A simple, rapid and efficient flame synthesis of ultrafine LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30): a versatile route to synthesize Ni-rich cathode materials for Li-ion batteries

Abstract

A simple, rapid, and efficient flame synthesis method was developed to fabricate LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30) cathode materials for Li-ion batteries. The ignition of metal nitrate precursors with a reddish flame was completed within 15–20 s. The complete procedure to obtain the LiNi_0.95−xCo_xTi_0.05O₂ precursor powder took 30–55 min, which is the simplest as well as the most rapid and efficient method among the synthesis methods reported thus far to obtain an analogous Ni-rich, LiNi_0.70Co_0.30O₂ cathode material. X-ray diffraction confirmed the formation of a single phase of LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30) calcined at 800 °C for 12 h. Scanning electron microscopy (SEM) and transmission electron microscopy revealed the ultrafine nature of the particles. SEM showed grains with smooth surfaces and size in the range of 200 nm–0.4 μm, and TEM displayed the particles with a size range of 90 nm–0.2 μm. The fabricated materials showed discharge specific capacities of 159 mA h g⁻¹ and 137 mA h g⁻¹ at a 0.1C rate for the initial cycle at room temperature for LiNi_0.95−xCo_xTi_0.05O₂ with x = 0.25 and 0.30, respectively. These results suggest that LiNi_0.95−xCo_xTi_0.05O₂ synthesized via this facile route shows promising electrochemical activity. Consequently, the flame synthesis technique can be very attractive for the facile fabrication of structurally similar lithiated nickel–cobalt oxides for use as a cathode material for Li-ion batteries that are suitable for mass production.

---

1. Introduction

Ni-rich, Li_1−xNi_1−yM_yO₂ (Co, Mn, Al etc.), systems are the most promising candidates for cathode materials with strong potential for commercial applications in advanced lithium-ion batteries for electric vehicles (EVs) and hybrid electric vehicles (HEVs).¹ In particular, the LiNi_1−xCo_xO₂ system has advantages over the pure isostructural layered oxide end member of LiCoO₂ and LiNiO₂ in terms of its structural and thermal stability, enhanced electrochemical properties with higher capacity, and better cyclability. Of compounds in the LiNi_1−xCo_xO₂ system, LiNi_0.7Co_0.3O₂ exhibits excellent capacity and good reversibility, but this solid solution shows poor intrinsic electronic conductivity, which contributes to diminished electrochemical performance during cycling. The electrochemical performance of cathode materials is strongly dependent on the fabrication process of the materials. Traditional solid-state techniques are generally used to fabricate lithiated nickel–cobalt oxides with the starting materials, Li₂CO₃ or LiOH·H₂O, NiO or NiCO₃, and Co₃O₄ or CoCO₃, with a stoichiometric amount calcined for a long duration with several intermediate grinding steps.^2,3 On the other hand, wet chemical synthesis provides a better option for the fabrication of lithiated transition-metal oxides because ultrafine powders with a small particle size of complex compositions with high purity and homogeneity can be prepared easily compared to solid-state techniques. In addition, ultrafine powders with small particle sizes and the reduction of Li/Ni disorder are considered an advantage for improving the electrochemical performance of LiNi_1−yCo_yO₂ systems.⁴ Various routes for the synthesis of lithiated nickel–cobalt oxides have been reported, such as the sol–gel method,^5–8 the co-precipitation method,^9,10 and the combustion method.^11–14 Among them, the combustion method for the synthesis of nanosized or ultrafine oxide materials has attracted considerable interest.¹⁵ The combustion method involves the use of metal nitrate together with the corresponding complexing agent, such as urea, hydrazine, citric, acetic acid, glycine, tartaric, or glycerol in aqueous medium, which results in complexation of the metal by poly-carboxylic acid to produce optimized combustion synthesis and requires a precalcination step for the intermediate decomposition and removal of excess organic precursors.^16,17 This article provides an effective method to fabricate LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30) materials with their spectroscopic characterization. In this proposed method, there is no need to add the complexing agents as a fuel for metal nitrate precursors of Li, Ni, and Co. Metal nitrates are soluble in 2-methoxyethanol, which allows the auto ignition of metal nitrates with a reddish flame to form the desired precursor powder within 15–20 s; a single phase of the material is obtained in a single calcination step. The whole process to obtain the LiNi_0.95−xCo_xTi_0.05O₂ precursor powder was completed within 30–55 minutes. The present method is the simplest synthesis technique to obtain the LiNi_0.95−xCo_xTi_0.05O₂ powder as well as the Ni-rich isostructural compounds of LiNi_1−yCo_yO₂ in comparison to other synthetic methods (see the ESI,† S1). In addition to the processing window, a new concept was also used to substitute a dopant at different sites for the same material, which allowed better substitution to enhance the electrochemical performance of the LiNi_0.7Co_0.3O₂ cathode material. Herein, Ti⁴⁺ ions were substituted at the Ni-site and the Co-site in LiNi_0.7Co_0.3O₂, and are abbreviated as LNTCO and LNCTO, respectively. Surprisingly, LNCTO exhibits better electrochemical performance by substitution of Ti⁴⁺ ions at the Co-site compared to that at the Ni-site. A comparison of the substitution of Ni and Co sites was made to examine the relationship between the substitution site and the electrochemical performance.

2. Methods

2.1 Material synthesis

LiNO₃ (98.0%, Junsei), Ni(NO₃)₂·6H₂O (98.50%, Aldrich), Co(NO₃)₂·6H₂O (98.0%, Junsei), TiO₂ (99.9%, Sigma Aldrich), and 2-methoxyethanol (99.0%, Alfa Aesar) were used to prepare LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30) precursor powders. First, an appropriate amount of LiNO₃, Ni(NO₃)₂·6H₂O, and Co(NO₃)₂·6H₂O were dissolved in a minimal amount of 2-methoxyethanol. Second, stoichiometric amounts of solid TiO₂ were added to the solution. The resulting heterogeneous mixture was sonicated for 10–15 min and heated on a hot plate using a magnetic stirrer at 100–120 °C to evaporate the organic solvent, until self-ignition took place, which exhausted a large amount of gases with a reddish flame and produced a fluffy mass of the LiNi_0.95−xCo_xTi_0.05O₂ precursor powder. This synthetic technique involves the mixing of metal nitrate in a non-aqueous solvent of 2-methoxyethanol at the ionic level except TiO₂. The organic solvent of 2-methoxyethanol acts as a fuel to induce a highly exothermal flame which can instantly provide ultrahigh thermal energy for the precursors of metal nitrates. They are decomposed in this environment to produce the desirable product. Finally, the resulting powder was calcined at 800 °C for 12 h in an O₂ atmosphere in a tubular electrical furnace. Fig. 1 presents a schematic diagram of the flame synthesis technique to obtain LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30). A video of the flame synthesis technique taken during the flame reaction is available in the ESI,† S2.

Fig. 1 Schematic diagram of the flame synthetic technique to prepare LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30).

2.2 Structural characterization

X-ray diffraction (XRD, Rigaku Ultima IV, Japan) with Cu K_α radiation was used to identify the phases of the synthesized material. Raman spectra (Ando Tech Sr-3031-A Raman spectrophotometer, Japan) were obtained for the qualitative structural analysis of the sample powder. The powders were analyzed in the range 400 up to 600 cm⁻¹ using the Rayleigh wavelength (λ = 784.5 nm) with a laser power of 16.5 mW as the excitation source with an integration time of 180 s.

⁷Li magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectra werecollected on a Bruker Avance III 300 with a 7.04 T magnet at room temperature. For the MAS NMR experiments, a 2.5 mm MAS probe was used with a zirconia rotor at 116.64 MHz as the resonance frequency of ⁷Li. The ⁷Li NMR spectra were referenced to the external 1 M LiCl. A single pulse for ⁷Li was used to produce the NMR signal. The sample spinning rate was 25 kHz. The ⁷Li spectra were acquired with a 90° pulse length of 1.4 μs, a repetition delay of 0.5 s, 256 transients, and a spectral width of 0.23 MHz. The isotropic chemical shifts were identified by varying the spinning rates.

X-ray photoelectron spectroscopy (XPS, K-Alpha, Thermo Scientific, USA) of the materials was performed in wide scan survey mode and high-energy resolution using Al K_α radiation (1486.6 eV). The morphology of the prepared materials was observed by field emission scanning electron microscopy (SEM, Jeol JSM6500F, Japan). The particle size and elemental mapping of the sample were analyzed using a transmission electron microscope (TEM, JEM-2100F) with an Oxford energy dispersive X-ray (EDX) analyzer attached to it. For TEM measurements, samples were dispersed in an alcohol and then the drops of dispersed samples were placed on the carbon-coated copper grid.

2.3 Electrochemical tests

The electrodes used to measure the electrochemical performance were prepared by slurry casting and pressing a mixture of 80 wt% of the synthesized cathode material, 10 wt% of carbon black (Super P), 10 wt% of polyvinylidene fluoride (PVDF, Aldrich), and 3 mL of N-methylpyrrolidone (NMP) on aluminum foil followed by drying in a vacuum oven at 110 °C for 2 h. Lithium metal foil, 15 mm in diameter and 0.30 mm in thickness, was used as the counter electrode. The circular disk electrodes with Cellguard 2400 separators were assembled into 2016 coin cells in an argon-filled glove box. Non-aqueous 1.15 M LiPF₆ in ethylene carbonate/dimethyl carbonate/diethyl carbonate with a 3 : 4 : 3 volume ratio was used as the electrolyte. The electrochemical cycle tests and cyclic voltammetry were performed at ambient temperature using a galvanostatic automatic battery cycler (WonATech WBCS3000, Korea). The cells for the rate test were cycled between 2.5 V and 4.5 V vs. Li/Li⁺ with a constant current–constant voltage (CC–CV) protocol in charging mode. Constant current rates at 0.1, 0.2, 0.5, 1.0, 2.0, and 3.0C were used. Cyclic voltammetry (CV) at room temperature for all samples was performed at a scan rate of 0.1 mV s⁻¹ over the voltage range, 2.5–4.5 V, for 5 h.

3. Results and discussion

Fig. 2 shows the XRD patterns of the prepared LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30), which are abbreviated as LNCTO and LNTCO, respectively. XRD revealed a highly crystalline and single phase with a hexagonal structure without a secondary phase. The XRD patterns clearly show the splitting of (006)/(102) and (108)/(110) as doublets, which exhibit a highly ordered layered structure with a homogeneous distribution of elements. The lattice parameters of LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30) were calculated using ‘cell software’ and the values are listed in Table 1. The results show that Ti substitution does not change the layered structure. However, it causes a minute increase in the unit cell volume of LiNi_0.70Co_0.25Ti_0.05O₂ than LiNi_0.70Co_0.25Ti_0.05O₂ due to the large ionic radii difference of the substituent in the same coordination number of 6. The intensity ratios of the XRD peaks, I₍₀₀₃₎/I₍₁₀₄₎, are considered an indicator of the ordering of lithium and other transition metal cations (Ni and Co). The I₍₀₀₃₎/I₍₁₀₄₎ of LNCTO was 1.57 larger than that of LNTCO (1.32), which indicates that the substitution of Ti⁴⁺ ions at the Co-site would suppress the occupancy of Li⁺ layers by Ni²⁺ and increase the conductivity, leading to better electrochemical performance compared to that of LNTCO, as discussed later.¹⁸

Fig. 2 XRD patterns of flame synthesized (a) LNCTO and (b) LNTCO.

Table 1 Lattice parameters and unit cell volume of LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30)

Sample	Lattice parameter (a) (Å)	Lattice parameter (c) (Å)	Lattice volume (Å^3)	Intensity ratio (003)/(104)
LNCTO	2.778	14.543	97.256	1.57
LNTCO	2.775	14.569	97.203	1.32

---

The phase and qualitative structural analyses of the LNCTO and LNTCO cathode materials were analyzed by Raman spectroscopy, as shown in Fig. 3. The spectra revealed Raman-active vibrational modes (A1_g and E_g), which are generally observed in a rock salt-layered structure with R3̄m group symmetry.¹⁹ Two well-defined bands were observed at 481 and 592 cm⁻¹, which are ascribed to the E_g (O–Co–O bending) and A1_g (Co–O stretching) modes,²⁰ respectively. This spectral result confirmed the formation of a rock salt-type structure, which is in good agreement with the previous studies of LiNi_0.7Co_0.3O₂.^21,22

Fig. 3 Raman spectra of flame synthesized (a) LNCTO and (b) LNTCO.

⁷Li magic angle spinning (MAS) nuclear magnetic resonance (NMR) experiments were carried out for further structural analysis of the LNCTO and LNTCO cathode materials, as shown in Fig. 4. A single isotropic resonance peak was observed at 0 ppm with broad spinning sideband manifolds marked as asterisks. The characteristic single peak at 0 ppm was observed, which is shown as a typical diamagnetic environment for Li atoms.²³ Therefore, the appearance of a single isotropic peak at 0 ppm in both cathode materials confirmed the presence of a single type of Li site that was well matched with the reported literature.^24,25

Fig. 4 ⁷Li MAS NMR spectra of flame synthesized (a) LNCTO and (b) LNTCO.

Fig. 5 shows the fitted results of XPS spectra for the synthesized LNCTO to provide information on the valence states of the elements. The spectrum of LNCTO revealed the presence of Li, Ni, Co, Ti, and O, respectively, which confirms the purity of the prepared material via flame synthesis. Fig. 5(a) shows the Li 1s peak located at 54.7 eV, which can be attributed to the presence of Li⁺ species in LNCTO.²⁶ The Ni 2p spectrum in Fig. 5(b) shows two well-defined peaks with binding energies of 855.1 and 872.8 eV, which were assigned to 2p_3/2 and 2p_1/2, respectively. This is consistent with the energy characteristics of the Ni valence state for Ni²⁺ in NiO²⁷ and Ni³⁺ in LiNiO₂,²⁸ respectively. Therefore, LNCTO revealed the existence of both Ni³⁺ and Ni²⁺ ions.

Fig. 5 XPS spectra of flame synthesized LNCTO for (a) Li 1s, (b) Ni 2p, (c) Co 2p, (d) Ti 2p, and (e) O 1s.

In contrast, the Co 2p spectrum in Fig. 5(c) was also characterized by two peaks at 779.9 and 794.9 eV for 2p_3/2 and 2p_1/2, respectively. Co ions in an oxygen environment represent the +3 oxidation state if the main peak positions are located around 780 and 795 eV.^29,30 The Ti 2p spectrum in Fig. 5(d) exhibited double peaks of Ti 2p_3/2 and Ti 2p_1/2 located at 457.7 eV and 463.8 eV, respectively, which are characteristic of the +4 oxidation state for Ti⁴⁺.^31,32 The O 1s spectrum in Fig. 5(e) exhibits a sharp peak centered at 531.3 eV, which suggests oxygen linked to metal ions as Li–O along with the Ni/Co–O for the layered structure cathode materials.³³ The observed binding energy suggests that the valence states of Ni, Co, and Ti are mainly +2/+3, +3, and +4, respectively, which are consistent with previous results.^26–32

Fig. 6(a) and (b) show SEM images of the LNCTO and LNTCO powders. The SEM images show the similar nature of morphology for both samples with the slight agglomeration of grains in LNTCO. SEM also revealed grains with smooth surfaces and particle sizes in the range of 200 nm–0.4 μm. The microstructure and surface morphology of the LNCTO and LNTCO materials were further characterized by TEM, as shown in Fig. 6(c) and (d). TEM revealed the ultrafine nature of the particles with a size range of 90 nm–0.2 μm, which is consistent with the SEM observations.

Fig. 6 FE-SEM images of flame synthesized (a) LNTCO and (b) LNCTO, and HR-TEM images of (c) LNTCO and (d) LNCTO.

Fig. 7 shows the EDX mapping images of LNCTO to determine the distribution of each element presented in the flame synthesized cathode materials. The EDX images of the LNCTO cathode showed a uniform distribution of the Ni, Co, Ti and O elements, as shown in Fig. 7. The EDX mapping image of Ti in Fig. 7(c) clearly showed a similar intensity and distribution to Ni, Co, and O, which indicates that all the elements are homogeneously mixed in the material. Although the percentage of Ti substitution in sample preparation was very low, the Ti in the sample was distributed uniformly, which highlights the merits of the suggested flame synthetic method. The purpose of this work was to develop a simple, rapid, and efficient chemistry-based synthetic method to prepare analogous materials of LiNi_0.70Co_0.30O₂. To simplify the procedure, inexpensive solid TiO₂ powder was also used as the titanium source to substitute Ti⁴⁺ in LiNi_0.70Co_0.30O₂ instead of the very expensive alkoxide, oxynitrate, or chloride of titanium, which are difficult to handle and are extremely sensitive to the environmental conditions, such as moisture, light, and heat.

Fig. 7 EDX elemental mapping images of flame synthesized LNCTO for elements (a) Ni, (b) Co, (c) Ti, and (d) O.

Fig. 8 displayed the electrochemical performance of LNCTO and LNTCO at various current rates of 0.1, 0.2, 0.5, 1, 2, and 3C in the potential range between 2.5 and 4.5 V vs. Li/Li⁺ at room temperature.

Fig. 8 (a) Initial charge/discharge profiles of LNCTO and LNTCO. (b) Charge/discharge profiles of LNCTO and (c) charge/discharge profiles of LNTCO at various C-rates, rate performances of (d) LNCTO (e) LNTCO, and (f) cyclic voltammetry (CV) curves of LNCTO and LNTCO.

As shown in Fig. 8(a), LNCTO delivers an initial discharge capacity of 159 mA h g⁻¹ at 0.1C rate, while LNTCO delivers 137 mA h g⁻¹. Both the cathode materials showed an irreversible capacity loss between the initial charge/discharge. The initial charge/discharge capacity loss has also been reported earlier³⁴ in the cathode materials containing Ni³⁺ which may arise due to the formation of electrochemically inactive regions in the cathode attributed to the oxidation of pre-existing Ni²⁺ ions occupying the Li layer in layered oxide materials.^35–37 In addition, it can be seen that the reversible capacity loss for both LNCTO and LNTCO as shown in Fig. 8(b) and (c) is decreased with increasing polarization between charge and discharge profiles upon increasing the C rate from 0.1 to 3C. The charge/discharge results showed that LNCTO exhibits much better electrochemical performance than LNTCO for all C rates.

Fig. 8(d) and (e) compare the rate capabilities of both LNCTO and LNTCO cathode materials. For the first cycle, charge/discharge capacities for LNCTO at current rates of 0.2, 0.5, 1, 2, and 3C in Fig. 8(d) are 152.5/149.1, 135.3/133.2, 122.5/119.4, 110.5/104.3, and 99.7/91.7 mA h g⁻¹, respectively, while those of LNTCO in Fig. 8(e) are 128.9/124.6, 112.1/110.4, 100.0/96.9, 86.7/81.4, and 77.5/70.1 mA h g⁻¹, respectively. As shown in Fig. 8(d) and (e), the specific capacities of LNCTO for all of C rates are delivered larger than those of LNTCO. In addition, the coulombic efficiency of the LNCTO of the initial cycle at a 1C rate is 97.4%, which is also larger than that of LNTCO of 96.9%. It is clear that LNCTO at all C rates demonstrates a much better rate capability and higher reversible capacity than those of LNTCO. Even for the high rate of 3C, LNCTO delivered an excellent discharge capacity of 91.7 mA h g⁻¹, which is much larger than that of LNTCO of 70.1 mA h g⁻¹. The high reversible capacity and coulombic efficiency of LNCTO as compared to those of LNTCO can be attributed to its higher ordered structure and reducing magnitude of Li/Ni disorder which reduced Ni content in the lithium layer as mentioned above in the XRD results.³⁸ In addition, the cycling performances of the flame synthesized LNCTO and LNTCO at the 1C rate for 50 cycles are shown in Fig. S3 (see the ESI,† S3). The initial discharge capacities of LNCTO and LNTCO are 119.3 and 103.1 mA h g⁻¹, and their discharge capacities 50 cycles remain 87.0 and 69.7 mA h g⁻¹, respectively. These correspond to the capacity retentions of 72.9 and 67.6% for LNCTO and LNTCO. It is shown that LNCTO has better reversibility than LNTCO. The electrochemical performance of flame synthesized LNCTO has also been compared with the previous analogous compound of LiNi_.07Co_0.3O₂ based on different synthetic routes as listed in Table 2.^39–43 It can be proved that the obtained specific discharge capacity of the flame synthesized LNCTO is comparable to those reported in the earlier literature.

Table 2 Comparison of specific capacities for the analogous structured LiNi_0.7Co_0.3O₂ prepared via different synthetic routes

System	Synthetic route	Current density/rate	Initial discharge capacity (mA h g^−1)	Ref.
LiNi0.65Co0.25Mn0.1O2	Combustion synthesis	0.05 C	140	39
LiNi0.65Mg0.05Co0.3O2	Sol–gel synthesis	0.1 C	178.9	40
LiNi0.70Co0.30O2	Solid-state route	200 μA cm^−2	127	41
LiNi0.7Co0.2Ti0.05Mg0.05O2	Solid-state route	0.1 C	149	42
LiNi0.70Co0.30O2	Precipitation route	15 mA g^−1	126	43
LiNi0.65Ti0.05Co0.30O2	Flame synthesis	0.1 C	159	This work

---

Fig. 8(f) shows the cyclic voltammetry (CV) curves of LNCTO and LNTCO in the potential range of 2.5–4.5 V vs. Li/Li⁺ at a scan rate of 0.1 mV s⁻¹. The CV curves for both LNCTO and LNTCO were characterized by two redox peaks. Highly symmetric peak profiles were observed for both materials, which indicate the good reversibility of Li⁺ in the extraction/insertion reactions. The anodic peaks were observed at 3.73 and 3.75 V for LNCTO and LNTCO, respectively, corresponding to the oxidation of Ni³⁺ to Ni⁴⁺, whereas the cathodic peaks at 3.68 and 3.69 V, corresponded to the reduction of Ni⁴⁺ to Ni³⁺ for LNCTO and LNTCO, respectively.^44,45 Furthermore, the potential difference between the oxidation and reduction peaks was smaller for LNCTO (0.05 V) than for LNTCO (0.06 V), which indicates that electrode polarization can be reduced and electrons participate actively in the redox reactions for LNCTO by the substitution of Ti⁴⁺ ions at the Co-site compared to the Ni-site. Therefore, the substitution of the dopant at different sites in the same material can influence the electrochemical performance of the cathode material.

4. Conclusion

A simple, rapid, and efficient flame synthetic method to prepare cathode materials for Li-ion batteries with highly ordered layered structures was developed. A new strategy was also used to determine the influence of the dopant at different sites in the same material on the electrochemical performance. Using the new flame synthetic route, ultrafine LiNi_0.95−xCo_xTi_0.05O₂ (x = 0.25 and 0.30) cathode materials that deviated from normal stoichiometry were synthesized by a single calcination step. These materials are generally difficult to synthesize due to the additional nickel ions on the lithium sites. The flame synthesized materials were characterized using a range of spectroscopic techniques, which revealed a single phase with a well-ordered hexagonal layered structure. The substitution of Ti⁴⁺ ions at the Co-site in LiNi_0.95−xCo_xO₂ showed better electrochemical performance than that at the Ni-site. Consequently, the developed flame synthetic technique is quite attractive for the facile fabrication of various lithium metal oxides, such as LiNi_0.8Co_0.2O₂, LiNi_0.76Co_0.24O₂, and LiNi_0.8Co_0.15Al_0.05O₂, (see the ESI,† S1) in the mass production of cathode materials for Li-ion batteries.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This study was supported by the National Research Foundation (NRF-2015R1D1A3A01019167 for Y. Lee and NRF-2015R1D1A4A01019630 for L. Singh) and Priority Research Centers Program (NRF-2009-0093818) funded by the Ministry of Education of the Korean Government.

References

1. W. Liu, P. Oh, X. Liu, M. J. Lee, W. Cho, S. Chae, Y. Kim and J. Cho, Angew. Chem., Int. Ed., 2015, 54, 4440–4457.
2. M. Y. Song, H. Rim and E. Bang, J. Appl. Electrochem., 2004, 34, 383–389.
3. J. Y. Liao, S. M. Oh and A. Manthiram, ACS Appl. Mater. Interfaces, 2016, 8, 24543–24549.
4. E. Lee, J. Blauwkamp, F. C. Castro, J. Wu, V. P. Dravid, P. Yan, C. Wang, S. Kim, C. Wolverton, R. Benedek, F. Dogan, J. S. Park, J. R. Croy and M. M. Thackeray, ACS Appl. Mater. Interfaces, 2016, 8, 27720–27729.
5. Z. Tang, J. Bao, Q. Du, Y. Shao, M. Gao, B. Zou and C. Chen, ACS Appl. Mater. Interfaces, 2016, 8, 34879–34887.
6. J. Cho and B. Park, J. Power Sources, 2001, 92, 35–39.
7. J. Cho, Chem. Mater., 2000, 12, 3089–3094.
8. C. J. Han, J. H. Yoon, W. I. Cho and H. Jang, J. Power Sources, 2004, 136, 132–138.
9. J. Su, Y. Su and M. Guo, J. New Mater. Electrochem. Syst., 2007, 10, 91–94.
10. V. Augustyn, S. Therese, T. C. Turner and A. Manthiram, J. Mater. Chem. A, 2015, 3, 16604–16612.
11. M. Song, I. Kwon and H. Kim, J. Appl. Electrochem., 2005, 35, 1073–1080.
12. P. Kalyani, N. Kalaiselvi, N. G. Renganathan and M. Raghavan, Mater. Res. Bull., 2004, 39, 41–54.
13. M. Dahbi, I. Saadounea and J. M. Amarilla, Electrochim. Acta, 2008, 53, 5266–5271.
14. P. Periasamy, H. S. Kim, S. H. Na, S. I. Moon and J. C. Lee, J. Power Sources, 2004, 132, 213–218.
15. A. Sutka and G. Mezinskis, Front. Mater. Sci., 2012, 6, 128–141.
16. A. Varma, A. S. Mukasyan, A. S. Rogachev and K. V. Manukyan, Chem. Rev., 2016, 116, 14493–14586.
17. L. Singh, U. S. Rai, K. D. Mandal and N. B. Singh, Prog. Cryst. Growth Charact. Mater., 2014, 60, 15–62.
18. R. Santhanam, P. Jones, A. Sumana and B. Rambabu, J. Power Sources, 2010, 195, 7391–7396.
19. C. Julien, Ionics, 1999, 5, 351–357.
20. R. B. Hadjean and J. P. P. Ramos, Chem. Rev., 2010, 110, 1278–1319.
21. C. Julien and S. S. Michael, Ionics, 1998, 4, 181–190.
22. D. M. Brandariz, M. A. S. Rodriguez, S. C. Garcia, M. A. C. Lopez and C. Julien, Ionics, 1999, 5, 345–350.
23. R. Stoyanova, S. Ivanova, E. Zhecheva, A. Samoson, S. Simova, P. Tzvetkovac and A. L. Barra, Phys. Chem. Chem. Phys., 2014, 16, 2499–2507.
24. C. Marichal, J. C. B. Hirschinger, P. Granger, M. Mknctrier, A. Rougier and C. Delmas, Inorg. Chem., 1995, 34, 1773–1778.
25. C. P. Grey and N. Dupre, Chem. Rev., 2004, 104, 4493–4512.
26. A. M. Andersson, D. P. Abraham, R. Haasch, S. Maclaren, J. Liu and K. Amine, J. Electrochem. Soc., 2002, 149, A1358–A1369.
27. A. P. Grosvenor, M. C. Biesinger, R. S. C. Smart and N. S. McIntyre, Surf. Sci., 2006, 600, 1771–1779.
28. A. W. Moses, H. G. G. Flores, J. G. Kim and M. A. Langell, Appl. Surf. Sci., 2007, 253, 4782–4791.
29. L. Dahéron, R. Dedryvère, H. Martinez, M. Ménétrier, C. Denage, C. Delmas and D. Gonbeau, Chem. Mater., 2008, 20, 583–590.
30. J. G. Kim, D. L. Pugmire, D. Battaglia and M. A. Langell, Appl. Surf. Sci., 2000, 165, 70–84.
31. L. Singh, I. W. Kim, W. S. Woo, B. C. Sin, H. Lee and Y. Lee, Solid State Sci., 2015, 43, 35–45.
32. L. Singh, I. W. Kim, B. C. Sin, U. S. Rai, S. H. Hyun and Y. Lee, Ceram. Int., 2015, l41, 12218–12228.
33. K. M. Shaju, G. V. S. Rao and B. V. R. Chowdari, Electrochim. Acta, 2002, 48, 145–151.
34. A. Manthiram, B. Song and W. Li, Energy Storage Mater., 2017, 6, 125–139.
35. I. Saadoune and C. Delmas, J. Mater. Chem., 1996, 6, 193–199.
36. C. Delmas, M. Menetrier, L. Croguennec, I. Saadoune, A. Rougier, C. Pouillerie, G. Prado, M. Grune and L. Fournes, Electrochim. Acta, 1999, 45, 243–253.
37. J. Cho, H. Jung, Y. Park, G. Kim and H. S. Lim, J. Electrochem. Soc., 2000, 147, 15–20.
38. K. S. Kang, Y. S. Meng, J. Greger, C. P. Grey and G. Ceder, Science, 2006, 311, 977–980.
39. I. Saadoune, M. Dahbi, M. Wikberg, T. Gustafsson, P. Svedlindh and K. Edström, Solid State Ionics, 2008, 178, 1668–1675.
40. H. Zhang, Adv. Mater. Sci. Eng., 2014, 2014, 746341.
41. H. Rim, J. Song and D. R. Mumm, Ceram. Int., 2014, 40, 3511–3516.
42. V. Subramanian and G. T. K. Fey, Solid State Ionics, 2002, 148, 351–358.
43. J. Su, Y. Su and M. Guo, J. New Mater. Electrochem. Syst., 2007, 10, 91–94.
44. D. Tong, Y. Luo, W. Chu, Y. He and X. Ji, Mater. Chem. Phys., 2007, 105, 47–52.
45. J. Ni, H. Zhou, J. Chen and X. Zhang, Electrochim. Acta, 2008, 53, 3075–3083.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7nj02679j

---