Abstract:Methyl phenyl bis-methoxydiethoxysilane (MPBMDS) was prepared and its effects were investigated as an additive in 1.0 mol dm −3 LiPF 6 -propylene carbonate (PC)/dimethyl carbonate (DMC) (1:1, v/v) electrolyte for lithium ion batteries. The electrochemical properties of the electrolyte with MPBMDS were characterized by discharge/charge tests, cyclic voltammetry, electrochemical impedance spectroscopy, scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS). The addition of MPBMDS can effectively prevent the decomposition and the co-intercalation of PC. In addition, burning tests showed that the addition of 4–13 wt.% MPBMDS to the bare PC-based electrolyte effectively reduces the flammability. This eco-friendly compound provides a new promising direction for the development of bi- or multi-functional additives for lithium ion batteries. Keywords Silane Lithium ion battery Film-forming Flame retardant Electrolyte additive 1 Introduction Lithium ion batteries have experienced growing acceptance in the consumer market since they were commercialized by Sony in the early 1990s, because of their favorable properties such as their high voltage, high energy density, and low self-discharge [1] . Their application in developing fields such as the space industry, polar research and electric vehicles results in a higher demand for low- and high-temperature performance lithium ion batteries. Use of these batteries, in conjunction with fuel cell and supercapacitor technology can contribute significantly to reduction of CO 2 emissions and protecting the environment. However, safety concerns are still challenging when scaled-up lithium-ion batteries are considered, especially at the overcharged state of the batteries. This is because the strongly oxidized cathodes may take part in a series of hazardous chemical reactions, leading to thermal runaway, fire or explosion [2–8] . The solvent of the electrolyte used in lithium ion batteries is a crucial factor influencing the low-temperature performance. Ethylene carbonate (EC)-based electrolytes are extensively used in current commercial lithium ion batteries, but they can only work well at above −20 °C because of the high melting point of EC (about 36 °C) [9,10] . In contrast, with such excellent properties as low melting point (−49 °C), high flash point (132 °C), high dielectric constant and wide electrochemical window, propylene carbonate (PC) is an attractive candidate as a solvent for nonaqueous electrolytes in lithium ion batteries at low temperatures and is also helpful in improving the safety of lithium ion batteries [11,12] . However, PC has been shown to have the tendency to co-intercalate together with lithium cations into the crystal structure of graphite, followed by its decomposition [12–14] . To improve safety and reliability, self-actuating overcharge protection, such as positive temperature coefficient resistors and integrated circuits, has been introduced. However, these methods are not completely reliable. The electropolymerization or coating of monomer additives on electrode surfaces is another possible protection mechanism for lithium-ion batteries [15–21] . In the case of PC, several ways have been explored to modify the surface of the graphite by means such as coating to prevent the direct contact of PC with the graphitic structure [22,23] . Several anode materials are currently pursued, such as titania-C [24] , silicon-C [25] , silicon-silica [26] , SnO 2 -CNT [27] , and MoO 2 [28] nanocomposites, but graphite is still the first choice for use in commercial batteries. However, graphite suffers extensive surface structural disordering upon prolonged cycling, especially at high charging rates and elevated temperatures [29] . When certain substances are added to PC-based electrolytes [30,31] , an effective solid electrolyte interface (SEI) film is formed prior to the decomposition of PC and intercalation of Li + into the graphite so that Li + can reversibly intercalate and de-intercalate [32–36] . For example, substances such as allyl ethyl carbonate [37] , acrylonitrile [38] , tetrachloroethylene [39] , ethylene sulfites [40] , vinylene carbonate (VC) [41] , propylene sulfite [10] , chloroethylene carbonate [42] , silanes [10,43] , triethyl orthoformate [44] , 2-phenylimidazole (PID) [45] and N -phenylmaleimide (NPM) [46] have been used. Silanes seem to behave similarly to crown ethers and cryptands as far as formation of complexes with metal cations is concerned [47–49] . Apparently, a unique component of the protective effect of polyoxasilanes, aside from electropolymerization, is condensation (polymerization) of podand molecules via silicon atoms, leading to the formation of a protective network on the graphite surface. In our own experience, phenyl tris-2-methoxydiethoxysilane (PTMS) suppresses the co-intercalation of PC by formation of a cross-polymerized network of Si–O–Si bonds on the graphite surface, thus restraining PC molecules from co-intercalating with lithium ions into graphite electrodes [30] . A specific capacity above 275 mAh g −1 is maintained using CMS electrode in 1 mol dm −3 LiPF 6 /PC:DMC (1:1, v/v) electrolyte. In addition, silanes can act as fire-retardant to improve the safety performance of lithium-ion batteries. When we added 3 wt.% PID to 1 mol dm −3 LiPF 6 /PC:DMC (1:1, v/v) electrolyte the co-intercalation of PC was suppressed and the decomposition of electrolyte during the first lithium intercalation was inhibited by formation of a favorable solid electrolyte interface (SEI) on the CMS graphite electrode, reaching a reversible capacity of 300 mAh g −1 [45] . NPM can electrochemically polymerize at the overcharge potential of 3.8–4.2 V (vs. Li/Li + ) and form a thin polymer film on the surface of the cathode, thus preventing voltage runaway. On the other hand, the use of 3–5% NPM as an overcharge protection electrolyte additive does not influence the normal performance of lithium-ion batteries, providing a specific capacity above 70 mAh g −1 for the LiFePO 4 electrode over many cycles [46] . One of the problems hindering the commercial application of large capacity lithium-ion batteries needed in the developing hybrid electric vehicle market is indeed the safety concerns arising from the use of volatile and flammable liquid electrolytes [50] . One promising strategy to address this safety issue is to employ solid electrolytes such as gelled polymer electrolytes as alternatives to the currently used organic carbonate electrolytes [51] . However, the improved flame retardation of these electrolytes is accompanied by reduced battery performance including significantly lower conductivity, especially at low temperature [52] . An alternative and popular approach is to add additives as flame retardants to increase the thermal stability of the electrodes and decrease the flammability of the electrolytes. In that case, the low temperature performance of lithium ion batteries can be maintained [53,54] . Burning tests showed that the addition of up to 10 vol% vinyl-tris-(methoxydiethoxy)silane (VTMS) to the currently used electrolyte could effectively reduce the flammability. As long as the added amount was kept below 10%, electrochemical performance such as reversible capacity and cycling showed little change [51] . In this paper, methyl phenyl bis-methoxydiethoxysilane (MPBMDS), a new film-forming additive, is examined for non-aqueous electrolytes-based lithium-ion batteries with graphite anodes. The performance of graphitic anodes was remarkably improved in PC-based electrolytes after adding MPBMDS. At the same time, MPBMDS is an environmental friendly flame retardant with high thermal stability and low viscosity. Here we report the performance of MPBMDS as a bi-functional additive: preventing the exfoliation of graphite for PC-based electrolyte and acting as a flame retardant. 2 Experimental Methyl phenyl bis-methoxydiethoxysilane (MPBMDS) was prepared by reaction of methyl phenyl dichlorosilane with anhydrous diethylene glycol monomethyl ether, according to the following chemical equation: Methyl phenyl dichlorosilane (15.02 g of 98% purity, 0.077 mol, 12.79 ml) in 15 ml anhydrous dichloromethane was added dropwise (under argon) over 4 h to a solution of anhydrous diethylene glycol monomethyl ether (18.44 g of 99% purity, 0.152 mol) in 15 ml anhydrous dichloromethane cooled to 0 °C, with continuous stirring. After warming to room temperature, the solvent was removed by evaporation and the mixture was heated at 90 °C under vacuum (15–20 mmHg) for about 6 h to remove gaseous HCl. The obtained product was purified by vacuum distillation (240 °C at 1 mmHg to remove the reactants) and the identity of the obtained compound was verified by FTIR and NMR spectroscopy. IR (neat): 3450, 3030, 2878, 1593, 1456, 1354, 1199–1108, 1083, 967 cm −1 . 1 H NMR (CDCl 3 , 300 MHz): δ = 0.23 (3H, s, Si– Me ), 3.19 (6H, s, O– Me ), 3.3–3.6 (16H, m, 8× – CH 2 –), 7.20 (3H, m, arom. ), 7.47 (2H, m, arom. ). The bare PC-based electrolyte, 1.0 mol dm −3 LiPF 6 -PC/DMC (1:1, v/v), was supplied by Guotai Huarong Co. Ltd., China, and used as the electrolyte without further treatment. Due to strong Van der Waals forces, the viscosity of pure PC is high. DMC was therefore added to reduce the viscosity of the electrolyte so that Li + can move faster. Different amounts of MPBMDS were added to the above bare electrolyte in an argon-filled glove box (water content less than 10 ppm). The bare EC-based electrolyte, 1.0 mol dm −3 LiPF 6 -EC/EMC/DMC (1:1:1, v/v/v) (Guotai Huarong Co. Ltd., China) was used as the control. Model cells were fabricated in an argon-filled glove box to investigate the effects of the additive on the electrochemical performance of graphite anodes. The anode was prepared by coating a mixture of CMS (an artificial spherical graphite, C content ≥99.0%, Shanshan Co. Ltd., Shanghai, China), PVDF (C.P.), and carbon black (with weight ratio = 80:10:10) on a piece of copper foil. After drying under ambient conditions, round discs of 1 cm diameter and 170 μm thickness were cut off and further dried under vacuum at 120 °C for 12 h. The amount of active material was about 5 mg. Lithium foil was employed as the counter and reference electrode, and Celgard 2400 as the separator. Cyclic voltammetry (CV) curves were recorded on a CHI604C Electrochemical Work Station. The scan rate was 0.3 mV/s in the potential range 1.8–0 V. Cycling performance was tested on a Land battery tester (Land Co. Ltd., China) at 90 mA g −1 in the potential range 0.01–2 V vs. Li/Li + . The AC impedance measurements were taken by means of the CHI604C Electrochemical Work Station. Scanning frequencies ranged from 100 kHz to 0.01 Hz, perturbation amplitude 5 mV. AC impedance was observed at open circuit voltage conditions with a coin type cell. A CMS electrode served as working electrode. Lithium metal foil was used as the counter and reference electrodes. Scanning electron microscopy (SEM) (Philips XL 300) was used to inspect the surface morphology of the electrodes. In addition, X-ray photoelectron spectroscopy (XPS) (Perkin Elmer PHI-5000C) was done using Mg K α monochromatic radiation. All the binding energies were calibrated using the C 1s peak at 284.6 eV for graphite as a reference. After five discharge–charge cycles, the model cell was disassembled in the argon-filled glove box. The test electrode was soaked and rinsed with propylene carbonate (PC) for 4 h, and then dried under vacuum in the glove box at room temperature to remove low molecular weight compounds prior to the SEM and XPS measurements. Flammability was measured with reference to the UL-94 vertical burning test [57,58] , executed with a CZF-3 instrument (Jiangning Analysis Instrument Factory) with a sample volume of 1.0 ml and a neutral wick of 80 mm length. 3 Results and discussion Fig. 1 shows CVs of the CMS electrode in 1 mol dm −3 LiPF 6 solution of PC/DMC (1:1, v/v) with and without 4 wt.% MPBMDS. For the bare electrolyte, a very large irreversible cathodic peak is observed near 0.5 V in the first cycle, which can be ascribed to the decomposition of PC and the exfoliation of graphite [9,10,59] . In the second scan, the current response of lithium ion intercalation into and de-intercalation from the CMS electrode becomes very weak, indicating that lithium ions could not effectively intercalate into and de-intercalate from the graphite in the case of bare PC-based electrolyte. However, for the electrolyte with MPBMDS, the CV in the first scan shows that the reduction peak at about 0.5 V almost disappeared and a new peak appeared at 1.13 V, which can be ascribed largely to the decomposition of MPBMDS. In the second scan, only one pair of reversible peaks at 0.2 V is visible, indicating that the formed film on the CMS electrode in the first cycle is efficient and enables reversible lithium intercalation/de-intercalation in the PC-based electrolyte. Fig. 2 (a) shows the first discharge–charge curves of the CMS electrode in 1.0 mol dm −3 LiPF 6 -PC/DMC (1:1, v/v) without MPBMDS and with 4 wt.% MPBMDS. In the first case there is a long plateau at about 0.70 V. It is well known that this long plateau is due to the co-intercalation of the Li + ions solvated by PC and PC decomposition [55,56] . There is no intercalation of lithium in the graphitic carbon because of the presence of sufficient PC. If the amount of PC-based electrolyte is small, a small capacity for graphite can be observed. However, a dramatic change is observed when 4 wt.% MPBMDS is added to the electrolyte. The plateau at about 0.70 V corresponding to the PC decomposition disappears almost completely. The typical plateaus for the lithium intercalation/de-intercalation are observed below 0.2 V, with a reversible capacity of 281 mAh g −1 and coulombic efficiency of 70%. This clearly supports the notion that the addition of MPBMDS effectively forms a surface film prior to the co-intercalation of solvated Li + ions and thus ensures the reversible intercalation and de-intercalation of lithium into/from the graphitic carbon. Due to the high viscosity of PC, the reversible capacity is a little lower than that in EC-based electrolyte. Fig. 2 (b) shows the cycling behavior at 90 mAh g −1 of the CMS anode in the PC-based electrolyte containing different amounts of MPBMDS. The cycling behavior of the CMS in EC-based electrolyte was used as a control. In the case of the bare EC-based electrolyte there is little capacity fading after 30 cycles, which illustrates the well-known fact that artificial graphite has good stability and cycling ability in EC-based electrolyte. With the PC-based electrolyte containing 4 wt.% MPBMDS, almost the same cycling performance as that of the bare EC electrolyte was achieved. This phenomenon indicates that the surface film formed after the addition of MPBMDS is very stable. However, when the amount of MPBMDS (1 wt.%) is too low, the formed surface film is ineffective and the CMS suffers from obvious capacity fading. On the other hand, when the amount is increased to 7 wt.%, the surface film apparently is too thick and the capacity decreases. Although the cycling behavior is not comparable with that in EC-based electrolyte, it is still very satisfactory. The decreasing capacity with increasing amount of additive may be due to the possibility that the surface film continues to grow and so leads to larger internal resistance. The impedance spectrum of CMS electrodes after 5 cycles in PC-based electrolyte with different amounts of MPBMDS added is presented in Fig. 3 . Generally, the high frequency semicircle region is attributed to SEI film and/or contact resistance and the Li + charge-transfer impedance on electrode/electrolyte interface. The inclined line at an approximate 45° angle to the real axis corresponds to the lithium-diffusion processes within electrodes [67] . A well-developed semicircle, which corresponds to the interfacial impedance of the surface film formed on the CMS electrode, is observed in Fig. 3 . It can be clearly seen that the diameters of the semicircles for the cycled electrode with MPBMDS are smaller than those of the bare electrode. This indicates that the electrodes with MPBMDS have lower interfacial impedances than those with no additives. In Fig. 3 , CMS electrode with 4 wt.% MPBMDS has the smallest semicircle. The diameter increases as the amount of MPBMDS increased. It is speculated that excessive additive can influence the ion migration process by enlarging interfacial impedance. As an exception, 1 wt.% MPBMDS exhibits larger impedance than 4 wt.%, which may be attributed to an incomplete SEI film network because of insufficient MPBMDS. The plot demonstrates the positive impact of MPBMDS on the electrode impedance behavior, illustrating faster Li-ion migration through the surface SEI film. To gain information on the surface morphology of the CMS electrode, SEMs of the CMS electrode before and after cycling are shown in Fig. 4 . It can be clearly seen that the original CMS has a spherical shape and smooth surface ( Fig. 4 a). However, the surface of the CMS spheres after cycling in 1 mol dm −3 LiPF 6 -PC/DMC (1:1, v/v) electrolyte containing 4 wt.% MPBMDS is uniformly covered by a layer of surface film ( Fig. 4 b). This is consistent with the above CV and electrochemical tests. On the other hand, the graphite layers of CMS electrode in bare PC-based electrolyte are severely exfoliated because of the co-intercalation of PC, which can be seen in Fig. 4 c. There has been much speculation about the exfoliation of graphite in PC-based electrolyte [1,8,11] ; Fig. 4 c provides clear evidence of exfoliation directly from SEM. The comparison also shows that the exfoliation of graphite in PC-based electrolyte is completely suppressed by the addition of MPBMDS. The components of the SEI film were further investigated by XPS, and the results are shown in Fig. 5 . The concentration of Si element is as high as 1.80%, which clearly shows that MPBMDS was reduced and became a part of the SEI film. There is an obvious broad peak in the range of the binding energy at 280–294 eV corresponding to carbon atoms. In the C 1s spectra, the binding energies at 287.8 and 290.6 eV correspond to the unsaturated and saturated carbon atoms in the RCH 2 OCOOLi compounds from the decomposition of the electrolyte, respectively. The binding energy at 290.4 eV corresponds to the C atoms existing in the Li 2 CO 3 . Moreover, the binding energy at 284.6 eV corresponds to the C atoms in CMS (sp 2 ). In the F 1s spectra, the main peak at 685 eV is assigned to LiF, which is the main component of the SEI layer [69,70] . For the P 2p spectra, the peak at 136 eV is attributed to LiPF 6 or its decomposition products [31,36,68] . In the case of Si, it shows that silicon on the surface of CMS after cycling in the electrolyte containing MPBMDS exists in the forms of Si–O–Si, Si–O x and C–Si–O, corresponding to binding energy peaks of 102.0, 102.5 and 103.5 eV, respectively [10,51,60] . From the O 1s spectra ( Fig. 5 b), the binding energy peak at 533.5 eV can be attributed unambiguously to the Si–O–Si bonding. Furthermore, the binding energy peaks at 532.8 and 531.4 eV correspond to C–O–C and Si–O–C, respectively. On the basis of the above results, it is clear that MPBMDS forms a stable network of Si–O–Si bonds on the graphite surface during the reduction process in the first cycle, which probably provides channels of some kind for the movement of Li + ions. Because of steric effects of the additive [61–65] , not all molecules are individually linked to the surface, some molecules combining with the graphite surface in the form of C–O–Si, Si–C and Si–O covalent bonds. The surface network is a part of the surface film, and can effectively prevent PC molecules from co-intercalating into the graphite electrodes and the subsequent exfoliation of graphite. Fig. 6 shows the flame propagation and burning-up time profiles of the electrolyte with different amounts of MPBMDS. Apparently, the bare electrolyte is highly flammable with short flame propagation and burning-up times, which are 26 and 55 s, respectively. With increasing content of MPBMDS, both the flame propagation time and burning-up time increase. When the content of MPBMDS reaches 13 wt.%, the flame propagation time of the electrolyte is 59 s, more than double that of the bare electrolyte, indicating that MPBMDS is effective in inhibiting the flame propagation. The burning-up time also increases with the MPBMDS content, which clearly shows that MPBMDS is a good flame retardant for the electrolytes of lithium ion batteries at a loading level of 4–13 wt.%. In addition, it is consistent with our former report on a silane acting as effective flame retardant [51] . 4 Conclusion Further development of rechargeable lithium-ion batteries remains a challenge [66] , especially with respect to safety, which requires a nonflammable electrolyte and/or a flame retardant additive(s) that can also develop a SEI layer to inhibit exfoliation. Methyl phenyl bis-methoxydiethoxysilane (MPBMDS) is the next development in the search for such a bi-functional additive for lithium-ion batteries. The added MPBMDS can greatly suppress the decomposition and the co-intercalation of PC and efficiently form a stable SEI film on the surface of the CMS electrode, as shown by the CV, XPS, and SEM images of the electrode after cycling, with little capacity loss. Addition of 4% MPBMDS has little effect on cycling behavior, with electrodes exhibiting reversible capacities above 280 mAh g −1 without fading after 30 cycles. MPBMDS is also an effective flame retardant at low concentrations, increasing both the flame propagation time and the burn-up time. Combining good cycling behavior with flame retardant functionality opens a new direction for the development of bi- or multi-functional additives for lithium ion batteries, which is an attractive prospect for future development. We believe that there is still scope for fine-tuning the effects by varying the substituents on silicon and the polyether chain length. Acknowledgements Financial support from the MOST Program (No. 2007CB209702 and 2010DFA61770), STCSM (09QH1400400) and the South African National Research Foundation (UID 67217) is greatly appreciated. References [1] H. Ota K. Shima M. Ue J. Yamaki Electrochim. Acta 49 2004 565 [2] D.D. Macneil J.R. Dahn J. Electrochem. Soc. 149 2002 A912 [3] K. Kumai H. Miyashiro Y. Kobayashi K. Takei R. Ishikawa J. Power Sources 81–82 1999 715 [4] J.R. Dahn E.W. Fuller M. Obrovac U. von Sacken Solid State Ionics 69 1994 265 [5] Z.H. Chen K. Amine Electrochem. Commun. 9 2007 703 [6] P. Biensan B. Simon J.P. Peres A. de Guibert M. Broussely J.M. Bodet F. Perton J. Power Sources 81–82 1999 906 [7] L.M. Moshuchak M. Bulinski W.M. Lamanna R.L. Wang J.R. Dahn Electrochem. Commun. 9 2007 1497 [8] L.F. Xiao X.P. Ai Y.L. Cao Y.D. Wang H.X. Yang Electrochem. Commun. 7 2005 589 [9] B. Wang H.P. Zhang L.C. Yang Q.T. Qu Y.P. Wu C.L. Gan D.L. Zhou Electrochem. Commun. 10 2008 1571 [10] G.H. Wrodnigg T.M. Wrodnigg J.O. Besenhard M. Winter Electrochem. Commun. 1 1999 148 [11] R. Chen F. Wu L. Li Y.B. Guan X.P. Qiu S. Chen Y.J. Li S.X. Wu J. Power Sources 172 2007 395 [12] C. Korepp H.J. Santner T. Fujii M. Ue J.O. Besenhard K.-C. Möller M. Winter J. Power Sources 158 2006 578 [13] M. Walkowiak D. Waszak G. Schroeder B. Gierczyk Electrochem. Commun. 10 2008 1676 [14] G. Schroeder B. Gierczyk D. Waszak M. Walkowiak Electrochem. Commun. 8 2006 1583 [15] L.F. Xiao X.P. Ai Y.L. Cao H.X. Yang Electrochim. Acta 49 2004 4189 [16] H. Mao, D.S. Wainwright, US Patent 6074,776, 2000. [17] J.N. Eeimers, B.M. Way, US Patent 6074,777, 2000. [18] H. Kim, US Patent 2006, 160,000, 2006. [19] F. Kazuyasu, US Patent 2006, 166,102, 2006. [20] K. Xu Chem. Rev. 104 2004 4303 [21] G.X. Wang S. Bewlay S.A. Needham H.K. Liu R.S. Liu V.A. Drozd J.F. Lee J.M. Chen J. Electrochem. Soc. 153 2006 A25 [22] J. Gao L.J. Fu H.P. Zhang T. Zhang Y.P. Wu H.Q. Wu Electrochem. Commun. 8 2006 1726 [23] J. Gao L.J. Fu H.P. Zhang L.C. Yang Y.P. Wu Electrochim. Acta 53 2008 2376 [24] L.J. Fu H. Liu H.P. Zhang C. Li T. Zhang Y.P. Wu R. Holze H.Q. Wu Electrochem. Commun. 8 2006 1 [25] N.H. Zhao L.J. Fu L.C. Yang T. Zhang G.J. Wang Y.H. Deng Y.P. Wu T. van Ree Pure Appl. Chem. 80 2008 2283 [26] T. Zhang J. Gao H.P. Zhang L.C. Yang Y.P. Wu H.Q. Wu Electrochem. Commun. 9 2007 886 [27] N.H. Zhao G.J. Wang Y. Huang B. Wang B.D. Yao Y.P. Wu Chem. Mater. 20 2008 2612 [28] L.C. Yang Q.S. Gao Y.H. Zhang Y. Tang Y.P. Wu Electrochem. Commun. 10 2008 118 [29] V.A. Sethuraman L.J. Hardwick V. Srinivasan R. Kostecki J. Power Sources 195 2010 3655 [30] Q. Xia B. Wang Y.P. Wu H.J. Luo S.Y. Zhao T. van Ree J. Power Sources 180 2008 602 [31] B. Wang Q.T. Qu Q. Xia Y.P. Wu X. Li C.L. Gan T. van Ree Electrochim. Acta 54 2008 816 [32] C. Korepp W. Kern E.A. Lanzer P.R. Raimann J.O. Besenhard M. Yang K.C. Möller D.T. Shieh M. Winter J. Power Sources 174 2007 628 [33] G. Schroeder B. Gierczyk D. Waszak M. Kopczyk M. Walkowiak Electrochem. Commun. 8 2006 523 [34] M. Walkowiak G. Schroeder B. Gierczyk D. Waszak M. Osinska Electrochem. Commun. 9 2007 1558 [35] K. Amine Q.Z. Wang D.R. Vissers Z.C. Zhang N. Rossi R. West Electrochem. Commun. 8 2006 429 [36] M. Lu H. Cheng Y. Yang Electrochim. Acta 53 2008 3539 [37] J.T. Lee Y.W. Lin Y.S. Jan J. Power Sources 132 2004 244 [38] H.J. Santer K.C. Moller J. Ivanco M.G. Ramsey F.P. Netzer S. Yamaguchi J.O. Besenhard M. Winter J. Power Sources 119–121 2003 368 [39] Y.S. Hu W.H. Kong Z.X. Wang X.J. Huang L.Q. Chen Solid State Ionics 176 2005 53 [40] G.H. Wrodnigg J.O. Besenhard M. Winter J. Electrochem. Soc. 146 1999 470 [41] R. Oesten U. Heider M. Schmidt Solid State Ionics 148 2002 391 [42] Z.X. Shu R.S. McMillan J.J. Murray J. Electrochem. Soc. 142 1995 L161 [43] B.A. Trofimov G.F. Myachina L.A. Oparina S.A. Korzhova N.K. Gusarova S.G. Doo M.D. Cho H. Kim J. Power Sources 147 2005 260 [44] L. Wang Y. Huang D. Jia Electrochim. Acta 51 2006 4950 [45] B. Wang Q.T. Qu L.C. Yang Q. Xia Y.P. Wu D.L. Zhou X.J. Gu T. van Ree J. Power Sources 189 2009 757 [46] B. Wang Q. Xia P. Zhang G.C. Li Y.P. Wu H.J. Luo S.Y. Zhao T. van Ree Electrochem. Commun. 10 2008 727 [47] G.W. Gokel O. Murillo Podands Compr. Supramol. Chem. 1996 1 [48] J.W. Steed J.L. Atwood Supramolecular Chemistry 2000 Wiley New York [49] J.L. Atwood J.W. Steed Encyclopedia of Supramolecular Chemistry 2004 Marcel Dekker Inc. New York [50] E.G. Shim T.H. Nam J.G. Kim H.S. Kim S.I. Moon Electrochim. Acta 54 2009 2276 [51] H.P. Zhang Q. Xia B. Wang L.C. Yang Y.P. Wu D.L. Sun C.L. Gan H.J. Luo A.W. Bebeda T. van Ree Electrochem. Commun. 11 2009 526 [52] S. Izquierdo-Gonzales W.T. Li B.L. Lucht J. Power Sources 135 2004 291 [53] S. Chen Z. Wang H.L. Zhao H.W. Qiao H.L. Luan L.Q. Chen J. Power Sources 187 2009 229 [54] H.F. Xiang H.Y. Xu Z.Z. Wang C.H. Chen J. Power Sources 173 2007 562 [55] M. Arakaw J. Yamaki J. Electroanal. Chem. 219 1987 273 [56] A.N. Dey B.P. Sullivan J. Electrochem. Soc. 117 1970 222 [57] Underwriters Laboratories, see e.g. http://www.boedeker.com/bpi-ul94.htm#VB , last accessed 18/10/2010. [58] Y.E. Hyung D.R. Vissers K. Amine J. Power Sources 119–121 2003 383 [59] M. Yoshio H.Y. Wang K. Fukuda Y. Hara Y. Adachi J. Electrochem. Soc. 147 2000 1245 [60] M. Roy J.K. Nelson R.K. MacCrone L.S. Schadler C.W. Reed R. Keefe W. Zenger IEEE Trans. Dielect. Elect. Insul. 12 2005 629 [61] B.J. Ravoo D.N. Reinhoudt S. Onclin Angew. Chem. Int. Ed. 44 2005 6282 [62] M.J. Stevens Langmuir 15 1999 2773 [63] P. Silberzan L. Leger D. Ausserre J.J. Benattar Langmuir 7 1991 1647 [64] C.P. Tripp M.L. Hair Langmuir 8 1992 1120 [65] C.P. Tripp M.L. Hair Langmuir 11 1995 1215 [66] J.B. Goodenough Y. Kim Chem. Mater. 22 2010 587 [67] J. Ufheil M.C. Baertsch A. Wursig P. Novak Electrochim. Acta 50 2005 1733 [68] X. Xie L. Chen W. Sun J. Xie J. Power Sources 174 2007 784 [69] D. Aurbach Y. Cohen J. Electrochem. Soc. 144 1997 3355 [70] H. Kanamura S. Takezawa Z. Shiraishi Takehara J. Electrochem. Soc. 144 1997 1900

Abstract:A safe electrolyte system is prepared from N -methyl- N -propylpiperidinium bis (trifluoromethanesulfonyl) imide (PP13TFSI), organic electrolyte (1 mol L −1 LiPF 6 /EC-DEC) and lithium bis (oxalato) borate (LiBOB). The additive of LiBOB enhances the stability of interface between electrolyte and anode. The LiBOB-containing mixed electrolytes show non-flammability and good compatibility with active materials. The performance of anode for lithium ion battery is successfully improved by LiBOB-containing mixed electrolytes, which shows 200 mA h g −1 reversible capacities at 0.3 C rate. The ionic conductivity and the lithium ion transference number in LiBOB-containing mixed electrolytes system also suits to application for lithium ion battery. Keywords Ionic liquid Solid electrolyte interphase Electrolyte Safety Li-ion battery 1 Introduction Lithium ion battery is thought as a promising power source for its high voltage, high cycle performance and high energy densities. The increase in improvement of safety is a key point for commercialization of lithium-ion batteries in the auto-mobile application. However, most lithium ion batteries use an organic solvent as electrolyte; the cell temperature at early stage of thermal runaway caused by organic electrolyte burning is thought to be the main reason for the thermal runaway of batteries. So a safe electrolyte needs to be designed in lithium-ion batteries [1,2] . In recent years, ionic liquids (ILs) are widely investigated due to their excellent properties, such as negligible volatility, non-flammability and high thermal stability. ILs also have higher ion conductivity (>10 −3 mS cm −1 ) and wider electrochemistry window, when the properties of high ionic conductivity and electrochemical stability occur simultaneously, ILs electrolytes will be the potential candidates for lithium ion batteries. Compared with tradition organic electrolytes, IL electrolytes have indeed an extraordinary safe character. The use of ILs is expected to provide the lithium battery with the level of safety which is required for large-scale applications such as electric vehicles (EVs) and hybrid electric vehicles (HEVs). One difficulty in utilizing ionic liquid for lithium ion batteries is the poor cathodic stability. The most common class of ILs is that formed by the imidazolium cation and TFSI − anion, such as 1-ethyl-3-methyl imidazolium bis(trifluoromethanesulfonyl) (EMITFSI) which has the higher ion conductivity (1.06 × 10 −2 S cm −1 at 303 K), but this kind IL is usually unstable toward lithium electrode, which is caused by EMI cation [3] . In recent years, pyrrolidine (PY) and piperidine (PP)-based ILs attract people's interest due to their stabilization to lithium [4,5] , the electrochemistry window of these kinds ILs is higher than 5 V and the cathodic potential is about −0.3 V (vs. Li/Li + ), but the poor performance on graphite caused by the unstable electrode/electrolyte interface formed by IL electrolytes, restricts their application in lithium ion batteries, so a stable solid electrolyte interphase (SEI) between the electrolyte and electrode is necessary for IL electrolytes to improve the anode performance, which could be accomplished by appending functional additives in IL-electrolytes system. LiBOB was proposed by Lischka et al. [6] as a promising and advantageous lithium salt for rechargeable lithium ion batteries, instead of the problematic LiPF 6 . LiBOB has many advantages compared with LiPF 6 , including superior thermal stability, increased safety and a more stable SEI. For instance, LiBOB can effectively stabilize the surface of graphite anode even in pure propylene carbonate-based electrolytes [7,8] . Mixed electrolytes comprised of ionic liquid and organic electrolyte have been reported in previous articles [9,10] and the proposed optimal concentration of ionic liquid in mixed electrolyte is 50 vol.%, which shows the no flaming character and good performance in lithium ion battery, but the rate capacity of the cell was still lower for application in practice. In this paper, we prepared an advanced electrolyte composed of 50 vol.% PP13TFSI, 50 vol.% organic electrolyte(1 mol L −1 LiPF 6 /EC-DEC, 1:1, v/v) which as a solvent to reduce the IL's viscosity and increase the system conductivity, 2 wt.% LiBOB as an additive to form SEI. It is found that this compound exhibits a relative high rate capacity with good cycling performance. 2 Experimental The structures of lithium salt and ionic liquid studied in this paper are shown in Fig. 1 . The IL here developed, PP13TFSI, was synthesized according to the route as followed: Dissolving LiTFSI and PP13Br (1:1 mole ratio, purchased from Aldrich) separately in deionized H 2 O and the combining gave pure liquid PP13TFSI, which phase separated from the aqueous layer (LiBr was dissolved in the aqueous layer). The PP13TFSI was washed several times with deionized H 2 O. The final aqueous layer was removed and ionic liquid was purified by activated carbon (Cabot). The mixture was dried under vacuum at 100 °C for 24 h and then 130 °C for 6 h. The Br − content was checked by 5 vol.% AgNO 3 solution and water content was below 40 ppm as measured by a Karl Fischer auto-titrator. The resulting PP13TFSI was a clear, colorless liquid at room temperature. The materials were stored and handled in a glove box, and then mixed with 50% volume of 1 mol L −1 LiPF 6 /EC-DEC (Tianci, Guangzhou), finally 2 wt.% LiBOB (99%, Aldrich) was added in the mixed electrolytes. A 3 μL portion of solution was placed into 20 μL stainless gold sample pan that was then crimp-sealed with a stainless steel lid, the thermal properties of the electrolytes studied in this paper were investigated by differential scanning calorimetric (DSC-Netzsch ST447F5) from 25 to 400 °C at the rate of 5 °C min −1 under argon condition. The electrochemistry window PP13TFSI was measured by cyclic voltammogram (CV) and linear sweep voltammogram. CV and Linear sweep voltammogram were measured on coin cell of Pt |electrolytes| Li at the scan rate of 2 mV s −1 in range of 0–3 V and 3–6 V, respectively. Cyclic voltammogram was measured on half cell of Li/electrolytes/hard carbon–graphite composition. Electrochemical impedance spectroscopy (EIS) was also measured on the same cell above mentioned by electrochemical analyzer CHI604B. The ion conductivity of mixed electrolytes with 2 wt.% LiBOB was measured by conductivity meter (Lei-Ci DDS 307). In this paper, a coin cell was used as a charge–discharge test cell. In order to test the electrochemical performance of the cell using 2 wt.% LiBOB-containing mixed electrolytes, LiFePO 4 cathode and hard carbon–graphite composite anode were used. The electrodes of cathode were prepared by casting slurry containing 80 wt.% LiFePO 4 (LiKai Taiwan), 10 wt.% acetylene black (AB) and 10 wt.% polyvinylydene fluoride on aluminum foil and dried at 120 °C for 10 h under vacuum, the film was punched into disk with area of 0.95 cm −2 and mass ∼2.0 mg for use as the cathode. Hard carbon as anode material has many merits, especially, hard carbon is stable during charge–discharge cycling in PC and ILs-based electrolytes, but hard carbon suffers from a high irreversible capacity compared with graphite for 1st cycle. In this paper, we used hard carbon and graphite composite material as anode; hard carbon–graphite composite anode was prepared by mechanically mixing the hard carbon (Kureha Corp., Japan) with graphite (MAGD, BTR) at the mass rate 1:2 in a planetary ball mill. The anode sheet consisted of 92 wt.% active materials (hard carbon–graphite composite), 5 wt.% acetylene black (AB), and 3 wt.% poly vinylydene fluoride; samples coated on Cu substrate and preparation of electrodes were also same as the cathode manner, finally, the electrodes of anode were assembled to Li/hard carbon–graphite composite half cells. Cells were subjected to battery test system (NewwareTC51), the charge–discharge voltage of anode materials was between 0.02 and 1.5 V, the charge–discharge voltage of cathode materials was between 2.7 and 4.0 V. The electrodes after 1 charge–discharge cycle were taken out and washed twice with the solvent DMC and then dried under vacuum overnight. A vacuum sample transporter was used to load the dried samples into the XPS sample chamber from glove box atmosphere. Surface analysis was then conducted on those electrodes with a PHI 5800 XPS system, where Al Ka excitation source was used. The resolution of the measurement generally falls within 0.3 eV and the elemental carbon peak at 284.5 eV was used as the internal reference to calibrate the energy position. The surface of electrode washed with DMC and dried in vacuum was observed with scanning electron microscopy (Hitachi S-4700). Fiberglass wicks (4 cm in length, 2 cm in width) were first immersed in the electrolyte, absorbing about 1 g electrolyte. A lighter was used to burn one end of the fiber. After 10 s, the light was removed from under the fiberglass wicks, the electrolyte was judged to be nonflammable if the electrolyte never ignited during the testing, or if the ignition of electrolyte ceased when the flame was removed. Each electrolyte was tested three times. 3 Results and discussion The structures of IL and LiBOB studied in this work are shown in Fig. 1 . Fig. 2 shows the DSC curves of electrolytes. The pure ionic liquid shows the well thermal ability above 400 °C, the peak temperature of electrolytes containing ionic liquid PP13TFSI is 18–24 °C higher than that of 1 mol L −1 LiPF 6 /EC-DEC. Because the decomposition temperature of LiBOB (314 °C) is lower than that of ionic liquid, the additive of 2 wt.% LiBOB makes the exothermic peaks shift lower in mixed electrolytes system. But compared with 1 mol L −1 LiPF 6 /EC-DEC, the thermal stability of system is improved by ionic liquid. Fig. 3 shows the CV and Linear sweep of platinum electrode in PP13TFSI electrolytes. From 0 V to 3 V, there is no oxidation and reduction peak were found during the CV process. Linear sweep volatmmogram shows the current keep stable between 3 V and 5.3 V. So we can say the electrochemical window of PP13TFSI is higher than 5 V, which is satisfied with the requirement of lithium ion battery. EIS is widely used to investigate electrolyte–electrode interface reactions [11,12] , such as the electrochemical lithium intercalation reaction in carbonaceous material within the carbonate electrolyte system and the subsequent formation of SEI films [13,14] . Fig. 4 a shows the equivalent circuit to fit the EIS of Li |LiBOB containing mixed electrolytes| composite anode cell at 1 V after 1 cycle, Here, the intercept at the Z Re axis in high frequency refers to R e , which includes electrolyte solution resistance, electric contacts resistance and ion conductive resistance. The semicircle in the high frequency range corresponds to the surface film resistance ( R f ); the semicircle in the middle frequency range reflects the charge transfer resistance ( R ct ); and the sloping line in the lower frequency represents lithium-ion diffusion resistance in bulk electrode. Fig. 4 b shows the EIS of Li/hard carbon–graphite composite anode cell using LiBOB-containing mixed electrolytes. It can be seen that EIS before and after cycling presents some obvious difference. Before cycling, here is only a depressed semicircle section in EIS. Since the electrode is not activated, the diameter of the semicircle is very large. After 1 cycle, EIS are composed of two partially overlapped and depressed semicircles and a straight slopping line at low frequency end. In the case of LiBOB-containing mixed electrolytes, the diameter of the semicircle in the range of high frequencies of the impedance spectra is small before the charge–discharge cycles; the value of R f is 137 Ω. However, R f significantly increases to 400 Ω after the first cycle, which demonstrates the formation of SEI by LiBOB on electrode surface [15,16] , the R f value does not change obviously with cycling, after 20 cycles, the value of R f is constant to 300 Ω, this proves well stability of SEI formed by LiBOB. Fig. 4 b inset shows the EIS of cell using LiBOB-free mixed electrolytes, the impendence of interface is increasing from 400 Ω (after 1 cycle) to 700 Ω (after 20 cycle), which is caused by co-interrelation of PP13 + and the continually electrochemical reactions between the electrolyte/active materials. The mechanism of forming SEI is also proved by the cyclic voltammogram [17] , Cell current and potential of the first two cycle of Li/hard carbon–graphite composite anode cells are shown in Fig. 5 . Two reduction peaks observed at ∼0.6 V and ∼0.2 V vs. Li/Li + are due to intercalation of PP13 + and Li + ions in the graphite for mixed electrolytes without LiBOB ( Fig. 5 a). In contrast, the small cathodic peak at 1.75 V is the SEI formation originated by LiBOB in Fig. 5 b. It is should be the decomposition of LiBOB forms a stable SEI film and reduces the decomposition of PP13TFSI and organic electrolytes, then prevents the co-intercalation of PP13 + . The columbic efficiencies of the first two cycles in two different systems are 66.16% and 78.49%, respectively. XPS technique is very useful in analysis of the composition of the solid surface and it has been used for the analysis of the composition and structure of SEI in the literatures [18,19] . To further analyze the composition and structure of SEI formed on the surface of anodes with different electrolytes; XPS was carried out for cells with LiBOB-containing and LiBOB-free mixed electrolytes. The C1s spectra of the anodes are shown in Fig. 6 a and b , the surface spectra between 282 and 290 eV were deconvoluted and the plot was embedded in the relevant plot. As we can see, the difference between LiBOB-containing sample and LiBOB-free one is obvious. Mixed electrolytes with 2 wt.% LiBOB have an abundance of ROCOOLi characterized by the peak corresponding to 289 eV in C1 s due to the ring-opening of the BOB − anion. This result is in well agreement with previous reports on the chemical composition of SEI formed with LiBOB [20,21] . In comparison with the S2p spectra ( Fig. 6 c and d), the most obvious difference between LiBOB-containing and LiBOB-free mixed electrolytes can be seen. In LiBOB-free mixed electrolytes, the S2p signal shows a rather high intensity, while the relative low peak is found in the spectra of the LiBOB-containing sample. So far as the component of electrolyte, TFSI − is the only source of elemental sulphur, so the serious decomposition of IL is thought to occur in LiBOB-free mixed electrolytes. In contrast, no higher significant decomposition of IL is detected in LiBOB-containing mixed electrolytes sample. It can be concluded that LiBOB salt can effectively prevent the ionic liquid-PP13TFSI from further decomposing on the surface of electrode. A pronounced B1s spectrum at 193 eV appears at SEI layers formed in LiBOB-containing mixed electrolytes ( Fig. 6 f), confirming LiBOB indeed takes part in the formation of SEI film. A semiquantitative analysis of the XPS spectra is carried out based on atom sensitivity factor. The comparison of atomic concentration of various elements in SEI layers formed in LiBOB-free and LiBOB-containing mixed electrolytes are shown in Fig. 7 . The atom concentrations of F, N, and S in SEI layer formed in LiBOB-containing electrolytes are obviously lower than that in LiBOB-free electrolytes, while the concentration of Li is higher than that in LiBOB-free mixed electrolytes. This has positive effect on the properties of SEI layer of anode. Fig. 8 shows SEM images of the hard carbon–graphite composite electrode surfaces after 10 cycles in different electrolytes. The surface morphology of LiBOB-free electrode is still homogeneous compared with the initial electrode (e). However, the anode platelets become thicker and rounder showing the expansion of the carbon particles ( Fig. 8 a and b), the morphology of electrode is similar to that in ILs-system [22,23] . Morphological LiBOB-containing electrode observation shows uniform; the surface of the electrode is covered with the compact film when the LiBOB salt is employed in the mixed electrolytes. Well stability of SEI is formation. (1) t + = i s R s b ( Δ V − i 0 R 0 film ) i 0 R 0 b ( Δ V − i s R s film ) Transference numbers were determined by the method reported by Bruce et al. [24] , which is a direct-current polarization method based on the dc polarization technique. Fig. 9 a shows the chronoamperometric curve of Li |2 wt.% LiBOB in mixed electrolytes| Li cell after 10 mV dc pulse. By fitting the curve, the values of the current can be obtained at the initial polarization time, I 0 and under steady-state conditions, I s . Fig. 9 b shows the impedance of the cell before and after polarization. The two curves practically overlap, which indicates that there is no great difference between the initial R 0 and the final resistance R s of the two Li interfaces. This is an additional confirmation of the stability of the lithium electrode in this kind electrolyte. By inserting the values of I 0 , I s , R 0 and R s in Eq. (1) , the lithium transference number results to be 0.64, i.e. a reasonable value for practical application in lithium batteries, which is higher than pure IL electrolyte-system (about 0.4). This may be the 1 mol L −1 LiPF 6 /EC-DEC decreases the viscosity of the system. We also tested the ion conductivity of 2 wt.% LiBOB containing PP13TFSI-1 mol L −1 LiPF 6 /EC-DEC mixed electrolytes at room temperature; the value is about 5.23 mS cm −1 , close to the value of 1 mol L −1 LiPF 6 /EC-DEC (8 mS cm −1 ). The cycling properties of Li/LiFePO 4 cell using mixed electrolytes with and without LiBOB at 0.1 C rate are shown in Fig. 10 . The high capacity and reasonable coulomb efficiency can be seen in the cell using LiBOB-containing mixed electrolytes. The reversible capacity of Li |2 wt.% LiBOB in mixed electrolytes| FeLiPO 4 half battery is 140 mA h g −1 (84.4% LiFePO 4 special capacity), the value is higher than the cell using LiBOB-free mixed electrolytes (reversible capacity is only 132 mA h g −1 at 0.1 C rate), from the second cycle, the coulomb efficiency increases higher than 99%. When the discharge rate increases to 0.15 C, the reversible capacity also can stabilize to 128 mA h g −1 , while the reversible capacity of cell without LiBOB is only 112 mA h g −1 at same discharge rate. We also tested the cell with the LiBOB-containing mixed electrolytes at 1 C rate, similar with other IL electrolytes, the reversible capacity decreases obviously, the decreased reversible capacity is correlation with the poor wetting ability of LiFePO 4 with electrolytes [25] , the additive LiBOB could improve the compatibility between the electrolytes with the LiFePO 4 cathode. The charge–discharge curves of Li |LiBOB-free mixed electrolytes| composite anode are shown in Fig. 11 a . The first charge–discharge capacities are 207 mA h g −1 and 306 mA h g −1 at rate of 0.15 C, coulomb efficiency is 68% for the first cycle. When the charge–discharge rate increases to 0.3 C, the capacity of the cell decreases obviously, the value is lower than 100 mA h g −1 . On the contrary, the initial discharge capacity of the LiBOB-containing cell is 296 mA h g −1 and the charge capacity is 257 mA h g −1 at 0.15 C, which results in coulomb efficiency 86% for the first cycle. Even the charge–discharge rate increases to 0.3 C, the reversible capacity is also higher than 200 mA h g −1 (see Fig. 11 b). Fig. 12 shows the capacity vs. number of cycle for the Li/hard carbon–graphite composite anode cell with different electrolytes. LiBOB-containing mixed electrolytes exhibits the better cycle performance and most stable coulomb efficiency after the first cycle, which indicating the decomposition of electrolytes was well hindered in subsequent cycles. The charge capacity is still 255.1 mA h g −1 even after 20 cycles and the capacity retention ratio is 99%, the coulomb efficiency in LiBOB-containing mixed electrolyte-system is higher than free-LiBOB system. The reversible capacity decreases obviously in LiBOB-free mixed electrolytes with cycle number, the charge capacity is only 180.7 mA h g −1 with the capacity retention ratio 87.4% after 20 cycles, the main reason is the poor SEI results in electrolytes decomposing on electrode. The well stability SEI formed by LiBOB is very important to the cell performance with mixed electrolytes. The burning tests on electrolytes are shown in Fig. 13 , the traditional organic electrolytes - 1 mol L −1 LiPF 6 /EC:DEC(1:1, v/v) ignited instantaneously and was completely engulfed in flame, it continued to be of inflammation condition more than 1 min after removal of the lighter. In contrast, the ionic liquid and 1 M LiPF 6 /EC-DEC (1:1, v/v) mixed electrolytes with 2 wt.% LiBOB never became engulfed in flame during the burning test, when the lighter was removed, they exhibited its self-extinguishing characters, non-flaming ability of ionic liquid and high thermal stability of LiBOB played the important role in the mixed electrolytes. The burning test well proved the safety of 2 wt.% LiBOB containing mixed electrolytes. 4 Conclusions A new electrolyte component was obtained through mixing LiBOB, PP13TFSI and organic electrolyte (1 mol L −1 LiPF 6 /EC-DEC). LiBOB additive effectively enhanced the interface stability of the battery. The 2 wt.% LiBOB-containing mixed electrolytes show higher conductivity (5.23 mS cm −1 ) and reasonable lithium ion transference number (0.63) at room temperature. When this kind electrolyte is used for LiFePO 4 /Li cell, the cell exhibits good performance, high reversible capacity (140 mA h g −1 ) with coulomb efficiency close to 100%, even cycling on composite anode materials (hard carbon mixed graphite) at charge–discharge rate 0.3 C, the battery also shows a reversible capacity of 200 mA h g −1 . The mixed electrolytes containing 2 wt.% LiBOB are found to have non-flammability property. Acknowledgements The authors would like to thank the anonymous reviewers and the finance support from 863 Foundation of China (No. 20009AA11A105). References [1] H. Matsumoto H. Sakaebe K. Tatsumi J. Power Sources 146 2005 45 [2] M. Galinski A. Lewandowski I. Stepniak Electrochim. Acta 51 2006 5567 [3] K. Tsunashima M. Sugiya Electrochem. Commun. 9 2007 2353 [4] A. Fernicola F. Croce B. Scrosati J. Power Sources 174 2007 342 [5] H. Sakaebe H. Matsumoto Electrochem. Commun. 5 2003 594 [6] U. Lischka, U. Wietelmann, M.W. German, Pat., DE 19829030C1 (1999). [7] Z. Chen W.Q. Lu J. Liu K. Amine Electrochim. Acta 51 2006 3322 [8] W.Q. Lu Z.H. Chen H. Joachin J. Power Sources 163 2007 1074 [9] A. Guerfi M. Dontigny P. Charest M. Petitclerc M. Lagace A. Vigh K. Zaghib J. Power Sources 195 2010 845 [10] H. Nakagawa Y. Fujino S. Kozono Y. Katayama T. Nukuda H. Sakaebe H. Matsumoto J. Power Sources 174 2007 1021 [11] K. Xu S.S. Zhang R. Jow J. Power Sources 143 2005 197 [12] D.P. Abraham M.M. Furczon S.H. Kang D.W. Dees A.N. Jansen J. Power Sources 180 2008 612 [13] M. Itagaki S. Yotsuda N. Kobari N. Koabari K. watanabe S. Kinoshita M. Ue Electrochim. Acta 51 2006 1629 [14] J.Y. Song H.H. Lee Y.Y. Wang J. Power Sources 111 2002 255 [15] J. WChoia G. Cheruvallya D.S. Kim J. Power Sources 183 2008 441 [16] L.B. Chen K. Wang X.H. Xie J.Y. Xie J. Power Sources 174 2007 538 [17] M. Lu H. Cheng Y. Yang Electrochim. Acta 53 2008 3539 [18] L.W. Zhao J.I. Yamaki M. Egashira J. Power Sources 174 2007 352 [19] N.S. Choi K.H. Yew H. Kim S.S. Kim J. Power Sources 172 2007 404 [20] K. Xu U. Lee S.S. Zhang Electrochem. Solid-State Lett. 6 2003 A144 [21] S. Santee A. Xiao L. Yang J. Power Sources 194 2009 1053 [22] H.H. Zheng B. Li Y.B. Fu Electrochim. Acta 52 2006 1556 [23] C.K. Chan R. Ruffo S.S. Hong Y. Cui J. Power Sources 189 2009 1132 [24] P.G. Bruce J. Evans C.A. Vincent Solid State Ionics 918 1988 28 [25] J. Hassoun A. Fernicola M.A. Navarra S. Panero B. Scrosati J. Power Sources 195 2010 574

2-Phenylimidazole As an Additive to Prevent the Co-Intercalation of Propylene Carbonate in Organic Electrolyte for Lithium-Ion Batteries

Methyl phenyl bis-methoxydiethoxysilane as bi-functional additive to propylene carbonate-based electrolyte for lithium ion batteries

Phenyl Tris-2-methoxydiethoxy Silane As an Additive to PC-based Electrolytes for Lithium-Ion Batteries

An In-Situ Synergistic Enhancement Strategy from G-C3n4 and PDOL Composite Solid Electrolyte on the Interface Stability of Solid-State Lithium Battery

SEI-Film-Suppression Additive for Enhancing Compatibility and Electrochemical Stability of Graphite Anode in PC-Based Electrolytes

Imidazolium-Based Ionic Liquid As a Solid Electrolyte Interphase- Forming Additive for Lithium Metal Anodes

Combined Stabilizing of the Solid-Electrolyte Interphase with Suppression of Graphite Exfoliation via Additive-Solvent Optimization in Li-Ion Batteries

Diethyl phenylphosphonite contributing to solid electrolyte interphase and cathode electrolyte interphase for lithium metal batteries

Additives-containing functional electrolytes for suppressing electrolyte decomposition in lithium-ion batteries

Influence of Li2CO3 Additive on Interface Features for Lithium-Ion Batteries Graphite Anode

The Role of Cesium Cation in Controlling Interphasial Chemistry on Graphite Anode in Propylene Carbonate-Rich Electrolytes.

Imidazole-Based Lithium Salt LiHDI as a Solid Electrolyte Interphase-Stabilising Additive for Lithium-Conducting Electrolytes

Molecular Adsorption‐Induced Interfacial Solvation Regulation to Stabilize Graphite Anode in Ethylene Carbonate‐Free Electrolytes

A bicomponent electrolyte additive towards stabilized interface for high-performance lithium-ion batteries

Bifunctional LiI additive for poly(ethylene oxide) electrolyte with high ionic conductivity and stable interfacial chemistry

Effects of Li_2CO_3 Additive on the Performance of Graphite Electrode

Improved thermal stability of graphite electrodes in lithium-ion batteries using 4-isopropyl phenyl diphenyl phosphate as an additive

Propylene sulfite as film-forming electrolyte additive in lithium ion batteries

The effects of LiBOB additive for stable SEI formation of PP13TFSI-organic mixed electrolyte in lithium ion batteries

Prop-1-ene-1,3-sultone As Interfacial-Engineering Additive in Nonflammable Electrolyte for 4.45 V LiCoO2/Si-Graphite 2 /Si-Graphite Lithium-Ion Batteries with High Performance

Ethylene Sulfite as Electrolyte Additive for Lithium‐Ion Cells with Graphitic Anodes