Synthesis of α-Fe₂O₃, Fe₃O₄ and Fe₂N magnetic hollow nanofibers as anode materials for Li-ion batteries

Abstract

α-Fe₂O₃ hollow nanofibers were synthesized via a facile electrospinning process followed by a post-calcination process, and for the first time, Fe₃O₄ and Fe₂N hollow nanofibers were successfully obtained via reduction and nitridation of the prepared α-Fe₂O₃ hollow nanofibers in the presence of NH₃ atmosphere at 350 °C and 400 °C, respectively. The crystal structure, morphology and compositions of the α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers were investigated by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and energy dispersive spectrometry (EDS). Electrochemical measurements show that the α-Fe₂O₃ and Fe₃O₄ hollow nanofibers electrodes deliver a high specific initial discharge capacity of 1314 and 1210 mA h g⁻¹, respectively, and a stable cycling performance (980 mA h g⁻¹ for α-Fe₂O₃ after 200 cycles and 572 mA h g⁻¹ for Fe₃O₄ after 300 cycles) at a current density of 100 mA g⁻¹. The Fe₂N hollow nanofibers electrode demonstrates a high initial discharge capacity, good cycling stability (438 mA h g⁻¹ at the 300^th cycle with a current density of 100 mA g⁻¹), high coulombic efficiency, and excellent rate capability. The superior electrochemical performances are attributed to the unique one-dimensional hollow nanostructure of the materials. The prepared hollow nanofibers are candidate anode materials for Li-ion batteries.

---

1. Introduction

Li-ion batteries (LIBs), a fast-developing technology in electric energy storage, are the dominant power source for a wide range of portable electronic devices.^1–3 Nevertheless, the currently most widely used anode material for commercial LIBs, graphite, has a low theoretical capacity of 372 mA h g⁻¹, which leads to a limited energy output of LIBs.⁴ In order to satisfy the demands of these electronic devices, it is essential to develop lightweight, nontoxic, durable LIBs with greater capacity.^5,6 Recently, it has been found that transition metal oxides such as Fe₂O₃,⁷ Fe₃O₄,⁸ NiO,⁹ Co₃O₄,¹⁰ CoO,¹¹ Mn₃O₄,¹² and MnO¹³ exhibit high reversible capacities, and have been exploited as anode materials for high performance LIBs. Among these promising anode materials, iron oxide (Fe₂O₃ and Fe₃O₄) have been investigated intensively due to their great advantages such as high theoretical capacities (1007 mA h g⁻¹ and 926 mA h g⁻¹), natural abundance, low cost, environmental friendliness, and high resistance to corrosion, which are expected to meet the requirements of future energy storage systems.^14–16 Nowadays, metal nitrides are gradually being exploited as LIBs anode materials because their excellent electrical conductivity, high capacity and their low and flat potentials close to that of lithium metal.¹⁷ Among these metal nitrides, Fe₂N holds great promise as anode material for LIBs since it has significant advantages of high capacity and excellent electrical conductivity.¹⁸ Despite these distinct advantages, the Fe₂O₃, Fe₃O₄ and Fe₂N electrodes still suffer from inferior cycling ability caused by the drastic volume changes during charge–discharge processes.^17,19

To improve the durability of Fe₂O₃, Fe₃O₄ and Fe₂N electrodes, various types of nano-architectures have been employed as anode materials for LIBs, such as nanospheres,²⁰ nanofibers,²¹ nanorods,²² nanotubes,²³ and some other more complex ones.²⁴ The results indicate that the initial discharge capacities increase obviously, and the capacity retention ratios are improved. Among the reported Fe₂O₃, Fe₃O₄ and Fe₂N nanostructures, one-dimensional (1D) nanomaterials are expected to have high performance in energy storage systems because they facilitate the electron transport along the long dimension and the two short dimensions ensure fast Li-ion insertion/extraction.²⁵ Electrospinning is an outstanding technique to process viscous solutions or melts into continuous fibers or ribbons with 1D nanostructure.^26,27 Electrospinning is widely used to fabricate solid nanofibers and hollow nanofibers. In our early work, we have fabricated LiFePO₄/C composite nanofibers and LiFePO₄/C/Ag composite hollow nanofibers by electrospinning technology and used the composites as the cathode materials for LIBs.^28,29 When the hollow-structure nanomaterials are explored as the electrode materials for LIBs, the underlying principle is that the void space within these hollow nanostructures could efficiently enhance their lithium storage properties by shortening the diffusion distance for Li-ion and reversibly accommodating large volume changes.³⁰ Jeong et al.²⁰ reported that hierarchical Fe₂O₃@PANI hollow spheres were prepared by a template-free sonochemical method, which exhibited high reversible capacity, excellent cycle stability and rate capability. Chen et al.³¹ reported that Fe₃O₄ beads exhibited superior performances with large reversible capacity and good cycling property benefiting from the unique hollow structures. Lei et al.³² synthesized coaxial Fe₃O₄@C hollow particles by confined nanospace pyrolysis, presenting enhanced electrochemical properties. All these works prove that hollow structure is desirable for achieving high capacity and cycling stability of electrode materials. However, it is still a great challenge to fabricate novel and well-defined hollow nanostructures of Fe₂O₃, Fe₃O₄ and Fe₂N with high lithium storage capacity and excellent cycling stability.

Herein, we employed electrospinning followed by thermal treatment to fabricate α-Fe₂O₃ hollow nanofibers, and then cleverly used the reduction and nitridation of NH₃ atmosphere for preparation of Fe₃O₄ and Fe₂N hollow nanofibers by the control of the calcination temperature, and then they were used as anode materials for LIBs. The morphology, structure and electrochemical performances of the obtained hollow nanofibers were systematically studied. Some new insights into the design and synthesis of electrodes have been proposed, which will be very helpful for the development of future superior performance electrodes.

2. Experimental section

2.1. Chemicals

The starting chemical reagents used in this work were as follows. Polyvinylpyrrolidone (PVP, M_w ≈ 90 000) and N,N-dimethylformamide (DMF) were purchased from Tianjin Tiantai Fine Chemical Reagents Co., Ltd. Ferric nitrate nonahydrate (Fe(NO₃)₃·9H₂O) was bought from Guangfu Fine Chemical Research Institute. Acetylene black, polyvinylidene fluoride (PVDF) and N-methyl-2-pyrrolidone (NMP) were purchased from Sinopharm Chemical Reagent Co., Ltd. All chemicals were of analytical grade and directly used as received without further purification.

2.2. Preparation of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers

2.2.1. Preparation of α-Fe₂O₃ hollow nanofibers. α-Fe₂O₃ hollow nanofibers were prepared by calcinating electrospun Fe(NO₃)₃/PVP composite nanofibers. First, 1.0 g of Fe(NO₃)₃·9H₂O was dissolved in 6.0 g of DMF, then 1.0 g of PVP was added into the above solution under magnetically stirring for 10 h to form homogeneous transparent spinning solution. In the spinning solution, the mass ratios of PVP, Fe(NO₃)₃·9H₂O, DMF were 1 : 1 : 6. Subsequently, the Fe(NO₃)₃/PVP composite nanofibers were prepared by electrospinning technique under a positive high voltage of 14 kV, distance between the capillary tip and the collector was 17 cm, and relative humidity was 40–60%. The collected electrospun composite nanofibers were then calcined at 450 °C in air for 2 h with a heating rate of 10 °C min⁻¹ to obtain α-Fe₂O₃ hollow nanofibers.

2.2.2. Fabrication of Fe₃O₄ and Fe₂N hollow nanofibers. α-Fe₂O₃ hollow nanofibers were heated to 350 °C in a quartz boat under a flow of gaseous ammonia with a heating rate of 1 °C min⁻¹ and maintained at that temperature for 6 h, and then calcination temperature was decreased to 100 °C with a cooling rate of 1 °C min⁻¹, followed by natural cooling down to room temperature, and thus Fe₃O₄ hollow nanofibers were successfully obtained. For preparation of Fe₂N hollow nanofibers, the preparative processes were the same as those for Fe₃O₄ hollow nanofibers except that the calcination temperature was 400 °C.

2.3. Characterization methods

Thermogravimetric (TG) and differential scanning calorimetric (DSC) curves were collected by using Pyris Diamond TG-DSC (TG, SDTQ600, Perkin Elmer Thermal Analyzer) with the heating rate of 20 °C min⁻¹ in flowing air. The structural characterization of the samples was investigated by an X-ray powder diffractometer (XRD, Bruker, D8FOCUS) in a two-theta range of 10–90°, and the working voltage and current were kept at 40 kV and 30 mA, respectively. The diameter, morphology and crystal structure of the samples were analyzed by using a field emission scanning electron microscope (FESEM, JSM-7610F, JEOL) and a transmission electron microscope (TEM, Tecnai G2 20 S-Twin, FEI). The elementary compositions of the samples were examined using OXFORD X-MaxN80 energy dispersive spectrometer (EDS) attached to FESEM. Then, the magnetic properties of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers were measured by a vibrating sample magnetometer (VSM, MPMS SQUID XL).

2.4. Electrochemical measurements

The anodes were manufactured using the above-obtained nanomaterials as the active materials, acetylene black as conductive additive, and PVDF as binder in the mass ratio of 75 : 15 : 10 dissolved in NMP. Then the mixed slurry was coated onto the Cu foil and dried in a vacuum oven at 120 °C for 24 h. Then the film was cut into discs with the loading mass of about 2 mg cm⁻². The electrochemical performances of the samples were tested by half cells assembled in an argon-filled glove box (H₂O, O₂ content < 1 ppm). Electrochemical measurements were carried out using 2032-type half cells with a pure lithium foil as the counter electrode, Celgard 2320 as the separator, and 1 mol L⁻¹ LiPF₆ in ethylene carbonate/dimethyl carbonate (1 : 1 v/v) solution as the electrolyte. The charge–discharge performances were performed at room temperature by a battery testing system (BTS-5 V/10 mA, Neware Technology Limited Corporation, China) with a cut-off potential of 0.01–3.0 V versus Li⁺/Li. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements were tested using an electrochemical workstation (CHI-760D, Shanghai Chenhua Instrument Limited Corporation, China). For the CV measurements, the voltage was fixed between 0.01 V and 3.0 V, and the scanning rate was fixed at 0.1 mV s⁻¹. For the EIS measurements, the amplitude of the alternating current signal to the cells was 10 mV and the frequency was between 0.01 Hz and 1 MHz.

3. Results and discussion

3.1. Structure and morphology of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers

In order to manifest the decomposition process of Fe(NO₃)₃/PVP composite nanofibers, TG and DSC analysis were carried out. As illustrated in Fig. 1a, it shows the thermal behavior of Fe(NO₃)₃/PVP composite nanofibers. The weight loss is involved in two stages in TG curve. The first weight loss is 40.83% before 255 °C, which is caused by the loss of the surface absorbed water and the rest of solvent DMF, and the decomposition of Fe(NO₃)₃·9H₂O in the composite nanofibers. The second obvious weight loss (55.41%) between 255 °C and 285 °C is mainly resulted from the decomposition and combustion of PVP, with an intensive exothermic peak around 284 °C in DSC curve. Above 400 °C, the TG curves is unvaried, the total weight loss is 96.24%. Therefore, in this study, α-Fe₂O₃ hollow nanofibers were prepared by calcining Fe(NO₃)₃/PVP composite nanofibers at 450 °C for 2 h.

Fig. 1 TG and DSC curves of Fe(NO₃)₃/PVP composite nanofibers (a); XRD patterns and PDF standard cards of the obtained α-Fe₂O₃ (b), Fe₃O₄ (c) and Fe₂N (d) hollow nanofibers.

Crystal structure and phase composition of α-Fe₂O₃ hollow nanofibers were confirmed by XRD measurements. As observed in Fig. 1b, all diffraction peaks are sharp and well-defined, implying that the sample is highly crystallized. It is seen that all the diffraction peaks are indexed as a pure rhombohedral phase of α-Fe₂O₃, which match well with those of the JCPDS standard card (PDF#33-0664), and no characteristic peaks are observed for other impurities, demonstrating the high purity of the prepared α-Fe₂O₃.

Typical XRD patterns of the obtained Fe₃O₄ hollow nanofibers are demonstrated in Fig. 1c. All diffraction peaks confirm that the obtained sample can be identified as cubic Fe₃O₄ with space group of Fm3̄m, in accordance with the JCPDS standard card (PDF#19-0629). Strong and sharp diffraction peaks are situated at 2θ values of 18.31°, 30.12°, 35.48°, 37.12°, 43.12°, 53.50°, 57.03° and 62.63°, corresponding to (111), (220), (311), (222), (400), (422), (511) and (440) crystallographic planes, indicating the formation of pure cubic crystalline Fe₃O₄.

The XRD patterns of sample obtained via nitriding α-Fe₂O₃ hollow nanofibers at 400 °C are shown in Fig. 1d. All diffraction peaks are well indexed to pure Fe₂N (PDF#76-0090) with orthorhombic structure, and the diffraction peaks are situated at 2θ values of 37.48°, 40.83°, 42.90°, 56.63°, 67.66° and 75.70°, corresponding to (110), (002), (−1−11), (−1−12), (300) and (−1−13) crystallographic planes of the orthorhombic Fe₂N, respectively. Notably, no other impurity phase can be detected, implying that the sample has high purity.

SEM was employed to investigate the morphologies. Fig. 2a demonstrates the SEM image of Fe(NO₃)₃/PVP composite nanofibers. It is observed that the fibers have smooth surface and fibrous morphology. SEM image of α-Fe₂O₃ hollow nanofibers is shown in Fig. 2b. One can see that the sample retains the morphology of fibers after calcination at 450 °C. The shrink of diameter of α-Fe₂O₃ hollow nanofibers after calcination is due to the decomposition of PVP and inorganic salts. Inset of Fig. 2b illustrates that α-Fe₂O₃ nanofibers possess hollow and well-defined tubular nanostructure. SEM image of the obtained Fe₃O₄ hollow nanofibers reduced at 350 °C is shown in Fig. 2c. After being reduced at 350 °C, the sample retains the morphology of hollow fibers. Inset of Fig. 2c indicates that Fe₃O₄ nanofibers possess hollow nanostructure. SEM image of the obtained Fe₂N hollow nanofibers nitrided at 400 °C is indicated in Fig. 2d. After being nitrided at 400 °C, the hollow nanofibers are broken, but still remain short fibrous structure.

Fig. 2 SEM images of electrospun Fe(NO₃)₃/PVP composite nanofibers (a), α-Fe₂O₃ hollow nanofibers (b), Fe₃O₄ hollow nanofibers (c) and Fe₂N hollow nanofibers (d).

Under the 95% confidence level, these nanofibers analyzed by Shapiro–Wilk method are normal distribution. Histograms of diameters distribution of the nanofibers are indicated in Fig. 3. The diameters of Fe(NO₃)₃/PVP composite nanofibers, α-Fe₂O₃ hollow nanofibers, Fe₃O₄ hollow nanofibers and Fe₂N hollow nanofibers are 236.81 ± 2.08 nm, 130.57 ± 0.90 nm, 129.18 ± 1.05 and 121.03 ± 1.39 nm, respectively.

Fig. 3 Histograms of diameters distribution of Fe(NO₃)₃/PVP composite nanofibers (a), α-Fe₂O₃ hollow nanofibers (b), Fe₃O₄ hollow nanofibers (c) and Fe₂N hollow nanofibers (d).

The elementary compositions of Fe(NO₃)₃/PVP composite nanofibers, α-Fe₂O₃ hollow nanofibers, Fe₃O₄ hollow nanofibers and Fe₂N hollow nanofibers were further confirmed by EDS, as revealed in Fig. 4. EDS spectra show that C, N, O, Fe are main elements in composite nanofibers, O and Fe elements exist in α-Fe₂O₃ and Fe₃O₄ hollow nanofibers. EDS spectrum reveals the presence of Fe and N elements in Fe₂N hollow nanofibers. The element C, in α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers, comes from the used conductive tape. The Pt peaks in the spectra come from Pt conductive film plated on the surface of the samples for SEM observation. No other elements are found in the samples, implying that the samples have high purity.

Fig. 4 EDS spectra of Fe(NO₃)₃/PVP composite nanofibers (a), α-Fe₂O₃ hollow nanofibers (b), Fe₃O₄ hollow nanofibers (c) and Fe₂N hollow nanofibers (d).

To provide further insights into the morphology and structure of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers, TEM investigations were carried out. Fig. 5a, c and e manifest the typical TEM images of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers, inset of Fig. 5a, c and e demonstrate the corresponding selected-area electron diffraction (SAED) patterns. TEM images in Fig. 5a and c clearly indicate that α-Fe₂O₃ and Fe₃O₄ nanofibers are hollow structure, which agree with SEM observations. Fig. 5e shows the TEM image of Fe₂N hollow nanofibers, which indicates that the sample is broken hollow nanofibers. Fig. 5b shows the HRTEM image of α-Fe₂O₃ hollow nanofibers. It is clearly demonstrated the well-textured and crystalline lattice with a distance of 0.27 nm matching very well with the lattice distance of (104) crystallographic plane of α-Fe₂O₃. The HRTEM image (Fig. 5d) of Fe₃O₄ hollow nanofibers displays clear crystal lattice with a spacing of 0.30 nm corresponding to the (220) crystallographic plane. The HRTEM image (Fig. 5f) exhibits clear lattice fringes with spacing of 0.21 nm, which correspond to the (−1−11) crystallographic plane of orthorhombic Fe₂N. Inset of Fig. 5a and c are the SAED patterns of α-Fe₂O₃ and Fe₃O₄ hollow nanofibers, they exhibit clear diffraction dots, demonstrating the single crystalline in nature. The SAED pattern (inset of Fig. 5e) of Fe₂N hollow nanofibers demonstrating the polycrystalline structure.

Fig. 5 TEM images (a, c and e) and HRTEM images (b, d and f) of α-Fe₂O₃ (a and b), Fe₃O₄ (c and d) and Fe₂N (e and f), the inset of (a, c and e) are the corresponding SAED patterns.

Based on above analyses, we propose a possible formation mechanism for α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers, as shown in Fig. 6. Firstly, PVP, residual DMF and Fe(NO₃)₃ were evenly dispersed in Fe(NO₃)₃/PVP composite nanofibers (step 1). During calcination process, nitrate was decomposed and formed the α-Fe₂O₃ crystallites, many crystallites were combined into nanoparticles, α-Fe₂O₃ nanoparticles moved to the surface of the composite fibers with evaporation of solvent DMF, then some nanoparticles were mutually connected to generate hollow-centered α-Fe₂O₃/PVP composite nanofibers (step 2). As the annealing temperature further increased, PVP was removed from the composite nanofibers via combustion. At the same time, α-Fe₂O₃ located near the surface of the composite nanofibers, and α-Fe₂O₃ hollow nanofibers were obtained (step 3). Then, α-Fe₂O₃ hollow nanofibers were heated to 350 °C in a quartz boat under a flow of NH₃ atmosphere, Fe₃O₄ hollow nanofibers were obtained (step 4). When nitridation temperature was 400 °C, α-Fe₂O₃ hollow nanofibers turned to Fe₂N hollow nanofibers, and Fe₂N hollow nanofibers were obtained (step 5).

Fig. 6 Possible formation mechanism for α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers.

3.2. Magnetic properties of α-Fe₂O₃, Fe₃O₄ and Fe₂N nanofibers

The magnetic properties of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers were investigated at room temperature in the applied magnetic field from −20 to 20 kOe. Typical hysteresis loops for α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers are shown in Fig. 7, and the magnetic properties data are summarized in Table 1. The curve in Fig. 7b indicates that Fe₃O₄ hollow nanofibers has a high saturation magnetization of 82.99 emu g⁻¹, and the remnant magnetization and coercivity are 39.27 emu g⁻¹ and 400.45 Oe, respectively. For α-Fe₂O₃ and Fe₂N hollow nanofibers, the saturation magnetization are 4.34 emu g⁻¹ and 2.07 emu g⁻¹, respectively, and neither remanence nor coercivity are detected.

Fig. 7 Hysteresis loops of α-Fe₂O₃ (a), Fe₃O₄ (b) and Fe₂N (c) hollow nanofibers.

Table 1 Magnetic properties of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers

Samples	Saturation magnetization (emu g^−1)	Remanence (emu g^−1)	Coercivity (Oe)
α-Fe2O3	4.34	—	—
Fe3O4	82.99	39.27	400.45
Fe2N	2.07	—	—

---

3.3. Electrochemical performance of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers

The discharge and charge profiles of α-Fe₂O₃ hollow nanofibers electrode in the voltage range from 0.01 to 3.00 V (vs. Li⁺/Li) at a current density of 100 mA g⁻¹ are shown in Fig. 8a. The plotted data reveal the 1^st, 2^nd and 200^th cycles of discharge and charge. The initial discharge curve of α-Fe₂O₃ hollow nanofibers electrode manifests two voltage plateaus, one voltage plateau is observed at 1.25 V due to lithium insertion into α-Fe₂O₃ with a structural change and the phase transformation from hexagonal α-Li_xFe₂O₃ to cubic Li_xFe₂O₃, the other voltage plateau is observed at 0.85 V indicating the complete reduction of iron from Fe(III) to Fe(0), which is typical of voltage trends for α-Fe₂O₃ electrodes.^33,34 For the charge curves, no obvious plateaus are found for the 1^st cycle, corresponding to the oxidation of Fe(0) to Fe(II) and further oxidation to Fe(III).²⁵ It is shown that the initial discharge and charge capacities are 1314 mA h g⁻¹ and 970 mA h g⁻¹, respectively. The phenomenon that the first discharge capacity considerably exceeds the theoretical capacity has been widely reported for transition metal oxides.^35,36 There is an irreversible capacity loss of 344 mA h g⁻¹ for the first cycle, which can be mainly attributed to the formation of the solid electrolyte interface (SEI), formation of gel-type polymeric layer and also partially due to the reduction of the solvent in the electrolyte.³⁷ During the second cycle, the discharge curve only indicates a slope at 0.82 V and the capacity decreases to 978 mA h g⁻¹. For the charge curves, no obvious plateaus are observed and the charge capacity is 930 mA h g⁻¹ for the 2^nd cycle. In the 200^th charge–discharge curves, discharge and charge capacities are 980 mA h g⁻¹ and 974 mA h g⁻¹, respectively, indicating superior cycling stability is obtained. The capacity retention over 200 cycles is 74.58%.

Fig. 8 Typical discharge and charge profiles (a), cycling performance and coulombic efficiency (b) of α-Fe₂O₃ hollow nanofibers electrode at a current density of 100 mA g⁻¹.

The cycling performance and coulombic efficiency of α-Fe₂O₃ hollow nanofibers electrode at a current density of 100 mA g⁻¹ are demonstrated in Fig. 8b. It is observed that the capacity of 980 mA h g⁻¹ at 100 mA g⁻¹ is retained after 200 cycles, indicating an excellent cycling stability is obtained. The superior electrochemical performances are attributed to the superfine hollow nanofibrous morphology which could enhance the contact between the electrolyte and electrode, facilitate the impregnation of the electrolyte into the electrode, shorten Li-ion diffusion path, the hollow core can buffer against the local volume change during charge–discharge and provide extra space for the storage of Li⁺, and hollow nanomaterials have big contacting area and fast diffusion rates along many grain boundaries existing in them.³⁸ More interestingly, accompanied with the cycle number increasing, the discharge capacity increases after 63 cycles and then gradually levels off. It is likely that Li-ion diffusion is activated and stabilized gradually during cycling.³⁹ The coulombic efficiency of the first cycle is poor as a result of the high irreversible capacity observed during the first discharge process. Nevertheless, after the first cycle, the coulombic efficiency of every cycle is more than 95.02%, implying a high capacity and charge–discharge reversibility of α-Fe₂O₃ hollow nanofibers electrode.

Fig. 9a reveals galvanostatic charge–discharge curves of Fe₃O₄ hollow nanofibers electrode in the 1^st, 2^nd and 300^th cycles at a current density of 100 mA g⁻¹. In the first discharge step, the voltage drops to 1.15 V, subsequently, the profile manifests a long voltage plateau at about 0.86 V corresponding to the transformation of Fe₃O₄ to Fe, followed by a sloping curve down to the cut-off voltage of 0.01 V, which are similar to literature descriptions for Fe₃O₄-based anodes.^40,41 The overall capacity at the end of the first discharge and charge cycles are 1210 mA h g⁻¹ and 915 mA h g⁻¹, respectively. There is an irreversible capacity loss of 295 mA h g⁻¹ for the first cycle, which can be mainly attributed to the formation of the SEI, formation of gel-type polymeric layer and also partially due to the reduction of the solvent in the electrolyte.³⁷ For the second discharge step, the discharge curve shows two slopes at 1.58 V and 0.80 V, and then drops in a smooth curve till the cut-off at 0.01 V. The total capacity at the end of the second discharge and charge cycles are 969 mA h g⁻¹ and 918 mA h g⁻¹, respectively. In the 300^th charge–discharge curves, discharge and charge capacities are 572 mA h g⁻¹ and 562 mA h g⁻¹, respectively, demonstrating high cycling stability. Fig. 9b reveals the cycling performance and coulombic efficiency of Fe₃O₄ hollow nanofibers electrode at a current density of 100 mA g⁻¹. It is noticeable that a high reversible capacity of about 572 mA h g⁻¹ is achieved after 300 cycles at 100 mA g⁻¹, implying an excellent cycling stability is obtained. Noticeably, the capacity increases gradually after the 60 cycle and then gradually levels off. This phenomenon has also been observed previously for metal oxide electrodes in long time cycles.⁴² For this phenomenon, we propose two possibilities as follows. (i) The decrement of capacity in the initial stage is accompanied with irreversible reactions that generate Fe nanoparticles. The metallic particles will increase the overall conductivity of the electrodes. As a result, the charge transfer kinetics will also be improved, resulting in the increment of the capacity in the following cycles. (ii) The decomposition of the electrolyte forms an gel-type polymeric SEI layer on the electrode surface. This SEI layer could improve the mechanical cohesion of the active materials without hindering the ions transfer. The SEI layer could also provide excess lithium ions storage sites by a so-called “pseudocapacitance-type” behavior, especially in the low potential region.⁴² The coulombic efficiency of the first cycle is poor as a result of the high irreversible capacity observed during the first discharge process. Nevertheless, after the first cycle, the coulombic efficiency is more than 89.82%, indicating a high capacity and charge–discharge reversibility of Fe₃O₄ hollow nanofibers.

Fig. 9 Typical discharge and charge profiles (a), cycling performance and coulombic efficiency (b) of Fe₃O₄ hollow nanofibers electrode at a current density of 100 mA g⁻¹.

The charge–discharge profiles of Fe₂N hollow nanofibers electrode at a current density of 100 mA g⁻¹ are manifested in Fig. 10a. In the first discharge curve, the sample exhibits long voltage plateau followed by a sloping curve down to 0.01 V, the long voltage plateau around 0.85 V. It is implied that the initial discharge and charge capacities are 750 mA h g⁻¹ and 563 mA h g⁻¹, respectively. There is an irreversible capacity loss of 187 mA h g⁻¹ for the first cycle, which can be mainly attributed to the formation of the SEI. The second discharge cycle has a different profile, the discharge curve shows two slopes at 1.53 V and 0.82 V, and then drops in a smooth curve till the cut-off at 0.01 V. The total capacity at the end of the second discharge and charge cycles are 580 mA h g⁻¹ and 553 mA h g⁻¹, respectively. In the 300^th charge–discharge curves, discharge and charge capacities are 438 mA h g⁻¹ and 437 mA h g⁻¹, respectively, demonstrating high cycling stability. Fig. 10b shows the cycling performance and coulombic efficiency of Fe₂N hollow nanofibers electrode at a current density of 100 mA g⁻¹. It can still remains above 438 mA h g⁻¹ after 300 cycles at a current density of 100 mA g⁻¹, indicating a good capacity retention is obtained. The coulombic efficiency of the first cycle is poor as a result of the high irreversible capacity observed during the first discharge process. Nevertheless, after the first cycle, the coulombic efficiency is more than 95.25%, implying a high capacity and charge–discharge reversibility of Fe₂N hollow nanofibers electrode.

Fig. 10 Typical discharge and charge profiles (a), cycling performance and coulombic efficiency (b) of Fe₂N hollow nanofibers electrode at a current density of 100 mA g⁻¹.

The electrochemical performances of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers electrodes are compared with Fe₂O₃, Fe₃O₄ and Fe₂N materials reported in the literatures,^{17,18,31,34,43–49} and summarized in Table 2. The as-prepared α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers electrodes demonstrate comparable or even better Li storage performance, which could be ascribed to their hollow nanostructure.

Table 2 Comparison of the electrochemical performances of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers electrodes with existing Fe₂O₃, Fe₃O₄ and Fe₂N electrodes reported in the literatures

Samples	Initial capacity (mA h g^−1)	Current density (mA g^−1)	Capacity retention (mA h g^−1)/(cycles)	References
α-Fe2O3 hollow nanofibers	1314	100	980 (200)	This work
α-Fe2O3 nanoparticles	1061	100	800 (70)	43
α-Fe2O3 nanoparticles	1500	200	115 (12)	44
Porous flower-like α-Fe2O3 nanostructures	1156	50	1069 (25)	45
Mesostructured 3D Fe2O3	1620	200	400 (100)	46
α-Fe2O3 nanorods	1515	50	1095 (50)	34
α-Fe2O3 submicron rods	1115	150	415 (30)	47
Fe3O4 hollow nanofibers	1210	100	572 (300)	This work
Fe3O4 nanoparticles	920	200	300 (80)	48
Fe3O4 hollow beads	1250	100	500 (50)	31
Fe3O4 mesoporous microspheres	1307	185	450 (110)	49
Fe2N hollow nanofibers	750	100	438 (300)	This work
Fe2N nanostructures	515	100	80 (100)	17
Fe2N power	240	6000	59 (300)	18

---

CV is a complementary technique to galvanostatic cycling and helps in understanding the nature and potentials of various structural transformations and redox reactions occurring during discharge–charge cycles.³⁴ Fig. 11a reveals the CV curves of α-Fe₂O₃ hollow nanofibers in the potential window of 0.01–3 V at the scan rate of 0.1 mV s⁻¹. It is clearly observed from the CV curves that there is substantial difference between the first and the subsequent cycles. During the first cycle, three cathodic peaks are observed at 1.63 V, 1.02 V and 0.72 V. The peaks at 1.63 V and 1.02 V correspond to formation of hexagonal α-Li_xFe₂O₃ and cubic Li_xFe₂O₃, without any crystal structure destruction.²⁵ The large peak at 0.72 V is due to the decomposition of Li_xFe₂O₃ and the crystal structure destruction accompanied by the complete reduction of Fe^2+/3+ to Fe⁰.³⁴ During the anodic scan, two peaks are located at 1.64 V and 1.85 V, corresponding to the oxidation of Fe⁰ to Fe²⁺ and further oxidation to Fe³⁺. The subsequent cycles are significantly different from that of the first cycle except a new cathodic peak appears at 0.82 V with decreasing intensity, while the anodic curve only showed one broad peak with small decrease in the intensity. The difference in the first and second cathodic curves is due to irreversible phase transformation during lithium insertion and extraction in the initial cycle. After the first cycle, the intensity of all the peaks remains almost the same, implying enhanced stability during lithiation and delithiation processes.

Fig. 11 CV curves of α-Fe₂O₃ (a), Fe₃O₄ (b) and Fe₂N (c) hollow nanofibers electrodes.

Fig. 11b shows the first three CV curves of Fe₃O₄ hollow nanofibers electrode between 0.01 V and 3.0 V at a scan rate of 0.1 mV s⁻¹. In agreement with literatures,^50,51 it is clear that the CV curve of the first cycle is quite different from those of subsequent cycles, especially for the discharge branch. In the first cathodic process, there is one intense peak located at around 0.65 V, corresponding to the initial reduction of Fe₃O₄ to Fe accompanying with the electrochemical formation of amorphous Li₂O. In the first anodic process, two peaks are located at around 1.65 and 1.87 V, which can be ascribed to the oxidation of Fe to Fe₃O₄ and the decomposition of Li₂O. The lithium storage mechanism of Fe₃O₄ electrode can be described by the electrochemical conversion reaction: .⁵² In the second cycle, both the cathodic and anodic peaks shift positively, which is ascribed to the polarization of the electrode materials in the first cycle. In comparison, the distinct peak appear at 0.70 V during discharge and at 1.92 V during charge from the second cycle onward. It is noteworthy that after the first cycle, the CV curves almost overlapped, suggesting a high reversibility of lithium storage.

The CV curves of Fe₂N hollow nanofibers electrode at a scan rate of 0.1 mV s⁻¹ are shown in Fig. 11c. During the first cathodic scan, the peaks observed at around 0.16 V and 0.67 V represent the lithiation reactions of Fe₂N, and another weak peak at around 0.45 V, corresponding to the formation of partially irreversible SEI layer. During the first anodic scan (Li extraction), the CV manifests two peaks at 1.63 V and 1.85 V. These peaks are the indication of decomposition of Li_3−xFe_xN and formation of Fe₂N. The lithium storage mechanism of Fe₂N electrode can be described by the electrochemical conversion reaction: .¹⁷ During the subsequent cycles, the main reduction peaks are shifted to 0.16 V, 0.81 V and 0.97 V, peak intensity drops significantly in the second cycle, indicating the occurrence of some irreversible processes in the electrode material in the first cycle. On the other hand, the oxidation peak at around 1.85 V in the anodic scan exhibits little change in the first three cycles, demonstrating a good reversibility of the electrochemical reaction.

In addition, Fe₂N hollow nanofibers electrode exhibits much better rate capability compared to α-Fe₂O₃ and Fe₃O₄ hollow nanofibers electrodes operated at various current density between 100 mA g⁻¹ and 1000 mA g⁻¹ (Fig. 12c), it can be found that the discharge remains stable and decreases regularly with an increased current density. After each 7 cycles at a specifical current density, the reversible capacities at 100, 200, 400, 600, 800 and 1000 mA g⁻¹ are about 567, 518, 478, 442, 410 and 395 mA h g⁻¹, implying that the rate cycling stability of Fe₂N hollow nanofibers electrode is excellent. Remarkably, when the current density is again reduced back to 100 mA g⁻¹, the discharge capacity can be recovered (even a little higher than the original capacity at 100 mA g⁻¹) (Fig. 12c). Whereas, the reversible capacities of α-Fe₂O₃ and Fe₃O₄ nanofibers electrodes rapidly drop with an increase current density, as implied in Fig. 12a and b.

Fig. 12 Rate capability of α-Fe₂O₃ (a), Fe₃O₄ (b) and Fe₂N (c) hollow nanofibers electrodes at various current densities.

To further understand the superior electrochemical performances of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers electrodes, an EIS measurement was performed with α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers electrodes. EIS technique is one of the widely used and powerful analytical techniques to investigate charge transfer and Li-ion diffusion kinetics in various electrode materials.⁵³ Fig. 13 indicates the Nyquist plots in the frequency range from 1 MHz to 0.01 Hz, with an amplitude of 10 mV. All of the measured values have similar Nyquist plots consisting of a semicircle at high frequency region, representing the charge transfer resistance between the electrode and electrolyte.^41,54 Apparently, the charge transfer resistance of Fe₂N (43.21 Ω) hollow nanofibers electrode is much lower than that of α-Fe₂O₃ (138.46 Ω) and Fe₃O₄ (91.63 Ω) hollow nanofibers electrodes, which can lead to rapid electron transport during the electrochemical lithium insertion/extraction reaction and thus result in significant improvement on the rate performance.

Fig. 13 Nyquist plots of α-Fe₂O₃, Fe₃O₄ and Fe₂N hollow nanofibers electrodes.

4. Conclusions

In summary, α-Fe₂O₃ magnetic hollow nanofibers were successfully synthesized via a facile electrospinning followed by post-calcination process, and then Fe₃O₄ and Fe₂N magnetic hollow nanofibers were fabricated by reduction and nitridation of α-Fe₂O₃ magnetic hollow nanofibers using NH₃ atmosphere through control of the calcination temperature. The α-Fe₂O₃, Fe₃O₄ and Fe₂N magnetic hollow nanofibers were used as anode materials for LIBs. α-Fe₂O₃ and Fe₃O₄ magnetic hollow nanofibers electrodes demonstrate extraordinary performance, such as high reversible capacity and excellent cycling stability. Fe₂N magnetic hollow nanofibers electrode exhibits excellent cycling performance, high coulombic efficiency and impressive rate capability. The superior electrochemical performances are attributed to the unique 1D hollow nanostructure of the materials. The synthetic method may be extended to the synthesis of other metal oxides and metal nitrides nanostructures as high performance anodes for LIBs.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (51573023, 50972020, 51072026, 21601018), the Science and Technology Development Planning Project of Jilin Province (20130101001JC, 20070402).

References

1. M. Armand and J. M. Tarascon, Nature, 2008, 451, 652–657.
2. J. M. Tarascon and M. Armand, Nature, 2001, 414, 359–367.
3. B. Wang, J. S. Chen, H. B. Wu, Z. Y. Wang and X. W. Lou, J. Am. Chem. Soc., 2011, 133, 17146–17148.
4. J. Ma, F. Yu, Z. H. Wen, M. X. Yang, H. M. Zhou, C. Li, L. Jin, L. Zhou, L. Chen, Z. W. Yuan and J. H. Chen, Dalton Trans., 2013, 42, 1356–1359.
5. J. Cabana, L. Monconduit, D. Larcher and M. R. Palacín, Adv. Mater., 2010, 22, E170–E192.
6. F. Y. Cheng, J. Liang, Z. L. Tao and J. Chen, Adv. Mater., 2011, 23, 1695–1715.
7. Y. Z. Jiang, D. Zhang, Y. Li, T. Z. Yuan, N. Bahlawane, C. Liang, W. P. Sun, Y. H. Lu and M. Yan, Nano Energy, 2014, 4, 23–30.
8. E. Kang, Y. S. Jung, A. S. Cavanagh, G. H. Kim, S. M. George, A. C. Dillon, J. K. Kim and J. Lee, Adv. Funct. Mater., 2011, 21, 2430–2438.
9. H. Liu, G. X. Wang, J. Liu, S. Z. Qiao and H. Ahn, J. Mater. Chem., 2011, 21, 3046–3052.
10. D. H. Ge, J. J. Wu, G. L. Qu, Y. Y. Deng, H. B. Geng, J. W. Zheng, Y. Pan and H. W. Gu, Dalton Trans., 2016, 45, 13509–13513.
11. F. D. Wu and Y. Wang, J. Mater. Chem., 2011, 21, 6636–6641.
12. J. Gao, M. A. Lowe and H. D. Abruña, Chem. Mater., 2011, 23, 3223–3227.
13. K. F. Zhong, B. Zhang, S. H. Luo, W. Wen, H. Li, X. J. Huang and L. Q. Chen, J. Power Sources, 2011, 196, 6802–6808.
14. M. Zhang, B. H. Qu, D. N. Lei, Y. J. Chen, X. Z. Yu, L. B. Chen, Q. H. Li, Y. G. Wang and T. H. Wang, J. Mater. Chem., 2012, 22, 3868–3874.
15. J. S. Chen, T. Zhu, X. H. Yang, H. G. Yang and X. W. Lou, J. Am. Chem. Soc., 2010, 132, 13162–13164.
16. Y. H. Chang, J. Li, B. Wang, H. Luo, H. Y. He, Q. Song and L. J. Zhi, J. Mater. Chem. A, 2013, 1, 14658–14665.
17. P. Yu, L. Wang, F. F. Sun, D. D. Zhao, C. G. Tian, L. Zhao, X. Liu, J. Q. Wang and H. G. Fu, Chem.–Eur. J., 2015, 21, 1–9.
18. M. S. Balohun, M. H. Yu, Y. C. Huang, C. Li, P. P. Fang, Y. Liu, X. H. Lu and Y. X. Tong, Nano Energy, 2015, 11, 348–355.
19. L. W. Ji, Z. K. Tan, T. R. Kuykendall, S. Aloni, S. Xun, E. Lin, V. Battaglia and Y. G. Zhang, Phys. Chem. Chem. Phys., 2011, 13, 7170–7177.
20. J. M. Jeong, B. G. Choi, S. C. Lee, K. G. Lee, S. J. Chang, Y. K. Han, Y. B. Lee, H. U. Lee, S. Kwon, G. Lee, C. S. Lee and Y. S. Huh, Adv. Mater., 2013, 19, 4505–4509.
21. L. W. Ji, O. Toprakci, M. Alcoutlabi, Y. F. Yao, Y. Li, S. Zhang, B. K. Guo, Z. Lin and X. W. Zhang, ACS Appl. Mater. Interfaces, 2012, 4, 2672–2679.
22. G. X. Gao, L. Yu, H. B. Wu and X. W. Lou, Small, 2014, 10, 1741–1745.
23. Y. Wu, Y. Wei, J. P. Wang, K. L. Jiang and S. S. Fan, Nano Lett., 2013, 13, 818–823.
24. M. Latorre-Sanchez, A. Primo and H. Garcia, J. Mater. Chem., 2012, 22, 21373–21375.
25. S. Chaudharia and M. Srinivasan, J. Mater. Chem., 2012, 22, 23049–23056.
26. D. Li and Y. N. Xia, Adv. Mater., 2004, 16, 1151–1170.
27. Z. Y. Hou, G. G. Li, H. Z. Lian and J. Lin, J. Mater. Chem., 2012, 22, 5254–5276.
28. D. Q. Shao, J. X. Wang, X. T. Dong, W. S. Yu, G. X. Liu, F. F. Zhang and L. M. Wang, J. Mater. Sci.: Mater. Electron., 2013, 24, 4263–4269.
29. D. Q. Shao, J. X. Wang, X. T. Dong, W. S. Yu, G. X. Liu, F. F. Zhang and L. M. Wang, J. Mater. Sci.: Mater. Electron., 2013, 24, 4718–4724.
30. Z. Wang, L. Zhou and X. W. Lou, Adv. Mater., 2012, 24, 1903–1911.
31. Y. Chen, H. Xia, L. Lu and J. M. Xue, J. Mater. Chem., 2012, 22, 5006–5012.
32. C. Lei, F. Han, Q. Sun, W. C. Li and A. H. Lu, Chem.–Eur. J., 2014, 20, 139–145.
33. J. S. Cho, Y. J. Hong and Y. C. Kang, ACS Nano, 2015, 9, 4026–4035.
34. C. T. Cherian, J. Sundaramurthy, M. Kalaivani, P. Ragupathy, P. S. Kumar, V. Thavasi, M. V. Reddy, C. H. Sow, S. G. Mhaisalkar, S. Ramakrishna and B. V. R. Chowdari, J. Mater. Chem., 2012, 22, 12198–12204.
35. B. T. Hang, T. Doi, S. Okada and J. Yamaki, J. Power Sources, 2007, 174, 493–500.
36. S. Liu, L. Zhang, J. Zhou, J. Xiang, J. Sun and J. Guan, Chem. Mater., 2008, 20, 3623–3628.
37. M. V. Reddy, G. Prithvi, K. P. Loh and B. V. R. Chowdari, ACS Appl. Mater. Interfaces, 2014, 6, 680–690.
38. S. Chaudhari and M. Srinivasan, J. Mater. Chem., 2012, 22, 23049–23056.
39. N. Venugopal, D. J. Lee, Y. J. Lee and Y. K. Sun, J. Mater. Chem. A, 2013, 1, 13164–13170.
40. S. M. Yuan, J. X. Li, L. T. Yang, L. W. Su, L. Liu and Z. Zhou, ACS Appl. Mater. Interfaces, 2011, 3, 705–709.
41. C. N. He, S. Wu, N. Q. Zhao, C. S. Shi, E. Z. Liu and J. J. Li, ACS Nano, 2013, 5, 4459–4469.
42. J. S. Luo, J. L. Liu, Z. Y. Zeng, C. F. Ng, L. J. Ma, H. Zhang, J. Y. Lin, Z. X. Shen and H. J. Fan, Nano Lett., 2013, 13, 6136–6143.
43. D. Mhamane, H. K. Kim, V. Aravindan, K. C. Roh, M. Srinivasan and K. B. Kim, Green Chem., 2015, 18, 1395–1404.
44. S. W. Hwang, A. Umar and S. H. Kim, J. Nanosci. Nanotechnol., 2015, 15, 5129–5134.
45. T. R. Penki, S. Shivakumara, M. Minakshi and N. Munichandraiah, Electrochim. Acta, 2015, 167, 330–339.
46. J. J. Wang, H. Zhou, J. Nanda and P. V. Braun, Chem. Mater., 2015, 27, 2803–2811.
47. L. H. Huang, Z. H. Min and Q. Y. Zhang, Mater. Res. Bull., 2015, 66, 39–44.
48. X. Y. Qin, H. R. Zhang, J. X. Wu, X. D. Chu, Y. B. He, C. P. Han, C. Miao, S. Wang, B. H. Li and F. Y. Kang, Carbon, 2015, 87, 347–356.
49. J. S. Xu and Y. J. Zhu, ACS Appl. Mater. Interfaces, 2012, 4, 4752–4757.
50. S. H. Lee, S. H. Yu, J. E. Lee, A. Jin, D. J. Lee, N. Lee, H. Jo, K. Shin, T. Y. Ahn, Y. W. Kim, H. Choe, Y. E. Sung and T. Hyeon, Nano Lett., 2013, 13, 4249–4256.
51. J. Z. Wang, C. Zhong, D. Wexler, N. H. Rdris, Z. X. Wang, L. Q. Chen and H. K. Liu, Chem.–Eur. J., 2011, 17, 661–667.
52. Y. He, L. Huang, J. S. Cai, X. M. Zheng and S. G. Sun, Electrochim. Acta, 2010, 55, 1140–1144.
53. B. G. Choi, S. J. Chang, Y. B. Lee, J. S. Bae, H. J. Kim and Y. S. Huh, Nanoscale, 2012, 4, 5924–5930.
54. X. L. Jia, Z. Chen, X. Cui, Y. T. Peng, X. L. Wang, G. Wang, F. Wei and Y. F. Lu, ACS Nano, 2012, 6, 9911–9919.

---