Binary iron sulfides as anode materials for rechargeable batteries: Crystal structures, syntheses, and electrochemical performance

Journal of Power Sources 379 (2018) 41–52

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Review article

Binary iron sulfides as anode materials for rechargeable batteries: Crystal structures, syntheses, and electrochemical performance

T

Qian-Ting Xu, Jia-Chuang Li, Huai-Guo Xue, Sheng-Ping Guo∗ School of Chemistry & Chemical Engineering, Yangzhou University, Yangzhou, Jiangsu 225002, PR China

H I G H L I G H T S

G RA P H I C A L AB S T R A C T

structures of iron sulfides are • The systematically summarized. synthetic methods of iron sulfides • The are discussed. progress on various binary iron • Recent sulfides' electrochemical applications is overviewed.

attention should be paid on their • More applications for post-lithium types of batteries.

A R T I C L E I N F O

A B S T R A C T

Keywords: Iron sulfide Rechargeable batteries Crystal structure Synthesis Electrochemical performance

Effective utilization of energy requires the storage and conversion device with high ability. For well-developed lithium ion batteries (LIBs) and highly developing sodium ion batteries (SIBs), this ability especially denotes to high energy and power densities. It's believed that the capacity of a full cell is mainly contributed by anode materials. So, to develop inexpensive anode materials with high capacity are meaningful for various rechargeable batteries' better applications. Iron is a productive element in the crust, and its oxides, sulfides, fluorides, and oxygen acid salts are extensively investigated as electrode materials for batteries. In view of the importance of electrode materials containing iron, this review summarizes the recent achievements on various binary iron sulfides (FeS, FeS2, Fe3S4, and Fe7S8)-type electrodes for batteries. The contents are mainly focused on their crystal structures, synthetic methods, and electrochemical performance. Moreover, the challenges and some improvement strategies are also discussed.

1. Introduction With the rapid development of society, nonrenewable natural resources are becoming scarcer and scarcer, such as coal, petroleum and natural gas. It is urgent to explore green and renewable energy resources, or effective utilization and storage of energy, as the remnant unrenewable resources can hardly meet the requirements of our life in the future. Effective utilization and storage of energy require high effective energy storage and conversion devices. To date, there are many types of such devices have been developed well, including various

∗

rechargeable batteries, super capacitors, solar cells, and so on. For energy storage devices, Li ion batteries (LIBs) show their great potential in various portable electronic products in view of their ideal energy densities [1–4]. A promising type of batteries needs electrodes with high capacities, low cost and safety. Among the main battery components, anode materials have received great attention as they are the main contributors for the batteries' capacities. Researchers wish to find a series of anode materials with much higher theoretical capacity than the commercial graphite electrode (theoretical capacity: 372 mAh/g). In the search of potential anode

Corresponding author. E-mail address: [email protected] (S.-P. Guo).

https://doi.org/10.1016/j.jpowsour.2018.01.022 Received 17 October 2017; Received in revised form 22 December 2017; Accepted 9 January 2018 0378-7753/ © 2018 Elsevier B.V. All rights reserved.

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

2. Crystal structures

materials, Si, Ge, Sn, SnO2, alloys, Li4Ti5O12, LixV3O8, transition-metal oxides, sulfides, and fluorides, have been extensively developed [5–8]. Among so many types of anode materials, iron-based materials are specifically studied in view of iron's abundance, low price, and nontoxicity. The most successful iron-based electrode material is olivine type LiFePO4 cathode, which has been commercialized [9,10]. It is well known that sulfides usually have much higher conductivities and smaller volume expansions during cycling than those of oxides, which means that sulfides have better ion transport mobilities than oxides [11,12]. In addition, the chemical bond between metal and sulfur is weaker than that of metal oxides, which can be advantageous to the conversion reactions. Moreover, metal sulfides usually have better mechanical and thermal stabilities than metal oxides [11]. Combining the advantages of iron and sulfides, it is valuable to develop iron sulfides as anode materials for rechargeable batteries. In fact, there are a large number of reports concerning with binary iron sulfides-type anodes. In view of the success of iron sulfides in the field of LIBs, more and more exploration of them in the field of sodium-ion batteries (SIBs) have been carried out in recent years. Apart from the well-known advantages, iron sulfides also demonstrate several disadvantages to restrict their practical applications, these including relatively large volume changes during cycling induced poor cyclability, and relatively difficult syntheses. To solve these issues and get valuable metal sulfide anode materials, a huge number of efforts are being made to explore and develop metal sulfide-based anode materials for rechargeable batteries via modified synthetic methods, and various morphology control and material composite techniques. Intrigued by iron sulfides' potential in the fields of LIBs and SIBs, it is necessary to present a summary on their crystal structures, syntheses, electrochemical properties, and future studies. To date, there are only limited review papers about nanostructures transition metal sulfides for LIBs and another review about metal sulfides/selenides for SIBs summarized, respectively [13–15]. As far as we know, there is not a specific review focusing on the applications of iron sulfides in LIBs and SIBs. In this review, the contents are addressed to firstly give the crystal structures of iron sulfides (FeS, FeS2, Fe3S4, and Fe7S8). It is well known that crystal structure is the decisive factor to influence a material's properties. Secondly, various synthetic methods for iron sulfides are discussed, mainly focused on the normally employed ones. Thirdly, the electrochemical properties for each type of iron sulfide are discussed, respectively. Relatively, FeS and FeS2 are extensively investigated, while Fe3S4 and Fe7S8 are rarely studied. Finally, the possible future of these iron sulfides is expected. This summary should give researchers not only a systematic summary of known iron sulfide-type anodes, but also make a discussion on their future.

2.1. FeS To date, there are totally eight polymorphs of FeS discovered as listed in Table 1. FeS can crystallize in the cubic, monoclinic, orthorhombic, tetrahedral, and hexagonal space groups, respectively. Their structures can be classified into pyrrhotite, troilite, mackinawite, and their modifications. Some polymorphs are rarely studied as they are high temperature stable phases. When carefully check the literature on the electrochemical application of FeS, it can be found that most of them crystallize in the hexagonal space group P6 2c (troilite) (Fig. 1a). This structure can be simply described as distorted FeS6 octahedra constructed a 3D structure, in which each FeS6 unit shares edges with its neighboring FeS6 units along on either a or c direction [16]. Apart from troilite one, pyrrhotite (hexagonal P63/mmc) and mackinawite (tetragonal P4/nmm) FeS materials also have been studies as anode materials for LIBs [17]. Pyrrhotite FeS demonstrates a 3D structure also built by FeS6 octahedra, in which these octahedra are also linked together via sharing edges (Fig. 1b). While for mackinawite FeS, it shows a layered structure parallel to the ab plane, in which intralayer FeS4 tetrahedra link their neighbors by sharing corners or edges (Fig. 1c). 2.2. FeS2 There are two polymorphs for FeS2, pyrite and marcasite, respectively. For the former, it crystallizes in the cubic space group Pa3, and its 3D structure contains FeS6 octahedra and S–S dimers. The distorted FeS6 octahedron has six identical Fe–S bond distances. Each S atoms have neighboring three Fe and one S atoms, and each FeS6 octahedron connects with the nearest twelve FeS6 octahedra via sharing corners (Fig. 2a). For marcasite FeS2, it crystallizes in the orthorhombic space group Pnnm, and also contains the structure building units FeS6 octahedron and S–S dimers (Fig. 2b). Different from pyrite one, the distorted FeS6 octahedra in marcasite FeS2 contains two groups of Fe–S bond lengths. Each FeS6 octahedron has ten neighboring FeS6 units, linked via sharing corners (eight) or edges (two). Compared with pyrite, marcasite has shorter Fe–S and longer S–S distances. It is interesting to find that almost all the studied FeS2 anode materials for batteries are pyrite. 2.3. Fe3S4 There are two polymorphs for Fe3S4, namely, cubic greigite (Fd3m, Pearson code: cF56) and trigonal smythite (R3m, Pearson code: hR21). The former with spinel structure has been extensively studied as electrode materials for batteries, while the latter has attracted little attention. For the cubic phase, the S atoms construct a face-centered cubic lattice, in which 1/8 of the tetrahedral cavities are occupied by Fe3+ ions, and 1/2 of the octahedral cavities are co-occupied by 1: 1 Fe2+

Table 1 Known polymorphs of FeS.a Crystal system

Space group

Unit cell parameters

Prototype

Modification

hexagonal tetragonal hexagonal orthorhombic orthorhombic cubic monoclinic hexagonal

P63/mmc P4/nmm P6 2c Pnma F222 F4 3m P21/c P63mc

3.445 × 3.445 × 5.763 3.674 × 3.674 × 5.033 5.965 × 5.965 × 11.759 5.739 × 3.377 × 5.807 5.195 × 5.487 × 5.540 5.42 × 5.42 × 5.42 8.110 × 5.667 × 6.483, β = 93.05 6.588 × 6.588 × 5.400

NiAs, hP4 PbO, tP4 FeS, hP24 FeAs, oP8 FeS, oF8 ZnS, cF8 FeS, mP24 Nb0.92S, hP16

pyrrhotite mackinawite troilite HTb LTb RTb blende –b HTb

a All the data from Pearson's Crystal Data (PCD) or Inorganic Crystal Structure Database (ICSD). Eight polymorphs have been identified via carefully checking their unit cell parameters and powder X-ray diffraction patterns. b HT, LT, RT denote high-, low-, and room temperature phases, respectively. “–” indicates no data available.

42

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Fig. 1. Crystal structures of troilite (a), pyrrhotite (b), and mackinawite (c) FeS. Fe and S atoms are marked with turquoise and yellow spheres, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

and Fe3+ ions (Fig. 3a) [18]. The trigonal structure shows tubular-like 1D structure along the a direction, in which both Fe2+ and Fe3+ ions connect with six neighboring S atoms to form FeS6 octahedra. Each Fe2+ ion-centered octahedron links with six Fe(II)S6 octahedra via sharing edges, meanwhile it connects with 14 Fe(III)S6 octahedra by sharing corners (12) or edges (2) (Fig. 3b).

3. Syntheses As mentioned above, the binary iron sulfides are usually obtained from their respective minerals via mining and separation. On the other hand, they also can be produced by heating the stoichiometric Fe and S elements by a suitable high-temperature heat treatment process according to the Fe–S binary phase diagram [19]. To achieve better electrochemical properties, many attempts have been made to solve the challengeable issues, such as nanostructures and surface engineering techniques. It is commonly believed that electrode materials with special nanostructures can better relieve the mechanical stress resulted from volume changes than bulk materials. For example, core-shell, yolk-shell, sandwich-like, hollow or porous nanostructures of iron sulfides are advantageous to show improved electrochemical performance than those of bulk iron sulfides. On the other hand, coating various conductive materials, including carbon nanofibers, carbon nanotubes,

2.4. Fe7S8 Fe7S8 can crystallize in two different space groups, monoclinic C2/c and hexagonal P3121. However, both of them belong to the 3D pyrrhotite type. There are four and six Fe atoms in their crystallographically independent units, respectively, and all the Fe atoms are sixfold-coordinated with S atoms (Fig. 3c and d).

Fig. 2. Crystal structures of pyrite (a, Pa3 ) and marcasite (b, Pnnm) FeS2. Fe and S atoms are marked with turquoise and yellow spheres, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

43

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Fig. 3. Crystal structures of greigite (a, Fd3 m) and smythite (b, R3 m) Fe3S4. For a, Fe3+, Fe2+/Fe3+, and S are marked with blue, red, and yellow spheres, respectively. For b, Fe2+ and Fe3+ ions are represented with blue and turquoise spheres, respectively [18]. @ 2014, American Chemical Society. Crystal structures of pyrrhotite Fe3S4. (c): C2/c; (d): P3121. Fe and S atoms are marked with turquoise and yellow spheres, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

supported carbon coated FeS on carbon cloth film (FeS@C/carbon cloth) using hydrothermal method followed by one step annealing process [24]. The carbon cloth film with interconnected structure possesses outstanding mechanical stability which is helpful for volume holding. Xing et al. prepared FeS microsheet networks via a hydrothermal method and got FeS microspheres by adjusting synthetic conditions. FeS with 2D layered structure grows directly on iron foil forming vertically aligned structures [25]. Wen et al. adopted one-pot hydrothermal method to get self-assembledFeS2 cubes fixed on rGO as anode for LIBs [26]. As shown in Fig. S3, 0.5–1 μm FeS2 microparticles self-assembled from 200 to 300 nm FeS2 cubes are dispersed regularly on the rGO nanosheets. These small FeS2 cubes provide large electrolyte/electrode interfaces to shorten electron/ion's transport path. Ordered arranged FeS2 microparticles are propitious to alleviate volume variation during ion insertion/extraction and capacity fading. He et al. used FeCl3·6H2O as iron source, sodium alkyl sulfate as sulfur source and 3D graphene via one-pot hydrothermal method to obtain cauliflower-likeFeS2@3DGF composite [27].

conductive polymers, graphene, and graphite, can avoid solvable polysulfide produced and diffusing from cathode by forming a stable SEI (solid electrolyte interface) layer during cycling, and are also more convenient for the transfer of electrons and ions [20]. To realize these structures and morphologies, various methods are introduced to synthesize binary iron sulfides. 3.1. Hydrothermal/solvothermal methods Hydrothermal method is a chemical reaction in water in a sealed pressure vessel, which is in fact a type of reaction at both high temperature and pressure. Solvothermal method is developed based on the hydrothermal synthesis. But the solvents for solvothermal method are organic ones. Compared with solid- or gaseous-state methods, liquid methods are better choices to synthesize materials with controlled compositions, morphologies, sizes, and structures by adjusting the thermodynamic and kinetic reaction parameters without harsh conditions such as high temperatures or high pressures [21]. Solvothermal and hydrothermal methods are especially selected to synthesize the electrode active materials. Wu et al. successfully prepared FeS embedded in carbon microspheres (FeS/CM) by solvothermal method with following heat treatment [22]. The SEM images (Fig. S1) indicate that the diameter of the composite with spherical structure is ca. 1.0–2.0 μm and the thickness of the compound whose framework formed by carbon nanosheets is ca. 2.0–5.0 nm. The composite spheres are meaningful to minimize the surface energy. Wang et al. obtained a novel 1D anataseTiO2 modified FeS nanosheets employing hydrothermal method [23]. To obtain the nanosheets assembled nanostructures, annealing temperature was increased to make atoms spread across the interface quickly. As FE-SEM (Fig. S2) evidenced, abundant wire-like nanostructures with tens of micrometers lengths were generated and assembled by FeS nanosheets.FeS@TiO2 nanostructures are beneficial to restrict the volume expansion of active material during cycling and electrochemical performance. Wei et al. prepared a flexible and self-

3.2. Electrospinning Electrospinning is a special technique to prepare fibers. Polymer solution or flux jets and spins under a strong electric field, in which the shapes of droplets at the syringe needle change from sphericity to conicalness, and stretch into filaments from the cone tip. Polymer filaments with nanometer diameter can be obtained in this way. Zhu et al. prepared FeS nanodots (ca. 5–10 nm) (Fig. S4) dispersed on carbon nanowires (around 100 nm) (Fig. S5) using electrospinning and following hydrothermal methods [28]. The void space of the complex, which is occupied by FeS nanodots and carbon nanowires, is greatly beneficial for the conversion reaction and lead to better reversibility. Fei et al. used polyacrylonitrile, iron acetate and sulfur to prepare iron sulfide nanoparticles scattered in electrospun carbon nanofibers [29]. Conductive 3D networks formed by lots of intertwining long nanofibers 44

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Fig. 4. (a) Voltage-dependent capacity curves of Fe-S-CM at 50 mA/g; (b) Cycling performance of pristine Fe-S-CM and Fe-S-CM at different rates [22]. © 2011 Royal Society of Chemistry.

nanoparticles disperse better, which can improve the ion/electron transport and long cycling stability. Zhang et al. obtained small size FeS2 powders via solid-state reaction by mixing FeC2O4·2H2O with S and further ball-milling [35]. FeS2@C/S-GAs was also prepared by Jin et al. via solid state reaction [36]. Apart from the methods motioned above, there are also many other strategies to obtain iron sulfides with nanostructures or coating conductive materials. For example, Xu, et al. obtained FeS@C ultrathin nanosheets by controlled soft-template synthesis [37]. Shangguan et al. used calcination method followed by coprecipitation process to get FeS/ C [38]. Xu et al. prepared FeS nanoparticles@C nanosheets via freezedrying/carbonization approach [39].

can not only avoid the nanoparticles' aggregation and decrease during cycling, but also have a capacity of the volume expansion and ensure electron transport easily. Besides, positive ions get a nearer route to the individual nanoparticle in the electrolyte with the existence of the macrospores among the nonwoven fibers. Cho et al. obtained porous FeS with hollow nanofiber structures by electrospinning method followed with sulfidation [30]. This method is driven by the nanoscale Kirkendall diffusion effect and located on the nanofiber making the structure more stable for batteries to storage energy. 3.3. Direct-precipitation method Direct-precipitation method is widely used to prepare ultrafine particles. The principle is to add precipitant into metal salt solution and generate precipitant in certain condition followed with washing, thermal decomposition process to obtain the superfine product. Different precipitation products can be got by different precipitants. The direct-precipitation method is simple to operate and difficult to introduce impurities because of its low technical requirement. It has many advantages such as high purity, good stoichiometry, low cost, and so on. However, it is hard to wash the anions in the original solution. Besides, the particles obtained vary in size and is less dispersed. Fei et al. successfully prepared FeS@rGO nanocomposite via direct-precipitation synthesis along with a post-annealing method [31]. The FeS@rGO nanocomposite is less uniformly distributed and the clusterization induced by rGO sheets is less pronounced than the FeS crystals. However, the diameter of the majority of the circled FeS nanoparticles wrapped in rGO (∼28 nm) (Fig. S6a) is much smaller than those without rGO (∼150 nm) (Fig. S6b). FeS nanoparticles are wrapped well in the rGO sheets (Fig. S6c), which restricts aggregation of the nanoparticles formed by the precipitation and the post annealing process. Shangguan et al. applied the direct-precipitation method to prepare FeS@rGO nanosheets [32], in which FeS nanoparticles are uniformly fixed on lots of rGO sheets. The rGO sheets with a large surface area can accommodate FeS nanoparticles and prevent its clusterization, and is also beneficial to the fast electrochemical reactions.

4. Electrochemical performance 4.1. FeS Iron sulfides as promising electrode materials for energy storage applications result from their abundant and inexpensive components in nature. They are extensively used as anodes for Li/Na storage devices due to their high Li/Na-insertion voltages, which can reduce electrolyte reduction and SEI layer's formation. Similar with many other types of anode materials, iron sulfides also have several disadvantages have to be resolved. To prepare various nanostructures and carbon coated iron sulfides have been widely introduced to enhance their electrochemical properties. 4.1.1. Applications for LIBs The theoretical capacity (609 mAh/g) of FeS is much higher than that of commercial graphite. Its lithiation and delithiation process can be described as FeS + 2Li+ + 2e– ←→Li2S+Fe. To date, there are many related studies about how to improve its electrochemical properties. Wu et al. prepared FeS microspheres embedded carbon nanosheets show a high capacity and nice high-rate capabilities (Fig. 4) [22]. The first capacity reaches to 1564 mAh/g and 736 mAh/g capacity can be maintained after 50 cycles at 50 mA/g. Its reversible capacities at 0.5, 1, and 5 A/g are 783, 734 and 541 mAh/g, respectively. The improved electrochemical data can be ascribed to the addition of carbon nanosheets. The dispersed carbon nanosheets can improve the surface contact between electrode and electrolyte, and the wettability of electrode. Furthermore, polysulfides are more difficult to dissolve in the electrolyte in the carbon nanosheet-wrapped FeS structure. The introduction of carbon nanosheets can effectively increase the conductivity (Fig. S8). Carbon nanosheets-coated FeS (FeS@C) obtained by Xu et al. delivers the first capacity of 1022 mAh/g [37]. With the formation of SEI layer in the first cycle, the subsequent charge capacity is 635 mAh/g. The small grains and ultrathin thickness of the composite reduce the Li-ions’ diffusion path but increase the interaction area between electrode material and electrolyte. In addition, it shows a stable cycle capacity of 623 mAh/g at the 20th cycle and remaining of 615 mAh/g at the 100th cycle. Wang et al. tried to introduce TiO2

3.4. Solid-state reactions Solid state reaction has the advantages of low-cost, high yield, simple preparation process, and easily scaled up to get target product compared with liquid or gas methods. It usually involves two processes: the reaction of the material at the interphase interface and the migration of the material. Fei et al. got NaCl-FeS2@C at 500 °C in a flowing N2 atmosphere by mixing ferrocene, sulfur, and NaCl, in which NaCl is used as the template [33]. Ferrocene and sulfur become vapor phases at elevated temperature with carbon produced, which can be easily sticked to the surface of template. As shown in Fig. S7, NaCl-FeS2@C have fewer aggregates than FeS2@C composites. Xiao et al. prepared Fe1-xS@CNTs by mixing Fe3C/Fe@CNTs with sulfur and annealing at 630 °C under Ar atmosphere [34]. With inclusion of CNTs, Fe1-xS 45

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

morphologies, in which the material and morphology, synthetic method, and electrochemical data of FeS are briefly summarized. Moreover, to compare the diverse electrochemical performance of FeS anode materials for LIBs more clearly, the recent main achievements selected from Table 2 are depicted as figure (Fig. 5).

nanoparticles into FeS nanosheets [23]. Compared with pure FeS, FeS@TiO2 nanostructures exhibit a much more stable cyclic performance of 87% capacity left and a high Columbic efficiency of 99% after 100 cycles. TiO2 phase enhances the Li+ ion insertion/extraction kinetics at the electrode/electrolyte interface and the diffusion rate of Li+ ion. Wei et al. used carbon coated FeS on carbon cloth without carbon black or current collector as anode directly [24]. At 1.2 C, the initial capacity is 453 mAh/g and about 300 mAh/g capacity remained after 200 cycles (Fig. S9a). When the current densities increased from 0.15 to 7.5 C, the capacities decreased from 560 to 370 mAh/g in 1.0–2.6 V (Fig. S9b). The carbon cloth improves the cycling stability and rate performance by providing larger contact area between electrode and electrolyte, and shortening the diffusion path for both electrons and ions. What's more, the carbon shell on the surface of FeS particles restricts the volume expansion during cycling, and rescues the particles' aggregation. Li et al. explored one-step solid-state approach to obtain FeS particles located in the 3D matrix of porous carbon [40]. The composite shows better electrochemical properties than pure FeS, a high first discharge capacity of 1428.8 mAh/g. After 150 cycles, 624.9 mAh/g capacity can be maintained. It also proves that the introduction of carbon material can improve the experimental capacity and protect the structure of composite. Besides, interconnected porous FeS/C composite was obtained by Xu et al. It can deliver a reversible capacity of around 703 mAh/g after 150 cycles at 1 A/g, and around 530 mA/g at 5 A/g (Fig. S10), the best performance for FeS anode materials until then [39]. All the above mentioned FeS materials are troilite. Rarely investigated pyrrhotite FeS also has been studied for its Li storage ability. Yu et al. synthesized carbon nanotubes filled with FeS nanoparticles [17]. The prepared Fe-S@CNTs delivered a capacity of 674 mAh/g at the 20th cycle and 670 mAh/g at the 65th cycle at 50 mA/g. At 2 A/g, 348 mAh/g capacity can be reached (Fig. S11). Employing in-situ TEM technique, the charge/discharge mechanism was also explored. The results indicated that CNTs can not only accommodate the volume expansion but also offer fast transport Li+ ions paths, and the transport speed of Li+ ions was measured to be around 33.3 nm/s along the CNT wall.

4.1.3. Applications for Li-S batteries Apart from the extensive investigation of FeS for LIBs and SIBs, its application for Li-S batteries is rarely studied. Recently, Liang, et al. prepared NiS2/FeS holey film using electrochemical anodic and CVD treatments, which exhibits nice electrochemical data for Li-S batteries, representing a novel pathway to explore potential electrode materials for Li-S battery. The special structure shows several advantages, including improved ion/mass diffusion, reduced electrode resistance and high cyclability stability [46]. It is expected that there are more related studies to develop various iron sulfides for this promising field. 4.2. FeS2 FeS2 has many advantages, such as abundance in nature, inexpensive cost, low toxicity, and so on. It is applicable for solar cells due to its suitable band gap (Eg = 0.95 eV). It also has been used for electrochemical energy storage for a long time. Especially, the Li/FeS2 primary battery has already commercialized as early as in the 1980's [47]. However, the weak electrochemical reversibility of FeS2 induces that it is unable to be used at ambient temperature commercially [48]. Furthermore, the voltage profiles of FeS2 as anode for LIBs obtained by different authors are largely different [49]. 4.2.1. Applications for LIBs FeS2 used for LIBs has a theoretical capacity as high as 894 mAh/g contributed by four Li+ ions storage per FeS2molecule by a reaction to Li2S and Fe (FeS2 + 4Li+ + 4e– → 2Li2S + Fe). Jin et al. obtained carbon coated FeS2 spheres on S doped graphene (FeS2@C/S-GAs) [50]. Fig. 6 exhibits the C-V curves of pristine FeS2. There is one reduction peak at around 1.15 V in the first anodic scan. In the first cathodic scan, two peaks (∼1.94 and 2.55 V) are observed. In the second reductionoxidation cycle, there are two reduction peaks at ∼1.82 and 1.30 V, and two oxidation peaks at ∼1.98 and 2.50 V. Compared with pristine FeS2, FeS2@C/S-GAs shows an additional peak at ∼0.64 V appeared only in the first cycle in view of the SEI layer's formation and irreversible wastage of electrolyte. Normally, pure FeS2 anode shows fast capacity fade and bad rate capability. To overcome its weak conductivity and structural pulverization, many efforts, such as carbon coating and preparation of nanostructure, have been tried. Fei et al. prepared watersoluble NaCl templated FeS2@C nanocomposites exhibiting great electrochemical performance [33]. It has the first discharge and charge capacities of 1446.7 and 1024.6 mAh/g, respectively, and remains a discharge capacity of 681.8 mAh/g after 100 cycles (Fig. S13a). The low initial Coulombic efficiency (70.8%) is resulted from the SEI layers' formation by the irreversible consumption of Li+ ions. However, the entire Coulombic efficiency among the 100 cycles is above 99% and the composite shows a capacity of 445.7 mAh/g after 100 cycles at 300 mA/g (Fig. S13b). Furthermore, it reveals good rate performance, capacities of ∼820, ∼720, ∼615, and ∼440 mAh/g at 200, 300, 500, and 1000 mA/g, respectively (Fig. S13c). The “sphere on mattress” architecture and the microstructure of FeS2 sandwiched between carbon of NaCl-FeS2@C composite can not only restrict volume changing and active material losing, but also enhance the conductivity and lessen the diffluence of polysulfide by absorbing and boxing it up inside. Du et al. synthesized cauliflower-like FeS2 microspheres wrapped by rGO using a solution route [51]. The anode has a reversible capacity of 1720 mAh/g after 700 cycles at 0.2 A/g, and 340 mAh/g after 800 cycles at 5.0 A/g and 85 °C. Besides, T. Takeuchi et al. studied improved cycling performance of FeS2 via using FeS2-Li2S composite instead [52]. To make clear the reaction mechanism and structure information for

4.1.2. Applications for SIBs Compared with Li resources, Na resources are much cheaper and abundant in nature. FeS has a high theoretical capacity of about 610 mAh/g as the anode for SIBs. But the operation at ambient temperature encounters an important issue as the shuttle effect of NaSx is heavy, resulting in low efficiency and fast capacity fade. Wang et al. synthesized yolk-shell FeS@carbon nanospheres [41]. Carbon shell results in high reversible capacity by improving the conductivity of the active material. In addition, the nanosized FeS slows down the disadvantageous influence of volume change during sodiation/desodiation processes, and improves the reaction rate by increasing the electrode/ electrolyte interface and decreasing the diffusion path of electron/ion. The initial charge and discharge capacity reach to 722 and 1029 mAh/ g, respectively, and it can keep stable capacity of 488 mAh/g after 300 cycles at 0.2 C (Fig. S12a). When the rate increased to 5 C, the capacity can still reach to 452 mAh/g (Fig. S12b). Cho et al. obtained porous FeS nanofibers [42]. Compared with hollow Fe2O3 nanofibers, the porous FeS nanofibers show better electrochemical performance, in which the discharge capacities for the 1st and 150th cycles are 561 and 592 mAh/ g at 500 mA/g, respectively. The capacity increases due to the formation of gel-like film on the active materials by electrolyte degradation. Even at 5 A/g, it has a discharge capacity of 353 mAh/g. The porous morphology of FeS makes good effects of stable structure and excellent Na+ ion storage. Apart from the above discussed FeS anode materials with special morphologies and excellent electrochemical performance, there are also many other studies. Table 2 lists various carbon materials coated or modified FeS materials with versatile hierarchical structures and 46

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Table 2 Electrochemical data of FeS for LIBs and SIBs.a Material and morphology

Synthetic method

Electrochemical data

Ref.

FeS/carbon microsphere FeS nanosheets/ultrathin carbon FeS/rGO nanoparticle

Solvothermal process Soft-template method Direct-precipitation method and post annealing Hydrothermal method Hydrothermal method

736 mAh/g after 50 cycles at 50 mA/g, and 541 mAh/g at 5 A/g 615 mAh/g after 100 cycles (100 mA/g), and 266 mAh/g at 10 C 978 mAh/g after 40 cycles at 100 mA/g, and 200 mAh/g at 1 A/g

[22] [37] [31] [23] [24]

Electrostatic spinning

635 mAh/g after 100 cycles at 100 mA/g, and 160 mAh/g at 4 A/g (Li)365 mAh/g after 100 cycles at 0.15 C, and 370 mAh/g at 7.5 C. (Na) 430 mAh/g after 50 cycles at 0.15 C, and 280 mAh/g at 7.5 C 400 mAh/g after 50 cycles at 61 mA/g, and 322 mAh/g at 10 C

Hydrothermal method Calcination method Electrostatic spinning Freeze-drying/carbonization method Chemical vapor deposition (CVD) system

677 mAh/g after 20 cycles at 100 mA/g, and 150 mAh/g at 2 C 322 mAh/g after 200 cycles at 300 mA/g, and 230 mAh/g at 1500 mA/g 535 mAh/g after 200 cycles at 1500 mA/g 703 mAh/g after 150 cycles at 1 A/g, and 532 mAh/g at 5 A/g 536 mAh/g after 100 cycles at 400 mA/g, and 498 mAh/g at 1 A/g

[25] [38] [29] [39] [49]

In-situ growth

[48]

Solid-state method Electrospinning and subsequent sulfidation

Initial capacity of 1106.9 mAh/g at 100 mA/g, 90% retained for 50 cycles at 500 mA/g 545 mAh/g over 100 cycles at 0.2 C, and 452 mAh/g at 5 C 592 mAh/g after 150 cycles at 500 mA/g, and 353 mAh/g at 5 A/g

Direct-precipitation approach

87.6% capacity left after 300 cycles at 2 C, and 220 mAh/g at 20 C

[32]

Preparation of metal organic matrix and followed carbonization Vulcanization of graphene@Fe3O4-graphene nanoribbons via CVD method One-pot solvothermal reaction Vapor deposition and followed heat treatment

1106.9 and 616.9 mAh/g at 0.1 and 1.1 A/g, respectively; 90% capacity retention for 50 cycles at 0.5 A/g 693 mAh/g at 0.1 A/g; 498 mAh/g after 25 cycles at 1.0 A/g

[43]

550 and 370 mAh/g discharge capacities for the 1st and 2nd cycles 674 mAh/g at the 20th cycle and 670 mAh/g at the 65th cycle at 50 mA/ g, 348 mAh/g capacity at 2 A/g

[45] [17]

FeS nanostructures/TiO2 FeS@C/carbon cloth (Li/Na) FeS nanodots@porous graphitic carbon nanowires FeS microsheet networks FeS/C (alkaline nickel-iron) FeS/carbon nanofiber networks FeS nanoparticles/carbon nanosheets Sandwich graphene-wrapped FeS/graphene nanoribbons FeS nanoparticles coated with N,S co-doped carbon nanostructures Yolk-shell FeS/C nanopheres (Na) Porous FeS nanofibers with numerous nanovoids (Na) FeS/rGO nanosheets (alkaline rechargeable batteries) FeS/N, S-doped carbon Sandwiched graphene@ FeS-graphene nanoribbons Carbon precursor coated FeS microcrystals Pyrrhotite Fe-S@CNTs

a

[28]

[41] [42]

[44]

If no specific noted, all the data were obtained for LIBs. The same below.

attention has been paid to marcasite FeS2, which may be attributed to its metastable feature. Fan et al., studied the electrochemical performance of marcasite FeS2 for the first time [56]. The FeS2/carbon nanofibers composite possesses a capacity of 1399.5 mAh/g after 100 cycles at 0.1 A/g and 782.2 mAh/g at 10 A/g (Fig. S14). The outstanding electrochemical behaviors are attributed to the specific hierarchical structure and the inclusion of carbon nanofibers, which is helpful to improve the redox kinetics and structure stability. All solid-state batteries are receiving increasing interest all over the world in view of their much improved safety [57,58]. To date, FeS2 also has been studied as the electrode material for solid-state LIBs. Whiteley et al. investigated nano-FeS2 domains imbedded into an amorphous LiTiS2 matrix to replace both the ionic conductor solid electrolyte and electronic conductor carbon, which can cycle for 500 cycles at 0.5 C, and the capacity can be fully recovered via simply reducing system rate [59]. Besides, Pelé, et al. studied FeS2 as the all-solid-state thin-film batteries [60]. Yersak et al. designed a solid-state battery architecture, which could realize four electron storage of pyrite [61]. Fig. 5. The electrochemical data of some FeS anode materials for LIBs.

4.2.2. Applications for SIBs FeS2 has a theoretical capacity of 894 mAh/g for SIBs according to the equation FeS2+4Na++4e–→2Na2S+Fe. Because of its semiconductor feature and morphology pulverization in the charging and discharging process, pristine FeS2 as electrode material for SIBs leads to fast capacity fade and poor rate capabilities. Moreover, the products of FeS2 decrease in pace with generating of polysulfides (Na2Sx, 2 < x < 8), which can be dissolved into non-aqueous electrolyte below 0.8 V [62]. To address these issues, it is necessary to obtain nanostructured FeS2 with coating carbon, but also control the discharge cut-off voltage [63,64]. Chen et al. synthesized FeS2 anchoring on rGO aerogel as anode [62]. The composite electrode shows great cyclic performance: remaining capacity is ∼58.03% after 800 cycles at 1 C (900 mA/g) (0.052% loss per cycle) (Fig. S15a). rGO aerogel plays an important role to avoid active materials losing. Fig. S15b exhibits that the capacities at

various batteries are very important, which can be helpful to design desirable battery systems. Butala et al. studied the reaction between FeS2 and Li using operando and ex-situ local structure methods since the local structures at different charge states are consistent with operando PDF and ex-situ XAS data [53], in which several new insights about the cycling mechanism for FeS2-Li batteries are proposed. Besides, Zhang, et al. investigated the mechanism and solution for the capacity fading of Li/FeS2 battery [54]. Yamaguchi et al. simulated an electrochemical energy profile on the cycle model of Li/FeS2 battery using density functional theory (DFT), and the results demonstrated that the reproduction of FeS2 is a more favorable scheme for the full-charge reaction at the cathode, indicating a relatively good cycling performance for the Li/FeS2 battery [55]. All the above mentioned FeS2 are pyrite type, however, little 47

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Fig. 6. C-V curves of pristine FeS2 (a) and FeS2@C/S-GAs (b) [50].

©

2015 Royal Society of Chemistry.

of 2980 mAh/g or 8063 mAh/cm3 for Al metal anode, together with the abundance and safety characters of Al [71]. Mori et al. studied FeS2 as electrode for Al batteries at 55 °C [72]. Several interesting information is given in their work. Firstly, the electronic structural transformation of FeS2 is studied using Fe and S K-edge XANES technique. Secondly, it is proposed that the S-3p orbitals for FeS2 is very important for the reactions. Lastly, after discharged, FeS2can be transformed to low crystalline FeS and amorphous Al2S3. Apart from the above discussion about the electrochemical behaviors of FeS2, there are also many other studies showing interesting or promising electrochemical properties. As listed in Table 3, the main achievements in recent years about FeS2 as anode materials for LIBs or SIBs are shortly summarized, the contents include material and morphology, synthetic method, and electrochemical data. To demonstrate the recent progresses about the lithium storage abilities of FeS2 visually, the cases with relatively better electrochemical data are shown in Fig. 7.

0.2 C are higher than that at 1 C, but performance degradation is worse than that at 1 C. At 0.2 C, there are more Na+ ions translating during cycling and the capacity is higher, but it causes more possibilities of volume change and relatively bad cyclic performance. As shown in Fig. S15c and S15d, compared with pure FeS2, there are several voltage plateaus for FeS2/rGO-A electrode at high rate due to the composite electrode owning higher conductivity. FeS2/rGO-A electrode has more excellent rate performance than pure FeS2 (Fig. S15e) because of its unique 3D conductive networks. In view of rGO-A's outstanding conductivity and unique structure for Na+ ion diffusing, the Rct of FeS2/ rGO-A is obviously decreased than that of FeS2. After the first discharge, produced NaxFeS2 has better conductivity than that of FeS2. Employing ether-based electrolyte, 1.0 M NaCF3SO3 in diglyme to replace traditional carbonate-type ones, FeS2 demonstrated much enhanced ability of Na storage [65]. More than 600 mAh/g capacity and 750 Wh/kg energy density could be achieved at 20 mA/g. When the rate increased to 60 mA/g, 530 mAh/g capacity could be reached and 450 mAh/g capacity remained after 100 cycles. Apart from the above investigations, A. Kitajou et al. also studied the discharge/charge mechanism of FeS2 for SIBs [66].

4.3. Fe3S4 Greigite Fe3S4 was discovered in silt and clay sediments in California in 1964 [100]. It didn't cause much attention due to its thermodynamic metastability [101]. As a semi-metallic-type magnetic material, Fe3S4 has been extensively applied in the field of magnetism [102,103]. Even though it has been studied extensively, the metastable nature of Fe3S4 limits its applications. Fe3S4 decomposes above 250 °C even in the atmosphere of argon [104]. Besides, the volume change during cycling and fast capacity fade are also important issues for its further researches.

4.2.3. Applications in other types of batteries In the periodic table, Mg and Li elements are in the diagonal positions. Mg ion battery (MIB) is potentially researched based on its abundance in nature (∼1100 times higher than that of Li) and cheap price (∼15 times cheaper than Li) [67]. Besides, MIB is much safer than LIB because of the formation of dendrites during the electro-deposition/ dissolution process in the LIB, which means that Mg can be directly used as anode material. However, MIB is still at the concept stage. Especially its separator and electrolyte materials have not been solved well and even the cathode materials research is still in the primary stage. Apart from the above applications, FeS2 also has been studied its potentials for Li-S and Mg-ion batteries [68,69]. Walter et al. synthesized FeS2 nanocrystals as cathode material in the Na-Mg hybrid battery [70]. Fig. S16a presents the cycling data and Coulombic efficiency for 40 cycles at 0.2 C in 0.4–1.95 V after initial discharge to 5 mV with 2 M NaBH4 + 0.2 M Mg(BH4)2 in diglyme as electrolyte. It offers the initial capacity of 225 mAh/g and retains an average capacity of 189 mAh/g after 40 cycles with high Coulombic efficiency of ∼99.8%. Its nice rate data is shown in Fig. S16b, capacities of 225, 180, 170, and 145 mAh/g at the rate of 0.2, 0.5, 1, and 2 C, respectively. Once the current recurs to 0.2 C, the capacity can be recovered to the initial capacity. There is a great potential for development and prospects according to the experiments above. Al secondary batteries are believed as one type of possible choice for large-scale energy storage devices in view of a high theoretical capacity

4.3.1. Applications for LIBs Compared with FeS and FeS2, Fe3S4 are relatively rarely explored its potentials for rechargeable batteries. To date, only less than ten studies happened on it. Fe3S4 has a theoretical capacity of 785 mAh/g as anode for LIBs, which is almost twice that of conventional anode graphite. Li et al. employed hydrothermal method to synthesize high-purity Fe3S4 microcrystals.18 The octahedral shape Fe3S4 microcrystals (∼1 μm) shows nice cycling performance in 0.005–3.0 V at 100 mA/g (Fig. 8a). It possesses the first discharge and charge capacities of 1161 and 1139 mAh/g, respectively. The higher first capacity than the theoretical one is considered to be contributed by a polymeric gel-like film, which is formed by kinetically controlled electrolyte degradation driven by active metal Fe on the particle electrode surface [105,106]. For the fifth cycle, the discharge and charge capacities decrease to 674 and 554 mAh/g, respectively, and the capacity rises again to 563 mAh/g after 100 cycles. Fe3S4 nanoparticles with the lateral dimension in 48

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Table 3 Electrochemical data of FeS2 for LIBs and SIBs. Material and morphology

Synthetic method

Electrochemical data

Ref.

FeS2 crystalline powder FeS2 nanocrystals (Li/Na)

Solid state reaction Solution-phase chemical synthesis

[34] [73]

FeS2/CPAN NaCl-FeS2/C nanocomposite FeS2 nanowires FeS2 nanocrystals FeS2 microspheres Microspheres FeS2/rGO Self-assembled cubes FeS2/rGO FeS2@C/S-GAs Nanocrystal FeS2@Resilient carbon Nanoparticles FeS2@carbon fiber electrode (Li/Na) Nanoparticles FeS2/CNT Core-shell nano-FeS2@N-doped graphene FeS2 decorated S-doped carbon fiber FeS2 nanocrystals in hierarchical porous carbon FeS2@C nanowires Cauliflower-like FeS2@3D graphene foams Cubic FeS2/N-doped graphene FeS2 particle (SIBs) FeS2(SIBs) FeS2 microspheres (∼10 μm) (SIBs) Crystalline FeS2 particles (SIBs) FeS2 nanospheres (SIBs) FeS2@rGO (SIBs) Microspheres FeS2/rGO aerogel (SIBs) FeS2 nanocrystals (SIBs/MIBs) FeS2@C yolk-shell nanoboxes Marcasite FeS2/carbon nanofibers

Solid state reaction Solid state method Solution synthesis followed with sulfidation Solvothermal method Solvothermal method One-step solvothermal method One-pot hydrothermal method Solid state reaction Wet-chemistry based conformal encapsulation Electrospinning

420 mAh/g after 30 cycles at 0.1 C (Na) 500 mAh/g after 400 cycles at 1 A/g, (Li) > 630 mAh/g after 100 cycles at 0.2 A/g 470 mAh/g after 50 cycles at 0.1 C 445.7 mAh/g after 100 cycles at 300 mA/g, and 440 mAh/g at 1000 mA/g 350 mAh/g after 50 cycles at 0.1 C 401.7 mAh/g after 400 cycles at 0.5 A/g, and 285 mAh/g at 5 A/g 540 mAh/g after 100 cycles at 1 A/g, and 318 mAh/g at 8 A/g 970 mAh/g after 300 cycles at 830 mA/g, and 237 mAh/g at 20 C 1001.41 mAh/g after 60 cycles at 100 mA/g 1000 mAh/g after 80 cycles at 0.1 C, and 455 mAh/g at 5 C 110 mAh/g after 50 cycles at 0.1 C (Li) 275 mAh/g at 30 mA/g, (Na) 150 mAh/g at 60 mA/g

Solvothermal method Solvothermal method

525 mAh/g after 1000 cycles at 2000 mA/g, and 345 mAh/g at 5000 mA/g 285 mAh/g at 5 A/g

[81] [82]

One-pot biotemplate synthesis Solvothermal method

689 mAh/g after 100 cycles at 100 mA/g, and 400 mAh/g at 2 A/g 720 mAh/g after 100 cycles at 1 C

[79] [83]

Hydrothermal method One-pot hydrothermal method

570 mAh/g after 100 cycles at 100 mA/g, and 552 mAh/g at 5 A/g 1080.3 mAh/g after 100 cycles at 0.2 C, and 615.1 mAh/g at 5 A/g

[84] [85]

Solution method Hydrothermal method Solid state method Hydrothermal method Solvothermal method One-step solvothermal method One-step hydrothermal method Hydrothermal method Hydrothermal method Etching method One-pot hydrothermal process

849 mAh/g after 100 cycles at 0.1 C First discharge capacity of 771 mAh/g and a reversible capacity of 521 mAh/g 415 mAh/g after 100 cycles at 20 mA/g, and 290 mAh/g at 200 mA/g ∼90% capacity after 20000 cycles, and 170 mAh/g at 20 A/g 460 mAh/g after 800 cycles at 200 mA/g 220 mAh/g after 5000 cycles at 2 A/g 240.5 mAh/g at 250 mA/g, and 192.9 mAh/g at 2 C 235 mAh/g at 5 C Initial discharge capacity of 225 mAh/h 511 mAh/g at 100 mA/g after 100 cycles, and 403 mAh/g at 5 A/g 1399.5 mAh/g after 100 cycles at 0.1 A/g and 782.2 mAh/g at 10 A/g, 573.4 mAh/g after 1000 cycles at 5 A/g 342.4 mAh/g after 800 cycles at 0.5 C, 0.02% capacity fade per cycle 80% capacity retention after 50 cycles, and an energy density of 542 Wh/kg

[86] [87] [88] [89] [90] [64] [91] [92] [70] [63] [56]

Solvothermal method Aerosol spray pyrolysis and followed vacuum sulfidation Chemical bath deposition, electro- deposition and glucose decomposition Solid state reaction Hydrothermal method

439, 340 and 256 mAh/g at 1, 2.5 and 5 C, respectively 1.0 M NaCF3SO3 in diglyme as electrolyte: 600 mAh/g and 750 Wh/kg at 20 mA/g; 530 and 450 mAh/g for the 1st and 100th cycles at 60 mA/g Initial 850 mAh/g at 0.25 °C and 539 mAh/g after 70 cycles at 0.25 °C

[95] [65]

495 mAh/g after 50 cycles Initial 486.1 mAh/g under 1.0–2.4 V at 1 C, and 367 mAh/g after 500 cycles

[97] [98]

Solvothermal method

540 mAh/g after 150 cycles at 1 A/g, and 220 mAh/g at 5 A/g

[99]

FeS2 nanochains FeS2 particles in TFSI–-based ionic liquid electrolyte FeS2@C porous nanooctahedra 100–600 nm FeS2 (SIBs) Porous Co/FeS2-C core/branch nanowire arrays Carbon-coated FeS2 200–300 nm FeS2 octahedral shape particles FeS2 nanocubes

Magnetic-field-assisted aerosol pyrolysis Ball milled commercial FeS2

[74] [32] [75] [80] [76] [77] [78] [36] [79] [80]

[93] [94]

[96]

100 nm have the capacity increasing from 32th cycle, and reach to 495 mAh/g after 100 cycles (Fig. S17) [107]. This phenomenon is considered to be caused by soaking of Fe3S4 in the electrolyte completely, namely, the process of electrode activation [108]. Fig. 8b shows the C-V curves of Fe3S4 microcrystals electrode [18]. In the first cycle, the reduction peak at ∼1.18 V with a shoulder at ∼1.51 V is contributed by the formation of Fe0-Li2S composite. The reduction peak at 0.72 V is related to the SEI film formation. In the subsequent oxidation scan, the material is developed into Li2FeS2 at ∼1.97 V and further into FeS at ∼2.52 V which disappears in the second cycle. In view of the structure change after the 1st cycle and great reversibility of reduced polarization in the process, the cathodic peaks are shifted to more-positive potentials from the second cycle onward. Zheng et al. prepared Fe3S4 mesoporous hollow spheres, which shows a reversible capacity of 750 mAh/g after 100 cycles at 0.2 A/g [109]. Paolella et al. studied the electrochemical performance of colloidal greigite nanoplatelets [110]. Their investigation indicates that greigite crystals can retain their structures during cycling. Besides, Fe3S4 synthesized by Li et al. also shows excellent discharge capacity of 720 mA/g at 1 A/g after 800 cycles and 462 mAh/g maintained at the high current density of 10 A/g. The nanoparticles show stable structure during cycling by wrapped in

Fig. 7. The electrochemical data of some FeS2 anode materials for LIBs.

49

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

Fig. 8. Galvanostatic charge−discharge (a) and C-V (b) curves of Fe3S4 [64].

©

2014 American Chemical Society.

under less than 0.8 V. According to the above contents, there are several feasible solutions can be employed. Nano-engineering of electrode materials is an applicable and useful way to improve powder density of rechargeable batteries. Based on the nanostructures, there are more electrode/electrolyte contacting areas and shorter paths for ions/electrons to transfer, which will result in faster charge/discharge rates and higher power. In addition, it is more advantageous to hold stable structure from volume changing during the ion insertion/extraction and some other related reactions to keep longer cycle life. Coating conductive materials is another available method to achieve better electrochemical performance. Especially, carbon components as composite electrodes can not only prevent the dissolution of polysulfides, but also increase the conductivity to obtain better cyclability. For the further studies of iron sulfides as electrode materials, more attention should be paid to design their structures and morphologies. Active materials with porous, fibroid or other hierarchical nanostructures are beneficial to keep stable capacities. By compositing with different forms of carbon sources, it can not only protect active materials' lose, but also increase the conductivities. In addition, choosing a suitable cut-off voltage is propitious to better cycling performance. Moreover, it is strongly proposed that more investigations should be performed on other rarely or totally not studied polymorphs for iron sulfides. If all these polymorphs can be profoundly explored, much more reasonable design iron sulfides-type electrode materials should be possible, and the relationship between their crystal structures and electrochemical properties can be discussed. Finally, it should be mentioned that these iron sulfides are relatively rare to be investigated their potentials in post-lithium types of batteries, which deserves to pay much attention in the future.

rGO via one-pot hydrothermal method [111]. 4.3.2. Applications for SIBs Similar with FeS and FeS2, nano-scale Fe3S4 active materials have been interested by researchers as anodes for SIBs [112]. Li et al. obtained 100–200 nm octahedral-like Fe3S4 [113]. It delivers the discharge capacity of 571 mAh/g, and the capacities of 401 mAh/g at 5 A/ g after 1000 cycles and 275 mAh/g at 20 A/g after 3500 cycles (Fig. S18). 4.4. Fe7S8 The investigation of Fe7S8 as electrode materials for batteries are also very limited. Huang et al. prepared carbon-film-coated Fe7S8 nanorods with core-shell architecture via hydrothermal method followed with heat treatment [114]. The capacity of Fe7S8/C is higher than the theoretical ones of FeS (609 mAh/g) and FeS2 (894 mAh/g) because of the polymeric gel-like film's formation and the insertion of Li+ ions [115,116]. Besides, Fe7S8/C exhibits a high first discharge capacity of 1072 mAh/g and1148 mAh/g capacity left after 170 cycles at 0.5 A/g. The increased capacity results from the rearrangement and decreasing nanosizes of active materials structures after the long terms of activation process, which can provide more active sites for Li+ ions to insert and extract [20]. In addition, core-shell Fe7S8/C nanospheres prepared by Zhang et al. also show the relatively stable cycle performance, 397 mAh/g capacity remained after 200 cycles at 0.1 A/g and 695 mAh/g left at the 50th cycle at 0.1 A/g [117]. The structure with 30 nm nanospheres is beneficial for Li+ ions diffusion. Furthermore, the direct contact between electrolyte and active materials can be reduced by the thickness of carbon shell and prevent the active materials being dissolved in the electrolyte. Recently, Li, et al. studied the electrochemical behaviors of carbon layer coated uniform hierarchical Fe7S8 nanostructures [118]. It exhibits capacities of 751 and 497 mAh/g after 100 cycles at 100 mA/g for LIBs and SIBs, respectively (Fig. S19).

Acknowledgements We gratefully acknowledge the financial support by the National Natural Science Foundation of China (21673203, 21771159), the Higher Education Science Foundation of Jiangsu Province (15KJB150031), Natural Science Foundation of Yangzhou (YZ2016122), Qing Lan Project and the Priority Academic Program Development of Jiangsu Higher Education Institutions.

5. Conclusions Rechargeable batteries are wonderful devices to storage energy in view of their high energy and power densities. Among various anode materials, iron sulfides demonstrate many advantages, such as abundance in nature, inexpensive cost, low toxicity, and high conductivities than iron oxides. Compared with FeS and FeS2, greigite Fe3S4 as electrode material has more stable cycle performance. However, its synthesis is more rigorous which should be very strict to control the reaction temperature and time. Besides, the cut-off voltage of Fe3S4 is restrained for better electrochemical performance. It is also need to control discharge cut-off voltage when FeS2 is used as electrode materials for NIBs in view of the formation of polysulfides (Na2Sx, 2 < x < 8), which can be dissolved into non-aqueous electrolyte in the conversion reaction

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.jpowsour.2018.01.022. References [1] J.M. Tarascon, M. Armand, Nature 414 (2001) 359–367. [2] W. Liu, M.S. Song, B. Kong, Y. Cui, Adv. Mater. 29 (2017) 1603436. [3] W.D. Li, B.H. Song, A. Manthiram, Chem. Soc. Rev. 46 (2017) 3006–3059.

50

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

[4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

[19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52]

[53]

[54] [55] [56]

[57] A. Manthiram, X.W. Yu, S.F. Wang, Nat. Rev. Microbiol. 2 (2017) 16103. [58] D. Larcher, J.M. Tarascon, Nat. Chem. 7 (2015) 19–29. [59] J.M. Whiteley, S. Hafner, S.S. Han, S.C. Kim, K.H. Oh, S.H. Lee, Adv. Eng. Mater. (2016) 1600495. [60] V. Pelé, F. Flamary, L. Bourgeois, B. Pecquenard, F.L. Cras, Electrochem. Commun. 51 (2015) 81–84. [61] T.A. Yersak, H.A. Macpherson, S.C. Kim, V.D. Le, C.S. Kang, S.B. Son, Y.H. Kim, J.E. Trevey, K.H. Oh, C. Stoldt, S.H. Lee, Adv. Eng. Mater. 3 (2013) 120–127. [62] W.H. Chen, S.H. Qi, L.Q. Guan, C.T. Liu, S.Z. Cui, C.Y. Shen, L.W. Mi, J. Mater. Chem. 5 (2017) 5332–5341. [63] Z.M. Liu, T.C. Lu, T. Song, X.Y. Yu, X.W. Lou, U. Paik, Energy Environ. Sci. 10 (2017) 1576–1580. [64] K. Zhang, M. Park, L.M. Zhou, G.H. Lee, J. Shin, Z. Hu, S.L. Chou, J. Chen, Angew. Chem. Int. Ed. 55 (2016) 12822–12826. [65] Y.J. Zhu, L.M. Suo, T. Gao, X.L. Fan, F.D. Han, C.S. Wang, Electrochem. Commun. 54 (2015) 18–22. [66] A. Kitajou, J. Yamaguchi, S. Hara, S. Okada, J. Power Sources 247 (2014) 391–395. [67] J. Emsley, Oxford University Press (2011). [68] K. Sun, C.A. Cama, R.A. DeMayo, D.C. Bock, X. Tong, D. Su, A.C. Marschilok, K.J. Takeuchi, E.S. Takeuchi, H. Gan, J. Electrochem. Soc. 164 (2017) A6039–A6046. [69] Y. Zhang, J.J. Xie, Y.L. Han, C.L. Li, Adv. Funct. Mater. 25 (2015) 7300–7308. [70] M. Walter, K.V. Kravchyk, M. Ibáñez, M.V. Kovalenko, Chem. Mater. 27 (2015) 7452–7458. [71] Q.F. Li, N.J. Bjerrum, J. Power Sources 110 (2002) 1–10. [72] T. Mori, Y. Orikasa, K. Nakanishi, C. Kezheng, M. Hattori, T. Ohta, Y. Uchimoto, J. Power Sources 313 (2016) 9–14. [73] M. Walter, T. Zünd, M.V. Kovalenko, Nanoscale 7 (2015) 9158–9163. [74] S.B. Son, T.A. Yersak, D.M. Piper, S.C. Kim, C.S. Kang, J.S. Cho, S.S. Suh, Y.U. Kim, K.H. Oh, S.H. Lee, Adv. Eng. Mater. 4 (2014) 1300961. [75] L.S. Li, M. Cabán-Acevedo, S.N. Girard, S. Jin, Nanoscale 6 (2014) 2112–2118. [76] Z. Hu, K. Zhang, Z.Q. Zhu, Z.L. Tao, J. Chen, J. Mater. Chem. 3 (2015) 12898–12904. [77] H.T. Xue, Y.W. Yu, J. Qing, X. Yang, J. Xu, Z.P. Li, M.L. Sun, W.P. Kang, Y.B. Tang, C.S. Lee, J. Mater. Chem. 3 (2015) 7945–7949. [78] X. Wen, X.L. Wei, L.W. Yang, P.K. Shen, J. Mater. Chem. 3 (2015) 2090–2096. [79] T.S. Yoder, M. Tussing, J.E. Cloud, Y.G. Yang, J. Power Sources 274 (2015) 685–692. [80] Y.J. Zhu, X.L. Fan, L.M. Suo, C. Luo, T. Gao, C.S. Wang, ACS Nano 10 (2015) 1529–1538. [81] L. Xu, Y.J. Hu, H.X. Zhang, C.Z. Li, ACS Sustain. Chem. Eng. 4 (2016) 4251–4255. [82] R. Tan, J.L. Yang, J.T. Hu, K. Wang, Y. Zhao, F. Pan, Chem. Commun. (J. Chem. Soc. Sect. D) 52 (2016) 986–989. [83] X. Xu, T.W. Cai, Z. Meng, H.J. Ying, Y. Xie, X.L. Zhu, W.Q. Han, J. Power Sources 331 (2016) 366–372. [84] F.F. Zhang, C.L. Zhang, G. Huang, D.M. Yin, L.M. Wang, J. Power Sources 328 (2016) 56–64. [85] J.R. He, Q. Li, Y.F. Chen, C. Xu, K.R. Zhou, X.Q. Wang, W.L. Zhang, Y.R. Li, Carbon 114 (2017) 111–116 1. [86] W.D. Qiu, J. Xia, H.M. Zhong, S.X. He, S.H. Lai, L.P. Chen, Electrochim. Acta 137 (2014) 197–205. [87] Z. Shadike, Y.N. Zhou, F. Ding, L. Sang, K.W. Nam, X.Q. Yang, Z.W. Fu, J. Power Sources 260 (2014) 72–76. [88] Y.J. Zhu, L.M. Suo, T. Gao, X.L. Fan, F.D. Han, C.S. Wang, Electrochem. Commun. 54 (2015) 18–22. [89] Z. Hu, Z.Q. Zhu, F.Y. Cheng, K. Zhang, J.B. Wang, C.C. Chen, J. Chen, Energy Environ. Sci. 8 (2015) 1309–1316. [90] K.Y. Chen, W.X. Zhang, L.H. Xue, W.L. Chen, X.H. Xiang, M. Wan, ACS Appl. Mater. Interfaces 9 (2017) 1536–1541. [91] W.H. Chen, S.H. Qi, M.M. Yu, X.M. Feng, S.Z. Cui, Electrochim. Acta 230 (2017) 1–9. [92] W.H. Chen, S.H. Qi, L.Q. Guan, C.T. Liu, S.Z. Cui, C.Y. Shen, L.W. Mi, J. Mater. Chem. 5 (2017) 5332–5341. [93] J. Yang, M.N. Liu, Z.H. Wei, Z.H. Pan, Y.C. Qiu, F.M. Ye, Y.L. Yang, X.L. Zhao, L.M. Sheng, Y.G. Zhang, Electrochim. Acta 211 (2016) 671–678. [94] T. Evans, D.M. Piper, S.C. Kim, S.S. Han, V. Bhat, K.H. Oh, S.H. Lee, Adv. Mater. 26 (2014) 7386–7392. [95] J. Liu, Y.R. Wen, Y. Wang, P.A. van Aken, J. Maier, Y. Yu, Adv. Mater. 26 (2014) 6025–6030. [96] F. Cao, G.X. Pan, J. Chen, Y.J. Zhang, X.H. Xia, J. Power Sources 303 (2016) 35–40. [97] D. Zhang, Y.J. Mai, J.Y. Xiang, X.H. Xia, Y.Q. Qiao, J.P. Tu, J. Power Sources 217 (2012) 229–235. [98] S. Cheng, J. Wang, H.Z. Lin, W.F. Li, Y.C. Qiu, Z.Z. Zheng, X.L. Zhao, Y.G. Zhang, J. Mater. Sci. 52 (2017) 2442–2451. [99] W.L. Liu, X.H. Rui, H.T. Tan, C. Xu, Q.Y. Yan, H.H. Hng, RSC Adv. 4 (2014) 48770–48776. [100] B.J. Skinner, R.C. Erd, F.S. Grimaldi, Am. Mineral. 49 (1964) 543–555. [101] I. Vasiliev, C. Franke, J.D. Meeldijk, M.J. Dekkers, C.G. Langereis, W. Krijgsman, Nat. Geosci. 1 (2008) 782–786. [102] Z. He, H. Yu, X. Zhou, X. Li, J. Qu, Adv. Funct. Mater. 16 (2006) 1105–1111. [103] I. Lyubutin, S. Starchikov, C.R. Lin, S.Z. Lu, M.O. Shaikh, K. Funtov, T. Dmitrieva, S. Ovchinnikov, I. Edelman, R. Ivantsov, J. Nanoparticle Res. 15 (2013) 1–13. [104] M.J. Dekkers, H.F. Passier, M.A.A. Schoonen, Geophys. J. Int. 141 (2000) 809–819.

J.W. Choi, D. Aurbach, Nat. Rev. Microbiol. 1 (2016) 16013. G. Chen, L.T. Yan, H.M. Luo, S.J. Guo, Adv. Mater. 28 (2016) 7580–7602. M.N. Obrovac, V.L. Chevrier, Chem. Rev. 114 (2014) 11444–11502. N. Nitta, G. Yushin, Part. Part. Syst. Char. 31 (2014) 317–336. N. Mahmood, T.Y. Tang, Y.L. Hou, Adv. Eng. Mater. (2016) 1600374. J.T. Hu, W. Li, Y.D. Duan, S.H. Cui, X.H. Song, Y.D. Liu, J.X. Zheng, Y. Lin, F. Pan, Adv. Eng. Mater. 7 (2017) 1601894. Y.H. Lee, J. Min, K. Lee, S. Kim, S.H. Park, J.W. Choi, Adv. Eng. Mater. 7 (2017) 1602147. X.Y. Yu, L. Yu, X.W. Lou, Adv. Eng. Mater. 6 (2016) 1501333. L.W. Ji, Z. Lin, M. Alcoutlabi, X.W. Zhang, Energy Environ. Sci. 4 (2011) 2682–2699. X.D. Xu, W. Liu, Y. Kim, J. Cho, Nano Today 9 (2014) 604–630. C.H. Lai, M.Y. Lu, L.J. Chen, J. Mater. Chem. 22 (2012) 19–30. Z. Hu, Q.N. Liu, S.L. Chou, S.X. Dou, Adv. Mater. (2017) 1700606. X.H. Rui, H.T. Tan, Q.Y. Yan, Nanoscale 6 (2014) 9889–9924. W.J. Yu, C. Liu, L.L. Zhang, P.X. Hou, F. Li, B. Zhang, H.M. Cheng, Adv. Sci. 3 (2016) 1600113. G.W. Li, B.M. Zhang, F. Yu, A.A. Novakova, M.S. Krivenkov, T.Y. Kiseleva, L. Chang, J.C. Rao, A.O. Polyakov, G.R. Blake, R.A.D. Groot, T.T.M. Palstra, Chem. Mater. 26 (2014) 5821–5829. H. Lux, Handbook of Preparative Inorganic Chemistry, in: G. Brauer (Ed.), second ed., Academic Press, 1963. C.D. Wang, M.H. Lan, Y. Zhang, H.D. Bian, M.F. Yuen, K. Ostrikov, J.J. Jiang, W.J. Zhang, Y.Y. Li, J. Lu, Green Chem. 18 (2016) 3029–3039. M.R. Gao, Y.F. Xu, J. Jiang, S.H. Yu, Chem. Soc. Rev. 42 (2013) 2986–3017. B. Wu, H.H. Song, J.S. Zhou, X.H. Chen, Chem. Commun. (J. Chem. Soc. Sect. D) 47 (2011) 8653–8655. X.F. Wang, Q.Y. Xiang, B. Liu, L.J. Wang, T. Luo, D. Chen, G.Z. Shen, Sci. Rep. 3 (2013) 2007. X. Wei, W.H. Li, J.A. Shi, L. Gu, Y. Yu, ACS Appl. Mater. Interfaces 7 (2015) 27804–27809. C.C. Xing, D. Zhang, K. Cao, S.M. Zhao, X. Wang, H.Y. Qin, J.B. Liu, Y.Z. Jiang, L. Meng, J. Mater. Chem. 3 (2015) 8742–8749. X. Wen, X.L. Wei, L.W. Yang, P.K. Shen, J. Mater. Chem. 3 (2015) 2090–2096. J.R. He, Q. Li, Y.F. Chen, C. Xu, K.R. Zhou, X.Q. Wang, W.L. Zhang, Y.R. Li, Carbon 114 (2017) 111–116. C.B. Zhu, Y.R. Wen, P.A.V. Aken, J. Maier, Y. Yu, Adv. Funct. Mater. 25 (2015) 2335–2342. L. Fei, B.P. Williams, S.H. Yoo, J.M. Carlin, Y.L. Joo, Chem. Commun. (J. Chem. Soc. Sect. D) 52 (2016) 1501–1504. J.S. Cho, J.S. Park, Y.C. Kang, Nano Rev. 10 (2017) 897–907. L. Fei, Q.L. Lin, B. Yuan, G. Chen, P. Xie, Y.L. Li, Y. Xu, S.G. Deng, S. Smirnov, H.M. Luo, ACS Appl. Mater. Interfaces 5 (2013) 5330–5335. E. Shangguan, L.T. Guo, F. Li, Q. Wang, J. Li, Q.M. Li, Z.R. Chang, X.Z. Yuan, J. Power Sources 327 (2016) 187–195. L. Fei, Y.F. Jiang, Y. Xu, G. Chen, Y.L. Li, X. Xu, S.G. Deng, H.M. Luo, J. Power Sources 265 (2014) 1–5. Y. Xiao, J.Y. Hwang, I. Belharouak, Y.K. Sun, ACS Energy Lett 2 (2017) 364–372. D. Zhang, J.P. Tu, J.Y. Xiang, Y.Q. Qiao, X.H. Xia, X.L. Wang, C.D. Gu, Electrochim. Acta 56 (2011) 9980–9985. F.Y. Jin, Y. Wang, J. Mater. Chem. 3 (2015) 14741–14749. C. Xu, Y. Zeng, X.H. Rui, N. Xiao, J.X. Zhu, W.Y. Zhang, J. Chen, W.L. Liu, H.T. Tan, H.H. Hng, Q.Y. Yan, ACS Nano 6 (2012) 4713–4721. E. Shangguan, F. Li, J. Li, Z.R. Chang, Q.M. Li, X.Z. Yuan, H.J. Wang, J. Power Sources 291 (2015) 29–39. Y.X. Xu, W.Y. Li, F. Zhang, X.L. Zhang, W.J. Zhang, C.S. Lee, Y.B. Tang, J. Mater. Chem. 4 (2016) 3697–3703. S.P. Guo, J.C. Li, Z. Ma, Y. Chi, H.G. Xue, J. Mater. Sci. 52 (2017) 2345–2355. Y.X. Wang, J. Yang, S.L. Chou, H.K. Liu, W.X. Zhang, D.Y. Zhao, S.X. Dou, Nat. Commun. 6 (2015) 8689. J.S. Cho, Y.C. Kang, Nano Rev. 10 (2017) 897–907. P. Ramakrishnan, S.H. Baek, Y. Park, J.H. Kim, Carbon 115 (2017) 249–260. L. Li, C.T. Gao, A. Kovalchuk, Z.W. Peng, G.D. Ruan, Y. Yang, H.L. Fei, Q.F. Zhong, Y.L. Li, J.M. Tour, Nano Rev. 9 (2016) 2904–2911. D.T. Tran, S.S. Zhang, J. Mater. Chem. 3 (2015) 12240–12246. K. Liang, K. Marcus, S.F. Zhang, L. Zhou, Y.L. Li, S.T. De Oliveira, N. Orlovskaya, Y.H. Sohn, Y. Yang, Adv. Eng. Mater. (2017) 1701309. S.S. Zhang, D.T. Tran, Electrochim. Acta 176 (2015) 784–789. F.F. Zhang, C.L. Wang, G. Huang, D.M. Yin, L.M. Wang, J. Power Sources 328 (2016) 56–64. S.S. Zhang, J. Mater. Chem. 3 (2015) 7689–7694. F.Y. Jin, Y. Wang, J. Mater. Chem. 3 (2015) 14741–14749. Y. Du, S.P. Wu, M.B. Huang, X.D. Tian, Chem. Eng. J. 326 (2017) 257–264. T. Takeuchi, H. Kageyama, K. Nakanishi, Y. Inada, M. Katayama, T. Ohta, H. Senoh, H. Sakaebe, T. Sakai, K. Tatsumi, H. Kobayashi, J. Electrochem. Soc. 159 (2012) A75–A84. M.M. Butala, M. Mayo, V.V.T. Doan-Nguyen, M.A. Lumley, C. Göbel, K.M. Wiaderek, O.J. Borkiewicz, K.W. Chapman, P.J. Chupas, M. Balasubramanian, G. Laurita, S. Britto, A.J. Morris, C.P. Grey, R. Seshadri, Chem. Mater. 29 (2017) 3070–3082. S.S. Zhang, D.T. Tran, J. Electrochem. Soc. 163 (2016) A792–A797. Y. Yamaguchi, T. Takeuchi, H. Sakaebe, H. Kageyama, H. Senoh, T. Sakai, K. Tatsumi, J. Electrochem. Soc. 157 (2010) A630–A635. H.H. Fan, H.H. Li, K.C. Huang, C.Y. Fan, X.Y. Zhang, X.L. Wu, J.P. Zhang, ACS Appl. Mater. Interfaces 9 (2017) 10708–10716.

51

Journal of Power Sources 379 (2018) 41–52

Q.-T. Xu et al.

11156–11165. [113] Q.D. Li, Q.L. Wei, W.B. Zuo, L. Huang, W. Luo, Q.Y. An, V.O. Pelenovich, L.Q. Mai, Q.J. Zhang, Chem. Sci. 8 (2017) 160–164. [114] W. Huang, S. Li, X.Y. Cao, C.Y. Hou, Z. Zhang, J.K. Feng, L.J. Ci, ACS Sustain. Chem. Eng. 5 (2017) 5039–5048. [115] B. Chen, E.Z. Liu, F. He, C.S. Shi, C.N. He, J.J. Li, N.Q. Zhao, Nanomater. Energy 26 (2016) 541–549. [116] E.Z. Liu, J.M. Wang, C.S. Shi, N.Q. Zhao, C.N. He, J.J. Li, J.Z. Jiang, ACS Appl. Mater. Interfaces 6 (2014) 18147–18151. [117] K.L. Zhang, T.W. Zhang, J.W. Liang, Y.C. Zhu, N. Lin, Y.T. Qian, RSC Adv. 5 (2015) 14828–14831. [118] S.Y. Li, B.H. Qu, H. Huang, P. Deng, C.H. Xu, Q.H. Li, T.H. Wang, Electrochim. Acta 247 (2017) 1080–1087.

[105] J.S. Xu, Y.J. Zhu, ACS Appl. Mater. Interfaces 4 (2012) 4752–4757. [106] Y. Yu, C.H. Chen, J.L. Shui, S. Xie, Angew. Chem. Int. Ed. 44 (2005) 7085–7089. [107] T.T. Li, H.H. Li, Z.N. Wu, H.X. Hao, J.L. Liu, T.T. Huang, H.Z. Sun, J.P. Zhang, H. Zhang, Z.X. Guo, Nanoscale 7 (2015) 4171–4178. [108] L. David, R. Bhandavat, G. Singh, ACS Nano 8 (2014) 1759–1770. [109] H.X. Huai, Q.F. Dong, M.S. Zheng, C.C. Wang, J. Mater. Chem. 2 (2014) 19882–19888. [110] A. Paolella, C. George, M. Povia, Y. Zhang, R. Krahne, M. Gich, A. Genovese, A. Falqui, M. Longobardi, P. Guardia, T. Pellegrino, L. Manna, Chem. Mater. 23 (2011) 3762–3768. [111] S.-P. Guo, J.-C. Li, J.-R. Xiao, H.-G. Xue, ACS Appl. Mater. Interfaces 9 (2017) 37694–37701. [112] A. Douglas, R. Carter, L. Oakes, K. Share, A.P. Cohn, C.L. Pint, ACS Nano 9 (2015)

52