Introduction

3d transition-metal oxides, which can be used as anode materials, such as iron oxide, cobalt oxide, and nickel oxide have attracted a great deal of attentions for their much higher capacity than that of conventional graphite (372 mAhg−1)1,2,3,4,5,6,7,8,9,10,11,12. For instance, the theoretical capacity of Co3O4 is about 890 mAhg−1, which is almost two and a half times higher than that of graphite. However, high price and toxicity of Co limit the application of Co3O413,14. Interestingly, Fe2O3 also exhibits high capacity (1007 mAhg−1) like Co3O4. More importantly, due to its low cost, nontoxicity and high resistance to corrosion, Fe2O3 has attracted special attentions in recent years15,16. For example, various morphologies of Fe2O3 such as nanoparticles, nanotubes, hollow structure and thin films have been studied as electrodes for lithium-ion batteries17,18,19,20,21,22,23,24,25. Lou et al. prepared a series of hollow microspheres of iron oxides which showed significantly improved lithium-storage capabilities24,25. In addition to the above-mentioned methods, many studies have proved that the crystal plane structure of electrode materials has a significant effect on the electrochemical properties. Islam et al. reported that the (010) plane of LiFePO4 is a favorable plane for fast Li+ transport26. Wei et al. found that the electrochemical performance of lithium rich material Li(Li0.17Ni0.25 Mn0.58)O2 with (010) and (100) planes have been greatly increased, exhibiting not only a high reversible capacity but also an excellent cycle stability27. Huang et al. found the facet-dependent electrochemical properties of Co3O4 toward heavy metal ions and found that the Co3O4 nanoplates with (111) facet performed better electrochemical sensing capability than the Co3O4 nanocubes with (001) facet28. Not long ago, we also reported the facet-dependent electrochemical capability of Co3O4 as anode material for Li-ion batteries and proved the Co3O4 octahedron with exposed (111) plane exhibited more excellent electrochemical properties than that of Co3O4 cube with exposed (001) plane and Co3O4 truncated octahedron with exposed (001) and (111) planes29. Therefore, studies on the crystal plane controllable synthesis of nanomaterials are of great interest and are actively being pursued. So, controlling the exposed crystal plane of Fe2O3 might also be an effective strategy to further improve the electrochemical performance of Fe2O3 as anode materials for lithium-ion batteries.

In this article, we successfully synthesized two kinds of Fe2O3 with exposed different crystal plane, including nanorods with (001) and (010) plane and nanosheets with the (001) plane. Interestingly, when used as anode materials in lithium-ion batteries, Fe2O3 nanosheets exhibit better cycle performance and rate capabilities than that of Fe2O3 nanorods. To be specific, the discharge capacity of Fe2O3 nanosheets could stabilize at 865 mAhg−1 at the rate of 0.2C (1 C = 1000 mAg−1) and 570 mAhg−1 at the rate of 1.2 C over 80 cycles, which increased by 90% and 79% compared with 456 mAhg−1 and 318 mAhg−1 of Fe2O3 nanorodes. Herein, the outstanding electrochemical performance of Fe2O3 nanosheets can be attributed to the highly exposed (001) planes. Crystal structure have revealed that the (001) plane possesses larger packing density than that of (010) plane, and the crystal effect is the crucial reason for the differences of electrochemical performance30. On the other hand, potentiostatic intermittent titration (PITT) results show that Fe2O3 nanosheets have higher diffusion coefficient of Li+ (DLi) and are more favorable for the diffusion of lithium ion.

To the best of our knowledge, we, for the first time, combined electrochemical experiment and crystal structure analysis to elucidate exposed crystal plane-electrochemical properties relationship of Fe2O3 as anode for rechargeable lithium ion batteries. Our results indicate the superior electrochemical performances of Fe2O3 nanosheets can be attributed to (1) the larger packing density of Fe3+ and O2− of (010) plane and (2) the higher diffusion coefficient of Li+ (DLi) of Fe2O3 nanosheets during discharge-charge process. Furthermore, our results provide a idea which we can design and synthesize electrode materials with high performances according to their structure features in future.

Results

In Fig. 1, the indexed X-Ray Diffraction (XRD) patterns of Fe2O3 samples show that the diffraction peaks match well with the standard PDF card (JCPDS no. 86–2368), indicating the purity of the products and the two kinds of Fe2O3 belong to the same space group. The exposed facets of nanosheets and nanorodes have been determined by high resolution transmission electron microscopy (HRTEM) characterization in Fig. 2c–h. The clear lattice spacing and fast Fourier transform selected-area electron diffraction (FFT-SAED) patterns indicate that Fe2O3 nanosheets and nanorods are single crystalline. Figure 2c shows the TEM image of a Fe2O3 nanosheet, and the corresponding SAED pattern is shown in Fig. 2d. It can be clearly seen that the exposed crystal facet is perpendicular to the (3000), (0300) and (0030) facets of Fe2O3 nanosheets, and the interlayer spacings of 0.252 nm inserted in Fig. 1c correspond to the (110) plane of the Fe2O3. Thus it can be concluded that the exposed facets of the nanosheets are (001). Figure 2e and f show the similar interlayer spacings and SAED pattern compared with the Fe2O3 nanosheets, which indicate that one of the exposed facets of the Fe2O3 nanorods are (001). Figure 2g shows the TEM image of another facet of Fe2O3 nanorodes and the corresponding SAED pattern is shown in Fig. 2h. The SAED pattern in Fig. 2h shows that the exposed crystal facet is perpendicular to the (300), (006) and (202) facets of the Fe2O3, and the interlayer spacings of 0.210 nm inserted in Fig. 2g correspond to the (202) plane of the Fe2O3.

Figure 1
figure 1

Powder X-ray diffraction patterns of Fe2O3 nanorods and nanosheets in a 2θ range of 10–80°.

Full size image
Figure 2
figure 2

TEM images of Fe2O3 nanosheets (a) and Fe2O3 nanorods (b); (c,d) The TEM image of a Fe2O3 nanosheet, inset shows the lattice fringes and the corresponding SAED pattern; (e,f) The TEM image of Fe2O3 nanorode with {001} plane, inset shows the lattice fringes and the corresponding SAED pattern; (g,h) The TEM image of Fe2O3 nanorode with {010} plane, inset shows the lattice fringes and the corresponding SAED pattern.

Full size image

So another exposed facets of the nanorodes are (010). The structural models of Fe2O3 nanorod is displayed in Fig. 3c, the exposed (001) and (010) crystal facets can be clearly shown and the models of Fe2O3 nanosheet is showed in Fig. 3f. Figure 3a,d show the SEM images of Fe2O3 nanorodes and nanosheets, respectively. It can be seen that the average length of Fe2O3 nanorods is about 500 nm, the width and thickness is about 50 and 15 nm, respectively. The average diameter and thickness of Fe2O3 nanosheets is about 200 nm and 15 nm, respectively. The Fe2O3 nanosheets and nanorods of nanosize can reduce the diffusion length of Li+ ions and increase reactivity of the material, which are very favorable for excellent electrochemical performances.

Figure 3
figure 3

The SEM images and crystal size distribution histogram of (a,b) Fe2O3 nanorods and (d,e) Fe2O3 nanosheets; Structural models of (c) Fe2O3 nanorods and (f) Fe2O3 nanosheets.

Full size image

Subsequently, the comparison galvanostatic discharge capacities of Fe2O3 nanorods and nanosheets in a potential range of 0.1–3.0 V (vs Li/Li+) at a rate of 0.2 C were comprehensively investigated and illustrated in Fig. 4a–c. It can be seen from Fig. 4a that Fe2O3 nanorod and Fe2O3 nanosheet electrodes deliverer approximately a discharge capacity of 1135 mAhg−1 and 1210 mAhg−1 in the first cycle, respectively. After that the discharge capacity decreases rapidly, and the reason can be ascribed to the change of structure during the initial charge-discharge process. From the beginning of the second cycle, two kinds of Fe2O3 electrodes exhibit good cycle stability until to the 20 cycles. Surprisingly, after 20 cycles, the discharge capacity of the Fe2O3 nanorods monotonically decline, while the discharge capacity of Fe2O3 nanosheets slightly increase. Similar phenomenon also has been found in the case of CoO and Co3O4 as well as other Fe2O3 reports in the literature, though a clear understanding has not been obtained31,32,33. As shown in Fig. 4a, the discharge capacity of the Fe2O3 nanosheets maintains at 865 mAhg−1 with a capacity retention of 95.3% after 80 cycles, in contrast, the discharge capacity of Fe2O3 nanorods maintains at 456 mAh/g with a capacity retention of 50.7% after 80 cycles.

Figure 4: N2 absorption curves of Fe2O3 nanosheets and Fe2O3 nanorods.
figure 4

(a) Comparison of cycling performance of the two kinds of Fe2O3 at the rate of 0.2 C; The charge/discharge curves of (b) Fe2O3 nanorods and (c) Fe2O3 nanosheets in the 1st, 2nd, 20th, 60th and 80th cycles at the rate of 0.2 C, respectively. (d) The rate performances of Fe2O3 nanorods and Fe2O3 nanosheets. (e) Plots of the specific discharge capacity vs. cycle number for the Fe2O3 nanostructure electrode at the rate of 1.2 C; The discharge curves of (f) Fe2O3 nanorods and (g) Fe2O3 nanosheets in the 1st, 20th, 60th, 100th and 150th cycles at the rate of 1.2 C, respectively.

Full size image

Figure 4d shows the rate performance of Fe2O3 nanorods and nanosheets. Specifically, the discharge capacity of Fe2O3 nanorods at 0.2, 0.4, 0.8, 1.2, 1.6, 2.0 and 2.4 C are 896, 763, 627, 544, 478, 430 and 385 mAhg−1, respectively. The corresponding values for the Fe2O3 nanosheets were 966, 832, 734, 667, 628, 586 and 550 mAhg−1, respectively. By comparing the discharge capacity of the two samples, Fe2O3 nanosheets display higher capacity than Fe2O3 nanorods at various charge−discharge rates from 0.2 to 2.4 C. Meanwhile, as the growth of charge–discharge current density, the gap between the discharge capacities of the Fe2O3 nanorods and nanosheets samples became larger. For instance, the discharge capacity of Fe2O3 nanosheets increase by 8% compared with that of Fe2O3 nanorods at 0.2 C, while the discharge capacity increase by 43% at the rate of 2.4 C. In addition, it should be noted when the rate was returned back to the 0.2 C, Fe2O3 nanosheets still show higher discharge capacity than that of Fe2O3 nanorods. At the recovery rate of 0.2 C, both Fe2O3 nanosheets and nanorods display lower discharge capacity compared with the initial capacity at 0.2 C. The phenomenon is due to the destruction of crystal structure of Fe2O3 during discharge-charge cycle process.

In order to research the cycle stability under high current density, the Fe2O3 nanorods and nanosheets are tested at the rate of 1.2 C, as shown in Fig. 4e–g. Obviously, the electrode of Fe2O3 nanosheets shows much higher discharge capacity than that of Fe2O3 nanorods at high rate. Especially, the discharge capacity of Fe2O3 nanosheets can reach 719 mAhg−1 after 150 cycles. This value is 71% higher than that of Fe2O3 nanorods, which only shows 419 mAhg−1 after 150 cycles.

SEM and TEM images of Fe2O3 nanosheets and nanorods samples after extensive cycling are shown in Fig. 5. It can be clearly seen in Fig. 5a,c that Fe2O3 nanosheets keep relatively complete sheet structure after extensive charge-discharge cycling. Similar to nanosheets, Fe2O3 nanorods also show well virgulate shape which can be seen in Fig. 5b,d. In addition, there is no significant change of the Fe2O3 particle size after charge-discharge cycling.

Figure 5
figure 5

SEM and TEM images of (a,c) Fe2O3 nanosheets and (b,d) Fe2O3 nanorods after 30 cycles at 0.2 C.

Full size image

It is reported that the Brunauer–Emmett–Teller (BET) surface areas of electrode materials play a improtant role on the electrochemical performance of lithium ions batteries34. Our nitrogen-sorption analysis reveals that the BET specific surface areas of Fe2O3 nanorods and Fe2O3 nanosheets were 26.81 and 18.25 m2/g, respectively (Fig. 6). The BET specific surface areas of Fe2O3 nanorods is larger than that of Fe2O3 nanosheets, whereas, the Fe2O3 nanosheets exhibit better electrochemical properties compared with Fe2O3 nanorods. So it can be concluded that the effect of specific surface areas of electrodes on the difference of electrochemical properties between Fe2O3 nanosheets and nanorodes can be overlooked.

Figure 6
figure 6

N2 absorption curves of Fe2O3 nanosheets and Fe2O3 nanorods.

Full size image

Evidently, the electrochemical performances of lithium ion batteries are related to the intrinsic crystal structure35. So the crystal structure of Fe2O3 is analyzed. For Fe2O3 samples, the (001) plane has been found possessing the larger packing density, in which Fe3+ and O2− ions pack layer by layer. Specifically, the packing densities of the Fe3+ and O2− are 9.11 nm−2 and 13.8 nm−2, respectively. In contrast, the packing densities of the (010) facets for ions are 2.89 nm−2 and 5.78 nm−2. Due to the high atomic density, more Fe3+ ions participate in the reaction, and lead to a high specific capacity28. The detailed crystal structure of Fe2O3 have been displayed in Fig. 7. Meanwhile, it can be seen from the model of Fe2O3 nanosheets and nanorodes in Fig. 7, that the proportion of (001) plane in nanosheet is almost 100%, while in nanorodes is about 23%. And the mainly exposed crystal plane is (010) facet in nanorodes, in which the proportion of (010) plane is about 77%. The results indicate that the Fe2O3 samples which exposed more (001) plane show a superior electrochemical capability.

Figure 7
figure 7

The surface atomic configurations in (a) the (001) plane and schematic hematite structure projected along {001}, (b) the (010) plane and schematic hematite structure projected along {010}. The Fe2O3 Crystallographic Information File (CIF) was taken from the NIST/FIZ FindIt Inorganic Crystal Structure Database.

Full size image

Figure 8 shows Nyquist plots of the two kinds of Fe2O3 electrode measured at the open circuit potential and an equivalent circuit proposed to fit the spectra. As can be seen from Table 1, the charge transfer resistances (Rct) for Fe2O3 nanosheets (53 Ω) is much smaller than that obtained from the Fe2O3 nanorodes (179 Ω) electrode. The electrochemical impedance spectroscopy (EIS) data indicates that Fe2O3 nanosheets possesses smaller lithium ion migration resistance and is more conducive to the rapid migration of lithium ions.

Figure 8
figure 8

(a) Nyquist plots for Fe2O3 nanorods and Fe2O3 nanosheets; (b) The corresponding equivalent circuit.

Full size image
Table 1 The fitted solution resistances (Rs) and charge transfer resistances (Rct) for the Nyquist plots of Fe2O3 nanorods and nanosheets.
Full size table

For the sake of confirming DLi in electrode materials, PITT measurement was performed. Figure 9a–c display the PITT results of Fe2O3 samples before discharge-charge cycle. It can be seen the DLi of Fe2O3 nanosheets is about one time higher than that of Fe2O3 nanorods. Figure 9d–f display the PITT results of Fe2O3 samples after a circle of discharge-charge cycle at the current density of 200 mA/g. It is obviously that DLi of Fe2O3 nanosheets are higher than that of Fe2O3 nanorods. For instance, the DLi average value of Fe2O3 nanosheets is 2.2 × 10−10 cm2 s−1 which increased by 15.7%, compared to 1.9 × 10−10 cm2 s−1 of Fe2O3 nanorods. The improved kinetic parameters DLi, indicate that Fe2O3 nanosheets possess higher lithium diffusion coefficient. For this reason, Fe2O3 nanosheets show better rate capability, as shown in Fig. 4d. Figure 4d show that Fe2O3 nanosheets with (001) planes possess higher discharge capacity not only at the low rate of 0.2 C but also at the high rate of 1.2 C. Additional, the Fe2O3 nanosheets with (001) planes exhibit better cycle stability and rate ability. Generally, the Fe2O3 nanosheets with (001) plane exhibit better electrochemical properties than that of the Fe2O3 nanorods with (010) and (001) planes.

Figure 9
figure 9

PITT curves of (a) Fe2O3 nanosheets and (b) Fe2O3 nanorods and (c) diffusion coefficients of Li+ (DLi) before discharge-charge cycle; Current-time transient plots of PITT for a potential step of 1.0–0.8 V after one cycle, (d) I vs. t and (e) ln(I) vs. t; (f) DLi of Fe2O3 nanorods and Fe2O3 nanosheets during discharge processe.

Full size image

In conclusion, we successfully synthesized two kinds of morphology of single crystal Fe2O3 with exposed different crystal plane, including nanorods with (001) and (010) plane and nanosheets with the (001) plane. Fe2O3 nanosheets exhibit better cycle performance and rate capabilities than that of Fe2O3 nanorods. The reasons can be attributed to that (1) the larger packing density of Fe3+ and O2− of (010) plane and (2) the higher diffusion coefficient of Li+ (DLi) of Fe2O3 nanosheets during discharge-charge process. Our studies indicate that the crystal structure has a very important influence on the electrochemical performances, which may be helpful for developing high performance lithium ion batteries.

Methods

Materials synthesis

The Fe2O3 nanorods were synthesized by FeOOH nanorods template. To prepare FeOOH nanorods precursors, 1.9 mmol of FeCl3·6H2O was put into 15 ml of deionized water to form a homogeneous solution, then added 15 ml of 0.8 M NaOH solution under stirring, quickly. After stirring for 10 min, the total solution was transferred into a 50 ml Teflon-lined stainless steel autoclave, sealed and heated at 180 °C for 4 h. The result product was collected by centrifugation, washed with deionized water and ethanol, then was dried at 80 °C and calcined at 250 °C for 2 h. The Fe2O3 nanosheets were synthesized based on the previous work28. 5.1 mmol of FeCl3·6H2O was dissolved in 35 ml of anhydrous ethanol, and 38.4 mmol of CH3COONa·3H2O was rapidly added into the solution with stirring. After about 5 min of stirring, all of the reactants were transferred into a 50 ml Teflon-lined stainless steel autoclave, sealed and heated at 200 °C for 22 h. The result product was collected by centrifuge, washed with deionized water and ethanol. Then dried at 80 °C and calcined at 250 °C for 2 h.

Characterization

XRD measurements were performed on a Persee XD2 X-ray diffractometer with Cu-Kα radiation (λ = 1.5418). The size and morphology of all of the samples were measured with a S-4800 HITACHI scanning electron microscope (SEM) and a JEM-2100 transmission electron microscope (TEM). The specific surface areas of the powders were collected by a Gemini V Brunauer-Emmett-Teller (BET).

Electrochemical Measurement

For electrochemical studies, working electrode was fabricated with mixing active material, acetylene black and polyvinylidene fluoride (PVDF) with weight ratio 2:1:1 using N-methylpyrrolidone (NMP) as solvent. The slurry was fully ground and pasted onto copper foil, and then the loaded copper foil was dried in a vacuum oven at 120 °C for 12 h. Lithium metal, celgard 2300 membrane and 1 M LiPF6 solution in DMC/EC (1: 1 in volume) were used as counter electrode, separator and electrolyte respectively to assemble coin cells in an Ar-filled glove box. The galvanostatic charge/discharge performance of the cells were tested on a battery testing system (BTS-5 V 5 mA, Neware) with the voltage between 0.1 and 3.0 V at the current density of 200, 400, 800, 1200, 1600, 2000, 2400 mA/g. The electrochemical spectroscopy (EIS) was tested with an (PGSTAT302N, Metrohm-Autolab) instrument using an amplitude of 5 mV and a frequency range from 100 KHz to 0.1 Hz. The PITT tests were also performed on the same instrument with EIS.

Additional Information

How to cite this article: Chen, M. et al. The Effect of Crystal Face of Fe2O3 on the Electrochemical Performance for Lithium-ion Batteries. Sci. Rep. 6, 29381; doi: 10.1038/srep29381 (2016).