The Effect of Crystal Face of Fe2O3 on the Electrochemical Performance for Lithium-ion Batteries

Fe2O3 nanorods exposing (001) and (010) plane as well as Fe2O3 nanosheets exposing (001) plane have been successfully synthesized. Fe2O3 nanosheets exhibit better cycle performance and rate capabilities than that of Fe2O3 nanorods. The discharge capacity of Fe2O3 nanosheets can stabilize at 865 mAh/g at the rate of 0.2 C (1C = 1000 mA/g) and 570 mAh/g at the rate of 1.2 C after 80 cycles, which increased by 90% and 79% compared with 456 mAh/g and 318 mAh/g of Fe2O3 nanorods. In comparison with (010) plane, the (001) plane of hematite possesses larger packing density of Fe3+ and O2−, which is responsible for the superior electrochemical performances of Fe2O3 nanosheets than that of Fe2O3 nanorods. In addition, potentiostatic intermittent titration (PITT) results show the diffusion coefficients of Li+ (DLi) of Fe2O3 nanosheets is higher than that of Fe2O3 nanorods. The higher diffusion coefficients of Li+ is favorable for the excellent lithium-storage capabilities and rate capability of Fe2O3 nanosheets. Inspired by our results, we can design and synthesize Fe2O3 or other electrodes with high performances according to their structure features in future.

In this article, we successfully synthesized two kinds of Fe 2 O 3 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, Fe 2 O 3 nanosheets exhibit better cycle performance and rate capabilities than that of Fe 2 O 3 nanorods. To be specific, the discharge capacity of Fe 2 O 3 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 Fe 2 O 3 nanorodes. Herein, the outstanding electrochemical performance of Fe 2 O 3 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 performance 30 . On the other hand, potentiostatic intermittent titration (PITT) results show that Fe 2 O 3 nanosheets have higher diffusion coefficient of Li + (D Li ) 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 Fe 2 O 3 as anode for rechargeable lithium ion batteries. Our results indicate the superior electrochemical performances of Fe 2 O 3 nanosheets can be attributed to (1) the larger packing density of Fe 3+ and O 2− of (010) plane and (2) the higher diffusion coefficient of Li + (D Li ) of Fe 2 O 3 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 Fe 2 O 3 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 Fe 2 O 3 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 Fe 2 O 3 nanosheets and nanorods are single crystalline. Figure 2c shows the TEM image of a Fe 2 O 3 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 Fe 2 O 3 nanosheets, and the interlayer spacings of 0.252 nm inserted in Fig. 1c correspond to the (110) plane of the Fe 2 O 3 . 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 Fe 2 O 3 nanosheets, which indicate that one of the exposed facets of the Fe 2 O 3 nanorods are (001). Figure 2g shows the TEM image of another facet of Fe 2 O 3 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 Fe 2 O 3 , and the interlayer spacings of 0.210 nm inserted in Fig. 2g correspond to the (202) plane of the Fe 2 O 3 . So another exposed facets of the nanorodes are (010). The structural models of Fe 2 O 3 nanorod is displayed in Fig. 3c, the exposed (001) and (010) crystal facets can be clearly shown and the models of Fe 2 O 3 nanosheet is showed in Fig. 3f. Figure 3a   been obtained [31][32][33] . As shown in Fig. 4a, the discharge capacity of the Fe 2 O 3 nanosheets maintains at 865 mAhg −1 with a capacity retention of 95.3% after 80 cycles, in contrast, the discharge capacity of Fe 2 O 3 nanorods maintains at 456 mAh/g with a capacity retention of 50.7% after 80 cycles. Figure 4d shows the rate performance of Fe 2 O 3 nanorods and nanosheets. Specifically, the discharge capacity of respectively. By comparing the discharge capacity of the two samples, Fe 2 O 3 nanosheets display higher capacity than Fe 2 O 3 nanorods at various charge− discharge rates from 0.2 to 2.4 C. Meanwhile, as the growth of chargedischarge current density, the gap between the discharge capacities of the Fe 2 O 3 nanorods and nanosheets samples became larger. For instance, the discharge capacity of Fe 2 O 3 nanosheets increase by 8% compared with that of Fe 2 O 3 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, Fe 2 O 3 nanosheets still show higher discharge capacity than that of Fe 2 O 3 nanorods. At the recovery rate of 0.2 C, both Fe 2 O 3 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 Fe 2 O 3 during discharge-charge cycle process.
In order to research the cycle stability under high current density, the Fe 2 O 3 nanorods and nanosheets are tested at the rate of 1.2 C, as shown in Fig. 4e-g. Obviously, the electrode of Fe 2 O 3 nanosheets shows much higher discharge capacity than that of Fe 2 O 3 nanorods at high rate. Especially, the discharge capacity of Fe 2 O 3 nanosheets can reach 719 mAhg −1 after 150 cycles. This value is 71% higher than that of Fe 2 O 3 nanorods, which only shows 419 mAhg −1 after 150 cycles.
SEM and TEM images of Fe 2 O 3 nanosheets and nanorods samples after extensive cycling are shown in Fig. 5. It can be clearly seen in Fig. 5a,c that Fe 2 O 3 nanosheets keep relatively complete sheet structure after extensive charge-discharge cycling. Similar to nanosheets, Fe 2 O 3 nanorods also show well virgulate shape which can be seen in Fig. 5b,d. In addition, there is no significant change of the Fe 2 O 3 particle size after charge-discharge cycling.
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 batteries 34 . Our nitrogen-sorption analysis reveals that the BET specific surface areas of Fe 2 O 3 nanorods and Fe 2 O 3 nanosheets were 26.81 and 18.25 m 2 /g, respectively (Fig. 6). The BET specific surface areas of Fe 2 O 3 nanorods is larger than that of Fe 2 O 3 nanosheets, whereas, the Fe 2 O 3 nanosheets exhibit better electrochemical properties compared with Fe 2 O 3 nanorods. So it can be concluded that the effect of specific surface areas of electrodes on the difference of electrochemical properties between Fe 2 O 3 nanosheets and nanorodes can be overlooked.
Evidently, the electrochemical performances of lithium ion batteries are related to the intrinsic crystal structure 35 . So the crystal structure of Fe 2 O 3 is analyzed. For Fe 2 O 3 samples, the (001) plane has been found possessing the larger packing density, in which Fe 3+ and O 2− ions pack layer by layer. Specifically, the packing densities of the Fe 3+ and O 2− 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 Fe 3+ ions participate in the reaction, and lead to a high specific capacity 28 . The detailed crystal structure of Fe 2 O 3 have been displayed in Fig. 7. Meanwhile, it can be seen from the model of Fe 2 O 3 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 Fe 2 O 3 samples which exposed more (001) plane show a superior electrochemical capability. Figure 8 shows Nyquist plots of the two kinds of Fe 2 O 3 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 (R ct ) for Fe 2 O 3 nanosheets (53 Ω) is much smaller than that obtained from the Fe 2 O 3 nanorodes (179 Ω) electrode. The electrochemical impedance spectroscopy (EIS) data indicates that Fe 2 O 3 nanosheets possesses smaller lithium ion migration resistance and is more conducive to the rapid migration of lithium ions.
For the sake of confirming D Li in electrode materials, PITT measurement was performed.   Fig. 4d. Figure 4d show that Fe 2 O 3 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 Fe 2 O 3 nanosheets with (001) planes exhibit better cycle stability and rate ability. Generally, the Fe 2 O 3 nanosheets with (001) plane exhibit better electrochemical properties than that of the Fe 2 O 3 nanorods with (010) and (001) planes.
In conclusion, we successfully synthesized two kinds of morphology of single crystal Fe 2 O 3 with exposed different crystal plane, including nanorods with (001) and (010) plane and nanosheets with the (001) plane. Fe 2 O 3 nanosheets exhibit better cycle performance and rate capabilities than that of Fe 2 O 3 nanorods. The reasons can be attributed to that (1) the larger packing density of Fe 3+ and O 2− of (010) plane and (2) the higher diffusion coefficient of Li + (D Li ) of Fe 2 O 3 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.     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 LiPF 6 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.