carbonaceous matrix with long-life performance for high-rate lithium ion batteries

Accepted Manuscript Three-dimensional Fe3O4/carbonaceous matrix with long-life performance for highrate lithium ion batteries Yang Yu, Wei Wang, Zhihong Jing, Xiaofeng Zheng, Xiaoling Geng, Jingfa Li PII:

S0925-8388(16)32268-X

DOI:

10.1016/j.jallcom.2016.07.237

Reference:

JALCOM 38406

To appear in:

Journal of Alloys and Compounds

Received Date: 28 June 2016 Revised Date:

19 July 2016

Accepted Date: 21 July 2016

Please cite this article as: Y. Yu, W. Wang, Z. Jing, X. Zheng, X. Geng, J. Li, Three-dimensional Fe3O4/ carbonaceous matrix with long-life performance for high-rate lithium ion batteries, Journal of Alloys and Compounds (2016), doi: 10.1016/j.jallcom.2016.07.237. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Three-dimensional Fe3O4/carbonaceous matrix with long-life performance for high-rate lithium ion batteries

RI PT

Yang Yu a,b,∗, Wei Wang a, Zhihong Jing a, Xiaofeng Zheng a, Xiaoling Geng a, Jingfa Li c

School of Chemistry and Chemical Engineering, Qufu Normal University,

SC

a

M AN U

273165, Shandong, PR China

Qufu

b Technology Research and Development Center, Shandong Sacred Sun Power Sources CO., LTD, Qufu 273100, Shandong, PR China

c School of Physics and Optoelectronic Engineering, Nanjing University of

Abstract

TE D

Information Science and Technology, Nanjing 210044, Jiangsu, PR China

EP

Three-dimensional (3D) Fe3O4/carbonaceous matrix composites are successfully

AC C

assembled via a facile hydrothermal treatment of precursor followed by heat-decomposition process. Cellulose acetate (CA), a kind of polymer, as carbon source, can form a flexible and uniform 3D carbon matrix after carbonation at 500 ◦C, in addition of the highly degree uniformity with active Fe3O4 particles. When applied as an anode for lithium-ion batteries, these 3D Fe3O4/C composites exhibit superior electrochemical performance after prolonged cycling at high rates (~790 mA h g-1 at * Corresponding author: Phone: 86-0537-445-6301. E-mail address: [email protected] (Y. Yu).

ACCEPTED MANUSCRIPT 2C after 1000 cycles) and good rate capability at changing rates of 0.1C–3C–0.1C. The excellent performance observed from the electrochemical measurement can be attributed to the 3D architecture and the micro-carbonaceous matrix, both of which

RI PT

play important buffering role in the balance of volume variation and provide an excellent conductive matrix for high rate charge-discharge process.

Lithium ion battery, Carbon matrix, Anode, Iron oxides, Long life

SC

Keywords:

1.

M AN U

performance

Introduction

Nowadays, rechargeable lithium ion batteries (LIBs) are widely used as green power source for portable electronics, electric vehicles and hybrid electric vehicles

TE D

owing to their stable output voltage, high energy density and environmental benignity [1-3]. With the ever-growing demands for highly improved lithium storage, the

EP

traditional graphite anode has been unable to meet the demand because of its inherent theoretical limitation (372 mA h g-1) and the lower operating voltage [4-6]. Over the

AC C

past decade, various novel transition metal oxides with high specific capacities (~600–1000 mAh g-1) have been developed as promising anode candidates [7-10], Among these materials, iron oxide (Fe3O4) has attracted considerable attention for its high theoretical capacity (~924 mA h g-1, Fe3O4 + 8Li+ + 8e− ↔ 3Fe + 4Li2O), as well as eco-friendly, low processing cost, and relatively high electronic conductivity [11-13]. The practical implementation of Fe3O4 based anodes, however, is mainly frustrated by great volume variation during the cycling process, which may result in

ACCEPTED MANUSCRIPT pulverization of electrodes and rapid capacity decay. Much effort has been devoted to resolve these issues from designing nanostructures to constructing hybrid composites [14,15]. One effective strategy is to design three-dimensional (3D) micro-nano

RI PT

architecture by combining Fe3O4 active particles with micro-carbonaceous conducting matrix, including thin carbon layer [16], porous carbon matrix [17,18], sheet-like grapheme [19], etc. According to reported literatures, this 3D architecture can

SC

effectively maintain the structural integrity and improve the electric conductivity

M AN U

[19-22]. The micro-carbonaceous materials act as buffer layers to mitigate volume change, and as conductive matrix for electron transfer. Meanwhile, it also provides active sites for Li+ and prevents aggregation of the actives [23-25]. However, most of the hybrid composites need a complicated preparation process with multi-steps under

TE D

demanding conditions, resulting in a hardly realized mass production. It would be urgently needed to synthesize 3D Fe3O4/C structure using refined and effective methods, so in order to fully utilize the merits of the micro-nano architecture and

EP

conductive carbon.

AC C

In this work, we successfully synthesized a 3D Fe3O4/C composite by a facile hydrothermal treatment

of precursor followed by heat-decomposition and

carbonization process. In the obtained composite, the dispersed Fe3O4 particles are embedded uniformly in the 3D micro-carbonaceous matrix. The carbon content is 5.11 wt %, with this weight the carbon matrix as a conductive network can accommodate volume variation, as well as no much effect on the tap density of the composite. It is noted that cellulose acetate (CA), a kind of polymer with a molecular

ACCEPTED MANUSCRIPT weight of ~20000, is used as carbon source. This polymer can form a flexible and uniform 3D structure after carbonation at 500℃, in addition of the highly degree conformity with active particles. Thus, this synthetic process is applicable for large

RI PT

scale production of carbon encapsulated materials. When used as anode in LIBs, these 3D Fe3O4/C composites exhibit greatly improved cycling stability with a high capacity of ~790 mA h g-1 at 2C after 1000 cycles and satisfactory performance at rates

of

0.1C–3C–0.1C

within 50

cycles.

It

shows

SC

changing

that

the

M AN U

Fe3O4/carbonaceous matrix composites could be used as a desired anode material for higher energy storage and longer cycle lifetimes.

2.

Experiments

TE D

2.1. Sample preparation 2.1.1. Preparation of precursor

In a typical synthesis process, FeCl2·4H2O (5 mmol), CO(NH2)2 (0.3 g) and

EP

NH4HCO3 (0.2 g) were dissolved in distilled water (40 mL) simultaneously and

AC C

stirred for 30 min, then the solution was transferred into a 50 mL Teflon-lined autoclave and heated to 120 oC for 12 h. A brown precursor was obtained after magnetic separation and washed several times with ethanol and deionized water to remove impurities.

2.1.2. Preparation of Fe3O4/carbonaceous matrix composite The obtained precursor powders were milled with CA (precursor/CA = 2:1, w/w) in a ball mill (400 r s-1, 10 h). Then, the mixtures were annealed in a quartz tube with a

ACCEPTED MANUSCRIPT slow ramping rate of 2 ◦C min-1 to 500 ◦C for 5 h in Ar atmosphere (Scheme 1). Then, the black Fe3O4/carbonaceous matrix composite powders were obtained. 2.1.3. Preparation of Fe3O4 particles

RI PT

The obtained precursor powders were directly annealed in a quartz tube with a slow ramping rate of 2 ◦C min-1 to 500 ◦C for 5 h in Ar atmosphere. Then, the pure Fe3O4 particles were obtained.

SC

2.2. Characterization

M AN U

The phase identification of the products was accomplished using powder X-ray diffraction (XRD) employing a Philips X’pert X-ray diffractometer with Cu Kα radiation (λ= 1.54178 Å). Raman spectrum was carried out on a JYLABRAM-HR Confocal Laser Micro-Raman spectrometer with 514.5 nm from an argon laser at

TE D

room temperature. The scanning electron microscopy (SEM) images were taken with a JEOL-JSM-6700F field emission scanning electron microscope. The transmission electron microscopy (TEM) images were recorded on a Hitachi Model H-7650

EP

transmission electron microscope, using an electron kinetic energy of 100 kV.

AC C

Thermogravimetric analysis (TGA) was collected with a TGA-2050 (TA Corp.) under the flowing air. A Vario ELIII measured the carbon content. 2.3. Electrochemical measurements The working electrode was made of active materials (65 wt %), acetylene black (25

wt %) and poly (vinylidene difluoride) (PVDF) (10 wt %) slurry coated onto a copper foil substrate and dried at 100 ºC for 12 h in vacuum. Electrochemical measurements were carried out using coin-type 2032 cells. The coin cells were assembled in an

ACCEPTED MANUSCRIPT argon-filled glovebox with lithium foil as the counter electrode, Celgard 2400 as the separator, and a solution of 1.0 M LiPF6 in ethylene carbonate (EC)/diethyl carbonate (DEC) (1:1 by volume) as the electrolyte. The galvanostatic charge-discharge tests

RI PT

were performed on a Land automatic battery tester (Wuhan, China) for a voltage window of 0.01–3.00 V (vs. Li+/Li) at room temperature. Electrochemical impedance spectroscopy (EIS) was carried out on an electrochemical analyzer (CHI660D, Taisite.

Results and discussion

M AN U

3.

SC

China).

The schematic illustration of the preparation of 3D Fe3O4/C composite has been show in Scheme 1. During the ball-milling, the precursor and CA are well mixed

TE D

using little volatile organic solvent. Then, the mixture is placed in an argon atmosphere. As the temperature rise, the precursor gradually decomposes to Fe3O4 particles and CA molecules carbonize to micro-carbon matrix. Thus, the Fe3O4

EP

particles disperse naturally in the matrix and a 3D architecture with an amorphous

AC C

micro-carbonaceous matrix is formed. The crystallographic structures of the precursor and Fe3O4/C composite are identified by XRD. Fig. 1a displays XRD pattern of the precursor. All the peaks could be indexed to FeCO3 (JCPDS Card No.83-1764), except four extra peaks at 2θ = 30.091, 35.443, 56.964 and 62.553, which belong to (220), (311), (511) and (440) peaks of Fe2.886O4 (JCPDS Card No. 86-1342), respectively. Fig. 1b is the typical XRD pattern of the Fe3O4/C composite, which can be perfectly indexed as face-centered cubic structure Fe3O4 (JCPDS Card No.

ACCEPTED MANUSCRIPT 79-0419). No diffraction peak for graphitic carbon is observed, suggesting the amorphous nature of the carbon content. To insure the complete transformation of FeCO3 precursor to Fe3O4 phase, the appropriate decomposition temperature is

RI PT

determined by thermogravimetric analysis (TGA). As shown in Fig. 1c, the precursor decomposed from ~200 ◦C to 450 ◦C, with a weight loss of about 31.97 %, mainly

process may be expressed as:

(1)

M AN U

FeCO3 →Fe3O4 + CO↑ + CO2↑

SC

corresponding to the release of gaseous contents from the sample. The decomposition

On the basis of TGA results, we choose 500 ◦C as treatment temperature to ensure the completely decomposition of the precursor. The Brunauer-Emmett-Teller (BET) gas-sorption measurement of the Fe3O4/C composites was also performed (Fig. 1d).

TE D

The N2 adsorption-desorption isotherm can be categorized as type IV with hysteresis loops [26], and the insets show the Barrett-Joyner-Halenda (BJH) pore-size distribution. The composites have a relatively higher BET surface area of 67.296 m2

EP

g-1 than that of the pure Fe3O4 particles (25.414 m2 g-1, Fig. 1e). The big surface area

AC C

of the Fe3O4/C composite could provide more electrochemical active sites and be helpful for permeation of the electrolyte. From the inset of Fig. 1d, two peaks centered at 3.8 and 5 nm in the pore size distribution may be ascribed to the mesoporous structure of the amorphous carbon formed from carbon source (CA). The size distribution in the range of 3-5 nm is favorable for electrochemical reactions [27,28], because the mesoporous structure can provide the high contact area between composites and the additives (acetylene black and PVDF), and facilitate better

ACCEPTED MANUSCRIPT diffusion of Li+ and electrons. From these viewpoints, the Fe3O4/C composite could be used as electrode materials for long-cycling and high-rate LIBs. To confirm the presence of carbon, the Raman spectrum of the as-obtained Fe3O4/C composite is

RI PT

investigated. As shown in Fig. 1f, the two peaks at 1350 and 1596 cm-1 are the characteristic peaks corresponding to the D-band and G-bond of graphite [29]. D-band is associated with the vibration of carbon atoms with dangling bonds in-plane

SC

terminations of disordered graphite, and G-bond is related to the vibration of

M AN U

sp2-bonded carbon atoms in a 2D hexagonal lattice. Here, the two peaks show the poor crystallinity of carbon. The amorphous carbon has also been confirmed by Elementar vario EL cube, which gives the carbon content of the composite is ~ 5.11 wt % (Table 1).

TE D

The SEM and TEM images of the sample are shown in Fig. 2. As shown in Fig. 2a, the as-prepared precursor contains lots of irregular cubes with a side-length range of 0.5-1 µm, while a small amount of particles (50-100 nm) distributed around the cubes.

EP

The SEM image of the Fe3O4/C composite is shown in Fig. 2b. It can be observed that

AC C

the sintered cube has a porous structure, following some particles fall off the monolith. Under higher magnification (Fig. 2c), the cubes are shown to be composed of well-connected particles with a size of ~50 nm. Further investigation using TEM confirms the amorphous carbon in the composites, as shown in Fig. 2d. It is easily to identify Fe3O4 particles and amorphous carbon according to the difference of light and shade between particles. This characteristic is exactly in accordance with the Scheme 1. The schematic illustration shows that the CA chains anchored on the surface of

ACCEPTED MANUSCRIPT precursor by ball-milling process. With the decomposition of precursor in a gradually heating process, CA chains penetrate into its internal and finally form a micro-carbonaceous matrix, resulting in a 3D Fe3O4/C architecture.

RI PT

The electrochemical properties of samples were investigated by half coins using lithium foil as the counter electrode. To confirm the influence of the 3D micro-carbonaceous matrix to the composite, the rate capability and structure stability

SC

of the Fe3O4/C electrode and Fe3O4 electrode is first evaluated and compared at

M AN U

various discharge-charge rates (0.1C-0.2C-0.5C-1C-2C-2.5C-3C-1C). Fig. 3 shows that the Fe3O4/C sample is superior to the Fe3O4 sample at any current rates with excellent rate capability and relative stable capacity. It exhibits a high specific capacity and stable cycling performance with an increase in the current rate from 0.1C

TE D

to 3C. The relative discharge capacities of the Fe3O4/C electrode at the higher rates are 875 (1C), 707 (2C), 603 (2.5C) and 530 mA h g-1 (3C), which are 94.7%, 76.5%, 65.3% and 57.2% of the theoretical capacity, respectively. In contrast, the Fe3O4

EP

electrode exhibit a poor capacity of ~ 400 mA h g-1 at 3C rate. Note that when the rate

AC C

is back to 0.1C from 3C, the discharge capacity of the composite recovers rapidly. This result illustrates that the structure stability of the composites is well maintained. It is reported that rate capability is determined by the internal resistances, which is the sum of ohmic and charge-transfer resistances [30]. As mentioned above, the larger BET surface area, the excellent dispersion of Fe3O4 particles in the matrix and the high conductivity of the carbon matrix could improve the structure stability and prevent the pulverization and aggregation of the active particles, thus make high

ACCEPTED MANUSCRIPT discharge capacity and superior rate capability of the composites. The discharge capacity and long-term cycling performance of the Fe3O4/C electrodes at 1C and 2C are both evaluated at room temperature. The galvanostatic

RI PT

charge-discharge voltage profiles of the electrodes tested at 1C (1A g-1) and 2C (2A g-1) between 0.01 and 3.00 V are shown in Fig. 4. The electrodes exhibit a lithiation voltage plateau at 0.8 V in the first cycle, and the corresponding delithiation voltage

SC

plateau at ~1.75 V. The enhanced first discharge capacity (1191 mA h g-1) is probably

M AN U

attributed to the further lithium consumption via interfacial reactions at the low voltage and the formation of the SEI film [30,31]. From the following curves for the 200th, 500th, 990th/1060th cycles, however, the discharge voltages take place at a lower voltage with almost no plateau and the capacities are added steadily from 200th to

TE D

990th/1060th cycles, which can be clearly observed from the insets of Fig. 4. The inset of Fig. 4a shows the cycling performance of Fe3O4/C electrode at 1C. It can be observed that the specific capacity decreases rapidly from 1191 (1st cycle) to 403 mA

EP

h g-1 (77th cycle) with a capacity loss rate of 10 mA h g-1 per cycle. Then the capacity

AC C

recovers and rises to 665 mA h g-1 after 500 cycles with a capacity rising rate of 0.62 mA h g-1 per cycle. Compared with the 1C performance, the capacity of 2C rate decreases faster (inset of Fig. 4b), the reversible capacity sharply decreases from 1210 (1st cycle) to 325 mA h g-1 (77th cycle), and its value recovers to 563 mA h g-1 after 500 cycles with a capacity rising rate of 0.56 mA h g-1 per cycle. Overall, the cycling performances (insets of Fig. 4a and b) are similar at 1C and 2C, firstly the discharge capacity fading till 77th cycles, then recover and rise steadily at the following cycles.

ACCEPTED MANUSCRIPT The initial capacity fading is generally attributed to the structure re-organization of the carbon coatings [32]. With the activation of the electrodes, the capacity recovered steadily, after ~1000 cycles, 1C cell rises to 740 mA h g-1, while 2C cell has a higher

RI PT

capacity of 790 mA h g-1. These phenomenon could be explained as: the higher current rate makes less activated materials work, so that the reversible capacity of 2C is lower than that of 1C in the previous process; at the following cycles, the gradual

SC

increasing capacities of 1C and 2C are attributed to the reversible growth of a

M AN U

polymeric gel-like film resulting from activated electrolyte degradation at the higher rate and long repeated cycling process. This result is well-documented in the literatures [33,34] and similar results have been reported for many transition metal oxides [35,36]. It is note that the coulombic efficiencies of the Fe3O4@C composites

TE D

cycled at 1C and 2C (the blue dots in insets of Fig. 4a and b) maintain at ~99% during the charge-discharge process. The high capacities and excellent coulombic efficiency of the electrodes indicate the 3D core-shell Fe3O4/C composites have great potential

EP

as a long-cycle anode material for high-rate lithium storage.

AC C

To further test the excellent cyclic stability of the composite at a much higher rate, a new cell is first cycled at 0.1C for five cycles, and then evaluated at 10 C (10 A g-1) for 1000 cycles. As shown in Fig. 5a, the initial discharge and charge capacities are 1670 and 1134 mA h g-1, respectively, corresponding to a Columbic efficiency of 67.90%. After the initial capacity fading, the Columbic efficiency (the violet dots in Figure 5a) maintains consistently ＞99.0% at following cycles, and the discharge capacity stabilizes and retains at a high capacity of 280 mA h g-1, which is about

ACCEPTED MANUSCRIPT 30.3% of the theoretical capacity. This is mainly because the amorphous micro-carbon plays as a perfect buffering area which can prevent the huge volume expansion and structure collapse of the Fe3O4/C electrodes. Meanwhile, the micro-carbon as a

RI PT

conductive matrix, can make rapid transmission of electrons among the active particles, so reduce the loss of the electrons and improve the Columbic efficiency. To

gain

further

understanding

of

the

micro-carbonaceous

matrix,

the

SC

electrochemistry impedance spectra (EIS) of the Fe3O4/C electrodes and pure Fe3O4

M AN U

particles are performed with a frequency range of 100 kHz to 0.01 Hz (Fig. 5b). Note that Fe3O4 particles are prepared by the same experimental method without CA. The typical Nyquist plots consist of an intercept at high frequency region corresponds to the ohmic resistance (Re), the semicircle in the middle frequency indicates the

TE D

charge-transfer resistance (Rct), and the inclined straight line relates to the Warburg impedance (Zw) [37]. As shown in Figure 5b, the semicircle diameter of the Fe3O4/C electrode is much smaller than that of Fe3O4 electrode, which means Fe3O4/C

EP

composites have lower charge-transfer impedance and faster charge transfer process

AC C

than pure Fe3O4. It indicates that the carbonaceous matrix plays an important role in reducing the Rct of cells during charge-discharge process. The carbon matrix formed from polymer has a flexible, uniform structure, in addition of the highly degree conformity with active particles, thus, this synthetic process is applicable for large scale production of other carbon encapsulated materials. The 3D architecture and carbon matrix have several advantages: improve the intimate connection of active materials and electrolyte and the buffering role of the carbon matrix to prevent

ACCEPTED MANUSCRIPT volume expansion.

4. Conclusions

RI PT

In this work, three-dimensional Fe3O4/C composites are successfully assembled by a facile hydrothermal preparation of precursor followed by heat-decomposition process. These Fe3O4 nanoparticles are uniformly dispersed in the micro-carbon

SC

matrix. The galvanostatic charge-discharge tests give the cycle performance and rate

M AN U

capability of the Fe3O4/C composites. The Fe3O4/C electrodes show a stable capacity retention of ~790 mA h g-1 at 2C after prolonged cycles and satisfactory performance at changing rates of 0.1C–3C–0.1C within 50 cycles. The analysis from the electrochemical measurement exhibits the role of micro-carbon as a buffering area

TE D

and a conductive matrix in the composite. It is note that this carbon matrix formed from polymer has a flexible, uniform structure, thus, this synthetic process is

EP

applicable for large scale production of other carbon encapsulated materials.

AC C

Acknowledgements

This work was supported by the National Natural Science Foundation of China

(21501108

and

21271115),

the

Doctoral Foundation of Shandong Province

(BS2014NJ015 and BS2014CL015), the Scientific Research Staring Foundation of Qufu Normal University (BSQD20130118) and the Jiangsu Provincal Natural Science Foundation (BK20160941).

ACCEPTED MANUSCRIPT References [1] J.E. Lee, S.H. Yu, D.J. Lee, D.C. Lee, S.I. Han, Y.E. Sung, T. Hyeon, Energy Environ. Sci. 5 (2012) 9528-9533.

Zhi, ACS Nano 7 (2013) 1437-1445. [3] J.M. Tarascon, M. Armand, Nature 414 (2001) 359-367.

RI PT

[2] B. Wang, X. Li, X. Zhang, B. Luo, M. Jin, M. Liang, S.A. Dayeh, S.T. Picraux, L.

SC

[4] N.A. Kaskhedikar, J. Maier, Adv. Mater. 21 (2009) 2664-2680.

M AN U

[5] J. Luo, X. Xia, Y. Luo, C. Guan, J. Liu, X. Qi, C.F. Ng, T. Yu, H. Zhang, H.J. Fan, Adv. Energy Mater. 3 (2013) 737-743.

[6] D. Chen, G. Ji, Y. Ma, J.Y. Lee, J. Lu, ACS Appl. Mater. Interfaces 3 (2011) 3078-3083.

TE D

[7] P. Poizot, S. Laruelle, S. Grugeon, L. Dupont, J. M. Tarascon, Nature 407 (2000) 496-499.

EP

[8] Z.Y. Wang, D.Y. Luan, F.Y.C. Boey, X.W. (David) Lou, J. Am. Chem. Soc. 133 (2011) 4738-4741.

AC C

[9] J. Chen, L.N. Xu, W.Y. Li, X.L. Gou, Adv. Mater. 17 (2005) 582-586. [10] H.B. Rong, G.T. Xie, S. Cheng, Z.H. Zhen, Z.Q. Jiang, J.L. Huang, Y. Jiang, B.H. Chen, Z.J. Jiang, J. Alloys Comp. 679 (2016) 231-238. [11] S. Ito, K. Nakaoka, M. Kawamura, K. Ui, K. Fujimoto, N. Koura, J. Power Sources 146 (2005) 319-322. [12] H. Liu, G. X. Wang, J. Z. Wang, D. Wexler, Electrochem. Commun. 10 (2008) 1879-1882.

ACCEPTED MANUSCRIPT [13] T. Muraliganth, A.V. Murugan, A. Manthiram, Chem. Commun. 47 (2009) 7360-7362. [14] F.S. Ke, L. Huang, B. Zhang, G.Z. Wei, L.J. Xue, J.T. Li, S.G. Sun, Electrochim.

RI PT

Acta 78 (2012) 585-591. [15] F.X. Ma, H.B. Wu, C.Y. Xu, L. Zhen, X.W. (David) Lou, Nanoscale 7 (2015) 4411-4414.

SC

[16] C. Lei, F. Han, D. Li, W.C. Li, Q. Sun, X.Q. Zhang, A.H. Lu, Nanoscale 5 (2013)

M AN U

1168-1175.

[17] S. Wu, Z.Y. Wang, C.N. He, N.Q. Zhao, C.S. Shi, E.Z. Liu, J.J. Li, J. Mater. Chem. A 1 (2013) 11011-11018.

[18] D. Li, X.W. Li, S.Y. Wang, Y.X. Zheng, L. Qiao, D.Y. He, ACS Appl. Mater.

TE D

Interfaces 6 (2014) 648-654.

[19] K. Zhu, Y. Zhang, H.L. Qiu, Y. Meng, Y. Gao, X. Meng, Z.M. Gao, G. Chen, Y.J. Wei, J. Alloys Comp. 675 (2016) 399-406.

EP

[20] Y.R. Wang, L. Zhang, Y. L. Wu, Y. J. Zhong, Y. Hu, X.W. (David) Lou, Chem.

AC C

Commun. 51 (2015) 6921-6924. [21] H.J. Yin, S.L. Zhao, J.W. Wan, H.J. Tang, L. Chang, L.C. He, H.J. Zhao, Y. Gao, Z.Y. Tang, Adv. Mater. 25 (2013) 6270-6276. [22] X.L. Fan, J. Shao, X.Z. Xiao, L.X. Chen, X.H. Wang, S.Q. Li and H.W. Ge, J. Mater. Chem. A 2 (2014) 14641-14648. [23] L. Li, A.R.O. Raji, J.M. Tour, Adv. Mater. 25 (2013) 6298-6302. [24] A.K. Rai, L.T. Anh, J. Gim, V. Mathew, J. Kang, B.J. Paul, N.K. Singh, J.J. Song,

ACCEPTED MANUSCRIPT J. Kim, J. Power Sources 244 (2013) 435-441. [25] L. Li, Z.P. Guo, A.J. Du, H.K. Liu, J. Mater. Chem. 22 (2012) 3600-3605. [26] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol,

[27] P. Simon, Y. Gogotsi, Nat. Mater. 7 (2008) 845-854.

RI PT

T. Siemieniewska, Pure Appl. Chem. 57 (1985) 603-619.

Qian, J. Mater. Chem. A 1 (2013) 15292-15299.

SC

[28] J.F. Li, J.Z. Wang, D. Wexler, D.Q. Shi, J.W. Liang, H.K. Liu, S.L. Xiong, Y.T.

M AN U

[29] A.C. Ferrari, J. Robertson, Phys. Rev. B 61 (2000) 14095-14107.

[30] J. Kim, M.K. Chung, B.H. Ka, J.H. Ku, S. Park, J. Ryu, S.M. Oh, J. Electrochem. Soc. 157 (2010) A412-A417.

[31] Z.P. Zeng, H.L. Zhao, J. Wang, P.P. Lv, T.H. Zhang, Q. Xia, J. Power Sources 248

TE D

(2014) 15-21.

[32] J.S. Gnanaraj, M.D. Levi, E. Levi, G. Salitra, D. Aurbach, J.E. Fischer, A. Claye, J. Electrochem. Soc. 148 (2001) A525-A536.

EP

[33] S. Grugeon, S. Laruelle, L. Dupont, J.M. Tarascon, Solid State Sci. 5 (2003) 895.

AC C

[34] J.S. Do, C.H. Weng, J. Power Sources 146 (2005) 482-486. [35] S.L. Chou, J.Z. Wang, D. Wexler, K. Konstantinov, C. Zhong, H.K. Liu, S.X. Dou, J. Mater. Chem. 20 (2010) 2092-2098. [36] L.L. Xu, B.G. Lu, Electrochim. Acta 210 (2016) 456-461. [37] C.H. Chen, J. Liu, K. Amine, J. Power Sources 96 (2001) 321-328.

ACCEPTED MANUSCRIPT Caption for figures Scheme 1. Schematic illustration of the formation of 3D Fe3O4/C composite. Table 1. The amount of N, C, H elements in Fe3O4/C composites (Elementar vario EL

RI PT

cube). Fig. 1. (a) XRD pattern of the precursor, (b) XRD pattern of the Fe3O4/C composites, (c) TGA and DrTGA curves for the precursor in flowing nitrogen, N2

SC

adsorption/desorption isotherms and BJH pore size distribution plots of (d) the

M AN U

Fe3O4/C composite and (e) Fe3O4 particles, (f) Raman spectrum of the 3D Fe3O4/C composite.

Fig. 2. Low and high magnification SEM images of (a) precursor and (b and c) Fe3O4/C composites. (d) TEM images of the Fe3O4/C composites.

TE D

Fig. 3. Cycling performance of Fe3O4/C electrode and Fe3O4 electrode measured at various discharge-charge rates.

Fig. 4. Charge-Discharge profiles and cycle performances (inset) of Fe3O4/C

EP

electrodes at (a) 1C and (b) 2C.

AC C

Fig. 5. (a) Cycle performance and Coulomb efficiency of Fe3O4/C electrode at 10C. (b) Nyquist plots of the Fe3O4 and Fe3O4/C electrodes.

RI PT

ACCEPTED MANUSCRIPT

Weight（mg） 2.046

N (%)

C (%)

H (%)

0.09

5.12

0.64

0.10

5.10

0.51

5.11

0.57

TE D

2.978

M AN U

SC

Scheme 1. Schematic illustration of the formation of 3D Fe3O4/C composite.

Average Value

0.09

AC C

EP

Table 1. The amount of N, C, H elements in Fe3O4/C composites (Elementar vario EL cube)

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 1. (a) XRD pattern of the precursor, (b) XRD pattern of the Fe3O4/C composites, (c) TGA and DrTGA curves for the precursor in flowing nitrogen, N2

adsorption/desorption isotherms and BJH pore size distribution plots of (d) the Fe3O4/C composite and (e) Fe3O4 particles, (f) Raman spectrum of the 3D Fe3O4/C composite.

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 2. Low and high magnification SEM images of (a) precursor and (b and c)

AC C

EP

TE D

M AN U

Fe3O4/C composites. (d) TEM images of the Fe3O4/C composites.

Fig. 3. Cycling performance of Fe3O4/C electrode and Fe3O4 electrode measured at various discharge-charge rates.

RI PT

ACCEPTED MANUSCRIPT

SC

Fig. 4. Charge-Discharge profiles and cycle performances (inset) of Fe3O4/C

AC C

EP

TE D

M AN U

electrodes at (a) 1C and (b) 2C.

Fig. 5. (a) Cycle performance and Coulomb efficiency of Fe3O4/C electrode at 10C. (b) Nyquist plots of the Fe3O4 and Fe3O4/C electrodes.

ACCEPTED MANUSCRIPT ·Fe3O4 particles are embedded in the three-dimensional carbonaceous matrix. ·Cellulose acetate can form a flexible and uniform 3D carbon architecture after

carbonation.

RI PT

·The carbon matrix in the Fe3O4/C enhanced the conductivity and the cyclic stability.

AC C

EP

TE D

M AN U

SC

·Fe3O4/C composite maintained a reversible capacity of 790 mA h g-1 at 2000 mA g-1 after 1000 cycles.