Electrochemical characteristics of pyrrhotine as anode material for lithium-ion batteries

Journal of Alloys and Compounds 661 (2016) 483e489

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom

Electrochemical characteristics of pyrrhotine as anode material for lithium-ion batteries Xiaodong Zheng a, b, * a b

College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao, 266042, China Department of Chemical Engineering, Binzhou University, Binzhou, 256603, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 October 2015 Received in revised form 26 November 2015 Accepted 27 November 2015 Available online 2 December 2015

Micron-sized pyrrhotine particles with a standard hexagonal structure were prepared by simple precipitation and subsequent heat treatment and were investigated as anode material for lithium-ion batteries. The as-prepared Fe7S8 electrode delivered a highly reversible capacity of 604.1 mAh/g with a voltage range of 2.5e0.05 V. X-ray diffraction, scanning electronic microscopy and X-ray photoelectron spectroscopy were employed to characterize the reaction products at different stages, and a possible two-step reaction mechanism was proposed. The electrochemical test results demonstrated that the discharge/charge voltage range had a remarkable influence on the capacity retention and coulombic efficiency, which could be associated with the decomposability of lithiation products and volume change. Additionally, the dissolution of Li2Sx (2 < x < 8) in electrolyte was found to cause severe capacity loss as well. Thus a proper coating layer and cycling voltage range were essential for this kind of electrode material to achieve practical application in lithium-ion batteries. © 2015 Elsevier B.V. All rights reserved.

Keywords: Lithium-ion batteries Anode Pyrrhotine

1. Introduction

FeS þ 2Liþ þ 2e 4 Li2S þ Fe0

Recently, various kinds of alternative materials for lithium-ion battery anodes have been extensively studied because of the low specific capacity and poor high-rate performance of graphite which is seriously falling behind the demand for some applications, such as electric vehicles and hybrid electric vehicles. As new-generation anode materials, several transitional metal compounds (i.e., MnXm, where M ¼ transitional metal; X ¼ O, S, F or P) based on the conversion reaction mechanism have gained considerable attention because of their much higher theoretical capacity and energy densities than graphite electrode. Among them, iron monosulphide has recently been considered to be a promising anode material for lithium-ion batteries and has attracted much attention because of its excellent electrochemical attributes, low environmental effect, and natural abundance [1e3]. However like other transitional metal compounds, iron monosulphide (denoted as FeS) exhibits poor capacity retention and low coulombic efficiency. The chemical reaction process proceeds as follows:

According to previous studies [4e6], the key to realize the reversibility of the conversion reaction is the nanosized transition metal particles, Fe0, which has been verified to be electrochemically active in decomposing the Li2S matrix. However, given the drastic volume expansion after lithiation and poor electron conductivity of Li2S, the reverse reaction can be inhibited, thereby resulting in rapid capacity degradation [3,7,8]. Furthermore, the lithiation process of metal sulfides tends to generate the polysulfides Li2Sx (2 < x < 8), which can easily dissolve into organic electrolyte and result in poor capacity retention [1,9e11]. To improve the electrochemical performance of the FeS material, many effective methods have been proposed, most of which focus on the synthesis of nanosized particles with different structures [12e14] or nanocomposites with high-conductivity materials [15e17]. Nanosized materials can better accommodate strains caused by volume change and reduce the lithium-ion diffusion path. The coating layer of conductive carbon or metals can enhance the conductivity of the electrode, thereby improving coulombic efficiency and reversible capacity. As a main natural form of iron monosulphide, pyrrhotine powder has a broad size distribution and is often referred to as Fe7S8 because of the deficient part of Fe2þ. Nanosized pyrrhotine/

* College of Chemical Engineering, Qingdao University of Science and Technology, Qingdao, 266042, China. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.jallcom.2015.11.210 0925-8388/© 2015 Elsevier B.V. All rights reserved.

(1)

X. Zheng / Journal of Alloys and Compounds 661 (2016) 483e489

carbon composite has been reported to exhibit almost the same electrochemical properties as FeS; the reversible reaction mainly occurs in a voltage range of 2.5e0.9 V [18]. However, studies on other similar anode materials (e.g., hematite) have identified particle size and morphology as important factors that influence their electrochemical performance, even the electrochemical reaction process [19e23]. To the best of our knowledge, no report exists on micron-sized pyrrhotine particles that account for most of the natural powder. For this purpose, we synthesized micron-scale pyrrhotine particles without any modification and investigated the electrochemical properties as anode materials for lithium-ion batteries. The results reveal that micron-sized pyrrhotine particles exhibit a reaction process that differs from the reported reaction process of nano or submicron-sized particles, and can deliver an initial reversible capacity as high as the theoretical capacity when discharged/charged in the voltage range of 2.5e0.05 V. 2. Experimental

(206) Intensity(a.u.)

484

(200) (203)

25

30

35

(220)

40

45

50

55

/degree Fig. 1. X-ray diffraction pattern of as-prepared pyrrhotine sample.

2.1. Preparation of sample In the experiment, an aqueous solution of Fe(NH4)2$(SO4)2$6H2O (0.44 mol/l) was added dropwise in equivoluminal Na2S (0.50 mol/l) solution while stirring at room temperature. The resulting precipitate was repeatedly rinsed in distilled water to remove inorganic ions and was rinsed with ethanol for drying. After drying under vacuum at 80  C, the product was heat-treated at 600  C for 3 h under a constant flow of argon to obtain the final Fe7S8 powder with a hexagonal lattice structure. 2.2. Characterization The crystalline structure of the electrodes was measured by Xray diffraction (XRD) (X'pert PRO, Panalytical) at a scanning rate of 10 /min. The morphology of the samples was observed with a scanning electron microscope (SEM) (S4800, Hitachi). The surface functional groups were measured with X-ray photoelectron spectroscopy (XPS) using a Quantum 2000 Scanning ESCA Microprobe with monochromatic Al Ka radiation (1486.6 eV). 2.3. Electrochemical characterization The electrochemical performance was measured with CR2025 coin-type cells. A slurry was first prepared by ball-milling a mixture of active materials (80 wt.%), conductive carbon black (Super-P, 10 wt.%) and polyvinylidene fluoride (PVDF, 10 wt.%) in an adequate amount of N-methyl-2-pyrrolidone (NMP) for 2 h. The resulting slurry was used as coating for a copper foil with an area of 1 cm2. The coated copper foil was dried at 100  C under vacuum for at least 10 h to obtain an electrode for electrochemical measurement. The cells were assembled in an argon-filled glove box (Etelux 2000, China), where both moisture and oxygen levels were kept at less than 1 ppm. The electrolyte used was LiPF6 (1 mol/L) in ethylene carbonate and dimethyl carbonate (EC-DMC, 1:1 v/v), and the lithium foil was used as the counter electrode. The charge/ discharge cycles were performed using a battery test instrument (Neware®, China). 3. Results and discussion The XRD pattern of the as-prepared Fe7S8 sample is presented in Fig. 1. The reflection of the sample has sharp diffraction peaks at 29.9 , 33.7, 43.6 and 53.0 , corresponding to the (200), (203), (206), and (220) faces, respectively, which are in good agreement with the standard hexagonal structure of Fe7S8 (JCPDS reference

code 00-025-0411). Additionally, no other obvious peaks are found in the pattern indicating the high purity of the product. Fig. 2(a) shows the first discharge/charge voltage profiles of Fe7S8 electrode at a 0.1C rate (1C ¼ 660 mA/g) in the potential range of 2.5e0.9 V, which is the region of the as-known lithiation/delithiation reaction of iron monosulphide. The initial dischargecharge curve exhibits a typical discharge plateau at 1.32 V and charge plateau at 1.78 V, which are consistent with the behavior of FeS or nanosized Fe7S8 [12e14,16,18]. Nevertheless, the initial discharge and charge capacities are only 416.7 and 361.2 mAh/g, respectively, which are far below the theoretical capacity (660 mAh/g for Fe7S8). When the discharge cut-off voltage extends to 0.05 V (Fig. 2(b)), another discharge plateau at approximately 0.7 V is found. Similar anomalous electrochemical activities have also been reported in the literature on FeS [12,13], and these activities are generally attributed to the formation of a solid electrolyte interlayer (SEI) film because of the extremely low corresponding reversible capacity. Obviously, the discharge plateau found in our test results cannot be simply ascribed to the formation of the SEI film because in the voltage range the discharge and charge capacities remarkably increases to 793.2 and 604.1 mAh/g, respectively. This phenomenon indicates that the lithiation process can involve a two-step mechanism to actualize the formation of Fe0 and Li2S, which is quite different from the reported single-step reaction process of FeS or nanosized Fe7S8. The reason for this phenomenon may be due to the formation of the intermediate, Li2FeS2 [24,25]. During the lithiation process at 1.3 V, the conversion reaction first occurs on the surface of the particles. Given the relatively longer diffusion path of Liþ, the pristine pyrrhotine inside the particles cannot undergo full lithiation, thereby resulting in the formation of Li2FeS2. The proposed reaction mechanism is shown below: Fe7S8 þ 8Liþ þ 8e / 4Li2FeS2 þ 3Fe0, (2.5e0.9 V)

(2)

4Li2FeS2 þ 8Liþ þ 8e / 8Li2S þ 4Fe0, (0.9e0.05 V)

(3)

During the subsequent de-lithiation process, the charge curve in Fig. 2(a) is much more steady with a potential plateau observed at around 1.8 V than that in Fig. 2(b), and the latter exhibits a steep slope before a potential plateau appears, indicating a different delithiation process. For a clear understanding of the reaction, the differential capacities versus potential (dQ/dV versus V) curves are provided (Fig. 2(c) and (d), respectively), which are considered to provide information similar to that of cyclic voltammograms. In

X. Zheng / Journal of Alloys and Compounds 661 (2016) 483e489

485

Fig. 2. First-cycle discharge-charge curves and differential discharge-charge versus voltage plots of pyrrhotine electrode at 0.1C, (a) and (c) between 2.5 and 0.9 V, (b) and (d) between 2.5 and 0.05 V.

Fig. 2(c), a pair of nearly reversible redox peaks at 1.28 and 1.78 V can be clearly observed corresponding to the voltage plateaus of Fig. 2(a). In Fig. 2(d), apart from similar redox peaks at 1.26 and 1.81 V, a pair of cathodic and anodic peaks at 0.71 and 1.63 V is found as well, though the anodic peak is very broad and less evident. This indicates that the charge process also includes a twostep reaction between 2.5 and 0.05 V. The second-step reaction corresponds to a poor de-lithiation process that may be related to the insulativity of Li2S, which also influences the reversibility of the first-step reaction (the gap between the cathodic and anodic peaks in Fig. 2(c) and (d) is 0.5 and 0.55 V, respectively). Fig. 3 shows the typical SEM images of the Fe7S8 electrodes at different potential. The pristine Fe7S8 particles are observed to have a smooth surface (Fig. 3(a)). After lithiation at 0.9 V (Fig. 3(b)), a distinct change in the Fe7S8 electrode occurs. The volume of the particles severely expands, thereby resulting in the agglomeration of the reaction product. When it is charged back to 2.5 V from 0.9 V, the layer of the lithiation product almost disappears (Fig. 3(c)), thereby indicating the good reversibility of the first-step reaction. Fig. 3(d) presents the SEM image of the electrode discharged to 0.05 V, which shows that a polymeric gel-like layer covers the whole surface of the electrode, which is a widely reported phenomenon for many conversion electrodes [26e28]. When charged back to 2.5 V, most of the lithiation product can be discomposed reversibly (Fig. 3(e)). By comparing Fig. 3(c) and (e), the residue of the reaction product in the first cycle between 2.5 and 0.05 V is significantly greater than those in the voltage range of 2.5e0.9 V, indicating that the reversibility of the overall reaction is much worse than that of the first-step reaction.

To further characterize the lithiation products at different voltages, we also performed an ex-situ XRD analysis of Fe7S8 electrode discharged to 0.9 and 0.05 V. In Fig. 4, the sharp diffraction peaks of the pristine Fe7S8 in Fig. 1 completely disappear and the lithiation products both show an amorphous feature apart from a broad diffraction peak at 44.6 , which matches well with the Fe0 nanoparticle. Speculations on this XRD pattern are difficult to make in the presence of Li2FeS2 or Li2S. However, a difference exists between the two diffraction peaks of the Fe0 nanoparticle. According to Scherrer's equation, the calculated average Fe0 nanoparticle size of a half and fully lithiated state is 2.1 and 2.6 nm, respectively. Therefore, although they are inconclusive, the XRD results tend to confirm the assumed electrochemical process in accordance with Eqs. (2) and (3). A further reaction in the range of 0.9e0.05 V increases the size growth of the Fe0 nanoparticles, which may also be one of the factors leading to the poor reversibility of Eq. (3). As a powerful method providing information on the element valence state, XPS was also employed to analyze the surface of the electrodes. The XPS spectra of Fe(2p3/2) and S(2p3/2) of the pristine sample and the electrodes lithiated at 0.9 and 0.05 V are shown in Fig. 5, and the corresponding fitted results are presented in Table 1. From the spectroscopy of Fe(2p3/2) (Fig. 5(a)), the Fe ion species on the surface of the pristine sample mainly consists of Fe(II)eS and Fe(III)eO, which can be attributed to the surface oxidation of the pyrrhotine sample. After lithiation at 0.9 V, the peak of Fe(III)eO disappears and is replaced by that of Fe(II)eO, and the peak for Fe metal is found as well. By comparing the relative peak areas (Table 1), the presence of Fe metal may be a result of the conversion of Fe(II)eS, and the peaks corresponding to Fe(II)eS and Fe(II)eO

486

X. Zheng / Journal of Alloys and Compounds 661 (2016) 483e489

Fig. 3. SEM images of (a) pristine sample and pyrrhotine electrodes discharged/charged in different voltage range, (b) and (c) between 0.9 and 2.5 V, (d) and (e) between 0.05 and 2.5 V.

can be ascribed to the intermediate compounds, Li2FeS2 and Li2Fe2O3, respectively [29e31]. With further lithiation, a bulk SEI layer gradually forms and covers the surface of the electrode. As a consequence, the spectroscopy of the Fe(2p3/2) cannot been observed at 0.05 V. Fig. 5 (b) presents the S(2p3/2) spectra and shows that, on the surface of the pristine sample, the peak corre2 sponding to S0 (or/and Sn) and S2 2 has been found aside from S , which may be formed according to [32,33]: Fe7S8 þ O2 þ H2O / FeOOH þ S0 (or Sn)

(4)

Fe7S8 þ S0 (or Sn) / FeS2

(5)

At 0.9 V, although the peak area is greatly reduced (i.e., from 59.5% to 24.1%), the peak of S2 does not disappear completely. Distinguishing S2 from the Li2S produced by lithiation and the unreacted Fe7S8 is difficult. However, the peak area of S2 2 is found to increase and exceed that of S2, suggesting the existence of an intermediate Li2FeS2 compound and supporting the ex-situ XRD results discussed above that the conversion reaction is not completed after lithiation at 0.9 V. In addition, the peak of S2 n is observed, which can be attributed to Li2Sx, indicating the occurrence of a side reaction. Fig. 6 provides the discharge/charge profiles in the two voltage ranges for five cycles at a 0.1C rate (1C ¼ 660 mA/g). The figure shows that, in the limited voltage range (2.5e0.9 V), the electrode

X. Zheng / Journal of Alloys and Compounds 661 (2016) 483e489

487

Table 1 Binding energies (BE), peak full width at half maximum (FWHM) and peak areas for Fe(2p3/2) and S(2p3/2). XPS peaks

Species

Fe(2p3/2)

FWHM

Area (%) Pristine

0.9 V

0.05 V

2þ

Fe eS Fe2þeS(satellite) Fe3þeO Fe2þeO Fe0

710.0 713.8 711.6 709.3 706.8

2.0 2.0 2.0 2.0 2.0

60.2 4.6 35.2 e e

16.7 5.2 e 37.2 39.8

e e e e

S2 S2 2 S(0) S2 n

161.5 162.5 163.6 164.2

1.2 1.2 1.2 1.2

59.5 32.5 9.0 e

24.1 38.8 17.7 19.2

e e e e

S(2p3/2)

activity of Fe0 nanoparticles owing to their larger size. Additionally, in both Fig. 6(a) and (b), the first discharge plateau is much lower than other discharge plateaus, which is similar to the reported FeS electrode, and can be ascribed to the transformation from the crystalline to amorphous phase [2,12,16,18]. Moreover, the change in the two discharge plateaus in Fig. 6(b) further confirms that the conversion process is a two-step reaction. To investigate the cyclic and rate performances of the Fe7S8 electrode, the cells are tested for extended cycles with different chargeedischarge current densities, and the results are shown in Fig. 7. For the first ten cycles (0.1/0.1C and 0.5/0.1C for 5 cycles,

Fig. 4. Ex-situ XRD patterns of pyrrhotine electrodes discharged to 0.9 V and 0.05 V.

exhibits better cycling performance (i.e., after 5 cycles, the capacity retention of Fe7S8 electrodes in Fig. 6(a) and (b) is 74.29% and 67.16%, respectively). The worse cycle stability in the range of 2.5e0.05 V may be a result of the comprehensive effects involving more severe volume expansion and the aggravated electrochemical

(a)

BE (eV)

F e ( II) S

(b)

F e ( II I) O

S

2-

F e ( II) S ( s a t e l lit e )

S2

2-

S (0 )

Pristine Pristine F e (II)-O

(s a te llite )

Fe

0.9V lithiated

S (0 ) Sn

S2

2-

2-

S

2-

0.9V lithiated

0.05V lithiated 725

0

Intensity(a.u.)

Intensity(a.u.)

F e (II)-S F e (II)-S

720

0.05V lithiated 715

710

Binding Energy (eV)

705

165

164

163

162

Binding Energy (eV)

Fig. 5. XPS spectra of (a) Fe(2p3/2) and (b) S(2p3/2) of pristine, 0.9 V lithiated and 0.05 V lithiated electrodes.

161

160

488

X. Zheng / Journal of Alloys and Compounds 661 (2016) 483e489

(a)

2.5

1000

1st

5th

Capacity (mAh/g)

Voltage (V)

800 2.0

1.5

1.0

Discharge

0

50

100

150

200

250

300

350

400

0.1/0.1C

600

0.5/0.1C

400

0 0

5

10

Cycle Number

3.0

5th

(b)

0.1/0.1C

450

Capacity (mAh/g)

2.5

0.1/0.5C

200

1st

5th

Disharge Capacity (2.5-0.05V) Charge Capacity (2.5-0.05V) Discharge Capacity (2.5-0.9V) Charge Capacity (2.5-0.9V)

(a)

Charge

1st

15

20

130

Charge

(b) 120

Coulombic Efficiency /%

Voltage (V)

2.0

1.5

1.0

0.5

Discharge 0.0

5th 0

100

200

300

400

500

1st 600

700

110

100

90

80

2.5-0.9 V 2.5-0.05 V

70 800

Capacity (mAh/g) Fig. 6. The discharge/charge profiles of as-prepared pyrrhotine electrodes at 0.1C rate for 5 cycles in the voltage range of (a) 0.9e2.5 V and (b) 0.05e2.5 V.

respectively) the specific capacities gradually decease and the coulombic efficiency increases with increasing cycle numbers. However, when the discharge/charge current density changes from 0.5/0.1C to 0.1/0.5C (i.e., the 11th cycle), the irreversible capacities significantly increase (the coulombic efficiency drops to 80.3% and 88.0%, respectively). When the discharge/charge current density changes back to 0.1/0.1C (i.e., the 16th cycle), both coulombic efficiencies are much higher than 100% (i.e., increase to 121.7% and 110.1%, respectively). This characteristic means that the reversible capacity of the Fe7S8 electrode is mainly determined by its delithiation process. Under the condition of high charge current densities, the lithiated lithium-ions cannot be completely de-lithiated and therefore becomes “dead lithium”, but can deliver at lower current densities. This phenomenon depends on the extent of lithiation; the deeper the discharging voltage range, the more obvious the phenomenon, which may be related to the decomposition difficulty of lithiation products at different phases. In addition, Fig. 7 shows that after a dozen cycles the capacity retention of the electrode tested between 2.5 and 0.9 V severely decreases, whereas the electrode tested in the range of 2.5e0.05 V progressively exhibits a more stable cycle life and higher coulombic efficiency. This result is perhaps associated with the SEI films, which can effectively reduce the contact area between particles and electrolyte, thereby preventing the dissolution of Li2Sx in electrolyte. To verify the effect of SEI films on the cycling performance, the Fe7S8 electrode was pre-treated by cycling between 2.5 and 0.05 V for the first three times to form stable SEI film and was then tested between 2.5 and 0.9 V. Fig. 8 compares the chargeedischarge capacities to the cycle numbers for Fe7S8 electrode with and without

60 0

2

4

6

8

10

12

14

16

18

20

Cycle Number Fig. 7. (a) The discharge/charge capacity vs. cycle number plots and (b) the coulombic efficiency vs. cycle numbers plots of pyrrhotine electrodes at different current rates.

SEI film. The electrode with SEI film obviously presents a relatively better cyclic stability; the reversible capacity reaches 110 mAh/g after 50 cycles. Compared with the sample without SEI film, this improvement in capacity retention is significant, which indicates that a coating treatment is necessary for Fe7S8 electrode to improve its electrochemical performance. Therefore, as previously mentioned, the improvement of the electrochemical performance of micron-sized Fe7S8 electrode materials depends on the reversibility of the conversion reaction and the inhibition of the dissolution of Li2Sx. Surface coating with highconductivity materials (e.g., carbon, conductive organic polymer, and conductive metals, etc.) can be considered as an effective method, which can serve a significant function in decreasing the loss of Li2Sx, improving electronic conductivity and depressing volume expansion. Based on this, limiting the cycling voltage range, by which the volume change and the size of nanosized Fe0 can be effectively controlled, is expected to be another effective means to enhance the reversibility of the reaction, thereby leading to better cycling stability and the coulombic efficiency. This proposed strategy may be suitable not only for micron-sized pyrrhotine but also for other similar monosulphide electrode materials.

4. Conclusions In summary, micron-sized pyrrhotine particles with a hexagonal structure were prepared by simple precipitation and subsequent heat treatment, and their electrochemical properties as anode

X. Zheng / Journal of Alloys and Compounds 661 (2016) 483e489

References

800

Discharge (without pre-treatment) Charge Dishcarge (with pre-treatment) Charge

700

Capacity (mAh/g)

489

600 500 400 300 200 100 0 0

10

20

30

40

50

Cycle Number Fig. 8. Comparison of cycle performance of pyrrhotine electrodes with and without pre-treatment tested at 0.1C rate between 2.5 and 0.9 V.

materials for lithium-ion batteries were investigated as well. The results reveal that the as-prepared Fe7S8 sample exhibits a two-step reaction process to realize the conversion reaction, which is different from the reported process of nano or submicron-sized particles. It can reach a high specific capacity of 604.1 mAh/g when tested in the voltage range of 2.5e0.05 V. Under different discharge/charge voltage ranges, the Fe7S8 electrodes exhibit different electrochemical performance, and the reversibility of the reaction process is associated with the lithiation extent of the sample. Additionally, the dissolution of the by-product, Li2Sx (2 < x < 8), in electrolyte has been verified to have a demonstrable effect on the cycling stability, and the SEI film formed at deepdischarge voltage can inhibit the loss of the lithium-ions, thereby contributing to improving the capacity retention. Therefore, for these kinds of electrode materials as Fe7S8, appropriate surface modification and limitation of cycling voltage range are essential for practical application in lithium-ion batteries.

Acknowledgments This work is financially supported by the Provincial Natural Science Foundation of Shandong Province (No. ZR2014EMP012) and the Project of Science and Technology Development Planning of Binzhou City (No. 2014ZC0216).

[1] X. Rui, H. Tan, Q. Yan, Nanoscale 6 (2014) 9889e9924. [2] Y. Kim, J.B. Goodenough, J. Phys. Chem. C 112 (2008) 15060e15064. [3] B.C. Kim, K. Takada, N. Ohta, Y. Seino, L. Zhang, H. Wada, T. Sasaki, Solid State Ionics 176 (2005) 2383e2387. [4] E. Strauss, D. Golodnitsky, E. Peled, Electrochim. Acta 45 (2000) 1519e1525. [5] K. Takada, Y. Kitami, T. Inada, A. Kajiyama, M. Kouguchi, S. Kondo, M. Watanabe, M. Tabuchi, J. Electrochem. Soc. 148 (2001) A1085eA1090. [6] T.A. Yersak, H.A. Macpherson, S.C. Kim, V.-D. Le, C.S. Kang, S.-B. Son, Y.-H. Kim, J.E. Trevey, K.H. Oh, C. Stoldt, S.-H. Lee, Adv. Energy Mater. 3 (2013) 120e127. [7] P.G. Bruce, S.A. Freunberger, L.J. Hardwick, J.M. Tarason, Nat. Mater. 11 (2012) 19e29. [8] Y. Yang, M.T. McDowell, A. Jackson, J. Cha, S. Hong, Y. Cui, Nano Lett. 10 (2010) 1486e1491. [9] Y. Kim, J.B. Goodenough, J. Phys. Chem. C 112 (2008) 15060e15064. [10] J. Cabana, L. Monconduit, D. Larcher, M.R. Palacin, Adv. Mater. 22 (2010) E170eE192. [11] D.T. Tran, S.S. Zhang, J. Mater. Chem. A 3 (2015) 12240e12246. [12] C. Xing, D. Zhang, K. Cao, S. Zhao, X. Wang, H. Qin, J. Liu, Y. Jiang, L. Meng, J. Mater. Chem. A 3 (2015) 8742e8749. [13] C. Xu, Y. Zheng, X. Rui, N. Xiao, J. Zhu, W. Zhang, J. Chen, W. Liu, H. Tan, H.H. Hng, Q. Yan, ACS Nano 6 (2012) 4713e4721. [14] C. Zhu, Y. Wen, P.A. van Aken, J. Maier, Y. Yu, Adv. Funct. Mater. 25 (2015) 2335e2342. [15] S.H. Kim, Y.J. Choi, D.H. Kim, S.H. Jung, K.W. Kim, H.J. Ahn, Surf. Rev. Lett. 15 (2008) 35e40. [16] C. Dong, X. Zheng, B. Huang, M. Lu, Appl. Surf. Sci. 265 (2013) 114e119. [17] B. Wu, H.-H. Song, J.-S. Zhou, X.H. Chen, Chem. Commun. 47 (2011) 8653e8655. [18] K. Zhang, T. Zhang, J. Liang, Y. Zhu, N. Lin, Y. Qian, RSC Adv. 5 (2015) 14828e14831. [19] M.V. Reddy, T. Yu, C.-H. Sow, Z.X. Shen, C.T. Lim, G.V. Subba Rao, B.V.R. Chowdari, Adv. Funct. Mater. 17 (2007) 2792e2799. [20] F. Han, D. Li, W.-C. Li, C. Lei, Q. Sun, A.-H. Lu, Adv. Funct. Mater. 23 (2013) 1692e1700. [21] P. Zhang, Z.P. Guo, H.K. Liu, Electrochim. Acta 55 (2010) 8521e8526. [22] S.-L. Chou, J.-Z. Wang, D. Wexler, K. Konstantinov, C. Zhong, H.-K. Liu, S.X. Dou, J. Mater. Chem. 20 (2010) 2092e2098. [23] X. Zheng, J. Li, Ionics 20 (2014) 1651e1663. [24] F. Liao, J. Swiatowska, V. Maurice, A. Seyeux, L.H. Klein, Appl. Surf. Sci. 283 (2013) 888e899. [25] J. Liu, Y. Wen, Y. Wang, P.A. van Aken, J. Maier, Y. Yu, Adv. Mater. 26 (2014) 6025e6030. [26] R. Malini, U. Uma, T. Sheela, M. Ganesan, N.G. Renganathan, Ionics 15 (2009) 301e307. [27] B.P. Jia, L. Gao, Cryst. Growth Des. 8 (2008) 1372e1376. [28] R. Dedryvere, S. Laruelle, S. Grugeon, P. Poizot, D. Gonbeau, J.M. Tarascon, Chem. Mater. 16 (2004) 1056e1061. [29] J. Chen, L. Xu, W. Li, X. Guo, Adv. Mater. 17 (2005) 582e586. [30] D. Larcher, C. Masquelier, D. Bonnin, Y. Chabre, V. Masson, J.B. Leriche, J.M. Tarascon, J. Electrochem. Soc. 150 (2003) A133eA139. [31] D. Larcher, D. Bonnin, I. Rivals, L. Personnaz, J.M. Tarascon, J. Electrochem. Soc. 150 (2003) A1643eA1650. [32] L.Y. Mikhlin, A.V. Kuklinskiy, N.I. Pavlenko, V.A. Varnek, I.P. Asanov, A.V. Okotrub, G.E. Selyutin, L.A. Solovyev, Geochim. Cosmochim. Acta 66 (2002) 4057e4067. [33] S. Boursiquot, M. Mullet, M. Abdelmoula, J.M. Genin, J.J. Ehrhardt, Phys. Chem. Miner. 28 (2001) 600e611.