Recent progress in conversion reaction metal oxide anodes for Li-ion batteries

Abstract

Transition metal oxides (TMOs) based on conversion reactions are attractive candidate anode materials for lithium-ion batteries (LIBs) because of their high theoretical capacity and safety characteristics. In this review, we have summarized recent progress in the rational design and efficient synthesis of TMOs with controllable morphologies, compositions, and micro-/nanostructures, along with their Li storage behaviors. Single metal oxides of manganese (Mn), iron (Fe), cobalt (Co), nickel (Ni), copper (Cu), ruthenium (Ru), chromium (Cr), molybdenum (Mo), and tungsten (W) and their common binary metal oxides have been discussed in this review. Finally, the less well-known merits of conversion reactions are put forward, and the design of metal oxide electrodes making full use of these merits has been proposed.

---

Kangzhe Cao

Kangzhe Cao received his BSc from Henan Normal University (Henan, China) in 2012 and his PhD from Nankai University (Tianjin, China) in 2017 under the supervision of Assoc. Prof. Lifang Jiao and Prof. Yijing Wang. His research interests focus on the development of advanced materials for electrochemical energy storage (Li/Na-ion batteries and Li–S batteries).

Ting Jin

Ting Jin was born in Shaanxi, China, in 1994. She received her BSc degree in materials chemistry from Northwest University (China) in 2015. Currently, she is a PhD student in the group of Assoc. Prof. Lifang Jiao at Nankai University (China). Her research interests focus on the design and fabrication of high-performance electrode materials for lithium-ion and sodium-ion batteries.

Lifang Jiao

Lifang Jiao is an Associate Professor at Nankai University, China. She received her PhD degree from Nankai University (China) in 2005. She has co-authored over 180 relevant peer-reviewed publications. Her current research is focused on energy conversion and storage (including lithium, sodium, and magnesium secondary batteries and supercapacitors), hydrogen storage materials and electrocatalytic hydrogen evolution.

---

1. Introduction

The increasing demand for clean and sustainable energy has facilitated the development of energy storage systems (ESSs) with large capacities.¹ Without these ESSs, many green energy sources, such as solar, wind and waves, cannot be fully utilized due to their instability. Moreover, portability and high energy density of these systems are becoming increasingly important for their application in electric vehicles (EVs), hybrid electric vehicles (HEVs), and small portable devices. Li ion batteries (LIBs) are considered to be the most promising ESSs; they have been used to power the abovementioned applications. Since rocking-chair configurations were commercialized by Sony in 1991, LIBs have dominated the consumer electronic market due to their high energy density, lack of memory effects, long life span, and low self-discharge; currently, they remain popular. However, the limited capacity of current LIBs cannot meet the intense demand for high capacity. A significant reason for this situation is that the theoretical capacity of the electrode materials is not sufficiently high. For example, the theoretical capacity of the graphite currently used in commercial LIBs is only 372 mA h g⁻¹. Therefore, it is highly necessary to develop novel anode materials with large capacities.²

The theoretical capacity of a material used for LIBs is related to the number of transferred electrons in its electrochemical reactions and can be calculated by eqn (1):

where Q is the gravimetric theoretical capacity, F is the Faraday constant (96 500 C mol⁻¹), n is the number of transferred electrons in the electrochemical reaction, and M_w is the equivalent molecular weight. It is clear that the theoretical capacity of an active material is in inverse proportion to the M_w and is proportional to the electron transfer number. Essentially, active materials with multiple-electron reactions and low M_w can achieve high specific capacities.

Upon discharging and charging, Li ions are shuttled between the anode and cathode through the electrolyte, and the electrons are transferred through the external circuit. It is clear that the electrons and Li ions should reach the same active site in the electrode material at the same moment to complete the energy transformation between chemical energy and electrical energy.³ This process determines the power density of LIBs, which involves a kinetic problem in the electrode. Li ion diffusion in the electrode mainly consists of two parts: diffusion in the electrolyte and solid-state diffusion in the electrode material. The latter is the key step. As depicted in eqn (2),

τ_eq = L²/D (2)

the mean diffusion time (τ_eq) is proportional to the square of the diffusion length (L) and in inverse proportion to the diffusion coefficient (D).⁴ This means that there are two approaches to enhance the kinetics of an electrode: (1) increasing D by introducing foreign atoms into the electrode material; (2) decreasing the diffusion length by fabricating the electrode material on the nanoscale.

In terms of the reaction mechanism, LIBs anode materials can be mainly divided into three types. (1) Li insertion/deinsertion materials. These materials, which react with Li by insertion and deinsertion processes, usually deliver limited capacity; they include graphite (372 mA h g⁻¹) and Li₄Ti₅O₁₂ (175 mA h g⁻¹). (2) Li-alloying materials. These materials can form Li-alloy compounds after lithiation and always exhibit high specific capacity by transferring multi-electrons. For example, Si anode delivers a theoretical specific capacity of 4211 mA h g⁻¹, and P anode can form Li₃P and delivers a capacity of 2594 mA h g⁻¹. However, extreme volume expansion cannot be avoided (about 400% volume expansion for the formation of Li₂₂Si₅),⁵ and this can destroy the electrode structure. (3) Conversion reaction materials. Trascon's group discovered this new Li storage mechanism and firstly used CoO particles as a concept to illustrate it.⁶

As depicted in Fig. 1a and b, the original CoO particles were highly crystallized, with dimensions of about 100 to 200 nm. After lithiation, the CoO particles disintegrated into 1 to 2 nm nanoparticles and lost their crystallinity, although their initial shapes were preserved. This phenomenon suggests that a nanosize effect exists in this conversion reaction, as indicated by Fig. 1c and d. Moreover, the nanosized character of the electrode is preserved in the following charge, as demonstrated by Fig. 1e and f. These results confirmed that the formed Co nanoparticles were dispersed in a lithia (Li₂O) matrix during discharge and were then reoxidated to CoO nanoparticles during charge. This pioneering work opens new territory for materials that hold Li beyond the Li insertion/deinsertion and Li-alloying processes. Since then, numerous efforts have been devoted to this field, and many other conversion reaction materials, such as sulfide, have been found.⁷

Fig. 1 TEM images (a, c and e) and their corresponding selected-area electron diffraction (SAED) patterns (b, d and f) of the starting CoO electrode (a and b), the fully lithiated CoO electrode (c and d), and the delithiated CoO electrode (e and f). Reproduced with permission.⁶ Copyright 2000, Nature Publishing Group (NPG).

The conversion reaction occurs based on a displacement reaction that can be generalized as the following reaction (eqn (3)):

M_aX_b + (b·n)Li⁺ + (b·n)e⁻ ↔ aM + bLi_nX (3)

where M = Mn, Fe, Co, Ni, Cu, Ru, Cr, Mo, W, etc., X = H, N, O, F, P, S, etc., and n is the oxidation state of X. All these transition metal compounds are fully replaced by lithium to form metal nanoparticles, which are dispersed in the Li_nX matrix. Obviously, these anode materials always transfer multi-electrons and deliver remarkably high specific capacities. Taking Fe₂O₃ as an example, 6 electrons are transferred per mole of material in the electrochemical reaction, delivering a corresponding theoretical specific capacity of 1007 mA h g⁻¹, which is two times larger than that of graphite. Moreover, the lithiation potentials of these electrodes are between almost 0.5 and 1.0 V (in addition to fluorides, which react at high potentials and can be used as cathode materials⁸); thus, they are much safer for avoiding the formation of Li dendrites at the anodes. Therefore, these conversion reaction materials are interesting for use as alternative LIB anodes.

Among these conversion reaction materials, transition metal oxides (TMOs) feature high corrosion resistance and facile operation in addition to their high specific capacities. Furthermore, many materials, such as Mn oxides⁹ and Fe oxides,¹⁰ are more attractive due to their high natural abundance and low price.¹ However, these TMO electrodes suffer from drawbacks that hinder their applications.

1.1. The challenges and solutions of conversion reaction electrodes

Low electronic conductivity, volume expansion during cycling, and voltage hysteresis are the three main issues facing TMO materials. The low electronic conductivity limits the transfer of electrons and hinders improvement of the rate capability of the electrode. Volume expansion damages the structures of the active materials, resulting in decayed capacity and inferior cycling stability. Voltage hysteresis, which is observed between discharging and charging potential, leads to low energy efficiency. However, whether this hysteresis originates from kinetics or thermodynamics contributions is still in debate.^11–14 Currently, researchers are mainly concentrating on improving electrochemical performance, such as the obtained specific capacity, cycling stability, and rate capability. Three approaches are usually adopted with joint application methods.

1.1.1. Optimize the components of the anode materials. TMOs are mostly semiconductors (except for RuO₂). To fabricate pathways for electron transfer, they are always hybridized with some conductive materials, such as carbon-based materials^15–17 (amorphous carbon, carbon nanotubes, graphene, etc.) and conductive polymers¹⁸ (polypyrrole, polyaniline, etc.). These conductive materials not only enhance the conductivity of the electrode but also function as a matrix to relieve volume change stress during cycling. Additionally, an increasing number of studies focus on introducing other inorganic materials with electrochemical activity to the composites, enabling exploitation of the synergistic effects of the different components.^19,20

1.1.2. Optimize the hierarchical micro-/nanostructures of the anode materials. As mentioned above, fabricating the active material at the nanoscale is an effective method to enhance the kinetics of an electrode. Active materials with nano-architectures always possess large electrode/electrolyte contact areas, short path distances for electron transfer and Li⁺ diffusion, and improved reactivity. Unfortunately, many undesirable phenomena, such as agglomeration, occur on the prepared materials; also, electrochemical agglomeration during cycling has side effects on the electrochemical performance owing to the high surface energy of the nanoparticles. Moreover, volume expansion cannot be prevented during cycling. Hierarchical micro-/nano-structures are assembled from low-dimensional nano-building blocks, resulting in micro-structures with high activity and stability. Voids between the nano-building blocks and well-designed porosity of the hierarchical architectures can effectively buffer volume expansion and relieve the structural strain associated with repeated Li⁺ insertion/extraction. Therefore, hierarchical architecture anodes, such as porous structures, hollow structures, core–shell and yolk–shell structures, nanowires, nanotubes, and nanofibers, have been developed to obtain high electrochemical activity and long-term stability.^21–23

1.1.3. Optimize the fabrication method of the as-prepared electrodes. In the preparation of conventional electrodes, insulating polymer binders, such as sodium carboxymethyl cellulose (CMC) and polyvinylidenedifluoride (PVDF), are adopted to integrate individual electroactive particles with each other and also to ensure their adherence to the current collector.²⁴ However, the insulating properties and electrochemical inactivity of these binders can reduce the conductivity of the electrodes, impeding efforts to improve their electrochemical performance. In terms of the integrated electrodes, the active materials are directly grown on the current collector and adhered without binder. In this configuration, more active sites as well as larger areas between the electrode and electrolyte are exposed, which is beneficial to the electrochemical reaction processes. Therefore, active materials with various morphologies and structures on 2D/3D current collectors (Ti foil, Cu foil, stainless steel, Ni foam, 3D graphene, 3D carbon nanofibers, etc.) are being fabricated and used as LIB-integrated electrodes, achieving satisfactory Li storage properties.²⁵

In this review, an overview of recent progress towards conversion reaction TMO electrodes for LIBs is provided, including single metal oxides and binary metal oxides. Specifically, we will focus on the rational structural design and enhanced electrochemical performance of these electrodes. Future prospects in the design of advanced anodes and other functional materials by taking advantage of the merits of conversion reactions are highlighted. All potentials in this review are given versus Li⁺/Li unless otherwise specified.

2. Single metal oxides

2.1. Manganese oxides

Manganese oxides, including MnO, Mn₃O₄, Mn₂O₃, and MnO₂, feature high specific capacities, low electrochemical motivation forces (1.032 V vs. Li⁺/Li), environmentally benign natures, and low cost; they are being considered for use as alternative LIB anode materials.^26–28 The general consensus is that Mn metal nanoparticles are formed after lithiation of these oxides; thus, the theoretical specific capacities of these MnO_x anodes are 756, 937, 1018, and 1232 mA h g⁻¹, respectively. However, the oxidation production of these materials are controversial. In terms of MnO and Mn₃O₄ anodes, the Mn nanoparticles can be reoxidated to their original oxidation states with high reversibility. The Li storage behaviors of these anodes can be described as follows:

MnO + 2Li⁺ + 2e⁻ ↔ Mn + Li₂O (4)

Mn₃O₄ + 8Li⁺ + 8e⁻ ↔ 3Mn + 4Li₂O (5)

Meanwhile, the reoxidized product of Mn₂O₃ anodes is in debate. One suggestion is that MnO is the oxidation product of Mn₂O₃ anodes at 3.0 V, as described in the following equations:^29,30

3Mn₂O₃ + 2Li⁺ + 2e⁻ → 2Mn₃O₄ + Li₂O (6)

Mn₃O₄ + Li⁺ + e⁻ → LiMn₃O₄ (7)

LiMn₃O₄ + Li⁺ + e⁻ → 3MnO + Li₂O (8)

MnO + 2Li⁺ + 2e⁻ ↔ Li₂O + Mn (9)

Another view is that Li⁺ storage behavior in Mn₂O₃ nanoplate anodes proceeds as follows:^31,32

2Li⁺ + 3Mn₂O₃ + 2e⁻ → 2Mn₃O₄ + Li₂O (10)

2Li⁺ + Mn₃O₄ + 2e⁻ → 3MnO + Li₂O (11)

2Li⁺ + MnO + 2e⁻ ↔ Li₂O + Mn (12)

xLi₂O + Mn ↔ 2xLi⁺ + MnO_x + 2xe⁻ (1.0 < x < 1.5) (13)

MnO₂ is known to have many polymorphs; they are often studied as LIB cathodes, catalysts, biosensors, and supercapacitors.^33–36 When used as LIB anodes, the oxidation product is vague, although it has been the subject of many studies.^37–39 Therefore, further research to detect Li⁺ storage behavior in Mn₂O₃ and MnO₂ anodes is urgent and meaningful.

Due to the limitation of low solid Li ion diffusion rates, Mn oxides with various morphologies and micro-/nano-structures have been rationally designed to facilitate ion diffusion in order to achieve satisfactory electrochemical performance. Porous structure electrodes, such as MnO nanowires,⁴⁰ Mn₂O₃ nanowires,⁴¹ Mn₂O₃ nanoplates,^31,42 Mn₂O₃ microspheres,⁴³ Mn₂O₃ derived from Mn-based MOFs,⁴⁴ Mn₂O₃ hollow boxes,⁴⁵ Mn₃O₄ nanotubes,⁴⁶ and hierarchical MnO₂/C hybrid spheres,³⁸ have been designed and fabricated as LIB anodes. By virtue of their numerous different-sized voids, which facilitate electrolyte infiltration and Li⁺ ion diffusion, these electrodes exhibit high electrochemical performance. For example, a porous Mn₂O₃ nanoplate anode exhibits a capacity of 813.7 mA h g⁻¹ after 50 cycles at a current density of 0.1 A g⁻¹ as reported by Zhang et al.³¹ A porous Mn₂O₃ anode using a MOF precursor delivers a reversible capacity of 705 mA h g⁻¹ at 1.0 A g⁻¹ even after 250 cycles.⁴⁴ Fabricating hollow structures is considered to be another effective way to alleviate structural strain on the electrode during cycling. Single shell hollow spheres,^47,48 multi shell hollow spheres,⁴⁹ and hollow porous MnO/C microspheres⁵⁰ have been synthesized and used as high-performance LIB anodes. Selected typical morphologies of MnO_x are listed in Fig. 2.

Fig. 2 SEM images of (a) Mn₂O₃ NWs, (b) Mn₂O₃ nanoplates, and (c) Mn₂O₃ microspheres; TEM images of (d) Mn₂O₃ nanoplates, (e) Mn₂O₃ porous hollow boxes, and (f) hierarchical MnO₂/C hybrid spheres. (a–f) are reproduced with permission. (a),⁴¹ Copyright 2014, ACS; (b),⁴² Copyright 2014, RSC; (c),⁴³ Copyright 2013, RSC; (d),³¹ Copyright 2014, Wiley; (e),⁴⁵ Copyright 2015, RSC; (f),³⁸ Copyright 2016, Wiley.

Recently, our group fabricated mini-hollow polyhedron Mn₂O₃ using Mn-based MOFs as precursors.²⁶ This mini-hollow structure differs from conventional hollow structures with large interiors, which endows the as-prepared polyhedron Mn₂O₃ with high performance in LIB anodes. As shown in Fig. 3a, the high-angle annular dark-field scanning transmission electron microscope (HAADF-STEM) image suggests that the hollow structure contains a mini-hollow cavity and thick shell, in contrast to familiar hollow structures with large interior hollow cavities and thin shells.^{47,48,51–53} The TEM image in Fig. 3b suggests that the mini-hollow cavity of the original polyhedron Mn₂O₃ disappears and the smooth shell becomes rough after cycling. Fig. 3c confirms that a reformation process has occurred in the mini-hollow polyhedron Mn₂O₃, and a hierarchical nanostructure consisting of reformatted nanoparticles is formed. The cycling performance in Fig. 3d displays that the mini-hollow Mn₂O₃ anode achieves super-long cycling stability and high capacity (a capacity of 819.8 mA h g⁻¹ is obtained after 1200 cycles at 1.0 A g⁻¹), while the bulk anode shows limited cycling performance (160 mA h g⁻¹ is retained after 250 cycles). The main reason for the outstanding electrochemical properties demonstrated by the mini-hollow Mn₂O₃ electrode is that the unique mini-hollow structure can utilize the merits induced by the nanosize effect. As typical features of conversion reactions, nanosize effects and volume expansion cannot be avoided. Fortunately, the small interior cavity offers space for inward volume expansion that is filled by the reformed nanoparticles, constructing a hierarchical nanostructure with a homogeneous dispersion of nanoparticles, as illustrated in Fig. 3e. Therefore, the reconstructed nanostructures possess the merits of short Li diffusion distance, more active sites, and stable structure, showing satisfactory electrochemical performance without any serious “electrochemical sintering” after long cycling. The size of the hollow cavity may not be optimal; however, as the authors point out, the concept of a mini-hollow structure can offer a new concept in designing rational structures to exploit the merits of nanosize effects.

Fig. 3 (a) HAADF-STEM and TEM images (inset) of the original mini-hollow Mn₂O₃; TEM images (b and c) and SEM image (inset in b) of mini-hollow Mn₂O₃ after the first cycle, (d) cycling performance of mini-hollow Mn₂O₃ and bulk Mn₂O₃ at 1.0 A g⁻¹, (e) schematic of the structure evolution of mini-hollow Mn₂O₃ electrodes with cycling. Reproduced with permission.²⁶ Copyright 2016, Wiley.

Mai and co-workers designed a manganese oxide/carbon yolk–shell structure to enhance the electronic conductivity of the composite and alleviate the effects of volume expansion of the active materials during cycling.⁵⁴ The hollow space around the manganese oxide allows it to expand without damaging the overall morphology during cycling, while the carbon shell is beneficial to fast electron transfer and the stability of the SEI layers. Thus, their yolk–shell anode exhibits a capacity of 634 mA h g⁻¹ after 900 cycles at 0.5 A g⁻¹. Jiang et al. reported a peapod-like MnO/C heterostructure anode.⁵⁵ The nano-peapod structure has an internal void space, which is sufficient to restrain volume expansion without cracking the carbon layer, hence preserving the structural integrity. Therefore, the peapod-like MnO/C anode exhibits outstanding cycling stability and rate capability (no capacity fading after cycling 1000 times at 2.0 A g⁻¹, and a capacity of 463 mA h g⁻¹ at 5.0 A g⁻¹). Additionally, MnO_x–C hybrid materials, in which MnO_x nanoparticles are embedded in porous carbon,^56–58 carbon nanosheets,⁵⁹ and carbon nanofibers,⁶⁰ have been fabricated and used as high performance LIB anodes. Interestingly, most of these electrodes show increasing capacities as cycling proceeds. Several factors account for this phenomenon, including the increased number of active sites introduced by nanosize effects, higher oxidation state products, interfacial Li storage (pseudocapacitance), and the growth of an electrochemical gel-like polymer layer.^26,54,56 Interestingly, the factor of increasing capacity could be adopted to design high performance electrodes. A new rGO–MnO–rGO sandwich nanostructure anode was fabricated by Yuan et al. and achieved outstanding rate capability (379 mA h g⁻¹ after 4000 cycles at 15 A g⁻¹, 331.9 mA h g⁻¹ at 40 A g⁻¹) with unprecedented cycle stability owing to the synergistic effect of the surface pseudocapacitance and diffusion-controlled lithium storage.⁶¹

MnO_x anodes not only show excellent electrochemical performance in half cells but also exhibit remarkable performance when applied in Li-ion full cells.^41,62,63 Wang et al. reported Mn-based flexible LIB full cells using an Mn₂O₃ anode and a LiMn₂O₄ cathode which were synthesized using MnOOH nanowires grown on Ti foil as the precursor.⁴¹ The Mn₂O₃ anode delivers a capacity of 502.3 mA h g⁻¹ after 100 cycles, while the LiMn₂O₄ cathode maintains a reversible capacity of 94.7 mA h g⁻¹ at 0.1 A g⁻¹ after 100 cycles in half cells. When a LIB full cell is fabricated using these two electrodes, a specific capacity of 99 mA h g⁻¹ is obtained based on the mass of the cathode material. The output voltage of the full cell can turn on a 3 V light emitting diode (LED, 10 mW), showing its potential practical applications. Wei's group fabricated a MnO@C‖LiMn₂O₄ full cell using a MnO@C core–shell nanowires anode.⁶² As depicted in Fig. 4a, the MnO cores are coated with uniform carbon shells, and many internal void spaces are retained in the hybrid MnO@C nanowires. It is believed that the uniform carbon shells enhance the conductivity and the internal void spaces along the 1D configuration can accommodate volumetric expansion of MnO during lithiation. The extraordinary cycling performance of the MnO@C core–shell nanowires anode is illustrated in Fig. 4b. Obviously, little capacity decay can be seen at a current density of 1.0 A g⁻¹ for 200 cycles. The full cell assembled from the as-prepared MnO/C anode and LiMn₂O₄ nanoparticle cathode is illustrated in Fig. 4c. The electrochemical reactions on the two electrodes at charge and discharge are believed to proceed as follows:

(x/2)MnO/C + LiMn₂O₄ ↔ (x/2)Li₂O + (x/2)Mn/C + Li_1−xMn₂O₄ (14)

Fig. 4 (a) TEM image and (b) cycling performance at 1.0 A g⁻¹ of MnO@C nanowires. (c) Schematic and (d) charge/discharge curves of the MnO@C‖LiMn₂O₄ full cell. (e) Cycling performance of the full cell at 0.2 A g⁻¹. The insets of (e) show the lighting of two red LEDs by the full cell. Reproduced with permission.⁶² Copyright 2016, Elsevier.

As shown in Fig. 4d, the output average voltage of the full cell is about 3.6 V, and a discharge capacity of 104 mA h g⁻¹ at 1.4 C (1 C = 148 mA h g⁻¹) is delivered. More importantly, the capacity retains 87% of its maximum value after 1000 cycles (Fig. 4e), indicating the outstanding cycling stability of the full cell. Encouragingly, the rate capability is also excellent. When the full cell is cycled at high rates of 6.8 C and 13.5 C, the capacity can still reach 72.7% and 63.6% of the capacity at 0.17 C. More importantly, the capacity of the full cell can be completely recovered if the current density is returned to 0.17 C. This excellent rate capability may have a positive effect on the practical application of Mn oxides. A flexible rGO/Mn₃O₄‖LiMn₂O₄ full battery was also assembled by this group.⁶³ They prepared long Mn₃O₄ nanowires and hybridized them with rGO using a facile vacuum filtration method. The strong interaction between the long Mn₃O₄ nanowires and large-area rGO not only enables the as-prepared rGO/Mn₃O₄ membrane to endure various mechanical deformations, such as bending, twisting, and even folding multiple times, but also offers a strong synergistic effect of enhanced electrochemical reaction kinetics by reducing electron/ion transport resistance and providing an enlarged electrode/electrolyte contact area. When coupled with a LiMn₂O₄/Al foil cathode, the assembled full cell can power a red light-emitting-diode (LED) in both flat and bended states. Additionally, the full cell exhibits good cycling performance, with a specific capacity of 79 mA h g⁻¹ being retained after 100 cycles.

2.2. Iron oxides

Iron oxides, including FeO, Fe₂O₃, and Fe₃O₄, have attracted particular attention for their abundance, non-toxicity, low processing cost, and high capacities when used as LIB anode materials. They achieve high specific capacities of 745, 1007, and 928 mA h g⁻¹ by delivering 2, 6, and 8 mol electrons per mole of material, respectively. Due to these merits, iron oxides are promising LIB anode materials; Lou's group summarized them in 2013.¹⁰ Recently, numerous papers on the material synthesis and electrochemical activity of FeO_x anodes have been presented.

In order to inhibit the volume expansion of FeO_x, various structures have been constructed, such as Fe₂O₃ nano-assembled spindles,⁶⁵ porous Fe₂O₃ nanocubes,⁶⁶ 3D mesostructured Fe₂O₃,¹³ Fe₂O₃ hollow microcubes,⁶⁷ multi-shell Fe₂O₃ hollow spheres,⁵³ self-assembled Fe₃O₄ nanoparticle clusters,⁶⁸ and hierarchical Fe₃O₄ hollow microspheres.^51,52 With this designed internal porosity to inhibit volume expansion, these electrodes always show enhanced cycling stability at low current densities. Chen et al. synthesized Fe₂O₃ nanotubes and nanorods and compared their lithium-storage performances.⁶⁹ They found that the electrochemical performance of the nanotube anode was much better than that of the nanorod electrode due to the tubular structure, which could tolerate enormous volume changes during cycling and shorten the pathways of Li ion transportation.

Wang's group proposed a simple method to synthesize hierarchical multi-shelled hollow microspheres.⁵³ Carbonaceous microspheres were chosen as the template, and ethanol was adopted to adjust the Fe³⁺ concentration within the template, as illustrated in Fig. 5a. When the carbonaceous microspheres soaked with Fe³⁺ were annealed in air, “thick and thin” multi-shelled hollow microspheres are produced, such as the thin double-shell and triple-shell Fe₂O₃ hollow microspheres in Fig. 5b and c. Due to the merits of thin multi-shell structures, such as high surface areas with sufficient active sites, short diffusion distances for Li ions, and sufficient gaps for buffering mechanical stresses, the thin triple-shelled Fe₂O₃ hollow microspheres anode exhibited an ultrahigh capacity of 1702 mA h g⁻¹ at 50 mA g⁻¹.

Fig. 5 (a) Illustration showing control of the number of multi-shells by adjusting the Fe³⁺ concentration in carbonaceous microsphere templates. TEM images of thin (b) double-shell and (c) triple-shell Fe₂O₃ hollow microspheres. Reproduced with permission.⁵³ Copyright 2014, RSC. (d) TEM image of Fe₂O₃–C composite nanofibers, (e) cycling performance and (f) EIS of Fe₂O₃–C composite nanofibers and pure Fe₂O₃ nanofiber electrodes. Reproduced with permission.⁶⁴ Copyright 2014, RSC. (g) TEM and (h) HR-TEM images of Fe₃O₄@PPy nanocomposite and its rate capability (i). Reproduced with permission.¹⁸ Copyright 2016, Wiley.

When these electrodes are coated with conductive materials, such as carbon and polypyrrole, the conductivity of the electrodes can be enhanced and their rate capabilities can also be improved.^{18,64,70–73} Lei et al. reported the concept of “confined nanospace pyrolysis” to fabricate coaxial Fe₃O₄@C hollow particles.⁷⁰ The obtained coaxial Fe₃O₄@C hollow particles delivered a high reversible capacity of 864 mA h g⁻¹ after 50 cycles, while hollow Fe₃O₄ (H-Fe₃O₄) showed a severely faded capacity of 738.5 mA h g⁻¹ under the same conditions. Moreover, the coaxial Fe₃O₄@C hollow particles electrode displayed outstanding rate capability compared to its counterparts, especially at high current density (above 1.5 A g⁻¹). Fan's group scattered Fe₂O₃ nanoparticles in carbon nanofibers by an electrospinning method.⁶⁴ As presented in Fig. 5d, the Fe₂O₃ nanoparticles are uniformly distributed in the nanofibers.

The capacity of the obtained Fe₂O₃–C composite nanofibers electrode was 820 mA h g⁻¹ after 100 cycles, while that of the pure Fe₂O₃ nanofiber electrode faded rapidly to 482 mA h g⁻¹ (Fig. 5e). As evidenced by electrochemical impedance spectroscopy (EIS, Fig. 5f), the charge-transfer resistance (R_ct) of the Fe₂O₃–C composite nanofibers electrode is much lower than that of the pure Fe₂O₃ nanofiber electrode, suggesting the importance of the introduction of carbon into the 1D composite nanofibers. Liu et al. constructed a hierarchical Fe₃O₄@polypyrrole (PPy) nano-cages LIB anode using a template-assisted interfacial reaction followed by in situ polymerization and reduction.¹⁸ The TEM images in Fig. 5g and h confirm that the 2D nanosheets are composed of crystalline Fe₃O₄ and an amorphous PPy component. The obtained electrode demonstrates a high value of 950 mA h g⁻¹ after 100 cycles at 0.2 A g⁻¹ and about 652 mA h g⁻¹ after 500 cycles at 2.0 A g⁻¹ (Fig. 5i). More importantly, this electrode still delivers a reversible capacity of about 490 mA h g⁻¹ at 5.0 A g⁻¹. These studies suggest that cooperation of the conducting material is an effective approach to achieve high rate capability by improving the electron transport in electrodes.

Recently, Paik's group reported an etching-in-a-box strategy to fabricate unique Fe₃O₄@carbon (Fe₃O₄@C) yolk–shelled nanocubes.⁷⁴ This structure not only has the merits of the yolk–shelled hollow structure but also adopts conductive carbon materials as buffer layers, endowing the as-prepared samples with high specific capacity, excellent rate capacity and ultralong cycling stability. They demonstrated that the Fe₂O₃@PDA core–shelled nanocubes were obtained by coating Fe₂O₃ with polydopamine (PDA) and were then converted to Fe₃O₄@C core–shelled nanocubes (Fe₃O₄@C-0, Fig. 6a) after being annealed. When these as-prepared Fe₃O₄@C-0 nanocubes were etched with HCl for different times, the Fe₃O₄ cores were partially etched and a series of yolk–shelled nanocubes with different void space sizes were obtained (Fe₃O₄@C-1, Fe₃O₄@C-2, and Fe₃O₄@C-3 were obtained with etching times of 1, 2, and 3 h, respectively). As depicted in the TEM images in Fig. 6b–d, the sizes of the inner cavities in the Fe₃O₄@C samples greatly increase as the etching proceeds. The electrochemical performances of these electrodes, shown in Fig. 6e, indicate that an optimized void space is vital for enhanced lithium storage performance. The Fe₃O₄@C-2 boxes anode could deliver a capacity of 470 mA h g⁻¹ after 8000 cycles at a rate of 10 C (1 C = 1.006 A g⁻¹) with an average coulombic efficiency of around 98% (Fig. 6f). This work suggests that an optimized void space in the electrode material for containing the volume expansion is vital to achieve long cycling stability.

Fig. 6 (a–d) TEM images of Fe₃O₄@C yolk–shelled nanocubes with different etching times (a, 0 h; b, 1 h; c, 2 h; d, 3 h); cycling performance of the four obtained samples at 0.5 A g⁻¹ (e) and the Fe₃O₄@C-2 yolk–shelled nanocubes at 10 A g⁻¹ (f). Reproduced with permission.⁷⁴ Copyright 2016, Wiley.

Carbon nanotubes (CNTs) are another important material used in conductive networks for electron transfer.^75–78 Jia et al. reported an aerosol spray drying process to construct CNT/Fe₃O₄ nanocomposites.⁷⁵ They dispersed as-prepared Fe₃O₄, CNTs, and sucrose into aqueous solution with the aid of surfactant and used them to generate aerosol droplets. After the annealing process, CNT/Fe₃O₄ nanocomposites were formed by these aerosol droplets in which CNTs were threaded through the particles and functioned as a conductive network. Thus, excellent rate capability was achieved by this hybrid electrode. Sun et al. prepared carbon-coated FeO_x/CNT composites by controlled pyrolysis of ferrocene. The composite electrode achieved an outstanding cycling stability (specific charge capacity retention of up to 84% after 2000 cycles under 2.0 A g⁻¹).⁷⁶ Cheng's group confined Fe₂O₃ nanoparticles inside carbon nanotubes and observed the structural evolution during lithium insertion/extraction by in situ TEM.⁷⁸ Fe₂O₃ particles with a mean diameter of about 10 nm filling the CNTs were obtained by a wet chemistry technique, and a Fe₂O₃ nanoparticle-coated CNT sample was also prepared. Taking advantage of in situ TEM, they observed that the nanograins formed inside the CNTs during the electrochemical process were much smaller than those coated on the outside. The cycling performance achieved by the Fe₂O₃ nanoparticle-filled CNTs was superior to that of Fe₂O₃ nanoparticle-coated CNTs after aggressive cycling at 7.5 A g⁻¹, suggesting a more stable structure of the former electrode. Moreover, the reasons for the high reversible capacity of the electrodes were discussed in detail. In addition to a capacity of 1007 mA h g⁻¹ contributed by the conversion reaction, enhanced interfacial lithium storage, the reversible reaction of LiOH to form LiH, and partial reversible formation and decomposition of the SEI layer were the vital factors contributing to the high reversible capacity.

In addition to constructing hybrid materials with carbon-based materials, other electrochemically active materials, such as Sn,⁷⁹ MoS₂,^80,81 SnO₂,^82,83 TiO₂,⁸⁴ Co₃O₄,¹⁹ ZnFe₂O₄,⁸⁵ CuFe₂O₄,⁸⁶ and NiCo₂O₄,⁸⁷ are employed to improve the performance of iron oxide anodes. An advantage of these configurations is that there is a synergistic effect between the active components. Zhu et al. reported a three-dimensional MoS₂@Fe₃O₄ nanohybrid.⁸⁰ In this structure, Fe₃O₄ particles were uniformly and densely decorated on tubular MoS₂. When the MoS₂@Fe₃O₄ was used as an LIB anode, the tubular MoS₂ inside it could function as a matrix, offering a hollow interior and mesopores to facilitate Li ion transfer and location and buffer the structure strain during cycling, while the Fe₃O₄ particulates could serve as a gasket to prevent aggregation of the tubular MoS₂. Hence, an encouragingly high reversible capacity of 1113 mA h g⁻¹ was obtained by this electrode, which is much higher than that of the tubular MoS₂ anode. Li₄Ti₅O₁₂ is known for its zero-strain behavior during Li⁺ intercalation–deintercalation and always shows long cycling stability.⁸⁸ Kang's group encapsulated Fe₃O₄ nanoparticles in Li₄Ti₅O₁₂ nanofibers (NFs) using an electrospinning process combined with subsequent controlled heat treatment to prepare an Fe₃O₄/Li₄Ti₅O₁₂ composite. Due to the combination of the stability of Li₄Ti₅O₁₂ and the high capacity of Fe₃O₄, the obtained C@Fe–Fe₃O₄/Li₄Ti₅O₁₂ hybrid NFs anode exhibited high capacity and long cycle life.²⁰ Rahman et al. reported a breathable structure constructed from clusters of Fe₂O₃ and SnO₂ nanoparticles which were dispersed along carbon black conductive chains.⁸³ The lithiation reactions of these two different materials proceeded at different voltages, which provides breathable aggregation and enables sequential expansion and contraction of the electrode, relieving the problems associated with volume changes. However, agglomeration of the electrode could not be avoided after long cycling because the dimensions of the nanoparticles are not sufficiently small and the mixture is not homogeneous; thus, the electrode exhibited capacity fade after long cycling. Therefore, the mixture electrode showed limited improvement in electrochemical performance compared to single-oxide electrodes.

The relationship between structure and function in hybrid electrodes was studied by Sultana et al.¹⁹ Taking a Co₃O₄−Fe₂O₃/C system as an example, they demonstrated that the cycling performance and rate capability were improved in the hybrid electrode compared to Co₃O₄−Fe₂O₃, Fe₂O₃/C, and Co₃O₄/C control electrodes. The results suggested that better management of stress resulting from the sequential reactivity of the components and the enhanced electronic conductivity provided by carbon chains were responsible for the superior electrochemical performance. Jiang et al. reduced the dimensions of the components to 2 to 10 nm and distributed them alternately by pulsed spray evaporation chemical vapor deposition (PSE-CVD).⁸² In their obtained SnO₂–Fe₂O₃–Li₂O film, the 2 to 10 nm nanoparticles with special interfacial relationships had a highly homogeneous distribution. During cycling, the metal oxide nanoparticles were largely locked at their original sites without atom migration. Thus, the structural integrity of the electrode could be well retained, which is critical for stable cycling performance. As a result, the SnO₂–Fe₂O₃–Li₂O nanocomposite anode exhibited high volumetric capacity and excellent rate capability. Impressively, a high volumetric capacity of 4704 mA h cm⁻³ (940.8 mA h g⁻¹) was obtained even when cycled at a current density of 20 A g⁻¹. These studies indicate that conversion reaction electrodes with high electrochemical performance could be fabricated by minimizing the particle size and distributing the composites in a highly homogeneous state.

Fabricating binder-free (integrated) electrodes is another method to improve electrochemical performance, as discussed above.^89–96 Luo et al. reported bicontinuous mesoporous Fe₃O₄ nanostructures on 3D graphene foams and used them as binder-free electrodes.⁹⁴ They firstly grew graphene on Ni foam by chemical vapor deposition (CVD) and removed the Ni template by etching. Secondly, the remaining 3D graphene foam was coated with a ZnO layer by atomic layer deposition (ALD). Subsequently, the prepared graphene foam was immersed in FeCl₃ solution mixed with glucose and then annealed to form bi-continuous mesoporous Fe₃O₄ nanostructures on graphene foam (GF@Fe₃O₄). The loading of Fe₃O₄ could be controlled by adjusting the ZnO layers. This obtained GF@Fe₃O₄ anode delivered a capacity of 190 mA h g⁻¹ at 60 C (1 C = 1007 mA g⁻¹), demonstrating its excellent rate capability. When their electrode was cycled at rates of 6 C and 10 C up to 500 cycles, the capacity remained at about 400 and 300 mA h g⁻¹, respectively. This outstanding electrochemical performance was ascribed to the well-designed electrode structure. The in situ fabrication method enabled uniform dispersion of the bi-continuous Fe₃O₄ nanocrystallites on the highly electrically conductive GF, ensuring efficient ion and electron transport and avoiding the aggregation of Fe₃O₄ particles. Huang et al. designed ordered hierarchically porous 3D electrodes with an entrapped active nanoparticle configuration.⁹⁵ They tailored the pore size of periodic porous carbon anchored on Ni foam and entrapped Fe₂O₃ inside the periodic porous carbon. This configuration provided low-resistant avenues for Li ion diffusion and electron transfer and prevented the active nanoparticles from aggregating. Therefore, ultrahigh rate capability and long-term cycling stability were achieved by this binder free electrode (a capacity of 644.71 mA h g⁻¹ after 1300 cycles at 4.0 A g⁻¹). Our group deposited 3D hierarchical Fe₂O₃ nanosheets on copper foil and evaluated it as a binder-free LIB anode.⁹⁶ The voids between the nanoparticles which composed the nanosheets and the intervals between the nanosheets offered avenues for Li⁺ diffusion and provided room for volume changes. Thus, the obtained binder-free electrode delivered long cycling stability and excellent rate capability (a capacity of 817 mA h g⁻¹ was attained at 2.01 A g⁻¹ after 1000 cycles, and a capacity of 433.2 mA h g⁻¹ was obtained at 20.1 A g⁻¹).

2.3. Cobalt oxides

The main phases of cobalt oxides used as LIB anodes are CoO and Co₃O₄. They react with Li by transferring 2 and 8 electrons per mole of material and deliver theoretical specific capacities of 715 and 890 mA h g⁻¹, respectively. In contrast to the well-accepted mechanism of the reaction of CoO with Li,^97–99

CoO + 2Li⁺ + 2e⁻ ↔ Co + Li₂O (15)

there are two different views on the reaction of Co₃O₄ with Li. One view insists that the reaction is highly reversible and that the reaction proceeds as follows:^100–104

Co₃O₄ + 8Li⁺ + 8e⁻ ↔ 3Co + 4Li₂O (16)

Meanwhile, the other view is that this reaction is not fully reversible and that the lithiation product Co nanograins should be reoxidated to CoO, not Co₃O₄.^105,106 Su et al. directly observed the dynamic structure changes of porous Co₃O₄ nanoplates/graphene in LIBs by in TEM.¹⁰⁶ They found that the highly crystalline Co₃O₄ nanoplates transformed into numerous Co nanograins and Li₂O during the first lithiation (eqn (17)). The following electrochemical reaction was a reversible conversion between Co nanograins and CoO nanograins (eqn (18)), not conversion between Co and Co₃O₄. The irreversible phase conversion of Co₃O₄ in the first cycle is responsible for the initial capacity fading. The whole reaction can be expressed as follows:

Co₃O₄ + 8Li⁺ + 8e⁻ → 3Co + 4Li₂O (first lithiation) (17)

Co + Li₂O ↔ CoO + 2Li⁺ + 2e⁻ (18)

In addition to the debate over the reaction mechanisms, volume changes and low conductivity are also drawbacks of these materials. Constructing hollow structures or multi-shelled hollow nanostructures organized by secondary subunits is considered to be an effective method to release the strain caused by volume changes.^21,107–110 Wang's group controlled the shell numbers of Co₃O₄ hollow microspheres accurately by adjusting the size and diffusion rate of the hydrated Co²⁺ as well as the absorption capability of the carbonaceous microspheres (CMSs).¹⁰⁷ As shown in Fig. 7a, single-shelled Co₃O₄ microspheres were finally obtained when water alone was used as the solvent. When the amount of ethanol in the aqueous solution was increased, the size of the hydrated Co²⁺ in the solution decreased owing to the lower coordination of H₂O and the increased diffusion rate of hydrated Co²⁺ in the CMS templates. As a result, double-shelled Co₃O₄ microspheres were formed when the CMCs were removed (Fig. 7b and e). When the diffusion rate of hydrated Co²⁺ was further accelerated by increasing the amount of ethanol up to 75% by volume and heating the solutions at higher temperatures, triple-shelled Co₃O₄ was obtained (Fig. 7c and f). Additionally, enlargement of the surface area and pore volume of the CMCs template by HCl treatment could enable more Co²⁺ to penetrate, achieving the formation of quadruple-shelled Co₃O₄ (Fig. 7d and g). In virtue of the multishell hollow structure, the prepared Co₃O₄ hollow microspheres delivered higher capacity than commercial Co₃O₄; the triple-shelled Co₃O₄ microspheres showed the highest specific capacity and best cycling performance, as shown in Fig. 7h.

Fig. 7 Mechanism for the controlled formation of multishelled Co₃O₄ hollow microspheres (a–d); (e–g) TEM images and (h) cycling performance of double-, triple-, and quadruple-shelled Co₃O₄ hollow microspheres. Reproduced with permission.¹⁰⁷ Copyright 2014, Wiley.

A 3D interconnected carbon-based network is an ideal structure to enhance the conductivity of a hybrid material and release the stress of volume changes.^97,111–114 Yao et al. developed a one-step multipurpose strategy to hybridize flexible graphene nanosheets with Co₃O₄ nanowires.¹¹¹ The resulting mesoporous Co₃O₄ nanowires confined by N-doped graphene aerogel anode feature enhanced electrochemical properties, delivering a high capacity of more than 1200 mA h g⁻¹ after 200 cycles. Self-adhesive Co₃O₄/expanded graphite paper¹¹⁵ and Co₃O₄ nanoparticles-embedded carbonaceous fibers^112,113 have been fabricated, and all these electrodes exhibited excellent electrochemical performance.

Metal–organic frameworks (MOFs), assembled from two main parts of inorganic vertices (metal ions or clusters) and organic linkers, are noted for their high surface areas and well-designed pore structures.¹¹⁶ This class of porous material has been demonstrated to provide promising templates or precursors to produce novel porous transition metal oxide nanostructures. The porous structure is beneficial for reducing the Li⁺ diffusion distance and relieving volume expansion strain. Moreover, the carbon matrix can be derived from the organic linkers under well-controlled conditions, which is important to enhance the electronic conductivity of the electrode material. Hou et al. prepared Co-based zeolitic imidazolate frameworks (ZIF-67) at room temperature and used them as precursors to obtain Co₃O₄/N-doped porous carbon dodecahedrons.¹¹⁷ In the obtained hybrid dodecahedrons, Co₃O₄ nanoparticles with dimensions of 15 to 30 nm were embedded in the N-doped porous carbon network which was derived from the organic linkers. The mean pore size and the specific surface area were calculated to be 36 nm and 97 m² g⁻¹, suggesting its mesoporous architecture. The unique rational structure that utilized the synergic merits of the highly conductive porous carbon matrix and well-defined Co₃O₄ nanoparticles could accommodate volume expansion and improve the electric and ionic conductivity, endowing the hybrid material with long cycling stability and excellent rate capability.

Wang's group inserted MWCNTs in MOFs to construct hierarchical porous MWCNTs/Co₃O₄ nanocomposites.¹⁰³ Carboxylic group (–COOH)-functionalized MWCNTs were used to adsorb Co²⁺ and served as nucleation centers for loading MOFs in 2-methylimidazolate methanol solution. As a result, the MWCNTs were self-inserted in situ into the ZIF-67 (Co) crystals.

Finally, MWCNTs/Co₃O₄ was obtained from MWCNTs/ZIF-67 by conducting heat treatment at 400 °C in air. The designed MWCNTs/Co₃O₄ integrated the excellent conductivity and strong mechanical/chemical stability of MWCNTs and the high theoretical capacity of Co₃O₄, displaying capacities of 813 mA h g⁻¹ at 0.1 A g⁻¹ and 541 mA h g⁻¹ at 1.0 A g⁻¹. Furthermore, their protocol could be further extended to the construction of MWCNTs/ZnCo₂O₄ polyhedrons using MWCNTs/ZIF-67 (Zn, Co) as a precursor. Recently, our group prepared hierarchical porous ZnO/ZnCo₂O₄ nanosheets using a Zn–Co-MOF nanosheets precursor.¹¹⁸ Qu et al. spread ZIF-67 on graphene oxide nanosheets and used them as precursors to obtain ultrafine Co₃O₄ nanocrystallites (<10 nm) grown on graphene.¹¹⁹ The obtained graphene/Co₃O₄ manifested long-term cycling stability and excellent rate behavior (a capacity of 877 mA h g⁻¹ at 5 A g⁻¹).

Lou's group designed hierarchical tubular structures composed of in situ-formed CNTs and Co₃O₄ hollow nanoparticles to address multiple issues at the same time.¹²⁰Fig. 8a illustrates the innovative fabrication of the hierarchical CNT/Co₃O₄ microtubes. They first prepared polyacrylonitrile (PAN)–cobalt acetate (Co(Ac)₂) composite nanofibers (PAN–Co(Ac)₂) and used them as the template and cobalt source for the growth of ZIF-67 nanotubes. After dissolving the extra PAN–Co(Ac)₂ core and annealing in Ar/H₂, the ZIF-67 tubular structures transformed into CNT/Co–carbon hybrids. Finally, hierarchical CNT/Co₃O₄ microtubes were obtained after the CNT/Co–carbon hybrids underwent controlled thermal annealing in air. As confirmed by the SEM and TEM images in Fig. 8b–d, the ZIF-67 tubular structures were totally converted to hierarchical CNT/Co₃O₄ microtubes, in which the sizes of the Co₃O₄ hollow nanoparticles ranged from 15 to 30 nm and the inner diameter of the CNTs was about 3 ± 2 nm. As a consequence, the hierarchical structure showed a high specific surface area of 93.9 m² g⁻¹ and small pore sizes (mostly below 15 nm). This configuration not only provides a short Li⁺ diffusion distance but also offers sufficient contact area for the charge-transfer reaction. Furthermore, the hollow structure could relieve volume expansion stress, and the CNTs integrated in the hierarchical tubular structures improved the electronic conductivity of the hybrid material. Consequently, their CNT/Co₃O₄ LIBs anode delivered a high reversible capacity of 1281 mA h g⁻¹ at 0.1 A g⁻¹ with a high rate capability and long cycle life over 200 cycles.

Fig. 8 (a) Formation of hierarchical CNT/Co₃O₄ microtubes. (b) SEM image of the as-prepared ZIF-67 microtubes. SEM (c) and TEM (d) images of the as-prepared CNT/Co₃O₄ microtubes. Reproduced with permission.¹²⁰ Copyright 2016, Wiley.

The electrodes mentioned above were prepared by a traditional slurry coating method, in which insulating binders are required to adhere the active materials to the current collector. The addition of binders not only may block the active sites but also decrease the electronic conductivity of the electrodes, impeding efforts to enhance the conductivity of the hybrid materials. Therefore, the rational fabrication of binder-free electrodes with various morphologies is meaningful. Zhang's group reported a novel strategy to fabricate binder-free CoO/graphene electrodes using an intelligent electrostatically induced spread growth method.¹¹⁴ As they pointed out, the as-prepared α-Co(OH)₂ nanosheets (Fig. 9a) are positively charged and the graphene oxide is negatively charged; therefore, mutual electrostatic interactions between these two materials can drive the assembly. Moreover, a large amount of ultrathin α-Co(OH)₂ nanosheets can be absorbed owing to the high surface area of GO. When the composites undergo reflux reactions under reduction atmosphere, α-Co(OH)₂ nanosheets are transformed to β-Co(OH)₂ and GO is reduced to graphene. The β-Co(OH)₂ nanosheets grow with the graphene surface at the same time and finally uniformly disperse on the graphene (Fig. 9b). Finally, a binder-free CoO/graphene anode (Fig. 9c) is constructed on Cu foil after the process of drop-coating, drying, and sintering of the β-Co(OH)₂ nanosheets and graphene composites. The capacity of the obtained binder-free CoO/graphene anode after 5000 cycles at a current density of 1.0 A g⁻¹ is 604 mA h g⁻¹, showing a high reversible capacity and outstanding cycling stability. As a comparison, the electrode with PVDF exhibits inferior cycling stability and lower specific capacity. The superior electrochemical performance of the binder-free electrode should be ascribed to its rationally designed structure. The thin CoO nanosheets can shorten the Li⁺ ion diffusion pathways, the encompassment of highly conductive graphene can ensure fast electron transfer, and the absence of insulating PVDF binder can maintain the high electronic conductivity of the electrode.

Fig. 9 TEM images of α-Co(OH)₂ nanosheets (a) and β-Co(OH)₂ nanosheets/graphene hybrid (b); image of the CoO/graphene hybrid (c). Reproduced with permission.¹¹⁴ Copyright 2013, Wiley. (d) TEM image of the Co₃O₄/carbon nanosheet array. Reproduced with permission.¹²¹ Copyright 2015, RSC. SEM (e) and TEM (f) images of CoO nanowire clusters. Reproduced with permission.⁹⁹ Copyright 2015, Wiley. (g) SEM image of Co₃O₄/3DNF. Reproduced with permission.¹²² Copyright 2016, Elsevier.

Hollow Co₃O₄ nanoparticles confined in thin carbon nanosheet arrays were fabricated on Ni foam and used as a binder-free electrode for LIBs by Peng et al.¹²¹ Co(OH)₂ nanosheet arrays were firstly electrodeposited on Ni foam; then, carbon layers were coated by a hydrothermal reaction using glucose as a green carbon source. Upon being annealed in H₂ and air atmosphere in sequence, hollow Co₃O₄ nanoparticles about 10 to 20 nm in size were formed due to the nanoscale Kirkendall effect and uniformly distributed in the thin carbon nanosheets (Fig. 9d). The as-prepared binder free anode features intimate contact between the active material and Ni foam current collector, high surface area, and an open framework, which are vital to achieve excellent electrochemical performance. CoO nanowire clusters grown on Cu foil were fabricated by our group and evaluated as a binder-free LIB anode.⁹⁹Fig. 9e and f suggest that the CoO nanowires are composed of small nanoparticles about 10 nm in size that are directly grown on the current collectors. More importantly, the voids between the nanoparticles and the spaces between the nanowires are helpful for electrolyte permeation as well as remission of stress caused by volume changes. Additionally, the small particles shorten the Li⁺ diffusion distances in the materials. As a consequence, the obtained electrode exhibits a high reversible capacity of 1516.2 mA h g⁻¹ at 1 C (1 C = 0.716 A g⁻¹), which is higher than the theoretical specific capacity. We discussed the origins of the additional capacity, and our results suggested that the higher-oxidation-state products and pseudocapacitive charges should be responsible for this phenomenon. Moreover, we fabricated nano-reactors for the CoO nanowires by coating them with amorphous silica shells. The shell worked as a stiff scaffold to prevent outward volume expansion, preserving the morphology of the nanowires and enhancing the cycling stability of the binder free electrodes. Fang et al. coated MOFs on 3D current collectors and used them as a precursor to prepare porous transition metal oxide binder-free electrodes.¹²² As a proof of concept, a Co₃O₄/3D nickel foam (Co₃O₄/3DNF) hybrid binder-free electrode was obtained (Fig. 9g). This electrode possesses the advantages of porous nanostructures and a 3D conductive substrate.

2.4. Nickel oxide

Nickel oxide (NiO) has received a great deal of attention due to its high theoretical capacity (718 mA h g⁻¹) and high volumetric density (6.67 g cm⁻³). However, the practical application of NiO in LIBs is still hindered by its poor cycling stability and very poor rate capability, which result from its intrinsically low conductivity and drastic volume variations during the continuous discharge/charge process. There are three main effective strategies to overcome these obstacles. The first significant approach is to construct nanostructured NiO, such as nanofibers,¹²³ nanosheets,^124,125 nanoparticles,¹²⁶ nanoflowers,^127,128 nanorods,¹²⁹ nanowires,¹³⁰ hierarchical microspheres,¹³¹ and multi-shelled hollow microspheres.¹³²

NiO with nanoscale structures efficiently shortens the length of Li⁺ diffusion and has improved structural stability and enhanced electrode/electrolyte surface contact area, contributing to enhanced electrochemical performance. For example, Bell et al. fabricated free-standing Ni–NiO nanofibers with a core–shell structure by a feasible process of electrospinning and thermal oxidation.¹²³ In Fig. 10a, it can be seen that the Ni–NiO nanofibers are continuous while the surface is rough, which is due to NiO layers attaching on the surface of the Ni nanofibers. When tested as an anode in LIBs, the Ni–NiO nanofibers demonstrated extraordinary performance, with a capacity of 1054 mA h g⁻¹ at 3 C (1 C = 718 mA g⁻¹), an ultralong cycle life of more than 1500 cycles, and excellent stability, with a coulombic efficiency of up to 99%. Additionally, ultrathin NiO nanosheets (4 to 5 nm in thickness) have been synthesized by Sun et al.¹²⁴ The TEM image of the ultrathin NiO nanosheets displayed in Fig. 10b clearly suggests their nanosheet-like structures. When evaluated as a LIB anode, the NiO nanosheets manifested high reversible capacities of 1242 and 250 mA h g⁻¹ at 0.2 and 15 A g⁻¹, respectively. Furthermore, 3D binder-free NiO nanorod-anchored Ni foam electrodes were fabricated by Yang et al.¹²⁹ Due to its favorable morphology and the abundant electrical contact between NiO and Ni foam, the obtained binder-free electrode exhibits high reversible capacity and superior rate performance. The second vital method is to fabricate novel composites combining NiO with conductive materials. Among these, graphene,¹³³ amorphous carbon,¹³⁴ MWCNTs,¹³⁵ metal,^136,137 and amorphous carbon nanotubes¹³⁸ are all excellent options. Lou's group constructed hybrid bowl-like structures by anchoring NiO nanosheets on flat carbon hollow particles.¹³⁹ The SEM and TEM images presented in Fig. 10c specifically indicate the hollow structure of this bowl-like NiO/C hybrid. This unique structure combines the merits of NiO sheet-like nano-building blocks and carbon layers, which effectively enhances the conductivity of the electrode and inhibits self-agglomeration of the active materials. Consequently, high specific capacity (1012 mA h g⁻¹ at 0.2 A g⁻¹), good cycling stability (93.8% capacity retention after 150 cycles at 0.2 A g⁻¹) and superior rate performance (608 mA h g⁻¹ at 1.6 A g⁻¹) were achieved. Shan et al. prepared NiO/graphene via a hydrothermal method and visualized the role of graphene during lithiation by in situ TEM.¹³³ Their results suggest that the addition of graphene enhances Li⁺ diffusion kinetics by increasing the Li⁺ diffusion rate, limits the expansion of NiO, and retains the tight electrical contact between NiO and graphene. Feng et al.¹³⁴ reported 2D sandwich-like NiO/C arrays on Ti foil through a hydrothermal method (Fig. 10d). The obtained sandwiched composites possess numerous active sites as well as enhanced contact area between the electrolyte and active materials. Furthermore, this configuration also protects the active nanoparticles from peeling off from the conductive substrate, ensuring excellent electrochemical performance (a capacity of ∼1458 mA h g⁻¹ at 0.5 A g⁻¹ was obtained, and a capacity of ∼95.7% was retained after 300 cycles) of this electrode. Susantyoko et al. sputtered NiO on vertically-aligned MWCNT arrays as an LIBs anode (MWCNT/NiO),¹³⁵ of which MWCNT/NiO presents a specific capacity of 864.8 mA h g⁻¹ at 143.6 mA g⁻¹ after 50 cycles. In addition, fabricating sophisticated and unique structures is a powerful method to improve the electrochemical performance of NiO. For example, a hierarchical hollow ball-in-ball NiO/Ni/graphene architecture (Fig. 10e) has been reported by Zou et al.¹⁴⁰ A high reversible specific capacity, excellent rate capability, and ultralong cycling stability are achieved when this architecture is evaluated as an LIB anode (Fig. 10f). This extraordinary performance benefits from the well-designed, unique structure of the NiO/Ni/graphene composites, which mitigates the volume changes of NiO and offers a continuous, highly conductive matrix to facilitate fast electron transfer and form stable SEI layers. Furthermore, “curved” NiO nanomembranes have been reported through a simple fabrication technique followed by a thermal treatment process by Sun et al.¹⁴¹ The “curved” structure was confirmed by the SEM image in Fig. 10g. The CV curves in Fig. 10h present an intense peak located at around 0.5 V in the cathodic scan during the first cycle, corresponding to the reduction of NiO to Ni, as well as the formation of SEI layers. Afterwards, this cathodic peak shifts to about 1.07 V and becomes weaker. The two broad anodic peaks located at about 1.4 and 2.2 V can be attributed to the decomposition of the SEI layer and the oxidation of Ni nanoparticles to NiO, respectively. The curved NiO nanomembranes deliver a high capacity of 721 mA h g⁻¹ at 1.5 C, an ultrafast power rate (50 C), and a long lifetime (1400 cycles) when evaluated as a LIB anode.

Fig. 10 (a) SEM image of core–shell nanofibers of NiO prepared by electrospinning. Reproduced with permission.¹²³ Copyright 2015, Elsevier. (b) TEM image and inset in (b) SAED patterns of NiO nanosheets. Reproduced with permission.¹²⁴ Copyright 2015, Elsevier. (c) SEM and TEM images (inset of c) of hybrid bowl-like structures obtained by anchoring NiO nanosheets on flat carbon hollow particles. Reproduced with permission.¹³⁹ Copyright 2015, RSC. (d) SEM image of 2D sandwich-like NiO/C arrays. Reproduced with permission.¹³⁴ Copyright 2016, Elsevier. (e) TEM images and (f) rate performance of hierarchical hollow ball-in-ball NiO/Ni/graphene nanostructured materials. Reproduced with permission.¹⁴⁰ Copyright 2016, ACS. (g) SEM image, (h) cyclic voltammetry curves of the curved NiO nanomembranes. Reproduced with permission.¹⁴¹ Copyright 2014, Wiley.

2.5. Copper oxide

When used as a LIB anode, CuO forms Cu metal nanoparticles embedded in an Li₂O matrix during the discharge process and re-forms as CuO during the charge process. CuO electrodes suffer from large volume expansion (174%), low conductivity (p-type semiconductor), and subsequent particle pulverization, resulting in rapid capacity fading and poor cycling performance. In order to efficiently address these issues, one effective way is to construct micro-/nano-structured CuO materials with various morphologies, mainly including nanowires (Fig. 11a),^142–144 nanosheets (Fig. 11b),¹⁴⁵ nanofibers (Fig. 11c),¹⁴⁶ nanorods,¹⁴⁷ nanoribbons,^148,149 3D hierarchical mesocrystals (Fig. 11d),¹⁵⁰ and hollow structures (Fig. 11e and f).^151,152 The nanostructured CuO can alleviate mechanical strain and reduce the diffusion length of Li⁺, leading to excellent performance for Li storage. For example, Zhang et al.¹⁴⁴ have fabricated vertically aligned single crystalline CuO nanowires on nickel foam by one-step thermal oxidation. The growth mechanism of CuO nanowires is regarded as stress-driven grain-boundary diffusion associated with surface diffusion of Cu atoms/ions. When used as a binder-free LIB anode, the CuO nanowires exhibit a high capacity of 692 mA h g⁻¹ after 50 cycles at 0.1 A g⁻¹ and a reversible capacity of up to 445 mA h g⁻¹ over 600 cycles, even at 1.0 A g⁻¹.

Fig. 11 (a) Nanowires, reproduced with permission.¹⁴⁴ Copyright 2014, RSC. (b) Nanosheets, reproduced with permission.¹⁴⁵ Copyright 2013, RSC. (c) Nanofibers, reproduced with permission.¹⁴⁶ Copyright 2016, RSC. (d) 3D hierarchical mesocrystals, reproduced with permission.¹⁵⁰ Copyright 2016, RSC. (e and f) Hollow structures, reproduced with permission.¹⁵² Copyright 2013, RSC. (g) Microspherical, (h) flower-like and (i) thorn-like CuO structures. Reproduced with permission.¹⁵³ Copyright 2014, Elsevier.

Furthermore, CuO nanowires coated with graphene quantum dots (GQD) grown on Cu foam have been designed and synthesized by Fan's group.¹⁴³ The unique GQD soft protection greatly increases the surface conductivity and stability of the nanowire arrays. Consequently, when the CuO–Cu–GQD triaxial nanowire arrays were applied to the LIB anode, they deliver a high reversible capacity (780 mA h g⁻¹ at 1/3 C), a superior rate capability (330 mA h g⁻¹ at 30 C) and excellent long-term cycling stability (almost no capacity decay after 1000 cycles). In addition, Wang et al.¹⁵³ controllably synthesized three types of micro/nanostructured CuO (CuO microspheres, flower-like CuO and thorn-like CuO) anodes for lithium-ion batteries (Fig. 11g–i). The CuO microspheres anode demonstrates much better electrochemical performance than the other two CuO samples in terms of cycling performance and rate capability. Another ideal strategy is to construct smart hybrid architectures by integrating CuO with conductive matrixes (such as carbon matrix,^154–158 metal¹⁵⁹ and other conductive substances¹⁶⁰), which not only improve the conductivity of CuO but also restrain the self-aggregation of CuO active materials during the cycling process. Liu's group encapsulated CuO nanoparticles in mesoporous carbon multi-yolk–shell octahedra through a solvothermal process followed by thermal treatment (Fig. 12a).¹⁵⁴ In the obtained samples, the multiple CuO nanoparticles are well-embedded in the compartments of octahedral carbon scaffolds. This novel multi-yolk–shell structure can effectively buffer volume variations, prevent the aggregation of CuO nanoparticles during the charge/discharge process and stabilize the SEI layers. Therefore, the CuO@C octahedra anode demonstrates high reversible capacity (598 mA h g⁻¹ at 0.25 A g⁻¹), excellent rate capacity and long-term cycling stability in LIBs. In addition, Ko et al. reported CuO/MWCNTs nanocomposites in which mesoporous CuO particles are threaded by MWCNTs in the long-axis direction. ¹⁵⁸ The high porosity of the CuO particles allows easy access of Li ions and acts as an elastic to buffer volume variations during the lithiation/delithiation process. The electrodes achieved remarkably improved electrochemical performance because the MWCNTs efficiently enhanced the electronic conductivity of the nanocomposites. Wang's group¹⁵⁵ fabricated carbon-coated CuO hollow spheres via a feasible aerosol spray pyrolysis method. In the hollow structure, a thin layer of CuO nanoparticles in the inner chamber was anchored to the carbon shell. The hollow spherical particles anode exhibited a high capacity (670 mA h g⁻¹ at 1 C) as well as outstanding rate capacity (a capacity of 400 mA h g⁻¹ at 50 C was retained after 300 cycles). The extraordinary electrochemical performance resulted from the structure of the carbon-coated hollow spheres. This unique structure improves the electronic conductivity of the hybrid material, decreases the diffusion distance of Li ions, and provides sufficient space for the volume expansion of CuO particles during the discharge/charge process.

Fig. 12 (a) The synthesis procedure of CuO-encapsulated mesoporous carbon multi-yolk–shell octahedra. Reproduced with permission.¹⁵⁴ Copyright 2016, Elsevier. (b) The formation mechanisms of cubic-Cu₂O and octa-Cu₂O. (c) Comparison of the cycling performances of cubic Cu₂O, octahedral Cu₂O and commercial Cu₂O. Reproduced with permission.¹⁶⁴ Copyright 2015, RSC.

In addition to CuO, Cu₂O has also aroused the interest of researchers as a LIB anode material due to its high reversible reaction with Li⁺, non-toxicity, and low cost. Unfortunately, inferior capacity and stability resulting from extreme volume expansion during the process of delithiation/lithiation have hindered the development of Cu₂O. Recently, many researchers have focused on fabricating nanostructured Cu₂O electrodes with controlled shapes, mainly including nanorod arrays,^161,162 porous hollow nanospheres,¹⁶³ and cubic Cu₂O,^164,165 for improving the performance of Cu₂O anodes. Cubic and octahedral Cu₂O nanostructures have been prepared by Kim et al.¹⁶⁴Fig. 12b schematically illustrates the formation mechanisms of cubic and octahedral Cu₂O. In the presence of PVP, the sample tends to form octahedral Cu₂O with dominant (111) facets because PVP mainly adsorbs on the (111) plane. In contrast, the sample tends to form cubic Cu₂O with more thermodynamically stable (100) facets in the absence of PVP. Hence, it is found that cubic Cu₂O and octahedral Cu₂O have dominant (100) and (111) facets, respectively. The electrochemical performance of octahedral Cu₂O is superior to that of cubic Cu₂O (Fig. 12c), regardless of the achieved specific capacity or the rate capability; this may be because a crystal structure consisting of (111) facets facilitates lithium ion transport, as the authors declared.¹⁶⁴

2.6. Ruthenium oxide

Ruthenium dioxide, RuO₂, features a tetragonal rutile structure and metallic-type electronic conduction (σ is about 10³ to 10⁴ S cm⁻¹ at 300 K). As a precious metal, RuO₂ is not a desirable candidate as a LIB anode material. It is often chosen as a prototype material to study the conversion reaction mechanism due to its high coulombic efficiency and good mass transport properties.^166–171 More importantly, it does not form superparamagnetic nanoparticles like Fe and Co systems, which enables the application of many techniques. Theoretically, RuO₂ can accept four electrons per formula unit when used as an LIB anode material, delivering a theoretical capacity of 806 mA h g⁻¹. As with other conversion reaction materials, the measured capacity is always larger than the theoretical capacity. Grey's group chose RuO₂ as a prototype material to investigate the origin of this additional capacity.¹⁶⁹ They employed X-ray absorption spectroscopy (XAS), in situ pair distribution function methods, high-resolution multinuclear/multidimensional solid-state NMR techniques, and other electrochemical measurements to identify the origins of the additional capacity. As they summarized, the formation of LiOH and the subsequent reversible formation of Li₂O and LiH were confirmed as the main origins of the additional capacity in their system. Additionally, the reversible formation of SEI layers and Li adsorption also contribute to the additional capacity. Kim et al. investigated the structural changes and electrochemical behavior of RuO₂ by in situ XRD and X-ray absorption spectroscopy combined with electrochemical techniques.¹⁷⁰ They found that intermediate phase LiRuO₂ was initially formed at the start of discharge and further decomposed to Ru metal and Li₂O by a conversion reaction. They also probed the source of the additional capacity by TEM, X-ray photoelectron spectroscopy, and the galvanostatic intermittent titration technique. They proposed that the additional capacity results from Li storage in the grain boundary between the Ru nanoparticles and Li₂O.

2.7. Chromium oxide

Cr₂O₃ has been evaluated as LIB anode due to its high theoretical capacity of 1058 mA h g⁻¹ and relatively low lithium insertion potential of 1.08 V. However, Cr₂O₃ has the drawbacks of very poor electronic conductivity and severe volume variation during the cycling process, which result in rapid deterioration of cyclic performance. To address these issues, many strategies, mainly including the fabrication of micro-/nano-structured materials and hybrid composites such as mesoporous Cr₂O₃,^172,173 Cr₂O₃/graphene,^174,175 Cr₂O₃/CNTs,¹⁷⁶ and carbon-coated graphene–Cr₂O₃,¹⁷⁷ have been reported. Liu et al. constructed highly ordered mesoporous Cr₂O₃ materials by a vacuum-assisted route using 2-dimensional (2D) hexagonal SBA-15 and 3D cubic KIT-6 silica as templates.¹⁷²Fig. 13a and b present TEM images of 2D hexagonal Cr₂O₃ (denoted as H-Cr₂O₃), from which we can observe that H-Cr₂O₃ is dominated by a wire-like array shape. Additionally, Fig. 13c and d clearly indicate that C-Cr₂O₃ possesses large scales of irregular spherical particles with highly ordered mesopores. Owing to the high surface area, thin crystal walls, and narrow pore size distribution on these unique ordered mesoporous materials (Fig. 13e), the mesoporous H-Cr₂O₃ and C-Cr₂O₃ electrodes demonstrate enhanced discharge capacities of 521 mA h g⁻¹ and 540 mA h g⁻¹ after 100 cycles compared to a bulk Cr₂O₃ (B-Cr₂O₃) electrode (Fig. 13f). In addition, the intercalation-transformation method has been developed to fabricate sandwich-like Cr₂O₃–graphite intercalation composites (Cr₂O₃–GICs).¹⁷⁸ The resulting Cr₂O₃–GICs manifests a reversible capacity as high as 500 mA h g⁻¹ and extremely stable cycling performance with over 100% capacity retention after 1000 cycles.

Fig. 13 (a) TEM and (b) HRTEM images of H-Cr₂O₃. (c) TEM and (d) HRTEM images of C-Cr₂O₃. (e) Nitrogen sorption isotherms and inset in (e) corresponding pore size distribution. (f) Cycle performance of B-Cr₂O₃, H-Cr₂O₃ and C-Cr₂O₃. Reproduced with permission.¹⁷² Copyright 2012, RSC.

2.8. Molybdenum oxide

It is well known that MoO₃ exists in three different forms: orthorhombic (α-MoO₃), monoclinic (β-MoO₃), and hexagonal (h-MoO₃). α-MoO₃ is the most stable phase; its anisotropic layered structure makes it a suitable candidate for LIBs. Based on the conversion reaction mechanism, a high theoretical specific capacity of 1117 mA h g⁻¹ is delivered by MoO₃ anodes. However, poor conductivity and tremendous volumetric variations during the discharge/charge process lead to very poor cyclic stability and rapid capacity fading. Similar to other transition metal oxides, numerous efforts have been devoted to constructing nanostructured MoO₃, such as nanorods,¹⁷⁹ nanobelts,¹⁸⁰ and nanowires.¹⁸¹ Qian's group prepared mesoporous orthorhombic MoO₃ nanowire bundles from single-crystal α-MoO₃·H₂O nanorods via a feasible method of vacuum topotactic transformation.¹⁸¹ The as-prepared MoO₃ nanowire bundles exhibited a high reversible capacity of 954.8 mA h g⁻¹ after 150 cycles when tested as a LIB anode. Another efficient method to improve the cycling stability and rate performance of MoO₃ anode materials is the construction of self-supported nanostructure materials directly on current collectors. Additionally, coating a conductive layer and confining the MoO₃ active material in a selected matrix are regarded as effective strategies to improve the conductivity and relieve the structure degradation of MoO₃.^181–189 For example, an anion-exchange reaction has been developed to fabricate a tailored heterostructure of MoO₃@MoS₂ (MoS₂ nanosheets grown on MoO₃ nanowires) by Liu et al.¹⁸² Due to the abundant Li⁺ diffusion channels resulting from layered MoS₂ nanosheets grown on MoO₃ nanowires and the synergistic effect between the MoO₃ nanowires and MoS₂ nanosheets, MoO₃@MoS₂ exhibits a high discharge capacity (1510 mA h g⁻¹ was obtained at 0.1 A g⁻¹) and good cycle performance (a discharge capacity of 781 mA h g⁻¹ was retained after 100 cycles). More interestingly, many complicated and subtle structures of MoO₃ have been proposed in recent years.^190,191 Kang's group reported a new MoO₃ structure, named “ant-cave-microball” by spraying pyrolysis using a colloidal spray solution with polystyrene nanobeads, molybdenum salt, and sucrose.¹⁹¹ As depicted in the SEM image in Fig. 14a, the “ant-cave-structured” MoO₃-C microballs possess many open nanochannels which are uniformly distributed inside the microballs. Meanwhile, the MoO₃ and carbon components are distributed homogeneously in the microballs. As illustrated in Fig. 14b, the channels facilitate electrolyte immersion and shorten the Li⁺ diffusion distance. Moreover, the carbon component can improve the electronic conductivity of the microballs and function as a buffering layer for volume changes during the lithiation and delithiation processes. Owing to the synergistic effect of the nanochannels and carbon components of this unique structure, an “ant-cave-structured” MoO₃-C microball anode delivered superior cycling performance (a capacity of 733 mA h g⁻¹ after 300 cycles) and high rate performance (a capacity of 679 mA h g⁻¹ even at 3 A g⁻¹). Li et al.¹⁹² systematically investigated the lithium adsorption and diffusion behavior in MoO₃ with different dimensions by density functional theory computations. Their computational results suggest that MoO₃ monolayer nanosheets and nanoribbons have more attractive properties, such as fast Li⁺ diffusion, high operating voltage, excellent electronic conductivity, and large energy density. Due to these properties, MoO₃ nanosheets and nanoribbons have become promising high-rate LIB anodes.

Fig. 14 (a) SEM image of ant-cave-structured MoO₃-C. (b) Illustration of electrolyte penetration in ant-cave-structured MoO₃-C composite powders. Reproduced with permission.¹⁹¹ Copyright 2013, ACS. (c) The sequential self-templating mechanism for formation of triple-shelled Mo-PDA hollow spheres. Reproduced with permission.²⁰² Copyright 2016, Wiley.

Among molybdenum-based oxides, MoO₂ has also received substantial attention. Diverse MoO₂ nanomaterials, such as nanotubes,¹⁹³ nanorods,¹⁹⁴ nanobelts,¹⁹⁵ nanosheets,¹⁹⁶ porous nanostructures and yolk–shell microspheres,¹⁹⁷ as well as various MoO₂/carbon hybrids such as MoO₂@carbon hollow nanospheres,¹⁹⁸ exfoliated graphene oxide/MoO₂,¹⁹⁹ MoO₂/C,^200–205 and MoO₂/MWCNT,²⁰⁶ have been reported. Mai's group encapsulated ultrathin MoO₂ nanosheets in a carbon matrix.¹⁹⁶ The resulting MoO₂/C nanosheets exhibited superior Li storage capacity, retaining 1051 mA h g⁻¹ at 0.5 A g⁻¹ after 100 cycles. Triple-shelled MoO₂-C hollow spheres have been synthesized by Lou's group.²⁰² Benefiting from their proposed sequential self-templating mechanism, the shell numbers of Mo-polydopamine (Mo-PDA) precursor can be controlled by adjusting the etching conditions of the Mo-glycerate (MoG) solid spheres, as depicted in Fig. 14c. Finally, the inner solid core is etched, forming a triple-shelled Mo-PDA hollow sphere. After thermal treatment, triple-shelled MoO₂/carbon composite hollow spheres are obtained. This unique structure offers more electrochemical active sites and adequate space to relieve volume changes, thus exhibiting high specific capacities and excellent cycling stability (a capacity of about 580 mA h g⁻¹ was obtained at 0.5 A g⁻¹ after 200 cycles). Mo₂C is a highly conductive material with a high specific conductance of 1.02 × 10² S cm⁻¹; however, it is electrochemically inactive for LIBs. Zhang et al. used Mo₂C to modify MoO₂ material by fabricating MoO₂/Mo₂C heteronanotubes for the first time, in which Mo₂C distinctly increased the electronic conductivity and structural stability of the heteronanotubes.²⁰⁷ Taking advantage of the combination of the hierarchical porous structure and appropriate addition of Mo₂C, the as-prepared MoO₂/Mo₂C heteronanotubes anode exhibited superior cycling performance and outstanding rate performance (a capacity of 510 mA h g⁻¹ was retained at 1.0 A g⁻¹ after 140 cycles).

2.9. Tungsten oxide

In its conversion reaction with Li, tungsten trioxide (WO₃) delivers a specific capacity of 693 mA h g⁻¹ and 6 electrons are transferred, as shown in eqn (19):^208,209

WO₃ + 6Li⁺ + 6e⁻ ↔ W + 3Li₂O. (19)

Prior to the conversion reaction, a Li intercalation step was suggested to occur in the first lithiation process.^210,211

Recently, He et al. confirmed the intercalation-initiated conversion reaction on WO₃ by in situ HRTEM. The intercalation process is expressed as follows:²¹²

WO₃ + xLi⁺ + xe⁻ ↔ Li_xWO₃ (20)

They carried out the experiment on a WO₃ single crystal model electrode and revealed that an intercalation step occurred prior to the conversion reaction. As depicted in Fig. 15, Li⁺ diffused and intercalated into the WO₃ unit cells and partially reduced W⁶⁺. As the lithium concentration increased, the intercalation phase gradually evolved from monoclinic WO₃ to cubic Li_xWO₃. Meanwhile, Li atoms bonded with oxygen atoms, which reduced Wⁿ⁺. The WO₃ framework was destabilized and gradually shrank and distorted. Finally (full lithiation), the Wⁿ⁺ ions were fully reduced to W⁰, and the WO₃ framework collapsed into W metal and a Li₂O matrix. In order to restrain volume changes during cycling, various nanostructures of WO₃-based materials have been developed. WO₃ hollow nanospheres,^213–215 nanowire arrays on carbon cloth,²⁰⁹ yolk–shell WO₃–carbon composites,²¹⁶ and hexagonal ultrathin WO₃ nano-ribbons,²¹⁷ hierarchical chrysanthemum-like WO₃·0.33H₂O,²¹⁸ and hierarchical WO₃@SnO₂ core–shell nanowires²¹⁹ were prepared and evaluated as LIB anodes. Li et al. synthesized a nanocomposite of polyaniline (PANI)/mesoporous tungsten trioxide (m-WO₃) by coating polymerized aniline onto ordered m-WO₃.²²⁰ For the m-WO₃ anode, the main reduction peak emerged at about 0.73 V; also, a small cathodic current starting from 2.5 V was observed and was ascribed to the Li intercalation process. The broad oxidation peak at about 1.24 V (vs. Li⁺/Li) is attributed to the oxidation reaction of W metal nanograins to WO₃. When PANI is coated, the reduction peak at 0.73 V positively shifts to 0.79 V and the oxidation peak negatively shifts to 1.21 V, suggesting that the electrochemical activity of m-WO₃ improves after it is coated with PANI. The PANI/m-WO₃ anode exhibits a capacity of 883 mA h g⁻¹ at 180 mA g⁻¹ and retains a reversible capacity of 803 mA h g⁻¹ after 100 cycles. Meanwhile, the m-WO₃ anode shows fast capacity fading, decreasing from 449 mA h g⁻¹ to 303 mA h g⁻¹ after 50 cycles. Wang et al. synthesized an off-stoichiometric hollow tungsten trioxide (WO₃-s) by a selective leaching strategy.²¹⁵ Yolk–shell WO₃-s could be obtained from the hollow structure by a further intervallic leaching process. They found that the electrochemical performances of hollow structure and yolk–shell structure WO₃-s were superior to that of commercial densely structured WO₃-s. Moreover, the yolk–shell structure exhibited the most stable cycling performance owing to its additional spatial elasticity, which could buffer volume changes.

Fig. 15 (a) HAADF images showing WO₃ structure evolution during Li insertion (lithiation). Scale bar, 50 nm. (b) Nano beam diffractions (NBD) from the pristine sample and across the conversion front. (c) Structure and symmetry evolution with increasing Li concentration in the Li_xWO₃ lattice as indicated by the corresponding NBDs. Reproduced with permission.²¹² Copyright 2016, Wiley.

3. Binary metal oxides

Binary metal oxides (A_xB_yO_z, A = Mn, Fe, Co, Ni, and Cu; B = Mn, Fe, Co, Ni, or Cu; A ≠ B) based on conversion reactions have two electrochemical active transition metal ions and show exciting electrochemical properties as LIB anodes. In the first lithiation, two metal nanoparticles (A and B metals) are reduced and dispersed in the Li₂O matrix. After full delithiation, a mixture of two metal oxides (AO_x and BO_y) is formed, replacing the original binary metal oxide (A_xB_yO_z). The relevant electrochemical reactions can be summarized as follows:

A_xB_yO_z + 2zLi⁺ + 2ze⁻ → xA + yB + zLi₂O (first lithiation) (21)

A + B + (x + y)Li₂O → AO_x + BO_y + 2(x + y)Li⁺ + 2(x + y)e⁻ (first delithiation) (22)

AO_x + BO_y + 2(x + y)Li⁺ + 2(x + y)e⁻ ↔ A + B + (x + y)Li₂O (23)

In this section, binary metal oxides are divided to three parts: AB₂O₄ type (AMn₂O₄, AFe₂O₄, and ACo₂O₄) spinel structure oxides, ABO₄ type (AMoO₄), and other binary metal oxide types (ABO₂, ABO₃, and A₂B₃O₈). The structure and electrochemical performance of these materials are reviewed as follows.

3.1. AB₂O₄ type

AB₂O₄ type binary metal oxides always have spinel structures. They have attracted considerable interest owing to the complementarity and synergy of the two transition metal oxides during discharge–charge processes. These oxides usually exhibit better electrical conductivity and higher electrochemical activity than mono-metal oxides; thus, they are more attractive in energy storage and conversion applications.^221,222 When used as LIB anodes, AB₂O₄ type binary metal oxides deliver high theoretical specific capacity (≈1000 mA h g⁻¹) by transferring 8 electrons. However, the enormous volume expansion during lithiation often leads to poor cycling performance.

3.1.1. AMn₂O₄ (A = Fe, Co, Ni, Cu). Manganese-based oxides can offer higher output voltages and energy densities due to their lower operating voltages (1.2 to 1.4 V for lithium extraction). Furthermore, manganese is highly abundant in nature and has the features of environmental benignity and relatively high safety. Considering the abovementioned advantages, numerous studies have been reported on compounds of AMn₂O₄ (A = Co, Ni).

Diverse structures, such as nanoparticles,²²³ nanorods,²²⁴ nanofibers,²²⁵ hollow microcubes,²²⁶ multi-shelled hollow spheres,²²⁷ microspheres,^228–230 hierarchical nanostructures,^231,232 and multi-level nanotubes,²³³ have been prepared. A facial method of electrospinning followed by controlled heat treatment has been applied to the synthesis of shrinkable CoMn₂O₄ tube-in-tube nanotubes.²³³ As observed in Fig. 16a, the shrinkable tube-in-tube nanotubes are composed of inner tubes and outer tubes. The HR-TEM image in Fig. 16b indicates that this tube-in-tube architecture is composed of uniform nanoparticles (∼10 nm in diameter). The corresponding SAED pattern (inset of Fig. 16b) suggests their polycrystallinity. These multilevel nanotubes possess large specific surface areas, fast mass transport, good strain accommodation, and high packing density. Benefiting from these advanced structural features, the CoMn₂O₄ tube-in-tube nanotubes deliver a high discharge capacity of ∼565 mA h g⁻¹ at an extremely high current density of 2.0 A g⁻¹ (Fig. 16c) and maintain a capacity of 89% after 500 cycles when tested as a LIB anode. Furthermore, carbon-coated CoMn₂O₄ triple-shelled hollow spheres have been fabricated by Lou's group (Fig. 16d).²²⁷ A specific capacity of 726.7 mA h g⁻¹ and a capacity retention of almost 100% after 200 cycles were achieved by their as-prepared electrode (Fig. 16e). This superior performance is attributed to the unique multi-shelled structure with a porous, hollow interior, in which the high surface area leads to an enhanced electrode–electrolyte contact area, the hollow interior provides additional voids to alleviate volume expansion, and the multi-shelled structure provides highly stable structural support to preserve the integrity of the electrodes. Kang et al. reported porous hierarchical NiMn₂O₄/C tremella-like nanostructures obtained through a simple solvothermal and calcination method.²³⁴ The TEM image in Fig. 16f confirms that NiMn₂O₄/C contains porous hierarchical tremella-like nanostructures. The SAED pattern (inset in Fig. 16f) presents its polycrystalline nature. For Li storage, the porous NiMn₂O₄/C nanostructures exhibit a superior specific capacity (2130 mA h g⁻¹ within 350 cycles at 1000 mA g⁻¹) as well as excellent long-term cycling performance. Ma et al.²³⁵ constructed Fe-doped NiMn₂O₄via a facile method of nanocasting followed by an annealing process. A stabilized mesoporous cubic phase was prepared by tailoring the metal precursor ratio; as a result, the tetragonal sites of NiMn₂O₄ were preferentially occupied by iron. The preparation procedure of Fe-doped NiMn₂O₄ is schematically illustrated in Fig. 16g. The procedure begins with the amination of commercial BPPO (bromomethylated poly(2,6-dimethyl-1,4-phenylene oxide)) grafted to the polymer surface with hydroxyl groups, followed by a modified breath figure (BF) method. Subsequently, functionalized BPPO as the template is mixed with cations (Fe²⁺, Ni²⁺, and Mn³⁺). Finally, the metal hydroxides are converted into metal oxides by oxidative decomposition of the template. The obtained Fe-doped NiMn₂O₄ exhibits a high reversible capacity of 620 mA h g⁻¹ after 250 cycles at 0.2 A g⁻¹, good rate capability up to 2.0 A g⁻¹, and an ultralong cycle life stability for 1200 cycles. This outstanding Li-storage performance can be attributed to the iron doping, which improves the electrical conductivity and structural robustness.

Fig. 16 (a and b) TEM images, inset in (b) SAED patterns, and (c) rate performance of shrinkable CoMn₂O₄ tube-in-tube nanotubes in LIBs. Reproduced with permission.²³³ Copyright 2016, Springer. (d) TEM image and (e) cycling performance and corresponding coulombic efficiency at a current density of 200 mA g⁻¹ of carbon-coated CoMn₂O₄ triple-shelled hollow spheres. Reproduced with permission.²²⁷ Copyright 2014, Wiley. (f) TEM image, inset in (f) SAED patterns of porous hierarchical tremella-like NiMn₂O₄/C. Reproduced with permission.²³⁴ Copyright 2015, RSC. (g) Schematic of the synthesis of Fe-doped NiMn₂O₄. Reproduced with permission.²³⁵ Copyright 2015, ACS.

3.1.2. AFe₂O₄ (A = Mn, Co, Ni, Cu). The spinel AFe₂O₄ series (A = Mn, Co, Ni, and Cu) has received an upsurge of interest due to their natural abundance, non-toxicity and cost efficiency. It is highly anticipated that the presence of other metal cations can effectively offset the drawbacks of single Fe-based oxides. Unfortunately, rapid capacity fading is still observed in AFe₂O₄. In order to improve the electrochemical performance of AFe₂O₄, micro-/nano-structured materials with diverse morphologies have been fabricated. AFe₂O₄ anodes with nanowires,²³⁶ nanoparticles,²³⁷ nanoplates,²³⁸ nanotubes,²³⁹ nanofibers,²⁴⁰ hollow structures,²⁴¹ porous ball-in-ball hollow spheres²⁴² and hierarchical structures^243,244 have been comprehensively researched and show enhanced electrochemical performance. For example, mesoporous CoFe₂O₄ nanowire arrays grown on flexible carbon fabric have been prepared through a simple method of a hydrothermal process followed by thermal treatment.²³⁶ The SEM image shown in Fig. 17a clearly demonstrates the morphology of the CoFe₂O₄ nanowire arrays. This nanostructure electrode not only facilitates fast electron transport and ion diffusion but also enlarges the electrode–electrolyte contact area. Accordingly, a high reversible capacity of 1185.75 mA h g⁻¹ at 0.2 A g⁻¹ and a superior rate capability are obtained. Furthermore, Xu and co-workers fabricated hierarchical porous ball-in-ball hollow spheres of CoFe₂O₄ and NiFe₂O₄.²⁴² The SEM and TEM images shown in Fig. 17b and c demonstrate that these porous ball-in-ball hollow CoFe₂O₄ spheres have an outer ball with a diameter of 200 to 500 nm and a small interior hollow ball with a diameter of about 150 nm that form interesting hierarchical structures. Moreover, numerous nanoparticles and pores are observed to be homogeneously distributed in the walls of the CoFe₂O₄ hollow spheres. The structure of NiFe₂O₄ is similar to that of CoFe₂O₄. The obtained CoFe₂O₄ and NiFe₂O₄ anodes both exhibit extraordinary rate performance and high reversible specific capacity owing to these unique structural features. Yan's group synthesized porous hollow CoFe₂O₄ and NiFe₂O₄ nanoarchitectures via a MOF self-sacrificial template strategy.²⁴¹ The general synthetic process of the spinel AFe₂O₄ (A = Co, Ni) hollow structures is illustrated schematically in Fig. 17d. Firstly, metal ion (A⁺) is mixed with Fe-based MOFs Fe₄(Fe(CN)₆)₃ (Prussian blue, PB). In the subsequent annealing process, A⁺ and the outer surface of PB react with oxygen, forming the AFe₂O₄ shell. When the inner core PB further diffuses out, small voids are generated and new oxide layers are formed. Finally, the cubic hollow structure forms after all the PB is consumed. The NiFe₂O₄ cubes possess a uniform size of around 1 μm; the surface of the cube is rough because it is composed of many nanocrystals (Fig. 17e). The TEM image presented in Fig. 17f validates that the wall of the hollow cube is around 250 nm in thickness. Because the nanoparticles in the hollow cubes offer an enlarged electrode/electrolyte contact area and supply short ion diffusion distances, NiFe₂O₄ and CoFe₂O₄ exhibit high capacities of 636 and 380 mA h g⁻¹ at a high current density of 8.0 A g⁻¹, respectively. Additionally, NiFe₂O₄ exhibits high specific capacity and superior rate performance.

Fig. 17 (a) SEM image of CoFe₂O₄ nanowire arrays. Reproduced with permission.²³⁶ Copyright 2015, RSC. (b) SEM and (c) TEM images of hierarchical porous ball-in-ball hollow spheres of CoFe₂O₄. Reproduced with permission.²⁴² Copyright 2014, Springer. (d) Synthetic process of the spinel AFe₂O₄ (A = Co, Ni) hollow structures. (e) SEM and (f) TEM images of NiFe₂O₄ cubes. Reproduced with permission.²⁴¹ Copyright 2015, ACS. (g) SEM image of CoFe₂O₄/graphene nanocomposites. Reproduced with permission.²⁵¹ Copyright 2014, Elsevier.

In order to enhance the conductivity of electrodes, conductive materials such as graphene,^{243–248,251} CNT²⁴⁹ and carbon nanofiber²⁵⁰ have been combined with AFe₂O₄. Mesoporous CoFe₂O₄ nanospheres cross-linked with CNTs (CFO/CNT) have been prepared by a feasible one-pot solvothermal method.²⁴⁹ Firstly, the CNTs, as common metal ion adsorbents, adsorb Co²⁺ and Fe²⁺. Subsequently, these metal ions accumulate on the surface of the CNTs with increasing autoclave temperature. As a result, small crystals are obtained and then aggregate into CFO nanoparticles, which are further assembled into CFO nanospheres. With stable electronic and ionic transfer channels and short Li⁺ ion diffusion lengths supplied by the CFO/CNT composites, they manifest excellent rate performance. Li et al. prepared CoFe₂O₄/graphene nanocomposites via a facile one-step method, as shown in Fig. 17g.²⁵¹ It is noteworthy that almost all CoFe₂O₄ nanoparticles can be encapsulated within graphene sheets, which efficiently prevents aggregation of the particles and avoids direct contact between the CoFe₂O₄ nanoparticles and the electrolyte. Hence, CoFe₂O₄/graphene delivers enhanced rate capability and long cycling stability.

In contrast to the numerous studies on NiFe₂O₄ and CoFe₂O₄, there are fewer reports on MnFe₂O₄ and CuFe₂O₄. The research on CuFe₂O₄ mainly focuses on the preparation of nanostructures as well as some hybrid materials, such as nanofibers,^252,253 hollow spheres,²⁵⁴ CoFe₂O₄/graphene,²⁵⁵ and polypyrrole-coated CuFe₂O₄.²⁵⁶ Hollow CuFe₂O₄ spheres encapsulated in carbon shells have been reported by Jin et al.²⁵⁴ When used as an anode for LIBs, the electrode based on hollow CuFe₂O₄ spheres encapsulated in carbon shells exhibits high specific capacity, good cycling stability, and excellent rate performance. Even after 70 cycles, a high specific capacity of 550 mA h g⁻¹ is retained. Zhao et al. have fabricated CuFe₂O₄ hollow fibers by a simple electrospinning method followed by heat treatment.²⁵³ Qian's group prepared porous MnFe₂O₄ microrods as a promising LIB anode.²⁵⁷ The microrods are composed of many interconnected nanoparticles, forming a porous structure. Due to this unique structure, the porous MnFe₂O₄ electrode exhibits outstanding cycling stability (a capacity of 630 mA h g⁻¹ is retained after 1000 cycles at 1.0 A g⁻¹) as well as extraordinary rate performance.

3.1.3. ACo₂O₄ (A = Mn, Fe, Ni, and Cu). ACo₂O₄ (Mn, Fe, Ni, and Cu) type binary metal oxides originate from one atom replacing the Co in Co₃O₄, which possesses an isostructural nature of Co₃O₄, followed by a conversion reaction with lithium. Various micro-/nano-structures and hybrids combining NiCo₂O₄ with conductive materials such as hollow microspheres,^258,259 nanosheets,^260,261 nanoneedles,²⁶² nanocubes,²⁶³ rGO/NiCo₂O₄,^264,265 NiCo₂O₄ nanowires/carbon textiles,²⁶⁶ NiCo₂O₄/CNT,²⁶⁷ and NiCo₂O₄/CMK-3²⁶⁸ have been reported. A simple method involving low-temperature solutions and thermal treatment has been developed to grow hierarchical NiCo₂O₄ nanosheets on rGO by Lou's group;²⁶⁵ the two-step synthesis procedure is schematically illustrated in Fig. 18a. The TEM image of the rGO/NiCo₂O₄ nanocomposite presented in Fig. 18b reveals its high porosity. This hybrid nanostructure can not only effectively improve the charge transport but also buffer volume variations during ultralong discharge/charge cycles. As a result, the rGO/NiCo₂O₄ nanocomposites deliver high reversible capacity (954.3 at 0.2 A g⁻¹ after 50 cycles) as well as remarkable rate performance (Fig. 18c). Additionally, they fabricated NiCo₂O₄ complex hollow structures with three-layer core-in-double-shell hollow spheres.²⁵⁸ This complicated structure can efficiently enlarge the surface area, tolerate volume changes, and improve the structural integrity of the electrode. As a consequence, these NiCo₂O₄ hollow sphere anodes exhibit superior Li storage performance (the specific capacities are 834 and 533 mA h g⁻¹ at current densities of 0.3 and 2.0 A g⁻¹, respectively). Additionally, mesoporous NiCo₂O₄ nanowire arrays grown on carbon textiles were fabricated by a simple surfactant-assisted hydrothermal method combined with a subsequent annealing treatment.²⁶⁶Fig. 18d indicates the structural features of the NiCo₂O₄ nanowire arrays. It is worth mentioning that the morphology and array features of the nanowires can benefit the penetration of the electrolyte. The TEM image in Fig. 18e further reveals that the typical NiCo₂O₄ nanowires are actually composed of small nanoparticles with diameters of 10 to 20 nm and possess numerous pores. Benefiting from this unique structure, a NiCo₂O₄/carbon textile binder-free electrode demonstrates a high specific capacity (854 mA h g⁻¹ after 100 cycles).

Fig. 18 (a) The synthesis procedure, (b) TEM image, and (c) rate performance of rGo/NiCo₂O₄. Reproduced with permission.²⁶⁵ Copyright 2014, Wiley. (d) SEM and (e) TEM images of NiCo₂O₄/carbon textiles. Reproduced with permission.²⁶⁶ Copyright 2014, Wiley. (f) SEM image of MnCo₂O₄ microspheres. Reproduced with permission.²²⁹ Copyright 2013, RSC. (g) SEM image of FeCo₂O₄ nanoneedles array. Reproduced with permission.²⁷⁹ Copyright 2016, ACS.

In addition, MnCo₂O₄ has aroused a great deal of interest. Different morphologies of MnCo₂O₄ have been fabricated, such as nanoparticles,²⁶⁹ mesoporous spheres,^{228–230,270,271} nanosheets,²⁷² nanoflakes,²⁷³ core–shell structures,²⁷⁴ and MnCo₂O₄/PPy.²⁷⁵ Qian's group synthesized porous MnCo₂O₄ quasi-hollow spheres for Li storage.²²⁹ The hollow interiors of MnCo₂O₄ microspheres are depicted in the SEM image shown in Fig. 18f. It should be noted that the walls of the hollow structures are extremely thick. When evaluated as an anode for LIBs, the MnCo₂O₄ electrode maintains a stable capacity of 610 mA h g⁻¹ at 0.4 A g⁻¹ after 100 cycles.

Recently, porous CuCo₂O₄ nanocubes,²⁷⁶ hollow FeCo₂O₄ nanospheres,²⁷⁷ FeCo₂O₄ nanoflakes,²⁷⁸ porous FeCo₂O₄ nanoneedle arrays on nickel foam,²⁷⁹ and self-assembled mesoporous FeCo₂O₄²⁸⁰ were also investigated. Liu et al.²⁷⁹ have fabricated porous FeCo₂O₄ monocrystalline nano-needle arrays on nickel foam by a hydrothermal technique combined with subsequent calcination. The SEM image presented in Fig. 18g reveals that the FeCo₂O₄ nanoneedle arrays present an open framework. Owing to the merits of effective conductive transport paths, shorter ion transfer distances, and binder-free electrodes by direct growth on nickel foam, the obtained FeCo₂O₄ nanoneedle arrays display a high specific capacity of 1962 mA h g⁻¹ at 0.1 A g⁻¹ and superior rate capability (a capacity of 875 mA h g⁻¹ at 2.0 A g⁻¹).

3.2. ABO₄ (AMoO₄, A = Mn, Fe, Co, Ni)

ABO₄ type binary metal oxides are mainly Mo-containing mixed oxides (MnMoO₄,²⁸⁴ FeMoO₄,^281,285 CoMoO₄,^{282,286–291} and NiMoO₄^{283,292–294}). They are attractive for their high oxidation state as ‘‘hosts’’ for reacting with lithium, delivering 8 mol electrons per mol of material and achieving a high theoretical capacity of about 1000 mA h g⁻¹. Guan and coworkers prepared MnMoO₄ and coated it with carbon by a hydrothermal treatment.²⁸⁴ Their MnMoO₄@C composite anode achieved a high reversible capacity of 1050 mA h g⁻¹ even after 200 cycles at 0.1 A g⁻¹ with the reversible reactions between the Mo and Mn nanoparticles and MnO and MoO₃. Monodisperse FeMoO₄ nanocubes with sizes of about 100 nm were prepared by Qian's group via a one-pot hydrothermal method, as depicted in Fig. 19a.²⁸¹ They observed the morphology changes during the growth of FeMoO₄ nanocubes and found that the defined single-crystalline cubes were self-aggregated by nanosheets. They studied the Li storage behavior in FeMoO₄ (Fig. 19b) and proposed the electrochemical reactions as follows:

FeMoO₄ + xLi⁺ + xe⁻ → Li_xFeMoO₄ (24)

Li_xFeMoO₄ + (8 − x)Li⁺ + (8 − x)e⁻ → Fe⁰ + Mo⁰ + 4Li₂O (25)

Fe⁰ + Li₂O ↔ FeO + 2e⁻ + 2Li⁺ (26)

Mo⁰ + 3Li₂O ↔ MoO₃ + 6e⁻ + 6Li⁺ (27)

MoO₃ + yLi⁺ + ye⁻ → Li_yMoO₃ (28)

Li_yMoO₃ + (6 − y)Li⁺ + (6 − y)e⁻ → Mo + 3Li₂O (29)

Fig. 19 TEM image (a) and CV curves (b) of FeMoO₄ nanocubes; reproduced with permission.²⁸¹ Copyright 2015, Wiley. SEM image (c) and schematic of the lithiation–delithiation process (d) in porous MoO₃−CoMoO₄ hybrid microspheres (P-Mo–Co-HMSs); reproduced with permission.²⁸² Copyright 2016, ACS. TEM image (e) and CV curves (f) of the NiMoO₄@3DGF composites. Reproduced with permission.²⁸³ Copyright 2015, RSC.

Electrochemical characterization suggested that the FeMoO₄ nanocubes anode achieves a high reversible capacity of 926 mA h g⁻¹ at 0.1 A g⁻¹ as well as excellent rate performance. A FeMoO₄ nanorods anode was also prepared and exhibited a high reversible capacity of 1110 mA h g⁻¹ at 1 C (1 C = 992.3 mA g⁻¹) after 1000 cycles.²⁸⁵ Its excellent electrochemical performance indicates that FeMoO₄ is a promising LIB anode material.

CoMoO₄ has a high theoretical specific capacity of 980 mA h g⁻¹ when used as a LIB anode material; however, its low electronic conductivity and volume changes during cycling result in rapid capacity loss and inferior rate capability, hampering its practical applications. Ajayan and coworkers fabricated nanosized CoMoO₄ particles of 3 to 5 nm and dispersed them on reduced graphene oxide (rGO).²⁹¹ Yang et al. prepared a hybrid composite of CoMoO₄ nanorods and rGO.²⁸⁸ Using an electrospinning method,²⁹⁵ lotus root-like CoMoO₄@graphene nanofibers have been prepared.²⁸⁶ The high surface area and multi-level porous structure of rGO not only improves the conductivity of the CoMoO₄/rGO composites but also shortens the Li⁺ diffusion distance during cycling, endowing the hybrid materials with good electrochemical performance. Chen et al. constructed 3D hierarchical CoMoO₄/polypyrrole (PPy) core–shell nanowire (NW) arrays on conductive and flexible carbon cloth and evaluated them as binder-free LIB anodes for half-cells and full cells.²⁸⁹ The structural integrity and mechanical strength of the binder-free electrode were guaranteed by the direct growth method. The outside PPy layers not only enhanced the conductivity of the electrode but also effectively accommodated the volume expansion of the CoMoO₄ NWs during cycling. Due to its well-designed nanostructure, their CoMoO₄/PPy hybrid anode achieved a stable reversible capacity of 764 mA h g⁻¹ after 1000 cycles at 1.2 A g⁻¹. When coupled with commercial cathode LiFePO₄, the full cell using CoMoO₄/PPy hybrid as the anode achieved a high cycling stability (stable capacity was exhibited at 0.4 A g⁻¹ even after 1000 cycles).

Recently, porous MoO₃−CoMoO₄ hybrid microspheres (P-Mo−Co-HMSs) with homogeneously dispersed MoO₃ and CoMoO₄ subunits were synthesized by Wang et al.²⁹² As shown in Fig. 19c, the surface of the microspheres possesses an obvious pore structure, which is particularly beneficial for creating more active sites, enlarging the contact area of the active material with the electrolyte, and shortening the Li ion diffusion distance. Benefiting from these appealing structural and component features, the P-Mo−Co-HMSs anode exhibits excellent electrochemical performance, such as high capacity (1598.7 mA h g⁻¹ after 100 cycles at 0.2 A g⁻¹), superior cycling stability, and excellent rate capability (640 mA h g⁻¹ at 3.0 A g⁻¹). The introduction of CoMoO₄ is believed to be responsible for the superior lithium storage performance. As depicted in Fig. 19d, the in situ formed Co nanoclusters from CoMoO₄ not only promoted the reversible formation and decomposition of Li₂O but also functioned as an excellent buffer matrix for the MoO₃ electrode, preventing the aggregation of Mo nanograins and buffering volume changes and structural stress of the electrode during cycling. Thus, excellent electrochemical performance was achieved by the composite electrode.

NiMoO₄ has been investigated as anode material for LIBs. The two main polymorphs, α-NiMoO₄ and β-NiMoO₄, are both monoclinic but have different Mo coordinations. Mo⁶⁺ ions occupy octahedral sites in α phase and tetrahedral sites in β phase.²⁹⁶ The α phase is stable at room temperature, while the β phase begins to crystallize between 550 °C and 670 °C; it is difficult to prepare pure β phase. Ahn et al. prepared phase-pure β-NiMoO₄ yolk–shell spheres using nickel nitrate, ammonium molybdate, and sucrose as the starting material by a one-pot spray pyrolysis method.²⁹⁴ Amorphous dense spheres and α/β-NiMoO₄ yolk–shell spheres were also prepared by controlling the post-annealing temperature. Their results suggested that the electrochemical performance of the phase-pure β-NiMoO₄ yolk–shell spheres is significantly superior to those of other samples. Honeycomb-like NiMoO₄ ultrathin nanosheet arrays on Ni foam²⁹³ were fabricated and evaluated as binder-free LIB anodes; a high capacity of 680 mA h g⁻¹ was obtained at 1.0 A g⁻¹ after 200 cycles. Wang et al. constructed 3D NiMoO₄ nanowire arrays²⁹² and network-like porous NiMoO₄ nanoarchitectures (NiMoO₄@3DGF)²⁸³ on a 3D graphene network combining a chemical vapor deposition (CVD) method and a hydrothermal route. As the TEM image in Fig. 19e depicts, the interconnected network-like nanostructures of the porous NiMoO₄ nanoarchitectures consist of ultrathin crooked nanosheets. The adjacent nanosheets are self-assembled into continuous, interconnected porous architectures. When these nanoarchitectures were evaluated as a LIB anode, two obvious reduction broad peaks appeared at 0.33 and 1.09 V as Li intercalation proceeded in the NiMoO₄@3DGF anode, as shown in Fig. 19f. These two peaks can be assigned to the reduction of NiMoO₄ to metallic Ni and Mo, accompanying the formation of the SEI layer. The anodic peaks at 1.38 and 1.7 V are attributed to the oxidation of Mo and Ni nanograins to Mo⁶⁺ and Ni²⁺, respectively. In subsequent cycles, two pairs of redox peaks were induced by the reversible reactions of MoO₃ and NiO between Mo and Ni metal. The Li storage behavior in the as-prepared NiMoO₄@3DGF composite anode can be expressed as follows:

NiMoO₄ + 8Li⁺ + 8e⁻ → Ni + Mo + 4Li₂O (30)

Ni + Li₂O → NiO + 2Li⁺ + 2e⁻ (31)

Mo + 3Li₂O → MoO₃ + 6Li⁺ + 6e⁻ (32)

Ni + Mo + 4Li₂O ↔ NiO + MoO₃ + 8Li⁺ + 8e⁻ (33)

3.3. Other types of binary metal oxides (ABO₂, ABO₃, and A₂B₃O₈)

In addition to AB₂O₄ and ABO₄ types, other binary metal oxides have been studied as LIB anodes, such as ABO₂ type (CoNiO₂²⁹⁷ and CuFeO₂^298,299), ABO₃ type (FeMnO₃^300,301), and A₂B₃O₈ type (Fe₂Mo₃O₈,³⁰² Mn₂Mo₃O₈,^303,304 and Co₂Mo₃O₈,^305,306). Suffering from low conductivity and volume changes, electrodes with various hierarchical structures combined with carbon-based materials have been rationally designed and synthesized, achieving high reversible capacities and long cycling stabilities. The electrochemical performances of these materials are summarized in Table 1. Among these, the lithiation and delithiation products of Mn₂Mo₃O₈ are ambiguous, although a high capacity was achieved. As reported by Chowdari,^303,307 only Mo in Mn₂Mo₃O₈ can be reduced to Mo metal, while the Mn remains as MnO; this is different from the mechanism of other binary metal oxides. This is interesting and worthy of clarification.

Table 1 Electrochemical properties of nanostructured mixed transition metal oxides

Materials	First discharge/charge capacity (mA h g^−1)	Capacity (mA h g^−1)/cycle number, (current density, A g^−1)	Capacity (mA h g^−1)/current density (A g^−1)	Ref.
Mesoporous CoNiO2	941.3/617.1	449.3/50th (0.1)	66/5	297
CuFeO2/G composites	1571/944	670/100th (0.14)	440/0.7	298
FeMnO3 particles	—/—	984/500th (1.0)	282/8.0	300
MnFeO3/MWCNT	1524/1049	840/50th (0.2)	410/10.0	301
Fe2Mo3O8–rGO	1275.2/923.5	835/40th (0.2)	574.8/3.0	302
Mn2Mo3O8–graphene	921/—	950/40th (0.2)	671/1.5	304
Co2Mo3O8/rGO	1834/1170	954/60th (0.06)	503/1.0	305

---

4. Conversion-alloying metal oxides

Conversion-alloying metal oxides, mainly including SnO₂, Sn₃O₄, GeO₂, and ZnO, are achieved in two steps of conversion and alloying reactions when reacting with lithium.

SnO₂ is an outstandingly attractive LIB anode candidate due to its abundance, environmental benignity, and high theoretical capacity. The key obstacle encountered by SnO₂ anodes is the large volume change (≈358%) during the charge/discharge process, which leads to poor cycling stability and rate performance. It is widely believed that the lithium storage mechanism of SnO₂ anodes involves two steps, as follows:

SnO₂ + 4Li⁺ + 4e⁻ → Sn + 4Li₂O (first lithiation) (34)

Sn + 4.4Li⁺ + 4.4e⁻ ↔ Li_4.4Sn (35)

In this mechanism, the first step is irreversible and the second step is reversible.^308–310 From this standpoint, SnO₂ anode delivers a theoretical specific capacity of 780 mA h g⁻¹ by transferring 4.4 moles of Li ions per mole of Sn. However, some researchers consider that both steps are reversible when SnO₂ possesses a nanoscale structure.^311–313 In this case, the theoretical specific capacity for SnO₂ is 1494 mA h g⁻¹ (transferring 8.4 moles of Li ions per mole of SnO₂). Hu et al. insisted that suppressing in situ formed coarse Sn, constructing numerous Sn/Li₂O interfaces, and maintaining durable nanostructures in the electrode are greatly significant to obtain highly reversible SnO₂ electrodes.³¹³ In order to confirm this point, they synthesized a thin-film SnO₂ electrode with large numbers of interfaces and grain boundaries by a magnetron sputtering method. Aided by XRD, electron energy loss spectroscopy (EELS), XPS and in situ surface-enhanced Raman scattering (SERS), they concluded that all the metallic Sn in the lithiated film electrode was completely re-oxidized to amorphous SnO₂. This work confirms that the particle size of SnO₂ influences the reversibility of the conversion reaction. Therefore, researchers have concentrated on decreasing the size of the SnO₂ particles,^311,314,315 constructing nanostructured SnO₂^316–318 and preparing SnO₂-conductive material hybrids.^{312,319–322} Wang and co-workers prepared N-doped SnO₂ nanoparticles through a laser-assisted pyrolysis method.³¹¹ The unreactive Sn–N bonds and small particle sizes resulted in exceptional electrochemical performance (a reversible capacity of 1192 mA h g⁻¹ after 500 cycles at 1.4 A g⁻¹). Polydopamine-coated SnO₂ nanocrystals were synthesized by a polymer-templated method.³¹² The establishment of a passivating SEI layer resulted from the introduction of PDA; also, the corn-like nanostructure contributed to the outstanding rate capability and superior long-term stability. Hollow structure materials are widely used in LIBs due to their hollow interiors, which supply additional volume to alleviate structural strain, contributing to superior electrochemical performance. Uniform multi-shelled SnO₂ hollow microspheres were synthesized by Wang's group via a sequential hard template method.³¹⁰ They prepared single-, double-, triple-, and quadruple-shelled SnO₂. Fig. 20a and b display the SEM images of triple- and quadruple-shelled SnO₂, respectively. Fig. 20c shows the CV curves of the multi-shelled SnO₂ hollow microsphere in the first three cycles. In the first cycle, an intensive cathodic peak around 0.73 V is associated with the reduction of SnO₂ to Sn and the formation of SEI layers. The peak near 0.15 V is ascribed to the formation of Li_xSn. The peaks at about 0.1 V and 0.53 V that appeared in the anodic process are attributed to lithium dealloying from Li_xSn. The weak oxidation peaks at 1.25 V and 1.81 V may be due to the reversible reaction from Sn to SnO₂. In the subsequent cycles, the cathodic peaks at 0.92 V and 1.17 V, together with the anodic peaks at 1.25 V and 1.81 V, correspond to the partially reversible reaction between Sn and SnO₂. In virtue of the sufficient void space offered by the multi-shelled structure, the volume change during cycling could be effectively buffered and more lithium storage sites are exposed. Hence, the multi-shelled SnO₂ hollow microsphere anode demonstrates superior rate capacity, good cycling performance, and excellent structural stability.

Fig. 20 TEM images of triple-shelled SnO₂ (a) and quadruple-shelled SnO₂ (b); (c) cyclic voltammetry profiles of quadruple-shelled SnO₂ hollow microspheres at a scan rate of 2 mV s⁻¹. Reproduced with permission.³¹⁰ Copyright 2016, RSC. (d) SEM, (e) TEM and inset in (e) SAED pattern of SnO₂@N-CNF; (f) cycling performance of SnO₂@N-CNF and commercial SnO₂ nanoparticles. Reproduced with permission.³¹⁴ Copyright 2016, Wiley. (g) SEM image, (h) TEM image and (i) rate performance of hierarchical flower-like Sn₃O₄. Reproduced with permission.³²⁵ Copyright 2015, Elsevier.

Fabricating nanostructures in carbon matrices can effectively restrain the aggregation of SnO₂ particles and improve the conductivity.^{314,321–323} Qiao et al.³¹⁴ encapsulated SnO₂ nanoparticles in flexible nitrogen-doped carbon nanofiber films (denoted as SnO₂@N-CNF) by electrospinning. The fibrous morphology of SnO₂@N-CNF can be observed in Fig. 20d. Furthermore, the TEM image shown in Fig. 20e clearly indicates that SnO₂ nanoparticles with sizes of about 7.5 nm are well dispersed in the carbon nanofibers. Extraordinary rate performance and long-term cycling stability were obtained, which can be attributed to the unique nanostructure of SnO₂. As shown in Fig. 20f, the cycling performance of SnO₂@N-CNF is much better than that of commercial SnO₂ nanoparticles. Specifically, graphene, an attractive conductive material, is usually used to improve the conductivity of SnO₂ by forming SnO₂/graphene.^{315,319,320,324} SnO₂ quantum dots (QDs) supported by graphene nanosheets (GNSs) were fabricated via a hydrothermal method and electron-beam irradiation (EBI) strategies.³¹⁹ The EBI technique can induce the exfoliation of GNSs and increase their interlayer spacing, leading to increases in GNS amorphization, disorder, and defects. The ultrasmall SnO₂ quantum dots (3 to 5.5 nm) and the addition of graphene are beneficial to the utilization and conductivity of the active material.

There are few studies of Sn₃O₄ as a LIB anode. Chen et al.³²⁵ reported the use of hierarchical flower-like Sn₃O₄ as a LIB anode material. The flower-like Sn₃O₄ was assembled with single-crystalline Sn₃O₄ nanosheets, which can be clearly observed in the SEM and TEM images shown in Fig. 20g and h, respectively. The unique structure possesses a large inner space and a high surface area, which not only facilitate liquid electrolyte diffusion but also provide abundant room to retain volume expansion. A reversible capacity of 505.2 mA h g⁻¹ was obtained after 50 cycles. Additionally, superior rate performance was also obtained, as shown in Fig. 20i. The electrochemical reactions may proceed as follows:

Sn₃O₄ + 8Li⁺ + 8e⁻ → 3Sn + 4Li₂O (first lithiation) (36)

Sn + xLi⁺ + xe⁻ ↔ Li_xSn (x ≤ 4.4) (37)

Germanium oxide (GeO₂) is another conversion-alloying anode material whose Li storage behavior resembles that of SnO₂. Owing to its high theoretical capacity (1125 mA h g⁻¹), low operation voltage (0.7 V to 0 V vs. Li⁺/Li) and higher thermal stability, GeO₂ anode has attracted the attention of researchers.^332–337 For example, Guo's group synthesized Ge/GeO₂/C by calcinating GeO₂ in C₂H₂/Ar at 650 °C.³³⁵ When used as a LIB anode, the kinetics of the conversion reaction of the as-prepared GeO₂/Ge/C anode was largely activated and improved by the combination of fine nanparticles, carbon coating, and the elemental Ge in the composite. As a consequence, the Ge/GeO₂/C nanocomposites anode delivered a capacity of 1680 mA h g⁻¹ at 10C (C = 2.1 A g⁻¹) rate, demonstrating high capacity and good rate capability.

As an anode material, ZnO has a high theoretical capacity (987 mA h g⁻¹) and a higher lithium-ion diffusion coefficient.³²⁶ However, it usually demonstrates poor electrical conductivity and large volume variations during cycling, leading to limited rate performance. The electrochemical reactions of ZnO electrode are proposed as follows:

ZnO + 2Li⁺ + 2e⁻ ↔ Zn + Li₂O (38)

Zn + xLi⁺ + xe⁻ ↔ Li_xZn (x ≤ 1) (39)

Recently, a great deal of progress on ZnO anodes has been reported.^326–331 Among these, ZnO@ZnO quantum dots/C core–shell nanorod arrays (ZnO@ZnO QDs/C NRAs) grown on carbon cloth as an anode for LIBs have been synthesized by Zhang et al.³²⁶ Benefiting from the shortened Li⁺ diffusion distance, superior conductivity, and good structural stability of the nanorod arrays, ZnO@ZnO QDs/C NRAs manifest a remarkable rate capability and excellent cycling performance.

5. Summary and outlook

We have summarized recent progress in the use of conversion reaction metal oxides as LIB anodes. The typical characteristics of the materials, including theoretical capacity, cathodic voltage, anodic voltage, and their formed products are listed in Tables 2 and 3 according to the consensus of reported studies. Obviously, the usage of conversion reaction metal oxides is promising to obtain advanced Li-ion batteries with large capacities. Unfortunately, several obstacles remain to be overcome before this promise can become reality. Concentrating on the low electronic conductivity and volume expansion, various micro/nanostructure electrodes are currently being rationally designed by combining two or more of the three approaches mentioned in the introduction; these electrodes exhibit satisfied reversible capacity and cycling stability. It should be pointed out that currently, the reversible specific capacities of the electrodes are mostly calculated based on the mass loading of the active materials, ignoring their volume-specific capacities. It is well known that nanoscale powder materials have relatively low packing densities, which severely lowers the volumetric capacities of electrodes. Assuming that the capacity and tapping density of the graphite anode are 372 mA h g⁻¹ and 2.0 g cm⁻³, respectively, the volume specific capacity of graphite is calculated to be 744 mA h cm⁻³. This means that only oxides with tapping densities of at least 1.0 g cm⁻³ and specific capacities of 744 mA h g⁻¹ can compete with graphite anode. Therefore, it is necessary to consider tapping density during the design and synthesis of oxide anodes. Additionally, the voltage hysteresis observed between the discharge and charge plateaus is also always overlooked, which severely diminishes the electrode round-trip efficiency. Unfortunately, the origins of the hysteresis are ambiguous, and the controversy regarding traditional kinetic and thermodynamics factors remains. A similar situation troubles Li–air batteries. The cleavage of stable Li–O bonds is a common factor between Li-ion batteries based on conversion reaction metal oxides and Li–air systems; the rational design and incorporation of catalysts may be an alternative to address the voltage hysteresis observed in these electrodes.

Table 2 Typical characteristics of conversion reaction single metal oxide anodes

Material	Theoretical capacity (mA h g^−1)	Reduction potential (V, vs. Li^+/Li)	Reduction products	Oxidation potential (V, vs. Li^+/Li)	Oxidation products	Ref.
Cr2O3	1058	0.7	Cr	1.1	Cr2O3	176
MnO	755	0.2–0.4	Mn	1.2–1.3	MnO	56, 62 and 338
Mn3O4	936	0.2–0.4	Mn	1.2–1.4	Mn3O4	63, 339 and 340
Mn2O3	1018	0.2–0.4	Mn	1.2–1.4	MnO, Mn3O4	26, 44 and 45
Fe2O3	1007	0.8–1.0	Fe	1.7–1.9	Fe2O3	96 and 341
Fe3O4	924	0.8–1.0	Fe	1.6–1.9	FeO, Fe3O4	74, 92 and 342
CoO	716	1.1–1.5	Co	1.9–2.2	CoO	99, 343 and 344
Co3O4	954	0.9–1.4	Co	2.0–2.3	CoO, Co3O4	106, 111, 112 and 345
NiO	718	0.9–1.2	Ni	2.–1–2.3	NiO	139, 140 and 346
Cu2O	375	1.0–1.1	Cu	2.3–2.5	Cu2O	164 and 347
CuO	674	1.17, 0.72	Cu2O, Cu	2.4–2.5	CuO	143 and 154
MoO2	837	1.55, 1.26, 0.29	LiMoO2, Mo	1.47, 1.72	MoO2	196 and 199
MoO3	1117	2.3, 0.3	LixMoO2, Mo	1.5	MoO3	191 and 348
WO3	693	0.6–0.9	W	1.2–1.3	WO3	208 and 220

---

Table 3 Typical characteristics of conversion reaction binary metal oxide anodes

Material	Theoretical capacity (mA h g^−1)	Reduction potential (V, vs. Li^+/Li)	Reduction products	Oxidation potential (V, vs. Li^+/Li)	Oxidation products	Ref.
AMn2O4, A =
Co	921	0.47, 1.25	Mn, Co	1.44, 2.03	MnO, CoO	225 and 226
Ni	922	0.42, 1.13	Mn, Ni	1.32, 1.98	MnO/Mn3O4, NiO	234 and 235
AFe2O4, A =
Mn	930	0.76	Mn, Fe	1.67–2.02	MnO, Fe3O4	257
Co	914	0.75	Co, Fe	1.6–1.9	CoO, Fe2O3	244
Ni	915	0.84, 0.98	Fe, Ni	1.68	NiO, Fe2O3	241 and 349
Cu	897	0.88	Fe, Cu	1.7–1.9	CuO, Fe2O3	252 and 350
ACo2O4, A =
Mn	906	0.85	Mn, Co	2.27	MnO, CoO	229
Fe	902	0.72, 1.57	Fe, Co	1.68, 2.25	FeO, Co3O4	278 and 279
Ni	892	0.81–1.25	Ni, Co	1.5–1.7, 2.1–2.3	NiO, Co3O4	265 and 267
Cu	874	0.8–1.3	Cu, Co	2.0–2.3	CuO, Co3O4	351 and 352
AMoO4, A =
Mn	998	—	Mn, Mo	—	MnO, MoO3	284
Fe	994	0.67, 1.37	Fe, Mo	1.5, 2.0	FeO, MoO3	281
Co	980	0.64, 1.55	Co, Mo	1.4, 1.8	MoO3, CoO	286
Ni	980	1.0, 1.75	Ni, Mo	1.48, 2.25	MoO3, NiO	293
CoNiO2	717	1.06	Co, Ni	2.15	CoO, NiO	297
CuFeO2	709	1.15	Cu, Fe	1.68	Cu2O, Fe2O3	298
FeMnO3	1013	0.3, 0.8	Mn, Fe	1.2, 1.7	Mn3O4, Mn2O3, Fe2O3, Fe3O4	300

---

In addition, researchers are currently focusing on the side effects of conversion reactions, such as volume expansion, pulverization and structural collapse, ignoring their favorable aspects. The nanosize effect is a typical characteristic of conversion reactions; this effect further reduces the dimensions of active materials, enlarges the contact area between the active material and electrolyte, and shortens the Li⁺ diffusion distance. These features are beneficial to enhance the electrode kinetics. Recent progress on structural changes in conversion reaction anode materials (such as Co₃O₄ nanocubes, FeF₂ nanoparticles, RuO₂ nanowires, and NiO nanosheets) using in situ TEM confirm that a conductive network is reversibly generated in conversion reactions that can provide pathways for electron transfer.^{105,168,353,354} When the reduction and oxidation potentials of these metal oxides are analysed, it is clear that the redox potential of MnO_x is much lower than those of other oxides, such as FeO_x, CoO_x, and CuO_x. If MnO_x is mixed with other high redox potential oxides, two separate redox pairs will exist in the anode mixture. The earlier in situ formed metal network not only enables fast electron transfer but also functions as a matrix to prevent the aggregation of metal nanoparticles that form later during further lithiation processes. Upon delithiation, the metal network that was oxidated at high potential can play a dual role as a conductive network and matrix during the oxidation process of Mn metal. Meanwhile, the re-oxidated MnO_x can play the role of a matrix to guarantee good dispersion of the late-formed metal oxides in return. Therefore, a self-matrix and reversible in situ formed conductive network can be fabricated if the mixture oxides are small enough (1 to 5 nm) and have a highly homogeneous dispersion. However, it is difficult to form a thorough mixture on the nanoscale within a chemical or mechanical mixture.

Taking advantage of the nanosize effect of conversion reactions, good dispersion of two or more oxides can be achieved after lithiation–delithiation, which has been confirmed by the reported binary metal oxide anodes. Combining the reversible metal conductive network and the merits of nanosize effects, a self-matrix and conductive network can be fabricated by tailoring the components of the compounds, as illustrated by a reported FeMnO₃ anode.³⁰⁰ Other related studies have also remarked on this point recently.^19,281,282 More importantly, a reconstructed hierarchical structure with an in situ formed metal conductive network can be fabricated by controlling the lithiation or delithiation state. In these reformed structures, the small oxide nanoparticles and metal nanograins can function as active sites for catalysis. It is believed that the characteristics of the conversion reaction can not only be used to fabricate high performance electrode materials but can also be exploited in further applications, such as catalysts for the oxygen reduction reaction (ORR) and oxygen evolution reaction (OER).

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (51622102, 51231003, 51571124), the Ministry of Education of China (IRT-13R30), MOST(2016YFB0901502), and the 111 Project (B12015).

References

1. D. Larcher and J. M. Tarascon, Nat. Chem., 2015, 7, 19.
2. S. Chu, Y. Cui and N. Liu, Nat. Mater., 2017, 16, 16.
3. N. J. Dudney and J. Li, Science, 2015, 347, 131.
4. Y.-G. Guo, J.-S. Hu and L.-J. Wan, Adv. Mater., 2008, 20, 2878.
5. H. Wu, G. Yu, L. Pan, N. Liu, M. T. McDowell, Z. Bao and Y. Cui, Nat. Commun., 2013, 4, 1493.
6. P. Poizot, S. Laruelle, S. Grugeon, L. Dupont and J.-M. Tarascon, Nature, 2000, 407, 496.
7. X. Rui, H. Tan and Q. Yan, Nanoscale, 2014, 6, 9889.
8. F. Wang, S. W. Kim, D. H. Seo, K. Kang, L. Wang, D. Su, J. J. Vajo, J. Wang and J. Graetz, Nat. Commun., 2015, 6, 6668.
9. K. Zhang, X. Han, Z. Hu, X. Zhang, Z. Tao and J. Chen, Chem. Soc. Rev., 2015, 44, 699.
10. L. Zhang, H. B. Wu and X. W. Lou, Adv. Energy Mater., 2014, 4, 1300958.
11. D. Meggiolaro, G. Gigli, A. Paolone, P. Reale, M. L. Doublet and S. Brutti, J. Phys. Chem. C, 2015, 119, 17044.
12. R. Khatib, A. L. Dalverny, M. Saubanère, M. Gaberscek and M. L. Doublet, J. Phys. Chem. C, 2013, 117, 837.
13. J. Wang, H. Zhou, J. Nanda and P. V. Braun, Chem. Mater., 2015, 27, 2803.
14. M. R. Palacin, Chem. Soc. Rev., 2009, 38, 2565.
15. R. Raccichini, A. Varzi, S. Passerini and B. Scrosati, Nat. Mater., 2015, 14, 271.
16. Q. Wang, L. Jiao, H. Du, Y. Wang and H. Yuan, J. Power Sources, 2014, 245, 101.
17. L. Wen, F. Li and H. M. Cheng, Adv. Mater. Processes, 2016, 28, 4306.
18. J. Liu, X. Xu, R. Hu, L. Yang and M. Zhu, Adv. Energy Mater., 2016, 6, 1600256.
19. I. Sultana, M. M. Rahman, T. Ramireddy, N. Sharma, D. Poddar, A. Khalid, H. Z. Zhang, Y. Chen and A. M. Glushenkov, ACS Appl. Mater. Interfaces, 2015, 7, 20736.
20. M. R. Jo, J. Jung, G. H. Lee, Y. Kim, K. Song, J. Yang, J. S. Chae, K. C. Roh, Y. I. Kim, W. S. Yoon and Y. M. Kang, Nano Energy, 2016, 19, 246.
21. J. Qi, X. Lai, J. Wang, H. Tang, H. Ren, Y. Yang, Q. Jin, L. Zhang, R. Yu, G. Ma, Z. Su, H. Zhao and D. Wang, Chem. Soc. Rev., 2015, 44, 6749.
22. M. R. Lukatskaya, B. Dunn and Y. Gogotsi, Nat. Commun., 2016, 7, 12647.
23. J. Wang, H. Tang, H. Wang, R. Yu and D. Wang, Mater. Chem. Front., 2017, 1, 414.
24. B. Koo, H. Kim, Y. Cho, K. T. Lee, N. S. Choi and J. Cho, Angew. Chem., Int. Ed., 2012, 51, 8762.
25. F. Zhang and L. Qi, Adv. Sci., 2016, 3, 1600049.
26. K. Z. Cao, L. F. Jiao, H. Xu, H. Q. Liu, H. Y. Kang, Y. Zhao, Y. C. Liu, Y. J. Wang and H. T. Yuan, Adv. Sci., 2016, 3, 1500185.
27. Y. Deng, L. Wan, Y. Xie, X. Qin and G. Chen, RSC Adv., 2014, 4, 23914.
28. K. Zhong, X. Xia, B. Zhang, H. Li, Z. Wang and L. Chen, J. Power Sources, 2010, 195, 3300.
29. X. Fang, X. Lu, X. Guo, Y. Mao, Y.-S. Hu, J. Wang, Z. Wang, F. Wu, H. Liu and L. Chen, Electrochem. Commun., 2010, 12, 1520.
30. Y. Qiu, G.-L. Xu, K. Yan, H. Sun, J. Xiao, S. Yang, S.-G. Sun, L. Jin and H. Deng, J. Mater. Chem., 2011, 21, 6346.
31. Y. Zhang, Y. Yan, X. Wang, G. Li, D. Deng, L. Jiang, C. Shu and C. Wang, Chem. – Eur. J., 2014, 20, 6126.
32. S. Z. Huang, J. Jin, Y. Cai, Y. Li, Z. Deng, J. Y. Zeng, J. Liu, C. Wang, T. Hasan and B. L. Su, Sci. Rep., 2015, 5, 14686.
33. Y. Yuan, A. Nie, G. M. Odegard, R. Xu, D. Zhou, S. Santhanagopalan, K. He, H. Asayesh-Ardakani, D. D. Meng, R. F. Klie, C. Johnson, J. Lu and R. Shahbazian-Yassar, Nano Lett., 2015, 15, 2998.
34. Y. Yuan, C. Zhan, K. He, H. Chen, W. Yao, S. Sharifi-Asl, B. Song, Z. Yang, A. Nie, X. Luo, H. Wang, S. M. Wood, K. Amine, M. S. Islam, J. Lu and R. Shahbazian-Yassar, Nat. Commun., 2016, 7, 13374.
35. J. Qian, H. Jin, B. Chen, M. Lin, W. Lu, W. M. Tang, W. Xiong, L. Chan, S. P. Lau and J. Yuan, Angew. Chem., Int. Ed., 2015, 54, 6800.
36. S. Deng, L. Wang, T. Hou and Y. Li, J. Phys. Chem. C, 2015, 119, 28783.
37. X. Guo, J. Han, L. Zhang, P. Liu, A. Hirata, L. Chen, T. Fujita and M. Chen, Nanoscale, 2015, 7, 15111.
38. G. Wang, Y. Sun, D. Li, W. Wei, X. Feng and K. Muellen, Small, 2016, 12, 3914.
39. J.-Y. Liao, D. Higgins, G. Lui, V. Chabot, X. Xiao and Z. Chen, Nano Lett., 2013, 13, 5467.
40. Q. Sun, Z. J. Wang, Z. J. Zhang, Q. Yu, Y. Qu, J. Y. Zhang, Y. Yu and B. Xiang, ACS Appl. Mater. Interfaces, 2016, 8, 6303.
41. Y. H. Wang, Y. H. Wang, D. S. Jia, Z. Peng, Y. Y. Xia and G. F. Zheng, Nano Lett., 2014, 14, 1080.
42. X. Zhang, Y. Qian, Y. Zhu and K. Tang, Nanoscale, 2014, 6, 1725.
43. L. Chang, L. Mai, X. Xu, Q. An, Y. Zhao, D. Wang and X. Feng, RSC Adv., 2013, 3, 1947.
44. Z. C. Bai, Y. H. Zhang, Y. W. Zhang, C. L. Guo, B. Tang and D. Sun, J. Mater. Chem. A, 2015, 3, 5266.
45. R. Liu, S. Q. Zhao, M. M. Zhang, F. Feng and Q. Shen, Chem. Commun., 2015, 51, 5728.
46. Z. C. Bai, N. Fan, Z. C. Ju, C. L. Guo, Y. T. Qian, B. Tang and S. L. Xiong, J. Mater. Chem. A, 2013, 1, 10985.
47. J. Yue, X. Gu, L. Chen, N. Wang, X. Jiang, H. Xu, J. Yang and Y. Qian, J. Mater. Chem. A, 2014, 2, 17421.
48. G. Jian, Y. Xu, L.-C. Lai, C. Wang and M. R. Zachariah, J. Mater. Chem. A, 2014, 2, 4627.
49. Y. Qiao, Y. Yu, Y. Jin, Y.-B. Guan and C.-H. Chen, Electrochim. Acta, 2014, 132, 323.
50. Y. Xia, Z. Xiao, X. Dou, H. Huang, X. H. Lu, R. J. Yan, Y. P. Gan, W. J. Zhu, J. P. Tu, W. K. Zhang and X. Y. Tao, ACS Nano, 2013, 7, 7083.
51. F. X. Ma, H. Hu, H. B. Wu, C. Y. Xu, Z. Xu, L. Zhen and X. W. Lou, Adv. Mater., 2015, 27, 4097.
52. B. Wang, H. B. Wu, L. Zhang and X. W. Lou, Angew. Chem., Int. Ed., 2013, 52, 4165.
53. S. Xu, C. M. Hessel, H. Ren, R. Yu, Q. Jin, M. Yang, H. Zhao and D. Wang, Energy Environ. Sci., 2014, 7, 632.
54. Z. Cai, L. Xu, M. Yan, C. Han, L. He, K. M. Hercule, C. Niu, Z. Yuan, W. Xu, L. Qu, K. Zhao and L. Mai, Nano Lett., 2015, 15, 738.
55. H. Jiang, Y. Hu, S. Guo, C. Yan, P. S. Lee and C. Li, ACS Nano, 2014, 8, 6038.
56. D. M. Kang, Q. L. Liu, R. Si, J. J. Gu, W. Zhang and D. Zhang, Carbon, 2016, 99, 138.
57. S. Z. Huang, Y. Cai, J. Jin, J. Liu, Y. Li, Y. Yu, H. E. Wang, L. H. Chen and B. L. Su, Nano Energy, 2015, 12, 833.
58. W. Luo, X. Hu, Y. Sun and Y. Huang, ACS Appl. Mater. Interfaces, 2013, 5, 1997.
59. J. L. Liu, N. Chen and Q. M. Pan, J. Power Sources, 2015, 299, 265.
60. B. Liu, X. Hu, H. Xu, W. Luo, Y. Sun and Y. Huang, Sci. Rep., 2014, 4, 4229.
61. T. Z. Yuan, Y. Z. Jiang, W. P. Sun, B. Xiang, Y. Li, M. Yan, B. Xu and S. X. Dou, Adv. Funct. Mater., 2016, 26, 2198.
62. J. G. Wang, D. D. Jin, H. Y. Liu, C. B. Zhang, R. Zhou, C. Shen, K. Y. Xie and B. Q. Wei, Nano Energy, 2016, 22, 524.
63. J. G. Wang, D. D. Jin, R. Zhou, X. Li, X. R. Liu, C. Shen, K. Y. Xie, B. H. Li, F. Y. Kang and B. Q. Wei, ACS Nano, 2016, 10, 6227.
64. X. Zhang, H. H. Liu, S. Petnikota, S. Ramakrishna and H. J. Fan, J. Mater. Chem. A, 2014, 2, 10835.
65. A. Banerjee, V. Aravindan, S. Bhatnagar, D. Mhamane, S. Madhavi and S. Ogale, Nano Energy, 2013, 2, 890.
66. L. Zhang, H. B. Wu, R. Xu and X. W. Lou, CrystEngComm, 2013, 15, 9332.
67. H. Xiao, Y. Xia, W. Zhang, H. Huang, Y. Gan and X. Tao, J. Mater. Chem. A, 2013, 1, 2307.
68. S. H. Lee, S.-H. Yu, J. E. Lee, A. Jin, D. J. Lee, N. Lee, H. Jo, K. Shin, T.-Y. Ahn, Y.-W. Kim, H. Choe, Y.-E. Sung and T. Hyeon, Nano Lett., 2013, 13, 4249.
69. L. Chen, H. Y. Xu, L. Li, F. F. Wu, J. Yang and Y. T. Qian, J. Power Sources, 2014, 245, 429.
70. C. Lei, F. Han, Q. Sun, W. C. Li and A. H. Lu, Chem. – Eur. J., 2014, 20, 139.
71. J. L. Liu, W. W. Zhou, L. F. Lai, H. P. Yang, S. H. Lim, Y. D. Zhen, T. Yu, Z. X. Shen and J. Y. Lin, Nano Energy, 2013, 2, 726.
72. F. Han, D. Li, W. C. Li, C. Lei, Q. Sun and A. H. Lu, Adv. Funct. Mater., 2013, 23, 1692.
73. W. Huang, X. Xiao, C. Engelbrekt, M. Zhang, S. Li, J. Ulstrup, L. Ci, J. Feng, P. Si and Q. Chi, Mater. Chem. Front., 2017, 1, 1185.
74. Z. Liu, X.-Y. Yu and U. Paik, Adv. Energy Mater., 2016, 6, 1502318.
75. X. Jia, Y. Cheng, Y. Lu and F. Wei, ACS Nano, 2014, 8, 9265.
76. Z. Sun, E. Madej, A. Genc, M. Muhler, J. Arbiol, W. Schuhmann and E. Ventos, Chem. Commun., 2016, 52, 7348.
77. N. Yan, X. H. Zhou, Y. Li, F. Wang, H. Zhong, H. Wang and Q. W. Chen, Sci. Rep., 2013, 3, 3392.
78. W. J. Yu, L. L. Zhang, P. X. Hou, F. Li, C. Liu and H. M. Cheng, Adv. Energy Mater., 2016, 6, 1501755.
79. H. Y. Zhang, R. Z. Hu, H. Liu, W. Sun, Z. C. Lu, J. W. Liu, L. C. Yang, Y. Zhang and M. Zhu, J. Mater. Chem. A, 2016, 4, 10321.
80. X. D. Zhu, K. X. Wang, D. J. Yan, S. R. Le, R. J. Ma, K. N. Sun and Y. T. Liu, Chem. Commun., 2015, 51, 11888.
81. Y. Chen, B. Song, X. Tang, L. Lu and J. Xue, Small, 2014, 10, 1536.
82. Y. Jiang, Y. Li, W. Sun, W. Huang, J. Liu, B. Xu, C. Jin, T. Y. Ma, C. Wu and M. Yan, Energy Environ. Sci., 2015, 8, 1471.
83. M. M. Rahman, A. M. Glushenkov, T. Ramireddy, T. Tao and Y. Chen, Nanoscale, 2013, 5, 4910.
84. H. Xu, X. D. Zhu, K. N. Sun, Y. T. Liu and X. M. Xie, Adv. Mater. Interfaces, 2015, 2, 1500239.
85. D. Zhao, Y. Xiao, X. Wang, Q. Gao and M. H. Cao, Nano Energy, 2014, 7, 124.
86. F. X. Ma, P. P. Wang, C. Y. Xu, J. Yu, H. T. Fang and L. Zhen, J. Mater. Chem. A, 2014, 2, 19330.
87. G. Huang, L. L. Zhang, F. F. Zhang and L. M. Wang, Nanoscale, 2014, 6, 5509.
88. C. Chen, Y. Huang, C. An, H. Zhang, Y. Wang, L. Jiao and H. Yuan, ChemSusChem, 2015, 8, 114.
89. D. Z. Kong, C. W. Cheng, Y. Wang, B. Liu, Z. X. Huang and H. Y. Yang, J. Mater. Chem. A, 2016, 4, 11800.
90. Y. Huang, Z. X. Lin, M. B. Zheng, T. H. Wang, J. Z. Yang, F. S. Yuan, X. Y. Lu, L. Liu and D. P. Sun, J. Power Sources, 2016, 307, 649.
91. G. B. Zeng, N. Shi, M. Hess, X. Chen, W. Cheng, T. X. Fan and M. Niederberger, ACS Nano, 2015, 9, 4227.
92. Y. Z. Wan, Z. W. Yang, G. Y. Xiong and H. L. Luo, J. Mater. Chem. A, 2015, 3, 15386.
93. X. B. Hu, M. H. Ma, M. Q. Zeng, Y. Y. Sun, L. F. Chen, Y. H. Xue, T. Zhang, X. P. Ai, R. G. Mendes, M. H. Rummeli and L. Fu, ACS Appl. Mater. Interfaces, 2014, 6, 22527.
94. J. S. Luo, J. L. Liu, Z. Y. Zeng, C. F. Ng, L. J. Ma, H. Zhang, J. Y. Lin, Z. X. Shen and H. J. Fan, Nano Lett., 2013, 13, 6136.
95. X. Huang, H. Yu, J. Chen, Z. Lu, R. Yazami and H. Hoon Hng, Adv. Mater., 2014, 26, 1296.
96. K. Cao, L. Jiao, H. Liu, Y. Liu, Y. Wang, Z. Guo and H. Yuan, Adv. Energy Mater., 2015, 5, 1401421.
97. G. Binitha, A. G. Ashish, D. Ramasubramonian, P. Manikandan and M. M. Shaijumon, Adv. Mater. Interfaces, 2016, 3, 1500419.
98. C. C. Chen, Y. Huang, H. Zhang, X. F. Wang, Y. J. Wang, L. F. Jiao and H. T. Yuan, J. Power Sources, 2016, 314, 66.
99. K. Cao, L. Jiao, Y. Liu, H. Liu, Y. Wang and H. Yuan, Adv. Funct. Mater., 2015, 25, 1082.
100. B. Wang, X.-Y. Lu, K. Y. Wong and Y. Tang, Mater. Chem. Front., 2017, 1, 468.
101. S. J. Sun, X. Y. Zhao, M. Yang, L. L. Wu, Z. Y. Wen and X. D. Shen, Sci. Rep., 2016, 6, 19564.
102. N. S. Marzuki, N. U. Taib, M. F. Hassan and N. H. Idris, Electrochim. Acta, 2015, 182, 452.
103. G. Huang, F. F. Zhang, X. C. Du, Y. L. Qin, D. M. Yin and L. M. Wang, ACS Nano, 2015, 9, 1592.
104. Y. Kim, J. H. Lee, S. Cho, Y. Kwon, I. Inss, J. Lee, N. H. You, E. Reichmanis, H. Ko, K. T. Lee, H. K. Kwon, D. H. Ko, H. Yang and B. Park, ACS Nano, 2014, 8, 6701.
105. L. Luo, J. Wu, J. Xu and V. P. Dravid, ACS Nano, 2014, 8, 11560.
106. Q. M. Su, J. Zhang, Y. S. Wu and G. H. Du, Nano Energy, 2014, 9, 264.
107. J. Wang, N. Yang, H. Tang, Z. Dong, Q. Jin, M. Yang, D. Kisailus, H. Zhao, Z. Tang and D. Wang, Angew. Chem., Int. Ed., 2013, 52, 6417.
108. L. Yu, H. Hu, H. B. Wu and X. W. Lou, Adv. Mater., 2017, 29, 1604563.
109. J. Wang, H. Tang, L. Zhang, H. Ren, R. Yu, Q. Jin, J. Qi, D. Mao, M. Yang, Y. Wang, P. Liu, Y. Zhang, Y. Wen, L. Gu, G. Ma, Z. Su, Z. Tang, H. Zhao and D. Wang, Nat. Energy, 2016, 1, 16050.
110. L. Zhou, Z. Zhuang, H. Zhao, M. Lin, D. Zhao and L. Mai, Adv. Mater., 2017, 29, 1602914.
111. X. Yao, G. L. Guo, Y. Zhao, Y. Zhang, S. Y. Tan, Y. F. Zeng, R. Q. Zou, Q. Y. Yan and Y. L. Zhao, Small, 2016, 12, 3849.
112. J. Sun, D. H. Li, Y. Z. Xia, X. Y. Zhu, L. Zong, Q. Ji, Y. Jia and D. J. Yang, Chem. Commun., 2015, 51, 16267.
113. S. Abouali, M. A. Garakani, B. Zhang, H. Luo, Z. L. Xu, J. Q. Huang, J. Q. Huang and J. K. Kim, J. Mater. Chem. A, 2014, 2, 16939.
114. X. L. Huang, R. Z. Wang, D. Xu, Z. L. Wang, H. G. Wang, J. J. Xu, Z. Wu, Q. C. Liu, Y. Zhang and X. B. Zhang, Adv. Funct. Mater., 2013, 23, 4345.
115. Y. Zhao, C. L. Ma, Y. Li, H. L. Chen and Z. P. Shao, Carbon, 2015, 95, 494.
116. S.-L. Li and Q. Xu, Energy Environ. Sci., 2013, 6, 1656.
117. Y. Hou, J. Y. Li, Z. H. Wen, S. M. Cui, C. Yuan and J. H. Chen, Nano Energy, 2015, 12, 1.
118. X. Xu, K. Cao, Y. Wang and L. Jiao, J. Mater. Chem. A, 2016, 4, 6042.
119. Q. T. Qu, T. Gao, H. Y. Zheng, X. X. Li, H. M. Liu, M. Shen, J. Shao and H. H. Zheng, Carbon, 2015, 92, 119.
120. Y. M. Chen, L. Yu and X. W. Lou, Angew. Chem., Int. Ed., 2016, 55, 5990.
121. L. Peng, Y. Y. Feng, Y. J. Bai, H. J. Qiu and Y. Wang, J. Mater. Chem. A, 2015, 3, 8825.
122. G. Fang, J. Zhou, C. Liang, A. Pan, C. Zhang, Y. Tang, X. Tan, J. Liu and S. Liang, Nano Energy, 2016, 26, 57.
123. J. Bell, R. Ye, K. Ahmed, C. Liu, M. Ozkan and C. S. Ozkan, Nano Energy, 2015, 18, 47.
124. W. P. Sun, X. H. Rui, J. X. Zhu, L. H. Yu, Y. Zhang, Z. C. Xu, S. Madhavi and Q. Y. Yan, J. Power Sources, 2015, 274, 755.
125. X. H. Wang, L. Qiao, X. L. Sun, X. W. Li, D. K. Hu, Q. Zhang and D. Y. He, J. Mater. Chem. A, 2013, 1, 4173.
126. J. H. Jang, B. M. Chae, H. J. Oh and Y. K. Lee, J. Power Sources, 2016, 304, 189.
127. Y. P. Pang, J. Zhang, D. R. Chen and X. L. Jiao, RSC Adv., 2016, 6, 30395.
128. A. Guo, Y. Li, K. Liu and W. J. Ruan, RSC Adv., 2015, 5, 89269.
129. W. Yang, G. Cheng, C. Dong, Q. Bai, X. Chen, Z. Peng and Z. Zhang, J. Mater. Chem. A, 2014, 2, 20022.
130. K. Matsubara, S. Y. Huang, M. Iwamoto and W. Pan, Nanoscale, 2014, 6, 688.
131. Z. C. Bai, Z. C. Ju, C. L. Guo, Y. T. Qian, B. Tang and S. L. Xiong, Nanoscale, 2014, 6, 3268.
132. H. Li, H. Ma, M. Yang, B. Wang, H. Shao, L. Wang, R. Yu and D. Wang, Mater. Res. Bull., 2017, 87, 224.
133. X. Y. Shan, G. M. Zhou, L. C. Yin, W. J. Yu, F. Li and H. M. Cheng, J. Mater. Chem. A, 2014, 2, 17808.
134. Y. Y. Feng, H. J. Zhang, W. X. Li, L. Fang and Y. Wang, J. Power Sources, 2016, 301, 78.
135. R. A. Susantyoko, X. Wang, Q. Xiao, E. Fitzgerald and Q. Zhang, Carbon, 2014, 68, 619.
136. W. Wen, J. M. Wu and M. H. Cao, Nano Energy, 2013, 2, 1383.
137. W. Wen, J. M. Wu and M. H. Cao, J. Mater. Chem. A, 2013, 1, 3881.
138. X. Xu, H. Tan, K. Xi, S. Ding, D. Yu, S. Cheng, G. Yang, X. Peng, A. Fakeeh and R. V. Kumar, Carbon, 2015, 84, 491.
139. J. Liang, H. Hu, H. Park, C. Xiao, S. Ding, U. Paik and X. W. Lou, Energy Environ. Sci., 2015, 8, 1707.
140. F. Zou, Y. M. Chen, K. W. Liu, Z. T. Yu, W. F. Liang, S. M. Bhaway, M. Gao and Y. Zhu, ACS Nano, 2016, 10, 377.
141. X. Sun, C. Yan, Y. Chen, W. Si, J. Deng, S. Oswald, L. Liu and O. G. Schmidt, Adv. Energy Mater., 2014, 4, 1300912.
142. R. Zhang, J. Liu, H. Guo and X. Tong, Mater. Lett., 2015, 139, 55.
143. C. Zhu, D. Chao, J. Sun, I. M. Bacho, Z. Fan, C. F. Ng, X. Xia, H. Huang, H. Zhang, Z. X. Shen, G. Ding and H. J. Fan, Adv. Mater. Interfaces, 2015, 2, 1400499.
144. Q. Zhang, J. Wang, D. Xu, Z. Wang, X. Li and K. Zhang, J. Mater. Chem. A, 2014, 2, 3865.
145. H. Huang, Y. Liu, J. Wang, M. Gao, X. Peng and Z. Ye, Nanoscale, 2013, 5, 1785.
146. H. Zhang, G. H. Zhang, Z. Q. Li, K. Qu, L. Wang, W. Zeng, Q. F. Zhang and H. G. Duan, J. Mater. Chem. A, 2016, 4, 10585.
147. Q. Zhang, D. Xu, X. Zhou, X. Wu and K. Zhang, Small, 2014, 10, 935.
148. Y. F. Yuan, Y. B. Pei, J. Fang, H. L. Zhu, J. L. Yang and S. Y. Guo, Mater. Lett., 2013, 91, 279.
149. K. Chen and D. Xue, Phys. Chem. Chem. Phys., 2013, 15, 19708.
150. X. Duan, H. Huang, S. Xiao, J. Deng, G. Zhou, Q. Li and T. Wang, J. Mater. Chem. A, 2016, 4, 8402.
151. J. Wang, Y. Liu, S. Wang, X. Guo and Y. Liu, J. Mater. Chem. A, 2014, 2, 1224.
152. R. Wu, X. Qian, F. Yu, H. Liu, K. Zhou, J. Wei and Y. Huang, J. Mater. Chem. A, 2013, 1, 11126.
153. C. Wang, D. Higgins, F. Wang, D. Li, R. Liu, G. Xia, N. Li, Q. Li, H. Xu and G. Wu, Nano Energy, 2014, 9, 334.
154. T. Chen, Y. Hu, B. Cheng, R. Chen, H. Lv, L. Ma, G. Zhu, Y. Wang, C. Yan, Z. Tie, Z. Jin and J. Liu, Nano Energy, 2016, 20, 305.
155. Y. Xu, G. Jian, M. R. Zachariah and C. Wang, J. Mater. Chem. A, 2013, 1, 15486.
156. G. D. Park and Y. C. Kang, Chem. – Eur. J., 2015, 21, 9179.
157. Y. Liu, W. Wang, L. Gu, Y. Wang, Y. Ying, Y. Mao, L. Sun and X. Peng, ACS Appl. Mater. Interfaces, 2013, 5, 9850.
158. S. Ko, J.-I. Lee, H. S. Yang, S. Park and U. Jeong, Adv. Mater., 2012, 24, 4451.
159. Z. Yuxin, S. Mu, W. Sun, Q. Liu, Y. Li, Z.-F. Yan, Z. Huo and W. Liang, Nanoscale, 2016, 8, 19994.
160. W. Zhang, Z. Zhou, W. Zhao, Z. Yang and X. Yang, J. Mater. Chem. A, 2014, 2, 5800.
161. D. Rehnlund, M. Valvo, C. W. Tai, J. Angstrom, M. Sahlberg, K. Edstrom and L. Nyholm, Nanoscale, 2015, 7, 13591.
162. J. H. Shin, S. H. Park, S. M. Hyun, J. W. Kim, H. M. Park and J. Y. Song, Phys. Chem. Chem. Phys., 2014, 16, 18226.
163. S. Shilpa, P. Rai and A. Sharma, RSC Adv., 2016, 6, 105231.
164. M.-C. Kim, S.-J. Kim, S.-B. Han, D.-H. Kwak, E.-T. Hwang, D.-M. Kim, G.-H. Lee, H.-S. Choe and K.-W. Park, J. Mater. Chem. A, 2015, 3, 23003.
165. A. Paolella, R. Brescia, M. Prato, M. Povia, S. Marras, L. De Trizio, A. Falqui, L. Manna and C. George, ACS Appl. Mater. Interfaces, 2013, 5, 2745.
166. O. Delmer, P. Balaya, L. Kienle and J. Maier, Adv. Mater., 2008, 20, 501.
167. A. Ponrouch, P.-L. Taberna, P. Simon and M. Rosa Palacin, Electrochim. Acta, 2012, 61, 13.
168. K. E. Gregorczyk, Y. Liu, J. P. Sullivan and G. W. Rubloff, ACS Nano, 2013, 7, 6354.
169. Y.-Y. Hu, Z. Liu, K.-W. Nam, O. J. Borkiewicz, J. Cheng, X. Hua, M. T. Dunstan, X. Yu, K. M. Wiaderek, L.-S. Du, K. W. Chapman, P. J. Chupas, X.-Q. Yang and C. P. Grey, Nat. Mater., 2013, 12, 1130.
170. Y. Kim, S. Muhammad, H. Kim, Y.-H. Cho, H. Kim, J. M. Kim and W.-S. Yoon, ChemSusChem, 2015, 8, 2378.
171. A. S. Hassan, K. Moyer, B. R. Ramachandran and C. D. Wick, J. Phys. Chem. C, 2016, 120, 2036.
172. H. Liu, X. Du, X. Xing, G. Wang and S. Z. Qiao, Chem. Commun., 2012, 48, 865.
173. Z. Q. Cao, M. L. Qin, B. R. Jia, L. Zhang, Q. Wan, M. S. Wang, A. A. Volinsky and X. H. Qu, Electrochim. Acta, 2014, 139, 76.
174. S. Khamlich, Z. Y. Nuru, A. Bello, M. Fabiane, J. K. Dangbegnon, N. Manyala and M. Maaza, J. Alloys Compd., 2015, 637, 219.
175. G. X. Zhao, T. Wen, J. Zhang, J. X. Li, H. L. Dong, X. K. Wang, Y. G. Guo and W. P. Hu, J. Mater. Chem. A, 2014, 2, 944.
176. S. M. Abbas, N. Ahmad, R. Ataur, U. A. Rana, S. U.-D. Khan, S. Hussain and K.-W. Nam, Electrochim. Acta, 2016, 212, 260.
177. W. B. Yue, S. S. Tao, J. M. Fu, Z. Q. Gao and Y. Ren, Carbon, 2013, 65, 97.
178. F. Wang, W. Li, M. Hou, C. Li, Y. Wang and Y. Xia, J. Mater. Chem. A, 2015, 3, 1703.
179. B. Ahmed, M. Shahid, D. H. Nagaraju, D. H. Anjum, M. N. Hedhili and H. N. Alshareef, ACS Appl. Mater. Interfaces, 2015, 7, 13154.
180. Y. Sun, J. Wang, B. Zhao, R. Cai, R. Ran and Z. Shao, J. Mater. Chem. A, 2013, 1, 4736.
181. Z. Yuan, L. Si, D. Wei, L. Hu, Y. Zhu, X. Li and Y. Qian, J. Phys. Chem. C, 2014, 118, 5091.
182. H. Liu, X. Chen, L. Deng, M. Ding, J. Li and X. He, J. Mater. Chem. A, 2016, 4, 17764.
183. L. Noerochim, J.-Z. Wang, D. Wexler, Z. Chao and H.-K. Liu, J. Power Sources, 2013, 228, 198.
184. Y. F. Dong, S. Li, H. M. Xu, M. Y. Yan, X. M. Xu, X. C. Tian, Q. Liu and L. Q. Mai, Phys. Chem. Chem. Phys., 2013, 15, 17165.
185. J. Y. C. Qiu, Z. X. Yang and Y. Li, J. Mater. Chem. A, 2015, 3, 24245.
186. M. Xu, J. Tang, H. Wu and G. F. Zheng, RSC Adv., 2014, 4, 29586.
187. B. Mendoza-Sanchez and P. S. Grant, Electrochim. Acta, 2013, 98, 294.
188. J. F. Ni, G. B. Wang, J. Yang, D. L. Gao, J. T. Chen, L. J. Gao and Y. Li, J. Power Sources, 2014, 247, 90.
189. C. Villevieille, X. J. Wang, F. Krumeich, R. Nesper and P. Novak, J. Power Sources, 2015, 279, 636.
190. G. Y. Zhao, C. L. Li, L. Zhang, J. X. Lv, Y. N. Niu, Y. Du and K. N. Sun, J. Mater. Chem. A, 2015, 3, 22547.
191. Y. N. Ko, S. Bin Park, K. Y. Jung and Y. C. Kang, Nano Lett., 2013, 13, 5462.
192. F. Y. Li, C. R. Cabrera and Z. F. Chen, J. Mater. Chem. A, 2014, 2, 19180.
193. H. J. Zhang, J. Shu, K. X. Wang, X. T. Chen, Y. M. Jiang, X. Wei and J. S. Chen, J. Mater. Chem. A, 2014, 2, 80.
194. B. Guo, X. Fang, B. Li, Y. Shi, C. Ouyang, Y.-S. Hu, Z. Wang, G. D. Stucky and L. Chen, Chem. Mater., 2012, 24, 457.
195. L. Yang, L. Liu, Y. Zhu, X. Wang and Y. Wu, J. Mater. Chem., 2012, 22, 13148.
196. J. F. Ni, Y. Zhao, L. Li and L. Q. Mai, Nano Energy, 2015, 11, 129.
197. X. F. Zhang, X. X. Song, S. Gao, Y. M. Xu, X. L. Cheng, H. Zhao and L. H. Huo, J. Mater. Chem. A, 2013, 1, 6858.
198. X. Liu, D. Wu, W. Ji and W. Hou, J. Mater. Chem. A, 2015, 3, 968.
199. S. Petnikota, K. W. Teo, L. Chen, A. Sim, S. K. Marka, M. V. Reddy, V. V. Srikanth, S. Adams and B. V. Chowdari, ACS Appl. Mater. Interfaces, 2016, 8, 10884.
200. Y. Sun, X. Hu, W. Luo and Y. Huang, J. Mater. Chem., 2012, 22, 425.
201. L. X. Zeng, C. Zheng, C. L. Deng, X. K. Ding and M. D. Wei, ACS Appl. Mater. Interfaces, 2013, 5, 2182.
202. Y. Wang, L. Yu and X. W. Lou, Angew. Chem., Int. Ed., 2016, 128, 14888.
203. A. L. Chen, C. X. Li, R. Tang, L. W. Yin and Y. X. Qi, Phys. Chem. Chem. Phys., 2013, 15, 13601.
204. J. Besnardiere, X. Petrissans, C. Surcin, V. Buissette, T. Le Mercier, M. Morcrette, D. Portehault and S. Cassaignon, RSC Adv., 2014, 4, 21208.
205. G. L. Xia, D. Liu, F. C. Zheng, Y. Yang, J. W. Su and Q. W. Chen, J. Mater. Chem. A, 2016, 4, 12434.
206. A. Bhaskar, M. Deepa and T. N. Rao, ACS Appl. Mater. Interfaces, 2013, 5, 2555.
207. H. J. Zhang, K. X. Wang, X. Y. Wu, Y. M. Jiang, Y. B. Zhai, C. Wang, X. Wei and J. S. Chen, Adv. Funct. Mater., 2014, 24, 3399.
208. J. Lee, C. Jo, B. Park, W. Hwang, H. I. Lee, S. Yoon and J. Lee, Nanoscale, 2014, 6, 10147.
209. L. Gao, X. Wang, Z. Xie, W. Song, L. Wang, X. Wu, F. Qu, D. Chen and G. Shen, J. Mater. Chem. A, 2013, 1, 7167.
210. S. A. Pervez, D. Kim, C.-H. Doh, U. Farooq, H.-Y. Choi and J.-H. Choi, ACS Appl. Mater. Interfaces, 2015, 7, 7635.
211. X. Gu, F. Wu, B. Lei, J. Wang, Z. Chen, K. Xie, Y. Song, D. Sun, L. Sun, H. Zhou and F. Fang, J. Power Sources, 2016, 320, 231.
212. Y. He, M. Gu, H. Xiao, L. Luo, Y. Shao, F. Gao, Y. Du, S. X. Mao and C. Wang, Angew. Chem., Int. Ed., 2016, 55, 6244.
213. M. Sasidharan, N. Gunawardhana, M. Yoshio and K. Nakashima, Nano Energy, 2012, 1, 503.
214. H. Tong, Y. Xu, X. Cheng, X. Zhang, S. Gao, H. Zhao and L. Huo, Electrochim. Acta, 2016, 210, 147.
215. W. Wang, X. Meng, K. Zhang, P. Li, D. Choi, J. H. Park and Y. Son, Electrochim. Acta, 2016, 215, 466.
216. C. M. Sim, Y. J. Hong and Y. C. Kang, ChemSusChem, 2013, 6, 1320.
217. C. Lian, X. Xiao, Z. Chen, Y. Liu, E. Zhao, D. Wang, C. Chen and Y. Li, Nano Res., 2016, 9, 435.
218. J. Yang, L. Jiao, Q. Zhao, Q. Wang, H. Gao, Q. Huan, W. Zheng, Y. Wang and H. Yuan, J. Mater. Chem., 2012, 22, 3699.
219. L. Gao, F. Qu and X. Wu, J. Mater. Chem. A, 2014, 2, 7367.
220. B. Li, X. Li, W. Li, Y. Wang, E. Uchaker, Y. Pei, X. Cao, S. Li, B. Huang and G. Cao, ChemNanoMat, 2016, 2, 281.
221. Y. Zhang, L. Li, H. Su, W. Huang and X. Dong, J. Mater. Chem. A, 2015, 3, 43.
222. C. Yuan, H. B. Wu, Y. Xie and X. W. Lou, Angew. Chem., Int. Ed., 2014, 53, 1488.
223. M. Bijelic, X. Liu, Q. Sun, A. B. Djurisic, M. H. Xie, A. M. C. Ng, C. Suchomski, I. Djerdj, Z. Skoko and J. Popovic, J. Mater. Chem. A, 2015, 3, 14759.
224. L. Wang, B. Liu, S. Ran, L. Wang, L. Gao, F. Qu, D. Chen and G. Shen, J. Mater. Chem. A, 2013, 1, 2139.
225. G. Yang, X. Xu, W. Yan, H. Yang and S. Ding, Electrochim. Acta, 2014, 137, 462.
226. L. Zhou, D. Zhao and X. W. Lou, Adv. Mater., 2012, 24, 745.
227. G. Zhang and X. W. Lou, Angew. Chem., Int. Ed., 2014, 53, 9041.
228. G. Li, L. Xu, Y. Zhai and Y. Hou, J. Mater. Chem. A, 2015, 3, 14298.
229. J. Li, S. Xiong, X. Li and Y. Qian, Nanoscale, 2013, 5, 2045.
230. C. Fu, G. Li, D. Luo, X. Huang, J. Zheng and L. Li, ACS Appl. Mater. Interfaces, 2014, 6, 2439.
231. L. Yu, L. Zhang, H. B. Wu, G. Zhang and X. W. Lou, Energy Environ. Sci., 2013, 6, 2664.
232. L. Zhang, G. He, S. Lei, G. Qi, H. Jiu and J. Wang, J. Power Sources, 2016, 326, 505.
233. J. Meng, C. Niu, X. Liu, Z. Liu, H. Chen, X. Wang, J. Li, W. Chen, X. Guo and L. Mai, Nano Res., 2016, 9, 2445.
234. W. Kang, Y. Tang, W. Li, X. Yang, H. Xue, Q. Yang and C.-S. Lee, Nanoscale, 2015, 7, 225.
235. Y. Ma, C.-W. Tai, R. Younesi, T. Gustafsson, J. Y. Lee and K. Edstrom, Chem. Mater., 2015, 27, 7698.
236. B. Wang, S. Li, X. Wu, B. Li, J. Liu and M. Yu, Phys. Chem. Chem. Phys., 2015, 17, 21476.
237. P. R. Kumar and S. Mitra, RSC Adv., 2013, 3, 25058.
238. G. Huang, X. Du, F. Zhang, D. Yin and L. Wang, Chem. – Eur. J., 2015, 21, 14140.
239. X. Zhang, Y. Xie, Y. Sun, Q. Zhang, Q. Zhu, D. Hou and J. Guo, RSC Adv., 2015, 5, 29837.
240. C. T. Cherian, J. Sundaramurthy, M. V. Reddy, P. S. Kumar, K. Mani, D. Pliszka, C. H. Sow, S. Ramakrishna and B. V. R. Chowdari, ACS Appl. Mater. Interfaces, 2013, 5, 9957.
241. H. Yu, H. Fan, B. Yadian, H. Tan, W. Liu, H. H. Hng, Y. Huang and Q. Yan, ACS Appl. Mater. Interfaces, 2015, 7, 26751.
242. S. Li, A. Li, R. Zhang, Y. He, Y. Zhai and L. Xu, Nano Res., 2014, 7, 1116.
243. B. Wang, S. Li, J. Liu, M. Yu, B. Li and X. Wu, Electrochim. Acta, 2014, 146, 679.
244. B. Wang, G. Wang, Z. Lv and H. Wang, Phys. Chem. Chem. Phys., 2015, 17, 27109.
245. J. Xiao, G. Xu, S.-G. Sun and S. Yang, Part. Part. Syst. Charact., 2013, 30, 893.
246. E. K. Heidari, B. Zhang, M. H. Sohi, A. Ataie and J.-K. Kim, J. Mater. Chem. A, 2014, 2, 8314.
247. G. Zeng, N. Shi, M. Hess, X. Chen, W. Cheng, T. Fan and M. Niederberger, ACS Nano, 2015, 9, 4227.
248. P. Zhu, S. Liu, J. Xie, S. Zhang, G. Cao and X. Zhao, J. Mater. Sci. Technol., 2014, 30, 1078.
249. Z. Zhang, Y. Wang, M. Zhang, Q. Tan, X. Lv, Z. Zhong and F. Su, J. Mater. Chem. A, 2013, 1, 7444.
250. S. Ren, X. Zhao, R. Chen and M. Fichtner, J. Power Sources, 2014, 260, 205.
251. S. Li, B. Wang, J. Liu and M. Yu, Electrochim. Acta, 2014, 129, 33.
252. L. Luo, R. Cui, H. Qiao, K. Chen, Y. Fei, D. Li, Z. Pang, K. Liu and Q. Wei, Electrochim. Acta, 2014, 144, 85.
253. J. Zhao, Y. Cheng, X. Yan, D. Sun, F. Zhu and Q. Xue, CrystEngComm, 2012, 14, 5879.
254. L. Jin, Y. Qiu, H. Deng, W. Li, H. Li and S. Yang, Electrochim. Acta, 2011, 56, 9127.
255. Y. Dong, Y.-S. Chui, R. Ma, C. Cao, H. Cheng, Y. Y. Li and J. A. Zapien, J. Mater. Chem. A, 2014, 2, 13892.
256. H. Xu, Y. Wang, L. Zheng, X. Duan, L. Wang, J. Yang and Y. Qian, J. Energy Chem., 2014, 23, 354.
257. N. Wang, X. Ma, Y. Wang, J. Yang and Y. Qian, J. Mater. Chem. A, 2015, 3, 9550.
258. L. Shen, L. Yu, X.-Y. Yu, X. Zhang and X. W. Lou, Angew. Chem., Int. Ed., 2015, 54, 1868.
259. X. Yao, C. Zhao, J. Kong, D. Zhou and X. Lu, RSC Adv., 2014, 4, 37928.
260. Y. Chen, J. Zhu, B. Qu, B. Lu and Z. Xu, Nano Energy, 2014, 3, 88.
261. Y. Zhu and C. Cao, Electrochim. Acta, 2015, 176, 141.
262. S. Liu, J. Wu, J. Zhou, G. Fang and S. Liang, Electrochim. Acta, 2015, 176, 1.
263. H. Guo, L. Liu, T. Li, W. Chen, J. Liu, Y. Guo and Y. Guo, Nanoscale, 2014, 6, 5491.
264. Y. Chen, M. Zhuo, J. Deng, Z. Xu, Q. Li and T. Wang, J. Mater. Chem. A, 2014, 2, 4449.
265. G. Gao, H. B. Wu and X. W. Lou, Adv. Energy Mater., 2014, 4, 1400422.
266. L. Shen, Q. Che, H. Li and X. Zhang, Adv. Funct. Mater., 2014, 24, 2630.
267. S. Abouali, M. A. Garakani, Z.-L. Xu and J.-K. Kim, Carbon, 2016, 102, 262.
268. Z. Fan, B. Wang, Y. Xi, X. Xu, M. Li, J. Li, P. Coxon, S. Cheng, G. Gao, C. Xiao, G. Yang, K. Xi, S. Ding and R. V. Kumar, Carbon, 2016, 99, 633.
269. C. Chen, B. Liu, Q. Ru, S. Ma, B. An, X. Hou and S. Hu, J. Power Sources, 2016, 326, 252.
270. J. Li, J. Wang, X. Liang, Z. Zhang, H. Liu, Y. Qian and S. Xiong, ACS Appl. Mater. Interfaces, 2014, 6, 24.
271. A. K. Mondal, D. Su, S. Chen, A. Ung, H.-S. Kim and G. Wang, Chem. – Eur. J., 2015, 21, 1526.
272. X. Hou, X. Wang, B. Liu, Q. Wang, T. Luo, D. Chen and G. Shen, Nanoscale, 2014, 6, 8858.
273. Y. Zhang, X. Wang, Q. Zhao, Y. Fu, H. Wang and H. Shu, Electrochim. Acta, 2015, 180, 866.
274. G. Huang, S. Xu, Z. Xu, H. Sun and L. Li, ACS Appl. Mater. Interfaces, 2014, 6, 21325.
275. R. Jin, Y. Meng, Y. Ma, H. Li, Y. Sun and G. Chen, Electrochim. Acta, 2016, 209, 163.
276. W. Kang, Y. Tang, W. Li, Z. Li, X. Yang, J. Xu and C. S. Lee, Nanoscale, 2014, 6, 6551.
277. L. Liu, Z. Hu, L. Sun, G. Gao and X. Liu, RSC Adv., 2015, 5, 36575.
278. S. G. Mohamed, C.-J. Chen, C. K. Chen, S.-F. Hu and R.-S. Liu, ACS Appl. Mater. Interfaces, 2014, 6, 22701.
279. L. Liu, H. Zhang, Y. Mu, J. Yang and Y. Wang, ACS Appl. Mater. Interfaces, 2016, 8, 1351.
280. H. Zhu, Y. Sun, X. Zhang, L. Tang and J. Guo, Mater. Lett., 2016, 166, 1.
281. Z. Ju, E. Zhang, Y. Zhao, Z. Xing, Q. Zhuang, Y. Qiang and Y. Qian, Small, 2015, 11, 4753.
282. W. Wang, J. Qin, Z. Yin and M. Cao, ACS Nano, 2016, 10, 10106.
283. B. Wang, S. Li, X. Wu, W. Tian, J. Liu and M. Yu, J. Mater. Chem. A, 2015, 3, 13691.
284. B. Guan, W. Sun and Y. Wang, Electrochim. Acta, 2016, 190, 354.
285. Z. Zhang, W. Li, T.-W. Ng, W. Kang, C.-S. Lee and W. Zhang, J. Mater. Chem. A, 2015, 3, 20527.
286. J. Xu, S. Gu, L. Fan, P. Xu and B. Lu, Electrochim. Acta, 2016, 196, 125.
287. Y. Wang, Y. Sun, X. Zhang, Y.-h. Wen and J. Guo, RSC Adv., 2016, 6, 51710.
288. T. Yang, H. Zhang, Y. Luo, L. Mei, D. Guo, Q. Li and T. Wang, Electrochim. Acta, 2015, 158, 327.
289. Y. Chen, B. Liu, W. Jiang, Q. Liu, J. Liu, J. Wang, H. Zhang and X. Jing, J. Power Sources, 2015, 300, 132.
290. H. Yu, C. Guan, X. Rui, B. Ouyang, B. Yadian, Y. Huang, H. Zhang, H. E. Hoster, H. J. Fan and Q. Yan, Nanoscale, 2014, 6, 10556.
291. J. Yao, Y. Gong, S. Yang, P. Xiao, Y. Zhang, K. Keyshar, G. Ye, S. Ozden, R. Vajtai and P. M. Ajayan, ACS Appl. Mater. Interfaces, 2014, 6, 20414.
292. B. Wang, S. Li, X. Wu, J. Liu and W. Tian, Phys. Chem. Chem. Phys., 2016, 18, 908.
293. K. Xiao, L. Xia, G. Liu, S. Wang, L.-X. Ding and H. Wang, J. Mater. Chem. A, 2015, 3, 6128.
294. J. H. Ahn, G. D. Park, Y. C. Kang and J.-H. Lee, Electrochim. Acta, 2015, 174, 102.
295. X. Wang, Y. Liu, Y. Wang and L. Jiao, Small, 2016, 12, 4865.
296. J. Haetge, I. Djerdj and T. Brezesinski, Chem. Commun., 2012, 48, 6726.
297. Y. Liu, Y. Zhao, Y. Yu, J. Li, M. Ahmad and H. Sun, New J. Chem., 2014, 38, 3084.
298. Y. Dong, C. Cao, Y.-S. Chui and J. A. Zapien, Chem. Commun., 2014, 50, 10151.
299. L. Lu, J.-Z. Wang, X.-B. Zhu, X.-W. Gao and H.-K. Liu, J. Power Sources, 2011, 196, 7025.
300. K. Cao, H. Liu, X. Xu, Y. Wang and L. Jiao, Chem. Commun., 2016, 52, 11414.
301. C. S. Bongu, J. Ragupathi and K. Nallathamby, Inorg. Chem., 2016, 55, 11644.
302. Y. Sun, X. Hu, W. Luo, J. Shu and Y. Huang, J. Mater. Chem. A, 2013, 1, 4468.
303. S. Petnikota, S. K. Marka, V. V. S. S. Srikanth, M. V. Reddy and B. V. R. Chowdari, Electrochim. Acta, 2015, 178, 699.
304. Y. Sun, X. Hu, W. Luo and Y. Huang, J. Mater. Chem., 2011, 21, 17229.
305. S. K. Marka, S. Petnikota, V. V. S. S. Srikanth, M. V. Reddy, S. Adams and B. V. R. Chowdari, RSC Adv., 2016, 6, 55167.
306. B. Das, M. V. Reddy, S. Tripathy and B. V. R. Chowdari, RSC Adv., 2014, 4, 33883.
307. B. Das, M. V. Reddy, C. Krishnamoorthi, S. Tripathy, R. Mahendiran, G. V. S. Rao and B. V. R. Chowdari, Electrochim. Acta, 2009, 54, 3360.
308. M. Xie, X. Sun, S. M. George, C. Zhou, J. Lian and Y. Zhou, ACS Appl. Mater. Interfaces, 2015, 7, 27735.
309. C. Kim, J.-W. Jung, K. R. Yoon, D.-Y. Youn, S. Park and I.-D. Kim, ACS Nano, 2016, 10, 11317.
310. J. Zhang, H. Ren, J. Wang, J. Qi, R. Yu, D. Wang and Y. Liu, J. Mater. Chem. A, 2016, 4, 17673.
311. L. P. Wang, Y. Leconte, Z. Feng, C. Wei, Y. Zhao, Q. Ma, W. Xu, S. Bourrioux, P. Azais, M. Srinivasan and Z. J. Xu, Adv. Mater., 2017, 29, 1603286.
312. B. Jiang, Y. He, B. Li, S. Zhao, S. Wang, Y.-B. He and Z. Lin, Angew. Chem., Int. Ed., 2017, 56, 1869.
313. R. Hu, D. Chen, G. Waller, Y. Ouyang, Y. Chen, B. Zhao, B. Rainwater, C. Yang, M. Zhu and M. Liu, Energy Environ. Sci., 2016, 9, 595.
314. L. Xia, S. Wang, G. Liu, L. Ding, D. Li, H. Wang and S. Qiao, Small, 2016, 12, 853.
315. K. Zhao, L. Zhang, R. Xia, Y. Dong, W. Xu, C. Niu, L. He, M. Yan, L. Qu and L. Mai, Small, 2016, 12, 588.
316. V. Etacheri, G. A. Seisenbaeva, J. Caruthers, G. Daniel, J.-M. Nedelec, V. G. Kessler and V. G. Pol, Adv. Energy Mater., 2015, 5, 1401289.
317. Y. Liu, X. Yan, J. Lan, Y. Yu, X. Yang and Y. Lin, Mater. Chem. Front., 2017, 1, 1331.
318. H. Wang, X. Lu, L. Li, B. Li, D. Cao, Q. Wu, Z. Li, G. Yang, B. Guo and C. Niu, Nanoscale, 2016, 8, 7595.
319. R. Huang, L. Wang, Q. Zhang, Z. Chen, Z. Li, D. Pan, B. Zhao, M. Wu, C. M. L. Wu and C.-H. Shek, ACS Nano, 2015, 9, 11351.
320. H.-P. Cong, S. Xin and S.-H. Yu, Nano Energy, 2015, 13, 482.
321. C. Ma, W. Zhang, Y.-S. He, Q. Gong, H. Che and Z.-F. Ma, Nanoscale, 2016, 8, 4121.
322. B. Huang, X. Li, Y. Pei, S. Li, X. Cao, R. C. Masse and G. Cao, Small, 2016, 12, 1945.
323. L. Shen, F. LiU, G. Chen, H. Zhou, Z. Le, H. B. Wu, G. Wang and Y. Lu, J. Mater. Chem. A, 2016, 4, 18706.
324. Z.-F. Li, O. Liu, Y. Liu, F. Yang, L. Xin, Y. Zhou, H. Zhang, L. Stanciu and J. Xie, ACS Appl. Mater. Interfaces, 2015, 7, 27087.
325. X. Chen, Y. Huang, K. Zhang, X. Feng and C. Wei, J. Alloys Compd., 2017, 690, 765.
326. G. Zhang, S. Hou, H. Zhang, W. Zeng, F. Yan, C. C. Li and H. Duan, Adv. Mater., 2015, 27, 2400.
327. Z. Li, X. Yu, Y. Liu, W. Zhao, H. Zhang, R. Xu, D. Wang and H. Shen, J. Mater. Chem. A, 2016, 4, 19123.
328. X. Shen, D. Mu, S. Chen, R. Huang and F. Wu, J. Mater. Chem. A, 2014, 2, 4309.
329. Shilpa, B. M. Basavaraja, S. B. Majumder and A. Sharma, J. Mater. Chem. A, 2015, 3, 5344.
330. A. Kushima, X. H. Liu, G. Zhu, Z. L. Wang, J. Y. Huang and J. Li, Nano Lett., 2011, 11, 4535.
331. S. Li, Y. Xiao, X. Wang and M. Cao, Phys. Chem. Chem. Phys., 2014, 16, 25846.
332. X. Li, J. Liang, Z. Hou, Y. Zhu, Y. Wang and Y. Qian, Chem. Commun., 2014, 50, 13956.
333. H. Jia, R. Kloepsch, X. He, J. P. Badillo, M. Winter and T. Placke, J. Mater. Chem. A, 2014, 2, 17545.
334. A. Jahel, A. Darwiche, C. M. Ghimbeu, C. Vix-Guterl and L. Monconduit, J. Power Sources, 2014, 269, 755.
335. K. H. Seng, M.-h. Park, Z. P. Guo, H. K. Liu and J. Cho, Nano Lett., 2013, 13, 1230.
336. Y. Son, M. Park, Y. Son, J.-S. Lee, J.-H. Jang, Y. Kim and J. Cho, Nano Lett., 2014, 14, 1005.
337. N. Duc Tung, R. S. Kalubarme, H. T. T. Le, C.-N. Park and C.-J. Park, Nanoscale, 2015, 7, , 2552.
338. H. Jiang, Y. J. Hu, S. J. Guo, C. Y. Yan, P. S. Lee and C. Z. Li, ACS Nano, 2014, 8, 6038.
339. M. H. Alfaruqi, J. Gim, S. Kim, J. Song, P. T. Duong, J. Jo, J. P. Baboo, Z. Xiu, V. Mathew and J. Kim, Chem. – Eur. J., 2016, 22, 2039.
340. S. Z. Huang, J. Jin, Y. Cai, Y. Li, H. Y. Tan, H. E. Wang, G. Van Tendeloo and B. L. Su, Nanoscale, 2014, 6, 6819.
341. J. S. Cho, Y. J. Hong and Y. C. Kang, ACS Nano, 2015, 9, 4026.
342. W. Zhang, D. C. Bock, C. J. Pelliccione, Y. Li, L. Wu, Y. Zhu, A. C. Marschilok, E. S. Takeuchi, K. J. Takeuchi and F. Wang, Adv. Energy Mater., 2016, 6, 1502471.
343. X. Guan, J. Nai, Y. Zhang, P. Wang, J. Yang, L. Zheng, J. Zhang and L. Guo, Chem. Mater., 2014, 26, 5958.
344. W. W. Yuan, J. Zhang, D. Xie, Z. M. Dong, Q. M. Su and G. H. Du, Electrochim. Acta, 2013, 108, 506.
345. C. S. Yan, G. Chen, X. Zhou, J. X. Sun and C. D. Lv, Adv. Funct. Mater., 2016, 26, 1428.
346. N. S. Spinner, A. Palmieri, N. Beauregard, L. C. Zhang, J. Campanella and W. E. Mustain, J. Power Sources, 2015, 276, 46.
347. M. Valvo, D. Rehnlund, U. Lafont, M. Hahlin, K. Edstrom and L. Nyholm, J. Mater. Chem. A, 2014, 2, 9574.
348. G. D. Park, J. H. Kim, Y. J. Choi and Y. C. Kang, Electrochim. Acta, 2015, 173, 581.
349. P. Preetham, S. Mohapatra, S. V. Nair, D. Santhanagopalan and A. K. Rai, RSC Adv., 2016, 6, 38064.
350. Z. Xing, Z. Ju, J. Yang, H. Xu and Y. Qian, Electrochim. Acta, 2013, 102, 51.
351. F. Niu, N. Wang, J. Yue, L. Chen, J. Yang and Y. Qian, Electrochim. Acta, 2016, 208, 148.
352. Y. Sharma, N. Sharma, G. V. S. Rao and B. V. R. Chowdari, J. Power Sources, 2007, 173, 495.
353. F. Wang, H. C. Yu, M. H. Chen, L. Wu, N. Pereira, K. Thornton, A. Van der Ven, Y. Zhu, G. G. Amatucci and J. Graetz, Nat. Commun., 2012, 3, 1201.
354. F. Lin, D. Nordlund, T. C. Weng, Y. Zhu, C. Ban, R. M. Richards and H. L. Xin, Nat. Commun., 2014, 5, 3358.

---

Footnote

† These authors contributed equally.

---