Improved high-temperature capacity retention of the LiMn₂O₄ cathode lithium-ion battery by ion exchange polymer coating

Abstract

An ion exchange polymer coating on the LiMn₂O₄ cathode to overcome capacity fading of a lithium-ion battery at high temperatures is first demonstrated, and it shows very good capacity retention compared with the pristine LiMn₂O₄ cathode without coating.

---

Rechargeable lithium-ion batteries (LIBs) are expected to be used as the power source for hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (EVs) and electric vehicles because of their high intrinsic energy density and high voltage.¹ However, the current LIB technologies are not yet able to meet the requirement of efficiently storing alternative energy sources and/or powering hybrid or electric vehicles. Enhancement in safety and long life, cost reduction, and improvements in thermal stability and energy density are the main properties that need to be implemented. To meet the abovementioned requirements, many research groups have extensively and intensively investigated many possible cathode materials, such as LiCoO₂,² LiN_1−x−yCo_xMn_yO₂,³ LiFePO₄ (ref. 4) and LiMn₂O₄,^5–7 for application in EVs. Among the studied cathode materials, the LiMn₂O₄ (LMO) with the spinel framework has been considered as one of the most promising cathode material to be used for EVs due to several advantages desirable for large-scale LIB applications: (1) low cost and abundance of raw material; (2) high rate performance due to its three-dimensional (3D) channel structure, which facilitates efficient Li⁺ diffusion; (3) high safety and non-toxicity; and (4) a high operation potential (∼4.1 V versus Li/Li⁺).¹ However, serious capacity fading of the LIBs based on the LMO, particularly at high temperatures over 55 °C, is still a significant barrier to practical applications.^5–7 The capacity fading is known to be caused by the dissolution of Mn²⁺ from the LMO into the electrolyte, which is originated from the disproportionation reaction: 2Mn³⁺ (s) → Mn⁴⁺ (s) + Mn²⁺ (aq.).^5,6 Moreover, it is known that dissolution accelerated in the high-voltage range of >4.0 V versus Li/Li⁺ also in addition to the attack of hydrogen fluoride (HF), inevitably occurs by both thermal decomposition of lithium hexafluorophosphate (LiPF₆) and the reaction of the LiPF₆ with residual water in the electrolyte.^5–7 Mn²⁺ is soluble in the non-aqueous electrolyte, and Mn⁴⁺ tends to stay in the positive electrode in the form of λ-MnO₂, whereas the solvated Mn²⁺ is readily reduced at the negative electrode because the redox potential of Mn²⁺/Mn couple is much higher than the intercalation potential of the carbonaceous negative electrode materials. The deposition of Mn metal on the carbon surface will then lead to the significant impedance rise and capacity fading of the LIBs.

In order to suppress Mn²⁺ dissolution, considerable methods are reported, including: (1) partial Mn substitution with different transition metals such as Li⁺, Mg²⁺ and Al³⁺;⁸ (2) partial anion substitution O²⁻ with F⁻;⁹ (3) metal oxide surface coatings;¹⁰ (4) functional electrolyte additives;¹¹ (5) new polymer binders;¹² and (6) new non-fluorinated lithium salts, such as lithium bisoxalatoborate (LiBOB).¹³ Despite an improved cycling performance, many of these approaches also lead to additional drawbacks, such as the sacrifice of specific capacity, or involve more complex steps, which will increase the production costs.

Herein, we report a simple method to solve the Mn²⁺ dissolution problem by coating cation exchange polymer on the surface of the LMO cathode electrode. Through the ion exchange reaction, H⁺ ions of the cation exchange polymer will be replaced by the dissolved Mn²⁺ ions of LMO electrodes (shown in Fig. 1). Then the dissolved Mn²⁺ ions will be captured by the cation exchange polymer, resulting in improved cycle performance at high temperatures. In addition, the cation exchange polymers are normally considered as a single conductor (or ion-conducting electrolyte) used in lithium-ion¹⁴ and lithium–sulfur batteries.¹⁵ Thus, Li⁺ ion diffusion in the cell is not affected.

Fig. 1 (a) Schematic diagram for the principle of cation exchange polymer coating on the LiMn₂O₄ electrode to capture dissolved Mn²⁺ ions. (b) Ion exchange reaction between H⁺ ions on the Nafion ionomer and Mn²⁺ ions.

In our experiments, the Nafion ionomer made by DuPont, one of the best known cation exchange polymers, is used as the ion exchange polymer coating on the LMO electrodes. The LMO electrode was composed with LiMn₂O₄ powder (80 wt%), carbon black (10 wt%) and poly(vinylidene fluoride) (PVDF 10 wt%). The detailed experimental process is described in the ESI.† Nafion dispersion (Nafion dispersion D520, 5 wt% in alcohol–water) solution was coated on the LMO electrode and dried at room temperature to make a Nafion-coated electrode. The amount of Nafion on the electrode is about 0.56 mg cm⁻².

In order to verify the ion exchange reaction between the dissolved Mn²⁺ ions in electrolyte and H⁺ ions of cation exchange polymer, two experiments were carried out. First, both the LMO electrodes with and without the Nafion coating were put in a sealed bottle containing a 1.0 M LiPF₆-dissovled EC-DMC (1 : 1, v/v) electrolyte at 55 °C for 15 days. From Fig. 2a, it can be easily determined that the solution soaked in the LMO electrode with the Nafion coating is a transparent and colorless solution; however, the solution soaked in the LMO electrode without coating shows a pink color, which is generally the color of bivalent manganese ions. The manganese concentration of the electrolyte solution in the electrode with a Nafion coating (0.123 mg L⁻¹) is considerably smaller than that of the electrode without the Nafion coating (1.684 mg L⁻¹). These results indicate that the Nafion coating blocked the Mn²⁺ ions into the electrolyte, thereby inhibiting the disproportionation reaction of the Mn³⁺ ions. In addition, it also implies that the Nafion coating will act as an effective layer to prevent the LMO electrode by the attack of HF, formed by the decomposition of LiPF₆ salt in electrolyte. The occurrence of the ion exchange reaction was further supported by the FT-IR spectra of the Nafion film, obtained before and after storage in the manganese-dissolved electrolyte. In the Nafion with H⁺ form, the band at 972 cm⁻¹ was assigned to the –SO₃H group.¹⁶ However, the intensity of this band decreased when the part of H⁺ ions were exchanged by Mn²⁺ ions. The band at 1057 cm⁻¹ of the Nafion with H⁺ form was attributed to the –SO₃⁻ symmetric stretch.^16,17 Due to the interaction between Mn²⁺ and oxygen, this band shifted to 1060 cm⁻¹ of the Nafion-captured Mn²⁺ ions.

Fig. 2 (a) Image of solutions of the LiMn₂O₄ electrodes with and without the Nafion coating soaked in electrolytes at 55 °C for 15 days. (b) FT-IR spectra of the Nafion film obtained before (Nafion–H) and after (Nafion–Mn) storage in the manganese-dissolved electrolyte.

The typical morphologies of the LMO electrodes with and without the Nafion coating are shown in Fig. 3. Compared with the pristine electrode shown in Fig. 3a, it can be clearly seen that a visible Nafion layer is homogeneously formed on the LMO electrode after coating with the Nafion solution. The thickness of the Nafion layer was about 1 μm. The Nafion layer could have the advantage of not only blocking the Mn²⁺ ions into the electrolyte, but also preventing the active material by the attack of HF formed by the decomposition of LiPF₆. To analyze the morphology changes of the electrode after charge–discharge cycling measurements, the cells were disassembled in an argon-filled glove box. The electrodes were washed with anhydrous DMC several times to remove residual salts, and then dried in vacuum for more than 2 h at room temperature. It can be demonstrated from Fig. 3c that except for slight cracks, most of the Nafion coating retains its integrity after cycling measurements, implying the stability of the Nafion layer during charge–discharge cycling measurements. Fig. 3d shows the TG curve of the LMO electrodes with the Nafion coating before and after cycling measurements. In both electrodes, the weight loss between 300 and 550 °C is about 16.5 wt%, which is mainly due to the degradation of the PVDF binder and Nafion coating in both electrodes. TG curves of both electrodes before and after 100 cycling tests are almost the same, which indicates that the Nafion layer has good stability on the electrodes during charge–discharge cycling measurements.

Fig. 3 SEM images of the LiMn₂O₄ electrodes (a) without a Nafion coating, (b) with the Nafion coating and (c) after 100 cycles. TG curves of the LiMn₂O₄ electrodes with the Nafion coating (d) before and after 100 cycles.

The initial charge–discharge profiles of the cells using both the LMO electrodes at 55 °C, which were cycled under a voltage range of 3.0–4.5 V at a current density of 100 mA g⁻¹, were shown in Fig. 4a. Two pseudoplateaus at around 3.9 and 4.1 V that indicate the typical electrochemical behaviour of the spinel LiMn₂O₄, which are observed in both the charge and discharge curves.¹⁸ The performance of the first cycle of the cell using the pristine LMO electrode (charge capacity: 118.0 mA h g⁻¹, discharge capacity: 110.5 mA h g⁻¹, coulombic efficiency: 93.6%) was better than that of the cell using the LMO electrode with the Nafion coating (charge capacity: 109.6 mA h g⁻¹, discharge capacity: 102.1 mA h g⁻¹, coulombic efficiency: 93.2%). The main reason is the ohmic resistance based on AC impedance measurements shown in Fig. 4b, which increases from 3.74 Ω cm⁻² to 12.27 Ω cm⁻² with the Nafion coating on the electrodes due to the low conductivity of the Nafion film (about 10⁻⁵ S cm⁻¹ (ref. 15)). The detailed analysis of AC impedance results is described in the ESI.†

Fig. 4 Electrochemical performance of batteries at 55 °C. (a) Voltage profiles for the first cycle of the batteries using the LiMn₂O₄ electrodes with and without the Nafion coating measured between 3.0 and 4.5 V at a current density of 100 mA g⁻¹. (b) Nyquist plots of the LiMn₂O₄ electrodes with and without the Nafion coating after the first 100 mA g⁻¹ charged cell. (c) Cycling stability of batteries using the LiMn₂O₄ electrodes with and without the Nafion coating. Current density: 100 mA g⁻¹. (d) Charge–discharge curves of battery using the LiMn₂O₄ electrodes with the Nafion coating at different cycle numbers.

The cycling performance of the cells using both the LMO electrodes at 55 °C is shown in Fig. 4c. It can be found that the capacities of both cells decrease with the increment of cycle numbers. For the pristine LMO electrode, the capacity reduced rapidly after around 40 cycles and was decreased to 20.2 mA h g⁻¹ (18.3% of the initial capacity) after 100 cycles. Compared to the pristine LMO electrode, the battery using the LMO electrode with the Nafion coating showed a high capacity retention of 77.4 mA h g⁻¹ (75.8% of the initial capacity) after 100 cycles. The coulombic efficiency of the LMO electrode with the Nafion coating was almost 100% even after 100 cycles, which is also better than that of the battery using the pristine LMO electrode. In addition, the difference between the charge and discharge potentials of the LMO electrode with the Nafion coating (shown in Fig. 4d) does not significantly change, compared with the pristine LMO electrode (Fig. S3†). In general, there are two main factors leading to the capacity fading of the LMO electrodes at high temperatures, including the disproportionation reaction of LMO- and LiPF₆-based electrolyte.⁶ At room temperature, both the disproportionation reaction and thermal decomposition of LiPF₆ salt will not easily occur. Both cells, which were using both of the LMO electrodes, then show good cycling performance at room temperature (Fig. S4†). After 100 cycles, the capacity retention of both cells is almost same, and their capacity is about 100 mA h g⁻¹.

Fig. 5 shows the XRD patterns of the pristine and LMO electrodes after 100 cycles with and without the Nafion coating at 55 °C. All the peaks in the XRD patterns of the LMO electrodes were indexed as the spinel phase (JCPDS file no. 35-0782). It can be found that the intensities of all of the LiMn₂O₄ diffraction peaks of both LMO electrodes are weakened. This may be due to the decay of the spinel structure of LiMn₂O₄, which resulted from Mn²⁺ dissolution at elevated temperatures.⁷ However, the extent of weakening of the electrode without the coating is stronger than that of the electrode with coating. Fig. 5b shows the magnified patterns at 2θ = 18–37°. It can be found that both the diffraction peaks (111) and (311) of the electrode without coating shifted to higher angles after cycling. In addition, the value of full width at half maxima (FWHM) of both peaks also increased after cycling. These results suggest that the crystal lattice of the LiMn₂O₄ in the electrode without coating contracted after cycling.^6,7 Compared with the LMO electrode without coating, there were no obvious change of the FWHM value and no obvious shift were found in the diffraction peaks of the electrode with the Nafion coating even after 100 cycles, which indicates that the extent of degradation of the spinel structure of LiMn₂O₄ in the electrode without coating is much stronger than that of the electrode with coating. These results indicate that the Nafion-coated LMO electrode shows much better thermal stability than that of the pristine LMO electrode.

Fig. 5 XRD patterns of (a) the LiMn₂O₄ electrodes and (b) those magnified between 18° to 37° for 2θ: (i) the pristine LiMn₂O₄ electrodes with the Nafion coating, and (ii) after 100 cycles, (iii) the pristine LiMn₂O₄ electrodes without the Nafion coating and after 100 cycles (iv).

The surfaces of the LiMn₂O₄ electrodes with the Nafion coating before and after 100 cycles at 55 °C were analyzed by XPS. The C 1s, Mn 2p, O 1s and S 2p spectra of both electrodes are shown in Fig. 6. The C 1s spectra show slight difference between the LiMn₂O₄ electrodes with Nafion before and after testing (Fig. 6a). There are two characteristic peaks in both electrodes in the C 1s spectra: C–C bond in carbon black (284.1 eV),¹⁹ C–F bond in Nafion and PVDF (284.2 eV).²⁰ In addition, new peak attributing to Li₂CO₃ (288.7 eV) can be found in the LiMn₂O₄ electrodes after the test, which is due to SEI formation on the electrodes during the cycling measurements.^19,21

Fig. 6 XPS spectra of the LiMn₂O₄ electrodes with the Nafion coating before and after 100 cycles at 55 °C. (a) C 1s spectra, (b) Mn 2p spectra, (c) O 1s spectra and (d) S 2p spectra.

The Mn 2p spectra are shown in Fig. 6b. There are two main peaks in Mn 2p spectra in both electrodes, dominated by Mn 2p^3/2: Mn³⁺ in Mn₂O₃ or LiMn₂O₄ (641.7 eV), Mn⁴⁺ in MnO₂ or LiMn₂O₄ (642.9 eV).²² Moreover a new weak Mn 2p^1/2 peak (653.6 eV) appeared in the tested LiMn₂O₄ electrode.²² This shows that the Nafion film can capture of Mn²⁺ ions on the electrodes.

In the O 1s spectra of both electrodes shown in Fig. 6c, there are two main peaks: –SO₃H bonds in Nafion (535.7 eV)²⁰ and Li₂CO₃ (532.0 eV).²³ In addition, the Mn–O bond in –(SO₃)₂Mn (529.6 eV)²¹ can be found in the electrodes after testing, which indicates that the Nafion layer can capture the Mn²⁺ from the LiMn₂O₄ electrode.

The S 2p spectra are shown in Fig. 6d. There is only one main peak in S 2p spectra in the electrodes before testing, which is dominated by the –SO₃H bond (168.1 eV).²⁴ For the electrode after testing, a new peak attributing to –(SO₃)₂Mn (170.8 eV)²⁴ can be found, which is also indicates that the Nafion layer can capture the Mn²⁺ from the LiMn₂O₄ electrode.

In summary, a simple electrode modification method using ion exchange polymer coating is reported to overcome severe capacity fading of the LiMn₂O₄ cathode lithium-ion batteries at high temperatures. The Nafion coating layer will not only block the Mn²⁺ ions into the electrolyte but also act as an effective layer to prevent the LMO electrode by the attack of HF formed by the decomposition of the LiPF₆ salt in the electrolyte. Thus, the LMO electrodes after the Nafion coating shows significant high-capacity retention at 55 °C compared with the pristine LMO electrodes. We expect this new method may be readily applicable for commercial LMO cathode LIB products. Further work on the choice and synthesis of the ion exchange polymer, its loading optimization and coating process is under investigation.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (20904031), the Natural Science Foundation of Shanghai (14ZR1422100), and Shanghai Key Lab of Polymer Dielectrics (Shanghai Key Lab of Electrical Insulation and Thermal Aging). Thankful to Instrumental Analysis Center of Shanghai Jiaotong University.

Notes and references

1. J.-M. Tarascon and M. Armand, Nature, 2001, 414, 359.
2. E. Antolini, Solid State Ionics, 2004, 170, 159.
3. N. Yabuuchi and T. Ohzuku, J. Power Sources, 2003, 119–121, 171.
4. A. Deb, U. Bergmann, E. J. Cairns and S. P. Cramer, J. Phys. Chem. B, 2004, 108, 7046.
5. A. B1yr, C. Sigala, G. Amatucci, D. Guyomard, Y. Chabre and J. M. Tarascon, J. Electrochem. Soc., 1998, 145, 194.
6. F. T. Quinlan, K. Sano, T. Willey, R. Vidu, K. Tasaki and P. Stroeve, Chem. Mater., 2001, 13, 4207.
7. J. Shim, R. Kostecki, T. Richardson, X. Song and K. A. Striebel, J. Power Sources, 2002, 112, 222.
8. Y. K. Sun, C. S. Yoon, C. K. Kim, S. G. Youn, Y. S. Lee, M. Yoshioand and I. H. Oh, J. Mater. Chem., 2001, 11, 2519.
9. Y. K. Sun, M. J. Lee, C. S. Yoon, J. Hassoun, K. Amine and B. Scrosati, Adv. Mater., 2012, 24, 1192.
10. Y. C. Sun, Z. X. Wang, L. Q. Chen and X. J. Huang, J. Electrochem. Soc., 2003, 150, 1294.
11. I. H. Cho, S. S. Kim, S. C. Shin and N. S. Choi, Electrochem. Solid-State Lett., 2010, 13, 168.
12. M.-H. Ryou, S. Hong, M. Winter, H. Leeand and J. W. Choi, J. Mater. Chem. A, 2013, 1, 15224.
13. M. S. Whittingham, Chem. Rev., 2004, 104, 4271.
14. H. Liang, X. Qiu, S. Zhang, W. Zhu and L. Chen, J. Appl. Electrochem., 2004, 34, 1211.
15. Q. Tang, Z. Shan, L. Wang, X. Qin, K. Zhu, J. Tian and X. Liu, J. Power Sources, 2014, 246, 253.
16. R. Buzzoni, S. Bordiga, G. Ricchiardi, G. Spoto and A. Zecchina, J. Phys. Chem., 1995, 99, 1937.
17. M. Falk, Can. J. Chem., 1980, 58, 1495.
18. H.-W. Lee, P. Muralidharan, R. Ruffo, C. M. Mari, Y. Cui and D. K. Kim, Nano Lett., 2010, 10, 3852.
19. M. Q. Xu, L. S. Hao, Y. L. Liu, W. S. Li, L. D. Xing and B. Li, J. Phys. Chem. C, 2011, 115, 6087.
20. C. Chen, G. Levitin, D. W. Hess and T. F. Fuller, J. Power Sources, 2007, 169, 290.
21. Y. Kusachi, Z. C. Zhang, J. Dong and K. Amine, J. Phys. Chem. C, 2011, 115, 24013.
22. N. Treuil, C. Labrugere, M. Menetrier, J. Portier, G. Gampet, A. Deshayes, J. C. Frison, S. J. Hwang, S. W. Song and J. H. Choy, J. Phys. Chem. B, 1999, 103, 2100.
23. L. Zhou, S. Dalavi, M. Q. Xu and B. Lucht, J. Power Sources, 2011, 196, 8073.
24. M. Schulze, M. Lorenz, N. Wagner and E. Gülzow, Fresenius. J. Anal. Chem., 1999, 365, 106.

---

Footnote

† Electronic supplementary information (ESI) available: Details of experimental procedures. See DOI: 10.1039/c4ra07209j

---