Iron Fluoride Packaged into 3D Order Mesoporous Carbons as High-Performance Sodium-Ion Battery Cathode Material

The FeF 3 · 0.33H 2 O nanoparticles packaged into three-dimensional order mesoporous carbons (3D-OMCs) as cathode material of sodium-ion batteries (SIBs) was deliberately designed and fabricated by a facile nanocasting technique and mesoporous silica KIT-6 template. The structure, morphology, elemental distribution and electrochemical performance of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite are investigated by X-ray diffraction (XRD), ﬁeld emission scanning electron microscopy (FE-SEM), transmission electron microscope (TEM), energy-dispersive X-ray spectroscope (EDS), Raman spectroscopy and electrochemical measure- ment. The results show that the as-synthesized FeF 3 · 0.33H 2 O nanoparticles are perfectly packaged in 3D-OMCs matrix, and the size and morphology of FeF 3 · 0.33H 2 O nanoparticles can be effectively controlled. Furthermore, it has been found that the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite can deliver a high ﬁrst discharge capacity of 386 mAh g − 1 and excellent capacity reservation after 100 cycles at a rate of 20 mA g − 1 in the voltage range of 1.0–4.0 V. Especially, even up to 100 mA g − 1 , the discharge capacity is still as high as 201 mAh g − 1 , indicating a remarkable rate capability. The excellent electrochemical properties of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite can be because the 3D mesoporous structure of 3D-OMCs

In recently years, lithium-ion batteries (LIBs) have been generally used as an effective energy storage approach in the consumer electronics and electric vehicles (EVs) because of their high energy density, long cycle life and environmentally friendly. 1,2 However, the limited resource and high price of lithium salts restrict its further application and development, thus seeking alternative new energy materials have become a hot topic of concern and research. Sodium ion batteries (SIBs) have aroused considerable attention owing to the same high energy density as LIBs, abundant sodium resource, low price and appropriate redox potential for battery applications (E Na + /Na = −2.71 V only 0.3 V above that of lithium). [3][4][5] In addition, sodium, as the second lightest and smallest alkali metal next to lithium, is similar to lithium in physical and chemical properties. Therefore, there is no doubt that SIBs will occupie a major position in the energy storage system in the future.
Just like LIBs, the cathode material research is also the main constraint on development of SIBs. 6,7 Among the various SIBs cathode materials, iron-based fluorides, such as FeF 3 , FeF 3 · 0.5H 2 O, Fe 2 F 5 · H 2 O and FeF 3 · 0.33H 2 O, have caused tremendous interest due to high operational voltage (average ∼2.74 V vs. Na/Na + ), nontoxicity and low-cost. 8 Especially, in the family of iron-based fluorides, the nanoscale FeF 3 · 0.33H 2 O has a great prospect because its unique tunnel structure and high theoretical specific capacity. The tunnel structure is beneficial to electrolyte ion penetration, Na + transport, and interfacial processes. In addition, the high theoretical capacity of 237 mAh g −1 (1e − transfer) and 712 mAh g −1 (3e − transfer) of FeF 3 · 0.33H 2 O is clearly much higher than the current commercialized cathode materials such as LiMn 2 O 4 (∼120 mAh g −1 ), LiCoO 2 (∼140 mAh g −1 ), LiNi 1/3 Co 1/3 Mn 1/3 O 2 (∼160 mAh g −1 ) and LiFePO 4 (∼160 mAh g −1 ). [9][10][11][12][13][14][15] Despite these advantages, FeF 3 · 0.33H 2 O with low electronic and ionic conductivities, which is caused by the ionic nature of metalfluorine bonding, obstructs its application in practical energy storage devices. 16 Mixed with a variety of carbonaceous materials, such as graphene, 17,18 acetylene black, 19 carbon nanotube, 20 reduced graphene oxide (rGO), 21,22 and graphite, 23 is one of the major technologies to solve above obstacles because it can speed up the reaction kinetics through shortening the diffusion distances of electron and ion, z E-mail: wxianyou@yahoo.com and alleviate structure comminution due to volume expansion. 24 For example, Chung et al. 21 achieved a high discharge capacity of 266 mAh g −1 at 0.05 C and a capacity of 242 mAh g −1 after 100 cycles between 1.5 V and 4.5 V via non-aqueous precipitation method to fabricate FeF 3 · 0.5H 2 O/rGO with an open-framework nanostructure as a cathode material of SIBs. To further improve the rate capability, Sun et al. 17 prepared FeF 3 · 0.33H 2 O/GNS as the cathode material of LIBs, and obtained an outstanding rate capability of 115 mAh g −1 after 250 cycles at a high rate over 10 C. The prominent electrochemical performance can be attributed to addition of the graphene nanosheet, which can establish interleaved electron transfer channel and ensure the structural stability of the electrode material during charge/discharge cycling. Recently, Xiao et al. 25 synthesized a threedimensional (3D) hybrid nanocomposite of Fe 2 F 5 · H 2 O (FF) loaded in N-doped porous carbon (NPC) by a ionic liquid method as cathode material of LIBs, owing to the 3D structure and NPC matrix the high reversible capacity of 185 mAh g −1 at 0.25 C and 163 mAh g −1 at 0.5 C between 1.7 and 4.5 V (vs. Li/Li + ) at room temperature can be obtained. However, compared with reduced graphene oxide, N-doped porous carbon and other carbon materials, 3D-OMCs has the advantage of uniform pore size, high specific surface area and interconnected mesporous structure. 26 Accordingly, 3D-OMCs are a kind of promising conductive carbon matrix for optimizing the low conductivity electrode material performances.
In our previous studies, a series of iron-based fluoride cathode materials, such as FeF 3 · 0.33H 2 O/C nanocomposite, Ti-doped Fe 1-x Ti x F 3 · 0.33H 2 O (x = 0, 0.06, 0.08, 0.10) compounds and Fe 2 F 5 · H 2 O composite have been prepared and achieved excellent electrochemical performance. [27][28][29] Especially, MWCNTs-wired Fe 2 F 5 · H 2 O has been fabricated by an ionic liquid precipitation method as the cathode material of SIBs, and the discharge capacity can still keep about 115.0 mAh g −1 after 50 cycles even the current density increases up to 100 mA g −1 , and the corresponding capacity retention is 90.2%. 30 Herein, we reasonably designed and synthesized FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite by nanocasting technique and mesoporous silica KIT-6 template, in which FeF 3 · 0.33H 2 O nanoparticles were packaged into 3D-OMCs matrix.
[Bmim] [BF 4 ] is used as environmentally friendly and thermal stability fluorine source. 3D-OMCs as a carbon source not only can promote the transmission of electrons and shorten the diffusion path of Na + , but also can guarantee the structural stability and prevent nanoparticles aggregation during the Na + intercalation and deintercalation processes. Meanwhile, the 3D mesoporous structure also can offer a large surface area so that meet the demand for high active material loading and enhance the contact area of the electrolyte-electrode interface. The structures, morphologies, and electrochemical performances of the as-prepared FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite are studied in detail.

Material Preparation
Synthesis of mesoporous silica KIT-6 template.-All chemicals were used without further purification. 4.0 g of Pluronic P123 (EO 20 PO 70 EO 20 , Aldrich) was dissolved in 144 mL of deionized water and 6.8 mL of HCl (37 wt%) with vigorous stirring at 35 • C. Next, 4.9 mL of n-butanol was added into the solution and continuously stirred for 1 h. Subsequently, 9 mL of tetraethoxysilane (TEOS) was added into the above solution with vigorous stirring at 35 • C for 24 h. After the mixture was transferred into a 250 mL polytetrafluoroethylene beaker, it was continuously stirred for 24 h at 100 • C. The resultant product was filtered, washed, dried, and finally calcined at 550 • C for 6 h in air to obtain a mesoporous silica KIT-6 template. 31 Synthesis of 3D-ordered mesoporous carbons.-1.0 g of KIT-6 and 0.1537 g of toluene-p-sulfonic acid were dissolved in 20 mL ethanol solution, then 1.0 g of furfuryl alcohol was dispersed in the above solution with vigorous stirring 2 h at room temperature. Subsequently, the solution was kept in the oven at 80 • C and 160 • C for 2 h and 4 h, respectively. Finally, The resulted 3D-OMCs materials was completely carbonized at 600 • C for 3 h at a heating rate of 5 • C min −1 in an argon flow. The resulting black composite was stirred in a 5% HF water/ethanol solution at room temperature for 8 h to remove the KIT-6 template. 32 3 · 0.33H 2 O@3D-OMCs nanocomposite.-FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite was synthesized by nanocasting technique which 3D-OMCs as the template, and the synthetic route is shown in Scheme 1. Firstly, 0.5 g of 3D-OMCs powder was dispersed in concentrated HNO 3 (60 mL, 68%) solution and refluxed for 1 h at 80 • C to induce hydrophilicity. Secondly, 50 mg of 3D-OMCs were fully dispersed in 50 mL of alcohol by a strong ultrasonic treatment for 2 h. Subsequently, BmimBF 4 (1-Butyl-3-methylimidazolium tetrafluoroborate) (10 mL, 99%,) and

Synthesis of FeF
Fe(NO 3 ) 3 · 9H 2 O (1.0 g, 99.99%) were added to the above solution and stirred at 100 • C for 12 h. Finally, the resulting product was collected by centrifuged and washed with acetone for several times and dried under vacuum at 60 • C for 8 h, finally the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite sample is obtianed. The preparation process of the bare FeF 3 · 0.33H 2 O sample is the same as above procedure but no adding 3D-OMCs.
Material characterization.-The structure, crystallinity and phase composition of the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite were characterized by X-ray diffractometer (XRD, Model LabX-6000, Shimadzu, Japan) in the scattering angle (2θ) range of 10 • -80 • at a scan rate of 4 • min −1 . To analyze the size, morphology, and elemental distribution of the samples, field-emission scanning electron microscopy (FESEM, Hitachi S-4800) and energy dispersive X-ray spectroscopy (EDS) detector were implemented. The surface area (S BET ) of the materials was estimated via Brunauer-Emmett-Teller (BET) (TriStar II 3020, Micromeritics USA) with nitrogen as adsorption/desorption gas. Raman spectra of nanocomposites were finished by Raman spectrometer measurement with a 532 nm excitation line. The carbon content of the nanocomposite was determined by a Carbon-Sulfur elements Determinator.  X-ray diffraction of (a) FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs; The sketch map of (b) FeF 3 · 0.33H 2 O@3D-OMCs; The Raman spectra of (c) FeF 3 · 0.33H 2 O@3D-OMCs. measured using the same instrument. The disturbance amplitude was ±5 mV and the frequency range was from 10 mHz to 100 kHz. All the electrochemical measurements were performed at room temperature.

Results and Discussion
To confirm the phase composition and crystal structure of FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite, the XRD patterns are shown in Figure 1a. Apparently, all diffraction peaks correspond to the standard hexagonal-tungsten-bronze-type (HTB) FeF 3 · 0.33H 2 O (JCPDS: No. 76-1262, a = 0.7423, b = 1.273, c = 0.7526 nm) without any impurity signal. 33,34 In the crystal structure of FeF 3 · 0.33H 2 O (Figure 1b), six Fe octahedral are connected via corner-sharing to form a special huge hexagonal cavity, resulting in much larger cell volumes (∼0.710 nm 3 ), which are beneficial to a rapid pathway for Na + transport. 35 Besides, H 2 O molecules reside in the center of the cavity that can maintain stability of the enor-mous hexagonal cavity and avert structure disintegrate during Na + insertion/extraction process. 36 According to the results of the carbonsulfur elemental analysis, it can be determined that the carbon content is about 8.9 wt% in the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite. Raman spectroscopy is used to confirm the graphitization of the 3D-OMCs. As depicted in Figure 1c, the Raman spectra is consisted of intense broad bands positioned at 1338 and 1597 cm −1 , which are corresponding to the disordered D band and the graphitized G band, respectively. The G band at 1597 cm −1 is related to Raman-active E 2g mode, reflecting its symmetry and crystallinity. 37,38 The D-band at 1338 cm −1 is mainly ascribed to defects in the graphite structure. The relative intensity of D-bands to G-bands (I D /I G ) can reflect the degree of graphitization. It can be determined from the data in Figure 1c that the intensity ratio of I D /I G is 1.18, indicating that the 3D-OMCs in the composites has a high graphitization degree.
The microstructures of the 3D-OMCs and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite were examined by TEM. From the TEM image  of 3D-OMCs in Figure 2a can be clearly seen that the 3D-OMCs show an orderly honeycomb morphology, which is an reverse copy of KIT-6 template. This structural feature makes 3D-OMCs an ideal template for the synthesis of mesoporous carbon-hybrid composites. Comparing with the TEM image of 3D-OMCs, no obvious three-dimensional open-framework mesoporous structure is seen in the TEM image of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite (Figure 2b), it is probably because the FeF 3 · 0.33H 2 O nanoparticles have been filled into the mesoporous channels. In the same time, no obvious nanoparticles can also be observed on the external surface of the nanocomposite, indicating that FeF 3 · 0.33H 2 O nanoparticles perfectly infill into the mesoporous channel. The HRTEM image of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite is shown in Figure 2c, it can be seen the lattice fringe spacing of 0.32 nm which assigns to the (220) phase of FeF 3 · 0.33H 2 O. 39 Meanwhile, it can be clearly observed that the FeF 3 · 0.33H 2 O nanoparticles (labeled by a red arrow) with a size ranging between several nanometers and a dozen nanometers are encapsulated into the pore channels of the 3D-OMCs. The SAED image in the inset of Figure 2b exhibits two ring patterns related to (220) and (112), which proves further the polycrystalline nature of FeF 3 · 0.33H 2 O. 40,41 To analyze the distribution of Fe, F, O and C elements in the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite, the EDX was examined as presented in Figures 2d-2g that all the elements are evenly distribution in FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite. In contrast, it can be seen from Figures 2h-2i that the particle size of bare FeF 3 · 0.33H 2 O increases to ∼500 nm and agglomeration phenomenon is very serious. Therefore, the 3D-OMCs matrix cannot only limit the growth of FeF 3 · 0.33H 2 O, but also prevent the agglomeration during the crystallization process.
In order to further verify that the FeF 3 · 0.33H 2 O nanoparticles were embedded into the pore channels of 3D-OMCs, the nitrogen adsorption/desorption isotherm was measured. Figures 3a-3b exhibit the N 2 adsorption/desorption isotherms and pore-size distribution curves of 3D-OMCs and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite. The nitrogen adsorption-desorption isotherms of both samples exhibit the typical type-IV curves with a clear H1 hysteresis loop according to the International Union of Pure and Applied Chemistry (IUPAC) nomenclature, indicating that 3D-OMCs possess the ordered mesoporous structure. 42 The Brunauer-Emmett-Teller (BET) specific surface ar-eas and pore characteristics of 3D-OMCs and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite are catalogued in Table I. As illustrated in Table I, the specific surface area is observed to decrease remarkably from 1007.36 m 2 g −1 for 3D-OMCs to 67.35 m 2 g −1 for FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite. From the inset image of Figures 3a-3b, it is clearly found that both samples reveal narrow pore size distribution, in which the maximum peaks centered at about 4.86 and 3.71 nm for 3D-OMCs and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite, respectively. Pore-filling would significantly reduce the surface area and pore size. Consequently, the FeF 3 · 0.33H 2 O nanoparticles were filled into the pores of the 3D-OMCs. 43,44 Cyclic voltammetry (CV) measurement was applied to elaborate the energy storage mechanism and evaluate electrochemical performances of FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite. Figures 4a-4b show the first four CV curves of both electrodes between 1.0 and 4.0 V with a scan rate of 0.1 mV s −1 at room temperature. As shown in Figures 4a-4b, the both electrodes exhibit two oxidation peaks in the anodic process. The first anodic peak appears at ∼2.71 V, which is ascribed to the intercalation of Na + into the iron fluoride crystal (FeF 3 +Na→NaFeF 3 ); and the second anodic peak located in ∼2.05 V is mainly caused by the conversion reaction of Fe 2+ to Fe 0 (NaFeF 3 +2Na→Fe+3NaF). 45,46 It is noticed that the subsequent cycles show the consistent peak position of the redox reaction, suggesting the good reversibility of the electrode reaction. Moreover, the electrochemical reversibility of the electrode reaction is usually evaluated using the potential difference ( Ep) between the anode peak and the cathode peak. 47 It can be seen from Figures 4a-4b that the FeF 3 · 0.33H 2 O ( Ep = 0.66) is slightly higher than that of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite ( Ep = 0.62), implying that FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite has much better redox reversibility. Figures 4c-4d exhibit the charge-discharge profiles of the FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite for different cycles at 20 mA g −1 in the voltage range of 1.0-4.0 V. As being seen in Figure 4d, FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite reveals a higher first discharge capacity of 386 mAh g −1 and keeps a steady reversible capacity of 238.0 mAh g −1 after 100 cycles. On the contrary, the initial discharge capacity of the FeF 3 · 0.33H 2 O in Figure 4c fades from 238 mAh g −1 to 127 mAh g −1 after 100 cycle.  To further evaluate electrochemical properties of the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite as the cathode active material of SIBs, the cyclic life curves and charge/discharge curves of FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite at different current rate were compared in Figure 5. The discharge capacities at 20-100 mA g −1 rates were shown in Figures 5a-5b. As shown in Figures 5a-5b, the discharge capacities of the FeF 3 · 0.33H 2 O at 20, 40, 60, 80 and 100 mA g −1 are 244, 197, 164 and 148 mAh g −1 . In contrast, at the corresponding current density, the discharge capacities of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite are 392, 284, 262, 241 and 210 mAh g −1 , respectively.
The cycle stability of the electrode active materials is a significant index for its practical application. Figure 5c gives the cycle performances of both electrode active materials at a high rate of 60 mA g −1 between 1.0 V and 4.0 V. Noticeable, FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite delivers a good cycle performance and an exceptional discharge specific capacity of 183 mAh g −1 up to 100 cycles as well as an initial coulomb efficiency (CE) of 94% (which is significantly higher than that of FeF 3 · 0.33H 2 O (83%)). However, the discharge capacity of FeF 3 · 0.33H 2 O shows a fast capacity fading and only 125 mAh g −1 after 100 cycles. However, it is found from Figure 5c that the cycle stability of the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite shows some decays since the radius of sodium ion (0.102 nm) is greater than one of the lithium ion radius (0.076 nm), which can lead to structural collapse during Na + insertion/extraction processes. Besides, after FeF 3 · 0.33H 2 O was packaged into 3D-OMCs nanocomposite, the cycling stability of the material still appears some improvement to some extent. Apparently, FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite exhibits an outstanding electrochemical property that can be because 3D-OMCs in the nanocomposite not only act as a excellent conductive network, but also enlarge interlayer distance of FeF 3 · 0.33H 2 O nanoparticles that can render a larger space for Na + intercalation. Meanwhile, the 3D mesopores structures can facilitate the penetration of electrolyte ion and suppress the volume changes during Na + insertion/extraction processes, thus improving the cycle performance and keeping the structure stability of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite.
from 100 mA g −1 to 20 mA g −1 after 25 cycles, the discharge capacity of the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite can rapidly return to 346 mAh g −1 . Intelligible, the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite has better rate capability and structure stability than the FeF 3 · 0.33H 2 O. It can be mainly attributed to two points: (1) The 3D mesoporous structures of the nanocomposite can maintain unobstructed ordered paths for Na + transfer and provide abundant reaction sites during charge/discharge process. (2) The 3D-OMCs matrix can suppress the structural degeneration and consolidate the stability of the host structure.
To further investigate the electrochemical properties of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite for the application of NIBs, electrochemical impedance spectroscopy (EIS) measurements are carried out at around 3.2 V (room temperature), in which FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite were used as the working electrode versus Na after cycling for different cycles. The Nyquist plots of FeF 3 · 0.33H 2 O and FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite are illustrated in Figures 6a-6b. Obviously, all the Nyquist plots consists of one semicircle in the high frequency region and one oblique line in the low frequency region. The highfrequency region semicircle represents the charge transfer kinetics of the electrode and electrolyte interface, and the oblique line at low frequency is related to the diffusion kinetics of sodium ions in the solid material. Figure 6c shows the equivalent electrical circuit of the Nyquist curves, where R s is the ohmic resistance; R ct is the charge transfer resistance, which reflects the electrochemical kinetics of the cell reactions; W is the Warburg impedance caused by the diffusion of Na + in the electrode material; and CPE is the constant phaseangle element associated with the surface layer and the electric double layer. 48,49 Nyquist plots are fitted as the red smooth curves (Figures  6a-6b), and the fitting impedance data are good consistent with the experimental EIS data. As being seen from Table II, FeF 3 · 0.33H 2 O has a high charge transfer impedance, and the R ct values at the different cycle numbers are 97, 192, 314 at 1 st , 50 th , 100 th , reespectively. However, the R ct values for FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite after 1 st , 50 th , 100 th cycles are 58, 112, 168 , respectively. Usually, R s is composed of electrolyte resistance, current collector resistance and active material resistance. 50 The change of R s is mainly determined by the material resistance because the resistances for current collector and resistance have relatively fixed value. The reaction mechanism of FeF 3 · 0.33H 2 O eletrode in sodium ion battery can be expressed as following: Na + + FeF 3 → NaFeF 3 (4.0-1.2 V) [1] 2Na + + NaFeF 3 → 3NaF + Fe (1.2-1.0 V) [2] Reaction 1 is considered to be a conventional intercalation reaction, while Reaction 2 is a chemical conversion reaction. 51 The increased Rs of FeF 3 · 0.33H 2 O electrode during the charge-discharge Figure 6. Three-dimensional Nyquist plots of (a) FeF 3 · 0.33H 2 O and (b) FeF 3 · 0.33H 2 O@3D-OMCs electrodes after different cycles at 100 mA g −1 with a scan rate of 0.01 mV s −1 ; the equivalent circuit used for fitting the experimental EIS data (c); the linear relationship plots (d) between Z´and ω −1/2 at low frequency region and equivalent circuits. processes may be ascribed to the low electrochemical activity of the product NaF, which is electronic insulative and poor compatibility with the electrolyte. 52 However, the electrical conductive 3D-OMC framework can effectively improve the conductivity of FeF 3 · 0.33H 2 O during charge/discharge process, thus the R s of the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite is less than FeF 3 · 0.33H 2 O for diffenent cycles in Table II. In addition, the sloped line in the low frequency region is related to the sodium ion diffusion (D Na+ ) in the active material, which can be calculated by the following formula: 53 where R is the ideal gas constant, T is the absolute temperature (K), A is the electrode surface area, n is the number of electron(s) per molecule oxidized, F is the Faraday constant, C is the concentration of Na + in active material and σ w is the Warburg factor to be determined from the slope of the following formula: where σ w is the Warburg factor which is relative with Z , ω is the angular frequency of the low frequency region. Figure 6d shows the linear relationship between Z and ω −1/2 of both electrode materials in the low-frequency region, and the D Na + of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite was estimated to be ∼3.03 × 10 −14 , which is slightly higher than that of FeF 3 · 0.33H 2 O (2.71 × 10 −15 ).  46 and Fe (1-x) TixF 3 /C nanocomposites (1.37 × 10 −15 ), 8 the diffusion coefficients of the FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite is higher than that of others. Therefore, the improvement in electrochemical performance of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite could be related to the 3D-OMCs, which can increase the diffusion coefficient of sodium ions, enhance the electronic conductivity of low conductive material and result in high rate capability. Moreover, the 3D-OMCs cannot only maintain the stability of the material structure, but also take shape valid Na + diffusion channel. In addition, the mesoporous structure can also promote the full contact of the electrolyte with the electrode active material and effectively restricts the growth and agglomeration of FeF 3 · 0.33H 2 O nanocrystals during the chargedischarge process.

Conclusions
The FeF 3 · 0.33H 2 O nanoparticles packaged into 3D order mesoporous carbons as cathode material of sodium-ion batteries (SIBs) has been successfully designed and fabricated by a facile nanocasting technique and mesoporous silica KIT-6 template. The intimate contact between the 3D-OMCs and FeF 3 · 0.33H 2 O nanoparticles not only afford a highly conductive matrix for Na + insertion/extraction, but also restrain the agglomeration and growth of FeF 3 · 0.33H 2 O nanoparticles during the crystallization process. The FeF 3 · 0.33H 2 O nanoparticles bounded in mesoporous carbon channel exhibit the low resistance of electron and Na + transport diffusion. The 3D mesopores structures allow sufficient infiltration of electrolyte and repaid Na + diffusion, which can apparently improves the rate capability. The FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite exhibits supernormal electrochemical performance in the discharge capacity and cycling stability of SIBs. It can deliver a high first discharge capacity of 386 mAh g −1 and a high capacity retention after 100 cycles at 20 mA g −1 between 1.0 V and 4.0 V. Meanwhile, this cathode active material reveals high Na + diffusion velocity and excellent rate capability during long-term cycling. It can provide a high specific capacity of 201 mAh g −1 even at 100 mA g −1 . Therefore, the synthesis strategy of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite provides a new approach to preparation of low conductive metal fluoride cathode active material for the application of high-performance SIBs. Of course, although the discharge capacity and electrochemical performances of FeF 3 · 0.33H 2 O@3D-OMCs nanocomposite in this work have been enhanced to some extent, the cycling stability needs still to be further improved before it is commercially used as cathode materials of sodium ion battery.