Electrochemical properties of Li₂MnO₃ nanocrystals synthesized using a hydrothermal method

Abstract

In this article, we report the synthesis of Li₂MnO₃ nanocrystals using a hydrothermal process. XRD, SEM, and TEM analyses are performed, and the electrochemical properties of the resultant nanocrystals are investigated. Adding the oxidant KMnO₄ influences the phase purity, size, and shape of the Li₂MnO₃ nanocrystals. An amount of KMnO₄ that exceeds 6% of the total Mn source enhances the formation of the pure monoclinic Li₂MnO₃ phase. The effect of the amount of KMnO₄ on Li₂MnO₃ grain size can be divided into three ranges. Li₂MnO₃ crystals with a size around 28.7 nm and plate morphology are obtained at less than 6% of KMnO₄ content; those with a size of 28.7 nm to 9.8 nm and mixed morphology of plates and rods are obtained at 6% to 11% KMnO₄ content, and those with a size around 9.8 nm and rod morphology are mainly obtained at KMnO₄ content exceeding 11%. Discharge capacities increase with decreasing size of Li₂MnO₃ nanocrystals in a linear relationship. The voltage of the first charge shows a 4.55 V plateau for Li₂MnO₃ nanocrystals with 28.7 nm size, becoming 4.4 V with a decrease in size to 12.2 nm, and splitting into two plateaus at around 3.8 and 4.4 V with a further decrease in size to 9.8 nm. The XRD and XANES results of the Li₂MnO₃ electrodes obtained after charge/discharge experiments show that smaller sizes are more beneficial in maintaining a layered structure than larger nanocrystals.

---

Introduction

Rapid developments in electric and hybrid cars, as well as large power tools, in recent years have resulted in increased demand for higher energy and power density.^1,2 Currently available cathode materials, such as LiMn₂O₄, LiNi_1/3Co_1/3Mn_1/3O₂, and LiFePO₄, demonstrate specific capacities lower than 200 mA h g⁻¹, severely limiting the energy density of lithium (Li) batteries.^3,4 Given that Li-rich manganese (Mn)-based materials present high specific capacities (>250 mA h g⁻¹), low raw material costs, and several other advantages, these compounds have attracted increasing research attention.⁵ Li₂MnO₃ is an important component of Li-rich material; thus, analyzing its electrochemical reaction mechanism is beneficial for structure change studies and composition design of Li-rich Mn-based solid solutions.

Li₂MnO₃ can be written as Li(Li_1/3Mn_2/3)O₂ and presents an α-NaFeO₂ layered structure (monoclinic C2/m space group) similar to that of LiMO₂. Li atoms occupy the 3a position of rock salt structures, while 1/3 Li and 2/3 Mn occupy the 3b position of transition metal layers, thereby forming a Li, Mn ordered super structure.^6–8 Mn⁴⁺ ions in Li₂MnO₃ are electrochemically inactive; however, decreasing the particle size to sub-micrometer scale causes reversible electrochemical activation.^9,10 The electrochemical activation of this material may be attributed to the mechanisms of oxidation (removal of oxygen to balance charges) and proton exchange.^11–16 The electrochemical activation of Li₂MnO₃, which depend on the two mechanisms, is influenced by grain morphology and size, as well as structure. Grains that are smaller than submicrometer size show electrochemical activity, and their activation increases with decreasing size. Therefore, controlling the size and morphology of Li₂MnO₃ grains is the most important method in increasing electrochemical properties. The synthesis methods of controlling Li₂MnO₃ nanocrystals with particularly high crystallinity have been featured in several reports.

In most reports, Li₂MnO₃ synthesis involves calcination (solid-state reaction and annealing of precursors from soft chemistry methods) to increase crystallinity, but this route renders controlling grain size and morphology difficult.^17–20 Thus, several researchers have focused on hydrothermal synthesis. The hydrothermal treatment of transition metal hydroxides and δ-MnO₂ gives rise to nanosized Li₂MnO₃ and Li₂MnO₃–LiMO₂ (M = Fe, Ni, Co, etc.) plates, wires, and particles.^21–26 Taking α and δ MnO₂ as hydrothermal precursors, Baek obtained hierarchically assembled 2D nanoplates and 0D nanoparticles, respectively, wherein the samples with larger surface area of 113 m² g⁻¹ delivered larger discharge capacity of 280 mA h g⁻¹.²⁵ In Huang's report Li₂MnO₃ and Li₂MnO₃–LiMnO₂ were obtained by a one-step hydrothermal process; the Li₂MnO₃ and LiMnO₂ content varied with the employed amount of the oxidant.²⁶ However, studies on the synthesis of Li₂MnO₃ nanocrystals with continuous change in size have not been reported thus far.

The particle size of Li₂MnO₃ also influences the structure change in the charge/discharge process. Numerous studies have examined the atomic rearrangement of Li₂MnO₃ and Li₂MnO₃–LiMO₂ (M = Ni, Co) during activation, which is accompanied by changes in the Mn oxidation state and structural transformations to LiMnO₂ and LiMn₂O₄.^28–33 Wang observed the atomic rearrangement directly in Li₂MnO₃ particles and LiMnO₂ domains by STEM after electrochemical delithiation. In his study, Mn atoms were mobile and partially migrated to the Li layer.²⁹ Amalraj et al. found that the voltage plateau of 30 nm Li₂MnO₃ is 0.23 V lower than that of the micron-sized samples.³⁰ Boulineau et al. reported that Li deintercalation caused irreversible transformation to the spinel phase at the edge of the bulk material.³² The size of Li₂MnO₃ crystal hardly influences the structural change during charge/discharge process; however, rare reports studied the influence regularity of size change of Li₂MnO₃ nanocrystals on structure change.

The current article reports the preparation of ultrafine Li₂MnO₃ nanopowders with high crystallinity through a hydrothermal process. Introduction of KMnO₄ as a second oxidation agent allows for continuous size control of Li₂MnO₃ nanocrystals. Furthermore, the effect of grain size change on electrochemical properties is analyzed.

Experimental

Preparation of Li₂MnO₃nanocrystals

Li₂MnO₃ nanocrystals were prepared by hydrothermal method. MnSO₄, (NH₄)₂S₂O₈, LiOH, and KMnO₄ were purchased from Xilong Chemical Co. Ltd MnSO₄ (3 mmol) and (NH₄)₂S₂O₈ (2.4 mmol) were dissolved at a molar ratio of 1 : 1 in 10 mL of deionized water in a Teflon liner to obtain a mixed solution. Afterward, 20 mL of LiOH (3 mol L⁻¹) was trickled into the solution, which was stirred for 30 min to yield a δ-MnO₂ precursor precipitate.²⁵ The Teflon liner was transferred to an autoclave and maintained at 200 °C for 24 h. The as-prepared powders were filtered, washed, and dried at 80 °C for 24 h. To produce an oxidizing atmosphere during the precipitation and formation reactions, a determined amount of KMnO₄ was used as a raw material to substitute for MnSO₄. KMnO₄ was added at ratios of 0%, 3%, 6%, 7%, 9%, 11%, 14%, 16% and 20% relative to the total amount of the Mn source. The obtained samples were referred to as S₀, S₃, S₆, S₇, S₉, S₁₁, S₁₄, S₁₆ and S₂₀, respectively, according to the percentage of added KMnO₄.

Characterization of the Li₂MnO₃ products

The XPS Mn 3s region of the precursor was performed, using X-ray photoelectron spectroscopy (ESCALAB250Xi). XRD measurements were carried out using a Rigaku D/max-2600 PC with Cu Kα radiation (λ = 0.15406 nm). Grain sizes (D) were calculated by Scherrer formula (D = Kλ/B cos θ, K, B and θ were Scherrer constant, full width at half maximum and the incident angle). Particle morphologies were probed by SEM (S-4800) and TEM (JEM-2100F). Raman spectroscopy (InVia-Reflex) was carried out to analyze the microstructure of Li₂MnO₃ samples. Specific surface areas were measured using a specific surface area and pore size analyzer (ASAP 2020 M). Grain sizes (D) were calculated by specific surface areas (D = 6/BET × ρ, ρ was the theoretical density of a unit cell). Mn K-edge X-ray absorption spectra were obtained using a BL14W1 beamline at the Shanghai Synchrotron Radiation Facility.

Electrochemical measurement

The cathode was prepared by thoroughly mixing the active material, super P carbon black, and a polyvinylidene fluoride binder at a weight ratio of 8 : 1 : 1. Coin-type battery cells containing a cathode, a Li metal anode, and a microporous polyethylene separator were prepared in an argon-filled glove-box. Electrochemical measurements were conducted with button-type cells using 1.2 M LiPF₆ with ethyl methyl carbonate/ethylene carbonate (7 : 3, vol%) as electrolyte. The coin-type cells were tested at a constant current density of 20 mA g⁻¹ using a Land battery test system. CV curves were performed by Princeton 273A (0.1 mV s⁻¹, 2–4.8 V).

Results and discussion

Formation of Li₂MnO₃ nanocrystals

The phase compositions of the products with varying KMnO₄ contents were analyzed by XRD (Fig. 1). The XRD patterns of the samples showed a single Li₂MnO₃ phase when the KMnO₄ content exceeded 6% (i.e., S₆, S₉, S₁₁, and S₂₀). Without KMnO₄, a main phase of Li₂MnO₃ with minor O-LiMnO₂ (JPCDS card no. 72-0411) was observed (S₀). These results indicated that KMnO₄ enhanced the formation of a single Li₂MnO₃ phase. The diffraction lines of Li₂MnO₃ were indexed to a monoclinic unit cell with a space group symmetry of C2/m (JPCDS card no. 84-1634). The diffraction lines showed decreased density and increased broadness with increasing KMnO₄ content because of the decreases in the grain size of the Li₂MnO₃ products.

Fig. 1 XRD patterns of the obtained samples of (a) for S₀, (b) for S₆, (c) for S₉, (d) for S₁₁ and (e) for S₂₀.

(NH₄)₂S₂O₈ was used as an oxidant to synthesize the δ-MnO₂ precursor at room temperature. Mn valence of the precursor was +3.57, calculated from the XPS Mn 3s region in Fig. S1 of the ESI.† ²⁷ Mn³⁺ precipitate led to LiMnO₂ in the hydrothermal process. The XRD patterns of C₁–C₄ samples (comparison experiment without (NH₄)₂S₂O₈) showed pure Li₂MnO₃ phase only when the average Mn valence exceeded +4 in Fig. S2.† KMnO₄, which took effect in the hydrothermal process, raised average Mn valence of the precursor and produced a weak oxidizing environment. While average Mn valence of the precursor was close to tetravalence, LiMnO₂ phase disappeared. A small amount of KMnO₄ could achieve pure Li₂MnO₃; proper excess oxidant did not affect phase purification.

The image in Fig. 2 illustrated the Raman spectra of the samples. The Raman spectrum of S₆ exhibited four significant bands at 369, 435, 487, and 635 cm⁻¹. These bands belonged to Mn–O bonding of Li₂MnO₃ and in agreement with previous reports.^28,29 The bands were assigned to phonon vibrations, and the peak at 635 cm⁻¹ was attributed to the A_1g mode of Mn–O bonding. As the KMnO₄ ratio increased, the A_1g peak position remained essentially the same but showed broadening.

Fig. 2 Raman spectra of the obtained samples of (a) for S₆, (b) for S₉, (c) for S₁₁, and (d) for S₂₀.

Morphology of the Li₂MnO₃ products

The morphologies of the Li₂MnO₃ products with different KMnO₄ contents were observed by SEM and TEM (Fig. 3). The addition of KMnO₄ resulted in decreased grain sizes and gradual transformation from plates to rods. S₀ and S₆ [Fig. 3(a) and (b)] revealed square and hexagonal nanoplates, respectively, with well-distributed particle sizes. The sharp morphologies showed high crystallinity. Nanorods and nanoplates coexisted in S₉ [Fig. 3(c)]. Increasing the KMnO₄ ratios to 11% and 20% (S₁₁ and S₂₀) resulted in further decreases in particle size, and the morphologies substantially transformed from nanoplates to nanorods [Fig. 3(d) and (e) and 3(f) and (g), respectively]. The addition of KMnO₄ decreased the size of the products and affected their morphology. By contrast, the samples without (NH₄)₂S₂O₈ did not show similar change while the average Mn valence varied from 3.5 to 4.25 in Fig. S3.†

Fig. 3 SEM images of (a) for S₀, (b) for S₆, (c) for S₉, (d) for S₁₁ and (f) for S₂₀, and TEM images of (e) for S₁₁, (g) for S₂₀.

The orientation of Li₂MnO₃ products with different sharpness was observed by high-resolution TEM images (Fig. 4). S₆ showed square crystals with sharp corners and edges, whereas S₁₁ showed unclear edges and reductions in size. Ten fringe spacing of the two crystal faces of S₆ of approximately 4.74 and 2.07 nm were observed, corresponding to the (001) and (022) planes of Li₂MnO₃, respectively. These planes were indexed to the [100] axis. Spacing of the clear crystal face around 0.47 nm were observed for S₁₁, corresponding to the (001) plane, in Fig. 4(b). Thus, the c-axis lay along the thickness direction. S₆ and S₁₁ samples might experience different growth processes.

Fig. 4 TEM images of (a) for S₆ and (b) for S₁₁.

Specific surface areas and particle sizes

Table 1 summarized the morphologies, BET surface area, grain sizes derived from XRD Scherrer equation and BET calculation and the first discharge capacities of S₆–S₂₀ samples. Specific surface areas gradually increased from 80.8 m² g⁻¹ to 81.4 m² g⁻¹ for S₆ with the addition of 6% KMnO₄, and instantaneously to 150.2 m² g⁻¹ for S₉ with the addition of 9% KMnO₄. The specific surface areas maintained an approximate linear increase with increasing amount of added KMnO₄ until 236.2 m² g⁻¹ for S₂₀.

Table 1 Morphologies, specific surface areas, grain sizes and discharge capacities of the samples

Sample	S0	S6	S9	S11	S20
Morphology	Plate	Plate	Plate and rod	Plate and rod	Rod
Specific surface area (m^2 g^−1)	80.8	81.4	150.2	191.5	236.2
Grain sizes from BET (nm)	28.9	28.7	15.6	12.2	9.8
Grain sizes from XRD (nm)	33	30	21	10	5
Discharge capacity (mA h g^−1)		196	229	230	234

---

Grain sizes calculated from the (001) plane of the XRD patterns by Scherrer formula and from BET surface areas are shown in Fig. 5. The results obtained from these two methods were fairly similar. Changes in size with the amount of added KMnO₄ could be classified as follows: KMnO₄ ≤ 6%, 6% < KMnO₄ ≤ 11%, and KMnO₄ > 11%. Particle sizes displayed minor changes at KMnO₄ contents of ≤6% and ≥11% areas, whereas significant decreases were observed at KMnO₄ contents of 6% < KMnO₄ < 11%. Grain sizes of around 28.7 nm with only nanoplate shape were obtained at KMnO₄ ≤ 6%, whereas those around 9.8 nm with mainly nanorods were presented at KMnO₄ > 11%. KMnO₄ addition evidently resulted in continuous decreases in size between 28.7 and 12.2 nm at 6% < KMnO₄ ≤ 11%. These results indicated that growth mechanism varied among these samples. The growth mechanism of Li₂MnO₃ nanocrystals might be oriented attachment.³³ The amount of crystal nucleus controlled changes in grain size, morphology, and orientation. KMnO₄, as a Mn source for Li₂MnO₃, accelerated the dissolution of the intermediate phase and enhanced amount of crystal nucleus of the samples. KMnO₄ contents between 6% and 11% enabled rapid increases in the amount of crystal nucleus of the samples. At KMnO₄ ≥ 11%, the amount of crystal nucleus reached a certain limit, such that grain size hardly changed thereafter.

Fig. 5 Grain size of the obtained samples calculated from Scherrer formula and BET surface area.

In conclusion, adding the amount of KMnO₄ raised average Mn valence of the precursor (be close to tetravalence) and introduced a weak oxidizing environment. Thus trivalent Mn compound disappeared. O₂ decomposed from KMnO₄, accelerated the dissolution of the intermediate phase and enhanced amount of crystal nucleus, which led to the decreasing grain sizes and the morphology change from nanoplates to nanorods. Increasing average Mn valence of precursor gave rise to the morphology transformation from nanoplates to nanorods and larger discharge capacities. Meanwhile (NH₄)₂S₂O₈ was significant for the δ-MnO₂ precursor, without which grain sizes could not be controlled (shown in Fig. S3†).

Effects of particle size on the electrochemical properties of Li₂MnO₃ nanocrystals

The image in Fig. 6(a) demonstrated the first charge/discharge voltage profiles of S₆, S₉, S₁₁, and S₂₀ at a constant current density of 20 mA g⁻¹. The first discharge capacities were 196, 229, 230, and 234 mA h g⁻¹ for S₆, S₉, S₁₁, and S₂₀, respectively. S₆, with a size of 28.7 nm, showed a relatively low capacity of 196 mA h g⁻¹. Capacity considerably increased to 229 mA h g⁻¹ for S₉ with 15.6 nm. Then, the capacities showed small change for S₁₁ and S₂₀, with 12.2 and 9.8 nm, respectively. The capacities showed a linear change with the change in grain size in Fig. 6(b). First-cycle coulombic efficiencies of S₆–S₂₀ samples were 57.5%, 69.1%, 62.3% and 62.2% respectively, indicating approximate 40% Li⁺ could not insert into the original position. Minor improvement of the initial coulombic efficiencies was related to the grain size decreases. Meanwhile, decreasing grain sizes led to larger initial discharge capacities and lower capacity retentions after 50 cycles in Fig. S5.† Electrodes charged to 4.3 V, 4.6 V and 4.8 V and discharged to 3 V and 2 V were selected for analyzing the phase transformation in the first cycle.

Fig. 6 First charge and discharge curves at a rate of 20 mA g⁻¹ (a), the relation between grain size and discharge capacity of the S₆, S₉, S₁₁ and S₂₀ samples (b).

The downtrend of voltage plateaus were relevant with the decreasing grain sizes. To determine the voltage plateaus, corresponding dQ/dV curves were calculated, which were displayed in Fig. 7. Li₂O extraction peaks shifted to lower voltages with decreasing grain sizes. S₆, S₉, and S₁₁ showed oxidation peaks at 4.55, 4.4, and 4.4 V respectively, corresponding to Li₂O extraction from Li₂MnO₃. By contrast, S₂₀ demonstrated two peaks at 3.8 and 4.4 V. The new peak at 3.8 V might result from Li extraction of the outer structure in Li₂MnO₃ crystals. Stacking faults on the surface could provide electrochemical reaction points and facilitate Li extraction from outer structures.^30,35 Reorganization of the outer structure subsequently occurred because of oxygen evolution, followed by Li deintercalation of inner Li₂MnO₃. In addition, the evident peaks at 4.7 V to 4.8 V might be ascribed to interfacial reactions. It might result in the poor cycle performance in the samples with smaller grain sizes in Fig. S5.† The discharge curves of S₆, S₉, S₁₁, and S₂₀ showed reduction peaks at 3.05 V, which were identified as Mn³⁺–Mn²⁺ reactions.

Fig. 7 dQ/dV of charge and discharge curves of the samples of (a) for S₆, (b) for S₉, (c) for S₁₁, and (d) for S₂₀.

The charge and discharge curves of the second cycle demonstrated minor difference with each other in Fig. S4.† Corresponding dQ/dV curves were similar in all samples. The redox peaks appeared at around 4 V in the second cycle for all samples, being identified as Mn⁴⁺–Mn³⁺ reactions of the spinel phase. In all samples, the redox peaks at 3 V shifted to lower values, implying the voltage drop. Furthermore, the samples with smaller particle size demonstrated less voltage fade in the 3 V regime.

Evidently, larger surface area brought about lower voltage plateaus. Samples with more defects exhibited higher energies and decreased the potential barrier for Li₂O extraction, thereby resulting in voltage plateau drops.³⁴ The strong redox peak located at 4.55 V corresponded to 28.7 nm Li₂MnO₃ nanocrystals. Samples with 15.6 nm to 12.2 nm size showed voltage plateaus at 4.4 V. Two oxidation peaks at 3.8 and 4.4 V appeared in the 9.8 nm sample. The above conclusion was mainly consistent with the CV curves in Fig. S6 of the ESI.†

Effects of size on structural changes in the Li₂MnO₃ nanocrystals during charging/discharging

Representative S₆ and S₁₁ electrodes were obtained to analyze structural evolution using a series of XRD profiles. These electrodes were initially charged and discharged at different potentials, and results were shown in Fig. 8(a) and (b). The diffraction peaks of S₆ and S₁₁ at various voltages matched with monoclinic Li₂MnO₃ (C2/m) well. Charging S₆ and S₁₁ to 4.6 V led to the gradual disappearance of superlattice peaks around 2θ = 22°. About 67% Li ions were removed from Li₂MnO₃, and no other phases appeared at the 4.6 V charge state of S₁₁ samples. LiMnO₂ formed with further charging from 4.6 V to 4.8 V. According to the obtained dQ/dV curves, strong interfacial reactions were observed during charging from 4.6 V to 4.8 V in S₁₁. During this process, LiMnO₂ was formed because of interfacial reactions and electrolyte erosion. The spinel phase generated during the initial discharge. The (440) plane of the spinel phase was observed at 3 and 2 V of the first discharge.

Fig. 8 XRD patterns of the samples of S₆ (a) and S₁₁ (b), magnified (−202) peak of S₆ and S₁₁ (c), and the change of FWHM₍₋₂₀₂₎ (d) after the initial discharge.

The image in Fig. 8(c) showed the fitted peak near 45°. Here, the (440) spinel peaks [I_(sp440)] of the S₆ electrodes were stronger than those of S₁₁ electrodes. This characteristic indicates fewer spinel phases in S₁₁ electrodes, which might be related to particle size. Li extraction and reorganization at low potential prevented phase transformation in S₁₁.

In both S₆ and S₁₁, initial charging to 4.8 V led to a broad (001) peak and a sharp (−202) peak, which gradually recovered after discharge. The full width at half-maximum (FWHM) of the (−202) peak [FWHM₍₋₂₀₂₎] was shown at various points of initial charging and discharging [Fig. 8(d)]. This phenomenon might involve microstress during Li deintercalation. Stress of Li removal engendered the migration of transition metal ions into the Li layers and the formation of LiMn₂O₄.³¹ However, whether microstress affected phase transformation in the cycles required further study.

The image in Fig. 9 showed the Mn K-edge XANES spectra of S₆ and S₁₁. Absorption peaks at 6560 and 6575 eV were assigned to 1s to 3d (t_eg) and 1s to 3d (e_g) transitions, respectively. The edges of pristine S₆ and S₁₁ electrodes only partially overlapped, although their absorption peaks were both at 6561 eV. The intermediate bump at 6552 eV that was observed in S₆ disappeared in S₁₁ samples, being described as the disordered arrangement of MnO₆ octahedra in this sample, which was due to stacking defects.²⁰ Higher peak intensities near the edges of S₁₁ indicated that Mn was in a high-energy state. Therefore, more structure distortion and stacking faults were found in S₁₁ samples. Charging S₆ and S₁₁ to 4.8 V caused the edges shifting toward low energies, implying evident Mn³⁺ compound formation. According to the XRD patterns, the result indicated that only minor Mn⁴⁺ amounts were transformed into LiMnO₂ after charging to 4.8 V in both S₆ and S₁₁ samples.

Fig. 9 Mn K-edge XANES spectra of the pristine simple and after charging to 4.8 V.

Conclusions

In this study, pure Li₂MnO₃ nanocrystals are synthesized via hydrothermal method. By adding KMnO₄ as oxidation in the reaction system, pure Li₂MnO₃ nanocrystals with controlled size and morphology are obtained. The particle sizes remarkably influence the electrochemical performances of the nanocrystals. The discharge capacities decrease with decreasing size of Li₂MnO₃ nanocrystals in a linear relationship. The plateau voltage of the first charge decreases from 4.5 V to 4.4 V with the decrease in Li₂MnO₃ nanocrystals from 28.7 nm to 12.2 nm and splits to two plateaus around 3.8 and 4.4 V for nanocrystals of 9.8 nm size. Li₂MnO₃ crystals with smaller sizes are more beneficial for maintaining a layered structure than larger nanocrystals.

Acknowledgements

This work was supported by the National Nature Science Foundation of China (NSFC No. 51372114), the Research Fund of State Key Laboratory of Mechanics and Control of Mechanical Structures (Nanjing University of Aeronautics and astronautics) (Grant No. 0514Y01), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Notes and references

1. J. M. Tarascon and M. Armand, Nature, 2001, 414, 359.
2. M. Armand and J. M. Tarascon, Nature, 2008, 451, 652.
3. V. Etacheri, R. Marom, R. Elazari, G. Salitra and D. Aurbach, Energy Environ. Sci., 2011, 4, 3243.
4. H. J. Yu and H. S. Zhou, J. Phys. Chem. Lett., 2013, 4, 1268.
5. M. M. Thackeray, S. H. Kang, C. S. Johnson, J. T. Vaughey, R. Benedek and S. A. Hackney, J. Mater. Chem., 2007, 30, 3053.
6. Y. Koyama, N. Yabuuchi, I. Tanaka, H. Adachi and T. J. Ozhuku, J. Electrochem. Soc., 2005, 152, A1434.
7. A. Boulineau, L. Croguennec, C. Delmas and F. Weill, Chem. Mater., 2009, 21, 4216.
8. J. Bareno, C. H. Lei, J. G. Wen, S. H. Kang, I. Petrov and D. P. Abraham, Adv. Mater., 2010, 22, 1122.
9. C. S. Johnson, S. D. Korte, J. T. Vaughey, M. M. Thacheray, T. E. Bofinger, Y. Shao-Horn and S. A. Hachney, J. Power Sources, 1999, 81, 491.
10. S. H. Park, Y. Sato, J. K. Kim and Y. S. Lee, Mater. Chem. Phys., 2007, 102, 225.
11. Z. Lu, Z. Chen and J. R. Dahn, Chem. Mater., 2003, 15, 3214.
12. J. Jiang, K. W. Eberman, L. J. Krause and J. R. Dahn, J. Electrochem. Soc., 2005, 152, A1874.
13. A. R. Armstrong, M. Holzapfel, P. Novák, C. S. Johnson, S. H. Kang, M. M. Thackeray and B. G. Bruce, J. Am. Chem. Soc., 2006, 128, 8694.
14. A. D. Robertson and P. G. Bruce, Chem. Commun., 2002, 2790.
15. A. R. Armstrong, A. D. Robertson and P. G. Bruce, J. Power Sources, 2005, 146, 275.
16. W. Tang, H. Kanoh, X. Yang and K. Ooi, Chem. Mater., 2000, 12, 3291.
17. J. K. Ngala, S. Alia, A. Dobley, V. M. B. Crisostomo and S. L. Suib, Chem. Mater., 2007, 19, 229.
18. G. Jain, J. Yang, M. Balasubramanian and J. J. Xu, Chem. Mater., 2005, 17, 3850.
19. P. Lanz, C. Villevieille and P. Novák, Electrochim. Acta, 2013, 109, 426.
20. G. Z. Wei, X. Lu, F. S. Ke, L. Huang, J. T. Li, Z. X. Wang, Z. Y. Zhou and S. G. Sun, Adv. Mater., 2010, 22, 4364.
21. J. Cho, Y. Kim and M. G. Kim, J. Phys. Chem. C, 2007, 111, 3192.
22. Y. Lee, M. G. Kim and J. Cho, Nano Lett., 2008, 8, 957.
23. A. Choi, K. Palanisamy, Y. Kim, J. Yoon, J. H. Park, S. W. Lee, W. S. Yoon and K. B. Kim, J. Alloys Compd., 2014, 591, 356.
24. M. Tabuchi, Y. Nabeshima, T. Takeuchi, K. Tatsumi, J. Imaizumi and Y. Nitta, J. Power Sources, 2010, 195, 834.
25. J. Y. Baek, H. W. Ha, I. Y. Kim and S. J. Hwang, J. Phys. Chem. C, 2009, 113, 17392.
26. X. Huang, Q. Zhang, H. Chang, J. Gan, H. Yue and Y. Yang, J. Electrochem. Soc., 2009, 153, 162.
27. M. G. Verde, H. Liu, K. J. Carroll, L. Baggetto, G. M. Veith and Y. S. Meng, ACS Appl. Mater. Interfaces, 2014, 6, 18868.
28. W. Fang, H. Kanoh, K. Ooi and Y. Wang, J. Mater. Sci. Lett., 2000, 19, 1361.
29. R. Wang, X. Q. He, L. H. He, F. W. Wang, R. J. Xiao, L. Gu, H. Li and L. Q. Chen, Adv. Eng. Mater., 2013, 3, 1358.
30. F. Amalraj, D. Sharon, M. Talianker, C. M. Julien, L. Burlaka, R. Lavi, E. Zhecheva, B. Markovsky, E. Zinigrad, D. Kovacheva, R. Stoyanova and D. Aurbach, Electrochim. Acta, 2013, 97, 259.
31. F. Amalraj, L. Burlaka, C. M. Julien, A. Mauger, D. Kovacheva, M. Talianker, B. Markovsky and D. Aurbach, Electrochim. Acta, 2014, 123, 395.
32. A. Boulineau, L. Simonin, J. F. Colin, E. Canévet, L. Daniel and S. Patoux, Chem. Mater., 2012, 24, 3558.
33. R. L. Penn and J. F. Banfield, Science, 1998, 281, 969.
34. Y. Okamoto, J. Electrochem. Soc., 2012, 159, A152.
35. D. Y. W. Yu, K. Yanagida, Y. Kato and H. Nakamura, J. Electrochem. Soc., 2009, 156, A417.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra08387g

---