CHEMSUSCHEM COMMUNICATIONS DOI: 10.1002/cssc.201402177
Paving the Way for Using Li2S Batteries Rui Xu,[a, c] Xiaofeng Zhang,[a] Cun Yu,[d] Yang Ren,[d] James C. M. Li,[c] and Ilias Belharouak*[a, b] In this work, a novel lithium–sulfur battery was developed comprising Li2S as the cathode, lithium metal as the anode and polysulfide-based solution as the electrolyte. The electrochemical performances of these Li2S-based cells strongly depended upon the nature of the electrolytes. In the presence of the conventional electrolyte that consisted of lithium bis(trifluoromethanesulfonyl)-imide (LiTFSI) salt dissolved in a solvent combination of dimethoxyethane (DME)/1,3-dioxolane (DOL), the Li/Li2S cells showed sluggish kinetics, which translated into poor cycling and capacity retention. However, when using small amounts of polysulfides in the electrolyte along with a shuttle inhibitor the Li2S cathode was efficiently activated in the cell with the generation of over 1000 mAh g1 capacity and good cycle life.
Li-S batteries are attractive for high-energy electrochemical storage devices, because sulfur is abundant, nontoxic and lowcost.[1–5] A rechargeable Li–S battery is based on the reversible redox process between sulfur and lithium via the electrochemical reaction: S8 + 16 Li + + 16 e$8 Li2S. Conventional Li-S batteries consist of elemental sulfur as the cathode, a non-aqueous liquid as the electrolyte, and lithium metal as the anode. The sulfur cathode offers superior theoretical capacity (1672 mAh g1) compared to all Li-ion battery cathode chemistries (300 mAh g1).[6–8] Despite these advantages, capacity decline and low coulombic efficiency have presented significant challenges to the development of practical Li–S batteries. The issue is related to the dissolution of lithium polysulfides (Li2Sx) (x = 4 to 8), from the sulfur electrode into the electrolyte, which leads to a redox shuttle reaction in the cell.[9–14] In particular, all up-to-date Li-S battery configurations require the use
[a] Dr. R. Xu, Dr. X. Zhang, Dr. I. Belharouak Chemical Sciences and Engineering Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA) E-mail: [email protected] [b] Dr. I. Belharouak Qatar Environment and Energy Research Institute Qatar Foundation P.O. Box 5825, Doha (Qatar) [c] Dr. R. Xu, Prof. J. C. M. Li Materials Science Program Department of Mechanical Engineering University of Rochester Rochester, NY 14627 (USA) [d] C. Yu, Dr. Y. Ren Advanced Photon Source Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA)
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
of metallic lithium and this poses significant risk to the development of practical batteries. Recently, lithium disulfide (Li2S) cathode (theoretical capacity 1166 mAh g1) has gained great attention due to the potential of using forms of anodes other than lithium metal, such as silicon and metal alloys.[15, 16] However, Li2S does not provide the full extent of its capacity in the cell because it is barely active due to its highly insulating character.[17, 18] Therefore, researchers have devoted efforts to improve the conductivity of the Li2S electrode and mitigate sulfur dissolution. As an example, Li2S powder has been ball-milled or mixed with different carbon forms or metals.[9, 10, 19–24] Others have suggested the use of all-solid-state cells.[25, 26] As a result, improvements have been made in terms of initial capacities; however, issues such as poor cycle life and cell efficiency remain unsolved. Herein, we describe the use of a non-conventional electrolyte in a Li2S battery. Inspired by our recent work on sulfur batteries, we developed a Li/Li2S cell comprising a catholytebased electrolyte that is made of polysulfides. This electrolyte significantly improved the electrochemical activity of Li2S by delivering high capacity and reasonable cycling. Figure 1 a and b show typical voltage profile and cycling of the Li2S cell using an electrolyte consisting of 1 m LiTFSI (lithium bis(trifluoromethanesulfonyl)-imide) in dimethoxyethane/ 1,3-dioxolane (DME/DOL). In the initial charge (Figure 1 a), the voltage quickly reached 3.0 V and then declined to 2.6 V followed by a plateau and then rose to 3.0 V. These abnormal features during the first charge were also observed by others,[18, 22, 23] and thus a higher cut-off voltage (3.2 V) was applied in order to activate Li2S. This barrier did not appear in the subsequent cycles. The initial discharge capacity of the cell was in the range of 400 mAh g1 which dropped to about 170 mAh g1 after 20 cycles. In general, the cell showed a low coulombic efficiency (Figure 1 b). The redox shuttle reaction started from the second charge and reached its maximum in the 3rd charge in which 1260 mAh g1 could be achieved which resulted in a very low coulombic efficiency of 23 % (Figure 1 b). Typical voltage profile and cycling of the Li2S cell using an electrolyte of 0.2 m Li2S9-0.5 m LiNO3 in DME are shown in Figure 1 c and 1 d. Unlike the cell filled with the LiTFSI-based electrolyte there was no initial barrier in the charge process. Based on the Li2S content in the electrode, the initial charge and discharge capacities were 400 and 1300 mAh g1, respectively. The capacities remained at the level of 1100 mAh g1 during the subsequent cycles and slightly dropped below 1000 mAh g1 after 65 cycles. The coulombic efficiency of the cell was 96 % except for the first couple of cycles. As the used electrolyte contained polysulfides which are catholyte species that can contribute to the overall cell capacity, we recalculated ChemSusChem 2014, 7, 2457 – 2460
2457
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
Figure 1. Charge–discharge voltage profiles for Li2S cathode using (a) 1 m LiTFSI dissolved in DOL and DME solvents (1:1 v/v) as the electrolyte, and (c) 0.2 m Li2S9 and 0.5 m LiNO3 dissolved in DME as the electrolyte, and cycling performance for Li2S cathode using (b) the LiTFSI-based electrolyte, and (d) the polysulfide-based electrolyte.
the specific capacity based on the total mass of Li2S in the electrode in addition to the mass of lithium polysulfides in the electrolyte, as shown in Figure 1 d. With this taken into account, the cell’s capacity was still in the range of 550 mAh g1. These results clearly demonstrate the excellent capacity, cycling stability, and high coulombic efficiency for Li2S electrodes using a polysulfide electrolyte. Another advantage of using the polysulfide electrolyte with the Li2S electrode is that the electrode itself does not require any special treatment, such as ball milling,[22] carbon coating,[20–23] or thinning.[24] The potential barrier in the initial charge of Li2S when using the LiTFSI electrolyte is believed to be linked to an activation process during which the oxidation of Li2S to Li2Sx needs extra driving energy to facilitate the nucleation of a new phase.[18] In our case, the use of the Li2Sxbased electrolyte likely eliminates this initial energy barrier because of the synergetic presence of the reduced and oxidized forms of Li2S. Li2S can spontaneously oxidize to low-order Li2Sx when in contact with the high-order Li2Sx in the polysulfide electrolyte. This phenomenon is very much similar to what happens during the synthesis of the polysulfide electrolyte in which the oxidation of Li2S (dissolution) only takes place in the presence of sulfur and/or high-order Li2Sx in the DME solvent. This chemical reaction may produce a layer of lithium polysulfides at the Li2S/electrolyte interface even before the initiation of the electrochemical charge; thus no further energy is needed in the nucleation of polysulfides and/or sulfur during the initial charge. Another main role of the polysulfide species 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
is that they assist in establishing equilibrium between their dissolution and precipitation at the Li2S/electrolyte interface, especially in the presence of LiNO3 that played the role of the shuttle-reaction suppressor. The Li2Sx–LiNO3 couple in DME led to the maximization of sulfur utilization, which was translated by the good capacity retention upon cycling of the cell. Furthermore, the Li2Sx–LiNO3 electrolyte also improved the overall capacity of the Li2S cell while preserving its main role as a medium for Li + -ion conduction in the electrolyte. To quantify the capacity contribution of the polysulfide species in the electrolyte, a cell was assembled with an electrode comprising carbon as the positive electrode, Li metal as the negative electrode and the polysulfide-based solution as the electrolyte. The specific capacity of this cell was calculated based on the mass of
Figure 2. Performance of a Li–S cell comprised of carbon electrode, Li-metal anode, and polysulfide electrolyte: (a) charge–discharge voltage profiles and (b) capacity and coulombic efficiency vs. cycle number at C/10.
ChemSusChem 2014, 7, 2457 – 2460
2458
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
the polysulfides contained in the electrolyte (1.2 mg). As illustrated in Figure 2 a, the initial charge curve showed a small capacity (10 mAh g1) when the cell was charged to 2.6 V because most of the lithium polysulfide species in the electrolyte are in the form of Li2S9. The subsequent charge/discharge curves exhibited two plateaus in agreement with a previous report.[27] The higher discharge voltage plateau (~ 2.3–2.4 V) is associated to the generation of lithium polysulfides with longer chains while the lower discharge voltage plateau (~2.0 V) is related to the further reduction of these polysulfides to Li2S2 and/or Li2S. Figure 2 b shows that the capacity of the carbon cell progressively increased from the second cycle and kept steady at around 150 mAh g1 for the remainder of cycling. This value Figure 3. Synchrotron X-ray diffraction patterns of the Li2S electrodes: (a) original electrode; (b) point A1 in Figure 1 (a) using the LITFSI-based elecwas much lower than the capacity of the Li2S cell (Figure 1 c trolyte; (c) point A2 in Figure 1 (c) using the polysulfide-based electrolyte; and d). As discussed earlier, even based on the total active ma(d) end of charge, using the LiTFSI-based electrolyte; (e) end of charge, terial amount contained in both the cathode and electrolyte, using the polysulfide based electrolyte; (f) Li2S powders; (g) sulfur powders. the capacity of the Li2S cell was 500–600 mAh g1. This result confirms that despite its electrochemical activity the polysulthe active material loss was pronounced in the case of the fide electrolyte does not significantly contribute to the capaciLiTFSI-based electrolyte. ty of the cell. We rather believe that the chemical activation of In summary, the electrochemical performance of a Li2S batLi2S in the presence of the polysulfides is the main reason for the high capacity and facile activation of the Li2S cells. tery has been significantly improved in the presence of a polysulfide formulated electrolyte. We first showed that the use of Ex situ synchrotron X-ray diffraction was used to study the the LiTFSI-based electrolyte in conjunction with Li2S has resultphase evolution of Li2S cathode during the initial charge. Diffraction patterns at different states of charge of Li2S using the ed in low capacity and coulombic efficiency and poor cycle life. However, the use of the polysulfide electrolyte eliminated two electrolytes are plotted in Figure 3. Regardless of the electhe initial energy barrier for the oxidation of Li2S and demontrolyte, we observed that the main diffraction peaks of Li2S decreased in intensity during the charge process (points A1 and A2 in Figure 1 a and c), which is possibly due to the creation of the polysulfide interface. At the end of the charge process (4.0 V), we only observed broad peaks which correspond to illcrystalline sulfur. Figure 4 illustrates the morphology of the pristine Li2S electrode and the electrodes recovered from cells charged using the two electrolytes. Compared to the pristine Li2S electrode (Figure 4 a), the charged electrode in the presence of the LiTFSI-based electrolyte (Figure 4 b) lost some active material because of the dissolution of Li2S that created more porosity in the electrode. In contrast, the porosity of the electrode using the polysulfide electrolyte (Figure 4 c) appeared to be unchanged. To confirm this observation, the EDS analysis conductFigure 4. SEM images showing morphology of (a) a fresh Li2S electrode; (b) Li2S electrode after first charge using ed on the pristine, charged and the LiTFSI-based electrolyte; (c) Li2S electrode after first charge using the polysulfide-based electrolyte and (d) EDS cycled electrodes confirms that quantitative analysis on sulfur element for the Li2S electrodes. 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 2457 – 2460
2459
CHEMSUSCHEM COMMUNICATIONS strated a superior high capacity (1100 mAh g1) and almost 100 % coulombic efficiency with good cycle life. Although we use lithium metal to demonstrate the cell performance of Li2S, the results are geared toward paving the way for the application of Li2S cathode combined with a lithium-free anode in practical sulfur batteries.
Experimental Section Li2S, sulfur, and lithium nitrate (LiNO3), DME, DOL, and lithium bis(trifluoromethanesulfonyl)-imide (LiTFSI) were purchased from Sigma Aldrich. The polysulfide electrolyte was prepared by dissolving stoichiometric amounts of Li2S and S (1:8) in DME at 60 8C. The concentration of Li2S9 in DME was 0.2 m. Subsequently, LiNO3 salt was dissolved into the above solution at the concentration level of 0.5 m. The resulting polysulfide-based electrolyte solution was dark red. The electrochemical tests were performed on CR2032-type coin cells (1.6 cm2). The positive electrode consisted of Li2S (40 wt %), carbon (50 wt %), and pol(vinylidene fluoride) binder (10 wt %). Cell preparation and assembly were done inside a glove box filled with ultrahigh-purity argon. The loading of Li2S in the electrode was around 0.8 mg cm2. Two electrolytes were prepared: (1) 1 m LiTFSI in DOL and DME (1:1 v/v) solvents and (2) 0.2 m Li2S9 and 0.5 m LiNO3 in DME solvent. The cells were assembled with lithium metal as the anode. The amount of electrolyte used in each cell was 0.02 mL which contains about 1.2 mg of Li2S9 dissolved species. In the presence of the LiTFSI-based electrolyte, the voltage range was 1.5–3.2 V for the first cycle and then 1.5–2.7 V for the rest of cycles. In the presence of the polysulfide-based electrolyte the cells were tested between 1.5 and 2.7 V. The current density was 120 mA g1, corresponding to the C/10 rate based on Li2S. High-energy synchrotron X-ray powder diffraction (115 keV, l = 0.108 ) was conducted at beamline 11-ID-C at the Advanced Photon Source at Argonne National Laboratory. Fit2D software was applied to convert the 2D diffraction patterns to 1D diffraction patterns. The obtained XRD patterns were converted to equivalent patterns with the CuKa wavelength (l = 1.54 ). The morphology of the electrodes and energy dispersive spectroscopy (EDS) quantitative information was investigated by SEM using a Hitachi S-4700II microscope in the Electron Microscopy Center at Argonne National Laboratory.
Acknowledgements This research was funded by the U.S. Department of Energy, Freedom CAR, and Vehicle Technologies Office. The electron microscopy was accomplished at the Electron Microscopy Center for Materials Research at Argonne National Laboratory, a U.S. Department of Energy Office of Science Laboratory operated under Contract No. DE-AC02-06CH11357 by UChicago Argonne, LLC.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org Keywords: electrolyte · energy storage · lithium disulfide · lithium polysulfide · lithium sulfur batteries
[1] J. Hassoun, B. Scrosati, Adv. Mater. 2010, 22, 5198 – 5201. [2] J. R. M. Akridge, Y. V. Mikhaylik, N. White, Solid State Ionics 2004, 175, 243 – 245. [3] X. Ji, L. F. Nazar, J. Mater. Chem. 2010, 20, 9821 – 9826. [4] Y. V. Mikhaylik, J. R. Akridge, J. Electrochem. Soc. 2003, 150, A306 – A311. [5] R. Xu, I. Belharouak, J. C. M. Li, X. Zhang, I. Bloom, J. BareÇo, Adv. Energy Mater. 2013, 3, 833 – 838. [6] M. S. Whittingham, Chem. Rev. 2004, 104, 4271 – 4301. [7] Y.-K. Sun, S. T. Myung, B. C. Park, J. Prakash, I. Belharouak, K. Amine, Nat. Mater. 2009, 8, 320 – 324. [8] P. G. Bruce, S. A. Freunberger, L. J. Hardwick, J. M. Tarascon, Nat. Mater. 2012, 11, 19 – 29. [9] Y. V. Mikhaylik, J. R. Akridge, J. Electrochem. Soc. 2004, 151, A1969 – A1976. [10] Y. Diao, K. Xie, S. Xiong, X. Hong, J. Electrochem. Soc. 2012, 159, A421 – A425. [11] C. Barchasz, F. Molton, C. Duboc, J. C. Lepretre, S. Patoux, F. Alloin, Anal. Chem. 2012, 84, 3973 – 3980. [12] R. Dominko, R. Demir-Cakan, M. Morcrette, J.-M. Tarascon, Electrochem. Commun. 2011, 13, 117 – 120. [13] V. S. Kolosnitsyn, E. V. Kuzmina, E. V. Karaseva, S. E. Mochalov, J. Power Sources 2011, 196, 1478 – 1482. [14] J. Nelson, S. Misra, Y. Yang, A. Jackson, Y. Liu, H. Wang, H. Dai, J. C. Andrews, Y. Cui, M. F. Toney, J. Am. Chem. Soc. 2012, 134, 6337 – 6343. [15] Y. Yang, M. T. McDowell, A. Jackson, J. J. Cha, S. S. Hong, Y. Cui, Nano Lett. 2010, 10, 1486 – 1491. [16] J. Hassoun, B. Scrosati, Angew. Chem. Int. Ed. 2010, 49, 2371 – 2374; Angew. Chem. 2010, 122, 2421 – 2424. [17] M. N. Obrovac, J. R. Dahn, Electrochem. Solid-State Lett. 2002, 5, A70 – A73. [18] Y. Yang, G. Zheng, S. Misra, J. Nelson, M. F. Toney, Y. Cui, J. Am. Chem. Soc. 2012, 134, 15387 – 15394. [19] J. Hassoun, Y. K. Sun, B. Scrosati, J. Power Sources 2011, 196, 343 – 348. [20] S. Jeong, D. Bresser, D. Buchholz, M. Winter, S. Passerini, J. Power Sources 2013, 235, 220 – 225. [21] T. Takeuchi, H. Kageyama, K. Nakanishi, M. Tabuchi, H. Sakaebe, T. Ohta, H. Senoh, T. Sakai, K. Tatsumi, J. Electrochem. Soc. 2010, 157, A1196 – A1201. [22] J. Guo, Z. Yang, Y. Yu, H. D. AbruÇa, L. A. Archer, J. Am. Chem. Soc. 2013, 135, 763 – 767. [23] K. Cai, M. K. Song, E. J. Cairns, Y. Zhang, Nano Lett. 2012, 12, 6474 – 6479. [24] Y. Zhou, C. Wu, H. Zhang, X. Wu, Z. Fu, Electrochim. Acta 2007, 52, 3130 – 3136. [25] A. Hayashi, R. Ohtsubo, T. Ohtomo, F. Mizuno, M. Tatsumisago, J. Power Sources 2008, 183, 422 – 426. [26] M. Nagao, A. Hayashi, M. Tatsumisago, J. Mater. Chem. 2012, 22, 10015 – 10020. [27] K. Kumaresan, Y. Mikhaylik, R. E. White, J. Electrochem. Soc. 2008, 155, A576 – A582. Received: March 13, 2014 Published online on July 8, 2014
ChemSusChem 2014, 7, 2457 – 2460
2460
Paving the Way for Using Li2S Batteries Rui Xu,[a, c] Xiaofeng Zhang,[a] Cun Yu,[d] Yang Ren,[d] James C. M. Li,[c] and Ilias Belharouak*[a, b] In this work, a novel lithium–sulfur battery was developed comprising Li2S as the cathode, lithium metal as the anode and polysulfide-based solution as the electrolyte. The electrochemical performances of these Li2S-based cells strongly depended upon the nature of the electrolytes. In the presence of the conventional electrolyte that consisted of lithium bis(trifluoromethanesulfonyl)-imide (LiTFSI) salt dissolved in a solvent combination of dimethoxyethane (DME)/1,3-dioxolane (DOL), the Li/Li2S cells showed sluggish kinetics, which translated into poor cycling and capacity retention. However, when using small amounts of polysulfides in the electrolyte along with a shuttle inhibitor the Li2S cathode was efficiently activated in the cell with the generation of over 1000 mAh g1 capacity and good cycle life.
Li-S batteries are attractive for high-energy electrochemical storage devices, because sulfur is abundant, nontoxic and lowcost.[1–5] A rechargeable Li–S battery is based on the reversible redox process between sulfur and lithium via the electrochemical reaction: S8 + 16 Li + + 16 e$8 Li2S. Conventional Li-S batteries consist of elemental sulfur as the cathode, a non-aqueous liquid as the electrolyte, and lithium metal as the anode. The sulfur cathode offers superior theoretical capacity (1672 mAh g1) compared to all Li-ion battery cathode chemistries (300 mAh g1).[6–8] Despite these advantages, capacity decline and low coulombic efficiency have presented significant challenges to the development of practical Li–S batteries. The issue is related to the dissolution of lithium polysulfides (Li2Sx) (x = 4 to 8), from the sulfur electrode into the electrolyte, which leads to a redox shuttle reaction in the cell.[9–14] In particular, all up-to-date Li-S battery configurations require the use
[a] Dr. R. Xu, Dr. X. Zhang, Dr. I. Belharouak Chemical Sciences and Engineering Division Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA) E-mail: [email protected] [b] Dr. I. Belharouak Qatar Environment and Energy Research Institute Qatar Foundation P.O. Box 5825, Doha (Qatar) [c] Dr. R. Xu, Prof. J. C. M. Li Materials Science Program Department of Mechanical Engineering University of Rochester Rochester, NY 14627 (USA) [d] C. Yu, Dr. Y. Ren Advanced Photon Source Argonne National Laboratory 9700 South Cass Avenue, Argonne, IL 60439 (USA)
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
of metallic lithium and this poses significant risk to the development of practical batteries. Recently, lithium disulfide (Li2S) cathode (theoretical capacity 1166 mAh g1) has gained great attention due to the potential of using forms of anodes other than lithium metal, such as silicon and metal alloys.[15, 16] However, Li2S does not provide the full extent of its capacity in the cell because it is barely active due to its highly insulating character.[17, 18] Therefore, researchers have devoted efforts to improve the conductivity of the Li2S electrode and mitigate sulfur dissolution. As an example, Li2S powder has been ball-milled or mixed with different carbon forms or metals.[9, 10, 19–24] Others have suggested the use of all-solid-state cells.[25, 26] As a result, improvements have been made in terms of initial capacities; however, issues such as poor cycle life and cell efficiency remain unsolved. Herein, we describe the use of a non-conventional electrolyte in a Li2S battery. Inspired by our recent work on sulfur batteries, we developed a Li/Li2S cell comprising a catholytebased electrolyte that is made of polysulfides. This electrolyte significantly improved the electrochemical activity of Li2S by delivering high capacity and reasonable cycling. Figure 1 a and b show typical voltage profile and cycling of the Li2S cell using an electrolyte consisting of 1 m LiTFSI (lithium bis(trifluoromethanesulfonyl)-imide) in dimethoxyethane/ 1,3-dioxolane (DME/DOL). In the initial charge (Figure 1 a), the voltage quickly reached 3.0 V and then declined to 2.6 V followed by a plateau and then rose to 3.0 V. These abnormal features during the first charge were also observed by others,[18, 22, 23] and thus a higher cut-off voltage (3.2 V) was applied in order to activate Li2S. This barrier did not appear in the subsequent cycles. The initial discharge capacity of the cell was in the range of 400 mAh g1 which dropped to about 170 mAh g1 after 20 cycles. In general, the cell showed a low coulombic efficiency (Figure 1 b). The redox shuttle reaction started from the second charge and reached its maximum in the 3rd charge in which 1260 mAh g1 could be achieved which resulted in a very low coulombic efficiency of 23 % (Figure 1 b). Typical voltage profile and cycling of the Li2S cell using an electrolyte of 0.2 m Li2S9-0.5 m LiNO3 in DME are shown in Figure 1 c and 1 d. Unlike the cell filled with the LiTFSI-based electrolyte there was no initial barrier in the charge process. Based on the Li2S content in the electrode, the initial charge and discharge capacities were 400 and 1300 mAh g1, respectively. The capacities remained at the level of 1100 mAh g1 during the subsequent cycles and slightly dropped below 1000 mAh g1 after 65 cycles. The coulombic efficiency of the cell was 96 % except for the first couple of cycles. As the used electrolyte contained polysulfides which are catholyte species that can contribute to the overall cell capacity, we recalculated ChemSusChem 2014, 7, 2457 – 2460
2457
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
Figure 1. Charge–discharge voltage profiles for Li2S cathode using (a) 1 m LiTFSI dissolved in DOL and DME solvents (1:1 v/v) as the electrolyte, and (c) 0.2 m Li2S9 and 0.5 m LiNO3 dissolved in DME as the electrolyte, and cycling performance for Li2S cathode using (b) the LiTFSI-based electrolyte, and (d) the polysulfide-based electrolyte.
the specific capacity based on the total mass of Li2S in the electrode in addition to the mass of lithium polysulfides in the electrolyte, as shown in Figure 1 d. With this taken into account, the cell’s capacity was still in the range of 550 mAh g1. These results clearly demonstrate the excellent capacity, cycling stability, and high coulombic efficiency for Li2S electrodes using a polysulfide electrolyte. Another advantage of using the polysulfide electrolyte with the Li2S electrode is that the electrode itself does not require any special treatment, such as ball milling,[22] carbon coating,[20–23] or thinning.[24] The potential barrier in the initial charge of Li2S when using the LiTFSI electrolyte is believed to be linked to an activation process during which the oxidation of Li2S to Li2Sx needs extra driving energy to facilitate the nucleation of a new phase.[18] In our case, the use of the Li2Sxbased electrolyte likely eliminates this initial energy barrier because of the synergetic presence of the reduced and oxidized forms of Li2S. Li2S can spontaneously oxidize to low-order Li2Sx when in contact with the high-order Li2Sx in the polysulfide electrolyte. This phenomenon is very much similar to what happens during the synthesis of the polysulfide electrolyte in which the oxidation of Li2S (dissolution) only takes place in the presence of sulfur and/or high-order Li2Sx in the DME solvent. This chemical reaction may produce a layer of lithium polysulfides at the Li2S/electrolyte interface even before the initiation of the electrochemical charge; thus no further energy is needed in the nucleation of polysulfides and/or sulfur during the initial charge. Another main role of the polysulfide species 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
is that they assist in establishing equilibrium between their dissolution and precipitation at the Li2S/electrolyte interface, especially in the presence of LiNO3 that played the role of the shuttle-reaction suppressor. The Li2Sx–LiNO3 couple in DME led to the maximization of sulfur utilization, which was translated by the good capacity retention upon cycling of the cell. Furthermore, the Li2Sx–LiNO3 electrolyte also improved the overall capacity of the Li2S cell while preserving its main role as a medium for Li + -ion conduction in the electrolyte. To quantify the capacity contribution of the polysulfide species in the electrolyte, a cell was assembled with an electrode comprising carbon as the positive electrode, Li metal as the negative electrode and the polysulfide-based solution as the electrolyte. The specific capacity of this cell was calculated based on the mass of
Figure 2. Performance of a Li–S cell comprised of carbon electrode, Li-metal anode, and polysulfide electrolyte: (a) charge–discharge voltage profiles and (b) capacity and coulombic efficiency vs. cycle number at C/10.
ChemSusChem 2014, 7, 2457 – 2460
2458
CHEMSUSCHEM COMMUNICATIONS
www.chemsuschem.org
the polysulfides contained in the electrolyte (1.2 mg). As illustrated in Figure 2 a, the initial charge curve showed a small capacity (10 mAh g1) when the cell was charged to 2.6 V because most of the lithium polysulfide species in the electrolyte are in the form of Li2S9. The subsequent charge/discharge curves exhibited two plateaus in agreement with a previous report.[27] The higher discharge voltage plateau (~ 2.3–2.4 V) is associated to the generation of lithium polysulfides with longer chains while the lower discharge voltage plateau (~2.0 V) is related to the further reduction of these polysulfides to Li2S2 and/or Li2S. Figure 2 b shows that the capacity of the carbon cell progressively increased from the second cycle and kept steady at around 150 mAh g1 for the remainder of cycling. This value Figure 3. Synchrotron X-ray diffraction patterns of the Li2S electrodes: (a) original electrode; (b) point A1 in Figure 1 (a) using the LITFSI-based elecwas much lower than the capacity of the Li2S cell (Figure 1 c trolyte; (c) point A2 in Figure 1 (c) using the polysulfide-based electrolyte; and d). As discussed earlier, even based on the total active ma(d) end of charge, using the LiTFSI-based electrolyte; (e) end of charge, terial amount contained in both the cathode and electrolyte, using the polysulfide based electrolyte; (f) Li2S powders; (g) sulfur powders. the capacity of the Li2S cell was 500–600 mAh g1. This result confirms that despite its electrochemical activity the polysulthe active material loss was pronounced in the case of the fide electrolyte does not significantly contribute to the capaciLiTFSI-based electrolyte. ty of the cell. We rather believe that the chemical activation of In summary, the electrochemical performance of a Li2S batLi2S in the presence of the polysulfides is the main reason for the high capacity and facile activation of the Li2S cells. tery has been significantly improved in the presence of a polysulfide formulated electrolyte. We first showed that the use of Ex situ synchrotron X-ray diffraction was used to study the the LiTFSI-based electrolyte in conjunction with Li2S has resultphase evolution of Li2S cathode during the initial charge. Diffraction patterns at different states of charge of Li2S using the ed in low capacity and coulombic efficiency and poor cycle life. However, the use of the polysulfide electrolyte eliminated two electrolytes are plotted in Figure 3. Regardless of the electhe initial energy barrier for the oxidation of Li2S and demontrolyte, we observed that the main diffraction peaks of Li2S decreased in intensity during the charge process (points A1 and A2 in Figure 1 a and c), which is possibly due to the creation of the polysulfide interface. At the end of the charge process (4.0 V), we only observed broad peaks which correspond to illcrystalline sulfur. Figure 4 illustrates the morphology of the pristine Li2S electrode and the electrodes recovered from cells charged using the two electrolytes. Compared to the pristine Li2S electrode (Figure 4 a), the charged electrode in the presence of the LiTFSI-based electrolyte (Figure 4 b) lost some active material because of the dissolution of Li2S that created more porosity in the electrode. In contrast, the porosity of the electrode using the polysulfide electrolyte (Figure 4 c) appeared to be unchanged. To confirm this observation, the EDS analysis conductFigure 4. SEM images showing morphology of (a) a fresh Li2S electrode; (b) Li2S electrode after first charge using ed on the pristine, charged and the LiTFSI-based electrolyte; (c) Li2S electrode after first charge using the polysulfide-based electrolyte and (d) EDS cycled electrodes confirms that quantitative analysis on sulfur element for the Li2S electrodes. 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 2457 – 2460
2459
CHEMSUSCHEM COMMUNICATIONS strated a superior high capacity (1100 mAh g1) and almost 100 % coulombic efficiency with good cycle life. Although we use lithium metal to demonstrate the cell performance of Li2S, the results are geared toward paving the way for the application of Li2S cathode combined with a lithium-free anode in practical sulfur batteries.
Experimental Section Li2S, sulfur, and lithium nitrate (LiNO3), DME, DOL, and lithium bis(trifluoromethanesulfonyl)-imide (LiTFSI) were purchased from Sigma Aldrich. The polysulfide electrolyte was prepared by dissolving stoichiometric amounts of Li2S and S (1:8) in DME at 60 8C. The concentration of Li2S9 in DME was 0.2 m. Subsequently, LiNO3 salt was dissolved into the above solution at the concentration level of 0.5 m. The resulting polysulfide-based electrolyte solution was dark red. The electrochemical tests were performed on CR2032-type coin cells (1.6 cm2). The positive electrode consisted of Li2S (40 wt %), carbon (50 wt %), and pol(vinylidene fluoride) binder (10 wt %). Cell preparation and assembly were done inside a glove box filled with ultrahigh-purity argon. The loading of Li2S in the electrode was around 0.8 mg cm2. Two electrolytes were prepared: (1) 1 m LiTFSI in DOL and DME (1:1 v/v) solvents and (2) 0.2 m Li2S9 and 0.5 m LiNO3 in DME solvent. The cells were assembled with lithium metal as the anode. The amount of electrolyte used in each cell was 0.02 mL which contains about 1.2 mg of Li2S9 dissolved species. In the presence of the LiTFSI-based electrolyte, the voltage range was 1.5–3.2 V for the first cycle and then 1.5–2.7 V for the rest of cycles. In the presence of the polysulfide-based electrolyte the cells were tested between 1.5 and 2.7 V. The current density was 120 mA g1, corresponding to the C/10 rate based on Li2S. High-energy synchrotron X-ray powder diffraction (115 keV, l = 0.108 ) was conducted at beamline 11-ID-C at the Advanced Photon Source at Argonne National Laboratory. Fit2D software was applied to convert the 2D diffraction patterns to 1D diffraction patterns. The obtained XRD patterns were converted to equivalent patterns with the CuKa wavelength (l = 1.54 ). The morphology of the electrodes and energy dispersive spectroscopy (EDS) quantitative information was investigated by SEM using a Hitachi S-4700II microscope in the Electron Microscopy Center at Argonne National Laboratory.
Acknowledgements This research was funded by the U.S. Department of Energy, Freedom CAR, and Vehicle Technologies Office. The electron microscopy was accomplished at the Electron Microscopy Center for Materials Research at Argonne National Laboratory, a U.S. Department of Energy Office of Science Laboratory operated under Contract No. DE-AC02-06CH11357 by UChicago Argonne, LLC.
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org Keywords: electrolyte · energy storage · lithium disulfide · lithium polysulfide · lithium sulfur batteries
[1] J. Hassoun, B. Scrosati, Adv. Mater. 2010, 22, 5198 – 5201. [2] J. R. M. Akridge, Y. V. Mikhaylik, N. White, Solid State Ionics 2004, 175, 243 – 245. [3] X. Ji, L. F. Nazar, J. Mater. Chem. 2010, 20, 9821 – 9826. [4] Y. V. Mikhaylik, J. R. Akridge, J. Electrochem. Soc. 2003, 150, A306 – A311. [5] R. Xu, I. Belharouak, J. C. M. Li, X. Zhang, I. Bloom, J. BareÇo, Adv. Energy Mater. 2013, 3, 833 – 838. [6] M. S. Whittingham, Chem. Rev. 2004, 104, 4271 – 4301. [7] Y.-K. Sun, S. T. Myung, B. C. Park, J. Prakash, I. Belharouak, K. Amine, Nat. Mater. 2009, 8, 320 – 324. [8] P. G. Bruce, S. A. Freunberger, L. J. Hardwick, J. M. Tarascon, Nat. Mater. 2012, 11, 19 – 29. [9] Y. V. Mikhaylik, J. R. Akridge, J. Electrochem. Soc. 2004, 151, A1969 – A1976. [10] Y. Diao, K. Xie, S. Xiong, X. Hong, J. Electrochem. Soc. 2012, 159, A421 – A425. [11] C. Barchasz, F. Molton, C. Duboc, J. C. Lepretre, S. Patoux, F. Alloin, Anal. Chem. 2012, 84, 3973 – 3980. [12] R. Dominko, R. Demir-Cakan, M. Morcrette, J.-M. Tarascon, Electrochem. Commun. 2011, 13, 117 – 120. [13] V. S. Kolosnitsyn, E. V. Kuzmina, E. V. Karaseva, S. E. Mochalov, J. Power Sources 2011, 196, 1478 – 1482. [14] J. Nelson, S. Misra, Y. Yang, A. Jackson, Y. Liu, H. Wang, H. Dai, J. C. Andrews, Y. Cui, M. F. Toney, J. Am. Chem. Soc. 2012, 134, 6337 – 6343. [15] Y. Yang, M. T. McDowell, A. Jackson, J. J. Cha, S. S. Hong, Y. Cui, Nano Lett. 2010, 10, 1486 – 1491. [16] J. Hassoun, B. Scrosati, Angew. Chem. Int. Ed. 2010, 49, 2371 – 2374; Angew. Chem. 2010, 122, 2421 – 2424. [17] M. N. Obrovac, J. R. Dahn, Electrochem. Solid-State Lett. 2002, 5, A70 – A73. [18] Y. Yang, G. Zheng, S. Misra, J. Nelson, M. F. Toney, Y. Cui, J. Am. Chem. Soc. 2012, 134, 15387 – 15394. [19] J. Hassoun, Y. K. Sun, B. Scrosati, J. Power Sources 2011, 196, 343 – 348. [20] S. Jeong, D. Bresser, D. Buchholz, M. Winter, S. Passerini, J. Power Sources 2013, 235, 220 – 225. [21] T. Takeuchi, H. Kageyama, K. Nakanishi, M. Tabuchi, H. Sakaebe, T. Ohta, H. Senoh, T. Sakai, K. Tatsumi, J. Electrochem. Soc. 2010, 157, A1196 – A1201. [22] J. Guo, Z. Yang, Y. Yu, H. D. AbruÇa, L. A. Archer, J. Am. Chem. Soc. 2013, 135, 763 – 767. [23] K. Cai, M. K. Song, E. J. Cairns, Y. Zhang, Nano Lett. 2012, 12, 6474 – 6479. [24] Y. Zhou, C. Wu, H. Zhang, X. Wu, Z. Fu, Electrochim. Acta 2007, 52, 3130 – 3136. [25] A. Hayashi, R. Ohtsubo, T. Ohtomo, F. Mizuno, M. Tatsumisago, J. Power Sources 2008, 183, 422 – 426. [26] M. Nagao, A. Hayashi, M. Tatsumisago, J. Mater. Chem. 2012, 22, 10015 – 10020. [27] K. Kumaresan, Y. Mikhaylik, R. E. White, J. Electrochem. Soc. 2008, 155, A576 – A582. Received: March 13, 2014 Published online on July 8, 2014
ChemSusChem 2014, 7, 2457 – 2460
2460
