Tin negative electrodes using an FSA-based ionic liquid electrolyte: improved performance of potassium secondary batteries

Takayuki Yamamoto * and Toshiyuki Nohira *
Institute of Advanced Energy, Kyoto University, Uji 611-0011, Japan. E-mail: yamamoto.takayuki.2w@kyoto-u.ac.jp; nohira.toshiyuki.8r@kyoto-u.ac.jp

Received 9th January 2020 , Accepted 3rd February 2020

First published on 3rd February 2020

In this study, submicron-sized tin particles were used as the negative electrode material for potassium secondary batteries. With a bis(fluorosulfonyl)amide-based ionic liquid electrolyte, K[FSA]–[C3C1pyrr][FSA], the tin negative electrodes showed improved capacity retention of over 170 mA h (g-Sn)−1 after 100 cycles at room temperature.

To make progress toward our ultimate goal of achieving and implementing sustainable energy systems without the use of conventional fossil fuels, closer collaboration between renewable energy resources and energy storage devices is essential to ensure a stable power supply. One possible solution is the wider distribution of large-scale batteries in households, buildings, industrial plants, etc. Although lithium secondary batteries currently installed in portable electric devices are promising candidates, the scarcity of lithium and cobalt resources and the flammability of organic-solvent-based electrolytes increase the likelihood of a future price hike and the risk of fatal accidents. Thus, in recent decades, tremendous efforts have been devoted to the development of alternative batteries that use abundant resources and safer electrolytes.1–5

We are now focusing on potassium secondary batteries using ionic liquid (IL) electrolytes. Potassium resources are plentiful in the Earth's crust and seawater, resulting in their low cost and wider distribution of large-scale battery systems. Moreover, inexpensive aluminum current collectors can be used for the negative electrodes of potassium secondary batteries, whereas the more costly copper is needed for their lithium counterparts. Moreover, ILs are known to be safer electrolytes because of their negligible volatility and nonflammability; they are widely studied as electrolytes in batteries, fuel cells, capacitors, and other devices. Several groups, including ours, have reported the physicochemical properties of IL electrolytes for potassium secondary batteries.6–9 One important advantage is the wider electrochemical window of potassium-based IL electrolytes compared with those of lithium- and sodium-based ones.7,9 For example, the redox potentials of M+/M (M = Li, Na, K) were compared in the M[FSA]–[C3C1pyrr][FSA] IL (FSA = bis(fluorosulfonyl)amide, C3C1pyrr = N-methyl-N-propylpyrrolidinium).7 The redox potential of K+/K is more negative than that of Na+/Na by 0.35 V and even that of Li+/Li by 0.25 V, whereas the anodic decomposition potentials are almost equal within experimental error. Furthermore, the use of this potassium-based IL is beneficial in terms of stability against potassium metal. In some types of organic-solvent-based electrolytes, potassium metal does not work effectively as a counter electrode in a two electrode-type cell.10 In the K[FSA]–[C3C1pyrr][FSA] IL, on the other hand, we recently evaluated the charge–discharge behavior of alloy-based negative electrodes with two-electrode coin cells using a potassium metal counter electrode and obtained reasonably high performance.11 These results imply the superiority of this IL in long-term durability, especially in the negative potential region.

In this study, tin negative electrodes prepared from commercial submicron-sized tin particles are used with the IL electrolyte, and their fundamental electrochemical behavior is investigated via galvanostatic charge–discharge (GCD) tests, galvanostatic intermittent titration technique (GITT), X-ray diffraction (XRD), and field emission scanning electron microscopy (FE-SEM).

The K[FSA]–[C3C1pyrr][FSA] IL (x(K[FSA]) = 0.20, where x(K[FSA]) = molar fraction of K[FSA]; C(K+) = ca. 1 mol dm−3) was prepared by mixing K[FSA] and [C3C1pyrr][FSA]. The tin negative electrodes were composed of tin powder (particle size: 100–200 nm) as the active material, acetylene carbon black (AB) powder as a conductive agent, and polyamide–imide (PAI) as a binder, with a weight ratio of 80/10/10. To prepare the electrodes (denoted as Sn/AB/PAI electrodes) the active material, conductive agent, and binder were mixed in N-methyl-2-pyrrolidone (NMP) and the resultant slurry was coated onto aluminum foil. The NMP was then removed at 353 K for 3 h in a vacuum, and the electrodes were punched out of the foil as circles with diameter 10 mm. The electrodes were dried at 363 K overnight in a vacuum of <1 Pa. The tin loading masses were controlled within the range of 2.0–2.6 (mg-Sn) cm−2.

Fig. 1 shows representative charge–discharge curves of the Sn/AB/PAI electrode in the K[FSA]–[C3C1pyrr][FSA] (x(K[FSA]) = 0.20) IL at a charge–discharge rate of 20 mA (g-Sn)−1. In the 1st charge (potassiation) process, several plateaus and slopes are observed in the voltage region below 1.5 V. The capacity from a small plateau at approximately 1.35 V and a slope region at around 1.0 V may correspond to the irreversible formation of a solid electrolyte interphase (SEI), because amide-based ILs give similar behavior with carbon-based negative electrodes.12,13 According to previous reports of K–Sn alloy formation in organic-solvent-based electrolytes,14–22 a slope region at around 0.5 V and a long plateau at 0.17 V include the potassiation reaction of tin. In contrast, the 1st discharge curve consists of multiple short plateaus, suggesting the existence of several K–Sn alloy phases. The charge and discharge capacities obtained in the 1st cycle are 328 and 167 mA h (g-Sn)−1, respectively. Table S1 (ESI) summarizes the K–Sn alloy phases reported in papers on the phase diagram and on structure determination. The 1st discharge capacity lies between the theoretical capacities for K2Sn3 and KSn formation. Except for the 1st charge, the charge–discharge curves in the initial 5 cycles almost overlap with each other, indicating stable charge–discharge cycling. This is also supported by morphological observation by FE-SEM, as shown in Fig. 2. On the surfaces of the Sn/AB/PAI electrodes, only minor morphological change was confirmed after 5 cycles. Additionally, rate capability was investigated as shown in Fig. S1 (ESI). The tin electrode maintains a capacity of 117 mA h (g-Sn)−1 at a rate of 100 mA (g-Sn)−1.

image file: d0cc00209g-f1.tif
Fig. 1 Charge–discharge curves of the Sn/AB/PAI electrode in K[FSA]–[C3C1pyrr][FSA] IL electrolyte for 5 cycles at 298 K. Charge–discharge rate: 20 mA (g-Sn)−1.

image file: d0cc00209g-f2.tif
Fig. 2 Representative FE-SEM images of Sn/AB/PAI electrodes (a) before the test and (b) after 5 charge–discharge cycles.

The formation of K–Sn alloy phases was confirmed by X-ray diffraction on the fully charged state. Fig. 3 shows the diffraction pattern of the fully charged Sn/AB/PAI electrode prepared by the combination of galvanostatic charging and open-circuit potential relaxation. Several peaks at diffraction angles 2θ ≈ 14.5° (index: 112), 28.6° (321), and 29.7° (215) are ascribed to the KSn phase. However, the intensities of other diffraction peaks are very low, owing to the low crystallinity of the electrochemically formed K–Sn alloy. Similar behavior has been observed in the electrochemical formation of Na–Sn alloys.23 In addition, β-Sn peaks are also present, either because the active material remained partially unreacted, or because the K–Sn alloys produced are decomposed into pure tin and potassium hydroxide by incomplete airtightness of the supposedly air-tight XRD cell. According to the phase diagram, the most potassium-rich phase is K2Sn, whose crystal structure is unknown. However, in some reports on tin-based negative electrodes for potassium secondary batteries using organic-solvent-based electrolytes,15,17,20,21 KSn is the richest phase in the electrochemical reaction. In the present study on the FSA-based IL electrolyte, the fully charged state of the Sn/AB/PAI electrode is reasonably considered to be KSn, based on the reversible capacities and the X-ray diffraction results.

image file: d0cc00209g-f3.tif
Fig. 3 X-ray diffraction pattern of the fully charged Sn/AB/PAI electrode.

Further investigation was conducted by the galvanostatic intermittent titration technique (GITT), as shown in Fig. S2 (ESI). The GITT measurements after 1 charge–discharge cycle indicate the existence of several slope and plateau regions during the charging process, as observed in the galvanostatic charge–discharge tests. The capacity at the end of the slope corresponds to the theoretical capacity of K8Sn46 (K4Sn23). The subsequent long plateau at 0.35 V vs. K+/K continues over the compositional range 0.2 < x < 0.7 of KxSn. Judging from the list of K–Sn alloy phases (Table S1, ESI), this plateau is attributed to the equilibrium between K8Sn46 (x = 0.17) and K2Sn3 (x = 0.67). The final slope reaches almost 0 V vs. K+/K at the composition where x = 1.19 (270 mA h (g-Sn)−1), which slightly exceeds that of the K–Sn alloy phase (KSn; x = 1) observed by X-ray diffraction. This phenomenon may have two possible explanations: (1) irreversible decomposition of the electrolyte at lower potential, or (2) the existence of a solid-solution compositional region at around x = 1. In contrast, multiple plateaus are confirmed in the discharging process (Fig. S2c, ESI), which is consistent with the results of the charge–discharge test (see Fig. 1 and Fig. S2a, ESI). Since dealloying of K–Sn alloys begins from the second plot at 0 V vs. K+/K, we tentatively assume that this point corresponds to the most potassium-rich composition of the K–Sn alloy, i.e., xmax = 1.19. The first sloping region between 0 and 0.5 V is typical behavior for solid solutions, implying the possibility (2) proposed above in the discussion of the charging GITT measurements. In the compositional range x < 1, the profile of the open circuit potentials is roughly divided into several parts, as described by the broken lines. Although this assignment almost agrees with the reported K–Sn alloy phases, further detailed examination is needed in the future. Finally, the proportion (x) of potassium in tin comes back to nearly zero, indicating that the fully discharged state is pure metallic tin.

The prolonged charge–discharge performance was evaluated for Sn/AB/PAI electrodes at room temperature. According to the charge–discharge curves shown in Fig. 4a, the charging (potassiation) profiles are stable for 100 cycles, except for the 1st cycle. Concerning the discharge (depotassiation) process, the polarization of plateaus at around 1.0 V becomes slightly larger after the 10th cycle, whereas no significant change is confirmed in the middle plateau region at around 0.7 V. Such a different trend is possibly explained by the degree of volume contraction for each reaction step. When the discharge reactions are expressed with K–Sn alloy phases whose structures are known (eqn (1) and (2)), the contraction ratios are calculated to be 0.50 (=16.29/32.67) for region (A) at ca. 1.0 V and 0.72 (=32.67/45.58) for region (B) at ca. 0.7 V, using the molar volumes of K–Sn alloys summarized in Table S1 (ESI).

(Region (A) at ca. 1.0 V) K4Sn9 → 9Sn + 4K+ + 4e(1)
(Region (B) at ca. 0.7 V) 9KSn → K4Sn9 + 5K+ + 5e(2)

image file: d0cc00209g-f4.tif
Fig. 4 (a) Charge–discharge curves, and (b) cycling properties of specific capacities and coulombic efficiencies of the Sn/AB/PAI electrode in K[FSA]–[C3C1pyrr][FSA] IL electrolyte at 298 K. Charge–discharge rate: 20 mA (g-Sn)−1.

Since the contraction ratio is lower for region (A), the degradation will progress faster for that reaction, leading to large polarization in the earlier cycles. Similar behavior was also confirmed in our previous studies on Na–Sn negative electrodes.24Fig. 4b shows cycling properties of the Sn/AB/PAI electrode for 100 cycles at room temperature. The discharge capacity at the 100th cycle is 173 mA h (g-Sn)−1, corresponding to a capacity retention ratio of 93% with respect to the 1st cycle. The average coulombic efficiency (except for the 1st cycle) is 97.5% over 100 cycles. In addition, a charge–discharge test at elevated temperature was conducted to study the improvement in reversible capacities. As shown in Fig. S3 (ESI), although the capacity increased to nearly 200 mA h (g-Sn)−1 at 313 K, the production of phases more potassium-rich than KSn is unlikely. Thus, further studies are necessary to elucidate the K–Sn formation mechanism.

Table 1 summarizes the present and previous studies on tin negative electrodes for potassium secondary batteries, in which organic-solvent-based electrolytes were utilized in all cases except ours. One important point to be mentioned is the operating voltage range. Most studies adopt upper cut-off voltages higher than 1.5 V;14,16–19,21,22 however, such operation ranges wider than 1.5 V are not realistic conditions for practical batteries. In addition to the reported initial discharge capacities, the estimated initial capacities and nth cycle capacities, based on a realistic upper cut-off voltage of 1.2 V, are given in the middle and right columns of the discharge capacity. Several previous studies achieved relatively high initial capacities of over 200 mA h g−1 even for a realistic voltage range, but a large deterioration in capacity was observed after prolonged cycling. In the present study, however, the highest capacity of 173 mA h g−1 was retained even after 100 cycles. Simple commercial tin powders with no carbon coating exhibit reasonably high capacities and decent cycling properties in the FSA-based IL electrolyte, because of the stability of the IL in negative potential regions. In other words, stable SEI film formation may give superior durability of tin negative electrodes, as with the case of negative electrodes for sodium secondary batteries.5,13

Table 1 Comparison of present and previous studies on tin negative electrodes for potassium secondary batteries
Type of tin electrode (particle size) Electrolyte Discharge capacity/mA h (g-active material)−1 Charge–discharge rate/mA g−1 Active material Binderb Ref.
1st cycle (cut-off voltage) 1st cycle (<1.2 V) nth cycle (<1.2 V)
a The capacities were recalculated based on the composition of the tin electrode. b PAI = polyamide–imide, CMC = carboxymethyl cellulose, SA = sodium alginate, PVdF = polyvinylidene difluoride, SBR = styrene butadiene rubber.
Sn powder (100–200 nm) 1 mol dm−3 K[FSA]–[C3C1pyrr][FSA] 186 (0.005–1.2 V) 186 173 (100th) 20 Sn PAI This study
Sn–C composite 0.75 mol dm−3 KPF6 in EC–DEC 140 (0.01–2.0 V) ∼100 ∼80 (15th) 25 Sn + C CMC 14
200 nm thin Sn film 1 mol dm−3 KPF6 in EC–DEC ∼225 (0.01–1.2 V) ∼225 85 (20th) 25 Sn 15
Sn powder (70–350 nm) 0.8 mol dm−3 KPF6 in EC–DEC 197 (0.01–2 V) ∼180 <50 (10th) 20 Sn SA 16
Sn/C composite 0.8 mol dm−3 KPF6 in EC–DEC ∼200 (0.01–2 V) ∼150 <50 (30th) 50 Sn CMC 17
Porous carbon/Sn composite 0.8 mol dm−3 KClO4 in EC–DEC 385 (0.01–3 V) ∼190 ∼140 (100th) 50 Sn + C PVdF 18
Sn powder (70–500 nm) 0.5 mol dm−3 KPF6 in EC–DEC 229 (0.005–2.0 V) ∼210 <50 (20th) 25 Sn CMC–SBR 19
Sn powder 0.8 mol dm−3 KFSA in EC–DEC ∼240 (0.0–1.1 V) ∼240 (0.0–1.1 V) ∼110 (50th) 23 Sna CMC 20
Sn@RGO 0.8 mol dm−3 KPF6 in EC–DEC ∼240 (0.001–3.0 V) ∼140 ∼115 (50th) 100 Sn + C CMC 21
Sn nanoparticles anchored on N doped porous carbon 1 mol dm−3 KPF6 in EC–DEC 286 (0.01–2.6 V) ∼125 ∼90 (100th) 50 Sn + C PVdF 22

In conclusion, the charge–discharge behavior of tin negative electrodes has been evaluated using the K[FSA]–[C3C1pyrr][FSA] IL electrolyte. The Sn/AB/PAI electrodes showed high capacities of over 170 mA h (g-Sn)−1, stable over 100 cycles at room temperature, owing to the electrochemical stability of the IL. The electrochemical alloying/dealloying mechanism was partly clarified by a combination of the galvanostatic intermittent titration technique and X-ray diffraction. To further enhance the performance of potassium secondary batteries, the mismatch between phase diagrams and observed capacities needs to be resolved in future studies.

This study was partly supported by JSPS KAKENHI grant (JP18K14320) and the research grant of Izumi Science and Technology Foundation. We thank Nippon Shokubai Co., Ltd for supplying K[FSA]. The polyamide–imide binder was provided by Prof. R. Hagiwara and K. Matsumoto in Kyoto University.

Conflicts of interest

The authors declare no competing financial interest.

Notes and references

  1. N. Yabuuchi, K. Kubota, M. Dahbi and S. Komaba, Chem. Rev., 2014, 114, 11636–11682 CrossRef CAS .
  2. D. Larcher and J.-M. Tarascon, Nat. Chem., 2015, 7, 19–29 CrossRef CAS PubMed .
  3. M. Watanabe, M. L. Thomas, S. Zhang, K. Ueno, T. Yasuda and K. Dokko, Chem. Rev., 2017, 117, 7190–7239 CrossRef CAS PubMed .
  4. A. Eftekhari, Z. Jian and X. Ji, ACS Appl. Mater. Interfaces, 2017, 9, 4404–4419 CrossRef CAS PubMed .
  5. K. Matsumoto, J. Hwang, S. Kaushik, C.-Y. Chen and R. Hagiwara, Energy Environ. Sci., 2019, 12, 3247–3287 RSC .
  6. K. Beltrop, S. Beuker, A. Heckmann, M. Winter and T. Plack, Energy Environ. Sci., 2017, 10, 2090–2094 RSC .
  7. T. Yamamoto, K. Matsumoto, R. Hagiwara and T. Nohira, J. Phys. Chem. C, 2017, 121, 18450–18458 CrossRef CAS .
  8. T. Masese, K. Yoshii, Y. Yamaguchi, T. Okumura, Z.-D. Huang, M. Kato, K. Kubota, J. Furutani, Y. Orikasa, H. Senoh, H. Sakaebe and M. Shikano, Nat. Commun., 2018, 9, 3823 CrossRef PubMed .
  9. K. Yoshii, T. Masese, M. Kato, K. Kubota, H. Senoh and M. Shikano, ChemElectroChem, 2019, 6, 3901–3910 CrossRef CAS .
  10. T. Hosaka, S. Muratsubaki, K. Kubota, H. Onuma and S. Komaba, J. Phys. Chem. Lett., 2019, 10, 3296–3300 CrossRef CAS PubMed .
  11. Y. Domi, H. Usui, E. Nakabayashi, T. Yamamoto, T. Nohira and H. Sakaguchi, Electrochemistry, 2019, 87, 333–335 CrossRef CAS .
  12. A. Fukunaga, T. Nohira, R. Hagiwara, K. Numata, E. Itani, S. Sakai, K. Nitta and S. Inazawa, J. Power Sources, 2014, 246, 387–391 CrossRef CAS .
  13. T. Yamamoto, T. Yamaguchi, T. Nohira, R. Hagiwara, A. Fukunaga, S. Sakai and K. Nitta, Electrochemistry, 2017, 85, 391–396 CrossRef CAS .
  14. I. Sultana, T. Ramireddy, M. M. Rahman, Y. Chen and A. M. Glushenkov, Chem. Commun., 2016, 52, 9279–9282 RSC .
  15. T. Ramireddy, R. Kali, M. K. Jangid, V. Srihari, H. K. Poswal and A. Mukhopadhyay, J. Electrochem. Soc., 2017, 164, A2360–A2367 CrossRef CAS .
  16. Q. Wang, X. Zhao, C. Ni, H. Tian, J. Li, Z. Zhang, S. X. Mao, J. Wang and Y. Xu, J. Phys. Chem. C, 2017, 121, 12652–12657 CrossRef CAS .
  17. W. Zhang, J. Mao, S. Li, Z. Chen and Z. Guo, J. Am. Chem. Soc., 2017, 139, 3316–3319 CrossRef CAS PubMed .
  18. K. Huang, Z. Xing, L. Wang, X. Wu, W. Zhao, X. Qi, H. Wang and Z. Ju, J. Mater. Chem. A, 2018, 6, 434–442 RSC .
  19. M. Shimizu, R. Yatsuzuka, T. Koya, T. Yamakami and S. Arai, ACS Appl. Energy Mater., 2018, 1, 6865–6870 CrossRef CAS .
  20. V. Gabaudan, R. Berthelot, M. T. Sougrati, P.-E. Lippens, L. Monconduit and L. Stievano, J. Mater. Chem. A, 2019, 7, 15262–15270 RSC .
  21. H. Wang, Z. Xing, Z. Hu, Y. Zhang, Y. Hu, Y. Sun, Z. Ju and Q. Zhuang, Appl. Mater. Today, 2019, 15, 58–66 CrossRef .
  22. Y. Yang, D. Li, J. Zhang, G. Suo, Q. Yu, L. Feng, X. Hou, X. Ye, L. Zhang and W. Wang, Mater. Lett., 2019, 256, 126613 CrossRef CAS .
  23. T. Yamamoto, T. Nohira, R. Hagiwara, A. Fukunaga, S. Sakai, K. Nitta and S. Inazawa, J. Power Sources, 2013, 237, 98–103 CrossRef CAS .
  24. T. Yamamoto, T. Nohira, R. Hagiwara, A. Fukunaga, S. Sakai, K. Nitta and S. Inazawa, J. Power Sources, 2012, 217, 479–484 CrossRef CAS .


Electronic supplementary information (ESI) available: Experimental details, summary table of K–Sn alloy phases, GITT measurements, charge–discharge test at 313 K, and rate capability at 298 K. See DOI: 10.1039/d0cc00209g

This journal is © The Royal Society of Chemistry 2020