Electrochemical performance of flowerlike CaSnO3 as high capacity anode material for lithium-ion batteries

Electrochemical performance of flowerlike CaSnO3 as high capacity anode material for lithium-ion batteries

Electrochimica Acta 55 (2010) 3891–3896 Contents lists available at ScienceDirect Electrochimica Acta journal homepage: www.elsevier.com/locate/elec...

956KB Sizes 0 Downloads 22 Views

Electrochimica Acta 55 (2010) 3891–3896

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Electrochemical performance of flowerlike CaSnO3 as high capacity anode material for lithium-ion batteries Sen Zhao, Ying Bai, Wei-Feng Zhang ∗ Key Laboratory of Photovoltaic Materials of Henan Province and School of Physics & Electronics, Henan University, Kaifeng 475001, PR China

a r t i c l e

i n f o

Article history: Received 4 November 2009 Received in revised form 30 January 2010 Accepted 6 February 2010 Available online 12 February 2010 Keywords: CaSnO3 Anode materials Electrochemical performance Lithium-ion batteries

a b s t r a c t Nanosized CaSnO3 is synthesized by a hydrothermal process and characterized by X-ray diffraction (XRD), Raman spectroscopy, and scanning electron microscopy (SEM). The SEM observation shows the sample has a porous flowerlike morphology. The electrochemical results exhibit that the stable and reversible capacity of 547 mAh g−1 is obtained after 50 cycles at 60 mA g−1 (0.1 C) and the corresponding charge capacity is determined to be 316 mAh g−1 at the current density of 2.5 C. Cyclic voltammetry and electrochemical impedance spectroscopy data are analyzed to complement the galvanostatic results. The observed excellent performance is attributed to the porous structure and large surface area of flowerlike CaSnO3 . © 2010 Elsevier Ltd. All rights reserved.

1. Introduction Lithium-ion batteries (LIBs) have long been considered as the most promising chemical power sources for a wide variety of modern portable information technology equipment because of their high energy storage density, long cycle life, little memory effect, poisonous-metal free, and so on [1–4]. Among kinds of anode materials for lithium-ion batteries, graphite carbons have been commercialized owing to its excellent cycling behavior upon repeated charge and discharge cycles. However, the graphite anode has some disadvantages such as electrical disconnection, structural deformation, initial loss of capacity, and limitation of theoretical capacity of 372 mAh g−1 [5,6]. Nowadays, the increasing demand for higher energy density battery systems has motivated an ongoing search for new storage electrodes with high charge capacity and excellent cyclability. Many intermetallic-based anode materials have been studied recently to improve the capacity and cyclability of batteries [7–9]. In the past few years, Sn-based binary and ternary oxides with various morphologies have become attractive anode materials for LIBs due to their feasible low potentials for Li+ insertion and high reversible storage capacities [10–13]. The Sn-based binary oxides such as SnO2 [14], SnO2 /C [4], SnO2 /Al2 O3 [15] have been widely studied. In addition, the Sn-based ternary oxides, especially, a number of stannates, ASnO3 (A = Ca, Sr, and Ba) [12,16–18] and M2 SnO4 (M = Mg, Mn, Co, Zn) [19–22] have been investigated as anode mate-

rials for LIBs. Recently, nanomaterials have attracted much interest as anodes for Li-ion batteries because of their larger reversible capacity, higher Li+ diffusion coefficients, and better rate capability than conventional micrometer materials. It was reported that the nano-CaSnO3 synthesized by sol–gel [12,17] and coprecipitation methods [18] show high reversible capacities of 380 and 450 mAh g−1 with good cycling stability up to 50 cycles in 0.005–1.0 V 60 mA g−1 . It is known that the electrochemical performance of electrode materials is close related to their microstructures and morphologies. Generally, a porous structure with large surface area facilitates the electrochemical lithium-reaction properties. The hydrothermal method is a facile way to prepare electrode materials, by which peculiar morphologies can be obtained. In this work, therefore, we first synthesized CaSn(OH)6 precursors with three-dimensional (3D) symmetrically distributed eight pods via a hydrothermal route and then converted the precursors into flowerlike CaSnO3 with similar morphologies by heat treatment. The preparation method is convenient, low-cost and easily adapts to mass production. The flowerlike CaSnO3 samples exhibit excellent electrochemical properties. In particular, the sample obtained by annealing the CaSn(OH)6 precursor at 1000 ◦ C shows a high charge capacity of 547 mAh g−1 after 50 cycles in 0.005–1.0 V at 60 mA g−1 (0.1 C). Interestingly, its capacity is still able to stabilize at a favorable high value at 2.0 C, equaling to the theoretical capacity of graphite (372 mAh g−1 ). 2. Experimental

∗ Corresponding author. Tel.: +86 378 3881 940; fax: +86 378 3880 659. E-mail addresses: [email protected], [email protected] (W.-F. Zhang). 0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2010.02.018

All the reagents were of analytical grade and were used without further purification. In a typical experiment, certain ratio of


S. Zhao et al. / Electrochimica Acta 55 (2010) 3891–3896

Table 1 Impedance parameters obtained by fitting the impedance spectra of CaSnO3 (vs. Li/Li+ ) to the equivalent circuit elements during the first discharge. Potential vs. (Li/Li+ ) (V)










Rsf+ct (±5) () CPEsf+ct (±2) (␮F)

547 15

728 12

748 13

750 12

670 13

697 20

828 19

1478 13

1722 12

Ca(NO3 )2 ·4H2 O and Na2 SnO3 ·3H2 O were separately put into two beakers and dissolved in deionized water to form transparent solutions with stirring. Then, Ca(NO3 )2 solution was added into that of Na2 SnO3 drop by drop to form homogeneous white suspension. Afterwards, the mixture was transferred into a Teflon-lined stainless steel autoclave and sealed tightly. The autoclave was maintained at 160 ◦ C for 10 h and cooled naturally to room temperature. After thoroughly rinsed and dried at 70 ◦ C for 12 h in air, the CaSn(OH)6 precursors were obtained. Then the precursors were further annealed in a muffle furnace at 700 (sample 1), 800 (sample 2), 900 (sample 3) and 1000 ◦ C (sample 4) separately and cooled down to room temperature, finally yielding the white powders. The products were ground into a tiny powder for characterization. X-ray diffraction (XRD) patterns were recorded on a DX-2500 diffractometer (Fangyuan, Dandong) with Cu K␣ radiation of  = 0.154145 nm. The Raman spectra were collected on a laser Raman Spectrometer (RM-1000, Renishaw) with 457.5 nm solidstate laser. Morphological characterizations were carried out by scanning electron microscopy (SEM) (FE-SEM, X-650). The surface area of the samples was determined by BET measurement (Micromeritics-3000, micrometrics, USA) on nitrogen adsorption at 77 K after the pretreatment at 573 K for 2 h. The electrochemical properties of the CaSnO3 samples were measured at room temperature. The working electrodes were fabricated by compressing the mixture of 70 wt.% active materials (CaSnO3 ), 20 wt.% acetylene black (Super-P, MMM Carbon), and 10 wt.% binder (polyvinylidene fluoride, PVDF) dissolved in N-methyl-pyrolline(NMP) on a 10 ␮m thick copper foil. After thoroughly dried, the films were cut into circular disks of 15 mm in diameter as working electrode. The electrolyte was 1 M LiPF6 dissolved in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) with the volume ratio of 1:1. The assembly process was conducted in an argon-filled glove-box with the content of H2 O and O2 less than 1 ppm. Before electrochemical tests, the batteries were aged for 24 h to ensure good soakage of the CaSnO3 particles in the electrolyte. The cells were charged and discharged on a battery tester (LAND BT110, China) between 0.005 and 1.0 V (vs. Li/Li+ ). Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) investigation were performed using an electrochemical workstation with three electrode systems (CHI660C, Shanghai). The CV curves were recorded between 0.005 and 1.0 V at a scan rate of 58 ␮V s−1 . The EIS measurements were performed over a frequency range from 100 KHz to 5 mHz.

Fig. 1. XRD patterns of (a) as-prepared CaSn(OH)6 and CaSnO3 samples annealed at different temperatures of (b) 700 (sample 1), (c) 800 (sample 2), (d) 900 (sample 3), and (e) 1000 ◦ C (sample 4).

products are well-crystallized. The crystallite size of CaSnO3 was calculated using the Scherrer’s formula, D = K/ˇcos  [23], where K is a constant,  the wavelength of Cu K␣ radiation in Å, ˇ the full width at half maximum (FWHM) in radians obtained using Jade software, and  the scattering angle. The estimated crystallite sizes of the four CaSnO3 samples are 27, 29, 32 and 37 nm, respectively. Raman spectroscopy is a more sensitive probe of structural distortions, short-range order, and symmetry in solids in comparison with XRD, which usually reveals structural information on longrange order of materials. Fig. 2 displays the Raman spectra of CaSnO3 . All the peaks locate at 181, 277, 356, 443 and 699 cm−1 for the CaSnO3 samples obtained at different temperatures. The bands at 356 and 443 cm−1 are related to Sn-O torsional modes, and the band at 699 cm−1 is ascribed to the Sn-O stretching mode [24]. The Raman investigation indicates that the four CaSnO3 samples obtained at different temperatures are of a well-crystallized perovskite phase, and, not only long-range order but also shortrange order. It is also seen from Fig. 2 that the Raman peaks are

3. Results and discussion 3.1. Physical characterizations Fig. 1 illustrates the XRD patterns of CaSn(OH)6 precursor and CaSnO3 obtained at different annealing temperatures. All the diffraction peaks in Fig. 1(a) can be indexed to the pure cubic phase of CaSn(OH)6 (JCPDS 74-1823). The corresponding diffraction peaks of the four CaSnO3 samples in Fig. 1(b)–(e) can be all indexed to the pure orthorhombic perovskite structure CaSnO3 (JCPDS 77-1797). No peaks of secondary phase are observed in the XRD patterns, suggesting that the CaSn(OH)6 precursors have been completely converted into a single-phase CaSnO3 . In addition, the strong diffraction peaks indicate that the as-obtained CaSnO3

Fig. 2. Raman spectra of (a) sample 1, (b) sample 2, (c) sample 3 and (d) sample 4.

S. Zhao et al. / Electrochimica Acta 55 (2010) 3891–3896


Fig. 3. SEM images of (a) CaSn(OH)6 precursor and (b) sample 1, (c) sample 2, (d) sample 3 and (e) sample 4.

strengthened and get narrow with increasing annealing temperature, indicating the improvement of crystallization with annealing. The SEM images of the CaSn(OH)6 precursors and CaSnO3 samples are shown in Fig. 3. Fig. 3(a) displays the CaSn(OH)6 precursors are composed of the secondary particles (∼4 ␮m), which are formed by tight accumulating the primary particles (∼400 nm) along threedimensional directions of symmetrically distributed eight horns which extend outward from the center [25,26]. It shows the shape of an open four petals is symmetrically distributed and centers of the secondary particles were deep holes with diameter of 400 nm from the upside of the secondary particles. It can be seen in Fig. 3(b)–(e) that the morphologies of the four CaSnO3 samples obtained by annealing CaSn(OH)6 at different temperatures are similar to that shown in Fig. 3(a). With increasing annealing temperature, however, the edges and corners of the primary particles are getting intangible, and the growth direction tends to spherical growth. Simultaneously, four petals of the secondary particles stretch and open gradually. It is observed among the secondary particles in Fig. 3(d) and (e) that the size and amount of holes get larger as the annealing temperature increases from 700 to 1000 ◦ C. The observations in Fig. 3(b)–(e) indicate that the disparate growth process of the primary particles at different temperatures

results in the difference and structural evolution of the morphologies of the four CaSnO3 samples. Evolution of the BET surface area with annealing for the CaSnO3 samples is given as follows: 3.7 (sample 1), 5.0 (sample 2), 6.8 (sample 3), 7.3 (sample 4) m2 g−1 . The surface area increases about two times as the annealing temperature increases. The obvious rise in BET surface area mainly results from the emerging holes, as can be clearly seen from the SEM images in Fig. 3. 3.2. Galvanostatic cycling Fig. 4 displays the charge and discharge profiles at the first cycle for the four CaSnO3 electrodes at a current density 60 mA g−1 (0.1 C). The first discharge profiles of the samples 1–4 (at 700, 800, 900 and 1000 ◦ C, respectively) show a short plateau at 0.7 V (vs. Li/Li+ ) with an onset at 1.2 V followed by a sloping profile until a capacity of approximate 300 mAh g−1 is reached and then a sloping profile from ∼0.2 V to the lower cutoff potential, 0.005 V. In the calculation of the capacity of active material CaSnO3 , the contribution of certain proportion of acetylene black was considered and deducted. The four first charge capacities are getting higher from the samples 1–4 and the capacity of sample 4 reaches 515 mAh g−1 , which is the highest among the four samples. This may be ascribed


S. Zhao et al. / Electrochimica Acta 55 (2010) 3891–3896

Fig. 4. The first galvanostatic potential profiles for the four CaSnO3 samples at 60 mA g−1 (0.1 C, 1 C = 600 mA g−1 ).

Fig. 5. Cyclability for the four CaSnO3 samples during the 2–50 cycles between 0.005 and 1.0 V at 60 mA g−1 (0.1 C).

to the fact that sample 4 has many holes and the largest BET surface area, as shown in Figs 3(e) and 4, which can provide more active sites for electrochemical reactions. The electrochemical process during the first discharge leads to the formation of electrochemically inactive CaO matrix and active Sn metal (see Eq. (1)) in the case of CaSnO3 . The freshly formed Sn metal reacts with Li+ to form the intermetallic alloy, Li4.4 Sn at ∼0.2 V according to the following reaction of Eq. (2) [12,17,18].

capacities at the 50th cycle of the samples 1–4 are 418, 458, 512 and 547 mAh g−1 , respectively, with a coulombic efficiency larger than 99%. The excellent cycling performance is attributed to the special morphology of flowerlike CaSnO3 . The inactive CaO matrix with good buffering ability can relieve the strain associated with the volume change and prevent the aggregation of the active regions [30,31]. Fig. 6 shows the rate performance of CaSnO3 between 0.005 and 1.0 V (vs. Li/Li+ ). The charge capacities of these cells slowly decrease with increasing the current density and are determined to be 369, 380, 426 and 448 mAh g−1 , respectively, at the current density of 0.8 C for the samples 1–4, with a coulombic efficiency larger than 99%. The four samples present excellent capacity retention at low rate while the capacity declines rapidly when the current density is larger than 0.8 C. However, it is observed that the capacity of the sample 4 at 2.0 C is about equal to the theoretical capacity of graphite (372 mAh g−1 ), meaning that sample 4 shows a stable and high capacity at 2.0 C up to 140 cycles. At high current densities, high Li+ insertion-flux density and slow Li+ transport in the electrode materials with traditional morphologies (e.g., cube, oblong and sphere-like etc.) result in concentration polarization of Li+ , and thus rapid decline of capacity. However, for the porous flowerlike CaSnO3 , increasing the electrode–electrolyte contact area and decreasing the diffusion

CaSnO3 + 4Li+ + 4e− → CaO + 2Li2 O + Sn


4.4Li+ + 4.4e− + Sn ↔ Li4.4 Sn


The above equations show that the overall consumption of 8.4 mol of Li per mole of CaSnO3 constitutes 1089 mAh g−1 as the first theoretical discharge capacity. In the present work, the experimentally observed capacities are 1467 (11.3 mol of Li per mole of sample 1), 1550 (12 mol of Li per mole of sample 2), 1665 (12.8 mol of Li per mole of sample 3) and 1725 (13.3 mol of Li per mole of sample 4) mAh g−1 for samples 1–4, respectively, which are all higher than the theoretically values. Taking sample 4 for instance, the first discharge capacity is 1725 mAh g−1 at current density 60 mA g−1 . The experiment value is about 630 mAh g−1 higher than the first theoretical discharge capacity. The extra capacities can be mainly attributed to the solid electrolyte interphase (SEI) film formation due to the reaction of Li+ with the solvents of the electrolyte [12,13,17,27,28]. The thickness of SEI is usually 4–5 nm, however, we noticed the previous report about that SEI thickness of catalytic activity material Cr2 O3 was 30–90 nm [29]. Sn metal is catalytic activity material also, so SEI of CaSnO3 is thicker than that of general materials causing consumption of more Li+ . Besides, side reaction consumes inevitably more Li+ than the electrode materials with traditional morphologies for special morphology of flowerlike CaSnO3 . The different capacities in the first galvanostatic profile among the four samples can be possibly attributed to that bigger surface area will not only consume more Li+ to form SEI film but also ensure sufficient reaction at particle surface. To compare the annealing effect on electrochemical properties, the charge profiles of the four CaSnO3 /Li cells were conducted. The plots of the charge capacity vs. cycle number of the samples 1–4 between 0.005 and 1 V (vs. Li/Li+ ) at 60 mA g−1 (0.1 C) are demonstrated in Fig. 5. The charge capacity shows a noticeable increasing tendency in the first 15 cycles, which could be due to the activation of metallic Sn. The capacity slowly increases with cycling and is gradually stabilized at the 15th cycles. The four samples have stable reversibility in repeated cycles. For the different annealing temperatures, the cycling shows an obvious different behavior. The charge

Fig. 6. Comparison of the rate performances for the four CaSnO3 samples between 0.005 and 1.0 V.

S. Zhao et al. / Electrochimica Acta 55 (2010) 3891–3896


Fig. 7. Cyclic voltammograms for the four CaSnO3 samples at scan rates of 58 ␮V s−1 in the potential window 0.005–1.0 V for the first three cycles.

distance can lead to a decrease of concentration polarization at high current densities, and consequently, the cycling capability can be markedly improved [32]. The sample 4 has the optimal maintenance ability of capacity, as shown in Fig. 6. It can be inferred that the porous structures and large surface area conduce to sufficient contact between the electrode materials and electrolyte and to shorten the diffusion distance of Li+ . The sample 4 has the most reactive Sn-sites and shortest diffusion length among these four samples, which can enhance the reaction shown in Eq. (2) at the surface [17]. 3.3. Cyclic voltammetry CV measurements were carried out to further clarify the origin of excellent electrochemical performances of the flowerlike CaSnO3 . Fig. 7 shows the first three complete cycles performed between 0.005 and 1.0 V (vs. Li/Li+ ) at the slow scan rate of 58 ␮V s−1 . The first discharge shows the two-step reactions for the samples obtained at different temperatures. A broad shoulder at ∼0.75 V was observed, which can be ascribed to the reaction that CaSnO3 is reduced to Sn metal embedded in the inactive CaO and Li2 O matrix (forward reaction of Eq. (1)). After the first cycle, the cathodic peak of 0.75 V is disappeared, so this step is irreversible reaction, corresponding to Eq. (1). In the first three cycles, the cathodic peaks is located at ∼0.1 V, corresponding to the formation of Li4.4 Sn alloy (forward reaction of Eq. (2)) [17,18]. The redox couple at 0.1/0.6 V is attributed to the reversible reactions in Eq. (2). There is no cathodic peak observed in the region below 1.0 V, corresponding to the formation of SEI from Fig. 7(a)–(d). It is estimated that SEI cathodic peak, which is a part source of the large irreversible capacity at the first cycle, is submerged by the alloying Sn cathodic peak in the first cycle. This will be confirmed by the EIS test described below. As can be fitted from the CV curve in Fig. 7, the interval of reversible cathodic/anodic peaks for sample 4 is 0.54 V, lower than those for the samples 1–3. Because the lower peak potential interval

indicates that the lithium can react with active material more easily [33,34]. Sample 4 has more favorable reaction kinetics and cyclic reversibility. This result is consistent with the excellent cycling performance of sample 4 obtained at high temperature. Therefore, it can be concluded that the CV results support the galvanostatic cycling data. 3.4. Electrochemical impedance spectroscopy EIS measurement was carried out to understand the electrode reaction kinetics [35,36] and the SEI cathodic peak is submerged in the first cycle, which cannot be observed in the CV profiles. The CaSnO3 /Li cells were aged for two days to reach equilibrium before ac impedance spectra were recorded. And, to ensure the SEI is formed just at the preconcerted potential, for instance 0.01 V (vs. Li/Li+ ), the battery was discharged at a constant current density of 60 mA g−1 to 0.01 V, and then followed by a constant potential of 0.01 V (vs. Li/Li+ ) until the current decreased to 2 ␮A. Fig. 8(a) shows the Nyquist plots (z vs. −z ) performed over a frequency range from 100 kHz to 5 mHz at different potentials, where z and −z refer to the real and imaginary parts of the cell impedance. The semicircle in the high-frequency region of the Nyquist plot is mainly the contribution of the solid electrolyte interphase (SEI) film on the electrode [37,38]. It is seen qualitatively that SEI film forms mainly between 0.1 and 0.005 V (see Fig. 8a). This potential is close to that of the Li–Sn alloying/de-alloying, thus the SEI film corresponding peak is submerged in the first CV profile by the bigger Sn alloying peak. Based on the EIS measurement, we ascertain the potential of SEI film formation, and verify the previous conjecture for the SEI cathodic peak in the CV measurements. In addition, the equivalent circuit was depicted in Fig. 8(b), where CPE, R and W represent the double-layer capacitance, resistance and Warburg impedance, respectively. The cell impedance is a summation of the resistances offered by the electrolyte (e), surface film (sf), charge transfer (ct), bulk (b) and Warburg impedance (W). The electrolyte resistance (Re ) value was found to be 3.3 (±0.3) . The


S. Zhao et al. / Electrochimica Acta 55 (2010) 3891–3896

Acknowledgements This work was supported by the National Natural Science Foundation of China (60476001) and the Project of Cultivating Innovative Talents for Colleges & Universities of Henan Province (2002006). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] Fig. 8. (a) Family of Nyquist plots (z vs. −z ) during the first discharge from open circuit potential (OCV∼3.0 V) to 0.005 V at different potential vs. Li/Li+ . (b) The equivalent circuit.

cathodic peak is located at ∼0.75 V, corresponding to the formation of Sn metal (Eq. (1)). In the potential range 0.75–0.2 V (vs. Li/Li+ ), Rsf+ct shows a decreasing trend (Table 1). This possibly results from the formation of metallic Sn which can lead to an increase in the electronic conductivity of the in situ formed composite electrode. Below 0.1 V, Rsf+ct shows a drastic rise trend, which can be mainly attributed to the SEI film formation and certain extent of contact loss among the active particles (caused by a huge volume change due to the formation of Li–Sn alloys). 4. Conclusions

[15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27]

Porous flowerlike CaSnO3 with three-dimensional directions of symmetrically distributed were synthesized by a hydrothermal process. The samples showed significant reversible capacities in the charge–discharge constant current and rate experiments. This is attributed to the special porous structures which can provide more active sites, enhance the conductivity, and shorten Li+ diffusion distance. The stable capacity increased with increasing annealing temperature. The cycling performance of the sample annealed at 1000 ◦ C is the best among these four samples. In addition, its reversible capacity reached 547 mAh g−1 after 50 cycles at the current rate 60 mA g−1 (0.1 C), and is about equal to the theoretical capacity of graphite (372 mAh g−1 ) at 2.0 C with the Coulombic efficiency larger than 99%. These results indicate that the porous flowerlike CaSnO3 is a promising anode material for rechargeable lithium-ion batteries.

[28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38]

J.M. Tarascon, M. Armand, Nature 414 (2001) 359. B. Scrosati, Nature Nanotechnol. 2 (2007) 598. J. Tollefson, Nature 456 (2008) 436. W.M. Zhang, J.S. Hu, Y.G. Guo, S.F. Zheng, L.S. Zhong, W.G. Song, L.J. Wan, Adv. Mater. 20 (2008) 1160. F. Disma, L. Aymard, L. Dupond, J.M. Tarascon, J. Electrochem. Soc. 143 (1996) 3959. F. Salver-Disma, C. Lenain, B. Beaudoin, L. Aymard, J.M. Tarascon, Solid State Ionics 98 (1997) 145. J. Yang, Y. Takeda, N. Imanishi, J.Y. Xie, O. Yamamoto, Solid State Ionics 133 (2000) 189. J.O. Besenhard, J. Yang, M. Winter, J. Power Sources 68 (1997) 87. G. Derrien, J. Hassoun, S. Panero, B. Scrosati, Adv. Mater. 19 (2007) 2336. Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa, T. Miyasaki, Science 276 (1997) 1395. Z. Ying, Q. Wan, H. Cao, Z.T. Song, S.L. Feng, Appl. Phys. Lett. 87 (2005) 113108. N. Sharma, K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, J. Power Sources 139 (2005) 250. J. Zhu, Z. Lu, S.T. Aruna, D. Aurbach, A. Gedanken, Chem. Mater. 12 (2000) 2557. L. Yuan, Z.P. Guo, K. Konstantinov, H.K. Liu, S.X. Dou, J. Power Sources 159 (2006) 345. M. Jayalakshmi, N. Venugopal, K. Phani Raja, M. Mohan Rao, J. Power Sources 158 (2006) 1538. N. Yogesh Sharma, G.V. Sharma, Subba Rao, B.V.R. Chowdari, J. Power Sources 192 (2009) 627. N. Sharma, K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, Electrochem. Commun. 4 (2002) 947. N. Yogesh Sharma, G.V. Sharma, B.V.R. Subba Rao, Chowdari, Chem. Mater. 20 (2008) 6829. P.A. Connor, J.T.S. Irvine, J. Power Sources 97–98 (2001) 223. F. Belliard, P.A. Connor, J.T.S. Irvine, Solid State Ionics 135 (2000) 163. A. Rong, X.P. Gao, G.R. Li, T.Y. Yan, H.Y. Zhu, J.Q. Qu, D.Y. Song, J. Phys. Chem. B 110 (2006) 14754. X.J. Zhua, L.M. Geng, F.Q. Zhang, Y.X. Liu, L.B. Cheng, J. Power Sources 189 (2009) 828. P. Jha, P.R. Arya, A.K. Ganguli, Mater. Chem. Phys. 82 (2003) 355. W.F. Zhang, J.W. Tang, J.H. Ye, J. Mater. Res. 22 (2007) 1863. H. Cheng, Z.G. Lu, Solid State Sci. 10 (2008) 1042. C.H. Fan, X.Y. Song, H.Y. Yu, Z.L. Yin, H.Y. Xu, G.X. Cao, D.S. Zheng, S.X. Sun, Mater. Lett. 61 (2007) 1589. I. Sandu, T. Brousse, D.M. Schleich, M. Danot, J. Solid State Chem. 177 (2004) 4332. F. Huang, Z. Yuan, H. Zhan, Y. Zhou, J. Sun, Mater. Chem. Phys. 83 (2004) 16. J. Hu, H. Li, X.J. Huang, L.Q. Chen, Solid State Ionics 177 (2006) 2791. N. Sharma, K.M. Shaju, G.V. Subba Rao, B.V.R. Chowdari, Electrochim. Acta 49 (2004) 1035. N. Sharma, G.V. Subba Rao, B.V.R. Chowdari, Electrochim. Acta 50 (2005) 5305. Y.K. Zhou, L. Cao, F.B. Zhang, B.L. He, H.L. Li, J. Electrochem. Soc. 150 (2003) A1246. P. Krtil, D. Fattakhova, L. Kavan, S. Burnside, M. Gratzel, Solid State Ionics 135 (2000) 101. H. Lindstrom, S. Sodergren, A. Solbrand, H. Rensmo, J. Hjelm, A. Hagfeldt, S.E. Lindquist, J. Phys. Chem. B 101 (1997) 7717. D. Aurbach, A. Nimberger, B. Markovsky, E. Levi, E. Sominski, A. Gedanken, Chem. Mater. 14 (2002) 4155. N. Sharma, J. Plevert, G.V. Subba Rao, B.V.R. Chowdari, T.J. White, Chem. Mater. 17 (2005) 4700. S.S. Zhang, K. Xu, T.R. Jow, Electrochem. Solid-State Lett. 5 (2002) A92. Y. Bai, Y.F. Yin, N. Liu, B.K. Guo, H.J. Shi, J.Y. Liu, Zh.X. Wang, L.Q. Chen, J. Power Sources 174 (2007) 332.