Enhancing the safety and electrochemical performance of ether based lithium sulfur batteries by introducing an efficient flame retarding additive

Abstract

Lithium sulfur batteries have been considered as a promising candidate for use as next generation high energy power sources. However, safety problems could be one key problem that hinders the development of lithium sulfur batteries. In this study, a novel flame retarding additive, hexafluorocyclotriphosphazene (HFPN), is investigated for the construction of an ether based (1,3-dioxolane and dimethoxyethane) nonflammable electrolyte for lithium sulfur batteries. A 20% addition could render the electrolyte nonflammable. Moreover, this additive could enhance the electrochemical properties of lithium sulfur batteries by reducing the solubility of polysulfides and reducing the electrode interphase resistance. These results suggest that HFPN could be considered as a useful additive for safer lithium sulfur batteries.

---

1. Introduction

Lithium sulfur batteries have been considered as next generation batteries for portable electronics and large-scale electric vehicles, because of the abundance and green character of sulfur, and their high energy density (2567 W h kg⁻¹) and high theoretical specific capacity (1672 mA h g⁻¹).^1–8 However, safety concerns remain as one key hindrance in the further application of lithium sulfur batteries. State-of-the-art lithium batteries make use of flammable, voltage sensitive electrolyte solvents, and highly reactive lithium metal and sulfur electrodes, which might cause serious hazards such as fire and explosion under abusive conditions such as short-circuiting, heating, crashing, overcharging, etc.^9–11 . Nonflammable electrolytes could be the most effective way to solve these problems. A large number of brilliant work has been done on safer electrolytes for lithium sulfur batteries, such as solid state electrolytes, polymer based electrolytes and ionic liquids.^12–14 However, owing to their high ionic conductivity, high polysulfide solubility and chemical stability, ether based solvents such as 1,3-dioxolane (DOL), dimethoxyethane (DME) and tetra(ethylene glycol)dimethyl ether (TEGDME) still remain the domain of research into lithium sulfur electrolytes.^15–20 Reports on ether based nonflammable electrolyte solvents are rare.^21–23

In this study, several kinds of flame retarding additives are probed to develop nonflammable electrolytes for the popular 1,3-dioxolane (DOL) and dimethoxyethane (DME) based lithium sulfur electrolytes, and four intensely studied nonflammable solvents for lithium ion batteries, tri(2,2,2-trifluoroethyl)phosphate (TFEP), (ethoxy)pentafluorocyclotriphosphazene-1,1,2,2-tetrafluoroethyl-2,2,3,3-tetrafluoropropylether (HFE-458) and hexafluorocyclotriphosphazene (HFPN), are selected as flame retarding additives for lithium sulfur batteries.^24–32 It is found that HFPN is an efficient flame retarding additive for ether based electrolytes. The electrochemical performance of sulfur in this electrolyte is also discussed.

2. Experimental

All solvents were of analytical grade and used as received, HFPN was bought from TCI. The electrolytes were prepared by mixing HFPN with 1 M LiTFSI DOL/DME (1/1 v/v) at different weight ratios. To examine the flame retarding ability of HFPN, we measured the self-extinguishing time (SET) of the electrolyte solutions, using a similar method to that described in ref. 33 and 34. The typical procedure for SET measurements is to use a 2032 coin cell to absorb 0.5 g of electrolyte, and then to burn the electrolyte and record the burning time. Each test was repeated four times and the burning times were averaged. To study the change in the morphology of the electrodes after overcharging, the electrode films were washed using tetrahydrofuran followed by drying under vacuum at 50 °C.

The electrochemical window of the electrolytes was examined using cyclic voltammetry (CV), using a stainless steel electrode as the working electrode and a lithium sheet as both the counter electrode and reference electrode. The data acquisition and analysis were performed on a CHI electrochemical workstation (CHI660e, China). The conductivity of the electrolyte was measured using a conductivity meter (DDS-11A, Leici Co. Ltd, Shanghai, China).

The electrochemical compatibility of the electrolytes was examined using 2016 coin type cells. The sulfur electrode contained 50% elemental sulfur, 40% Ketjen Black (EC 600 JD) and 10% sodium carboxymethyl cellulose (CMC) binder. The slurry was cast on aluminum foil and vacuum dried at 70 °C for 24 hours. All the cells were assembled in a glove box with a water/oxygen content lower than 1 ppm and tested at ambient temperature. Cyclic voltammetry measurements were carried out using two electrode half cells at a scanning rate of 0.1 mV s⁻¹ using a CHI660E electrochemical workstation (Chenhua, Shanghai, China). The charge–discharge measurements were carried out using a computer-controlled programmable battery charger (BTS-0518001 type, Shenzhen, China). The sulfur half cell was measured between 1.5–3 V (vs. Li/Li⁺) at a rate of 100 mA g⁻¹.

3. Results and discussion

3.1. Physical properties and characterization of HFPN based electrolyte

The structures and flame retarding abilities towards 1 M LiTFSI DOL/DME (1/1 v/v) of HFPN, EFPN, HFE-458 and TFEP are displayed in Fig. 1. From the figure we can see that HFPN shows the best flame retarding ability; 20% HFPN could render the electrolyte nonflammable. In contrast, the electrolyte with a 60% addition of EFPN, HFE-258 and TFEP is still flammable. It is well established that the most favorable flame retarding mechanism is oxygen isolation and radical elimination in the vapor phase, so the boiling point of flame retarding additives should be close to the flammable components. By comparing the boiling points, we can see that the boiling points of EFPN (125 °C), HFE-458 (93 °C) and TFEP (186 °C) are much higher than the boiling points of DOL (74 °C) and DME (83 °C), while the boiling point of HFPN is about 51 °C. When ignited, the HFPN could evaporate together with DOL and DME to dilute the oxygen content and eliminate the flammable H radical. Thus, HFPN has the best flame retarding ability, and we choose HFPN as the flame retarding additive whose physical and electrochemical properties were measured.³⁵

Fig. 1 Chemical structures and flame retarding ability of HFPN, EFPN, HFE-258 and TFEP on 1 M LiTFSI DOL/DME (1/1 v/v) electrolyte.

We chose 1 M LiTFSI DOL/DME (1 : 1 v/v) with 2 wt% LiNO₃ as the base electrolyte. The effect of the HFPN content on the SET and conductivity is shown in Fig. 2a. From the figure we can see that the SET of the base electrolyte drops to zero when the HFPN content is increased to 20%. The most popular flame retarding mechanism of P and F containing organics is the “radical elimination mechanism”. When ignited, HFPN decomposes and generates F and P radicals which could combine with H radicals and stop flame propagation. Moreover, the N element in HFPN could consume and dilute oxygen in the atmosphere to prevent the combustion progress.¹⁰ Fig. 2a also shows the conductivity change of 1 M LiTFSI DOL/DME (1 : 1 v/v) as a function of the HFPN content. From the figure we can see that the conductivity increases with the addition of HFPN, which may due to the low viscosity of HFPN. With the further increase in HFPN content, the conductivity slowly decreases, which can be attributed to the low polarity of HFPN. Further research may focus on the optimization of the electrolyte component. Fig. 2b shows the burn test of the electrolyte with and without 20% HFPN; from the figure we can see that the base electrolyte is highly flammable, while the electrolyte with 20% HFPN cannot be ignited, illustrating excellent flame retarding ability.³⁵

Fig. 2 The SET test data and the ionic conductivity of electrolytes containing different ratios of HFPN (a) and the ignition test on blank and 20% HFPN containing electrolyte (b).

3.2. Electrochemical behavior of HFPN containing electrolyte

To study the electrochemical character of a sulfur electrode in blank and HFPN containing electrolyte, CV analysis was also performed. The results are shown in Fig. 3a and b, respectively. From the figure we can see that the sulfur electrode in HFPN containing electrolyte shows a similar shape to the blank electrolyte, with two reduction peaks in the cathodic scan and one oxidation peak in the anodic scan. The first reduction peak is ascribed to the formation of Li₂S_n (8 > n > 4) and the second reduction peak is ascribed to the formation of Li₂S_n (4 > n > 2), while the oxidation peak near 2.5 V is caused by oxidation to lithium polysulfide (Li₂S_n, n > 2). This result suggests that sulfur works well in HFPN containing electrolyte.^35,36

Fig. 3 CV measurements of sulfur electrodes in blank (a) and 20% HFPN containing electrolyte (b) at a scan rate of 0.5 mV s⁻¹.

Fig. 4a and b show initial charge/discharge profiles of sulfur electrodes in 20% HFPN and base electrolytes at the rate of 100 mA g⁻¹ between 1.5–3 V (vs. Li/Li⁺). For the sulfur electrode in 20% HFPN containing electrolyte (Fig. 4a), two distinct voltage plateaus were observed at around 2.4 and 2.1 V, which is consistent with the blank electrolyte (Fig. 4b) and previous results. It is generally accepted that the plateau at about 2.4 V corresponds to the reduction of elemental sulfur to soluble, long-chain polysulfides. The lower plateau at 2.1 V reflects the reduction of short-chain polysulfides, which results in the formation of insoluble products, such as Li₂S₂ and Li₂S.^36,37 Fig. 4c shows the cycling performance of sulfur electrodes with 20% HFPN and blank electrolytes. The reversible capacity of the sulfur electrode in HFPN-containing electrolyte is higher (360 mA h g⁻¹) than in blank electrolyte (290 mA h g⁻¹) after 100 cycles, exhibiting a better cycle ability.

Fig. 4 The 1^st, 2^nd, and 100^th charge–discharge curves of electrolyte with 20% HFPN (a), and blank (b) electrolyte, the cycling performance of sulfur/Li half cells in blank and 20% HFPN containing 1 M LiTFSI DOL/DME (1/1 v/v) + 2 wt% LiNO₃ electrolyte (c) at a rate of 100 mA g⁻¹ and the EIS results of cells before cycling, and after the 10^th, 20^th and 50^th cycle in HFPN containing (d) and blank (inset) electrolyte.

To better understand the different cycling performance of sulfur electrodes in different electrolytes, electrochemical impedance spectroscopy (EIS) studies (Fig. 4d) were carried out using a fresh half-cell and the 10^th, 20^th and 50^th cycle half-cells at 1000 mA g⁻¹ to study the kinetic mechanisms of lithium storage. The Nyquist type plots were recorded in the frequency range of 0.01 Hz to 1 MHz at an amplitude of 5 mV. In the EIS study, the sulfur electrodes were used as the working electrodes with lithium foil as both the anode and reference electrodes. The results are typical Nyquist plots with one compressed semicircle at high frequency and a straight line at low frequency.¹⁷ The semicircle is believed to be attributed to surface film impedance at high frequency and charge transfer resistance at middle frequencies. From Fig. 4d, we can see that the resistance of the fresh cell is much higher than the cycled cells, which may be due to the dissolution of polysulfides. However, from the 10^th to 50^th cycles, the resistance increases gradually, which may be due to side reactions between the lithium metal and the polysulfides. The total resistance increase of the HFPN containing cell is slower than the cell with blank electrolyte. These results suggest that HFPN addition could lower the total resistance.³⁵

The morphology of the fresh and cycled sulfur electrodes was also investigated using SEM. Fig. 5c and d show low and high magnification SEM images of the sulfur electrode with 20% HFPN containing electrolyte after 100 cycles. The surface of the cathode was in a more porous state than the original electrode (Fig. 5a and b), which may facilitate the lithium redox reaction and reduce polysulfide dissolution. However, the electrode cycled in blank electrolyte (Fig. 5e and f) is partly covered with thick films, which may inhibit the full utilization of sulfur and increase the resistance. The phenomenon further confirmed that HFPN addition has a positive effect on the normal charge–discharge behavior, which may benefit from the F element in HFPN.^38,39 The high solubility of polysulfides in conventional DOL/DME (1/1 v/v) based electrolytes results in the dissolution of the polysulfide species into the electrolyte, and their side reactions at the anode may cause a shuttling phenomenon and the loss of active materials.^2,38 To confirm the effect of HFPN on the cell performance, a solubility test was conducted by placing 0.1 g of Li₂S_x (6 < x < 8) in 3 ml of HFPN or 1 M LiTFSI DOL/DME (1 : 1 v/v) solution. The results are shown in the inset of Fig. 5. It can be seen that Li₂S_x (6 < x < 8) dissolved in the 1 M LiTFSI DOL/DME (1 : 1 v/v) electrolyte and formed a dark-brown solution. In contrast, Li₂S₈ is almost insoluble in HFPN. The low solubility of polysulfides may account for the better cell performance in HFPN containing electrolyte.^38–40

Fig. 5 SEM images of the sulfur cathode before cycling (a and b), and after 100 cycles in HFPN containing (c and d), and blank (e and f) electrolyte, and the solubility test of Li₂S_x (x = 6–8) in pure HFPN and 1 M LiTFSI DOL/DME (1/1 v/v) + 2 wt% LiNO₃ electrolyte (inset).

4. Conclusions

In summary, we have demonstrated a nonflammable ether based electrolyte for general lithium sulfur batteries with a novel high efficiency flame retarding additive – HFPN. 20% HFPN addition could render the 1 M LiTFSI DOL/DME (1 : 1 v/v) electrolyte nonflammable. Moreover, HFPN has a negligible effect on the conductivity and could enhance the electrochemical performance of lithium sulfur batteries. This study can be useful for the design of highly safe lithium sulfur battery systems.

Acknowledgements

This work was supported by the Key Research Plan of Shandong Province (2015GGE27286), the National Natural Science Foundation of China (No. 21371108), the Shandong Provincial Natural Science Foundation for Distinguished Young Scholars (JQ201304), and the Project of the Taishan Scholar (No. ts201511004).

References

1. B. Dunn, H. Kamath and J. M. Tarascon, Science, 2011, 334, 928–935.
2. J. Schuster, G. He, B. Mandlmeier, T. Yim, K. T. Lee, T. Bein and L. F. Nazar, Angew. Chem., Int. Ed., 2012, 51, 3591–3595.
3. Z. Y. Wang, Y. F. Dong, H. J. Li, Z. B. Zhao, H. B. Wu, C. Hao, S. H. Liu, J. S. Qiu and X. W. Lou, Nat. Commun., 2014, 5, 5002.
4. L. Zhen, L. Ma, K. E. Hendrickson, S. Y. Wei and L. A. Archer, Nano Today, 2015, 10, 315–338.
5. Y. X. Yin, S. Xin, Y. G. Guo and L. J. Wan, Angew. Chem., Int. Ed., 2013, 52, 13186–13200.
6. R. J. Chen, T. Zhao and F. Wu, Chem. Commun., 2015, 51, 18–33.
7. G. Y. Xu, B. Ding, P. Nie, L. F. Shen and X. G. Zhang, J. Mater. Chem. A, 2014, 2, 12662–12676.
8. Y. M. Huang, L. X. Yuan, Z. X. Hao and Y. H. Huang, Carbon, 2015, 92, 41–63.
9. R. G. Cao, W. Xu, D. P. Lv, J. Xiao and J. G. Zhang, Adv. Energy Mater., 2015, 5, 1402273.
10. J. K. Feng, Y. L. An, L. J. Ci and S. L. Xiong, J. Mater. Chem. A, 2015, 3, 14539–14544.
11. P. G. Balakrishnan, R. Ramesh and T. Prem Kumar, J. Power Sources, 2006, 155, 401–411.
12. A. Rosenman, E. Markevich, G. Salitra, D. Aurbach, A. Garsuch and F. F. Chesneau, Adv. Energy Mater., 2015, 5, 1500212.
13. J. K. Feng, L. J. Ci and S. L. Xiong, RSC Adv., 2015, 5, 96649–96652.
14. G. Y. Xu, B. Ding, J. Pan, P. Nie, L. F. Shen and X. G. Zhang, J. Mater. Chem. A, 2014, 2, 12662–12676.
15. J. Scheers, S. Fantini and P. Johansson, J. Power Sources, 2014, 255, 204–218.
16. F. J. Lin, J. L. Wang, H. Jia, C. W. Monroe, J. Yang and T. N. Nuli, J. Power Sources, 2013, 223, 18–22.
17. J. L. Wang, F. J. Lin, H. Jia, J. Yang, C. W. Monroe and Y. N. Nuli, Angew. Chem., Int. Ed., 2013, 52, 13186–13200.
18. J. L. Wang, J. Yang, J. Y. Xie and N. X. Xu, Adv. Mater., 2002, 14, 963–965.
19. S. Y. Wei, L. Ma, K. E. Hendrickson, Z. Y. Tu and L. A. Archer, J. Am. Chem. Soc., 2015, 137, 12143–12152.
20. C. Luo, Y. J. Zhu, Y. Wen, J. J. Wang and C. S. Wang, Adv. Funct. Mater., 2014, 24, 4082–4089.
21. J. Gao, M. A. Lowe, Y. Kiya and H. D. Abruna, J. Phys. Chem. C, 2015, 115, 25132–25137.
22. C. Barchasz, J. C. Leprêtre, S. Patoux and F. Alloin, Electrochim. Acta, 2013, 89, 737–743.
23. J. K. Feng, Z. Zhang, L. F. Li, J. Yang, S. L. Xiong and Y. T. Qian, J. Power Sources, 2015, 284, 222–226.
24. S. S. Zhang, K. Xu and T. R. Jow, J. Power Sources, 2003, 113, 166–172.
25. J. K. Feng, X. P. Gao, L. J. Ci and S. L. Xiong, RSC Adv., 2016, 6, 7224–7228.
26. Z. Q. Zeng, B. B. Wu, L. F. Xiao, X. Y. Jiang, Y. Chen, X. P. Ai, H. X. Yang and Y. L. Cao, J. Power Sources, 2015, 279, 6–12.
27. J. K. Feng, X. J. Sun, X. P. Ai, Y. L. Cao and H. X. Yang, J. Power Sources, 2008, 184, 570–573.
28. L. Xia, Y. G. Xia and Z. P. Liu, J. Power Sources, 2014, 278, 190–196.
29. J. K. Feng, Y. L. An, L. J. Ci and S. L. Xiong, J. Mater. Chem. A, 2015, 3, 14539–14544.
30. K. Naoi, E. Iwama, N. Ogihara, Y. Nakamura, H. Segawa and Y. Ino, J. Electrochem. Soc., 2009, 156, A272–A276.
31. G. Nagasubramanian and K. Fenton, Electrochim. Acta, 2013, 101, 3–10.
32. L. Kapicka, D. Dastych, V. Richterova, M. Alberti and P. Kubacek, Magn. Reson. Chem., 2005, 43, 294–301.
33. J. K. Feng and L. Lu, J. Power Sources, 2013, 243, 29–32.
34. J. K. Feng, P. Ma, H. X. Yang and L. Lu, Electrochim. Acta, 2013, 114, 688–692.
35. K. Xu, Chem. Rev., 2004, 104, 4303–4417.
36. P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J. M. Tarascon, Nat. Mater., 2012, 11, 19–29.
37. J. Ye, F. He, J. Nie, Y. L. Cao, H. X. Yang and X. P. Ai, J. Mater. Chem. A, 2015, 3, 7406–7412.
38. N. Azimi, Z. Xue, I. Bloom, M. L. Gordin, D. H. Wang, T. Daniel, C. Takoudis and Z. Z. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 9169–9177.
39. L. C. Zeng, Y. Jiang, J. Xu, M. Wang, W. H. Li and Y. Yu, Nanoscale, 2015, 7, 10940–10949.
40. X. Liang, C. Hart, Q. Pang, A. Garsuch, T. Weiss and L. F. Nazar, Nat. Commun., 2015, 6, 5682.

---