Hierarchical-Structured Fe2O3 Anode with Exposed (001) Facet for Enhanced Lithium Storage Performance

The hierarchical structure is an ideal nanostructure for conversion-type anodes with drastic volume expansion. Here, we demonstrate a tin-doping strategy for constructing Fe2O3 brushes, in which nanowires with exposed (001) facets are stacked into the hierarchical structure. Thanks to the tin-doping, the conductivity of the Sn-doped Fe2O3 has been improved greatly. Moreover, the volume changes of the Sn-doped Fe2O3 anodes can be limited to ~4% vertical expansion and ~13% horizontal expansion, thus resulting in high-rate performance and long-life stability due to the exposed (001) facet and the unique hierarchical structure. As a result, it delivers a high reversible lithium storage capacity of 580 mAh/g at a current density of 0.2C (0.2 A/g), and excellent rate performance of above 400 mAh/g even at a high current density of 2C (2 A/g) over 500 cycles, which is much higher than most of the reported transition metal oxide anodes. This doping strategy and the unique hierarchical structures bring inspiration for nanostructure design of functional materials in energy storage.


Introduction
With the development of portable electronics and electric vehicles, advanced lithiumion batteries (LIBs) have become urgently desirable [1][2][3][4][5]. The key issue for high performance batteries (high energy/power density) is developing advanced anode materials with high specific capacity and high electronic/ionic conductivity. Anode materials play an important role in LIBs. The property of anode, such as the specific capacity, stability, conductivity, and diffusion of Li + ions, is associated with the excellent performance of the batteries. Graphite was the earliest anode material used in the commercial lithium-ion batteries. In the development process of lithium-ion batteries, graphite plays a crucial role, and it continues to play an irreplaceable role in the anode material of present energy storage batteries [6]. The traditional graphite anode has a lower specific capacity (only 372 mAh/g) and poor diffusion of Li + ions, which cannot be satisfied with the development of the advanced batteries. The lithium metal anode has been widely studied due to its high specific capacity (3860 mAh/g), low redox potential (−3.04 V vs. SHE), and low weight density (0.534 g/cm). The uneven deposition of lithium metal can induce the severe lithium dendritic growth and the drastic volume changes, leading to the low Coulombic efficiency and potential safety hazards [7]. In recent years, strategies such as improving solid-state electrolyte interface (SEI) films [8], constructing artificial layers [9], and modifying diaphragms [10] have been proposed to stabilize the issues with the lithium metal anode. However, the stability and safety have not been well solved, restricting the commercial application of lithium metal. Si-based anodes (mainly Si and SiO x ) also have very high theoretical specific capacity (4200 mAh/g), while the significant volume exchanges during the charging/discharging cycles lead to the continuous destruction of the electrode materials and the repeated break of SEI films, leading to a serious capacity fading [11]. Transition metal oxides (TMOs) have attracted extensive attention recently due to their high theoretical specific capacities, good safety performance, and their cheap and easy preparation [12]. In 2000, researchers first proposed that TMOs should be used as negative electrodes for lithium batteries [13]. Since then, research on TMO materials has been increasing and TMOs have become one of the important candidate materials in the field of anode materials. Actually, TMO anodes have the same problems as Si-based anodes. For example, the cyclic stability and the rate capacity of TMO anodes are not very good because of the volume effects and the low charge diffusion. Among numerous TMOs, Fe 2 O 3 has been regarded as one of the most promising candidates for advanced LIBs because of its high theoretical capacity (1005 mAh/g), which is~2.7 times greater than that of commercial graphite (372 mAh/g) [14][15][16][17][18]. However, the increased capacity based on its conversion-type lithium storage is generally accompanied by many challenges, such as low electron conductivity, low ion diffusion, and huge volume change [19][20][21]. The conductivity and the stability of Fe 2 O 3 electrodes are equally important to the high performance of batteries, which means that the excellent electron/ion transport can contribute to the high-rate performance, and the super stable electrode can contribute to the long-cycling performance [22][23][24][25][26][27][28][29]. In contrast, the poor conductivity of electrodes will prevent electrons and ions diffusing rapidly in the solid bulk, which causes the serious polarization and the uneven lithification of the Fe 2 O 3 electrode. Finally, the power density and the safety of batteries will be problems due to the poor conductivity of the material [30,31]. Additionally, large volume changes of Fe 2 O 3 anodes during the lithification/delithification processes will lead to the collapse of electrode structure, thus leading to the destruction of the solid electrolyte interface (SEI) films [32][33][34]. Therefore, constructing well-formed Fe 2 O 3 anodes with fast ionelectron diffusion and with accommodation for volumetric expansion are guarantees for high performances of batteries.
Many efforts have been made to improve the conductivity and the stability of Fe 2 O 3 anodes, and many efficient strategies, such as reducing size, conductive agent coating, conductive component composition, heterogeneous element doping, and novel nanostructure designs, have been proposed to tackle the problems associated with the low rate capability and the poor cyclic stability [30,[35][36][37][38]. Recent studies have proved that the crystal planes and the hierarchical structures of electrode materials have significant effects on the electrochemical properties. For example, it has been confirmed that the energy barrier for Li-ion transfer across the (001) facet of Fe 2 O 3 is much lower than that across the (110) facet [39]. Again, the gradient-structured Fe 3 O 4 /C nanospheres can effectively relieve the stress concentration caused by drastic volume changes and can derive an excellent stability during fast charging/discharging processes [40]. Therefore, studies on the controllable synthesis of the given crystal planes and on the design of the unique structures for Fe 2 O 3 materials are of great interest. Many reports indicate that element doping can tune the facets of Fe 2 O 3 , especially for the exposed (001) facet tuning of Fe 2 O 3 [7,17,36]. Based on the understanding of the literature, the Sn-doped Fe 2 O 3 samples can more easily expose the (001) facet than pure Fe 2 O 3 due to the Sn-doping. Moreover, when Sn atoms are introduced into Fe 2 O 3 , an impurity state can be induced in the bandgap of pure Fe 2 O 3 , resulting in the improved conductivity of Sn-doped Fe 2 O 3 . In addition, the Sn doping can induce the formation of belts with an exposed (001) facet, which can be stacked into ordered hierarchical structures [38,39]. Thus, we use a tin-doping strategy here to easily fabricate Fe 2 O 3 material with the dual structural features of an exposed (001) facet and hierarchical structures. The exposed (001) facet provides channels for rapid Li + ion diffusion, and the hierarchical structures accommodate the large volume expansion.

Materials Synthesis
Sn-doped Fe 2 O 3 samples were prepared using an improved hydrothermal reaction [41]. The typical procedure is as follows: 150 mL 0.5 mol/L FeCl 3 aqueous solution and 50 mL 0.012 mol/L K 2 SnO 3 aqueous solution were prepared, respectively. Here, for the convenience of expression, one solution containing Fe 3+ is indicated by Fe-sol and another solution containing SnO 3 2− is indicated by Sn-sol. Firstly, the Sn-sol was added dropwise to the Fe-sol while stirring vigorously for 30 min to obtain a clear mixed solution. Then, the mixed system was incubated at 180 • C for 5 h in a Teflon-lined autoclave and cooled down to room temperature after the hydrothermal reaction. Finally, the products were collected through a succession of processes including centrifuging, washing, and drying. Pure Fe 2 O 3 samples were also prepared using the same process of the Sn-doped Fe 2 O 3 samples except for the absence of SnO 3 2− ions for comparison.

Structure Characterization
The morphologies of the samples were obtained using a field emission scanning electron microscope (SEM, JSM-5610LV, Tokyo, Japan), and the electron diffraction spectrum was performed on a field emission transmission electron microscope (TEM, FEI Talos F200X, Hillsborough, NC, USA). The energy dispersive spectrum and the elemental mappings of the samples were tested using X-ray photoelectron spectroscopy (XPS, Thermo ESCALAB 250Xi, Waltham, MA, USA). The crystallographic information for the samples was investigated with X-ray powder diffraction (XRD, Bruker D8 Advance, Karlsruhe, Germany).

Electrochemical Tests
The electrochemical measurements were carried out by using CR 2025 coin-type cells. The working electrode was prepared using an electrode slurry on a copper current collector (11 µm thick). The slurry was composed of active material (Sn-doped Fe 2 O 3 ), carbon (Super-P-Li), and polymer binder (sodium carboxymethyl cellulose, CMC, 3%) with a weight ratio of 8:1.5:0.5. The total mass loaded on each electrode was about 1.5 mg. Lithium wafer was used as both the counter electrode and the reference electrode. The cell was assembled in a high-purity argon filled glove box. A microporous polypropylene membrane (Cellgard 2300, Charlotte, NC, USA) was used as the separator, and the electrolyte was LiPF 6 (1.0 mol/L) in a 1:1 (w/w) mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC). Charge/discharge tests were performed using a Land testing system (CT2001A, China) at different current densities with voltage windows of 0.05-3 V (vs. Li + /Li) and the electrochemical data were obtained using an electrochemical workstation (CHI 650E, China).

Calculation
The DFT calculations were carried out by Vienna Ab initio Simulation Package (VASP) [42]. Core electrons were described using pseudopotentials generated from the projector augmented wave method [43], and valence electrons were expanded in a plane-wave basis set with an energy cutoff of 440 eV. The Perdew-Burke-Ernzerh (PBE) exchange correlation functional with the on-site Coulomb Repulsion U term was used for Fe 2 O 3 . The value of U was selected to be 4.0 eV for Fe atoms. The Fe 2 O 3 (001) surface was simulated using a slab model with a (1 × 1) Fe 2 O 3 unit. A vacuum layer of 16 Å along with Z-axis direction was applied to avoid interaction between images caused by periodic boundary conditions. The Sn-doped Fe 2 O 3 was constructed by replacing one Fe atom with Sn atom in the Fe 2 O 3 cell. The convergence criteria of force and energy were set as 0.01 eV/Å and 10 −4 eV, respectively. The surface free energies of the Fe 2 O 3 (001) surface and Sn-doped Fe 2 O 3 (001) surface were calculated using the following equation: where E bulk is the per formula unit energy of bulk Fe 2 O 3 (Sn-doped Fe 2 O 3 ), E slab is the energy of the surface, and S is the area of the surface. one Fe atom with a Sn atom in the Fe 2 O 3 cell, and the Sn-doping concentration is set as 6% (wt%) to be consistent with the experimental values. Compared to pure Fe 2 O 3 (Figure 1a), the surface energy of the Sn-doped Fe 2 O 3 (Figure 1b,c) is decreased by about 64 KJ/m 2 (surface doping) and 66 KJ/m 2 (internal doping), respectively. Clearly, the Sn-doping can reduce the surface energy of the (001) facet greatly. The calculation results indicate that a small amount of Sn-doping in Fe 2 O 3 can easily make the (001) facet exposed. Based on this understanding, we prepared Sn-doped Fe 2 O 3 powders experimentally with a low Sn-doping concentration used as anode materials with the (001) facet exposed for LIBs. Pure Fe 2 O 3 powders were also prepared using the same process as the Sn-doped Fe 2 O 3 powders for comparison to investigate the further effects of the Sn-doping on the microstructures. As shown in Figure 1d,e, two samples have similar morphologies to a peanut-like contour but a significant difference in the secondary structure. Clearly, pure Fe 2 O 3 samples are dense aggregates formed by the accumulation of nanoparticles, while Sn-doped Fe 2 O 3 samples are loose stacks formed by one-dimensional (1D) nanowires. For electrode materials with volume expansion, the hierarchical structures with 1D wires and 3D stack brushes should be ideal structures for accommodating the expanded volume and releasing stress. Therefore, the electrode structure will be relatively stable during the charging/discharging processes compared to the dense structures. Moreover, the loose structure is conducive to the contact between the electrolyte and the active material, and the diffusion rate of Li + ions will be faster in loose structures than that in dense structures, hence the better rate performance of electrode materials.

Results and Discussion
where Ebulk is the per formula unit energy of bulk Fe2O3 (Sn-doped Fe2O3), Eslab is the energy of the surface, and S is the area of the surface. Figure 1 shows the optimized structures of (001) facets and the corresponding surface energy of Fe2O3 and Sn-doped Fe2O3. The Sn-doped Fe2O3 was constructed by replacing one Fe atom with a Sn atom in the Fe2O3 cell, and the Sn-doping concentration is set as 6% (wt%) to be consistent with the experimental values. Compared to pure Fe2O3 (Figure 1a), the surface energy of the Sn-doped Fe2O3 (Figure 1b,c) is decreased by about 64 KJ/m 2 (surface doping) and 66 KJ/m 2 (internal doping), respectively. Clearly, the Sn-doping can reduce the surface energy of the (001) facet greatly. The calculation results indicate that a small amount of Sn-doping in Fe2O3 can easily make the (001) facet exposed. Based on this understanding, we prepared Sn-doped Fe2O3 powders experimentally with a low Sn-doping concentration used as anode materials with the (001) facet exposed for LIBs. Pure Fe2O3 powders were also prepared using the same process as the Sn-doped Fe2O3 powders for comparison to investigate the further effects of the Sn-doping on the microstructures. As shown in Figure 1d,e, two samples have similar morphologies to a peanut-like contour but a significant difference in the secondary structure. Clearly, pure Fe2O3 samples are dense aggregates formed by the accumulation of nanoparticles, while Sn-doped Fe2O3 samples are loose stacks formed by one-dimensional (1D) nanowires. For electrode materials with volume expansion, the hierarchical structures with 1D wires and 3D stack brushes should be ideal structures for accommodating the expanded volume and releasing stress. Therefore, the electrode structure will be relatively stable during the charging/discharging processes compared to the dense structures. Moreover, the loose structure is conducive to the contact between the electrolyte and the active material, and the diffusion rate of Li + ions will be faster in loose structures than that in dense structures, hence the better rate performance of electrode materials.     (Figure 2b), many filamentous materials appear around the particles, and they are gradually aggregated and eventually stacked together to form peanut-like hierarchical structures (Figure 2c). The following processes are similar to Ostwald ripening (Figure 2d-f), in which the filamentous materials continue to stack on the original hierarchical structures forming larger particles. Therefore, the process of Sn-doped Fe 2 O 3 brushes forming can be illustrated as in Figure 2i. Firstly, Sn-doped Fe 2 O 3 nanoparticles are formed at the initial hydrothermal reaction. Then, many filaments appear around the nanoparticles and gather with each other. Finally, the excess nanobelts are aggregated and stacked orderly to form larger brushes. To represent the formation of stacked peanut-like brushes, the TEM images (Figure 2g) show that the stacked lines present ribbon-like structures. The HRTEM image and the FFT patterns (insets in Figure 2h) indicate that three sets of lattice fringes (0.25 nm) fit well to the α-Fe 2 O 3 (110), (−120), and (−210) facets, respectively [9,26], indicating the basal plane of the nanowire is the (001) facet.

Results and Discussion
Then, many filaments appear around the nanoparticles and gather with each other. Finally, the excess nanobelts are aggregated and stacked orderly to form larger brushes. To represent the formation of stacked peanut-like brushes, the TEM images (Figure 2g) show that the stacked lines present ribbon-like structures. The HRTEM image and the FFT patterns (insets in Figure 2h) indicate that three sets of lattice fringes (0.25 nm) fit well to the α-Fe2O3 (110), (−120), and (−210) facets, respectively [9,26], indicating the basal plane of the nanowire is the (001) facet.   The mass percentage of the Sn element in the prepared sample is about 7.4%, which is close to the concentration of 6.0% for the optimized structures (Figure 1b,c). The average content of the Sn element in the Sn-doped Fe 2 O 3 is 7.1%, which was detected using inductively coupled plasma mass spectrometry (ICP). The STEM-EDS elemental mappings in Figure 3c indicate that Fe, O, and Sn elements are uniformly distributed in the wires, which has also been confirmed by the XPS results in Figure 3d. percentage of the Sn element in the prepared sample is about 7.4%, which is close to the concentration of 6.0% for the optimized structures (Figure 1b,c). The average content of the Sn element in the Sn-doped Fe2O3 is 7.1%, which was detected using inductively coupled plasma mass spectrometry (ICP). The STEM-EDS elemental mappings in Figure 3c indicate that Fe, O, and Sn elements are uniformly distributed in the wires, which has also been confirmed by the XPS results in Figure 3d. Good conductivity is crucial to an electrode material for batteries. Many methods, such as element doping and carbon coating, can improve the conductivity of materials. Element doping to achieve bulk conductivity has become a popular method for preparation of the material. Based on theoretical calculations, we found that the Sn-doping not only tunes the surface morphology but also enhances the conductivity of Fe2O3. The density of states of pure Fe2O3 and Sn-doped Fe2O3 are shown in Figure 4. Clearly, pure Fe2O3 Good conductivity is crucial to an electrode material for batteries. Many methods, such as element doping and carbon coating, can improve the conductivity of materials. Element doping to achieve bulk conductivity has become a popular method for preparation of the material. Based on theoretical calculations, we found that the Sn-doping not only tunes the surface morphology but also enhances the conductivity of  Table 1 that the resistivity values have decreased from the order of magnitude of 10 6 to 10 4 and the conductivity values have increased from the order of magnitude of 10 −7 to 10 −5 , indicating a change by 2 orders of magnitude from pure Fe 2 O 3 to Sn-doped Fe 2 O 3 . For anode materials of lithium-ion batteries, good conductivity is not only benefitial for the cycling stability, but also for the rate performances during the charging/discharging cycles. It is well known that the electrochemical performances are not only based on the intrinsic crystallinity, but also related to the morphologies and assembled structures of anode materials. To understand the enhancement of this hierarchical structure to the performance of lithium storage, the as-prepared Sn-doped Fe 2 O 3 samples' electrochemical performances as anode materials for LIBs are evaluated by using a two electrode coin-type cell. The electrochemical performances of the as-prepared Sn-doped Fe 2 O 3 were investigated using cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements. The CV measurement for Sn-doped Fe 2 O 3 is shown in Figure 5a. In the first cathodic sweep, there is no typical reduction peak at~0.8V for Fe 3+ to Fe 0 compared to the CV curves of the pure Fe 2 O 3 ( Figure S2), while a broad irreversible peak can be observed at the low potential (0-0.5V), indicating that the side reactions of the electrode material are significant. Additionally, this process should be accompanied by the reduction and irreversible decomposition of the electrolyte to form solid electrolyte interphase (SEI) films [44]. Furthermore, the reduction reactions from Fe 2 O 3 to Fe 0 as shown at below in the first cathodic sweep should occur.      Figure 5b shows the Nyquist plots of EIS obtained from the Sn-doped Fe2O3 samples before cycling and after the 5th, the 10th, and the 15th cycle, respectively. The insert in Figure 5b is the equivalent circuit model and the partial enlargement of images of Nyquist plots. The corresponding fitted impedance parameters are listed in Table 2. In the circuit model, R1 is the ohmic contact resistance of the inside battery, R2 is the charge transfer impedance, R3 is the resistance of the formed SEI films after cycles, and W is the Warburg impedance. It can be seen from Table 2 that the values of R1 are very small in different charging and discharging states of batteries, indicating that the various components inside the battery are in good contact. The electrode materials are in an inactive state before the In the first anodic sweep, only a peak at 1.78 V can be observed, which is related to the oxidative reaction of Fe 0 to Fe 3+ as shown below.
In the subsequent charging/discharging processes, the main cathodic peaks and anode peaks are slightly shifted, and the peak intensity decreases significantly. The results indicate the capacity loss and some irreversible processes occurred during the lithium ions insertion and extraction in the first cycle. It is noteworthy that no typical peaks of Sn lithification/delithification can be observed on the CV curves, meaning that there may be two possibilities: one is that the doping amount of Sn in Fe 2 O 3 is too small to cause changes in the entire curves, and the other is that Sn atom does not participate in the lithium alloying reactions. Based on the literature, the lithium storage capacity contributed by the doped Sn element is very low due to the trace doping amount, which cannot be measured in the electrochemical testing curves [29]. Importantly, after the second cycle, both the CV peak positions and the integrated areas for Sn-doped Fe 2 O 3 sample remain almost unchanged, while the overlap of the CV curves for the pure Fe 2 O 3 in the first five cycles is not good ( Figure S2), suggesting good capacity retention and structural stability of the Sn-doped Fe 2 O 3 anode. Figure 5b shows the Nyquist plots of EIS obtained from the Sn-doped Fe 2 O 3 samples before cycling and after the 5th, the 10th, and the 15th cycle, respectively. The insert in Figure 5b is the equivalent circuit model and the partial enlargement of images of Nyquist plots. The corresponding fitted impedance parameters are listed in Table 2. In the circuit model, R1 is the ohmic contact resistance of the inside battery, R2 is the charge transfer impedance, R3 is the resistance of the formed SEI films after cycles, and W is the Warburg impedance. It can be seen from Table 2 that the values of R1 are very small in different charging and discharging states of batteries, indicating that the various components inside the battery are in good contact. The electrode materials are in an inactive state before the battery undergoes charging/discharging cycles, so the charge transfer resistance (R2) value of the materials is relatively larger, reaching 1135 Ω. After the battery has carried out several cycles, the electrode materials reach full activation, and correspondingly, the values of R2 are decreased significantly and decreased to 112.6 Ω, 163.1 Ω, and 168.7 Ω after the 5th, the 10th, and the 15th cycle, respectively. It is noteworthy that the fitted value of R3 before cycling is very small (almost negligible), while after charging/discharging cycles, it reaches more than 10 Ω. More importantly, after undergoing 5 cycles, 10 cycles, and 15 cycles, the values of R3 are not changed significantly. Based on the results, it can be inferred that the surfaces of Sndoped Fe 2 O 3 have changed greatly due to the Li + insertion/extraction and the formation of SEI films. Generally, it is inevitable that electrolyte is decomposed on the surface of anode materials with a lower lithium intercalation potential to form SEI films. Another valuable result obtained from the EIS experiment is that the values of R2 and W have scarcely changed after the 5th, the 10th, and the 15th charging/discharging, meaning that the formation of the SEI films does not affect the reactions of Li + ion insertion/extraction in Sn-doped Fe 2 O 3 . Therefore, the good cycling stability and high Li + ion storage of the Sndoped Fe 2 O 3 are mainly attributed to the (001) facet and the unique hierarchical structures of Sn-doped Fe 2 O 3 . These features are vital for the stability, the safe operation, and the rate capability of LIBs. The (001) facet has a shorter diffusion distance of lithium ions, which is beneficial to improving the rate performance, and the hierarchical structures can improve the stability and the specific capacity of Sn-doped Fe 2 O 3 anode.  Figure 6 shows the rate performances and the cycling performances of the Sn-doped Fe 2 O 3 samples. As a comparison, the results of the electrochemical tests for pure Fe 2 O 3 are shown in Figure S1. In the rate plots (Figure 6a), the average discharge capacities are about 779, 529, 431, 360, and 304 mAh/g at 0.2C, 0.5C, 1C, 1.5C, and 2C current densities, respectively. Except for a 32% reduction in capacity from 0.2C to 0.5C, in the following process, the capacity reduction is less than 20%, specifically, 19%, 16.5%, and 15.6% from 0.5C to 1C, 1C to 1.5C, and 1.5C to 2C, respectively. Compared with the pure Fe 2 O 3 samples ( Figure S1a), the improved rate performance for the Sn-doped Fe 2 O 3 samples is attributed to its better conductivity and the rapid Li + diffusion on the (001) facets [39]. The cycling stabilities of the Sn-doped Fe 2 O 3 samples are also outstanding (Figure 6a,b). The average discharge capacity can be maintained at 580 mAh/g at 0.2 C, and it can still be maintained above 400 mAh/g even at a high current density (2C) after 500 cycles, which is much higher than that of the pure Fe 2 O 3 at the same current density ( Figure S1b). The advantage of our hierarchical structure is ensuring sufficient clearance and space, which can gradually relieve the stress concentration caused by the drastic volume changes. The hierarchicalstructured brushes present a sustained buffering effect on the drastic volume changes of the active Fe 2 O 3 , thus retaining the completeness and the stability of the whole structure during the fast cyclic process ( Figure S3). Moreover, the hierarchical structure gradually relies on the void spaces to alleviate the volume changes in the following discharging processes. About 4% of the vertical increase and 13% of the horizontal increase for the hierarchical-structured Sn-doped Fe 2 O 3 brushes can be confirmed in comparison to the full charge/discharge states. The gram capacity is improved qualitatively, although a certain amount of volumetric energy density is sacrificed. Thus, the morphology of SEI layer and the whole electrode can be retained after hundreds of cycles, suggesting an ultra-long life of the hierarchical-structured Sn-doped Fe 2 O 3 anodes. a certain amount of volumetric energy density is sacrificed. Thus, the morphology of SEI layer and the whole electrode can be retained after hundreds of cycles, suggesting an ultra-long life of the hierarchical-structured Sn-doped Fe2O3 anodes.  Figure 6. Rate performances at different charging/discharging current densities (a) and cycling stability of Sn-doped Fe2O3 anode at larger charging/discharging current densities (b).

Conclusions
In conclusion, Sn-doped Fe2O3 materials have been prepared using a hydrothermal method. The prepared Fe2O3 samples have hierarchical structures constructed by stacked nanowires with exposed (001) facets. When used as anodes for lithium-ion batteries, the Sn-doped Fe2O3 materials exhibit excellent rate performance and good cycling stability. Consequently, the hierarchical structures of the Sn-doped Fe2O3 could be one of the potentially promising candidates as anodes for the next generation of high-power lithiumion batteries.
Supplementary Materials: The following supporting information can be downloaded at www.mdpi.com/xxx/s1, Figure S1: Rate performances at different charging/discharging current densities (a) and cycling stability of pure Fe2O3 anode at 0.2C current density (b); Figure S2: CV curves at a scanning rate of 2 mV/s of pure Fe2O3 anode; Figure S3: SEM images of Sn-doped Fe2O3

Conclusions
In conclusion, Sn-doped Fe 2 O 3 materials have been prepared using a hydrothermal method. The prepared Fe 2 O 3 samples have hierarchical structures constructed by stacked nanowires with exposed (001) facets. When used as anodes for lithium-ion batteries, the Sn-doped Fe 2 O 3 materials exhibit excellent rate performance and good cycling stability. Consequently, the hierarchical structures of the Sn-doped Fe 2 O 3 could be one of the potentially promising candidates as anodes for the next generation of high-power lithium-ion batteries.
Supplementary Materials: The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/nano13132025/s1, Figure S1: Rate performances at different charging/discharging current densities (a) and cycling stability of pure Fe 2 O 3 anode at 0.2C current density (b); Figure S2: CV curves at a scanning rate of 2 mV/s of pure Fe 2 O 3 anode; Figure S3: SEM images of Sn-doped Fe 2 O 3 anodes before cycling (a) and after the 5th discharge cycle (b). Inset: the statistics of particle size distribution in both horizontal and vertical directions.  Data Availability Statement: Data sharing is not applicable to this article.