Introduction

Rechargeable lithium–ion batteries (LIBs) have been the most widely used power source for portable electronic devices due to their long cycling life, high energy, and power density [1, 2]. As a key component, electrode material dominates the electrochemical properties of LIBs. Currently, graphite-based anode is utilized in most commercial rechargeable LIBs. However, it could not meet the increasing needs of power battery application because it can only deliver a theoretical capacity of about 372 mA h g−1 [2, 3]. Recently, it has been found that transition metal oxides, such as Fe2O3 [4–6], Fe3O4 [7], Co3O4 [8, 9], SnO2 [2, 10], and NiO [11] exhibit high reversible capacities, greatly spurring the rapid development of this field. Among these promising anode materials, Fe2O3 has long been regarded as an appealing host material for lithium storage due to its high theoretical capacity (1007 mA h g−1), nontoxicity, and low processing cost [5, 12]. However, in spite of these excellent characteristics, Fe2O3 suffers from poor conductivity and large volume change in the electrochemical reaction [13–15], which would lead to electrode pulverization and loss of contact with current collector, resulting in a large capacity loss, low initial coulombic efficiency, and poor rate and cycling stability.

Numerous measures are devoted to alleviate mechanical degradation and capacity fading. Among all the methods, an effective approach is structural modification to buffer volume change during the discharge and charge processes. In particular, three-dimensional (3D) hollow nanotube arrays’ electrode has attracted considerable attention due to its large internal void, high surface area, and outstanding permeation [16–18]. The thin shells in hollow nanostructures could provide a fast and efficient transport for both Li+ and electrons, which contributes to the improved reaction kinetics at the electrode surface [19]. Nevertheless, the electrode may degrade during long-term cycling because the hollow structure without protection would still expand out toward the electrolyte during lithiation, and shrink during delithiation [20]. As a consequence, the solid electrolyte interphase (SEI) film formed during Li+ insertion would be broken and decomposed completely [21, 22], resulting in the exposure of some fresh active materials to the electrolyte and repeated regeneration of the SEI film, which would cause consumption of abundant Li+ and a large irreversible capacity (i.e., low coulombic efficiency) [23, 24].

In general, one strategy to obtain the stable SEI film is to create core/shell nanostructures [25, 26]. Liu et al. have successfully prepared a 3D carbon/hematite (C/α–Fe2O3) nanotube arrays’ electrode, exhibiting better capacity retention and rate capability compared with the bare 3D Fe2O3 nanotube arrays. Nevertheless, the designed carbon/hematite composite anode is easily reduced to Fe2+ or Fe0 at 600 °C due to its poor thermal stability. Hence, the carbon precursors must be carefully chosen, and the time of carbonization should be prolonged to render moderate electrical conductivity of the components [27]. Recently, a great number of researches have proposed some electronically conducting polymers with high conductivity, chemical stability, and low cost as surface coating layer and/or active matrix [28–31]. Among them, polypyrrole (PPy) is regarded as one of the most attractive conducting polymers [29–31], which could not only lower the charge-transfer resistance in electrochemical reactions, but also prevent the direct contact between active materials and electrolytes. More importantly, the volume expansion of the electrode could be restricted into the PPy coating layer, which is helpful to protect the SEI film from rupturing during delithiation, thus ensuring high coulombic efficiency and long-term stable cycling [20].

Inspired by these needs, we fabricated vertically aligned Fe2O3 nanotube arrays on stainless steel substrate. Furthermore, a conductive PPy additive layer was coated on Fe2O3 via a chemical vapor-phase polymerization to prepare Fe2O3@PPy electrode. Due to the protection of PPy coating layer, the hollow Fe2O3 nanotube could expand into its internal void space and maintain a static outer surface, which is beneficial for the development of a stable SEI film and the maintenance of a long cycle life. As a result, the Fe2O3@PPy nanotube arrays’ electrode delivered higher reversible capacities and superior rate performances compared with the bare Fe2O3 electrode. We believe that the facile synthesis of Fe2O3@PPy nanotube arrays’ electrode could be favorably considered as a promising candidate for practical applications among lithium–ion anodes.

Experimental

Materials

All chemicals were analytical grade and used without further purification. Fe(NO3)3·9H2O (>99 %), sodium dodecyl sulfate (SDS), ammonium persulfate (APS) were obtained from Kewei Chemical Reagent Co. Ltd. of Tianjin University (Tianjin, China). Pyyole monomer (98 % reagent grade) was purchased from Sigma Aldrich PTE. Ltd, Singapore. The stainless steel foil (60 μm) was purchased from Shanghai Chemical Reagent Co. Ltd., pretreated by sonication in absolute ethanol and distilled water successively and dried in air at 40 °C.

Preparation of 3D Fe2O3@PPy nanotube arrays on stainless steel substrates

Conventional methods to prepare vertically aligned nanotube arrays have been proposed by template-directed growth with the utilization of ZnO nanorod arrays template as the growth director [32–34]. Herein, we used them to grow Fe2O3 nanotubes on a piece of stainless steel foil via a sacrificial template-accelerated hydrolysis method [35]. In brief, ZnO nanorod arrays [36] were grown on stainless steel foil and placed into a 40 mL aqueous solution containing 0.1 M Fe(NO3)3·9H2O. Then the sample was kept still at room temperature for 12 h. After immersion, it was taken out, dried in air, and treated at 450 °C in argon for 5 h to obtain Fe2O3 nanotube arrays. The PPy coating was conducted by chemical vapor-phase polymerization as follows: the resulting Fe2O3 nanotube arrays were put into a 0.1 mL solution composed of 3 mM APS as the oxidant and 0.15 mM SDS as the dopant, ultrasonicated for a moment and kept still for 30 min at room temperature. After the solution had been absorbed, the nanotube arrays were put into a beaker where a vial containing 20 μL pyrrole monomer was used for generate pyrrole vapor, without direct contact to the sample. After sealing, the beaker was left at room temperature for 15 min. Finally, the product was taken out and washed with deionized water, and dried at 80 °C overnight.

Materials characterization

The crystal structure of the product was determined by grazing incidence X-ray diffraction (GIXRD) using an automated RIGAKU/DMAX2500 X-ray diffractometer with monochromatic Cu Kα radiation, the 2 theta Bragg angles were scanned over a range of 20°–80° at a rate of 5.0° min−1. The size and morphology of the product were determined by a Hitachi S-4800 field-emission scanning electron microscope (FE-SEM) at an accelerating voltage of 5 kV and a high-resolution transmission electron microscope (HR-TEM, JEM-2100F) operated at 200 kV. Fourier transform infrared (FT-IR) spectra were recorded using an FT-IR analyzer (EQUINOX55). The valence state of iron was investigated by X-ray photoelectron spectroscopy (XPS) (PHI 1600 ESCA system). An Al Kα monochromatized radiation was employed as the X-ray source.

Electrochemical analyses

The electrochemical experiment was performed using coin cells (type CR2032) assembled in an argon-filled dry glove box by directly using the as-prepared bare Fe2O3 and Fe2O3@PPy nanotube arrays as the positive electrode. For comparison purpose, the pure PPy electrode was fabricated by coating the slurry comprising of 80 wt% PPy, 10 wt% acetylene black and 10 wt% polyvinylidene fluoride (PVDF) onto the stainless steel foil. Li metal foil was utilized as the counter electrode, 1 M LiPF6 in ethylene carbonate (EC) and dimethyl carbonate (DMC) (1:1 by volume) was used as the electrolyte and Celgard 2400 was used as the separator. The galvanostatic charge–discharge measurement and rate performance were performed over the potential range 0.01–3.00 V (versus Li+/Li) using a MTI Battery Testing System at room temperature. Electrochemical impedance spectroscopy measurements were performed in an alternating current frequency range from 1 to 1 MHz. The capacities of bare Fe2O3 and Fe2O3@PPy anode were reported based on the weight of Fe2O3.

Results and discussion

The crystallographic structures of bare Fe2O3 and Fe2O3@PPy nanotube arrays were investigated by grazing incidence X-ray diffraction analysis as shown in Fig. 1a. For curve (1), the diffraction peaks labeled with stars come from the stainless steel foil [37]. The peaks labeled with solid circles located at 24.1°, 33.1°, 35.6°, 40.8°, 49.4°, 54.0°, 57.6°, 62.4°, and 64.0° can be indexed to the standard hematite α-Fe2O3 (JCPDS 33-0664), which are assigned to the (012), (104), (110), (113), (024), (116), (018), (214), and (300) lattice planes, respectively. A wide diffraction peak is observed around 21.5°–28.5° in curve (2), which can be attributed to the amorphous PPy phase [38]. Fe2p XPS high-resolution spectrum was conducted to further confirm the valence state of Fe in the as-prepared nanotube arrays. As observed from Fig. 1b, Fe2p3/2 and Fe2p1/2 appeared at binding energies of about 710.9 and 724.8 eV with a shake-up satellite around 718.8 eV, which are attributed to hematite phase (α-Fe2O3) [39].

Fig. 1
figure 1

a XRD patterns of (i) bare Fe2O3 nanotube arrays and (ii) Fe2O3@PPy nanotube arrays; b XPS high-resolution spectrum of Fe2p core level for bare Fe2O3 nanotube arrays

Full size image

Figure 2a and b show the scanning electron microscopy (SEM) images of ZnO and Fe2O3 nanotube arrays, from which we can confirm that the array structure of ZnO is well maintained after the formation of Fe2O3 nanotubes. The columnar nanotubes have closed tips and are assembled from numerous nanoparticles, with outer diameters ranging from 150 to 250 nm (Fig. 2b). In order to measure the length of the nanotubes, the sample was rinsed with deionized water to present the cross section of array on the substrates before annealing at 450 °C. As shown in Fig. 2c, the length of the nanotubes is about 1.18 μm. More evidence about the structure of the product was given by transmission electron microscopy (TEM) analysis. Tubular structures with wall thickness in the range of 50–60 nm and inner diameter of 100 nm can be observed in the inset of Fig. 2b. HRTEM shown in Fig. 2d confirmed that the Fe2O3 nanotubes are composed of polycrystalline composition. The distances between the adjacent planes are 0.27, 0.25, and 0.23 nm corresponding to (104), (110), and (113) planes of α-Fe2O3, respectively [40]. The results are consistent with XRD.

Fig. 2
figure 2

a SEM image of ZnO nanorod arrays; b and c SEM images of bare Fe2O3 nanotube arrays, inset in b showing the TEM image of Fe2O3 nanotube; d HRTEM image of Fe2O3 nanotube

Full size image

A schematic depiction for the synthesis of the 3D Fe2O3@PPy nanotube arrays is presented in Fig. 3. The ZnO template was etched by H+ produced from Fe(NO3)3 hydrolysis, which in turn accelerated the hydrolysis and acted as a crucial step initiating the construction of the 3D nanotube arrays’ architecture. In order to improve the electronic conductivity and the stabilization of SEI film, we designed a thin layer of conducting PPy via a chemical vapor-phase polymerization, which was anchored on the surface of Fe2O3 nanotubes uniformly. Figure 4a shows scanning electron microscopy (SEM) images of the Fe2O3@PPy nanotube arrays. Obviously, after coating PPy onto the Fe2O3 nanotubes, some big bumps are generated on the surface. PPy is uniformly coated on the surface of each Fe2O3 nanotube, and there are some spaces between each pair of nanotubes. The outer diameter of Fe2O3@PPy nanotubes is about 200–300 nm. Upon combining with the SEM image of Fe2O3 nanotube arrays (Fig. 1c), we could deduce that the thickness of PPy layer is about 20 nm. The TEM image (Fig. 4b) clearly reveals that the nanotubes are uniformly and continuously coated with a thin PPy layer. A small part of the photo, which was denoted by the red rectangle, was chosen for EDS mapping analysis to further distinguish the compositions. As shown in Fig. 4c, C and N have a wider distribution than Fe and O, which is unambiguously illustrated. The molecular structures of the Fe2O3@PPy composite were characterized by FT-IR spectroscopy. The characteristic absorptions of the Fe2O3@PPy composites are similar to that of the PPy, as presented in Fig. 5, which demonstrates the existence of PPy in the composite. It can be clearly seen that the characteristic peaks of PPy are located at around 1610, 1470, 1398, 1235, 1047, 925, and 800 cm−1, which is in perfect accordance with the literature. The fundamental vibration bands of the PPy ring appear at around 1610, 1470, and 1398 cm−1 [41]. The peak at 1235 cm−1 is attributed to C–N stretching vibrations [42], at 1045 cm−1 to the =C–H in-plane vibrations, and at 925 and 800 cm−1 to C–H out-of-plane deformation vibrations of the ring [41].

Fig. 3
figure 3

Schematic illustration of the formation process of Fe2O3@PPy nanotube arrays

Full size image
Fig. 4
figure 4

a SEM image of Fe2O3@PPy nanotube arrays; b TEM image of Fe2O3@PPy nanotubes; c corresponding elemental mappings of Fe, O, C, and N of the selected area in b

Full size image
Fig. 5
figure 5

FT-IR spectra of pure PPy and Fe2O3@PPy nanotube arrays

Full size image

The electrochemical cycle performance of the Fe2O3@PPy nanotube arrays’ electrode was evaluated by galvanostatic charge/discharge cycling at a current density of 100 mA g−1 from 3.00 to 0.01 V on the basis of the mass of Fe2O3. For comparison, bare Fe2O3 nanotube arrays and pure PPy electrodes were also investigated under the same conditions. The galvanostatic discharge–charge curves of bare Fe2O3 and Fe2O3@PPy nanotube arrays are shown in Fig. 6a and b. The electrochemical behaviors of the two electrodes are similar, which is in general consistency with the literature [43, 44]. For the discharge curve in the first cycle, the voltage initially decreases quickly to approximately 1.5 V followed by a weak slope, corresponding to the initial lithium insertion into the Fe2O3 [Eq. (1)]. In the next step of lithium intercalation, a wide slope located at 1.2–0.8 V is observed, where there is a phase transformation from the hexagonal LixFe2O3 to the cubic Li2Fe2O3 [Eq. (2)]. An obvious plateau is observed at 0.8 V, indicating the complete reduction of iron from Fe2+ to Fe0 [Eq. (3)].

Fig. 6
figure 6

Discharge–charge curves of a bare Fe2O3 nanotube arrays’ electrode and b Fe2O3@PPy nanotube arrays' electrodes; c cycling performance; and d coulombic efficiencies of both Fe2O3@PPy nanotube arrays’ and control electrodes

Full size image
$$ {\text{Fe}}_{ 2} {\text{O}}_{ 3} + {\text{xLi}}^{ + } + {\text{xe}}^{ - } \to {\text{Li}}_{\text{x}} {\text{Fe}}_{ 2} {\text{O}}_{ 3} $$
(1)
$$ {\text{Li}}_{\text{x}} {\text{Fe}}_{ 2} {\text{O}}_{ 3} + \left( { 2- {\text{x}}} \right){\text{Li}}^{ + } + \left( { 2- {\text{x}}} \right){\text{e}}^{ - } \to {\text{Li}}_{ 2} {\text{Fe}}_{ 2} {\text{O}}_{ 3} $$
(2)
$$ {\text{Li}}_{ 2} {\text{Fe}}_{ 2} {\text{O}}_{ 3} + 4 {\text{Li}}^{ + } + 4 {\text{e}}^{ - } \to 2 {\text{Fe}}^{0} + 3 {\text{Li}}_{ 2} {\text{O}} $$
(3)

Continuation of the discharge reaction up to the lower cut-off (0.01 V) shows a sloping profile, which can be ascribed to the formation of SEI films along with the decomposition of the electrolyte and further lithium storage via interfacial charging at metal/Li2O interface [44, 45]. The first discharge and charge capacities of Fe2O3@PPy nanotube arrays are 1424 and 1088 mA h g−1, with a higher coulombic efficiency of 76 % compared with the bare Fe2O3 electrode (69 %). During the subsequent cycles, the discharge curves only show a slope at 1.0–0.75 V, and the capacities are reduced, which is caused by the incomplete conversion reaction and irreversible lithium loss due to the formation of the SEI film. The specific capacities from discharge profiles in the 30th and 50th cycles of the bare Fe2O3 electrode have dropped obviously, which is caused by the structural damages of the electrode and the continuous formation of the SEI films during cycling. By contrast, the specific capacity of Fe2O3@PPy electrode has basically reached a stable value after 30 cycles, and the discharge–charge profiles of the 30th and 50th cycles remain nearly unchanged, demonstrating the compositional and the structural stability of the PPy-coated electrode.

Figure 6c and d show the cycle performance of both Fe2O3@PPy and control electrodes. The capacity of the PPy electrode retains only 18 mA h g−1 after 10 cycles, in accordance with the literature [38]. Therefore, the low capacity of PPy can be ignored in analysis of Fe2O3@PPy anode. Although the initial capacity of bare Fe2O3 electrode is 1552 mA h g−1, it declines rapidly in the following cycles and maintains only about 390 mA h g−1 after 150 cycles. The poor cycle performance is caused by the unstable SEI film formed during Li+ insertion, which is decomposed easily, catalyzed by the iron metal formed during the reduction process [21]. On the contrary, the capacity of Fe2O3@PPy electrode exceeds that of bare Fe2O3 only after five cycles because of its smaller irreversible capacity. Even though the capacity decays gradually in the first 30 cycles, both charge and discharge capacities decrease slowly in the subsequent cycles and remain nearly constant with coulombic efficiency of higher than 97 %. After 150 cycles, the reversible capacity retains as high as 665 mA h g−1. The thin conductive PPy layer could not only prevent direct contact between Fe2O3 and electrolyte, but also inhibit the outward expansion of Fe2O3, thus improving the structural integrity of the electrode and facilitating the stability of SEI film formed on the surfaces, which is beneficial to the enhanced electrochemical performances.

Rate performances of the two samples are presented in Fig. 7. Fe2O3@PPy nanotube arrays deliver the discharge capacities of 861, 694, 592, and 455 mA h g−1 at current densities of 100, 200, 400, and 800 mA g−1 every 10 cycles, respectively. Even at a high current density of 1600 mA g−1, the specific capacity still retains a value of 333 mA h g−1 and rebounds to 678 mA h g−1 when the current density is reset to 100 mA g−1, indicating the excellent tolerance against the high current density of Fe2O3@PPy nanotube arrays’ electrode. In comparison, the capacity of bare Fe2O3 nanotube arrays decreases more dramatically to 167 mA h g−1 at a current density of 1600 mA g−1 after 50 cycles.

Fig. 7
figure 7

Rate performances of bare Fe2O3 and Fe2O3@PPy nanotube arrays’ electode

Full size image

It has been proposed that the cycle stability and high rate capability are mainly related to the thickness of SEI film, interfacial charge-transfer process, and lithium–ion diffusion in a composite. To gain further insight into the electrochemical performance, electrochemical impedance spectroscopy (EIS) measurements were carried out at room temperature on the cells after the first and 150 cycles, respectively, at the fully delithiation state. The complex plane plots for each sample can be divided into high-frequency semicircle and low-frequency sloping line (Fig. 8a, b). The intercepts at the real impedance (Z′) axis in the high-frequency region are generally considered as electrical conductivities of the electrode, the electrolyte, and the separator [46], and it remains almost constant after 150 cycles for both electrodes. The diameter of the semicircle in the high-frequency range is correlated to the resistance of SEI layer formed on the surface of electrode (R sf) and to the charge-transfer resistance (R ct) between the surface films and the active material [16]. No obvious difference of R (sf+ct) between the two sample can be seen after one cycle. However, the R (sf+ct) of bare Fe2O3 nanotube arrays is much higher than that of Fe2O3@PPy after 150 cycles, which could be attributed to the stable structure of Fe2O3@PPy nanotube arrays, demonstrating that PPy could promote charge transfer and improve the electrochemical performance of anode. Furthermore, in the low-frequency area, the phase angle for impedance plot of the Fe2O3@PPy nanotube arrays is much higher than that of the bare Fe2O3 nanotube arrays, indicating a faster Li+ diffusion in the Fe2O3@PPy nanotube’s structure during repeated charge/discharge processes.

Fig. 8
figure 8

The EIS spectra of bare Fe2O3 and Fe2O3@PPy nanotube arrays’ electrode after a the first and b 150 cycles, respectively

Full size image

To check the valence state of iron after the coulombic efficiency becomes roughly constant, the Fe2p3/2 XPS high-resolution spectra of bare Fe2O3 and Fe2O3@PPy electrodes after the 35th cycling at fully delithiation state were recorded. For both electrodes, the polymer layers on the surface were removed by scotch tape method in a glove box prior to analysis to expose the interior of the electrodes [47]. As shown in Fig. 9, the peaks at 710.9 and 709.7 eV are assigned to Fe3+ in an oxide matrix (Fe2O3 and/or Fe3O4); the peak at 708.3 eV is attributed to Fe2+ in an oxide matrix (Fe3O4); and the minor component at 714.3 eV is attributed to a Fe2+ 2p3/2 satellite [39, 48, 49]. As can be observed, although the peak intensity of Fe3+ is much larger than that of Fe2+ in both electrodes, the proportion of Fe3+–Fe2+ in Fe2O3@PPy electrode (Fig. 9b) is higher compared to that of the bare Fe2O3 electrode (Fig. 9a), indicating a lower content of Fe2+. The presence of Fe2+ in the electrode after cycling may be caused by the incompletely restored of hematite (Fe3+) during the delithiation process in the initial several cycles. Li+ would be trapped in the anode material along with the formation of Fe2+, and the capacity would decrease [37]. The lower content of Fe2+ in Fe2O3@PPy electrode is due to the faster transport of both Li+ and electrons during repeated charge/discharge processes, which is beneficial for the good reversibility of the electrochemical reactions.

Fig. 9
figure 9

XPS high-resolution spectra of Fe2p3/2 core level for a bare Fe2O3 and b Fe2O3@PPy nanotube arrays’ electrode after 35 cycles

Full size image

The enhanced structure and surface stabilization are demonstrated through SEM investigation. Figure 10 shows the SEM images of bare Fe2O3 and Fe2O3@PPy nanotube arrays after 150 cycles at a current density of 100 mA g−1. After the lithiation/delithiation cycles, the surface of Fe2O3@PPy nanotubes becomes rough with a modest increase of around 50 nm in diameter, but the morphology remains the same as before (Fig. 10b). On the contrary, the lithiated bare Fe2O3 nanotubes suffer a dramatic radial expansion with huge crack and pulverization phenomenon (Fig. 10a), leading to the formation of thick SEI films which would cause the consumption of more Li+ and a large irreversible capacity.

Fig. 10
figure 10

SEM images of a bare Fe2O3 and b Fe2O3@PPy nanotube arrays’ electrodes after 150 cycles

Full size image

Overall, several factors have contributed to the excellent electrochemical performance of hybrid nanotube arrays’ electrode. First, the well-defined hollow nanotube structure is capable of providing larger materials/electrolyte interface for faster reaction kinetics. Second, the introduction of PPy and the direct connection with the current collector can improve the electronic conductivity and provide an elastic protection to keep the integrity of the structure. Last but not the least, the homogenous coating of PPy is beneficial to form a stable SEI film, which could suppress the detrimental surface effects on the electrode, leading to the improved cycling characteristics.

Conclusions

We have designed and assembled a novel nanostructure of PPy coating on hollow Fe2O3 nanotube arrays for electrode material. In this well-engineered hybrid architecture, the preferred structures of each component are thoroughly conducive to guarantee highways for electron transport and maintain excellent structural/interfacial stability. By combining such fascinating advantages of ordered 3D Fe2O3 nanotube arrays and conductive PPy layer, the electrode exhibits superior electrochemical stability and high rate performance. Besides, the morphology is well preserved after repeated Li+ insertions and extractions. This 3D hybrid nanostructure might open new avenues for the design of other high-performance cathode and anode electrode materials.