals and the carbon spheres were 15 nm and 45 nm, respectively, which were responsible for the high interface area between the active material and electrolyte, thus shortening the migration distance of Li+ ions. The N-doped carbon shell formed a stable layer of a solid electrolyte interface (SEI), reducing the initial capacity loss and increasing the reversibility of Fe3O4 for Li ion storage. N-Doped urchin-like Fe3O4@C composites prepared by Chen and coworkers55 showed a reversible specic capacity of 800 mA h g1 at 500 mA g1 aer 100 cycles. Chen's group61 prepared ultrane Fe2O3 nanoparticles embedded in nitrogen-doped hollow carbon spheres (Fe2O3@N–C), with a capacity of 1573 mA h g1 aer 50 cycles at a current density of 100 mA g1. The reversible capacity could be maintained at 1142 mA h g1 aer 100 cycles at a high current density of 1 A g1. The unique structure of the ultrasmall Fe2O3 nanoparticles uniformly distributed in the shell of nitrogen-doped carbon spheres promoted rapid electrochemical kinetics and effectively prevented the aggregation of the Fe2O3 nanoparticles during delithiation. In addition, the hollow structure effectively buffered the volumetric change and prevented the spheres from crushing. Sulfur-doped carbon has also been used to form composite electrodes with iron oxide to improve the cycle stability and rate performance. For example, a pyrite modied sulfur-doped carbon (FeS2@S–C) electrode developed by Tao's team234 showed promising cycle stability and rate performance. Aer 100 cycles at a current density of 100 mA g1, a high capacity of 689 mA h g1 was retained. The FeS2@hierarchical porous carbon composite electrode prepared by Han's group233 displayed a specic capacity of 907 mA h g1 in the rst cycle and 720 mA h g1 aer 100 cycles at 1C current density.

9352 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

The synergy between graphene and iron-based nanoparticles has been used to improve the performance of iron oxide based electrodes.167,183,249 The Fe3O4 nanoparticles and graphene composites reported by Zapien's group166 had a specic capacity of 868 mA h g1 at 200 mA g1 aer 100 cycles. Aer 200 cycles at 1 A g1, the specic capacity was still as high as 539 mA h g1, while the coulombic efficiency was above 99%. The reduced graphene oxide-coated FeS2 (FeS2/rGO) microspheres prepared by Lee's group13 possessed a high capacity of 970 mA h g1 aer 300 cycles at a current density of 890 mA g1. The electrode capacity reached 380 mA h g1 even aer 2000 cycles at an ultrahigh current density of 8900 mA g1 (10C), as demonstrated in Fig. 18a and b. Due to the stability of the hybrid structure and the synergistic effect of the internal components of the nanosheets, the reduced graphene–Fe3O4–SnO2–C quaternary hybrid nanocomposites prepared by Sun's group212 showed a reversible capacity of 868.6 mA h g1, at a current density of 200 mA g1, aer 100 cycles. A reversible capacity of 414.7 mA h g1 was retained even at a high current density of 2 A g1. The Fe3O4@C/porous reduced microcrystalline graphene (PrMGO) composite prepared by Ma and co-workers34 showed a high reversible capacity and good cycle stability. The structure of the Fe3O4@C/PrMGO composites was stable against the high rate of charge/discharge processes. As the current density was increased from 100 to 3200 mA g1 and then decreased back to 100 mA g1, the capacity of the composites recovered to 946.1 mA h g1, which was even higher than that in the initial cycle. For the next 40 cycles, the reversible capacity of the composite only slowly increased to 1106.4 mA h g1. In addition, the specic capacity was maintained at 530 mA h g1 aer 200 cycles at a high current density of 1000 mA g1. When the current was reset to 100 mA g1, the reversible capacity immediately rose to 1015 mA h g1. In the next 30 cycles, the reversible capacity increased dramatically to 1216 mA h g1. The capacity retention of the 392nd cycle reached 135.8%, as observed in Fig. 18c and d. The high electrochemical performance could be ascribed to the synergistic effect of the Fe3O4@NP NPs and the PrMGO substrate. Aer the rate test, the structure of the composite material was further optimized, and the contact between the Fe3O4@C nanoparticles and the PrMGO nanosheet became more intimate. In addition, PrMGO has not only provided plentiful pleated microchips, but also offered appropriate void structures for strain relaxation during charging and discharging. Fe-based and carbon nanotube/ber composites demonstrated similar enhanced electrochemical performances. In the marcasite and carbon nanober composite (m-FeS2/CNFs) reported by Zhang's group,22 during Li+ insertion, m-FeS2 microparticles were converted to have a mimosa-like shape with open leaves. Once the Li+ ions were detached, the m-FeS2 particles would return to their closed mimosa shape, as illustrated in Fig. 19a. This change in morphology created an opportunity for Li+ ion insertion, thus resulting in enhanced electrochemical performance. The m-FeS2/CNFs electrode had a high reversible capacity of 1399.5 mA h g1 aer 100 cycles at a current density of 100 mA g1. It also exhibited excellent cycling performance with a reversible capacity of 573.4 mA h g1 aer 1000 cycles at

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

Journal of Materials Chemistry A

a high current density of 5 A g1. The m-FeS2/CNFs electrode possessed excellent rate capability. As the current density was increased from 100 mA g1 to 10 A g1, a capacity of 782.2 mA h g1 was still retained, as observed in Fig. 19b and c. The Fe2O3/SWCNT composite lms developed by Shang's group50 exhibited a reversible capacity of 1007.1 mA h g1 at a current density of 200 mA g1, together with a high rate capability of 384.9 mA h g1 at 5 A g1, while the discharge capacity was retained at a level of 567.1 mA h g1 aer 600 cycles at 2 A g1. The Fe3O4–Fe@bamboo-like carbon nanotubes prepared by Chen's team62 maintained a coulombic efficiency of over 97%, with a reversible capacity of nearly 800 mA h g1 aer 100 cycles at a current density of 300 mA g1. The Fe2O3@CNT composites prepared by Sun's group63 with atomic layer deposition showed a high reversible capacity of 859.7 mA h g1 aer 400 cycles at a current density of 500 mA g1. The Fe2O3@CNT electrode could achieve a high cycle capacity of 464.4 mA h g1 even at a high current density of 10 A g1.

Table 1

Nanocomposites based on iron oxide with other oxides have been demonstrated to have promising electrochemical performances.32,187,188,191 For instance, SnO2/Fe2O3 hollow sphere electrodes exhibited a high initial discharge capacity of 1726.6 mA h g1.189 SnO2–Fe2O3@C ternary nanocomposites reported by Guo et al.211 possessed reversible capacities of over 1000 mA h g1 aer 380 cycles at a current density of 400 mA g1, due to the synergistic effect of SnO2, Fe2O3 and C. The Fe3O4@C@Mn3O4 multilayer core–shell porous spheres synthesized by Zhang and co-workers210 had an initial specic capacity of 1261 mA h g1 at a current density of 0.1C. Also, the reversible capacity was retained to be 987 mA h g1 aer 200 cycles. The branched nanorods of MnO2/Fe2O3 prepared by Yang's group9 presented a reversible specic capacity of 1028 mA h g1 aer 200 cycles at a current density of 1 A g1. Even at a current density of 4 A g1, a reversible capacity of 881 mA h g1 was still retained. The Mn3O4/Fe3O4 nano-owers prepared by Zhao and co-workers32 showed excellent reversible capacity and cycle stability, due to

Representative Fe-based materials and their electrochemical performances as anodes of LIBs

Materials

Preparation method

Structure

Electrochemical performances

Year

Ref.

a-Fe2O3

Hydrothermal

705 mA h g1 aer 430 cycles at 100 mA g1

2013

105

g-Fe2O3 Fe2O3

Solvothermal e-Beam deposition + etching Hydrothermal Hydrothermal

Porous microspheres Nanospheres Nanomembranes

1077.9 mA h g1 aer 140 cycles at 100 mA g1 808 mA h g1 aer 1000 cycles at 2C, 530 mA h g1 aer 3000 cycles at 6C 877 mA h g1 aer 1000 cycles at 2C 953.2 mA h g1 aer 200 cycles at 100 mA g1, 678 mA h g1 aer 200 cycles at 1A g1 800 mA h g1 aer 100 cycles at 500 mA g1 1142 mA h g1 aer 100 cycles at 1 A g1 1300 mA h g1 aer 100 cycles at 100 mA g1 810 mA h g1 aer 1000 cycles at 0.2C

2015 2014

100 51

2015 2017

4 20

2016 2015 2017 2014

55 61 176 101

1399.5 mA h g1 aer 100 cycles at 100 mA g1, 573.4 mA h g1 aer 1000 cycles at 5 A g1 800 mA h g1 aer 100 cycles at 300 mA g1 859.7 mA h g1 aer 400 cycles at 500 mA g1

2017

22

2017 2017

62 63

1007.1 mA h g1 aer 100 cycles at 200 mA g1, 567.1 mA h g1 aer 600 cycles at 2A g1 720 mA h g1 aer 100 cycles at 1C 970 mA h g1 aer 300 cycles at 890 mA g1, 380 mA h g1 aer 2000 cycles at 8900 mA g1 1050 mA h g1 aer 300 cycles at 2 A g1, 690 mA h g1 aer 1000 cycles at 7 A g1 1028 mA h g1 aer 200 cycles at 1 A g1

2017

50

2016 2015

232 13

2014

3

2013

9

1000 mA h g1 aer 380 cycles at 400 mA g1

2014

211

Unchin-like

893 mA h g1 aer 100 cycles at 0.1C

2013

24

Mesoporousmicrosphere Yolk shell

1293 mA h g1 aer 100 cycles at 500 mA g1, 599 mA h g1 aer 100 cycles at 10 A g1 860 mA h g1 aer 500 cycles at 1 A g1, 500 mA h g1 aer 500 cycles at 2 A g1 832 mA h g1 aer 150 cycles at 500 mA g1 742 mA h g1 aer 400 cycles at 500 mA g1

2016

208

2016

47

2017 2017

12 145

a-Fe2O3 Ag/a-Fe2O3 Fe3O4@C Fe2O3@N–C Fe3O4/C FeOx@C

m-FeS2/CNFs Fe2O3/CNT Fe2O3@CNTs

Hydrothermal Pyrolysis Solvothermal Hydrothermal + hydrolysis + corrosion Hydrothermal

Microowers Hollow spheres Urchin-like Spheres Hollow spheres Yolk shell

Grape-like

FeS2@HPC FeS2/rGO

Detonation Atomic layer deposition Chemical vapor deposition Vulcanization Solvothermal

Composite Microspheres

Fe3O4/graphene

Spray pyrolysis

Nanospheres

MnO2/Fe2O3

Hydrothermal

SnO2–Fe2O3@C

Poly(ST-AN)/Fe3O4@C

Polymerization++ heat treatment Polymerization + chemical etching Hydrothermal

Branched nanorods Nanospheres

Fe3O4@void@N–C

St¨ ober

Fe3O4@N–C Fe3O4/VOx@C

Hydrothermal Hydrolysis + ion exchange

Fe2O3/SWNT

Fe2O3@polyaniline

This journal is © The Royal Society of Chemistry 2018

Bamboo-like Tube-like Tube-like

Yolk shell Microboxes

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9353

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

Fig. 18 (a) Cycling performances of the FeS2 microsphere/rGO and FeS2 electrodes and the corresponding CE of the FeS2 microsphere/ rGO electrode. (b) Cycling performances of the FeS2 microsphere/rGO electrode at 10C rate [reprinted with permission from ref. 13, copyright 2015, Royal Society of Chemistry]. (c) Rate capability and (d) cycling performance of the Fe3O4@C/PrMGO composite [reprinted with permission from ref. 34, copyright 2017, Elsevier].

the synergistic effect of the different oxide components and the interconnected voids in the ower-shaped structure. When the Mn3O4/Fe3O4 electrode was charged and discharged at 300 mA g1, the initial capacity was as high as 1510 mA h g1, and the reversible capacity was 1040 mA h g1 aer 200 cycles. Conductive polymers not only provide a high electrical conductivity but also enhance the mechanical strength and stability of composites. For example, Lee's group24 developed a hierarchical core–shell hollow structure of Fe2O3@polyaniline (PANI). Compared with urchin-like Fe2O3 without etching and a PANI coating (Fe2O3) and urchin-like hollow Fe2O3 without a PANI coating (h-Fe2O3), the Fe2O3@PANI exhibited a much higher reversible capacity and long-term cycling stability. The capacity of the Fe2O3@PANI electrode was 893 mA h g1 at a current rate of 0.1C up to 100 cycles, which is higher than 680 mA h g1 of the non-hollow Fe2O3 and 732 mA h g1 of the hFe2O3 based electrodes. As the current rate was raised from 0.1C to 10C, the capacity was still retained at 681 mA h g1, which is much higher than those of the non-hollow Fe2O3 electrode and the h-Fe2O3 electrode, as depicted in Fig. 19d. Due to the synergistic effect of the multiple components and the conductive polypyrrole layer, the MnOx/Fe2O3/polypyrrole nanotubes synthesized by Jin and co-workers37 maintained a high reversible specic capacity of 1060 mA h g1 at a current density of 200 mA g1 aer 100 cycles. The specic capacity could be kept at 630 mA h g1 even at a high current density of 5 A g1. The poly (ST-AN) (PSA)–Fe3O4@C mesoporous microspheres prepared by Zhang and co-workers208 had both high cyclability and high rate performance. The exible PSA@C conductive skeleton not only ensured high electrical conductivity, but also offered structural integrity to the material. Besides, the layered mesoporous structures promoted Li+ ion diffusion and provided more contact area with the electrolyte. At a current density of 500 mA g1, the reversible capacity of the PSA–

9354 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

Fig. 19 (a) Ex situ SEM images of the lithium insertion/extraction

process. (b) Cycling performances of the pristine m-FeS2, pure CNFs and m-FeS2/CNF composite at 0.1 A g1. (c) High-rate capabilities of m-FeS2/CNFs at various current densities [reprinted with permission from ref. 22, copyright 2017, American Chemical Society]. (d) Rate capability of Fe2O3, h-Fe2O3 and Fe2O3@PANI electrodes at different rates ranging from 0.1C to 10C [reprinted with permission from ref. 24, copyright 2013, WILEY].

Fe3O4@C electrode showed a slow increasing trend. Aer 100 cycles, the reversible capacity was as high as 1293 mA h g1. The reversible capacity was maintained at 928 mA h g1, 677 mA h g1 and 599 mA h g1, at current densities of 2 A g1, 8 A g1 and 10 A g1, respectively, aer 100 cycles. 7.1.2 Iron-based nanoelectrodes with intricate structures. The synthesis of iron-based materials with complex structures, especially hollow and microporous structures, is a hot research topic, which is a strategy to improve their electrochemical performances. In recent years, a variety of unique nanostructures have been developed, which can alleviate volume changes during the lithium insertion/extraction process, provide more lithium intercalation sites, optimize the diffusion paths of ions and electrons, shorten diffusion distances and reduce active material loss because of the repeated formation of the SEI layer.66–68,119 In this section, ve categories of nanostructured electrodes will be discussed, i.e., spherical, hollow spherical, yolk–shell, nanobox and ower-like nanostructures. Iron-based microspheres or mesoporous microspheres have good electrochemical properties.114,117 The unique structure of magnetite (g-Fe2O3) nanospheres prepared by Pan's group100 has been shown to be able to buffer the volumetric change during lithiation/delithiation. The electrode had an initial charge capacity of 1060 mA h g1 at a rate of 100 mA g1, with a steady cycling capacity of 1077.9 mA h g1 aer 140 cycles. Hierarchically porous Fe2O3 microspheres prepared by Zhang's group105 via a lysine-assisted hydrothermal process demonstrated excellent electrochemical performance, because of the hierarchical porosity and ordered microstructure. The initial specic capacity was 1079 mA h g1 at a current density of 100 mA g1, which remained at 705 mA h g1 aer 430 cycles. The Fe3O4/graphene spheres prepared by Choi and Kang3 achieved a reversible capacity of 981 mA h g1 aer 100 cycles at a current

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

density of 2 A g1. Due to the formation of a polymer gel lm on the active material and the decomposition of the electrolyte, the discharge capacity of the Fe3O4/graphene nanostructures gradually increased aer 100 cycles, reaching 1050 mA h g1 in the 300th cycle, as displayed in Fig. 20a and b. The reversible capacity of the Fe3O4/graphene was preserved at 690 mA h g1 aer 1000 cycles at a high current density of 7 A g1. At a very high current density of 30 A g1, the electrode still had a stable reversible capacity of 540 mA h g1. Hollow microspheres are promising candidates as electrode materials with high electrochemical performances. The Fe3O4/C hollow microspheres prepared by Wen's group181 through ultrasonic spray pyrolysis were constructed from mesoporous carbon nanosheets that covered Fe3O4 nanoparticles. The reversible capacity of the Fe3O4/C composite was maintained at 600 mA h g1 aer 200 cycles at 1.0 A g1, with a high value of 1030 mA h g1 at a low current density 0.1 A g1. The Fe3O4/C composite with hollow spheres synthesized by Yang et al.176 showed a stable electrochemical cycle life. The initial capacitance was 1450.1 mA h g1, together with a high reversible capacity of 1300 mA h g1 aer 100 cycles at 100 mA g1. Compared with Fe2O3 hollow spheres, the Ag–Fe2O3 hollow spheres prepared by Li's group20 showed a higher capacity of 953.2 mA h g1 at a current density of 100 mA g1 aer 200 cycles. The Ag–Fe2O3 electrode also exhibited a more stable capacity of 678 mA h g1 over 200 cycles at 1000 mA g1. It was found that the Ag–Fe2O3 electrode curve had three stages, as illustrated in Fig. 20c. In the rst stage, the Ag–Fe2O3 electrode showed a gradual increase in specic capacity before 60 cycles, due to the activation of the electrode material. In the second stage, the electrode decomposition dominated from the 60th to the 125th cycle, so that the electrode performance decreased. In the third stage, the electrode capacity gradually increased from the 125th to the 200th cycle. The formation and decomposition of an organic polymer gel-like lm provided additional sites for lithium storage. In the whole process, the external SEI layer may be damaged, spalled and reformed, resulting in an unstable SEI layer during cycling. When the structure is rened, a thin and stable SEI lm is formed without splitting, and the reactivated electrode will exhibit excellent cycle stability over a long period of time. This phenomenon, in which the electrode performance rst decreases and then increases, has been observed in other studies.47,250–252 As mentioned above, yolk–shell nanostructures have voids that can accommodate volumetric expansion during the lithiation process. The FeOx@C yolk–shell structure reported by Yu's group101 showed a high reversible capacity of 810 mA h g1 and excellent cycling stability (97.4% capacity retention at 100 cycles) at 0.2C. The Fe3O4@void@N-doped carbon yolk shell composites prepared by Yang's group47 exhibited a high electrochemical performance, with a capacity of 860 mA h g1 at a current density of 1 A g1 aer 500 cycles, which is more than double that of the Fe3O4@vacancies@carbon based electrode. The specic capacity of the Fe3O4@void@N–C rst decreased and then increased, as demonstrated in Fig. 20d. This is a typical characteristic of yolk–shell nanostructure electrodes. Accordingly, Chen's group12 well explained this phenomenon,

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

Fig. 20 (a and b) Cycle performances and long cycle performances of the Fe3O4-decorated hollow graphene ball powder at 2 A g1 and 7 A g1, respectively [reprinted with permission from ref. 3, copyright 2014, Elsevier]. (c) Cycling performance of the Ag–Fe2O3 composite electrode at 1 A g1 [reprinted with permission from ref. 20, copyright 2017, Nature]. (d) Cycling performance and corresponding columbic efficiency of the Fe3O4@void@N-doped carbon electrode at a current rate of 2000 mA g1 [reprinted with permission from ref. 47, copyright 2016, Elsevier]. (e) Cycling performance of the Fe3O4@N–C-700 at 500 mA g1 [reprinted with permission from ref. 12, copyright 2017, Elsevier]. (f) Rate capability of the Fe2O3 nanomembrane electrodes [reprinted with permission from ref. 51, copyright 2014, Nature].

by observing micro-morphological changes during the charging process. When the yolk–shell Fe3O4@nitrogen-doped carbon (Fe3O4@N–C) nanocapsules were used as the electrode, reversible capacity decreased during the rst 16 cycles and then rose slowly. The reversible capacity in the second cycle was 815.7 mA h g1 and reached the lowest level of 713 mA h g1 in the 16th cycle. However, aer 150 cycles, the reversible capacity was increased by 16.7% and remained at 832 mA h g1 at 500 mA g1, as seen in Fig. 20e. Fig. 21A shows the morphologies of the Fe3O4@N–C capsules aer the 50th, 100th and 150th cycles. Due to the volume expansion caused by the charge–discharge process, part of the Fe3O4 core disassembled into small particles aer the 50th cycle, as shown in Fig. 21A(a and b). The disassembly of the Fe3O4 core into Fe3O4 nanoparticles was helpful to increase the number of active sites in the electrode materials. With the process of cycling, the core continued to decompose, as illustrated in Fig. 21A(c and d). The small particles of Fe3O4 almost lled the entire carbon shell, while the internal voids were greatly shrunk, but the number of activation sites was greatly increased. Aer the 150th cycle, the Fe3O4 core decomposed into smaller nanoparticles and remained completely

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9355

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

enclosed within the carbon shell [Fig. 21A(e and f)]. The volume expansion of the charge–discharge process broke the core into small particles, which provided more activation sites and pores that further boosted the electrochemical performance. Due to the protection of the carbon shell, the spindle-like structure remained intact. Therefore, the electrode performance was not degraded due to the decomposition of the core; instead, the number of activation sites was increased. The morphological and volumetric change process of the Fe3O4@N–C nanocapsules during electrochemical cycling is shown schematically in Fig. 21B. Square hollow nanoboxes are another common hollow structure. In the Fe3O4/VOx@C hybrid hollow microboxes prepared by Xu's group,145 the outer carbon layer not only improved the conductivity of the electrode but also maintained the stability of the structure. Hollow structures and nanopores smoothen the Li+ ion diffusion path and reduce volume expansion during charge–discharge. The Fe3O4/VOx@C electrode showed a reversible capacity of 742 mA h g1 aer 400 charge/discharge cycles at a current density of 500 mA g1. Similarly, the hollow nitrogen-doped Fe3O4/carbon nanocages with hierarchical porosities prepared by Song's group149 had a specic capacity of 878.7 mA h g1 aer 200 cycles at 200 mA g1. The Fe3O4 microowers synthesized by Sun's team137 showed a reversible capacity of about 1 A h g1 at 100 mA g1 aer 50 cycles. He et al.26 anchored cauliower-like FeS2 on 3D graphene foam to form 3D GF–FeS2 composites, which can be used directly as a free-standing and binder-free anode for LIBs without polymeric binders, conductive additives or metal current collectors. The 3D GF–FeS2 electrode maintained a specic capacity of 1080.3 mA h g1 aer 100 cycles at

Review

a current density of 0.2C. Cao and co-workers4 prepared a similar ower-like structure in which 3D layered porous Fe2O3 nanosheets were stacked. The porous 3D layered nanostructures provided a 3D conductive network during cycling. The electrode exhibited excellent rate capability and high reversible capacity, with an initial discharge capacity of 1869.3 mA h g1. Aer 1000 cycles, the reversible capacity was stabilized at 877.7 mA h g1. With a free-standing structure, Fe2O3 nanomembrane electrodes prepared by Liu and co-workers51 showed an ultra-long cycling life and a high capacity, with a reversible capacity of 808 mA h g1 aer 1000 cycles at 2C. A high capacity of 530 mA h g1 was retained aer 3000 cycles at 6C. The rate capability of the Fe2O3 nanomembrane was examined by two rounds of rate capability measurements at various current rates, as shown in Fig. 20f. The specic capacitance decreased from 899 mA h g1 to 128 mA h g1 as the current was increased from 0.2C to 50C. However, when the current rate returned from 50C to 0.2C, the reversible capacity recovered to a high value of 946 mA h g1. The rate capacity of the Fe2O3 nanomembrane electrode in the second round rate capacity measurement process showed an almost the same level of that in the rst round. Liu et al. believed that the 2D structural characteristics of the Fe2O3 nanomembranes were the key factor to improve the electrochemical performance. First of all, the unique mechanical wrinkling and bending of the Fe2O3 nanomembrane effectively buffered the lithiation/delithiation induced strain. Due to the formation of the SEI layer, the thickness of the Fe2O3 nanomembranes increased and wrinkled under the strain of repeated lithiation/delithiation. The nanomembranes maintained a well layered structure without pulverization. Such a good mechanical stability of the nanomembranes can extend the cycling life. Secondly, numerous ‘mini-capacitors’ were formed by the parallel Fe2O3 nanomembranes and electrolyte, which provided excessive pseudocapacitive contribution. 2D nanomembrane structured electrodes can prolong the cycle life and improve the rate performance of high-rate batteries, which has been proved by Augustyn et al.92

7.2

Fig. 21 (A) TEM images of the Fe3O4@C–N-700 nanocapsules after the (a and b) 50th cycle, (c and d) 100th cycle and (e and f) 150th cycle. (B) Schematic illustration the morphological and volumetric change process of the Fe3O4@N–C nanocapsules during electrochemical cycling [reprinted with permission from ref. 12, copyright 2017, Elsevier].

9356 | J. Mater. Chem. A, 2018, 6, 9332–9367

Supercapacitors (SCs)

Basing on structural characteristics and applications, the ironbased supercapacitors that are discussed in this section are classied into three categories: nanocrystalline supercapacitors, asymmetric supercapacitors and transparent exible supercapacitors. Table 2 lists representative Fe-based materials and their electrochemical performances as the electrodes of SCs. 7.2.1 Iron-based nano-materials for supercapacitors. In recent years, many efforts have been made to explore iron-based materials for SCs. At present, the main challenge is their poor electrical conductivity and structural stability during the charging and discharging process. To solve these problems, nanostructural design of iron-based materials is the most common strategy. Nanostructures, such as nanorods,76,84,255 nanosheets,18,46,256 nanotubes14,59,257 and nanowires15,86 have been developed. The Fe3O4 nanoparticles synthesized by Yang's

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

Journal of Materials Chemistry A

group253 exhibited a specic capacitance of 207.7 F g1 at 0.4 A g1 and a good rate capability of 90.4 F g1 at 10 A g1. Using a sodium lauryl sulphate template, Chen et al.254 prepared pyrite FeS2 nanoribbons by using a hydrothermal method, which possessed a capacitance of 317.8 F g1 at 3 A g1. Combining iron-based nanomaterials with highly conductive materials, such as carbon layers,7 carbon nanotubes132,258 and graphene,30,160 is an effective method to improve the electrical conductivity of iron-based materials. Due to the synergistic effect of the high capacitance of iron-based materials and the excellent electrical conductivity of nanocarbon materials, these iron-based composites exhibited much better electrochemical performance than pristine iron-based ones. In addition, they also have the effect of inhibiting mutual agglomeration. He's team7 synthesized hollow C–C–Fe3O4 spheres, with a porous structure that allowed full contact of the active sites and thus promoted the interface redox reaction. The uniform distribution of Fe3O4 on the porous carbon walls made full use of the carbon conductive network. The C–C–Fe3O4 material achieved a maximum capacity of 1153 F g1 at a current density of 2 A g1. Even at a current density of 100 A g1, the capacity still remained at 514 F g1, with a rate capability retention of 45%. As the current density was increased to 5 A g1, the specic capacitance only decreased slightly from 914 F g1 to 886 F g1 aer 1000 cycles, with a loss of about 3.1%. It is believed that the hierarchically porous structure of the outer carbon layer facilitated ion diffusion and the reaction with Fe3O4 species, as shown schematically in Fig. 22a. The synergistic effect between the outer porous carbon layer and the uniformly distributed Fe3O4 species was responsible for the enhanced capacitive performance. Due to the large surface area and fast electron transport, the specic capacitance of the Fe2O3/graphene oxide paper (GP) prepared by Xie et al.160 was about 3.08 F cm2 at a current density of 5 mA cm2, which was about 14 times that of the Fe2O3/carbon paper (0.22 F cm2). The Fe2O3/GP hybrid

Table 2

electrode retained about 95% of its initial capacitance aer 5000 cycles. The Fe2O3 nanorods/Ag nanowires/coffee lter composites prepared by Hsu'a group258 had high reversible features and rate capabilities as an electrode of supercapacitors, with an excellent specic capacitance of 287.4 F g1, an energy density of 64.6 W h kg1 and a power density of 18 kW kg1. The N-doped graphene quantum dots@Fe3O4-halloysite nanotube composites fabricated by Doong's group132 exhibited excellent electrochemical performances. In a neutral electrolyte solution, the specic capacitance of the electrode reached 418 F g1 at a current density of 0.5 A g1. In addition, the nanocomposites also demonstrated a high energy density of 29 W h kg1 and high power density of 5.2 kW W1. Aer 3000 charge–discharge cycles, the original capacitance was retained by about 82%. Incorporation of graphene could improve the electrochemical performance of iron-based materials.259 For example, the amorphous FeOOH QDs/graphene hybrid nanosheets prepared by Xia's group30 showed an excellent pseudocapacitive effect, with a large specic capacitance of about 365 F g1. At the same time, the composite electrode had excellent cycle performance and excellent rate performance. Aer 20 000 cycles, the capacitance was still retained by about 89.7% at a current density of 4 A g1. At a current density of 128 A g1, the capacitance was still at a high level of 189 F g1. When the lower cut-off voltage was decreased to 1.0 and 1.25 V, the specic capacitances were increased to 403 and 1243 F g1, respectively, as seen in Fig. 22b. The Fe2O3/FeOOH nanorods were xed on the graphene sheets, so that both the agglomeration of the Fe2O3 nanorods and the re-stacking of the graphene sheets were prevented. The Fe2O3/graphene electrode prepared by Yang et al.161 showed a high performance, with a specic capacitance of 320 F g1 at 10 mA cm2 and ideal electrochemical stability (capacitance retained at 97% aer 500 cycles). Zhang's group218 synthesized ultrane a-FeOOH nanorods/GO composites as electrode materials for supercapacitors, which had a large specic capacitance of 127 F g1 at a current density of 10 A g1,

Representative Fe-based materials and their electrochemical performances as electrodes of SCs

Materials

Specic capacitance

Rate capability

Cycling stability

year

Ref.

Fe3O4 FeS2 C–C–Fe3O4 Fe2O3/GP

207.7 F g1 at 0.4 A g1 317.8 F g1 at 3 A g1 1153 F g1 at 2 A g1 3.08 F cm2 at 5 mA cm2 365 F g1 at 1 A g1 127 F g1 at 10 A g1 455 F g1 at 8 mV s1 1668 F g1 at 1 A g1 597 F g1 at 1 A g1 382.4 mF cm2 at 0.5 mA cm2 1.15 F cm2 at 1 mA cm2 418.7 F g1 at 10 mV s1 350.2 F g1 at a 0.5 A g1 3322 mF cm2 at 2 mA cm2

90.4 F g1 at 10 A g1 195.1 F g1 at 50 A g1 514 F g1 at 100 A g1 1.406 F cm2 at 10 mV cm2

100% aer 2000 cycles at 1 A g1 81.1% aer 1000 cycles 96.9% aer 1000 cycles at 5 A g1 95% aer 5000 cycles at 100 mV s1

2013 2016 2016 2017

253 254 7 160

189 F g1 at 128 A g1 100 F g1 at 20 A g1 317 F g1 at 27 mV s1 443 F g1 at 10 A g1 170.5 mF cm2 at 10 mV s1

89.7% aer 20 000 cycles at 4 A g1 85% aer 2000 cycles at 5 A g1 95% aer 9500 cycles at 3.8 A g1 93% aer 3000 cycles at 1 A g1 97.1% aer 2000 cycles at 1 A g1 97.2% aer 5000 cycles at 100 mV s1

2016 2017 2017 2016 2017 2017

30 218 28 200 35 25

0.77 F cm2 at 8 mA cm2

96% aer 30 000 cycles at 100 mV s1

2015

206

215.3 F g1 at 64 A g1 159.5 F g1 at 20 A g1 883 mF cm2 at 50 mA cm2

92.3% aer 5000 cycles at 100 mV s1 95.6% aer 10 000 cycles at 15 A g1 94.7% aer 6000 cycles at at 8 mA cm2

2017 2017 2017

39 36 222

FeOOH QDs/graphene a-FeOOH nanorods/GO Fe3O4/rGO RuO2–Fe2O3/OMC a-FeOOH@MnO2 Fe2O3/PPy Ti–Fe2O3@PEDOT NiNTAs@Fe2O3 Amorphous FeOOH/MnO2 Ti–FeOOH QD/GN/BC

This journal is © The Royal Society of Chemistry 2018

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9357

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

Fig. 22 (a) Schematic diagram of the hierarchically porous structure of the outer carbon layer to facilitate ion diffusion and reaction with the Fe3O4 in the inner core [reprinted with permission from ref. 7, copyright 2015, Elsevier]. (b) Specific capacitances of the FeOOH/40G electrode in different voltage windows of 1 to 0 V and 1.25 to 0 V at different scan rates [reprinted with permission from ref. 30, copyright 2015, Wiley]. (c) CV curves of the Fe3O4 NPs, 3D rGO NSs and 3D Fe3O4/rGO hybrids at a scan rate of 8 mV s1. (d) CV curves of the 3D Fe3O4/rGO hybrids at different scan rates. (e) Charge/discharge curves of the 3D Fe3O4/rGO hybrids at different current densities [reprinted with permission from ref. 28, copyright 2017, American Chemical Society].

an excellent rate performance (100 F g1 at 20 A g1) and a good cycle performance with a capacity retention of 85% aer 2000 cycles at 5 A g1. Quan et al.162 prepared 2D a-Fe2O3/reduced graphene oxide (rGO) nanocomposite electrodes, with a specic capacitance of 903 F g1 at a current density of 1 A g1, which was much higher than those of the bare a-Fe2O3 (347 F g1) and rGO (167 F g1). The Fe3O4 nanoparticle/reduced graphene oxide composite synthesized by Jiao's team259 had a high specic capacitance of 220.1 F g1 at 0.5 A g1 aer 3000 cycles. Compared with the spherical Fe3O4 nanoparticles, the contact area between the multi-faceted Fe3O4 nanoparticles and the rGO nanoparticles was enlarged and the contact was more intimate. Kumar et al.28 anchored 3D Fe3O4 nanoparticles to a network of reduced graphene oxide nanosheets (rGO NSs), with a large CV curve area at a scan rate of 8 mV s1, as illustrated in Fig. 22c–e. The Fe3O4 nanoparticles were anchored in open spaces where the rGO NSs were interconnected. The Fe3O4 nanoparticles inhibited the restacking of rGO NSs, while the Fe3O4 nanoparticles were separated by the rGO NSs, thus leading to the high capacitive

9358 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

performance. At different scan rates of 8, 12, 18 and 27 mV s1, the CV curves of the Fe3O4/rGO hybrid electrode had no signicant difference in shape except for the increase in area. The specic capacitance at 8 mV s1 was 455 F g1. The electrode exhibited an energy density of 102.4 W h kg1 and a power density of 2.027 kW kg1. The rGO NS structure relieved the volumetric expansion or contraction of the Fe3O4 particles and conned them to the voids during the redox reaction, which ensured a high cycle stability. The initial capacitance was retained by 95% aer 9500 cycles at a current density of 3.8 A g1. Examples of nanocomposites with other oxides to improve the electrochemical performance of iron-based materials are discussed as follows. For instance, the a-Fe2O3@Ni(OH)2 nanosheet hybrid composite prepared by Gao's group52 reached a specic capacitance of 356 F g1 aer 500 cycles at a current density of 16 A g1, with a capacity retention of 93.3%, due to the ample active sites and shortened ion diffusion distance. Xiang et al.200 embedded RuO2–Fe2O3 nanoparticles into ordered mesoporous carbon, which exhibited a maximum specic capacitance of 1668 F g1 at 1 A g1, much higher than that of the RuO2/OMC (1212 F g1). The addition of Fe2O3 promoted the uniform dispersion of the RuO2–Fe2O3 particles, thus reducing agglomeration and hence yielding good reversibility. Aer 3000 cycles, the RuO2–Fe2O3/OMC composite supercapacitor retained its initial capacity by 93%. The capacitance of the RuO2–Fe2O3/OMC composite electrode was contributed by both the double-layer capacitive and pseudocapacitive effects, thus having a promising capacitor energy density. The electrode had an energy density of 134 W h kg1 at a power density of 200 W kg1. Even at a power density of 6000 W kg1, the energy density of the electrode could still reach 80 W h kg1, which was nearly double that of the RuO2/OMC one (48 W h kg1). Zhong et al.260 synthesized a hierarchical heterostructure of Fe2O3 nanospheres on FeS2 nanosheets by using a hydrothermal treatment, which had a better capacitor performance than the bare Fe2O3 electrode. The composite electrode displayed a specic capacitance of 255 F g1 and good rate capability (145 F g1 at a current density of 8 A g1). 7.2.2 Asymmetric supercapacitors (ASCs). As discussed before, the low energy density has limited the development and application of supercapacitors. According to the relationship between energy density (E), specic capacitance (C) and voltage (V), E ¼ 1/2CV2, it is possible to increase the energy density by increasing the specic capacitance (C) and the voltage (V). Because the voltage can be increased at separate potential windows by conguring the positive and negative electrodes, the development of asymmetric supercapacitors (ASCs) has become an attractive strategy. ASCs are usually composed of a double-layer electrode as the power source and a battery-type Faraday electrode as the energy source, which not only operate over a wider voltage range, but also provide high energy density. As a result, ASCs are expected to be potential candidates for applications in the eld of hybrid vehicles, MEMS, handheld electronic devices and sensors. In recent years, great progress has been made in cathode materials, but anode materials are still under development. Currently, the main anode material is

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

carbon, whose specic capacity does not match that of the cathode. In addition, the working voltage of assembled ASCs is usually lower than 2 V due to the limitation of the cathodic– anodic potential window in aqueous electrolytes. Excitingly, the amorphous scale-shaped FeOOH nano-electrode synthesized by Wong's group219 showed a high pseudocapacitance of 1.11 F cm2 or 867 F g1. An asymmetric supercapacitor was assembled using FeOOH and Co–Ni double hydroxides as the anode and cathode, reaching a power density of 1831.6 W kg1 or 15.3 mW cm3. The asymmetric ultracapacitor could supply a high energy density of 86.4 W h kg1 or 0.7mW h cm3. Carbonaceous materials have been used as one of the cathode materials to construct ASCs. Shen et al.261 synthesized Fe3O4@carbon nanosheet composites by using ammonium ferric citrate as the precursor and graphene oxide as the structure-directing agent, which were used as the positive electrode, while porous carbon was employed as the negative electrode. The asymmetric supercapacitors achieved a maximum energy density of 18.3 W h kg1 at a power density of 351 W kg1, in the electrolyte of KOH/PVA gel. He's team7 reported an asymmetric supercapacitor, with double carbon-shelled C–C–Fe3O4 and activated carbon (AC) as the two electrodes. At current densities of 1 and 8 A g1, the corresponding specic capacitances were 117 and 50 F g1, respectively. When cycled at a current density of 1.5 A g1, the initial capacity could be retained by 96.7% aer 8000 cycles. The ASC device had energy densities in the range of 17–45 W h1, at power densities ranging from 400 to 8000 W kg1. The Fe3O4/reduced graphene oxide (rGO) composite electrode prepared by Lin et al.262 exhibited a high capacitance of 661 F g1 at 1 A g1. An asymmetric supercapacitor (ASC) was constructed with the Fe3O4/rGO as the anode and NiO/Ni3S2/3Dgraphene as the cathode, which had a maximum capacitance of 233 F g1 at a scan rate of 5 mV s1, corresponding to a maximum energy density of 82.5 W h kg1 at a power density of 930 W kg1. Due to 1D coaxial nanostructures and the presence of ne integrated conductive layers, the FeOOH@PPy electrode prepared by Wei et al.213 provided a specic capacitance of 1140 F g1 at a current density of 1 A g1, which was twice that of the bare FeOOH. The optimal energy density of the ASC device reached 39.1 W h kg1 at a power density of 800 W kg1. Similarly, the core–shell a-FeOOH@MnO2 heterostructures synthesized by Lv et al.35 showed much higher electrochemical performances than the a-FeOOH hollow-ber microspheres. The specic capacitance of the a-FeOOH@MnO2 electrode reached 597 F g1 at a current density of 1 A g1, which was more than twice that achieved by using the a-FeOOH electrode (232 F g1). When the current density was increased to 10 A g1, the capacitance was still retained by 74.2%, higher than 60% for the a-FeOOH electrode. Aer 2000 cycles, the capacitance retention was as high as 97.1%, also higher than 93.2% for the a-FeOOH electrode. The ASC device had a maximum energy density of 34.2 W h kg1 at a power density of 815 W kg1. MnO2 is another widely used cathode material for ASC devices. The g-FeOOH nanosheets developed by Chen et al.18 showed excellent pseudocapacitive performance, reaching an

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

extraordinary power density of 9000 W kg1. ASCs made of gFeOOH NSs and 2D MnO2 as the anode and cathode materials, respectively, delivered a maximum power density of 16 000 W kg1 and an energy density of 37.4 W h kg1, when charging/ discharging in a mild aqueous electrolyte at a maximum cell voltage of 1.85 V. The tectorum-like Fe2O3/polypyrrole (PPy) nanoarrays fabricated by Hu's group25 gave high electric-storage properties due to the intriguing hierarchical nanoarchitecture of the composite. The Fe2O3/PPy electrode had an area capacitance of 382.4 mF cm2 at a current density of 0.5 mA cm2, which was retained at 170.5 mF cm2 even at 10 mV s1. The initial capacity was retained by 97.2% aer 5000 cycles. A solid asymmetric supercapacitor made of the Fe2O3/PPy anode and MnO2 cathode exhibited a high energy density of 0.22 mW h cm3 at a power density of 165.6 mW cm3. Lu's group206 developed Ti-doped Fe2O3@poly(3,4-ethylenedioxythiophene) (PEDOT) nanorod array electrode, with a surface capacitance of 1.15 F cm2 at 1 mA cm2, much higher than that for the original Fe2O3 electrode (0.46 F cm2). Also, the device showed a very good cycling stability, with a retention of above 96% aer 30 000 cycles. The asymmetric supercapacitor was assembled with the Ti-doped Fe2O3@PEDOT nanorod arrays and MnO2 as the anode and cathode materials, which showed a maximum energy density of 0.89 mW h cm3 and an ultrahigh stability with 96.1% capacitance retention over 30 000 cycles. Hierarchical heterostructures and hybrid core structures have been shown to be able to provide rich active sites.75 For example, a hierarchical heterostrucure composite, NiCo2S4@Fe2O3 NRs, was prepared by Zhai's group,8 with CV curves to have an undistorted rectangular shape over a wide range of scan rates (5–800 mV s1), indicating that the composite electrode had good pseudocapacitive performance and high rate capability, as revealed in Fig. 23. Compared with the electrodes based on NiCo2S4 and Fe2O3, the NiCo2S4@Fe2O3 electrode exhibited superior capacitive performance, which was evidenced by the fact that the charge–discharge curve was more symmetrical. The NiCo2S4@Fe2O3 electrode delivered a high specic capacitance of 342 F g1 at a current density of 5 mV s1. The enhanced capacitive performance was attributed to the composite structure with highly conductive NiCo2S4 scaffolds and graed Fe2O3 NRs. According to the Nyquist plot, the NiCo2S4@Fe2O3 electrode possessed an electrode transfer resistance (Rct) of 0.76 U, implying the presence of a rapid

Fig. 23 (a) CV curves of the NiCo2S4@Fe2O3 electrode collected from 5 to 800 mV s1 in a potential window between 0 and 1 V vs. SCE. (b) Galvanostatic charge/discharge curves of the NiCo2S4, NiCo2S4@Fe2O3 and Fe2O3 electrodes collected at a current density of 2.5 A g1 [reprinted with permission from ref. 8, copyright 2016 Elsevier].

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9359

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

Faraday reaction between the NiCo2S4 and Fe2O3 particles. An ASC was constructed with the NiCo2S4@Fe2O3 nanoneedle arrays as the anode and MnO2 nanosheet arrays as the cathode. Because the voltage of the ASC device was increased to 2.3 V, the energy density could be signicantly increased. In a neutral aqueous solution, the ASC device provided a volumetric energy density of 2.29 mW h cm3 at an average power density of 196 mW cm3. Moreover, the ASC maintained an energy density of 1.08 mW h cm3 even at a high power density of 2063 mW cm3, demonstrating excellent power capacity. The nickel nanotube arrays (NiNTAs)@Fe2O3 electrodes prepared by Xia's team39 had excellent capacitive properties, with a specic capacitance of 418.7 F g1 at 10 mV s1, which was much higher than those of the Fe2O3 thin lm (76.4 F g1) and the Fe2O3 nanorod (177.1 F g1). The NiNTAs@Fe2O3 nanoneedle electrode retained an ultra-high specic capacitance of 215.3 F g1 even at a high current density of 64 A g1. Meanwhile, the electrode exhibited a minimum equivalent series resistance (Rs) of 2.3 U and a charge-transfer resistance (Rct) of 2.3 U. The ultrane and hollow Ni nanoframe structure is extremely advantageous for electron collection and fast Faraday reaction. At the same time, this type of special structure is also benecial to the cycle stability of the electrode. Aer 5000 cycles at 100 mV s1, the liquid-state ASC showed excellent cycling stability, with a capacitance retention of 92.3%, while the solidstate ASC had a retention of 79.3%. In addition, the CV curve of the ASC device was almost unchanged in shape, indicating its high reversibility and cycling stability. The NiNTA@Fe2O3 nanoneedle electrodes achieved maximum energy densities of 34.1 W h kg1 and 32.2 W h kg1, corresponding to power densities of 3197.7 W kg1 and 3199.5 W kg1, in an aqueous electrolyte and a quasi-solid gel electrolyte, respectively. 7.2.3 Transparent exible supercapacitors (TFS). Transparent supercapacitors have wide applications in the elds of transparent smartphones, implantable medical devices and smart/wearable devices.15 Previous studies have focused on carbon material based lms, such as carbon lms, 2D rGO lms, 2D CVD vinyl graphene lms, and 2D carbon nanotube lms.223,263,264 There is always a contradiction between the mass load and optical transmittance. A high mass load can produce a high capacitance but will result in decreased transparency. So far, the electrochemical performance of transparent supercapacitors is still far below that of opaque supercapacitors. In this case, transition metal oxides such as iron oxides are explored as potential electrode materials for transparent exible supercapacitors. Huang and co-workers15 fabricated a Fe2O3 nanowire network@graphene transparent lm (FNW@Gr-TF) electrode material, in which graphene wrapped the Fe2O3 nanowires, forming a fast 3D conductive network structure. The graphene layer also adhered the active material onto the ITO layer, greatly improving the stability of the electrode material. Due to the graphene encapsulation effect, the initial capacitance of the FNW@Gr-TF electrode could be retained by over 92% aer 10 000 cycles, much higher than 16.1% for FNW–TF without graphene. The FNW@Gr-TF electrode had a specic capacitance of 3.3 mF cm2 at a scan rate of 10 mV s1, at least 100 times

9360 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

higher than that of the carbon material electrode transparent device. The volumetric energy density of the FNW@Gr-TF electrode was 8 mW h cm3, which was equivalent to 18.6 times212 that of the transparent electrode capacitor based on FFT–GP (430 mW h cm3).264 The value was also much larger than that of the carbon lm (47 mW h cm3)264 and that of the graphene layer supercapacitor (50 mW h cm3).263 The transmissivity of the FNW–TF electrode was about 62.5% at a wavelength of 550 nm. Due to the polymer (PVA) electrolyte and the counter electrode, the assembled supercapacitor transmittance was decreased to 42.9%. There is no signicant change in the CV curve when the ultracapacitor device was bent at 0 , 30 , 60 and 90 , as seen in Fig. 24a. This outstanding exibility provides it with great potential in the wearable and folding appliance industries. The rapid growth of exible/wearable consumer electronics requires that energy storage devices not only have excellent electrochemical properties, but also have the tolerance to fold and bend.265,266 Flexible capacitors usually require the use of exible substrates, such as PET, textiles, carbon cloth, stainless steel sheets, paper and bers. The amorphous FeOOH/MnO2 composites prepared by Wu's group36 had high specic capacitance, good rate capability and cycling stability as transparent capacitor electrodes. The composite electrode delivered a specic capacitance of 350.2 F g1 at a current density of 0.5 A g1. The specic capacitance remained at 159.5 F g1 even as the current density was increased to 20 A g1. In addition, aer 10 000 cycles at a current density of 15 A g1, the capacitance was retained by 95.6%. SCs could be printed on different substrates, including PET, paper and textiles, which could power a 1.9 V yellow LED, with large bending and stretching, as demonstrated in Fig. 24b–d. Lokhande's group267 prepared MnO2 and Fe2O3 thin lms on highly exible stainless steel sheets. Asymmetric exible solid supercapacitors (FSS-SCs) were assembled by using the MnO2

(a) CV curves of the supercapacitors based on the FNW@GrTF in the bent states at a scanning rate of 100 mV s1 [reprinted with permission from ref. 15, copyright 2017, IOP]. (b) Photographs of all the solid-state flexible SC devices printed on PET, paper and textile substrates. (c) Photographs after bending. (d) Photograph of the SCs after bending in series to light up a yellow LED [reprinted with permission from ref. 36, copyright 2017, Elsevier]. Fig. 24

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

nanosheet as the anode and the Fe2O3 nanosheet as the cathode, with Na2SO4 and carboxymethylcellulose (CMC) gel as electrolyte and separator, respectively. These exible ASCs had a maximum specic capacitance of 92 F g1 and an energy density of 41.8 W h kg1, making their integration with portable electronics suitable. Wu's team193 synthesized two hybrid composites, a-Fe2O3@MnCo2O4 and a-Fe2O3@Co3O4, on a exible carbon cloth. Area capacitances of 490 mF cm2 and 1073 mF cm2 at 1 mA cm2 were observed for the a-Fe2O3@MnCo2O4 and a-Fe2O3@Co3O4 composites, respectively. Aer 6000 charge/discharge cycles, the capacities of the a-Fe2O3@MnCo2O4 and a-Fe2O3@Co3O4 composites were retained by 74.6% and 77.8%, respectively. The Ti-doped FeOOH QD/GN/BC exible anode prepared by Yuan's group222 exhibited a high area capacitance of 3322 mF cm2 and a specic capacitance of 235.6 F g1 at 2 mA cm2. Aer 6000 cycles, the capacitance retention reached 94.7%. ASC devices were assembled with the Ti-doped FeOOH QD/GN/BC and Mn3O4/GN/BC as the anodes and cathodes, which delivered an ultrahigh area energy density of 0.541 mW h cm2 and an ultrahigh volumetric energy density of 9.02 mW h cm3, over the operating voltage range of 0–1.8 V. At the same time, the capacitor had a good capacity retention in bent states. The graphene-coated FeOOH nanowire arrays and Co(OH)2 nanosheet thin-lm transparent electrodes prepared by Zhang et al.223 showed good transparency with a transmittance of 50.5% at 550 nm. When operated at 1.8 V, the asymmetric exible ultracapacitor made of Co@Gr-TF//Fe@Gr-TF exhibited a high specic capacitance of 5.5 mF cm2 and a high energy density of 1.04 mW h cm3, together with a capacity retention rate of 83.5% aer 10 000 cycles.

8. Conclusions and perspectives Here, we present a comprehensive review that highlights the signicant progress made in the design and preparation of ironbased nanomaterials over the last ve years. It includes 1D, 2D, and 3D nanostructural design and synthesis of Fe2O3/Fe3O4, nanocomposites of iron oxides with various carbonaceous supports, other metal oxides and polymers, and recent research hotspots—iron hydroxide (FeOOH) and iron suldes (FeS2). In addition, the application of iron-based nanomaterials in LIBs and SCs was reviewed according to the classication. For future research on iron-based electrode materials, it is believed that we should focus on the following aspects. (1) Design and synthesis of new nanostructured iron-based materials The preparation of new structures, such as grape-like structures and complex hollow spherical structures, has promoted the development of iron-based materials. Therefore, it is an important research direction for these materials. (2)

Formation mechanism of nanostructures

Although several new structures have been prepared, they are either accidentally obtained or relied on experiences. Therefore, the formation mechanisms are still elusive. For further

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

improvement and development of new structures, it is necessary to elucidate the formation mechanism, which will be a huge challenge and thus should be strengthened in the future. (3)

Electrochemical reaction mechanisms

Various advanced techniques have been used to study the electrochemical mechanism of electrode materials. For example, the growth of lithium metal dendrites in electrolyte was in situ observed by using an optical microscope. The morphological changes of the Si anode were monitored with an in situ scanning electron microscope (SEM). An atomic force microscope (AFM) was used to study the interface electrochemistry of silicon nanowires. In situ transmission electron microscopy (TEM) was used to directly observe the dynamic behavior and transformation mechanism of the Ti3Sn/NiTi alloy and Fe2O3/graphene. The formation of dendrites on the lithium metal anode and the passivation of the SEI layer were claried in self-discharge measurements. These advanced technologies could also be used to study the electrochemical reaction mechanism of iron-based materials, which will guide the design and synthesis of new nanostructures. (4) Applications of iron-based electrode materials in new areas It is well known that different applications put different requirements on the structure and performance of electrodes. For example, transparent exible electrodes for smart wearable devices require high light transmittance and mechanical exibility, while the electrochemical performance should not be compromised. (5) Optimization and simplication of synthetic methods for large scale applications At present, most of the synthetic processes are complicated and thus not cost-effective, which is not suitable for commercial applications. As a result, it is crucial to develop a simple, lowcost and scalable method to precisely control the chemical composition and physical structure. In short, the developments of iron-based nanomaterials and their applications as electrodes are still in the research stage, thus having a long way to go in terms of commercial applications. Hopefully, by comprehensively summarizing the recent research progress, this review will provide a reference for the follow-up research in this interesting and prospective eld and promote the development of iron-based electrode materials.

Conflicts of interest There are no conicts to declare.

Acknowledgements This work was supported by the Natural Science Fund of China (51762023, 51463018), the Jiangxi Provincial Department of Education, the Training Program of Outstanding Young Scientists in Jiangxi Province (20171BCB23070), Jingdezhen Science

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9361

View Article Online

Journal of Materials Chemistry A

and Technology Bureau, the Key Laboratory for Microstructural Control of Metallic Materials of Jiangxi Province (JW201523006), and Special Fund for Visiting Scholars in the Development Plan of Young Teachers in the Undergraduate Colleges and Universities of Jiangxi Province.

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

References 1 H. Geng, D. Ge, S. Lu, J. Wang, Z. Ye, Y. Yang, J. Zheng and H. Gu, Chemistry, 2015, 21, 11129–11133. 2 Z. H. Na, G. Huang, F. Liang, D. Yin and L. Wang, Chem.– Eur. J., 2016, 22, 12081–12087. 3 S. H. Choi and Y. C. Kang, Carbon, 2014, 79, 58–66. 4 K. Cao, L. Jiao, H. Liu, Y. Liu, Y. Wang, Z. Guo and H. Yuan, Adv. Energy Mater., 2015, 5, 1401421. 5 X. Wen, X. Wei, L. Yang and P. K. Shen, J. Mater. Chem. A, 2015, 3, 2090–2096. 6 Y. Jin, L. Dang, H. Zhang, C. Song, Q. Lu and F. Gao, Chem. Eng. J., 2017, 326, 292–297. 7 X. Li, L. Zhang and G. He, Carbon, 2016, 99, 514–522. 8 R. Jia, F. Zhu, S. Sun, T. Zhai and H. Xia, J. Power Sources, 2017, 341, 427–434. 9 X. Gu, L. Chen, Z. Ju, H. Xu, J. Yang and Y. Qian, Adv. Funct. Mater., 2013, 23, 4049–4056. 10 C. Liu, Y. Wang, P. Zhao, W. Li, Q. Wang, P. Sun, X. Chuai and G. Lu, J. Colloid Interface Sci., 2017, 505, 1039–1046. 11 P. Sun, B. Wang, L. Zhao, H. Gao, T. Wang, X. Yang, C. Liu and G. Lu, Sens. Actuators, B, 2017, 252, 322–329. 12 Q. Wu, R. Zhao, W. Liu, X. Zhang, X. Shen, W. Li, G. Diao and M. Chen, J. Power Sources, 2017, 344, 74–84. 13 H. Xue, D. Y. W. Yu, J. Qing, X. Yang, J. Xu, Z. Li, M. Sun, W. Kang, Y. Tang and C.-S. Lee, J. Mater. Chem. A, 2015, 3, 7945–7949. 14 L. Ma, H. Ma, N. Gao, J. Wang and X. Zhang, J. Mater. Sci., 2015, 51, 1959–1965. 15 X. Huang, H. Zhang and N. Li, Nanotechnology, 2017, 28, 075402. 16 M. Chen, E. Zhao, Q. Yan, Z. Hu, X. Xiao and D. Chen, Sci. Rep., 2016, 6, 29381. 17 J. S. Cho, Y. J. Hong, J. H. Lee and Y. C. Kang, Nanoscale, 2015, 7, 8361–8367. 18 Y. C. Chen, Y. G. Lin, Y. K. Hsu, S. C. Yen, K. H. Chen and L. C. Chen, Small, 2014, 10, 3803–3810. 19 B. Wang, H. B. Wu, L. Zhang and X. W. Lou, Angew. Chem., Int. Ed., 2013, 52, 4165–4168. 20 X. Lin, J. Zhang, X. Tong, H. Li, X. Pan, P. Ning and Q. Li, Sci. Rep., 2017, 7, 9657. 21 S. Li, M. Wang, Y. Luo and J. Huang, ACS Appl. Mater. Interfaces, 2016, 8, 17343–17351. 22 H. H. Fan, H. H. Li, K. C. Huang, C. Y. Fan, X. Y. Zhang, X. L. Wu and J. P. Zhang, ACS Appl. Mater. Interfaces, 2017, 9, 10708–10716. 23 C. Ding, Y. Zeng, L. Cao, L. Zhao and Q. Meng, ACS Appl. Mater. Interfaces, 2016, 8, 7977–7990. 24 J. M. Jeong, B. G. Choi, S. C. Lee, K. G. Lee, S. J. Chang, Y. K. Han, Y. B. Lee, H. U. Lee, S. Kwon, G. Lee, C. S. Lee and Y. S. Huh, Adv. Mater., 2013, 25, 6250–6255.

9362 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

25 L. Wang, Y. Huiling, X. Liu, R. Zeng, M. Li, Y. Huang and X. Hu, Angew. Chem., Int. Ed., 2017, 56, 1105–1110. 26 J. He, Q. Li, Y. Chen, C. Xu, K. Zhou, X. Wang, W. Zhang and Y. Li, Carbon, 2017, 114, 111–116. 27 R. Jin, L. Yang, G. Li and G. Chen, New J. Chem., 2014, 38, 5840–5845. 28 R. Kumar, R. K. Singh, R. Vaz Alfredo, R. Savu and S. A. Moshkalev, ACS Appl. Mater. Interfaces, 2017, 9, 8880–8890. 29 L. Guo, N. Xie, C. Wang, X. Kou, M. Ding, H. Zhang, Y. Sun, H. Song, Y. Wang and G. Lu, Sens. Actuators, B, 2018, 255, 1015–1023. 30 J. Liu, M. Zheng, X. Shi, H. Zeng and H. Xia, Adv. Funct. Mater., 2016, 26, 919–930. 31 H. Liang, W. Chen, Y. Yao, Z. Wang and Y. Yang, Ceram. Int., 2014, 40, 10283–10290. 32 D. Zhao, Q. Hao and C. Xu, Electrochim. Acta, 2015, 180, 493–500. 33 C. Gu, X. Song, S. Zhang, S. O. Ryu and J. Huang, J. Alloys Compd., 2017, 714, 6–12. 34 C. Ma, J. Shi, Y. Zhao, N.-j. Song and Y. Wang, Chem. Eng. J., 2017, 326, 507–517. 35 Y. Lv, H. Che, A. Liu, J. Mu, C. Dai, X. Zhang, Y. Bai, G. Wang and Z. Zhang, J. Appl. Electrochem., 2017, 47, 433–444. 36 Q. Lu, L. Liu, S. Yang, J. Liu, Q. Tian, W. Yao, Q. Xue, M. Li and W. Wu, J. Power Sources, 2017, 361, 31–38. 37 R. Jin, Q. Wang, H. Li, Y. Ma, Y. Sun and G. Li, Appl. Surf. Sci., 2017, 403, 62–70. 38 M. Y. Son, Y. J. Hong, J. K. Lee and Y. Chan Kang, Nanoscale, 2013, 5, 11592–11597. 39 Y. Li, J. Xu, T. Feng, Q. Yao, J. Xie and H. Xia, Adv. Funct. Mater., 2017, 27, 1606728. 40 W. Ma, X. Liu, X. Lei, Z. Yuan and Y. Ding, Chem. Eng. J., 2018, 334, 725–731. 41 C. Fu, A. Mahadevegowda and P. S. Grant, J. Mater. Chem. A, 2015, 3, 14245–14253. 42 H. Liang, B. Xu and Z. Wang, Mater. Chem. Phys., 2013, 141, 727–734. 43 Q. Q. Xiong, X. H. Xia, J. P. Tu, J. Chen, Y. Q. Zhang, D. Zhou, C. D. Gu and X. L. Wang, J. Power Sources, 2013, 240, 344–350. 44 Z. Li, X. Lai, H. Wang, D. Mao, C. Xing and D. Wang, Nanotechnology, 2009, 20, 245603. 45 H. Geng, Q. Zhou, Y. Pan, H. Gu and J. Zheng, Nanoscale, 2014, 6, 3889–3894. 46 X. Zheng, Z. Han, S. Yao, H. Xiao, F. Chai, F. Qu and X. Wu, Dalton Trans., 2016, 45, 7094–7103. 47 L. Yang, G. Guo, H. Sun, X. Shen, J. Hu, A. Dong and D. Yang, Electrochim. Acta, 2016, 190, 797–803. 48 Y. Xiang, W. Zhu, W. Guo, F. Lou, B. Deng, D. Liu, Z. Xie, J. Li, D. Qu and H. Tang, J. Alloys Compd., 2017, 719, 347– 352. 49 X. Gong, S. Li and P. S. Lee, Nanoscale, 2017, 9, 10794– 10801. 50 Y. Wang, J. Guo, L. Li, Y. Ge, B. Li, Y. Zhang, Y. Shang and A. Cao, Nanotechnology, 2017, 28, 345703.

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

51 X. Liu, W. Si, J. Zhang, X. Sun, J. Deng, S. Baunack, S. Oswald, L. Liu, C. Yan and O. G. Schmidt, Sci. Rep., 2014, 4, 7452. 52 H. Jiang, H. Ma, Y. Jin, L. Wang, F. Gao and Q. Lu, Sci. Rep., 2016, 6, 31751. 53 P. Poizot, S. Laruelle, S. Grugeon, L. Dupont and J. M. Tarascon, Nature, 2000, 407, 496–499. 54 Y. Sun, N. Liu and Y. Cui, Nat. Energy, 2016, 1, 16071. 55 M. Chen, X. Shen, K. Chen, Q. Wu, P. Zhang, X. Zhang and G. Diao, Electrochim. Acta, 2016, 195, 94–105. 56 F. Wu, R. Huang, D. Mu, B. Wu and S. Chen, ACS Appl. Mater. Interfaces, 2014, 6, 19254–19264. 57 C. T. Cherian, J. Sundaramurthy, M. Kalaivani, P. Ragupathy, P. S. Kumar, V. Thavasi, M. V. Reddy, C. H. Sow, S. G. Mhaisalkar, S. Ramakrishna and B. V. R. Chowdari, J. Mater. Chem., 2012, 22, 12198. 58 M.-S. Balogun, Z. Wu, Y. Luo, W. Qiu, X. Fan, B. Long, M. Huang, P. Liu and Y. Tong, J. Power Sources, 2016, 308, 7–17. 59 K. K. Lee, S. Deng, H. M. Fan, S. Mhaisalkar, H. R. Tan, E. S. Tok, K. P. Loh, W. S. Chin and C. H. Sow, Nanoscale, 2012, 4, 2958–2961. 60 Y. Park, M. Oh, J. S. Park, S.-H. Baek, M. Kim, S. Kim and J. H. Kim, Carbon, 2015, 94, 9–17. 61 F. Zheng, M. He, Y. Yang and Q. Chen, Nanoscale, 2015, 7, 3410–3417. 62 L. Du, C. Xu, J. Liu, Y. Lan and P. Chen, Nanoscale, 2017, 9, 14376–14384. 63 X. Lv, J. Deng, B. Wang, J. Zhong, T.-K. Sham, X. Sun and X. Sun, Chem. Mater., 2017, 29, 3499–3506. 64 Y. Zeng, M. Yu, Y. Meng, P. Fang, X. Lu and Y. Tong, Adv. Energy Mater., 2016, 6, 1601053. 65 L. Zhang, H. B. Wu and X. W. D. Lou, Adv. Energy Mater., 2014, 4, 1300958. 66 W. Zhou and L. Guo, Chem. Soc. Rev., 2015, 44, 6697–6707. 67 L. Zhou, Z. Zhuang, H. Zhao, M. Lin, D. Zhao and L. Mai, Adv. Mater., 2017, 29, 1602914. 68 L. Yu, H. Hu, H. B. Wu and X. W. Lou, Adv. Mater., 2017, 29, 1604563. 69 W. Shen, F. Lin, X. Jiang, H. Li, Z. Ai and L. Zhang, Chem. Eng. J., 2017, 308, 880–888. 70 X. Ding, S. Wang, W. Shen, Y. Mu, L. Wang, H. Chen and L. Zhang, Water Res., 2017, 112, 9–18. 71 C. Poonjarernsilp, N. Sano, N. Sawangpanich, T. Charinpanitkul and H. Tamon, Green Chem., 2014, 16, 4936–4943. 72 Y. Wang, Y. Wu, L. Xing, Q. Wang and X.-Y. Xue, J. Alloys Compd., 2016, 689, 655–661. 73 J. S. Cho, J. S. Park and Y. C. Kang, Sci. Rep., 2016, 6, 38933. 74 S. H. Choi and Y. C. Kang, Nanoscale, 2013, 5, 4662–4668. 75 X. Lu, Y. Zeng, M. Yu, T. Zhai, C. Liang, S. Xie, M. S. Balogun and Y. Tong, Adv. Mater., 2014, 26, 3148–3155. 76 S. Zhang, B. Yin, Z. Wang and F. Peter, Chem. Eng. J., 2016, 306, 193–203. 77 M. Zhang, J. Alloys Compd., 2015, 648, 134–138. 78 G. Z. Wei, X. Lu, F. S. Ke, L. Huang, J. T. Li, Z. X. Wang, Z. Y. Zhou and S. G. Sun, Adv. Mater., 2010, 22, 4364–4367.

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

79 X. Xiao, X. Liu, H. Zhao, D. Chen, F. Liu, J. Xiang, Z. Hu and Y. Li, Adv. Mater., 2012, 24, 5762–5766. 80 S. Grugeon, S. Laruelle, L. Dupont and J. M. Tarascon, Solid State Sci., 2003, 5, 895–904. 81 M. S. Islam, D. J. Driscoll, C. A. J. Fisher and P. R. Slater, Chem. Mater., 2005, 17, 5085–5092. 82 B. Lv, Z. Liu, H. Tian, Y. Xu, D. Wu and Y. Sun, Adv. Funct. Mater., 2010, 20, 3987–3996. 83 Z. Liu, B. Lv, Y. Xu and D. Wu, J. Mater. Chem. A, 2013, 1, 3040. 84 S. Chaudhari, D. Bhattacharjya and J.-S. Yu, RSC Adv., 2013, 3, 25120. 85 N. Shen, M. Keppeler, B. Stiaszny, H. Hain, F. Maglia and M. Srinivasan, Beilstein J. Nanotechnol., 2017, 8, 2032–2044. 86 G. Binitha, M. S. Soumya, A. A. Madhavan, P. Praveen, A. Balakrishnan, K. R. V. Subramanian, M. V. Reddy, S. V. Nair, A. S. Nair and N. Sivakumar, J. Mater. Chem. A, 2013, 1, 11698. 87 L.-Q. Y. Lei Liu, H.-W. Liang, H.-P. Cong, J. Jiang and S.-H. Yu, Nano, 2013, 7, 1368–1378. 88 L. Ji, O. Toprakci, M. Alcoutlabi, Y. Yao, Y. Li, S. Zhang, B. Guo, Z. Lin and X. Zhang, ACS Appl. Mater. Interfaces, 2012, 4, 2672–2679. 89 L. X. J. Chen, W. Li and X. Gou, Adv. Mater., 2005, 17, 582– 586. 90 Z. Wang, D. Luan, S. Madhavi, C. M. Li and X. W. Lou, Chem. Commun., 2011, 47, 8061–8063. 91 L. Zhang, J. Deng, L. Liu, W. Si, S. Oswald, L. Xi, M. Kundu, G. Ma, T. Gemming, S. Baunack, F. Ding, C. Yan and O. G. Schmidt, Adv. Mater., 2014, 26, 4527–4532. 92 V. Augustyn, J. Come, M. A. Lowe, J. W. Kim, P. L. Taberna, S. H. Tolbert, H. D. Abruna, P. Simon and B. Dunn, Nat. Mater., 2013, 12, 518–522. 93 S. Xiong, X. Jing, C. Di, W. Rongming, H. Xianluo, S. Guozhen and W. Zhong Lin, CrystEngComm, 2011, 13, 7114. 94 J. Lu, P. Qing, W. Zhongying, N. Caiyun, L. Lihong and L. Yadong, J. Mater. Chem. A, 2013, 1, 5232. 95 J. Cai, C. Suyue, J. Mei, H. Jun, M. Yurong and Q. Limin, CrystEngComm, 2014, 16, 1553–1559. 96 W. Cheng, J. He, T. Yao, Z. Sun, Y. Jiang, Q. Liu, S. Jiang, F. Hu, Z. Xie, B. He, W. Yan and S. Wei, J. Am. Chem. Soc., 2014, 136, 10393–10398. 97 G. S. D. Larcher, R. Patrice and J.-M. Tarascon, Chem. Mater., 2003, 15, 3543–3551. 98 R. Liu, Y. Jiang, Q. Lu, W. Du and F. Gao, CrystEngComm, 2013, 15, 443–446. 99 J. Zhu, Z. Yin, D. Yang, T. Sun, H. Yu, H. E. Hoster, H. H. Hng, H. Zhang and Q. Yan, Energy Environ. Sci., 2013, 6, 987. 100 L. L. Tian, M. J. Zhang, C. Wu, Y. Wei, J. X. Zheng, L. P. Lin, J. Lu, K. Amine, Q. C. Zhuang and F. Pan, ACS Appl. Mater. Interfaces, 2015, 7, 26284–26290. 101 H. Zhang, L. Zhou, O. Noonan, D. J. Martin, A. K. Whittaker and C. Yu, Adv. Funct. Mater., 2014, 24, 4337–4342. 102 L. S. Zhong, J. S. Hu, H. P. Liang, A. M. Cao, W. G. Song and L. J. Wan, Adv. Mater., 2006, 18, 2426–2431.

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9363

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

103 L. Zhang, H. B. Wu, S. Madhavi, H. H. Hng and X. W. Lou, J. Am. Chem. Soc., 2012, 134, 17388–17391. 104 P. Zhao, D. Li, S. Yao, Y. Zhang, F. Liu, P. Sun, X. Chuai, Y. Gao and G. Lu, J. Power Sources, 2016, 318, 49–56. 105 J. Zhang, Y. Sun, Y. Yao, T. Huang and A. Yu, J. Power Sources, 2013, 222, 59–65. 106 R. M. R. Y. Yin, C. K. Erdonmez, S. Hughes, G. A. Somorjai and A. P. Alivisatos, Science, 2004, 304, 711–714. 107 H. J. Fan, U. Gosele and M. Zacharias, Small, 2007, 3, 1660– 1671. 108 L. Hu, N. Yan, Q. Chen, P. Zhang, H. Zhong, X. Zheng, Y. Li and X. Hu, Chemistry, 2012, 18, 8971–8977. 109 M. K. H. J. Fan, R. Scholz, D. Hesse, K. Nielsch, M. Zacharias and U. G¨ osele, Nano Lett., 2007, 7, 993–997. 110 S. Jana, J. W. Chang and R. M. Rioux, Nano Lett., 2013, 13, 3618–3625. 111 X. Wang, X.-L. Wu, Y.-G. Guo, Y. Zhong, X. Cao, Y. Ma and J. Yao, Adv. Funct. Mater., 2010, 20, 1680–1686. 112 P. Sun, X. He, W. Wang, J. Ma, Y. Sun and G. Lu, CrystEngComm, 2012, 14, 2229. 113 X. Chai, C. Shi, E. Liu, J. Li, N. Zhao and C. He, Appl. Surf. Sci., 2015, 347, 178–185. 114 J. S. Xu and Y. J. Zhu, ACS Appl. Mater. Interfaces, 2012, 4, 4752–4757. 115 L. Chun, X. Wu, X. Lou and Y. Zhang, Electrochim. Acta, 2010, 55, 3089–3092. 116 J. B. L. J. M. Ma, X. C. Duan, X. D. Liu and W. J. Zheng, J. Phys. Chem. C, 2010, 114, 10671–10676. 117 H. Zhang, X. Sun, X. Huang and L. Zhou, Nanoscale, 2015, 7, 3270–3275. 118 R. M. Kore and B. J. Lokhande, J. Alloys Compd., 2017, 725, 129–138. 119 X. W. Lou and L. A. A. Zichao Yang, Adv. Mater., 2008, 20, 3987–4019. 120 L. Zhou, H. Xu, H. Zhang, J. Yang, S. B. Hartono, K. Qian, J. Zou and C. Yu, Chem. Commun., 2013, 49, 8695–8697. 121 Y. NuLi, P. Zhang, Z. Guo, P. Munroe and H. Liu, Electrochim. Acta, 2008, 53, 4213–4218. 122 X. Gou, G. Wang, J. Park, H. Liu and J. Yang, Nanotechnology, 2008, 19, 125606. 123 X.-H. Ma, X.-Y. Feng, C. Song, B.-K. Zou, C.-X. Ding, Y. Yu and C.-H. Chen, Electrochim. Acta, 2013, 93, 131–136. 124 T. R. P. S. Shivakumara and N. Munichandraiah, ECS Electrochem. Lett., 2013, 2, 60–62. 125 Y. Han, Y. Wang, L. Li, Y. Wang, L. Jiao, H. Yuan and S. Liu, Electrochim. Acta, 2011, 56, 3175–3181. 126 X. Zheng, X. Yan, Y. Sun, Y. Yu, G. Zhang, Y. Shen, Q. Liang, Q. Liao and Y. Zhang, J. Colloid Interface Sci., 2016, 466, 291–296. 127 E. Mitchell, R. K. Gupta, M.-D. Kwadwo, D. Kumar, K. Ramasamy, B. K. Gupta and P. Kahol, New J. Chem., 2014, 38, 4344–4350. 128 Y. Zhang, Y. Li, H. Li, Y. Zhao, F. Yin and Z. Bakenov, Electrochim. Acta, 2016, 216, 475–483. 129 T. Muraliganth, A. Vadivel Murugan and A. Manthiram, Chem. Commun., 2009, 7360–7362, DOI: 10.1039/b916376j.

9364 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

130 H. Xia, Y. Wan, G. Yuan, Y. Fu and X. Wang, J. Power Sources, 2013, 241, 486–493. 131 L. O'Neill, C. Johnston and P. S. Grant, J. Power Sources, 2015, 274, 907–915. 132 A. B. Ganganboina, A. D. Chowdhury and R.-an Doong, Electrochim. Acta, 2017, 245, 912–923. 133 H. Gu, Y. Zhang, M. Huang, F. Chen, Z. Yang, X. Fan, S. Li, W. Zhang, S. Yang and M. Li, Inorg. Chem., 2017, 56, 7657– 7667. 134 Q. Xin, L. Gai, Y. Wang, W. Ma, H. Jiang and Y. Tian, J. Alloys Compd., 2017, 691, 592–599. 135 X. H. Huang, J. B. Wu, Y. Lin and R. Q. Guo, Mater. Lett., 2016, 175, 199–202. 136 C. C. Yec and H. C. Zeng, J. Mater. Chem. A, 2014, 2, 4843– 4851. 137 X. Wang, Y. Liu, H. Arandiyan, H. Yang, L. Bai, J. Mujtaba, Q. Wang, S. Liu and H. Sun, Appl. Surf. Sci., 2016, 389, 240– 246. 138 W. Xie, L. Gu, X. Sun, M. Liu, S. Li, Q. Wang, D. Liu and D. He, Electrochim. Acta, 2016, 220, 107–113. 139 B.-Y. Jung, H.-S. Lim, Y.-K. Sun and K.-D. Suh, J. Power Sources, 2013, 244, 177–182. 140 H. Luo, D. Ji, Z. Yang, Y. Huang, G. Xiong, Y. Zhu, R. Guo and Y. Wan, Chem. Eng. J., 2017, 326, 151–161. 141 M. Hu, A. A. Belik, M. Imura and Y. Yamauchi, J. Am. Chem. Soc., 2013, 135, 384–391. 142 B. D. Anderson and J. B. Tracy, Nanoscale, 2014, 6, 12195– 12216. 143 M. B. Zakaria, M. Hu, M. Imura, R. R. Salunkhe, N. Umezawa, H. Hamoudi, A. A. Belik and Y. Yamauchi, Chem.–Eur. J., 2014, 20, 17375–17384. 144 P. Innocenzi, L. Malfatti and G. J. A. A. Soler-Illia, Chem. Mater., 2011, 23, 2501–2509. 145 Z. W. Zhao, T. Wen, K. Liang, Y. F. Jiang, X. Zhou, C. C. Shen and A. W. Xu, ACS Appl. Mater. Interfaces, 2017, 9, 3757– 3765. 146 Y. Zhao, J. Li, C. Wu, Y. Ding and L. Guan, ChemPlusChem, 2012, 77, 748–751. 147 Y. Chen, H. Xia, L. Lu and J. Xue, J. Mater. Chem., 2012, 22, 5006. 148 J. Du, Y. Ding, L. Guo, L. Wang, Z. Fu, C. Qin, F. Wang and X. Tao, Appl. Surf. Sci., 2017, 425, 164–169. 149 L. Wang, J. Wu, Y. Chen, X. Wang, R. Zhou, S. Chen, Q. Guo, H. Hou and Y. Song, Electrochim. Acta, 2015, 186, 50–57. 150 P. Wu, N. Du, H. Zhang, J. Yu, Y. Qi and D. Yang, Nanoscale, 2011, 3, 746–750. 151 M. S. Park, G. X. Wang, Y. M. Kang, D. Wexler, S. X. Dou and H. K. Liu, Angew. Chem., Int. Ed., 2007, 46, 750–753. 152 J. Luo, X. Xia, Y. Luo, C. Guan, J. Liu, X. Qi, C. F. Ng, T. Yu, H. Zhang and H. J. Fan, Adv. Energy Mater., 2013, 3, 737– 743. 153 W. Li, G. Li, J. Sun, R. Zou, K. Xu, Y. Sun, Z. Chen, J. Yang and J. Hu, Nanoscale, 2013, 5, 2901–2908. 154 X. Zheng, Y. Ye, Q. Yang, B. Geng and X. Zhang, Dalton Trans., 2016, 45, 572–578. 155 L. Li, F. He, S. Gai, S. Zhang, P. Gao, M. Zhang, Y. Chen and P. Yang, CrystEngComm, 2014, 16, 9873–9881.

This journal is © The Royal Society of Chemistry 2018

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Review

156 N. Padmanathan, S. Selladurai and K. M. Razeeb, RSC Adv., 2015, 5, 12700–12709. 157 L. Li, A. Kovalchuk, H. Fei, Z. Peng, Y. Li, N. D. Kim, C. Xiang, Y. Yang, G. Ruan and J. M. Tour, Adv. Energy Mater., 2015, 5, 1500171. 158 R. Madhuvilakku, S. Alagar, R. Mariappan and S. Piraman, Sens. Actuators, B, 2017, 253, 879–892. 159 M. Zhang, J. Sha, X. Miao, E. Liu, C. Shi, J. Li, C. He, Q. Li and N. Zhao, J. Alloys Compd., 2017, 696, 956–963. 160 S. Xie, M. Zhang, P. Liu, S. Wang, S. Liu, H. Feng, H. Zheng and F. Cheng, Mater. Res. Bull., 2017, 96, 413–418. 161 W. Yang, Z. Gao, J. Wang, B. Wang and L. Liu, Solid State Sci., 2013, 20, 46–53. 162 H. Quan, B. Cheng, Y. Xiao and S. Lei, Chem. Eng. J., 2016, 286, 165–173. 163 M. Aadil, W. Shaheen, M. F. Warsi, M. K. Shahid, A. Muhammad, Z. Ali, S. Haider and I. Shakir, J. Alloys Compd., 2016, 689, 648–654. 164 X. Qi, H. B. Zhang, J. Xu, X. Wu, D. Yang, J. Qu and Z. Z. Yu, ACS Appl. Mater. Interfaces, 2017, 9, 11025–11034. 165 Y. Yang, X. Fan, G. Casillas, Z. Peng, G. Ruan, G. Wang, M. J. Yacaman and J. M. Tour, ACS Nano, 2014, 8, 3939– 3946. 166 Y. Dong, R. Ma, M. Hu, H. Cheng, Q. Yang, Y. Y. Li and J. A. Zapien, Phys. Chem. Chem. Phys., 2013, 15, 7174–7181. 167 L. Zhao, M. Gao, W. Yue, Y. Jiang, Y. Wang, Y. Ren and F. Hu, ACS Appl. Mater. Interfaces, 2015, 7, 9709–9715. 168 S. Zhao, D. Xie, X. Yu, Q. Su, J. Zhang and G. Du, Mater. Lett., 2015, 142, 287–290. 169 Y. Gao, D. Wu, T. Wang, D. Jia, W. Xia, Y. Lv, Y. Cao, Y. Tan and P. Liu, Electrochim. Acta, 2016, 191, 275–283. 170 J. Ye, Q. Hao, B. Liu, Y. Li and C. Xu, Chem. Eng. J., 2017, 315, 115–123. 171 Z. Cao and B. Wei, J. Power Sources, 2013, 241, 330–340. 172 D. Xu, G. Luo, J. Yu, W. Chen, C. Zhang, D. Ouyang, Y. Fang and X. Yu, J. Alloys Compd., 2017, 702, 499–508. 173 C.-H. Xu, P.-Y. Shen, Y.-F. Chiu, P.-W. Yeh, C.-C. Chen, L.-C. Chen, C.-C. Hsu, I. C. Cheng and J.-Z. Chen, J. Alloys Compd., 2016, 676, 469–473. 174 H. Wang, X. Cheng, F. Yin, B. Chen, T. Fan and X. He, Electrochim. Acta, 2017, 232, 114–122. 175 T. Utschig, M. Schwarz, G. Miehe and E. Kroke, Carbon, 2004, 42, 823–828. 176 Z. Yang, D. Su, J. Yang and J. Wang, J. Power Sources, 2017, 363, 161–167. 177 Y. Ye, H. Zhang, Y. Chen, P. Deng, Z. Huang, L. Liu, Y. Qian, Y. Li and Q. Li, J. Alloys Compd., 2015, 639, 422–427. 178 S. Chou, J. Wang, D. Wexler, K. Konstantinov, C. Zhong, H. Liu and S. Dou, J. Mater. Chem., 2010, 20, 2041–2252. 179 M. Zou, L. Wang, J. Li, L. Guan and Z. Huang, Electrochim. Acta, 2017, 233, 85–91. 180 L. Liu, J. Wang, C. Wang and G. Wang, Appl. Surf. Sci., 2016, 390, 303–310. 181 W. Deng, S. Ci, H. Li and Z. Wen, Chem. Eng. J., 2017, 330, 995–1001. 182 Y. Liu, P. Li, Y. Wang, J. Liu, Y. Wang, J. Zhang, M. Wu and J. Qiu, J. Alloys Compd., 2017, 695, 2612–2618.

This journal is © The Royal Society of Chemistry 2018

Journal of Materials Chemistry A

183 X. Yi, W. He, X. Zhang, Y. Yue, G. Yang, Z. Wang, M. Zhou and L. Wang, Electrochim. Acta, 2017, 232, 550–560. 184 X. Qin, H. Zhang, J. Wu, X. Chu, Y.-B. He, C. Han, C. Miao, S. Wang, B. Li and F. Kang, Carbon, 2015, 87, 347–356. 185 Y. Yan, H. Tang, F. Wu, R. Wang and M. Pan, Energies, 2017, 10, 1296. 186 R. Li, X. Ren, F. Zhang, C. Du and J. Liu, Chem. Commun., 2012, 48, 5010–5012. 187 N. Du, Y. Chen, C. Zhai, H. Zhang and D. Yang, Nanoscale, 2013, 5, 4744. 188 W. Zhou, Y. Y. Tay, X. Jia, W. D. Y. Yau, J. Jiang, H. H. Hoon and T. Yu, Nanoscale, 2012, 4, 4459–4463. 189 R. Jin, Y. Guan, H. Liu, J. Zhou and G. Chen, ChemPlusChem, 2014, 79, 1643–1648. 190 Y. Li, Y. Hu, H. Jiang, X. Hou and C. Li, CrystEngComm, 2013, 15, 6715. 191 W. Xie, S. Li, S. Wang, S. Xue, Z. Liu, X. Jiang and D. He, ACS Appl. Mater. Interfaces, 2014, 6, 20334–20339. 192 G. Nie, X. Lu, M. Chi, Y. Zhu, Z. Yang, N. Song and C. Wang, Electrochim. Acta, 2017, 231, 36–43. 193 X. Zheng, Z. Han, F. Chai, F. Qu, H. Xia and X. Wu, Dalton Trans., 2016, 45, 12862–12870. 194 F. Yang, K. Xu and J. Hu, J. Alloys Compd., 2017, 729, 1172– 1176. 195 M. Armand and J. M. Tarascon, Nature, 2008, 451, 652–657. 196 Y. Ma, B. Ding, G. Ji and J. Lee, ACS Nano, 2013, 7, 10870– 10878. 197 L. Shen, C. Yuan, H. Luo, X. Zhang, K. Xu and F. Zhang, J. Mater. Chem., 2011, 21, 761–767. 198 L. Cheng, J. Yan, G.-N. Zhu, J.-Y. Luo, C.-X. Wang and Y.-Y. Xia, J. Mater. Chem., 2010, 20, 595–602. 199 M. Chen, W. Li, X. Shen and G. Diao, ACS Appl. Mater. Interfaces, 2014, 6, 4514–4523. 200 D. Xiang, L. Yin, C. Wang and L. Zhang, Energy, 2016, 106, 103–111. 201 T. Le Gall, K. H. Reiman, M. C. Grossel and J. R. Owen, J. Power Sources, 2003, 119–121, 316–320. 202 J. Huang, K. Wang and Z. Wei, J. Mater. Chem., 2010, 20, 1117–1121. 203 S. Wang, L. Hu, Y. Hu and S. Jiao, Mater. Chem. Phys., 2014, 146, 289–294. 204 F. Wu, J. Liu, L. Li, X. Zhang, R. Luo, Y. Ye and R. Chen, ACS Appl. Mater. Interfaces, 2016, 8, 23095–23104. 205 N. Casado, G. Hernandez, A. Veloso, S. Devaraj, D. Mecerreyes and M. Armand, ACS Macro Lett., 2016, 5, 59–64. 206 Y. Zeng, Y. Han, Y. Zhao, Y. Zeng, M. Yu, Y. Liu, H. Tang, Y. Tong and X. Lu, Adv. Energy Mater., 2015, 5, 1402176. 207 J. Liu, W. Zhou, L. Lai, H. Yang, S. Hua Lim, Y. Zhen, T. Yu, Z. Shen and J. Lin, Nano Energy, 2013, 2, 726–732. 208 W. Zhang, X. Hou, J. Shen, S. Hu, Q. Ru and K.-ho Lam, Electrochim. Acta, 2016, 188, 734–743. 209 L. Pan, X.-D. Zhu, X.-M. Xie and Y.-T. Liu, Adv. Funct. Mater., 2015, 25, 3341–3350. 210 Y. Zhang, K. Yue, H. Zhao, Y. Wu, L. Duan and K. L. Wang, Chem. Eng. J., 2016, 291, 238–243.

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9365

View Article Online

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

Journal of Materials Chemistry A

211 J. Guo, L. Chen, G. Wang, X. Zhang and F. Li, J. Power Sources, 2014, 246, 862–867. 212 G. Luo, Y. Lu, S. Zeng, S. Zhong, X. Yu, Y. Fang and L. Sun, Electrochim. Acta, 2015, 182, 715–722. 213 G. Wei, K. Du, X. Zhao, C. Li, J. Li, K. Ren, Y. Huang, H. Wang, S. Yao and C. An, J. Alloys Compd., 2017, 728, 631–639. 214 B. Wang, H. Wu, L. Yu, R. Xu, T. T. Lim and X. W. Lou, Adv. Mater., 2012, 24, 1111–1116. 215 L.-F. Chen, Zi-Y. Yu, J.-J. Wang, Q.-X. Li, Zi-Q. Tan, Y.-W. Zhu and S.-H. Yu, Nano Energy, 2015, 11, 119–128. 216 Y. X. Zhang, X. D. Hao, Z. P. Diao, J. Li and Y. M. Guan, Mater. Lett., 2014, 123, 229–234. 217 X. Xia, W. Lei, Q. Hao, W. Wang, Y. Sun and X. Wang, Electrochim. Acta, 2013, 113, 117–126. 218 Y. Wei, R. Ding, C. Zhang, B. Lv, Y. Wang, C. Chen, X. Wang, J. Xu, Y. Yang and Y. Li, J. Colloid Interface Sci., 2017, 504, 593–602. 219 J. Chen, J. Xu, S. Zhou, N. Zhao and C.-P. Wong, Nano Energy, 2016, 21, 145–153. 220 K. Jost, G. Dion and Y. Gogotsi, J. Mater. Chem. A, 2014, 2, 10776. 221 W. Zeng, L. Shu, Q. Li, S. Chen, F. Wang and X. M. Tao, Adv. Mater., 2014, 26, 5310–5336. 222 R. Liu, L. Ma, G. Niu, X. Li, E. Li, Y. Bai, Y. Liu and G. Yuan, Part. Part. Syst. Charact., 2017, 34, 1700213. 223 N. Li, C. Y. Zhi and H. Zhang, Electrochim. Acta, 2016, 220, 618–627. 224 T. A. Yersak, H. A. Macpherson, S. C. Kim, V.-D. Le, C. S. Kang, S.-B. Son, Y.-H. Kim, J. E. Trevey, K. H. Oh, C. Stoldt and Se-H. Lee, Adv. Energy Mater., 2013, 3, 120– 127. 225 L. Li, A. M. Caban, S. N. Girard and S. Jin, Nanoscale, 2014, 6, 2112–2118. 226 J. Liu, Y. Wen, Y. Wang, P. A. van Aken, J. Maier and Y. Yu, Adv. Mater., 2014, 26, 6025–6030. 227 P. Zhao, H. Cui, J. Luan, Z. Guo, Y. Zhou and H. Xue, Mater. Lett., 2017, 186, 62–65. 228 H. K. T. Takeuchi, K. Nakanishi, Y. Inada, M. Katayama, T. Ohta, H. Senoh, H. Sakaebe, T. Sakai, K. Tatsumi and H. Kobayashi, J. Electrochem. Soc., 2012, 159, 75–84. 229 D. Zhang, Y. J. Mai, J. Y. Xiang, X. H. Xia, Y. Q. Qiao and J. P. Tu, J. Power Sources, 2012, 217, 229–235. 230 F. Zhang, C. Wang, G. Huang, D. Yin and L. Wang, J. Power Sources, 2016, 328, 56–64. 231 T. S. Yoder, M. Tussing, J. E. Cloud and Y. Yang, J. Power Sources, 2015, 274, 685–692. 232 X. Xu, T. Cai, Z. Meng, H. Ying, Y. Xie, X. Zhu and W.-Q. Han, J. Power Sources, 2016, 331, 366–372. 233 V. Sridhar and H. Park, J. Alloys Compd., 2018, 732, 799–805. 234 Y. Gan, F. Xu, J. Luo, H. Yuan, C. Jin, L. Zhang, C. Fang, O. Sheng, H. Huang, Y. Xia, C. Liang, J. Zhang, W. Zhang and X. Tao, Electrochim. Acta, 2016, 209, 201–209. 235 L. Pei, Y. Yang, H. Chu, J. Shen and M. Ye, Ceram. Int., 2016, 42, 5053–5061. 236 K. Sivula, F. Le Formal and M. Gratzel, ChemSusChem, 2011, 4, 432–449.

9366 | J. Mater. Chem. A, 2018, 6, 9332–9367

Review

237 V. D. Nithya and N. S. Arul, J. Power Sources, 2016, 327, 297– 318. 238 S. Zheng, X. Li, B. Yan, Q. Hu, Y. Xu, X. Xiao, H. Xue and H. Pang, Adv. Energy Mater., 2017, 7, 1602733. 239 B. Y. Guan, X. Y. Yu, H. B. Wu and X. W. D. Lou, Adv. Mater., 2017, 29, 1703614. 240 J. Jiang, Y. Li, J. Liu, X. Huang, C. Yuan and X. W. Lou, Adv. Mater., 2012, 24, 5166–5180. 241 C. T. Fleaca, M. Scarisoreanu, I. Morjan, R. Alexandrescu, F. Dumitrache, C. Luculescu, I. P. Morjan, R. Birjega, A. M. Niculescu, G. Filoti, V. Kuncser, E. Vasile, V. Danciu and M. Popa, Appl. Surf. Sci., 2014, 302, 198–204. 242 N. Yabuuchi and S. Komaba, Sci. Technol. Adv. Mater., 2014, 15, 043501. 243 Y. L. Pang, S. Lim, H. C. Ong and W. T. Chong, Ceram. Int., 2016, 42, 9–34. 244 D. Fan, Y. Lan, P. G. Tratnyek, R. L. Johnson, J. Filip, D. M. O'Carroll, A. Nunez Garcia and A. Agrawal, Environ. Sci. Technol., 2017, 51, 13070–13085. 245 M. Mishra and D.-M. Chun, Appl. Catal., A, 2015, 498, 126– 141. 246 J. He, X. Yang, B. Men and D. Wang, J. Environ. Sci., 2016, 39, 97–109. 247 S. Haindl, S. Molatta, H. Hiramatsu and H. Hosono, J. Phys. D: Appl. Phys., 2016, 49, 345301. 248 P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J. M. Tarascon, Nat. Mater., 2011, 11, 19–29. 249 Y.-F. Mo, H.-T. Zhang and Y.-N. Guo, Mater. Lett., 2017, 205, 118–121. 250 W.-J. Yu, P.-X. Hou, F. Li and C. Liu, J. Mater. Chem., 2012, 22, 13756. 251 B. Jin, A.-H. Liu, G.-Y. Liu, Z.-Z. Yang, X.-B. Zhong, X.-Z. Ma, M. Yang and H.-Y. Wang, Electrochim. Acta, 2013, 90, 426– 432. 252 J. Luo, J. Liu, Z. Zeng, C. F. Ng, L. Ma, H. Zhang, J. Lin, Z. Shen and H. J. Fan, Nano Lett., 2013, 13, 6136–6143. 253 L. Wang, H. Ji, S. Wang, L. Kong, X. Jiang and G. Yang, Nanoscale, 2013, 5, 3793–3799. 254 J. Chen, X. Zhou, C. Mei, J. Xu, S. Zhou and C.-P. Wong, Electrochim. Acta, 2016, 222, 172–176. 255 K. Chen, X. Chen and D. Xue, CrystEngComm, 2015, 17, 1906–1910. 256 M.-S. Wu and R.-H. Lee, J. Electrochem. Soc., 2009, 156, A737. 257 K. Xie, J. Li, Y. Lai, W. Lu, Z. a. Zhang, Y. Liu, L. Zhou and H. Huang, Electrochem. Commun., 2011, 13, 657–660. 258 Y.-C. Chen, J.-H. Hsu, Y.-G. Lin and Y.-K. Hsu, J. Electroanal. Chem., 2017, 801, 65–71. 259 Q. Wang, L. Jiao, H. Du, Y. Wang and H. Yuan, J. Power Sources, 2014, 245, 101–106. 260 Y. Zhong, J. Liu, Z. Lu and H. Xia, Mater. Lett., 2016, 166, 223–226. 261 H. Fan, R. Niu, J. Duan, W. Liu and W. Shen, ACS Appl. Mater. Interfaces, 2016, 8, 19475–19483. 262 T. W. Lin, C. S. Dai and K. C. Hung, Sci. Rep., 2014, 4, 7274. 263 J. J. Yoo, K. Balakrishnan, J. Huang, V. Meunier, B. G. Sumpter, A. Srivastava, M. Conway, A. L. Reddy,

This journal is © The Royal Society of Chemistry 2018

View Article Online

Review

T. I. Kim, T. Song, K. Shigeta, S. Kang, C. Dagdeviren, I. Petrov, P. V. Braun, Y. Huang, U. Paik and J. A. Rogers, Nat. Commun., 2013, 4, 1543. 267 G. S. Gund, D. P. Dubal, N. R. Chodankar, J. Y. Cho, P. Gomez-Romero, C. Park and C. D. Lokhande, Sci. Rep., 2015, 5, 12454.

Published on 11 April 2018. Downloaded by Nanyang Technological University on 24/05/2018 02:49:05.

J. Yu, R. Vajtai and P. M. Ajayan, Nano Lett., 2011, 11, 1423– 1427. 264 H. Y. Jung, M. B. Karimi, M. G. Hahm, P. M. Ajayan and Y. J. Jung, Sci. Rep., 2012, 2, 773. 265 Y. Gogotsi, Nature, 2014, 509, 568–570. 266 S. Xu, Y. Zhang, J. Cho, J. Lee, X. Huang, L. Jia, J. A. Fan, Y. Su, J. Su, H. Zhang, H. Cheng, B. Lu, C. Yu, C. Chuang,

Journal of Materials Chemistry A

This journal is © The Royal Society of Chemistry 2018

J. Mater. Chem. A, 2018, 6, 9332–9367 | 9367