Synthesis and electrochemical performance of bud-like FeS2 microspheres as anode materials for rechargeable lithium batteries

Synthesis and electrochemical performance of bud-like FeS2 microspheres as anode materials for rechargeable lithium batteries

Materials Science and Engineering B 178 (2013) 483–488 Contents lists available at SciVerse ScienceDirect Materials Science and Engineering B journa...

971KB Sizes 1 Downloads 43 Views

Materials Science and Engineering B 178 (2013) 483–488

Contents lists available at SciVerse ScienceDirect

Materials Science and Engineering B journal homepage: www.elsevier.com/locate/mseb

Synthesis and electrochemical performance of bud-like FeS2 microspheres as anode materials for rechargeable lithium batteries Dong Zhang a,∗ , Guojian Wu a , Jiayuan Xiang c , Jun Jin a , Yubin Cai a , Guoliang Li b a b c

China National Product Quality Supervision Inspection Center of Builder’s Finish Hardware Material, Hangzhou 310019, China Architectural Design and Research Institute of Zhejiang Province, Hangzhou 310006, China Narada Power Source Co., Ltd, Hangzhou 311305, China

a r t i c l e

i n f o

Article history: Received 28 September 2012 Received in revised form 29 December 2012 Accepted 29 January 2013 Available online 20 February 2013 Keywords: Pyrite Bud-like microsphere Anode material Lithium-ion battery

a b s t r a c t Bud-like FeS2 powder was synthesized by a solvothermal method with the help of polyvinylpyrrolidone (PVP). The bud-like FeS2 microshperes with the diameters of 2.0–3.0 ␮m were consisted of the submicroflakes with 0.5–1␮m in width and length, and about 60 nm in thickness. As an anode material for Li-ion batteries, the bud-like FeS2 delivered initial specific discharge capacity of 773 and 749 mAh g−1 , and could sustain 387 and 368 mAh g−1 after 30 cycles at current densities of 45 and 89 mA g−1 , respectively, much higher than the solid one obtained without PVP. The bud-like FeS2 microshperes also showed large diffusion coefficient of Li-ions (DLi +) calculated by Galvanostatic intermittent titration (GITT). The improved electrochemical performance of bud-like FeS2 was due to the unique structure which provides large contact area between the FeS2 microspheres and electrolyte, decreased polarization and large DLi +, leading to enhanced electrode reaction kinetics. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Post transition metal dichalcogenides (FeS2 , CoS2 and NiS2 ) with pyrite structure have received special interest over the past decade because of their unique properties and potential applications in various technological fields [1–3]. Among those inorganic compounds, pyrite (FeS2 ) is an attractive anode material for lithium ion batteries associated with high theoretic capacity (890 mAh g−1 ), low environmental impact and affordable cost [4,5]. It is widely accepted that the morphology have great influence on the electrochemical performance of electrode material for lithium ion batteries [6–14]. Hence, in the past few years, many efforts have been devoted to synthesize FeS2 with various morphologies [15–18]. For instance, nanoflake-built pyrite FeS2 microspheres were synthesized using N,N-dimethytformamide and ethylene glycol as solvent and only first discharge capacity was investigated [15]. Recently, FeS2 microcubes and micro-octahedra were prepared by hydrothermal method but the electrochemical property was not mentioned [16]. Cubic FeS2 crystallites have been synthesized via a single-source approach under hydrothermal conditions and the first discharge capacity is 756 mAh g−1 with cutoff voltage of 1.0 V at a current density of 0.2 mA cm−2 [17]. Gao and co-workers established a novel self-decomposition single-source

∗ Corresponding author. Tel.: +86 571 88032738; fax: +86 571 88037538. E-mail addresses: zhangdong [email protected], [email protected] (D. Zhang). 0921-5107/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mseb.2013.01.024

precursor route for FeS2 2D nanowebs [18]. 1-D single crystalline pyrite FeS2 nanostructures have been synthesized using ethylenediamine (en, NH2 CH2 CH2 NH2 ) as solvent but the electrochemical property is not investigated [19]. However, as far as we know, there is little literature on electrochemical performances of FeS2 synthesized by solvothermal and hydrothermal method. We report the first synthesis of bud-like microspheres of FeS2 by PVP surfactant-assisted solvothermal processing. The morphology modification mechanism under the effect of PVP is discussed, and the enhanced electrochemical performances of FeS2 powders prepared with PVP are investigated. 2. Experimental The bud-like FeS2 microshperes were synthesized by solvothermal method. In a typical synthesis, 13.9 g of FeSO4 ·7H2 O and 3.8 g of NH2 CSNH2 were dissolved in 25 mL of deionized water and 25 mL of ethanol under magnetic stirring for 1 h. 1.2 g of S powder and 2 g of PVP were added to the resulting solution under vigorous magnetic stirring at room temperature for 1 h. Then this mixture was sealed in a Teflon-lined stainless steel autoclave (50% filled), maintained at 180 ◦ C for 24 h, and cooled to room temperature naturally. After that, the black solid product was collected by centrifugation, washed with absolute alcohol, CS2 and deionized water several times and finally dried at 100 ◦ C for 10 h in vacuum. The sample prepared without PVP was also obtained by the same method for comparison.

(333)

with PVP

(222) (023) (321)

(220)

(211)

(210)

Intensity (a.u.)

(111)

(311)

D. Zhang et al. / Materials Science and Engineering B 178 (2013) 483–488

(200)

484

without PVP

Pyrite

2 Theta (degree) Fig. 1. XRD patterns of FeS2 powders synthesized without and with PVP.

The as-prepared powders were characterized by X-ray diffraction (XRD, D/max 2550-PC) and field emission scanning electron microscopy (FESEM, FEI SIRION). The specific surface area of the powders were measured following the multipoint Brunauer–Emmett–Teller (BET) procedure from the N2 adsorption–desorption isotherms collected at liquid nitrogen temperature using an AUTOSORB-1-C gas sorption analyzer. Electrochemical performances of FeS2 were investigated in CR2025 coin-type cell. A metallic lithium foil served as the counter electrode. The working electrode consisted of as-prepared material (60 wt.%), acetylene black (25 wt.%) and polyvinylidene fluoride (15 wt.%) binder on aluminum foil. The cells were assembled in an argon-filled glove box using 1 M LiPF6 in ethylene carbonate (EC)-dimethyl carbonate (DMC) (1:1 in volume) as the electrolyte and a polypropylene micro-porous film (Cellgard 2300) as the separator. The galvanostatic charge–discharge tests were conducted

on LAND battery program-control test system (Wuhan, China) between 1.2 and 2.6 V by applying current densities from 45 to 890 mA g−1 at room temperature. Galvanostatic intermittent titration technique (GITT) tests were also conducted on this apparatus at room temperature in the voltage range of 1.2–2.6 V. The cells were discharged at a constant current flux (C/20) for 10 min followed by a relaxation period for 40 min. Cyclic voltammetry (CV) was performed on an electrochemical workstation (CHI 660 C) in the potential window of 1.1–2.7 V (vs. Li/Li + ) at a scan rate of 0.1 mV s−1 . The assembled cells were used for EIS tests, where lithium foils acted as both the counter and reference electrodes. The examined area of the counter-electrode and reference electrode is 2.25 cm2 . EIS measurements were performed over a frequency range of 100 kHz–10 mHz and the total number of points is 84. The measurements were under AC stimulus with 5 mV of amplitude and no applied voltage bias. The impedance data were fitted using the ZsimpWin computer program. All the tests were conducted at room temperature. These measurements have been repeated several times, so we hereby report a representative result over these experiments in the manuscript. 3. Results and discussion Fig. 1 shows XRD patterns of FeS2 powders synthesized with and without PVP. As can be seen, both the diffraction peaks of the powders correspond well with the cubic pyrite FeS2 (PDF card No. 42-1340) and no impurity phase can be detected. The strong and sharp diffraction peaks indicate that the as-obtained products are well-crystallized. Morphologies of FeS2 particles synthesized with and without PVP are compared in Fig. 2. The FeS2 particles obtained without PVP are solid with diameters of 2–3 ␮m (Fig. 2a and b). As PVP surfacant was added in the solution, the FeS2 particles show budlike microshperes with diameters of also about 2–3 ␮m (Fig. 2c). In addition, the bud-like microshperes are built by 2D submicroflakes

Fig. 2. SEM images of FeS2 powders prepared (a and b) without and (c and d) with PVP.

D. Zhang et al. / Materials Science and Engineering B 178 (2013) 483–488

(a)

without PVP with PVP

10

485

2 nd 3 rd

1 st

2.4

Potential (V)

Number (%)

8 6 4

2.0

0.93 V

1.6

2

1 st 3 rd

1.2

2 nd

0 0

2

4

6 Diameter ( μm)

8

200

400

600

800

-1

Spectific Capacity (mAh g )

10

(b)

Fig. 3. The particle size distributions of FeS2 powders.

2 nd

3 rd

1 st

contacting each other randomly (Fig. 2d). Each submicroflake is in the range of 0.5–1 ␮m in width and length, and about 60 nm in thickness. Besides, the particle size distributions of the powders are displayed in Fig. 3. The particle sizes of the FeS2 powders synthesized with and without PVP are both mainly in the range of 1.5–3 ␮m. In addition, the BET test show that the specific surface area of bud-like microshperes is 8.23 m2 g−1 , which is nearly four times larger than the solid ones prepared without using PVP (2.25 m2 g−1 ). It is consistent well with the SEM images. It is concluded that the adding of PVP has very little influence on the particle size, but large influence on the morphology and the surface condition of FeS2 powders. Based on the above results, we can clearly see that PVP is an effective surfactant for controlling the growth of FeS2 . The formation reaction for FeS2 can be summarized as follows [20]:

Potential (V)

2.4

2.0 0.90 V

1.6 1 st 2 nd

3 rd

1.2 0

200

400

600

800

-1

Specific Capacity (mAh g ) Fig. 4. The first three discharge–charge curves of FeS2 powders prepared (a) without and (b) with PVP at a current density of 45 mA g−1 between 1.2 and 2.6 V.

NH2 CSNH2 + S + FeSO4 + 2H2 O → 2NH3 ↑ +FeS2 + CO2 ↑ +H2 SO4

(1)

It is widely accepted that the preferential adsorption of capping agent molecules on different crystal facets is crucial to direct the growth of particles into various morphologies by controlling the growth rates along different crystal axes [16,21,22]. It is considered that the geometrical shape of a pyrite crystal is determined by the ratio (R) of I(2 0 0)/I(1 1 1) which means the ratio of growth rate along the [1 0 0] direction to that along the [1 1 1] direction [15,16]. In this work, the R value of the pyrite FeS2 powder prepared with PVP is about 3.01, which is lower than that without PVP (3.38) according to the XRD pattern. This is because C O group of PVP could adsorb on the {1 0 0} facets of FeS2 and slow down its growth along the [1 0 0] direction [23]. The lower ratio favors the formation of two-dimensional sheets which show {1 0 0} facets [24]. FeS2 nanosheets grow into nanoflakes at an early stage and small-sized nanoflakes are converted into large-sized submicroflakes to form skeleton spheres. Finally, the hierarchical bud-like microspheres are formed. Fig. 4 shows the first three discharge–charge curves of FeS2 electrodes between 1.2 and 2.6 V at a current density of 45 mA g−1 . Both the FeS2 electrodes exhibit one discharge plateau at 1.5 V or lower corresponding to the transformation from FeS2 to Li2 S and Fe, and two charge plateaus at 1.8 and 2.45 V indexed to the formation of Li2+x Fe1−x S2 , Fe1−x S and further to FeS, S and Li [25] in initial discharge curve. The following discharge curves show additional plateaus at about 2.0 V, indicating a change in the mechanism

of pyrite reduction. However, there are still some differences in discharge–charge curves of the two electrodes. The bud-like FeS2 microspheres exhibit higher initial capacity (773 mAh g−1 ) and initial coulombic efficiency (88.15%) than the irregular FeS2 particles at 45 mA g−1 (696 mAh g−1 and 83.76%). Besides, the bud-like FeS2 microspheres show a high discharge plateau and a low charge plateau comparing to the irregular FeS2 particles, indicating smaller polarization and lower internal resistance of bud-like FeS2 . It is due to the fact that a large number of submicroflakes on the surface of the microspheres increase the contact area between the active material and electrolyte, offering more sites to accommodate Li-ions and enhancing the utilization of the FeS2 powder. As a result, for the bud-like FeS2 microspheres, the first charge reaction can proceed to a high extent and thus results in the higher initial coulombic efficiency. Notice that the bud-like FeS2 microspheres show a high discharge plateau and a low charge plateau, indicating small potential hysteresis and low internal resistance of the cell [26,27]. The cycling performance for FeS2 electrodes at current densities of 45 and 89 mA g−1 is displayed in Fig. 5. Obviously, both the two electrodes show a decrease of the discharge capacity trend. It is well accepted that the degradation of cycle capability for FeS2 electrodes is associated with the structural irreversibility, dissolution of sulfur as polysulfides and the active material/electrolyte contact interface. For both the electrodes, the structural irreversibility and dissolution of sulfur as polysulfides is the main cause for the decrease of the discharge capacity. However, the bud-like FeS2 microspheres delivers the higher discharge capacities and better cycling performance

486

D. Zhang et al. / Materials Science and Engineering B 178 (2013) 483–488

400 -1

with PVP 45 mA g -1 with PVP 89 mA g -1 without PVP 45 mA g -1 without PVP 89 mA g

600

Re 300

500

400 58.8 % 50.64 %

300 50.40 %

200

5

10

15

20

25

CPE

CPE

Rsf1

Rsf2

Cdl

Rct

without PVP with PVP

58.95 %

-Z" (ohm)

-1

Discharge Capacity (mAh g )

700

Zw

0.01 Hz

200

100 100 kHz

30

Cycle Number Fig. 5. Cycling performances of FeS2 electrodes between 1.2 and 2.6 V at current densities of 45 and 89 mA g−1 for 30 cycles.

1 Hz

0 0

100

200

300

400

Z' (ohm) than the other sample. The irregular FeS2 particles and bud-like FeS2 microspheres can sustain 279 and 387 mAh g−1 at 45 mA g−1 after 30 cycles, and 275 and 368 mAh g−1 at 89 mA g−1 after 30 cycles. In addition, the bud-like FeS2 microspheres exhibit higher capacity retention. After 30 cycles, the bud-like FeS2 microspheres can sustain 58.95% and 58.80% fraction of the 2nd cycle at 45 and 89 mA g−1 , respectively. While, FeS2 particles prepared without PVP sustain only 50.64% and 50.40% after 30 cycles, respectively. The improved cycling performance is due to the large contact area between electrode and electrolyte, resulting in fast electrode kinetics. Moreover, even compared with the FeS2 materials reported in the literatures [28,29], a better cycling performance is achieved in this work. To better understand the electrochemical process, CV curves of the FeS2 electrodes between 1.1 V and 2.7 V for the first cycle are recorded in Fig. 6. Both the CV curves exhibit a single reduction peak and two oxidation peaks, which are consistent well with Fig. 4. For the bud-like FeS2 microspheres, a main reduction peak locates around 1.29 V and two oxidation peaks near 1.84 and 2.44 V. Compared to the irregular FeS2 , bud-like FeS2 microspheres show smaller interval between the oxidation and reduction peaks, indicating the weaker polarization and better reversibility. As is well known, the polarization is associated with the transferring delay of electrons on the active material/electrolyte interface [30]. It is considered that electrons and lithium ions can transfer more actively

2.44

1.84

0.5

b

-1

Current Density (A g )

1.0

a

0.0 -0.5 -1.0

b a

-1.5 1.0

1.29

1.5

2.0

2.5

Potential (V) Fig. 6. CV curves of FeS2 electrodes (a) without and (b) with PVP for the first cycle between 1.1 and 2.7 V (versus Li/Li+ ) at a scan rate of 0.1 mV s−1 .

Fig. 7. Nyquist plots of FeS2 electrodes from 0.01 Hz to 100 kHz after discharge the electrodes to 1.2 V at the 5th cycle.

in bud-like FeS2 microspheres due to the short path lengths for electron and Li-ions transport. Besides, bud-like FeS2 microspheres also exhibits the higher intensity of electrochemical peaks and larger area of these peaks than the other sample, suggesting more sufficient electrochemical reactions occur on the active surface of hierarchical bud-like microspheres. The difference in electrochemical behavior of the FeS2 electrodes is also explained by EIS analysis. Fig. 7 shows Nyquist plots of FeS2 from 0.01 Hz to 100 kHz after discharge the electrodes to 1.2 V at the 5th cycle. It is generally believed that the semicircle at middle frequency region is ascribed to the charge-transfer process of lithium ions at the FeS2 /electrolyte interface, while the line in the lower frequency region is corresponded to the diffusion of lithium ions in bulk FeS2 . The bud-like FeS2 electrode exhibits small semicircles compared with the other sample, which indicates a low charge transfer resistance at the FeS2 /electrolyte interface. This assumption is further confirmed by fitting to an equivalent circuit shown in Fig. 7. Aurbach et al. [31] demonstrated that the SEI had a multilayer structure due to difference in their composition as a function of the distance of the surface species precipitated from the active reductive electrode surface by means of XPS. Thus, a“Voigt”type equivalent analog of two R||C circuits in series (Rsfi and Csfi , i = 1 and 2) can be used to model the migration of Li+ through such surface films with a multilayer structure, in which each of these R||C circuits relates to Li+ migration through one of the layers composing the surface films covering the surface [32]. In equivalent circuit, a constant phase element (CPE) is instead of pure capacitance (due to the observation of a depressed semicircle). Rsf represents the electrolyte resistance and Rct is the charge transfer resistance of lithium ions at the interface between electrolyte and electrode. Zw and Cdl correspond to the solid state diffusion in FeS2 and intercalation capacitance, respectively [33,34]. Rsf1 and Rsf2 of the FeS2 electrode synthesized with PVP is 17.38 and 46.44 , respectively. Rsf1 and Rsf2 of the FeS2 electrode synthesized without PVP is 520.8 and 180 , which is nearly 30 times and 5 times larger than the FeS2 prepared with PVP, respectively. The small value of the FeS2 electrode synthesized with PVP indicates that the electrons and Li ions can transfer more quickly on the surface layer. As is well known, the better contact between active materials and electrolyte will lead to the lower charge transfer resistance. In order to further discuss the electrode reaction kinetics of budlike FeS2 microspheres, GITT is used to determine the diffusion

D. Zhang et al. / Materials Science and Engineering B 178 (2013) 483–488

487

Fig. 8. GITT measurements: (a) charge/discharge GITT curves for FeS2 electrodes, (b) a single titration step of FeS2 electrode without PVP at 2nd discharge, (c) the plot of transient voltage E vs. square root of titration time  1/2 and (d) chemical diffusion coefficients of Li-ions at 2nd discharge.

coefficient of lithium ions in FeS2 . Fig. 8a shows GITT curves of FeS2 electrodes in the initial cycles. The equilibrium potentials of FeS2 electrodes gradually decreased with Li-ions insertion but increased with Li-ions extraction. The first discharge curve shows a higher overpotential than the following curves because of the dislocations and other defects generated during the first discharge process [35]. As shown in Fig. 8a, the electrodes show similar curves, indicating that the morphology has minor effect on the voltage hysteresis, which is consistent with Ref. [28]. To investigate the kinetics of discharge, the chemical diffusion coefficient of Li-ions (DLi +) is calculated according to Eq. (1) derived by Weppner and Huggins [36]: DLi+ =

4 

 mV 2  M

MA

Es

2  √

(dE /d )

 << L2 DLi+

 (2)

where VM is the molar volume of the compound, M and m are the molecular weight and mass of FeS2 , respectively, A is the interface between the electrolyte and active material, and L is the radius of the active material particle. Es is the change of the steady-state (equilibrium) voltage,  is the time and E is the overpotential over a single galvanostatic titration. If E versus  1/2 shows a straight line behavior over the entire time period of current flux (as shown in Fig. 8b inset), Eq. (1) can be further simplified as [36]: DLi+ =

4 

 mV 2  E 2 M

MA

s

E

(3)

Fig. 8b shows a typical E versus  1/2 profile for single titration step of the FeS2 electrodes at 2nd discharge. Note that the transient potential exhibits a linear relationship with the square

root of titration time  1/2 almost in the entire period of current flux. Based on Eq. (2) and GITT measurement, we can obtain the diffusion coefficients of Li ions at varying potential during 2nd discharge, as shown in Fig. 8c. The diffusion coefficients are in the range of 10−6 –10−11 cm2 s−1 . It is found that the chemical diffusion coefficients strongly depend on the electrochemical process. It is reported that for synthetic FeS2 the 2.1 V lithiation step was attributed to the formation of pyrrhotite Fe1−x S and Li2+x Fe1−x S2 (0 < x < 0.33) solid solution as intermediate phases, and the 1.5 V reaction produced a mixture of Li2 S and Fe as final products [37]. It can be seen that the phase transitions cause a rapid decrease of the Li-ions diffusion coefficient. It is important to note that the values of DLi + are not reliable within the range where the two phases coexist in the plateau, because the Li-ions diffusion coefficient cannot be determined as a single value in the two-phase region. Thus, DLi + in the quasi-plateau at 1.5 and 2.1 V can only be considered as the apparent values [38,39]. For comparison, these values are calculated using the same methods and are given in Fig. 8c. Obviously, the bud-like FeS2 microspheres show larger DLi + than the other sample. It is reported that particle size has a remarkable effect on the reaction mechanism and phases involved upon Li-insertion [40,41]. The increase of the DLi + value is considered to be caused by the small particle size, which provides short path lengths for Li-ions and more sites to accommodate Li-ions, resulting in fast transferring of Li-ions. It is consistent well with the result of CV. The rate capability for FeS2 electrodes is presented in Fig. 9. The FeS2 electrode synthesized with PVP has an initial discharge capacity of 758, 451 and 322 mAh g−1 at 45, 450 and 890 mA g−1 , while the one synthesized without PVP has an initial discharge capacity of 701, 351 and 227 mAh g−1 at 45, 450 and 890 mA g−1 ,

488

D. Zhang et al. / Materials Science and Engineering B 178 (2013) 483–488

800

-1

Discharge Capacity (mAh g )

with PVP without PVP 45 mA g

600

-1

450 mA g

-1

400

-1

890 mA g

200

0

0

10

20

30

40

Cycle number Fig. 9. Rate capability of FeS2 electrodes between 1.2 and 2.6 V at current densities of 45, 450 and 890 mA g−1 for 45 cycles.

respectively. After 45 cycles, the FeS2 electrode synthesized with PVP can sustain 253 mAh g−1 , which is higher than the one synthesized without PVP. The good rate capability is attributed to the unique structure which provides large contact area and large DLi +, leading to enhanced electrode process kinetics. 4. Conclusions Hierarchical bud-like FeS2 microspheres synthesized by PVPassisted solvothermal reaction show improved initial specific capacity, high coulombic efficiency and improved rate capability. The initial discharge capacity of bud-like FeS2 microspheres is 773 and 749 mAh g−1 at 45 and 89 mA g−1 , respectively. The discharge capacity after 30 cycles decreases to 387 and 368 mAh g−1 at 45 and 89 mA g−1 , respectively. The results of CV, GITT and Nyquist plots confirm that the enhanced electrochemical performance is due to the large contact area between the FeS2 and electrolyte, large DLi + and decreased polarization. References [1] G. Ardel, D. Golodnitsky, K. Freedman, E. Peled, G.B. Appetecchi, P. Romagnoli, B. Scrosati, Journal of Power Sources 110 (2002) 152–162. [2] A. Débart, L. Dupont, R. Patrice, J.M. Tarascon, Solid State Science 8 (2006) 640–651. [3] D. Zhang, J.P. Tu, J.Y. Xiang, Y.Q. Qiao, X.H. Xia, X.L. Wang, C.D. Gu, Electrochimica Acta 56 (2011) 9980–9985. [4] E. Strauss, G. Ardel, V. Livshits, L. Burstein, D. Golodnitsky, E. Peled, Journal of Power Sources 88 (2000) 206–218. [5] E. Peled, D. Golodnitsky, E. Strauss, J. Lang, Y. Lavi, Electrochimica Acta 43 (1998) 1593–1599.

[6] C.V.S. Reddy, E.H. Walker, S.A. Wicker, Q.L. Williams, R.R. Kalluru, Journal of Solid State Electrochemistry 13 (2009) 1945–1949. [7] J. Santos-Pena, P. Soudan, C. Aren, G. Palomino, S. Franger, Journal of Solid State Electrochemistry 10 (2006) 1–9. [8] J.Y. Xiang, J.P. Tu, Y.Q. Qiao, X.L. Wang, J. Zhong, D. Zhang, Journal of Physical Chemistry C 115 (2011) 2505–2513. [9] K. Ragavendran, H.L. Chou, L. Lu, M.O. Lai, B.J. Hwang, R.R. Kumar, S. Gopukumar, B. Emmanuel, D. Vasudevan, D. Sherwood, Material Science and Engineering B 176 (2011) 1257–1263. [10] Y.J. Mai, J.P. Tu, X.H. Xia, C.D. Gu, X.L. Wang, Journal of Power Sources 196 (2011) 6388–6393. [11] J.Y. Xiang, J.P. Tu, L. Zhang, Y. Zhou, X.L. Wang, S.J. Shi, Electrochimica Acta 55 (2011) 1820–1824. [12] X.H. Huang, J.P. Tu, X.H. Xia, J.Y. Xiang, X.L. Wang, Journal of Power Sources 195 (2010) 1207–1210. [13] Y. Xia, W.K. Zhang, H. Huang, Y.P. Gan, C.G. Li, X.Y. Tao, Material Science and Engineering B 176 (2011) 633–639. [14] Y. Li, J.P. Tu, H.M. Wu, Y.F. Yuan, D.Q. Shi, Material Science and Engineering B 128 (2006) 75–79. [15] D.W. Wang, M.H. Wu, Q.H. Wang, T.M. Wang, J.A. Chen, Ionics 17 (2011) 163–167. [16] D.W. Wang, Q.H. Wang, T.M. Wang, CrystEngComm 12 (2010) 3797–3805. [17] X.Y. Chen, Z.H. Wang, X. Wang, J.X. Wan, J.W. Liu, Y.T. Qian, Inorganic Chemistry 44 (2005) 951–954. [18] P. Gao, Y. Xie, L.N. Ye, Y. Chen, Q.X. Guo, Crystal Growth and Design 6 (2006) 583–587. [19] S. Kar, S. Chaudhuri, Chemical Physics Letters 398 (2004) 22–26. [20] Y.H. Chen, Y.F. Zheng, X.G. Zhang, Y.F. Sun, Y.Z. Dong, Science in China Series G 48 (2005) 188–200. [21] Q.Q. Wang, G.L. Zhao, G.R. Han, Materials Letters 59 (2005) 2625–2629. [22] S.M. Lee, S.N. Cho, J. Cheon, Advanced Materials 15 (2003) 441–444. [23] A.S. Barnard, S.P. Russo, Journal of Physical Chemistry C 111 (2007) 11742–11746. [24] W.J. Dong, X.B. Wang, B.J. Li, L.N. Wang, B.Y. Chen, C.R. Li, X.A. Li, T.R. Zhang, Z. Shi, Dalton Transactions 40 (2011) 243–248. [25] R. Fong, J.R. Dahn, C.H.W. Jones, Journal of the Electrochemical Society 136 (1989) 3206–3210. [26] P.G. Bruce, B. Scrosati, J.M. Tarascon, Angewandte Chemie International Edition 47 (2008) 2930–2946. [27] J. Cabana, L. Monconduit, D. Larcher, M.R. Palacin, Advanced Materials 22 (2011) E170–E192. [28] L.A. Montoro, J.M. Rosolen, Solid State Ionics 159 (2003) 233–240. [29] Y.J. Choi, N.W. Kim, K.W. Kim, K.K. Cho, G.B. Cho, H.J. Ahn, J.H. Ahn, K.S. Ryu, H.B. Gu, Journal of Alloys and Compounds 485 (2009) 462–466. [30] D. Aurbach, B. Markovsky, I. Weissman, E. Levi, Y. Ein-Eli, Electrochimica Acta 45 (1999) 67–86. [31] D. Aurbach, I. Weissman, A. Schechter, H. Cohen, Langmuir 12 (1996) 3991–4007. [32] M.D. Levi, D. Aurbach, Journal of Physical Chemistry B 101 (1997) 4630–4640. [33] C. Lai, H.Z. Zhang, G.R. Li, X.P. Gao, Journal of Power Sources 196 (2011) 4735–4740. [34] M.D. Levi, D. Aurbach, Journal of Physical Chemistry B 101 (1997) 4641–4647. [35] Y.J. Zhu, C.S. Wang, Journal of Power Sources 196 (2011) 1442–1448. [36] W. Weppner, R.A. Huggins, Journal of the Electrochemical Society 124 (1977) 1569–1578. [37] Y. Shao-Horn, S. Osmialowski, Q.C. Horn, Journal of Electrochemical Society A 1499 (149) (2002). [38] J. Xie, N. Imanishi, T. Matsumura, A. Hirano, Y. Takeda, O. Yamamoto, Solid State Ionics 179 (2008) 362–370. [39] J. Xie, N. Imanishi, T. Zhang, A. Hirano, Y. Takeda, O. Yamamoto, Electrochimica Acta 54 (2009) 4631–4637. [40] J. Xie, N. Imanishi, A. Hirano, Y. Takeda, O. Yamamoto, X.B. Zhao, G.S. Cao, Solid State Ionics 181 (2010) 1611–1615. [41] M. Okubo, E. Hosono, J. Kim, M. Enomoto, N. Kojima, T. Kudo, H.S. Zhou, I. Honma, Journal of the American Chemical Society 129 (2007) 7444–7452.