Spherical LiCoBO₃ particles prepared via a molten salt method for lithium ion batteries

Abstract

LiCoBO₃/ketjen black composites were prepared at a moderate temperature of 450 °C by a molten salt method using eutectic mixtures of LiCl and KCl as the reaction medium and ketjen black as a carbon source. The monoclinic structure and spherical morphology are respectively confirmed by X-ray diffraction, scanning electron microscopy and transmission electron microscopy. The electrochemical behavior was studied by galvanostatic charge–discharge and cyclic voltammetry tests. At a rate of C/20, the composite delivered an initial specific capacity of 48 mA h g⁻¹ and a discharge specific capacity of 40 mA h g⁻¹ at the 25th cycle, indicating a stable cycling performance. However, rate capability for the composite needs to be further improved.

---

1. Introduction

Polyanionic intercalation compounds such as phosphates, silicates and fluorophosphates have been proposed as a new class of cathode materials for lithium ion batteries due to a lot of advantages since the first discovery of the electrochemical properties of olivine LiFePO₄ by Padhi et al.¹ However, the presence of spectator polyanion units impedes both the specific capacity and the specific energy to some degree. It is therefore essential to minimize this shortcoming. Considering this, a new family of lithium transition metal borates, namely LiMBO₃ (M = Mn, Fe or Co), is a ideal choice because of their comparably lower molecular weights,^2–13 which lead to a larger theoretical specific capacity (>210 mA h g⁻¹), about 30% larger than that of PO₄³⁻-based LiMPO₄. Among the LiMBO₃ borates, it is the excellent combination of a high theoretical capacity (ca. 215 mA h g⁻¹) and a high expected operating voltage (∼4 V) that makes LiCoBO₃ an ideal cathode.^14–18

LiCoBO₃ was usually synthesized by conventional solid-state reactions, which usually need high-temperature heating for a long time and lead to abrupt particle growth. For example, Yamashita developed a two-step solid-state synthesis, involving (i) formation of Co₂B₂O₅ intermediate phase followed by (ii) annealing an intimate mixture of Li₂CO₃ and Co₂B₂O₅ at 300–600 °C.¹⁴ Alternately, solid state synthesis have been reported by annealing the mixture of LiBO₂ and Co₃O₄ at 800–850 °C in air or by annealing LiBO₂ and CoO at 600 °C for 20 h in N₂ atmosphere.^15,16 In all cases, it is pivotal to maintain air/N₂ ambience avoiding reduction of Co-species to metallic Co⁰, indicating that synthesis of LiCoBO₃ is not trivial. To overcome these disadvantages, a sol–gel synthesis was developed by Afyon et al.¹⁶ Herein we present another alternative method, namely a molten salt method. The molten salt method shows accelerated reaction rate and controllable particle morphology compared to the solid-state reactions, because the salt melt exhibits higher ion diffusion rate and stronger dissolving capability.^19–21

In light of these aspects, we present the molten salt synthesis of LiCoBO₃ using eutectic mixtures of LiCl and KCl as the reaction medium. To enhance the surface electronic conductivity, ketjen black (KB) is adopted as a carbon source to coat the LiCoBO₃ particles. Additionally, the structure and the electrochemical performance of LiCoBO₃ were also characterized.

2. Experimental

Reagent-grade LiCl, KCl, CoO, H₃BO₃ and Li₂CO₃ were used in the study. LiCoBO₃ particles were prepared by a molten salt method. A 0.59LiCl–0.41KCl (molar ratio) flux was used as the eutectic composition with a melting point of 353 °C. After stoichiometric amounts of Li₂CO₃, CoO and H₃BO₃ were weighed and ground in a QM-3SP2 planetary ball mill, the mixture was mixed with chloride flux (molar proportion CoO/chloride flux 1 : 6) and a certain amount of KB as conductive agent (proportion by weight KB/LiCoBO₃ 1 : 9), and reground again, then sintered at 450 °C in Ar atmosphere for 10 h. The products were washed with de-ionized water, filtered, and dried in vacuum.

Powder X-ray diffraction (XRD) patterns were obtained on a D8 Advance type diffractometer equipped with Cu Kα radiation (operated at 40 kV and 40 mA). The XRD patterns were analyzed with MDI Jade 6.0 software to identify phase and calculate lattice parameters. The surface morphologies of the sample were observed using a JSM-6380 scanning electron microscopy (SEM) and a Tecnai G2 F20 S-TWIN transmission electron microscopy (TEM). The CHN elemental analyzer was used to measure the amount of carbon in the LiCoBO₃/C composite.

For electrochemical studies, after 80 wt% LiCoBO₃/KB composite and 10 wt% conductive carbon (Super-P) were mechanically mixed using a QM-3SP2 planetary ball mill for 4 h with a ball to powder ratio of 15 : 1, composite electrodes were fabricated with the composite, conductive carbon and polyvinylidene fluoride binder in the weight ratio 80 : 10 : 10 using N-methyl pyrrolidone as solvent. Electrodes were prepared using an etched aluminium foil as a current collector using the doctor-blade technique. After vacuum drying at 120 °C for 15 h, circular disks were punched out (ϕ = 12 mm), containing ∼25 μm thick active material with a loading of 3–5 mg cm⁻². Lithium metal foil, 1 M LiPF₆ solution in ethylene carbonate–diethyl carbonate (EC–DEC, 3/7 v/v) and Celgard 2502 membrane were used as counter electrode, electrolyte and separator respectively to assemble coin-type cells (size 2025) in an Ar-filled glove box (Mikrouna). Galvanostatic tests were done in the voltage range of 4.5–2.0 V. Cyclic voltammetry (CV) test was carried out at a sweep rate of 0.035 mV s⁻¹ within 4.5–2.0 V using a CHI660 electrochemical instrument. Electrochemical impedance spectroscopy (EIS) measurements were performed using a CHI 660 electrochemical instrument in the frequency range of 10 kHz to 10 mHz with an AC voltage of 5 mV.

3. Results and discussion

The XRD pattern of the LiCoBO₃/KB composite annealed at 450 °C for 10 h is shown in Fig. 1. The bars in the bottom of the graph are corresponding to the standard PDF card of monoclinic LiCoBO₃ (card no. 87-0512). The structural refinement of the XRD pattern was done based on a monoclinic structure using C2/c space group. The XRD results of the LiCoBO₃/KB composite, analyzed with MDI Jade 6.0 software, were shown in Table 1, which are consistent with the previous report (card no. 87-0512). It can be seen from the XRD pattern that the XRD peaks of the sample synthesized at 450 °C corresponds to monoclinic LiCoBO₃. No other common impurities such as CoO, Co and Li₆CoO₄ have been observed, which were easily produced in conventional solid-state reactions.^14,18,22 Since the molten salts as a reaction media could provide a liquid reaction environment for reactants, thereby facilitating the reaction and accelerating the crystallization process,^20,21 the well-crystallized LiCoBO₃ can be formed at a moderate temperature of 450 °C relatively rapidly. Although the residual carbon content in the LiCoBO₃/KB composite was 9.94% in weight measured by CHN analyzer, no diffraction peak related-carbon in Fig. 1 was observed.

Fig. 1 XRD of LiCoBO₃/KB composite.

Table 1 Cell parameters of LiCoBO₃ synthesized in this work

	a (Å)	b (Å)	c (Å)	β (°)	V (Å^3)
This work	5.1519	8.8334	10.12409	1.297	460.61
Card no. 87-0512	5.129	8.840	10.1009	1.36	457.81

---

The morphologies of the LiCoBO₃ samples observed via SEM are shown in Fig. 2. It can be found from Fig. 2 that the sample is composed of well shaped spherical particles with diameter ranging from 0.5 μm to 1 μm. The formation of spherical particles is much dependent on the effect of LiCl–KCl molten salt. According to dissolution–precipitation mechanism, the crystallization process of LiCoBO₃ undergoes a route of dissolution–precipitation controlled growth in the molten salt, in which the LiCoBO₃ particle shows an intense inclination to become spherical in shape to lower the surface energy under the effect of large interfacial intensity of LiCl–KCl molten salt.^23,24 Therefore, as the dissolution–precipitation process and sphericization process continue, the LiCoBO₃ particles gradually display spherical morphology. Furthermore, some floccus-like substance which adheres to the spherical particles and the connection of the spherical particles through KB are found in Fig. 2a. For comparison, a SEM micrograph of the original LiCoBO₃ sample synthesized by the same procedure is shown in Fig. 2b. It can be seen clearly that the spherical particles in Fig. 2b are smoother than in Fig. 2a, and there is no the connection between the particles. It has reported that the residual carbon may be useful as the conductive diluent in the electrode formulation.²⁵ In addition, the conductive carbon framework formation via KB could be expected to provide a mechanical stability within the composite electrode material. Therefore, the residual carbon may help to improve the conductivity and electrochemical properties of LiCoBO₃.

Fig. 2 SEM of LiCoBO₃/KB composite (a) and the original LiCoBO₃ sample (b).

TEM investigation was also conducted to examine the LiCoBO₃/KB composite. In the low-magnification TEM image of the LiCoBO₃ sample (Fig. 3a), the relatively darker locations appear to indicate LiCoBO₃ particles, and the brighter image seems to display a layer of carbon, suggesting that the LiCoBO₃ particles are coated by conductive carbon. The high-resolution TEM image (Fig. 3b) is applied to observe the fine structure of the LiCoBO₃/KB composite. Clear lattice fringes can be visible in Fig. 3b, indicating the LiCoBO₃ crystals with well crystallinity. Additionally, the presence of the carbon coating on the rim of LiCoBO₃ particles is evident from Fig. 3b.

Fig. 3 TEM of LiCoBO₃/KB composite.

The LiCoBO₃/KB composite was tested in standard Li-half cell architecture under ambient temperature. Fig. 4 shows the galvanostatic charge/discharge curves of the LiCoBO₃/KB electrode within 4.5–2.0 V at a rate of C/20 (1C = 215 mA h g⁻¹). As shown in Fig. 4, the LiCoBO₃/KB composite shows distinct electrochemical activity but with huge polarization, similar to the observations reported previously.^2,14–18 It should be mentioned that the tremendous polarization observed upon the first charge is less important during subsequent ones. The first charge exhibited a sloped region above 4.3 V and in the subsequent cycles, the sloped region for charge is found to be ∼3.9 V. A similar phenomenon was found in the cases of LiFeBO₃ and LiMnBO₃,^26–28 respectively. These differences could be explained by the partial amorphization of LiCoBO₃ due to the stress induced after the first cycle.¹⁶ The low first-cycle coulombic efficiency is also very evident similar to the observations in previous investigations, in which LiMnBO₃ and LiCoBO₃ was respectively employed as a cathode material for Li-ion batteries.^29–33 The low coulombic efficiency is probably attributed to the formation of solid electrolyte interphase (SEI) film on the surface of the electrode material and the reaction with the electrolyte, which seems to be more pronounced with the compound containing cobalt.^8,11,32–35 This may also be due to the fact that the CoBO₃ end member was found to be structurally/chemically unstable and prone to decomposition, as has been suggested by Yamada et al.¹⁴ Nonetheless, subsequent testing indicated that this coulombic inefficiency was restricted primarily to the first cycle because subsequent cycles showed significantly improved reversibility.

Fig. 4 The first charge/discharge profiles of LiCoBO₃/KB composite at C/20 rate in the voltage range 2.0–4.5 V.

The results of charge–discharge cycles at a rate of C/20 in the voltage range 4.5–2.0 V are shown in Fig. 5. From Fig. 5, it can be found that the first discharge capacity is ∼48 mA h g⁻¹, and the cycling is fairly stable with a discharge specific capacity of 45 mA h g⁻¹ at the 10th cycle. The discharge specific capacity of 40 mA h g⁻¹ can still be hold at the 25th cycle, indicating an average discharge capacity degradation of just ∼0.06% per cycle for the first 25 cycles. Even though the capacity we achieved for LiCoBO₃ is far from its theoretical specific capacity and still low for practical applications, the results are still encouraging since the overall cathode properties achieved in this study are much better than those previously obtained with solid state synthesis.^14,17,18 The reversible specific capacity obtained with solid state synthesis is within the limit of 30 mA h g⁻¹ at C/20 between 2.5 and 4.5 V. Even for the nano-LiCoBO₃ synthesized by sol–gel method, it just obtains an initial discharge specific capacity of 46 mA h g⁻¹ and a discharge specific capacity of 40 mA h g⁻¹ in the 25th cycle within 4.7–2.0 V at C/20 rate for charge and at C/40 rate for discharge with the use of reduced graphite oxide (RGO) as conductive additive.¹⁶

Fig. 5 Cycling performance of LiCoBO₃/KB composite at C/20 rate in the voltage range 2.0–4.5 V.

Further investigation into the sample's performance was carried out by examining the effect of current rate on the capacity of the composite electrode. Fig. 6 shows the discharge specific capacity of the composite obtained at different current rates. The discharge specific capacities are respectively in the range of ∼46, 37 and 23 mA h g⁻¹ for the rates of C/20, C/10 and C/5, comparable to the data reported by Afyon et al. for LiCoBO₃/RGO composite prepared by sol–gel.¹⁶ After the higher rate measurement, the current rate is reduced back to C/20, and 91% of the initial discharge specific capacity can be recovered. In addition, stable cycling performance was gotten for all the three rates. For example, 97% of the initial discharge specific capacity can be retained at the 5th cycle with C/10 rate. On the other hand, the rate capability of the LiCoBO₃/KB composite is unsatisfactory. As shown in Fig. 6, the discharge specific capacity is decreased significantly as the current increases, for instance, the value at C/5 is only equivalent to ∼50% of that at C/20, which suggests that carbon modification could enhance the surface electronic conductivity of LiCoBO₃, but not improve its bulk conductivity. Maybe, nanosizing, doping or enhancing the quality of conductive agent coating will be efficient in obtaining better rate performance.

Fig. 6 Rate performance of LiCoBO₃/KB composite at various rates after cycling five times at each rate between 2.0 and 4.5 V.

To understand the effect of KB coating on the electrochemical performance, EIS study was carried out for the original and KB-coated LiCoBO₃ samples, and the Nyquist plots are shown in Fig. 7. Each of the two samples exhibits a semicircle in the high frequency region and a straight line in the low-frequency one. The numerical value of the semicircle diameter on the Z′ axis (real axis) approximately corresponds to the charge transfer resistance (R_ct). After KB coating, R_ct is markedly decreased, indicating that the KB on the surface of particles can make the electrochemical reaction easier. Thus, KB coating is helpful for improving the electrochemical performances of LiCoBO₃ materials.

Fig. 7 Nyquist plots of the original and the KB coated LiCoBO₃ samples.

Considering the poor conducting nature of the LiCoBO₃, the foregoing results are significant improvements compared to the earlier work. The enhanced electrochemical properties of the LiCoBO₃/KB composite can be attributed to the following factors: (1) the presence of KB in the composite would provide good electronic contact between LiCoBO₃ particles, as shown in Fig. 6; (2) the KB which adhered to the LiCoBO₃ particles can prevent the undesirable reactions with by-products that could be formed during cycling by electrolyte decomposition. (3) LiCoBO₃ displays nice structural reversibility in the charged and discharged states, which was supported by results obtained from XRD patterns^14,16 and a computational study,¹² showing that remarkably small volume changes of Li_xCoBO₃ occur before and after delithiation, and these changes may facilitate lithium ion extraction or insertion with high reversibility.

As shown in Fig. 8, the CVs were recorded for the LiCoBO₃/KB electrode using the Li metal as counter and reference electrodes in the voltage range of 4.5–2.0 V. In all the four cycles, a broad reduction peak associated to lithium reinsertion (Co³⁺ reduced to Co²⁺) was observed between 3.8 and 2.4 V. During the first oxidation, voltage increases slowly to ∼4.0 V from the open circuit voltage (∼3.1 V), followed by a sharp increase in voltage to 4.5 V, which suggests a high polarization. In the subsequent oxidation processes, a very broad oxidation peak was found, ranging from 2.8 to 4.0 V.

Fig. 8 CV curves for LiCoBO₃/KB composite.

4. Conclusions

A molten salt method using eutectic mixtures of LiCl and KCl as the reaction medium and KB as a carbon source is developed for the synthesis of LiCoBO₃/KB composite at a moderate temperature of 450 °C. XRD studies confirmed the formation of monoclinic LiCoBO₃. The SEM image revealed that the sample is composed of well shaped spherical primary particles with diameter ranging from 0.5–1 μm. When cycled at a rate of C/20, the composite exhibited an initial specific capacity of 48 mA h g⁻¹ and a discharge specific capacity of 40 mA h g⁻¹ at the 25th cycle, indicating stable cycling performance. The discharge specific capacity, however, is decreased significantly as the current increases.

Acknowledgements

This work was financially supported by Scientific Research Fund of Hunan Provincial Education Department (No. 14A052), Hunan Provincial Innovation Foundation for Postgraduate (No. CX2015B538) and Hunan Provincial Natural Science Foundation of China (No. 12JJ2022).

Notes and references

1. A. K. Padhi, K. S. Nanjundaswamy and J. B. Goodenough, J. Electrochem. Soc., 1997, 144, 1188.
2. V. Legagneur, Y. An, A. Mosbah, R. Portal, A. Salle, A. Verbaere, D. Guyomard and Y. Piffard, Solid State Ionics, 2001, 139, 37.
3. J. C. Kim, D.-H. Seo, H. L. Chen and G. Ceder, Adv. Energy Mater., 2015, 8, 1401916.
4. J. C. Kim, X. Li, C. J. Moore, S. H. Bo, P. G. Khalifah, C. P. Greyand and G. Ceder, Chem. Mater., 2014, 26, 4200.
5. A. Muslim, T. Ma, Z. Su and N. Ili, Rare Met. Mater. Eng., 2014, 43, 2095.
6. A. P. Tang, D. H. He, Z. Q. He, G. R. Xu, H. S. Song and R. H. Peng, J. Power Sources, 2015, 275, 888.
7. S. Afyon, D. Kundu, F. Krumeich and R. Nesper, J. Power Sources, 2013, 224, 145.
8. K. Karthikeyan and Y. S. Lee, RSC Adv., 2014, 4, 31851.
9. Y. S. Lee and H. Lee, Mater. Lett., 2014, 132, 401.
10. S. H. Bo, F. Wang, Y. Janssen, D. L. Zeng, K. W. Nam, W. Q. Xu, L. S. Du, J. Graetz, X. Q. Yang, Y. M. Zhu, J. B. Parise, C. P. Grey and P. G. Khalifah, J. Mater. Chem., 2012, 22, 8799.
11. L. Chen, Y. M. Zhao, X. N. An, J. M. Liu, Y. Z. Dong, Y. H. Chen and Q. Kuang, J. Alloys Compd., 2010, 494, 415.
12. D.-H. Seo, Y.-U. Park, S.-W. Kim, I. Park, R. A. Shakoor and K. Kang, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 205127.
13. L. Tao, G. Rousse, J. N. Chotard, L. Dupont, S. Bruyère, D. Hanžel, G. Mali, R. Dominko, S. Levasseurf and C. Masquelier, J. Mater. Chem. A, 2014, 2, 2060.
14. Y. Yamashita, P. Barpanda, Y. Yamada and A. Yamada, ECS Electrochem. Lett., 2013, 2, A75.
15. S. H. Bo, G. M. Veith, M. R. Saccomanno, H. F. Huang, P. V. Burmistrova, A. C. Malingowski, R. L. Sacci, K. R. Kittilstved, C. P. Grey and P. G. Khalifah, ACS Appl. Mater. Interfaces, 2014, 6, 10840.
16. S. Afyon, C. Mensing, F. Krumeich and R. Nesper, Solid State Ionics, 2014, 256, 103.
17. B. L. Roux, C. Bourbon, J. F. Colin and V. Pralong, RSC Adv., 2015, 5, 72801.
18. B. L. Roux, C. Bourbon, O. I. Lebedev, J. F. Colin and V. Pralong, Inorg. Chem., 2015, 54, 5273.
19. L. Wen, Q. Lu and G. Xu, Electrochim. Acta, 2006, 51, 4388.
20. Y. N. NuLi, J. Yang, J. Wang and Y. Li, J. Phys. Chem. C, 2009, 113, 12594.
21. W. P. Tang, X. J. Yang, Z. H. Liu, S. Kasaishi and K. Ooi, J. Mater. Chem., 2002, 12, 2991.
22. L. Tao, J. R. Neilson, B. C. Melot, T. M. McQueen, C. Masquelier and G. Rousse, Inorg. Chem., 2013, 52, 11966.
23. X. Y. Yang, W. P. Tang, H. Kanoh and K. Ooi, J. Mater. Chem., 1999, 9, 2683.
24. G. Xie, Molten Salts: Principle and Application, Metallurgy, Industry Press, Beijing, 1998, p. 8.
25. J. Barker, M. Y. Saidi and J. L. Swoyer, J. Electrochem. Soc., 2004, 151, A1670.
26. A. Yamada, N. Iwane, Y. Harada, S. I. Nishimura, Y. Koyama and I. Tanaka, Adv. Mater., 2010, 22, 3583.
27. T. Ma, A. Muslim and Z. Su, J. Power Sources, 2015, 282, 95.
28. J. C. Kim, C. J. Moore, B. Kang, G. Hautier, A. Jain and G. Ceder, J. Electrochem. Soc., 2011, 158, A309.
29. S. L. Li, L. Q. Xu, G. D. Li, M. Wang and Y. J. Zhai, J. Power Sources, 2013, 236, 54.
30. K.-J. Lee, L.-S. Kang, S. Uhm, J. S. Yoon, D.-W. Kim and H. S. Hong, Curr. Appl. Phys., 2013, 13, 1440.
31. R. Ma, L. Y. Shao, K. Q. Wu, M. M. Lao, M. Shui, C. Chen, D. J. Wang, N. B. Long, Y. L. Ren and J. Shu, Ceram. Int., 2013, 39, 9309.
32. V. Aravindan, K. Karthikeyan, S. Amaresh and Y. S. Lee, Bull. Korean Chem. Soc., 2010, 31, 1506.
33. S. Afyon, D. Kundu, A. J. Darbandi, H. Hahn, F. Krumeich and R. Nesper, J. Mater. Chem. A, 2014, 2, 18946.
34. R. Dominko, M. Bele, A. Kokalj, M. Gaberscek and J. Jamnik, J. Power Sources, 2007, 174, 457.
35. V. Aravindan, K. Karthikeyan, S. Ravi, S. Amaresh, W. S. Kim and Y. S. Lee, J. Mater. Chem., 2010, 20, 7340.

---