Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries

A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries ARTICLE Received 28 Aug 2012 | Accepted 10 Jan 2013 | Published 12 Feb 2013 DOI: 10.1038/ncomms2513 A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries 1 1 1 1 1 Liumin Suo , Yong-Sheng Hu , Hong Li , Michel Armand & Liquan Chen Liquid electrolyte plays a key role in commercial lithium-ion batteries to allow conduction of lithium-ion between cathode and anode. Traditionally, taking into account the ionic con- ductivity, viscosity and dissolubility of lithium salt, the salt concentration in liquid electrolytes is typically less than 1.2 mol l . Here we show a new class of ‘Solvent-in-Salt’ electrolyte with ultrahigh salt concentration and high lithium-ion transference number (0.73), in which salt holds a dominant position in the lithium-ion transport system. It remarkably enhances cyclic and safety performance of next-generation high-energy rechargeable lithium batteries via an effective suppression of lithium dendrite growth and shape change in the metallic lithium anode. Moreover, when used in lithium–sulphur battery, the advantage of this electrolyte is further demonstrated that lithium polysulphide dissolution is inhibited, thus overcoming one of today’s most challenging technological hurdles, the ‘polysulphide shuttle phenomenon’. Consequently, a coulombic efficiency nearing 100% and long cycling stability are achieved. Key Laboratory for Renewable Energy, Beijing Key Laboratory for New Energy Materials and Devices, Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China. Correspondence and requests for materials should be addressed to Y.-S.H. (email: [email protected]). NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 1 & 2013 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ithium-based batteries as highly efficient energy storage The physicochemical properties of ‘Solvent-in-Salt’. For the devices have long been considered as promising power following discussion, we select an electrolyte system containing Lsupply for various electric vehicles and smart grid storage Li[CF SO ) N] (LiTFSI), one of the lowest lattice energy salts and 3 2 2 1–4 systems . However, presently available lithium-ion technology 1,3-dioxolane (DOL): dimethoxyethane (DME) (1:1 by volume) cannot satisfy the increasing demand for energy density. Metallic as solvent, resulting in what we show as one of the most pro- lithium batteries exhibit the highest theoretical energy densities mising electrolytes for Li–S batteries. The physicochemical among the secondary batteries . Among many possible systems, properties of this electrolyte with different ratios of salt-to-solvent lithium–sulphur (Li–S) and lithium–oxygen (Li–O ) batteries are are illustrated in Fig. 2. When the mole amount of salt reaches quite attractive as candidates for next-generation high-energy 4 mol in 1-l solvent, the electrolyte enters the D region of SIS by 5–22 density batteries . However, the application of metallic lithium weight, and beyond 5 mol salt in 1-l solvent, the salt begins to anode suffers from inevitable formation of lithium dendrites, which have a dominant role in either weight or volume ratio (Fig. 2a). are caused by uneven current distributions at the metal–electrolyte Arrhenius plots of the ionic conductivity of the electrolytes with interface during cycling. The formation of lithium dendrites will different salt concentrations in a temperature range of  20 to lead to poor cyclic performance and increase the probability of 60 C are shown in Fig. 2b and exhibit the typical curvature of the internal short circuit, resulting in safety issues. Thus, great effort Vogel–Tammann–Fulcher (VTF) equation (Supplementary has been paid to suppress the formation of lithium dendrites and Fig. S1). It can be seen that the ionic conductivity decreases reduce metallic lithium corrosion in liquid electrolytes .For with increasing salt concentration. The conductivity drops instance, the use of solid polymer or inorganic electrolytes could slowly in the high temperature region (20–60 C) but rapidly potentially suppress the formation of the lithium dendrites. at a low-temperature region (  20 to 20 C), resulting from a rise However, for solid-state electrolytes, the kinetic properties are in T . For a given electrolyte with fixed salt and solvent, ionic limited, due to both low conductivity at room temperature and conductivity depends on both viscosity and lithium-ion 24–28 high interfacial resistance . Recently, a breakthrough has been mobility. When increasing salt concentration, more and more achieved in inorganic sulphide-based electrolytes with quite high Li–ether complex pairs form due to incomplete solvation 2  1 room temperature conductivity of 10 Scm (ref. 29). The shell and the viscosity at room temperature increases markedly in application of Li S–P S glass–ceramic electrolyte in Li–S batteries the SIS region (Fig. 2c and Supplementary Fig. S2). At the same 2 2 5 30–32 has been demonstrated . To reduce the interfacial resistance, time, the lithium-ion transference number of SIS-7# electrolyte þ þ þ 30wt.% solid electrolyte and 35wt.% acetylene black were required increases to an unexpected high value (t ¼ 0.73, t ¼ s / Li Li Li to add into sulphur electrode . This leads to a decrease in energy (s þ s )) (see Fig. 2c and Supplementary Fig. S3), which is Li TFSI density of Li–S batteries. It seems that the strategy of physical much higher than that of traditional salt-in-solvent electrolytes mixing of active phase and solid electrolyte still suffers from point- (0.2–0.4). The conductivity of specific ion i is proportional to the 30–32 to-point contact , not like a complete surface-to-surface concentration of mobile ion (c ) and its mobility (m )(s ¼ nc m ). i i i i i wetting effect as liquid electrolyte. The mobility of an ion is determined by the viscosity (Z) Here we report a new class of non-aqueous liquid ‘Solvent-in- of the medium and radius of mobile ion (m ¼ 1/6pZr ) (ref. 23). i i Salt’ electrolytes and apply them in Li–S batteries. It is demon- In low-salt concentration electrolytes, lithium ions are strated that the use of ‘Solvent-in-Salt’ electrolyte inhibits the coordinated with ether oxygen and form a large solvation dissolution of lithium polysulphide, effectively protects metallic shell compared with anions, leading to relatively lower mobility lithium anodes against the formation of lithium dendrites and of solvated Li cations. In the SIS system, it is plausible that results in high lithium cycling efficiency, thus enhancing electro- the number of solvated Li cations is decreased and large chemical performance. anion (TFSI ) could be more seriously dragged than the small unsolvated cation (Li ) in this high viscosity system. Nevertheless, SIS-7# even with a high viscosity of 72 cP retains a Results conductivity of 0.814 mS cm at room temperature, which The concept of ‘Solvent-in-Salt’ electrolyte. In 1993, Angell remains superior to that of all-solid-state dry polymer or most of et al. proposed the innovative concept of ‘Polymer-in-Salt’ by the inorganic electrolytes, and especially, could form better reversing the ratio of solid polymer solvent to salt, in which glass interfacial contacts. transition (T ) was low enough to remain rubbery at room tempe- Differential scanning calorimeter (DSC) traces reveal distinct rature to preserve good conductivity and high electrochemical glass transition temperatures (Fig. 2d), which shows that all the stability. However, in practice the T remained above ambient electrolytes are glass-forming liquids, and their glass transition temperature and/or the system would crystallize. For temperatures shift from low to high, with increasing ratio of salt- conventional non-aqueous organic electrolytes, the salt to-solvent. For the pure solvent mixture without salt concentration is usually limited in a range of 1–2 mol l , (Supplementary Fig. S4), the glass transition temperature is which is a trade-off among ionic conductivity, viscosity and salt  138.6 C, justifying the excellent low temperature performance solubility. Thus, most of the studies focus on the region C of for the DOL–DME-based electrolyte. In the SIS-7# electrolyte, the Fig. 1, in which there is much less salt than solvent. There are few value of T is –77.3 C, which is much lower than typical reports of research in the A (yellow) or D (green) regions of ‘Polymer-in-Salt’ system (T 4–10 C) (ref. 33) and close to that Fig. 1, in which either the weight or volume ratio of salt-to- traditional commercial electrolyte systems (1 mol l LiPF in solvent exceeds 1.0 (refs 34–37). As a matter of fact, by choosing EC–DMC, T ¼ –67 C)) (ref. 40). The flexible hinge in the S–N– proper salt and solvent, we can move an electrolyte into those S bond of TFSI , specific to these imide anions, explains this regions (A, B and D) and also obtain some unexpected properties. ‘plasticizing effect’, which is also reflected in the low viscosities of To distinguish from traditional electrolytes, this new class of ionic liquids based on this anion. electrolyte is denoted by ‘Solvent-in-Salt’ (SIS). A similar system of hydrated molten salt composed of KNO and Ca(NO )  4H O, in which the water content is insufficient to Application in Li–S batteries. The power of the SIS electrolytes is 3 2 2 satisfy more than a first coordination sheath for the cation that demonstrated by their use in rechargeable metallic Li–S batteries. 38,39 was reported in 1965 by Angell . This kind of hydrated molten The Li–S battery using SIS-7# as electrolyte exhibits the best salts was usually used in heat storage but not called SIS. electrochemical performance (Fig. 3). 2 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. Lithium-ion transference Viscosity (cP) NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE Solid-state-based electrolytes Liquid-based electrolytes Salt content AB 1. Lithium ion conductor 100% Lithium molten salt 2. LiI/AI O 2 3 “Solvent-in-Salt” “Solvent-in-Salt” ‘Solvent-in-Salt’ by volume by volume and weight ([salt] > [solvent] fig.1b) “Polymer-in-Salt” ([salt] > [polymer]) 50% Plastic crystal Soggy sand electrolyte electrolyte CD Traditional non-aqueons Dry polymer Gel polymer “Solvent-in-Salt” electrolytes ([salt] < [polymer]) ([salt] < [solvent]) by weight 0 1 ∞ 100% 0% Weight ratio of salt-to-solvent Solid-phase content (%) Liquid phase 58–60 Figure 1 | A general concept for Solvent-in-Salt electrolyte. (a) The overview of the available electrolytes .(b) The distribution map of non-aqueous liquid electrolytes with the weight and volume ratios of salt-to-solvent. A, B and D regions are Solvent-in-Salt electrolyte, in which the ratio of salt-to- solvent is over 1.0 by volume or weight. C region is [solvent]4[salt] by weight and volume. 2.2 –1.6 C region D region B region 2.0 (fig. 1b) (fig. 1b) (fig. 1b) –2.0 1.8 Salt-in-Solvent 1.6 –2.4 1.4 –2.8 Volume ratio SIS-7# 1.2 1# Weight ratio –3.2 2# 1.0 3# 0.8 –3.6 4# 0.6 SIS-5# –4.0 Solvent-in-Salt 0.4 SIS-6# 25 °C –4.4 SIS-7# 0.2 0.0 123 4567 3.0 3.2 3.4 3.6 3.8 4.0 –1 Ratio of salt-to-solvent (mol per l solvent) 1,000/T (K ) 80 0.9 T = –77.3 °C SIS-7# 0.8 12 T = –83.6 °C SIS-6# 60 g 10 0.7 T = –91.9 °C SIS-5# 8 40 T = –98.9 °C 0.6 g 4# 6 30 T = –109 °C g 3# 0.5 4 20 2# 2 10 0.4 1# 0 0 0.3 123 4567 –150 –120 –90 –60 –30 0 30 Ratio of salt-to-solvent (mol per l solvent) Temperature (°C) Figure 2 | Physicochemical properties of Solvent-in-Salt electrolytes. (a) Weight and volume ratio of salt-to-solvent with different ratios of LiTFSI to DOL:DME (1:1 by volume). (b) Arrhenius plots of the ionic conductivity as a function of 1,000/T for electrolytes with different ratios of LiTFSI to solvent, (1#: 1 mol per l solvent, 2#: 2 mol per l solvent, 3#: 3 mol l solvent, 4#: 4 mol per l solvent, 5#: 5 mol per l solvent, SIS-6#: 6 mol l per solvent and SIS-7#: 7 mol l per solvent). (c) Viscosity, ionic conductivity and lithium-ion transference number at room temperature for the aforementioned different electrolytes. (d) DSC traces of the aforementioned different electrolytes. It shows an initial specific discharge capacity of still show excellent rate capability as shown in Fig. 3d. They can 1  1  1 1,041 mA h g at a current rate of 0.2C (that is, 335 mA g ) achieve capacities of 1,229, 988, 864, 744 and 551 mA h g of 1  1 and maintains a reversible capacity of 770 mA h g with sulphur at current rates of 0.2, 0.5, 1, 2 and 3C, respectively. capacity retention of 74% after 100 cycles (Fig. 3b). More When the current rate returns to 0.2C, a reversible capacity of 1  1 importantly, the coulombic efficiency reaches nearly 100% after 789 mA h g remains. The capacity still decays after rate the first cycle (it is 93.7% for the first cycle.) for the SIS-7# measurement, although the polysulphide dissolution is inhibited, electrolyte (also see Supplementary Fig. S5), which is higher than which is probably related to the unstable C/S electrode. In the 12–22 previous reports in a similar system , owing to effectively charge–discharge process, the electrode suffers from a large avoiding the polysulphide shuttle effect in the charging process volume change by a conversion reaction between S 3  3 (Fig. 3c) (ref. 41). In contrast, other less concentrated electrolytes (2.07 g cm ) and Li S (1.66 g cm ), which could result in exhibit an apparent coulombic efficiency greater than 100% not only the sulphur redistribution but also the structural damage (Fig. 3c and Supplementary Fig. S6), a sign of the ‘polysulphide of the carbon–sulphur composite. Thus, we can expect that shuttle effect’. The only shortcoming of the use of SIS-7# is that through optimization of cathode materials, the cyclic perfor- the polarization becomes slightly larger due to relatively higher mance will be further improved. viscosity compared with traditional electrolytes with low-salt To further prove the power of this new SIS electrolyte, the concentration (Fig. 3a). Yet Li–S batteries using SIS-7# electrolyte Ketjenblack without mesoporous structure was used as a support NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 3 & 2013 Macmillan Publishers Limited. All rights reserved. “Solvent-in-Salt’’ –1 Salt/solvent ratio σ (mS cm ) Exothermic Endothermic Volume ratio of salt-to-solvent –1 Log σ (S cm ) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 1,400 3.0 4# SIS-7# charge SIS-7# 1,200 2.7 SIS-7# discharge 2# 1,000 2.4 2.1 1.8 1.5 400 2# charge 1.2 4# 4# charge SIS-7# 2# discharge 4# discharge 2# 0.9 0 200 400 600 800 1,000 1,200 1,400 010 20 30 40 50 60 70 80 90 100 –1 Capacity (mA h g ) - sulphur Cycle number (N) 120 180 1,200 0.2C 140 0.5C 1,000 1C 110 2# 0.2C 4# 2C 3C 100 60 SIS-7# 40 SIS-7# charge SIS-7# discharge 0 102030405060708090 100 0 102030405060708090 100 Cycle number (N) Cycle number (N) Figure 3 | Electrochemical performance of lithium–sulphur batteries. (a) First discharge–charge profiles of C/S electrodes in electrolytes with different ratios of LiTFSI to DOL:DME (1:1 by volume). (b) Cyclic performance. (c) Coulombic efficiency at a current rate of 0.2 C (coulombic efficiency ¼ charge capacity/discharge capacity). (d) Rate capability with SIS-7# electrolyte. for sulphur, instead of highly ordered mesoporous carbon. It is the brittleness of such materials prevents their use in large-surface 47,48 shown that the electrochemical performance using SIS-7# practical systems . electrolyte is much better than that of 2# (Supplementary Our Li–S batteries using SIS-7# electrolyte exhibit excellent Fig. S7). The cycling performance is greatly improved and the electrochemical performance for all three aspects: (i) initial coulombic efficiency is nearly 100% even with non-porous carbons specific capacity of C/S composite with 60wt.% sulphur as the support. This further demonstrates the extraordinary loading: 625 mA h g , (ii) high capacity retention: 74% after properties of this new SIS electrolyte. 100 cycles and (iii) high coulombic efficiency: nearly 100% Currently, the electrochemical performance of Li–S batteries is (Supplementary Table S2). Additionally, it also shows excellent directly determined by the sulphur loading in C/S composites, the rate performance and works well in a wide temperature range even adsorbability of sulphur by mesoporous carbon, the solubility of to  20 C (Supplementary Fig. S8). lithium polysulphide in the electrolyte and the stability of the metallic lithium anode during cycling. Actually, to date, it has seemed almost impossible to simultaneously solve all these Discussion problems via a single solution. To improve cyclic performances, All those significant improvements derive from two reasons. First, one strategy is to decrease sulphur loading of the C/S composite the dissolution of lithium polysulphide is effectively reduced, thus (30–50wt.%), which not only increases electronic conductivity of avoiding lithium polysulphide shuttle phenomenon. In Li–S C/S composites, but also enhances the absorption effect for sulphur batteries, the soluble intermediate is mainly Li S (8rnr4) 2 n and polysulphide, finally improving reversible storage capacity, produced during initial discharge and final charge stages. It can cycling life and coulombic efficiency. However, according to our be considered to be a lithium salt, and its solubility has a certain estimate of practical energy density shown in Supplementary Table saturation degree for a given solvent. Thus, for the electrolyte S1, it can be seen that the sulphur loading should be higher than with ultrahigh lithium salt concentration such as SIS-7#, the 50wt.%; otherwise, it is difficult to achieve higher gravimetric concentration is almost close to saturation, in which case the energy density than that of advanced lithium-ion batteries. The soluble intermediate (Li S ) becomes hardly soluble, as further 2 n other strategy is to add strong absorption materials such as proven in Fig. 4. The colour of 7# (SIS-7#) is nearly unchanged nanosized mesoporous SiO (ref. 13) or Mg Ni O (ref. 42), after standing for 18 days. In contrast, all others show colour 2 0.6 0.4 Al O (ref. 43) into a sulphur electrode, which increases the change from dark-brown to yellow, which is consistent with the 2 3 coulombic efficiency to some extent; however, the dissolution of ultraviolet-visible (UV-Vis) spectra (Fig. 4b). After equal lithium polysulphide is not completely inhibited. proportional dilution, those three samples (2#, 4# and 7#) are The third strategy is to use LiNO as an effective additive to obviously different in UV-Vis light absorption. Among them, the 0 0 stabilize the metallic lithium anode, via an in situ formation of a curves of 7#, 4# and 7# are nearly superimposed in the whole 44,45 protective layer on the lithium anode surface . However, the region, which indicates that the dissolution of lithium poly- protective layer can only prevent further reaction between the sulphide is indeed inhibited in the SIS-7# electrolyte. During the lithium polysulphide and the lithium anode, it is unable to inhibit review process, we noticed that a very recent work simply showed the dissolution of lithium polysulphide into the electrolyte, which that a higher salt concentration electrolyte (but only up to 5 M) is 16,46 results in instability in terms of long cycling . Finally protecting helpful to decrease both the dissolution of lithium polysulphide the metallic lithium anode using inorganic solid electrolyte layer and the diffusion coefficient of bulky polysulphide in the prevents any soluble polysulphide reaching the lithium metal, but electrolyte, thus improving the coulombic efficiency . Actually, 4 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. Coulombic efficiency (%) Voltage (V) –1 –1 Capacity (mA h g ) - sulphur Capacity (mA h g ) - sulphur NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE Visible light region (400–760nm) 18 Days 2# 4# 7# 4#’ 7#′ –1 1 Day –2 150 300 450 600 750 900 1,050 Wavelength (nm) 1 h –1 –1 –1 –1 0 mol l 2 mol l 4 mol l 7 mol l Figure 4 | Lithium polysulphide dissolution experiments. (a) The colour changes of four samples with different salt concentrations containing the same amount of Li S were recorded by digital camera along with time: 0 mol per l solvent, 2#: 2 mol per l solvent, 4#: 4 mol per l solvent, 7#: 7 mol per l solvent. 2 8 (b) The ultraviolet-visible spectrophotometry. The 2#, 4# and 7# for UV-Vis measurement are the solutions from the standing samples for 18 days, which are then diluted with the corresponding electrolytes in a volume ratio of 1:7. The curves of 4#’ and 7#’ are the pure electrolytes of 4 mol per l solvent and 7 mol per l solvent for comparison, respectively. the dissolution of lithium polysulphide is not totally prohibited hand, high viscosity limits anion convection near deposition area, because in this case the highest concentration used was 5 M which is also helpful to deposit uniformly . LiTFSI in DME/DOL. Meanwhile, lithium cycling efficiencies in different electrolyte Second, a metallic lithium anode is more stable in the SIS-7# systems were investigated by means of a Li deposition-disso- lution experiment according to Aurbach et al. (2#: Fig. 5e, 4#: electrolyte than in other electrolytes. Compared with other kinds of rechargeable metallic lithium batteries, it is more complicated Fig. 5f and SIS-7#: Fig. 5g). In the commonly used low-salt to stabilize a metallic lithium anode in a Li–S battery owing to concentration electrolyte system, the lithium cycling efficiency is double damage from dendrite formation and the side reaction estimated to be below 50%, which means a large excess of lithium between lithium polysulphide and metallic lithium during as an anode needed in a real battery, thus decreasing the energy cycling. From the scanning electron microscopy (SEM) images density and increasing the cost. In contrast, interestingly, a shown in Fig. 5a–d, it is obvious that SIS-7# shows the lowest lithium cycling efficiency as high as 71.4% is obtained in the roughness and damage level of metallic lithium anode compared SIS-7# electrolyte system, which is much higher than that of with the other three samples (2#: Fig. 5b, 4#: Fig. 5c and SIS-7#: other LiTFSI-based non-aqueous electrolytes .Furthermore, from Fig. 5d), which demonstrates that the SIS electrolyte system can SEM images (Supplementary Fig. S9), a smoother and more effectively reduce the corrosion and suppress the formation of uniform lithium deposition was also clearly observed in the SIS-7# lithium dendrites owing to the ultrahigh lithium salt concentra- electrolyte system. This important issue for real applications has tion and high viscosity. It is generally accepted that dendrites start seldom been discussed in previous work on Li–S batteries. to grow in the non-aqueous liquid electrolyte when the anion is The different lithium cycling efficiencies in different electrolyte depleted in the vicinity of the electrode where plating occurs systems could be ascribed to the nature of solid electrolyte 50,51 according to Chazalviel model . In the case of SIS electrolyte interphase (SEI) formed on the surface of lithium electrodes. To with ultrahigh salt concentration, there is a mass of anion to keep confirm our conjecture, X-ray photoelectron spectroscopy the balance of cation (Li ) and anion (TFSI ) near metallic analysis of Li electrode surfaces was performed. Based on the 53–57 lithium anode, and the space charge that is created by TFSI previous work from Aurbach group and our experimental depletion is minimal, thus this is not a favourable condition for data, it could be concluded that the solvents of DME and DOL in dendrite growth. Furthermore, owing to both ultrahigh lithium electrolytes could possibly be reduced to ROLi species and salt concentration and high lithium-ion transference number oligomers with  OLi end group, and the salt of LiTFSI could (0.73), SIS electrolyte provides a large amount of available possibly be reduced to Li N, LiF, C F Li and sulphur-containing 3 2 x y lithium-ion flux and raises the lithium ionic mass transfer rate compounds such as Li S, Li S O ,Li SO or SO CF during the 2 2 2 4 2 3 2 x between electrolyte and metallic lithium electrode, thereby cycling process to form an SEI layer on the metallic lithium enhancing the uniformity of lithium deposition and dissolution surface (Supplementary Fig. S10). The composition of SEI for in charge/discharge process. Besides, influenced by high viscosity, both 2# and SIS-7# electrolytes seems to be similar; however, the on the one hand, it possibly increases the pressure from the thickness is different. In the case of SIS-7# electrolyte, it can be electrolyte to push back growing dendrites, resulting in a more seen that the lithium metal signal appears after 150 s sputtering uniform deposition on the surface of the anode. On the other with Ar ions, which etches the surface layer-by-layer from a NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 5 & 2013 Macmillan Publishers Limited. All rights reserved. Absorption ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 3.0 2# 2.5 2.0 1.5 Overpotential dissolution 1.0 0.5 0.0 2# 0.06 0.03 0.00 –0.03 –0.06 8:11:14 16:31:40 Time (h) 3.0 2.5 2.0 4# 1.5 Overpotential dissolution 1.0 0.5 0.0 4# 0.06 0.03 0.00 –0.03 –0.06 8:10:11 16:30:40 24:42:39 Time (h) 3.0 SIS-7# 2.5 2.0 1.5 1.0 0.5 0.0 SIS-7# 0.06 0.03 0.00 –0.03 –0.06 24:52:41 49:58:41 74:22:14 Time (h) Figure 5 | Typical scanning electron microscopy images of metallic lithium anodes and lithium deposition and dissolution experiments. (a) Fresh lithium metal, (b) Lithium metal with 2# electrolyte after 278 cycles. (c) Lithium metal with 4# electrolyte after 183 cycles. (d) Lithium metal with SIS-7# electrolyte after 280 cycles. The white scale bar represents 60mm for all the images; lithium deposition and dissolution experiments by electrochemically 2  2 depositing Li on Cu foil (Current: 0.1 mA cm ; the first deposition: 5 C cm ; cycling process: 10% depth of discharge and 20 cycles; final dissolution: charged to 3 V). (e) 2# electrolyte. (f) 4# electrolyte. (g) SIS-7# electrolyte. The lithium cycling efficiency is calculated according to Aurbach et al. lithium electrode. However, the signal still does not show up after DME:DOL ¼ 1:1 by volume (Ferro, Suzhou, China) were mixed by ratio of mole number to volume, which are 1#: 1 mol per l, 2#: 2 mol per l, 3#: 3 mol per l, 4#: 800 s sputtering for the lithium electrode cycled in the low-salt 4 mol per l, SIS-5#: 5 mol per l, SIS-6#: 6 mol per l and SIS-7#: 7 mol per l, concentration electrolyte (Fig. 6). This indicates that the thickness respectively. The mixtures were stirred for 24 h at room temperature. All the is different in these two electrolyte systems. experiments were performed in an argon-filled glove box. In summary, we propose a new class of SIS electrolyte for the The carbon/sulphur composite is prepared by a melt-diffusion method. According to Ji et al. , CMK-3 (Nanjing XF-nano, China) or Ketjenblack and next-generation high-energy rechargeable metallic lithium bat- sulphur were ground together, sealed in a glass tube containing argon, and heated teries and take the electrolyte system of LiTFSI and ether solvents at 155 C for 24 h. (see Supplementary Figs S11–S13 and Supplementary Table S3). as an example to demonstrate the power of SIS electrolyte. For the appealing application in Li–S batteries, it is demonstrated that SIS Characterizations. The conductivity of electrolyte system was measured with an electrolytes can not only inhibit the dissolution of lithium impedance analyser (Zahner Zennium, Germany) over a temperature range of polysulphide but also effectively protect a metallic lithium anode 20 to 60 C in a thermostated container. The lithium-ion transference numbers against the formation of lithium dendrites, which makes the cell were obtained by combining alternating-current (AC) impedance and direct-cur- exhibit both excellent electrochemical performance and high rent (DC) polarization measurements using a symmetric Li/electrolyte/Li cell. First, AC impedance test was performed to obtain a total resistance R . Then DC safety. As a matter of fact, we anticipate further conceptual design cell polarization was carried out to obtain a stable current I . The lithium-ion DC of other SIS electrolyte systems by selecting appropriate salt and transference number was calculated by the formulas (t ¼ R /R and R ¼ Li þ cell DC DC solvent components and adjusting these proportions according to V /I ). DSC measurements were performed with a NETZSCH DSC 200F3 by DC DC different requirements for different rechargeable metallic lithium sealing B10 mg of the sample in an aluminium pan. The pan and the electrolyte were first cooled to about  150 C with liquid nitrogen and then heated to 50 C batteries (for example Li-O batteries). It is expected that this at a heating rate of 10 Cmin under nitrogen flow. The viscosity (Z)ofthe class of SIS electrolyte would offer a new and important approach electrolytes was measured on a programmable viscometer (Brookfield, DV-iii þ)at for improving the electrochemical and safety performance for 25 C in a homemade dry chamber (H O o10 p.p.m.), and the temperature was metallic lithium batteries and further make us reconsider them accurately controlled to within 0.1 C using a Brookfield TC-502 oil bath. A SEM microscopy (Hitachi S-4800) equipped with a vacuum transfer box was without their past safety concerns. used to study the morphology of the samples. The samples of lithium metal anode were stored in vacuum transfer box in an argon-filled glove box and transferred Methods into SEM chamber without exposure to air. Synthesis. The SIS electrolytes are prepared as follows: lithium bis(tri- X-ray photoelectron spectroscopy analyses were performed in ESCALAB 250 fluoromethane sulphonyl) imide (LiN(SO CF ) , LiTFSI) (TCI, Japan) and purified using a monochromatized Al Ka source and equipped with an Ar ion sputtering 2 3 2 6 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. Current (mA) Voltage (V) Current (mA) Voltage (V) Current (mA) Voltage (V) NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE 3 k Experimental data Peak1-(58 ± 0.5 eV) Li-F Peak2-(55 ± 0.5 eV) Li-O 2 k Fit curve 1 k 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 Binding energy (eV) 0 s 30 s 90 s 150 s 500 s 800 s Li F S 0 30 90 150 500 800 Etching time (s) 3k Experimental data Peak1-(58 ± 0.5 eV) Li-F O Li Li Peak2-(55 ± 0.5 eV) Li-O 2k O Peak3-(53 ± 0.5 eV) Li Li Fit curve 1k 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 Binding energy (eV) 0 s 30 s 60 s 90 s 150 s 500 s 800 s Li 30 O 0 30 60 90 150 500 800 Etching time (s) Figure 6 | X-ray photoelectron spectroscopy (XPS) analysis of metallic lithium anodes. After 100 cycles at a current rate of 0.2C with 2# and SIS-7# electrolytes, (a) XPS Li spectra of metallic lithium electrode using 2# electrolyte before sputtering and after sputtering with different times. (b) Summary 1s of the atomic concentration of Li, C, O, S, N and F on the Li anode surface as a function of sputtering time using 2# electrolyte. (c) XPS Li spectra of 1s metallic lithium electrode using SIS-7# electrolyte before sputtering and after sputtering with different times. (d) Summary of the atomic concentration of Li, C, O, S, N and F on the Li anode surface as a function of sputtering time using SIS-7# electrolyte. gun (Thermo Fisher). Ar etching was conducted at an argon partial pressure of Electrochemistry. The electrode was fabricated by mixing C/S composite, carbon 10 Torr in the x–y scan mode at ion acceleration of 3 kV and ion beam current nanotube, poly(vinylidene difluoride) in weight ratio of 8:1:1 for cycle life mea- density of 1mAmm . surement and the other weight ratio of 7:2:1 for rate capability measurement. The The lithium polysulphide dissolution experiments were carried out in the slurry was cast on carbon-coated Al current collector and dried at 50 C in vacuum following: Li S and S with a mole ratio of 1:7 (49.5 and 224 mg) were mixed for 10 h. The coin cells CR2032 were assembled with the electrode, pure lithium foil and added in 2-ml pure solvent, 2 mol salt per 1-l solvent (2#), 4 mol salt per 1-l as counter electrode and a glass fibre separator in an argon-filled glove box. The solvent (4#) and 7 mol salt per 1-l solvent (SIS-7#), respectively. The colour discharge and charge measurements at room temperature and low-temperature changes of solutions with time were observed and recorded by digital camera. range of  20 to 0 C in a thermostated container were carried out on a Land The samples of 2#, 4# and SIS-7# after standing for 18 days were diluted BT2000 Battery Test System (Wuhan, China). with the corresponding electrolytes in the ratio of 1:7 for UV-Vis measurement, The average cycling efficiency of metallic lithium electrodes in various which was carried out on a Cary 5000 UV-Vis spectrophotometry (Varian, electrolytes (2#, 4#, SIS-7#) was performed by electrochemically depositing Li 0  2 America). The control samples are pure electrolytes without Li S (4# on Cu foil (5 C cm ) followed by Li stripping and deposition cycling (10% depth 2 8 0  2 and SIS-7# ). of discharge-DOD, 0.1 mA cm , 20 cycles). The residual Li was then dissolved X-ray powder diffraction analysis was characterized by X’Pert Pro MPD X-ray electrochemically (charged to 3 V) on a Land BT2000 Battery Test System. The diffractometer (Philips, Holland) using Cu–Ka radiation (1.5405 Å). average Li cycling efficiency was calculated according to Aurbach et al. NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 7 & 2013 Macmillan Publishers Limited. All rights reserved. Intensity/counts per second Atomic concentration (%) Intensity/counts per second Atomic concentration (%) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 References 32. Nagao, M., Hayashi, A. & Tatsumisago, M. Fabrication of favorable interface 1. Armand, M. & Tarascon, J. M. Building better batteries. Nature 451, 652–657 between sulfide solid electrolyte and Li metal electrode for bulk-type solid-state (2008). Li/S battery. Electrochem. Commun. 22, 177–180 (2012). 2. Goodenough, J. B. & Kim, Y. Challenges for rechargeable Li batteries. Chem. 33. Angell, C. A., Liu, C. & Sanchez, E. Rubbery solid electrolytes with Mater. 22, 587–603 (2010). dominant cationic transport and high ambient conductivity. Nature 362, 3. Scrosati, B., Hassoun, J. & Sun, Y. K. Lithium-ion batteries. A look into the 137–139 (1993). future. Energy Environ. Sci. 4, 3287–3295 (2011). 34. Hu, Y. S., Li, H., Huang, X. J. & Chen, L. Q. Novel room temperature molten 4. Choi, N. S. et al. Challenges facing lithium batteries and electrical double-layer salt electrolyte based on LiTFSI and acetamide for lithium batteries. capacitors. Angew. Chem. Int. Ed. 51, 9994–10024 (2012). Electrochem. Commun. 6, 28–32 (2004). 5. Zu, C. X. & Li, H. Thermodynamic analysis on energy densities of batteries. 35. Henderson, W. A. et al. Glyme-lithium bis(trifluoromethanesulfonyl)imide and Energy Environ. Sci. 4, 2614–2624 (2011). glyme-lithium bis(perfluoroethanesulfonyl)imide phase behavior and solvate 6. Ji, X. & Nazar, L. F. Advances in Li-S batteries. J. Mater. Chem. 20, 9821–9826 structures. Chem. Mater. 17, 2284–2289 (2005). (2010). 36. Tachikawa, N. et al. Reversibility of electrochemical reactions of sulphur 7. Bruce, P. G., Freunberger, S. A., Hardwick, L. J. & Tarascon, J.-M. Li-O and 2 supported on inverse opal carbon in glyme-Li salt molten complex electrolytes. Li-S batteries with high energy storage. Nat. Mater. 11, 19–29 (2012). Chem. Commun. 47, 8157–8159 (2011). 8. Evers, S. & Nazar, L. F. New approaches for high energy density 37. Yoshida, K. et al. Oxidative-stability enhancement and charge transport lithium-sulphur battery cathodes. Acc. Chem. Res. doi:10.1021/ar3001348 mechanism in glyme-lithium salt equimolar complexes. J. Am. Chem. Soc. 133, (2012). 13121–13129 (2011). 9. Manthiram, A., Fu, Y. Z. & Su, Y. S. Challenges and prospects of 38. Angell, C. A. Electrical conductance of concentrated aqueous solutions and molten salts: correlation through free volume transport model. J. Phys. Chem. lithium–sulphur batteries. Acc. Chem. Res. doi:10.1021/ar300179v (2012). 69, 2137 (1965). 10. Ji, X. L., Lee, K. T. & Nazar, L. F. A highly ordered nanostructured 39. Angell, C. A. A new class of molten slat mixtures the hydrated dipositive carbon-sulphur cathode for lithium-sulphur batteries. Nat. Mater. 8, 500–506 ion as an independent cation species. J. Electrochem. Soc. 112, 1224–1227 (2009). (1965). 11. Hassoun, J. & Scrosati, B. A high-performance polymer tin sulphur lithium ion 40. Kawamura, T., Kimura, A., Egashira, M., Okada, S. & Yamaki, J. I. Thermal battery. Angew. Chem. Int. Ed. 49, 2371–2374 (2010). stability of alkyl carbonate mixed-solvent electrolytes for lithium ion cells. 12. Elazari, R., Salitra, G., Garsuch, A., Panchenko, A. & Aurbach, D. J. Power Sources 104, 260–264 (2002). Sulfur-impregnated activated carbon fiber cloth as a binder-free cathode for 41. Mikhaylik, Y. V. & Akridge, J. R. Polysulfide shuttle study in the Li/S battery rechargeable Li-S batteries. Adv. Mater. 23, 5641–5644 (2011). system. J. Electrochem. Soc. 151, A1969–A1976 (2004). 13. Ji, X. L., Evers, S., Black, R. & Nazar, L. F. Stabilizing lithium-sulphur cathodes 42. Song, M. S. et al. Effects of nanosized adsorbing material on electrochemical using polysulphide reservoirs. Nat. Commun. 2, 235–241 (2011). properties of sulphur cathodes for Li/S secondary batteries. J. Electrochem. Soc. 14. Demir-Cakan, R. et al. Cathode composites for Li-S batteries via the 151, A791–A795 (2004). use of oxygenated porous architectures. J. Am. Chem. Soc. 133, 16154–16160 43. Choi, Y. J. et al. Electrochemical properties of sulphur electrode containing (2011). nano Al O for lithium/sulphur cell. Phys. Scr. 62, T129 (2007). 2 3 15. Wang, H. L. et al. Graphene-wrapped sulphur particles as a rechargeable 44. Mikhaylik, Y. A. Electrolytes for lithium sulphur cells. US Patent no. 0147891 lithium-sulphur battery cathode material with high capacity and cycling (2005). stability. Nano Lett. 11, 2644–2647 (2011). 45. Aurbach, D. et al. On the surface chemical aspects of very high energy 16. Zheng, G. Y., Yang, Y., Cha, J. J., Hong, S. S. & Cui, Y. Holow carbon nanofiber- density, rechargeable Li–sulphur batteries. J. Electrochem. Soc. 156, A694–A720 encapsulated sulphur cathodes for high specific capacity rechargeable lihtium (2009). batteries. Nano Lett. 11, 4462–4467 (2011). 46. Liang, X. et al. Improved cycling performances of lithium sulphur batteries with 17. Jayaprakash, N., Shen, J., Moganty, S. S., Corona, A. & Archer, A. Porous LiNO -modified electrolyte. J. Power Sources 196, 9839–9843 (2011). hollow carbon@sulphur composites for high-power lithium-sulphur batteriess. 47. Steven, J. V., Yevgeniy, S. N. & Bruce, D. K. Ionically conductive membranes Angew. Chem. Int. Ed. 50, 5904–5908 (2011). for protection of active metal anodes and battery cells. US Patent no. 0191617 18. Schuster, J. et al. Spherical ordered mesoporous carbon nanoparticles with high (2004). porosity for lithium-sulphurbatteries. Angew. Chem Int. Ed. 51, 3591–3595 48. Lutgard, D. J., Yevgeniy, S. N. & Steven, J. V. Alleviation of voltage delay in (2012). lithium-liquid depolarizer/electrolyte solvent battery cells. US Patent no. 674558 19. Schuster, J. & Nazar, L. F. Graphene-enveloped sulfur in a one pot reaction: a (2010). cathode with good coulombic efficiency and high practical sulfur content. 49. Shin, S. E., Kim, K., Oh, S. H. & Cho., W. I. Polysulfide dissolution control: the Chem. Commun. 48, 1233–1235 (2012). common ion effect. Chem. Commun. doi:10.1039/c2cc36986a (2012). 20. Su, Y. S. & Manthiram, A. Lithium-sulphur batteries with a microporous 50. Chazalviel, J. N. Electrochemical aspects of the generation of ramified metallic carbon paper as a bifunctional interlayer. Nat. Commun. 3, 1166 (2012). electrodeposits. Phys. Rev. A 42, 7355–7367 (1990). 21. Xia, L. F. et al. A soft approach to encapsulate sulphur: polyaniline nanotubes 51. Rosso, M. Electrodeposition from a binary electrolyte: new developments and for lithium-sulphur batteries with long cycle life. Adv. Mater. 24, 1176–1181 applications. Electrochim. Acta 53, 250–256 (2007). (2012). 52. Gonzalez, G., Marshall, G., Molina, F. V., Dengra, M. S. & Rosso, M. Viscosity 22. Yang, Y. et al. Improving the performance of lithium-sulphur batteries by effects in thin layer electrodeposition. J. Electrochem. Soc. 148, C479–C487 conductive polymer coating. ACS Nano 5, 9187–9193 (2011). (2001). 23. Xu, K. Nonaqueous liquid electrolytes for lithium-based rechargeable batteries. 53. Aurbach, D., Gofer, Y. & Langzam, J. The correlation between surface Chem. Rev 104, 4303–4417 (2004). chemistry, surface morphology, and cycling efficiency of lithium electrodes in a 24. Armand, M. B., Chabagno, J. M. & Duclot, M. J. In: Vashishta, P., few polar aprotic systems. J. Electrochem. Soc. 136, 3198–3205 (1989). Mundy, J.M. & Shenoy, G.K. (eds). Fast Ion Transport in Solids (Elsevier, 1979; 54. Aurbach, D. & Gront, E. The study of electrolyte solutions based on solvents pp 131–136). from the ‘glyme’ family (linear polyethers) for secondary Li battery systems. 25. Croce, F., Appetecchi, G. B., Persi, L. & Scrosati, B. Nanocomposite polymer Electrochim. Acta 42, 697–718 (1997). electrolytes for lithium batteries. Nature 394, 456–458 (1998). 55. Aurbach, D., Youngman, O., Gofer, Y. & Meitav, A. The electrochemical 26. Gadjourova, Z., Andreev, Y. G., Tunstall, D. P. & Bruce, P. G. Ionic behavior of 1,3-dioxolane-LiClO solutions. Electrochim. Acta 35, 625–638 conductivity in crystalline polymer electrolytes. Nature 412, 520–523 (2001). (1990). 27. Robertson, A. D., West, A. R. & Ritchie, A. G. Review of crystalline lithium-ion 56. Aurbach, D., Weissman, I. & Schechter, A. X-ray photoelectron conductors suitable for high temperature battery applications. Solid State Ionics spectroscoscopy studies of lithium surface prepared in several important 104, 1–11 (1997). electrolyte solution. A comparison with previous studies by fourier tansform 28. Kondo, S., Takada, K. & Yamamura, Y. New lithium ion conductors based on infrared spectroscopy. Langmuir 12, 3991–4007 (1996). Li S-SiS . Solid State Ionics 53, 1183–1186 (1992). 2 2 57. Schechter, A. & Aurbach, D. X-ray photoelectron spectroscoscopy study of 29. Kamaya, N. et al. A lithium superionic conductor. Nat. Mater. 10, 682–686 surface films formd on Li electrodes freshly prepared in alkyl carbonate (2011). solutions. Langmuir 15, 3334–3342 (1999). 30. Hayashi, A., Ohtomo, T., Mizuno, F., Tadanaga, K. & Tatsumisago, M. 58. Bhattacharyya, A. J. & Maier, J. Second phase effects on the conductivity of All-solid-state Li/S batteries with highly conductive glass-ceramic electrolytes. non-aqueous salt solutions: ‘Soggy sand electrolytes’. Adv. Mater. 16, 811–814 Electrochem. Commun. 5, 701–705 (2003). (2004). 31. Nagao, M., Hayashi, A. & Tatsumisago, M. Sulfide-carbon composite electrode 59. MacFarlane, D. R., Huang, J. H. & Forsyth, M. Lithium-doped plastic crystal for all-solid-state Li/S battery with Li S-P S solid electrolyte. Electrochim. Acta 2 2 5 electrolytes exhibiting fast ion conduction for secondary batteries. Nature 402, 56, 6055–6059 (2011). 792–794 (1999). 8 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE L.M.S. and Y.-S.H. wrote the paper; L.M.S., Y.-S.H., H.L., M.A. and L.Q.C. discussed the 60. Liang, C. C. Conduction characteristics of lithium iodide aluminum oxide solid results and participated in the preparation of the paper. electrolytes. J. Electrochem. Soc. 120, 1289–1292 (1973). Additional information Acknowledgements Supplementary Information accompanies this paper at http://www.nature.com/ We thank Profs Xuejie Huang and Zhibin Zhou for fruitful discussions and help on DSC naturecommunications and viscosity measurements. This work was supported by funding from the ‘863’ Project (2009AA033101), ‘973’ Projects (2009CB220104), NSFC (50972164, 51222210), CAS Competing financial interests: The authors declare no competing financial interests. project (KJCX2-YW-W26) and One Hundred Talent Project of the Chinese Academy of Sciences. Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ Author contributions Y.-S.H. conceived and designed this work; L.M.S. performed the experiment; L.M.S., How to cite this article: Suo, L. et al. A new class of Solvent-in-Salt electrolyte Y.-S.H., H.L. and L.Q.C. analysed the physicochemical properties of this new electrolyte; for high-energy rechargeable metallic lithium batteries. Nat. Commun. 4:1481 Y.-S.H. and M.A. analysed the inhibition mechanism of lithium dendrite formation; doi: 10.1038/ncomms2513 (2013). NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 9 & 2013 Macmillan Publishers Limited. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/a-new-class-of-solvent-in-salt-electrolyte-for-high-energy-tLy7qDNjUL

References (70)

Publisher
Springer Journals
Copyright
Copyright © 2013 by Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved.
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/ncomms2513
Publisher site
See Article on Publisher Site

Abstract

ARTICLE Received 28 Aug 2012 | Accepted 10 Jan 2013 | Published 12 Feb 2013 DOI: 10.1038/ncomms2513 A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries 1 1 1 1 1 Liumin Suo , Yong-Sheng Hu , Hong Li , Michel Armand & Liquan Chen Liquid electrolyte plays a key role in commercial lithium-ion batteries to allow conduction of lithium-ion between cathode and anode. Traditionally, taking into account the ionic con- ductivity, viscosity and dissolubility of lithium salt, the salt concentration in liquid electrolytes is typically less than 1.2 mol l . Here we show a new class of ‘Solvent-in-Salt’ electrolyte with ultrahigh salt concentration and high lithium-ion transference number (0.73), in which salt holds a dominant position in the lithium-ion transport system. It remarkably enhances cyclic and safety performance of next-generation high-energy rechargeable lithium batteries via an effective suppression of lithium dendrite growth and shape change in the metallic lithium anode. Moreover, when used in lithium–sulphur battery, the advantage of this electrolyte is further demonstrated that lithium polysulphide dissolution is inhibited, thus overcoming one of today’s most challenging technological hurdles, the ‘polysulphide shuttle phenomenon’. Consequently, a coulombic efficiency nearing 100% and long cycling stability are achieved. Key Laboratory for Renewable Energy, Beijing Key Laboratory for New Energy Materials and Devices, Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China. Correspondence and requests for materials should be addressed to Y.-S.H. (email: [email protected]). NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 1 & 2013 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ithium-based batteries as highly efficient energy storage The physicochemical properties of ‘Solvent-in-Salt’. For the devices have long been considered as promising power following discussion, we select an electrolyte system containing Lsupply for various electric vehicles and smart grid storage Li[CF SO ) N] (LiTFSI), one of the lowest lattice energy salts and 3 2 2 1–4 systems . However, presently available lithium-ion technology 1,3-dioxolane (DOL): dimethoxyethane (DME) (1:1 by volume) cannot satisfy the increasing demand for energy density. Metallic as solvent, resulting in what we show as one of the most pro- lithium batteries exhibit the highest theoretical energy densities mising electrolytes for Li–S batteries. The physicochemical among the secondary batteries . Among many possible systems, properties of this electrolyte with different ratios of salt-to-solvent lithium–sulphur (Li–S) and lithium–oxygen (Li–O ) batteries are are illustrated in Fig. 2. When the mole amount of salt reaches quite attractive as candidates for next-generation high-energy 4 mol in 1-l solvent, the electrolyte enters the D region of SIS by 5–22 density batteries . However, the application of metallic lithium weight, and beyond 5 mol salt in 1-l solvent, the salt begins to anode suffers from inevitable formation of lithium dendrites, which have a dominant role in either weight or volume ratio (Fig. 2a). are caused by uneven current distributions at the metal–electrolyte Arrhenius plots of the ionic conductivity of the electrolytes with interface during cycling. The formation of lithium dendrites will different salt concentrations in a temperature range of  20 to lead to poor cyclic performance and increase the probability of 60 C are shown in Fig. 2b and exhibit the typical curvature of the internal short circuit, resulting in safety issues. Thus, great effort Vogel–Tammann–Fulcher (VTF) equation (Supplementary has been paid to suppress the formation of lithium dendrites and Fig. S1). It can be seen that the ionic conductivity decreases reduce metallic lithium corrosion in liquid electrolytes .For with increasing salt concentration. The conductivity drops instance, the use of solid polymer or inorganic electrolytes could slowly in the high temperature region (20–60 C) but rapidly potentially suppress the formation of the lithium dendrites. at a low-temperature region (  20 to 20 C), resulting from a rise However, for solid-state electrolytes, the kinetic properties are in T . For a given electrolyte with fixed salt and solvent, ionic limited, due to both low conductivity at room temperature and conductivity depends on both viscosity and lithium-ion 24–28 high interfacial resistance . Recently, a breakthrough has been mobility. When increasing salt concentration, more and more achieved in inorganic sulphide-based electrolytes with quite high Li–ether complex pairs form due to incomplete solvation 2  1 room temperature conductivity of 10 Scm (ref. 29). The shell and the viscosity at room temperature increases markedly in application of Li S–P S glass–ceramic electrolyte in Li–S batteries the SIS region (Fig. 2c and Supplementary Fig. S2). At the same 2 2 5 30–32 has been demonstrated . To reduce the interfacial resistance, time, the lithium-ion transference number of SIS-7# electrolyte þ þ þ 30wt.% solid electrolyte and 35wt.% acetylene black were required increases to an unexpected high value (t ¼ 0.73, t ¼ s / Li Li Li to add into sulphur electrode . This leads to a decrease in energy (s þ s )) (see Fig. 2c and Supplementary Fig. S3), which is Li TFSI density of Li–S batteries. It seems that the strategy of physical much higher than that of traditional salt-in-solvent electrolytes mixing of active phase and solid electrolyte still suffers from point- (0.2–0.4). The conductivity of specific ion i is proportional to the 30–32 to-point contact , not like a complete surface-to-surface concentration of mobile ion (c ) and its mobility (m )(s ¼ nc m ). i i i i i wetting effect as liquid electrolyte. The mobility of an ion is determined by the viscosity (Z) Here we report a new class of non-aqueous liquid ‘Solvent-in- of the medium and radius of mobile ion (m ¼ 1/6pZr ) (ref. 23). i i Salt’ electrolytes and apply them in Li–S batteries. It is demon- In low-salt concentration electrolytes, lithium ions are strated that the use of ‘Solvent-in-Salt’ electrolyte inhibits the coordinated with ether oxygen and form a large solvation dissolution of lithium polysulphide, effectively protects metallic shell compared with anions, leading to relatively lower mobility lithium anodes against the formation of lithium dendrites and of solvated Li cations. In the SIS system, it is plausible that results in high lithium cycling efficiency, thus enhancing electro- the number of solvated Li cations is decreased and large chemical performance. anion (TFSI ) could be more seriously dragged than the small unsolvated cation (Li ) in this high viscosity system. Nevertheless, SIS-7# even with a high viscosity of 72 cP retains a Results conductivity of 0.814 mS cm at room temperature, which The concept of ‘Solvent-in-Salt’ electrolyte. In 1993, Angell remains superior to that of all-solid-state dry polymer or most of et al. proposed the innovative concept of ‘Polymer-in-Salt’ by the inorganic electrolytes, and especially, could form better reversing the ratio of solid polymer solvent to salt, in which glass interfacial contacts. transition (T ) was low enough to remain rubbery at room tempe- Differential scanning calorimeter (DSC) traces reveal distinct rature to preserve good conductivity and high electrochemical glass transition temperatures (Fig. 2d), which shows that all the stability. However, in practice the T remained above ambient electrolytes are glass-forming liquids, and their glass transition temperature and/or the system would crystallize. For temperatures shift from low to high, with increasing ratio of salt- conventional non-aqueous organic electrolytes, the salt to-solvent. For the pure solvent mixture without salt concentration is usually limited in a range of 1–2 mol l , (Supplementary Fig. S4), the glass transition temperature is which is a trade-off among ionic conductivity, viscosity and salt  138.6 C, justifying the excellent low temperature performance solubility. Thus, most of the studies focus on the region C of for the DOL–DME-based electrolyte. In the SIS-7# electrolyte, the Fig. 1, in which there is much less salt than solvent. There are few value of T is –77.3 C, which is much lower than typical reports of research in the A (yellow) or D (green) regions of ‘Polymer-in-Salt’ system (T 4–10 C) (ref. 33) and close to that Fig. 1, in which either the weight or volume ratio of salt-to- traditional commercial electrolyte systems (1 mol l LiPF in solvent exceeds 1.0 (refs 34–37). As a matter of fact, by choosing EC–DMC, T ¼ –67 C)) (ref. 40). The flexible hinge in the S–N– proper salt and solvent, we can move an electrolyte into those S bond of TFSI , specific to these imide anions, explains this regions (A, B and D) and also obtain some unexpected properties. ‘plasticizing effect’, which is also reflected in the low viscosities of To distinguish from traditional electrolytes, this new class of ionic liquids based on this anion. electrolyte is denoted by ‘Solvent-in-Salt’ (SIS). A similar system of hydrated molten salt composed of KNO and Ca(NO )  4H O, in which the water content is insufficient to Application in Li–S batteries. The power of the SIS electrolytes is 3 2 2 satisfy more than a first coordination sheath for the cation that demonstrated by their use in rechargeable metallic Li–S batteries. 38,39 was reported in 1965 by Angell . This kind of hydrated molten The Li–S battery using SIS-7# as electrolyte exhibits the best salts was usually used in heat storage but not called SIS. electrochemical performance (Fig. 3). 2 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. Lithium-ion transference Viscosity (cP) NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE Solid-state-based electrolytes Liquid-based electrolytes Salt content AB 1. Lithium ion conductor 100% Lithium molten salt 2. LiI/AI O 2 3 “Solvent-in-Salt” “Solvent-in-Salt” ‘Solvent-in-Salt’ by volume by volume and weight ([salt] > [solvent] fig.1b) “Polymer-in-Salt” ([salt] > [polymer]) 50% Plastic crystal Soggy sand electrolyte electrolyte CD Traditional non-aqueons Dry polymer Gel polymer “Solvent-in-Salt” electrolytes ([salt] < [polymer]) ([salt] < [solvent]) by weight 0 1 ∞ 100% 0% Weight ratio of salt-to-solvent Solid-phase content (%) Liquid phase 58–60 Figure 1 | A general concept for Solvent-in-Salt electrolyte. (a) The overview of the available electrolytes .(b) The distribution map of non-aqueous liquid electrolytes with the weight and volume ratios of salt-to-solvent. A, B and D regions are Solvent-in-Salt electrolyte, in which the ratio of salt-to- solvent is over 1.0 by volume or weight. C region is [solvent]4[salt] by weight and volume. 2.2 –1.6 C region D region B region 2.0 (fig. 1b) (fig. 1b) (fig. 1b) –2.0 1.8 Salt-in-Solvent 1.6 –2.4 1.4 –2.8 Volume ratio SIS-7# 1.2 1# Weight ratio –3.2 2# 1.0 3# 0.8 –3.6 4# 0.6 SIS-5# –4.0 Solvent-in-Salt 0.4 SIS-6# 25 °C –4.4 SIS-7# 0.2 0.0 123 4567 3.0 3.2 3.4 3.6 3.8 4.0 –1 Ratio of salt-to-solvent (mol per l solvent) 1,000/T (K ) 80 0.9 T = –77.3 °C SIS-7# 0.8 12 T = –83.6 °C SIS-6# 60 g 10 0.7 T = –91.9 °C SIS-5# 8 40 T = –98.9 °C 0.6 g 4# 6 30 T = –109 °C g 3# 0.5 4 20 2# 2 10 0.4 1# 0 0 0.3 123 4567 –150 –120 –90 –60 –30 0 30 Ratio of salt-to-solvent (mol per l solvent) Temperature (°C) Figure 2 | Physicochemical properties of Solvent-in-Salt electrolytes. (a) Weight and volume ratio of salt-to-solvent with different ratios of LiTFSI to DOL:DME (1:1 by volume). (b) Arrhenius plots of the ionic conductivity as a function of 1,000/T for electrolytes with different ratios of LiTFSI to solvent, (1#: 1 mol per l solvent, 2#: 2 mol per l solvent, 3#: 3 mol l solvent, 4#: 4 mol per l solvent, 5#: 5 mol per l solvent, SIS-6#: 6 mol l per solvent and SIS-7#: 7 mol l per solvent). (c) Viscosity, ionic conductivity and lithium-ion transference number at room temperature for the aforementioned different electrolytes. (d) DSC traces of the aforementioned different electrolytes. It shows an initial specific discharge capacity of still show excellent rate capability as shown in Fig. 3d. They can 1  1  1 1,041 mA h g at a current rate of 0.2C (that is, 335 mA g ) achieve capacities of 1,229, 988, 864, 744 and 551 mA h g of 1  1 and maintains a reversible capacity of 770 mA h g with sulphur at current rates of 0.2, 0.5, 1, 2 and 3C, respectively. capacity retention of 74% after 100 cycles (Fig. 3b). More When the current rate returns to 0.2C, a reversible capacity of 1  1 importantly, the coulombic efficiency reaches nearly 100% after 789 mA h g remains. The capacity still decays after rate the first cycle (it is 93.7% for the first cycle.) for the SIS-7# measurement, although the polysulphide dissolution is inhibited, electrolyte (also see Supplementary Fig. S5), which is higher than which is probably related to the unstable C/S electrode. In the 12–22 previous reports in a similar system , owing to effectively charge–discharge process, the electrode suffers from a large avoiding the polysulphide shuttle effect in the charging process volume change by a conversion reaction between S 3  3 (Fig. 3c) (ref. 41). In contrast, other less concentrated electrolytes (2.07 g cm ) and Li S (1.66 g cm ), which could result in exhibit an apparent coulombic efficiency greater than 100% not only the sulphur redistribution but also the structural damage (Fig. 3c and Supplementary Fig. S6), a sign of the ‘polysulphide of the carbon–sulphur composite. Thus, we can expect that shuttle effect’. The only shortcoming of the use of SIS-7# is that through optimization of cathode materials, the cyclic perfor- the polarization becomes slightly larger due to relatively higher mance will be further improved. viscosity compared with traditional electrolytes with low-salt To further prove the power of this new SIS electrolyte, the concentration (Fig. 3a). Yet Li–S batteries using SIS-7# electrolyte Ketjenblack without mesoporous structure was used as a support NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 3 & 2013 Macmillan Publishers Limited. All rights reserved. “Solvent-in-Salt’’ –1 Salt/solvent ratio σ (mS cm ) Exothermic Endothermic Volume ratio of salt-to-solvent –1 Log σ (S cm ) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 1,400 3.0 4# SIS-7# charge SIS-7# 1,200 2.7 SIS-7# discharge 2# 1,000 2.4 2.1 1.8 1.5 400 2# charge 1.2 4# 4# charge SIS-7# 2# discharge 4# discharge 2# 0.9 0 200 400 600 800 1,000 1,200 1,400 010 20 30 40 50 60 70 80 90 100 –1 Capacity (mA h g ) - sulphur Cycle number (N) 120 180 1,200 0.2C 140 0.5C 1,000 1C 110 2# 0.2C 4# 2C 3C 100 60 SIS-7# 40 SIS-7# charge SIS-7# discharge 0 102030405060708090 100 0 102030405060708090 100 Cycle number (N) Cycle number (N) Figure 3 | Electrochemical performance of lithium–sulphur batteries. (a) First discharge–charge profiles of C/S electrodes in electrolytes with different ratios of LiTFSI to DOL:DME (1:1 by volume). (b) Cyclic performance. (c) Coulombic efficiency at a current rate of 0.2 C (coulombic efficiency ¼ charge capacity/discharge capacity). (d) Rate capability with SIS-7# electrolyte. for sulphur, instead of highly ordered mesoporous carbon. It is the brittleness of such materials prevents their use in large-surface 47,48 shown that the electrochemical performance using SIS-7# practical systems . electrolyte is much better than that of 2# (Supplementary Our Li–S batteries using SIS-7# electrolyte exhibit excellent Fig. S7). The cycling performance is greatly improved and the electrochemical performance for all three aspects: (i) initial coulombic efficiency is nearly 100% even with non-porous carbons specific capacity of C/S composite with 60wt.% sulphur as the support. This further demonstrates the extraordinary loading: 625 mA h g , (ii) high capacity retention: 74% after properties of this new SIS electrolyte. 100 cycles and (iii) high coulombic efficiency: nearly 100% Currently, the electrochemical performance of Li–S batteries is (Supplementary Table S2). Additionally, it also shows excellent directly determined by the sulphur loading in C/S composites, the rate performance and works well in a wide temperature range even adsorbability of sulphur by mesoporous carbon, the solubility of to  20 C (Supplementary Fig. S8). lithium polysulphide in the electrolyte and the stability of the metallic lithium anode during cycling. Actually, to date, it has seemed almost impossible to simultaneously solve all these Discussion problems via a single solution. To improve cyclic performances, All those significant improvements derive from two reasons. First, one strategy is to decrease sulphur loading of the C/S composite the dissolution of lithium polysulphide is effectively reduced, thus (30–50wt.%), which not only increases electronic conductivity of avoiding lithium polysulphide shuttle phenomenon. In Li–S C/S composites, but also enhances the absorption effect for sulphur batteries, the soluble intermediate is mainly Li S (8rnr4) 2 n and polysulphide, finally improving reversible storage capacity, produced during initial discharge and final charge stages. It can cycling life and coulombic efficiency. However, according to our be considered to be a lithium salt, and its solubility has a certain estimate of practical energy density shown in Supplementary Table saturation degree for a given solvent. Thus, for the electrolyte S1, it can be seen that the sulphur loading should be higher than with ultrahigh lithium salt concentration such as SIS-7#, the 50wt.%; otherwise, it is difficult to achieve higher gravimetric concentration is almost close to saturation, in which case the energy density than that of advanced lithium-ion batteries. The soluble intermediate (Li S ) becomes hardly soluble, as further 2 n other strategy is to add strong absorption materials such as proven in Fig. 4. The colour of 7# (SIS-7#) is nearly unchanged nanosized mesoporous SiO (ref. 13) or Mg Ni O (ref. 42), after standing for 18 days. In contrast, all others show colour 2 0.6 0.4 Al O (ref. 43) into a sulphur electrode, which increases the change from dark-brown to yellow, which is consistent with the 2 3 coulombic efficiency to some extent; however, the dissolution of ultraviolet-visible (UV-Vis) spectra (Fig. 4b). After equal lithium polysulphide is not completely inhibited. proportional dilution, those three samples (2#, 4# and 7#) are The third strategy is to use LiNO as an effective additive to obviously different in UV-Vis light absorption. Among them, the 0 0 stabilize the metallic lithium anode, via an in situ formation of a curves of 7#, 4# and 7# are nearly superimposed in the whole 44,45 protective layer on the lithium anode surface . However, the region, which indicates that the dissolution of lithium poly- protective layer can only prevent further reaction between the sulphide is indeed inhibited in the SIS-7# electrolyte. During the lithium polysulphide and the lithium anode, it is unable to inhibit review process, we noticed that a very recent work simply showed the dissolution of lithium polysulphide into the electrolyte, which that a higher salt concentration electrolyte (but only up to 5 M) is 16,46 results in instability in terms of long cycling . Finally protecting helpful to decrease both the dissolution of lithium polysulphide the metallic lithium anode using inorganic solid electrolyte layer and the diffusion coefficient of bulky polysulphide in the prevents any soluble polysulphide reaching the lithium metal, but electrolyte, thus improving the coulombic efficiency . Actually, 4 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. Coulombic efficiency (%) Voltage (V) –1 –1 Capacity (mA h g ) - sulphur Capacity (mA h g ) - sulphur NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE Visible light region (400–760nm) 18 Days 2# 4# 7# 4#’ 7#′ –1 1 Day –2 150 300 450 600 750 900 1,050 Wavelength (nm) 1 h –1 –1 –1 –1 0 mol l 2 mol l 4 mol l 7 mol l Figure 4 | Lithium polysulphide dissolution experiments. (a) The colour changes of four samples with different salt concentrations containing the same amount of Li S were recorded by digital camera along with time: 0 mol per l solvent, 2#: 2 mol per l solvent, 4#: 4 mol per l solvent, 7#: 7 mol per l solvent. 2 8 (b) The ultraviolet-visible spectrophotometry. The 2#, 4# and 7# for UV-Vis measurement are the solutions from the standing samples for 18 days, which are then diluted with the corresponding electrolytes in a volume ratio of 1:7. The curves of 4#’ and 7#’ are the pure electrolytes of 4 mol per l solvent and 7 mol per l solvent for comparison, respectively. the dissolution of lithium polysulphide is not totally prohibited hand, high viscosity limits anion convection near deposition area, because in this case the highest concentration used was 5 M which is also helpful to deposit uniformly . LiTFSI in DME/DOL. Meanwhile, lithium cycling efficiencies in different electrolyte Second, a metallic lithium anode is more stable in the SIS-7# systems were investigated by means of a Li deposition-disso- lution experiment according to Aurbach et al. (2#: Fig. 5e, 4#: electrolyte than in other electrolytes. Compared with other kinds of rechargeable metallic lithium batteries, it is more complicated Fig. 5f and SIS-7#: Fig. 5g). In the commonly used low-salt to stabilize a metallic lithium anode in a Li–S battery owing to concentration electrolyte system, the lithium cycling efficiency is double damage from dendrite formation and the side reaction estimated to be below 50%, which means a large excess of lithium between lithium polysulphide and metallic lithium during as an anode needed in a real battery, thus decreasing the energy cycling. From the scanning electron microscopy (SEM) images density and increasing the cost. In contrast, interestingly, a shown in Fig. 5a–d, it is obvious that SIS-7# shows the lowest lithium cycling efficiency as high as 71.4% is obtained in the roughness and damage level of metallic lithium anode compared SIS-7# electrolyte system, which is much higher than that of with the other three samples (2#: Fig. 5b, 4#: Fig. 5c and SIS-7#: other LiTFSI-based non-aqueous electrolytes .Furthermore, from Fig. 5d), which demonstrates that the SIS electrolyte system can SEM images (Supplementary Fig. S9), a smoother and more effectively reduce the corrosion and suppress the formation of uniform lithium deposition was also clearly observed in the SIS-7# lithium dendrites owing to the ultrahigh lithium salt concentra- electrolyte system. This important issue for real applications has tion and high viscosity. It is generally accepted that dendrites start seldom been discussed in previous work on Li–S batteries. to grow in the non-aqueous liquid electrolyte when the anion is The different lithium cycling efficiencies in different electrolyte depleted in the vicinity of the electrode where plating occurs systems could be ascribed to the nature of solid electrolyte 50,51 according to Chazalviel model . In the case of SIS electrolyte interphase (SEI) formed on the surface of lithium electrodes. To with ultrahigh salt concentration, there is a mass of anion to keep confirm our conjecture, X-ray photoelectron spectroscopy the balance of cation (Li ) and anion (TFSI ) near metallic analysis of Li electrode surfaces was performed. Based on the 53–57 lithium anode, and the space charge that is created by TFSI previous work from Aurbach group and our experimental depletion is minimal, thus this is not a favourable condition for data, it could be concluded that the solvents of DME and DOL in dendrite growth. Furthermore, owing to both ultrahigh lithium electrolytes could possibly be reduced to ROLi species and salt concentration and high lithium-ion transference number oligomers with  OLi end group, and the salt of LiTFSI could (0.73), SIS electrolyte provides a large amount of available possibly be reduced to Li N, LiF, C F Li and sulphur-containing 3 2 x y lithium-ion flux and raises the lithium ionic mass transfer rate compounds such as Li S, Li S O ,Li SO or SO CF during the 2 2 2 4 2 3 2 x between electrolyte and metallic lithium electrode, thereby cycling process to form an SEI layer on the metallic lithium enhancing the uniformity of lithium deposition and dissolution surface (Supplementary Fig. S10). The composition of SEI for in charge/discharge process. Besides, influenced by high viscosity, both 2# and SIS-7# electrolytes seems to be similar; however, the on the one hand, it possibly increases the pressure from the thickness is different. In the case of SIS-7# electrolyte, it can be electrolyte to push back growing dendrites, resulting in a more seen that the lithium metal signal appears after 150 s sputtering uniform deposition on the surface of the anode. On the other with Ar ions, which etches the surface layer-by-layer from a NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 5 & 2013 Macmillan Publishers Limited. All rights reserved. Absorption ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 3.0 2# 2.5 2.0 1.5 Overpotential dissolution 1.0 0.5 0.0 2# 0.06 0.03 0.00 –0.03 –0.06 8:11:14 16:31:40 Time (h) 3.0 2.5 2.0 4# 1.5 Overpotential dissolution 1.0 0.5 0.0 4# 0.06 0.03 0.00 –0.03 –0.06 8:10:11 16:30:40 24:42:39 Time (h) 3.0 SIS-7# 2.5 2.0 1.5 1.0 0.5 0.0 SIS-7# 0.06 0.03 0.00 –0.03 –0.06 24:52:41 49:58:41 74:22:14 Time (h) Figure 5 | Typical scanning electron microscopy images of metallic lithium anodes and lithium deposition and dissolution experiments. (a) Fresh lithium metal, (b) Lithium metal with 2# electrolyte after 278 cycles. (c) Lithium metal with 4# electrolyte after 183 cycles. (d) Lithium metal with SIS-7# electrolyte after 280 cycles. The white scale bar represents 60mm for all the images; lithium deposition and dissolution experiments by electrochemically 2  2 depositing Li on Cu foil (Current: 0.1 mA cm ; the first deposition: 5 C cm ; cycling process: 10% depth of discharge and 20 cycles; final dissolution: charged to 3 V). (e) 2# electrolyte. (f) 4# electrolyte. (g) SIS-7# electrolyte. The lithium cycling efficiency is calculated according to Aurbach et al. lithium electrode. However, the signal still does not show up after DME:DOL ¼ 1:1 by volume (Ferro, Suzhou, China) were mixed by ratio of mole number to volume, which are 1#: 1 mol per l, 2#: 2 mol per l, 3#: 3 mol per l, 4#: 800 s sputtering for the lithium electrode cycled in the low-salt 4 mol per l, SIS-5#: 5 mol per l, SIS-6#: 6 mol per l and SIS-7#: 7 mol per l, concentration electrolyte (Fig. 6). This indicates that the thickness respectively. The mixtures were stirred for 24 h at room temperature. All the is different in these two electrolyte systems. experiments were performed in an argon-filled glove box. In summary, we propose a new class of SIS electrolyte for the The carbon/sulphur composite is prepared by a melt-diffusion method. According to Ji et al. , CMK-3 (Nanjing XF-nano, China) or Ketjenblack and next-generation high-energy rechargeable metallic lithium bat- sulphur were ground together, sealed in a glass tube containing argon, and heated teries and take the electrolyte system of LiTFSI and ether solvents at 155 C for 24 h. (see Supplementary Figs S11–S13 and Supplementary Table S3). as an example to demonstrate the power of SIS electrolyte. For the appealing application in Li–S batteries, it is demonstrated that SIS Characterizations. The conductivity of electrolyte system was measured with an electrolytes can not only inhibit the dissolution of lithium impedance analyser (Zahner Zennium, Germany) over a temperature range of polysulphide but also effectively protect a metallic lithium anode 20 to 60 C in a thermostated container. The lithium-ion transference numbers against the formation of lithium dendrites, which makes the cell were obtained by combining alternating-current (AC) impedance and direct-cur- exhibit both excellent electrochemical performance and high rent (DC) polarization measurements using a symmetric Li/electrolyte/Li cell. First, AC impedance test was performed to obtain a total resistance R . Then DC safety. As a matter of fact, we anticipate further conceptual design cell polarization was carried out to obtain a stable current I . The lithium-ion DC of other SIS electrolyte systems by selecting appropriate salt and transference number was calculated by the formulas (t ¼ R /R and R ¼ Li þ cell DC DC solvent components and adjusting these proportions according to V /I ). DSC measurements were performed with a NETZSCH DSC 200F3 by DC DC different requirements for different rechargeable metallic lithium sealing B10 mg of the sample in an aluminium pan. The pan and the electrolyte were first cooled to about  150 C with liquid nitrogen and then heated to 50 C batteries (for example Li-O batteries). It is expected that this at a heating rate of 10 Cmin under nitrogen flow. The viscosity (Z)ofthe class of SIS electrolyte would offer a new and important approach electrolytes was measured on a programmable viscometer (Brookfield, DV-iii þ)at for improving the electrochemical and safety performance for 25 C in a homemade dry chamber (H O o10 p.p.m.), and the temperature was metallic lithium batteries and further make us reconsider them accurately controlled to within 0.1 C using a Brookfield TC-502 oil bath. A SEM microscopy (Hitachi S-4800) equipped with a vacuum transfer box was without their past safety concerns. used to study the morphology of the samples. The samples of lithium metal anode were stored in vacuum transfer box in an argon-filled glove box and transferred Methods into SEM chamber without exposure to air. Synthesis. The SIS electrolytes are prepared as follows: lithium bis(tri- X-ray photoelectron spectroscopy analyses were performed in ESCALAB 250 fluoromethane sulphonyl) imide (LiN(SO CF ) , LiTFSI) (TCI, Japan) and purified using a monochromatized Al Ka source and equipped with an Ar ion sputtering 2 3 2 6 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. Current (mA) Voltage (V) Current (mA) Voltage (V) Current (mA) Voltage (V) NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE 3 k Experimental data Peak1-(58 ± 0.5 eV) Li-F Peak2-(55 ± 0.5 eV) Li-O 2 k Fit curve 1 k 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 Binding energy (eV) 0 s 30 s 90 s 150 s 500 s 800 s Li F S 0 30 90 150 500 800 Etching time (s) 3k Experimental data Peak1-(58 ± 0.5 eV) Li-F O Li Li Peak2-(55 ± 0.5 eV) Li-O 2k O Peak3-(53 ± 0.5 eV) Li Li Fit curve 1k 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 60 55 50 Binding energy (eV) 0 s 30 s 60 s 90 s 150 s 500 s 800 s Li 30 O 0 30 60 90 150 500 800 Etching time (s) Figure 6 | X-ray photoelectron spectroscopy (XPS) analysis of metallic lithium anodes. After 100 cycles at a current rate of 0.2C with 2# and SIS-7# electrolytes, (a) XPS Li spectra of metallic lithium electrode using 2# electrolyte before sputtering and after sputtering with different times. (b) Summary 1s of the atomic concentration of Li, C, O, S, N and F on the Li anode surface as a function of sputtering time using 2# electrolyte. (c) XPS Li spectra of 1s metallic lithium electrode using SIS-7# electrolyte before sputtering and after sputtering with different times. (d) Summary of the atomic concentration of Li, C, O, S, N and F on the Li anode surface as a function of sputtering time using SIS-7# electrolyte. gun (Thermo Fisher). Ar etching was conducted at an argon partial pressure of Electrochemistry. The electrode was fabricated by mixing C/S composite, carbon 10 Torr in the x–y scan mode at ion acceleration of 3 kV and ion beam current nanotube, poly(vinylidene difluoride) in weight ratio of 8:1:1 for cycle life mea- density of 1mAmm . surement and the other weight ratio of 7:2:1 for rate capability measurement. The The lithium polysulphide dissolution experiments were carried out in the slurry was cast on carbon-coated Al current collector and dried at 50 C in vacuum following: Li S and S with a mole ratio of 1:7 (49.5 and 224 mg) were mixed for 10 h. The coin cells CR2032 were assembled with the electrode, pure lithium foil and added in 2-ml pure solvent, 2 mol salt per 1-l solvent (2#), 4 mol salt per 1-l as counter electrode and a glass fibre separator in an argon-filled glove box. The solvent (4#) and 7 mol salt per 1-l solvent (SIS-7#), respectively. The colour discharge and charge measurements at room temperature and low-temperature changes of solutions with time were observed and recorded by digital camera. range of  20 to 0 C in a thermostated container were carried out on a Land The samples of 2#, 4# and SIS-7# after standing for 18 days were diluted BT2000 Battery Test System (Wuhan, China). with the corresponding electrolytes in the ratio of 1:7 for UV-Vis measurement, The average cycling efficiency of metallic lithium electrodes in various which was carried out on a Cary 5000 UV-Vis spectrophotometry (Varian, electrolytes (2#, 4#, SIS-7#) was performed by electrochemically depositing Li 0  2 America). The control samples are pure electrolytes without Li S (4# on Cu foil (5 C cm ) followed by Li stripping and deposition cycling (10% depth 2 8 0  2 and SIS-7# ). of discharge-DOD, 0.1 mA cm , 20 cycles). The residual Li was then dissolved X-ray powder diffraction analysis was characterized by X’Pert Pro MPD X-ray electrochemically (charged to 3 V) on a Land BT2000 Battery Test System. The diffractometer (Philips, Holland) using Cu–Ka radiation (1.5405 Å). average Li cycling efficiency was calculated according to Aurbach et al. NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 7 & 2013 Macmillan Publishers Limited. All rights reserved. Intensity/counts per second Atomic concentration (%) Intensity/counts per second Atomic concentration (%) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 References 32. Nagao, M., Hayashi, A. & Tatsumisago, M. Fabrication of favorable interface 1. Armand, M. & Tarascon, J. M. Building better batteries. Nature 451, 652–657 between sulfide solid electrolyte and Li metal electrode for bulk-type solid-state (2008). Li/S battery. Electrochem. Commun. 22, 177–180 (2012). 2. Goodenough, J. B. & Kim, Y. Challenges for rechargeable Li batteries. Chem. 33. Angell, C. A., Liu, C. & Sanchez, E. Rubbery solid electrolytes with Mater. 22, 587–603 (2010). dominant cationic transport and high ambient conductivity. Nature 362, 3. Scrosati, B., Hassoun, J. & Sun, Y. K. Lithium-ion batteries. A look into the 137–139 (1993). future. Energy Environ. Sci. 4, 3287–3295 (2011). 34. Hu, Y. S., Li, H., Huang, X. J. & Chen, L. Q. Novel room temperature molten 4. Choi, N. S. et al. Challenges facing lithium batteries and electrical double-layer salt electrolyte based on LiTFSI and acetamide for lithium batteries. capacitors. Angew. Chem. Int. Ed. 51, 9994–10024 (2012). Electrochem. Commun. 6, 28–32 (2004). 5. Zu, C. X. & Li, H. Thermodynamic analysis on energy densities of batteries. 35. Henderson, W. A. et al. Glyme-lithium bis(trifluoromethanesulfonyl)imide and Energy Environ. Sci. 4, 2614–2624 (2011). glyme-lithium bis(perfluoroethanesulfonyl)imide phase behavior and solvate 6. Ji, X. & Nazar, L. F. Advances in Li-S batteries. J. Mater. Chem. 20, 9821–9826 structures. Chem. Mater. 17, 2284–2289 (2005). (2010). 36. Tachikawa, N. et al. Reversibility of electrochemical reactions of sulphur 7. Bruce, P. G., Freunberger, S. A., Hardwick, L. J. & Tarascon, J.-M. Li-O and 2 supported on inverse opal carbon in glyme-Li salt molten complex electrolytes. Li-S batteries with high energy storage. Nat. Mater. 11, 19–29 (2012). Chem. Commun. 47, 8157–8159 (2011). 8. Evers, S. & Nazar, L. F. New approaches for high energy density 37. Yoshida, K. et al. Oxidative-stability enhancement and charge transport lithium-sulphur battery cathodes. Acc. Chem. Res. doi:10.1021/ar3001348 mechanism in glyme-lithium salt equimolar complexes. J. Am. Chem. Soc. 133, (2012). 13121–13129 (2011). 9. Manthiram, A., Fu, Y. Z. & Su, Y. S. Challenges and prospects of 38. Angell, C. A. Electrical conductance of concentrated aqueous solutions and molten salts: correlation through free volume transport model. J. Phys. Chem. lithium–sulphur batteries. Acc. Chem. Res. doi:10.1021/ar300179v (2012). 69, 2137 (1965). 10. Ji, X. L., Lee, K. T. & Nazar, L. F. A highly ordered nanostructured 39. Angell, C. A. A new class of molten slat mixtures the hydrated dipositive carbon-sulphur cathode for lithium-sulphur batteries. Nat. Mater. 8, 500–506 ion as an independent cation species. J. Electrochem. Soc. 112, 1224–1227 (2009). (1965). 11. Hassoun, J. & Scrosati, B. A high-performance polymer tin sulphur lithium ion 40. Kawamura, T., Kimura, A., Egashira, M., Okada, S. & Yamaki, J. I. Thermal battery. Angew. Chem. Int. Ed. 49, 2371–2374 (2010). stability of alkyl carbonate mixed-solvent electrolytes for lithium ion cells. 12. Elazari, R., Salitra, G., Garsuch, A., Panchenko, A. & Aurbach, D. J. Power Sources 104, 260–264 (2002). Sulfur-impregnated activated carbon fiber cloth as a binder-free cathode for 41. Mikhaylik, Y. V. & Akridge, J. R. Polysulfide shuttle study in the Li/S battery rechargeable Li-S batteries. Adv. Mater. 23, 5641–5644 (2011). system. J. Electrochem. Soc. 151, A1969–A1976 (2004). 13. Ji, X. L., Evers, S., Black, R. & Nazar, L. F. Stabilizing lithium-sulphur cathodes 42. Song, M. S. et al. Effects of nanosized adsorbing material on electrochemical using polysulphide reservoirs. Nat. Commun. 2, 235–241 (2011). properties of sulphur cathodes for Li/S secondary batteries. J. Electrochem. Soc. 14. Demir-Cakan, R. et al. Cathode composites for Li-S batteries via the 151, A791–A795 (2004). use of oxygenated porous architectures. J. Am. Chem. Soc. 133, 16154–16160 43. Choi, Y. J. et al. Electrochemical properties of sulphur electrode containing (2011). nano Al O for lithium/sulphur cell. Phys. Scr. 62, T129 (2007). 2 3 15. Wang, H. L. et al. Graphene-wrapped sulphur particles as a rechargeable 44. Mikhaylik, Y. A. Electrolytes for lithium sulphur cells. US Patent no. 0147891 lithium-sulphur battery cathode material with high capacity and cycling (2005). stability. Nano Lett. 11, 2644–2647 (2011). 45. Aurbach, D. et al. On the surface chemical aspects of very high energy 16. Zheng, G. Y., Yang, Y., Cha, J. J., Hong, S. S. & Cui, Y. Holow carbon nanofiber- density, rechargeable Li–sulphur batteries. J. Electrochem. Soc. 156, A694–A720 encapsulated sulphur cathodes for high specific capacity rechargeable lihtium (2009). batteries. Nano Lett. 11, 4462–4467 (2011). 46. Liang, X. et al. Improved cycling performances of lithium sulphur batteries with 17. Jayaprakash, N., Shen, J., Moganty, S. S., Corona, A. & Archer, A. Porous LiNO -modified electrolyte. J. Power Sources 196, 9839–9843 (2011). hollow carbon@sulphur composites for high-power lithium-sulphur batteriess. 47. Steven, J. V., Yevgeniy, S. N. & Bruce, D. K. Ionically conductive membranes Angew. Chem. Int. Ed. 50, 5904–5908 (2011). for protection of active metal anodes and battery cells. US Patent no. 0191617 18. Schuster, J. et al. Spherical ordered mesoporous carbon nanoparticles with high (2004). porosity for lithium-sulphurbatteries. Angew. Chem Int. Ed. 51, 3591–3595 48. Lutgard, D. J., Yevgeniy, S. N. & Steven, J. V. Alleviation of voltage delay in (2012). lithium-liquid depolarizer/electrolyte solvent battery cells. US Patent no. 674558 19. Schuster, J. & Nazar, L. F. Graphene-enveloped sulfur in a one pot reaction: a (2010). cathode with good coulombic efficiency and high practical sulfur content. 49. Shin, S. E., Kim, K., Oh, S. H. & Cho., W. I. Polysulfide dissolution control: the Chem. Commun. 48, 1233–1235 (2012). common ion effect. Chem. Commun. doi:10.1039/c2cc36986a (2012). 20. Su, Y. S. & Manthiram, A. Lithium-sulphur batteries with a microporous 50. Chazalviel, J. N. Electrochemical aspects of the generation of ramified metallic carbon paper as a bifunctional interlayer. Nat. Commun. 3, 1166 (2012). electrodeposits. Phys. Rev. A 42, 7355–7367 (1990). 21. Xia, L. F. et al. A soft approach to encapsulate sulphur: polyaniline nanotubes 51. Rosso, M. Electrodeposition from a binary electrolyte: new developments and for lithium-sulphur batteries with long cycle life. Adv. Mater. 24, 1176–1181 applications. Electrochim. Acta 53, 250–256 (2007). (2012). 52. Gonzalez, G., Marshall, G., Molina, F. V., Dengra, M. S. & Rosso, M. Viscosity 22. Yang, Y. et al. Improving the performance of lithium-sulphur batteries by effects in thin layer electrodeposition. J. Electrochem. Soc. 148, C479–C487 conductive polymer coating. ACS Nano 5, 9187–9193 (2011). (2001). 23. Xu, K. Nonaqueous liquid electrolytes for lithium-based rechargeable batteries. 53. Aurbach, D., Gofer, Y. & Langzam, J. The correlation between surface Chem. Rev 104, 4303–4417 (2004). chemistry, surface morphology, and cycling efficiency of lithium electrodes in a 24. Armand, M. B., Chabagno, J. M. & Duclot, M. J. In: Vashishta, P., few polar aprotic systems. J. Electrochem. Soc. 136, 3198–3205 (1989). Mundy, J.M. & Shenoy, G.K. (eds). Fast Ion Transport in Solids (Elsevier, 1979; 54. Aurbach, D. & Gront, E. The study of electrolyte solutions based on solvents pp 131–136). from the ‘glyme’ family (linear polyethers) for secondary Li battery systems. 25. Croce, F., Appetecchi, G. B., Persi, L. & Scrosati, B. Nanocomposite polymer Electrochim. Acta 42, 697–718 (1997). electrolytes for lithium batteries. Nature 394, 456–458 (1998). 55. Aurbach, D., Youngman, O., Gofer, Y. & Meitav, A. The electrochemical 26. Gadjourova, Z., Andreev, Y. G., Tunstall, D. P. & Bruce, P. G. Ionic behavior of 1,3-dioxolane-LiClO solutions. Electrochim. Acta 35, 625–638 conductivity in crystalline polymer electrolytes. Nature 412, 520–523 (2001). (1990). 27. Robertson, A. D., West, A. R. & Ritchie, A. G. Review of crystalline lithium-ion 56. Aurbach, D., Weissman, I. & Schechter, A. X-ray photoelectron conductors suitable for high temperature battery applications. Solid State Ionics spectroscoscopy studies of lithium surface prepared in several important 104, 1–11 (1997). electrolyte solution. A comparison with previous studies by fourier tansform 28. Kondo, S., Takada, K. & Yamamura, Y. New lithium ion conductors based on infrared spectroscopy. Langmuir 12, 3991–4007 (1996). Li S-SiS . Solid State Ionics 53, 1183–1186 (1992). 2 2 57. Schechter, A. & Aurbach, D. X-ray photoelectron spectroscoscopy study of 29. Kamaya, N. et al. A lithium superionic conductor. Nat. Mater. 10, 682–686 surface films formd on Li electrodes freshly prepared in alkyl carbonate (2011). solutions. Langmuir 15, 3334–3342 (1999). 30. Hayashi, A., Ohtomo, T., Mizuno, F., Tadanaga, K. & Tatsumisago, M. 58. Bhattacharyya, A. J. & Maier, J. Second phase effects on the conductivity of All-solid-state Li/S batteries with highly conductive glass-ceramic electrolytes. non-aqueous salt solutions: ‘Soggy sand electrolytes’. Adv. Mater. 16, 811–814 Electrochem. Commun. 5, 701–705 (2003). (2004). 31. Nagao, M., Hayashi, A. & Tatsumisago, M. Sulfide-carbon composite electrode 59. MacFarlane, D. R., Huang, J. H. & Forsyth, M. Lithium-doped plastic crystal for all-solid-state Li/S battery with Li S-P S solid electrolyte. Electrochim. Acta 2 2 5 electrolytes exhibiting fast ion conduction for secondary batteries. Nature 402, 56, 6055–6059 (2011). 792–794 (1999). 8 NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications & 2013 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms2513 ARTICLE L.M.S. and Y.-S.H. wrote the paper; L.M.S., Y.-S.H., H.L., M.A. and L.Q.C. discussed the 60. Liang, C. C. Conduction characteristics of lithium iodide aluminum oxide solid results and participated in the preparation of the paper. electrolytes. J. Electrochem. Soc. 120, 1289–1292 (1973). Additional information Acknowledgements Supplementary Information accompanies this paper at http://www.nature.com/ We thank Profs Xuejie Huang and Zhibin Zhou for fruitful discussions and help on DSC naturecommunications and viscosity measurements. This work was supported by funding from the ‘863’ Project (2009AA033101), ‘973’ Projects (2009CB220104), NSFC (50972164, 51222210), CAS Competing financial interests: The authors declare no competing financial interests. project (KJCX2-YW-W26) and One Hundred Talent Project of the Chinese Academy of Sciences. Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ Author contributions Y.-S.H. conceived and designed this work; L.M.S. performed the experiment; L.M.S., How to cite this article: Suo, L. et al. A new class of Solvent-in-Salt electrolyte Y.-S.H., H.L. and L.Q.C. analysed the physicochemical properties of this new electrolyte; for high-energy rechargeable metallic lithium batteries. Nat. Commun. 4:1481 Y.-S.H. and M.A. analysed the inhibition mechanism of lithium dendrite formation; doi: 10.1038/ncomms2513 (2013). NATURE COMMUNICATIONS | 4:1481 | DOI: 10.1038/ncomms2513 | www.nature.com/naturecommunications 9 & 2013 Macmillan Publishers Limited. All rights reserved.

Journal

Nature Communications – Springer Journals

Published: Feb 12, 2013

There are no references for this article.