Li[B(OCH2CF3)4]: synthesis, characterization and electrochemical application as a conducting salt for LiSB batteries.

DOI: 10.1002/cphc.201402680

Articles

Li[B(OCH2CF3)4]: Synthesis, Characterization and Electrochemical Application as a Conducting Salt for LiSB Batteries Michael Rohde,[a] Philipp Eiden,[a] Verena Leppert,[a] Michael Schmidt,[b] Arnd Garsuch,[b] Guenter Semrau,[b] and Ingo Krossing*[a] tate and THF at RT In DME (0.8 mol L1) it is 3.9 mS cm1, which is satisfactory for the use in lithium-sulfur batteries (LiSB). Cyclic voltammetry confirms the electrochemical stability of Li[B(OTfe)4] in a potential range of 0 to 4.8 V vs. Li/Li + . The performance of Li[B(OTfe)4] as conducting salt in a 0.2 mol L1 solution in 1:1 wt % DME/DOL is investigated in LiSB test cells. After the 40th cycle, 86 % of the capacity remains, with a coulombic efficiency of around 97 % for each cycle. This indicates a considerable performance improvement for LiSB, if compared to the standard Li[NTf2]/DOL/DME electrolyte system.

A new Li salt with views to success in electrolytes is synthesized in excellent yields from lithium borohydride with excess 2,2,2-trifluorethanol (HOTfe) in toluene and at least two equivalents of 1,2-dimethoxyethane (DME). The salt Li[B(OTfe)4] is obtained in multigram scale without impurities, as long as DME is present during the reaction. It is characterized by heteronuclear magnetic resonance and vibrational spectroscopy (IR and Raman), has high thermal stability (Tdecomposition > 271 8C, DSC) and shows long-term stability in water. The concentration-dependent electrical conductivity of Li[B(OTfe)4] is measured in water, acetone, EC/DMC, EC/DMC/DME, ethyl ace-

1. Introduction performance for LiSB.[8, 9] Moreover, the widely used conducting lithium salts for LIB, like Li[PF6], Li[BF4], Li[BOB] and Li[DFOB], are not suitable for LiSB, due to the nucleophilic attack of the polysulfide anion at the boron or phosphorus atom with formation of insoluble lithium fluoride,[7] as well as their incompatibility with DOL. The more costly Li[NTf2] dissolved in DOL/ DME has become the standard conducting salt in LiSB. Next to other approaches on the cathode[6, 10–12] and anode[9, 13–15] side, as well as using the Li[NO3] additive,[16–18] novel electrolyte salts that at least slow down the capacity fading induced by the shuttle mechanism might be another way to address the LiSB problems. Due to our long-standing experience with weakly coordinating anions (WCAs)[19] of the type [MIII(ORF)4][20] and [MV(ORF)6][21] (MIII = B, Al; MV = Nb, Ta, ORF = fluorinated alkoxide) and positive experience with this substance class to stabilize highly reactive cations,[22] weakly bound complexes,[23] as well as to induce high conductivities in low-polarity solvents,[24] we were interested to learn of the potential of the [B(ORF)4] type of ions in LiSB electrolyte applications. The known high thermal stability of the [MIII(ORF)4] salts should also be favorable when compared to Li[PF6], which decomposes above about 50 to 60 8C. Of those WCAs, the lithium salt of the tetrakis(2,2,2-trifluoroethoxy)aluminate (Li[Al(OTfe)4]) has been published as conducting salt in lithium-ion batteries.[25] Other symmetric lithium borates with haloacyloxy groups (Li[B(OCORX)4], X = CF3, C2F5, CClF2, CCl3) were published by Yamaguchi et al. in 2003.[26] Many related lithium borates containing aromatic chelate ligands are published, for example, lithium bis[1,2-benzenediolato-O,O’]borate, lithium bis[3-fluoro-1,2-benzenediola-

Lithium ion batteries (LIB) are one of the most used energy storage devices, because they provide lightweight high energy density and high power density.[1–4] The electrolyte medium, which allows the Li + ions movement between the anode and the cathode, can be liquid, polymeric, ceramic or a gel. A liquid electrolyte typically consists of solutions with a suitable lithium salt like Li[PF6] mixed with two or more non-aqueous organic carbonate solvents.[5] Enormous sulfur deposits, its low price, non-toxicity and the high specific capacity of 1675 Ah kg1 promise to make lithium-sulfur batteries (LiSB) the upcoming choice.[6] A LiSB has a high theoretical energy density of 2600 Wh kg1[7] and the cell voltage is only 2.2 V versus Li/Li + . However, the commercialization of the LiSB is hindered, due to the short cycle life, capacity fading, high selfdischarge and poor safety.[7] These problems originate from the polysulfides, which are formed during the charge and discharge cycles (the “polysulfide shuttle”[6]). Appropriate electrolyte solvents for LiSB are typically ethers, for example, 1,2-dimethoxyethane (DME) and 1,3-dioxolane (DOL). The combination of both ethers has demonstrated a good effect on the

[a] Dr. M. Rohde, Dr. P. Eiden, V. Leppert, Prof. Dr. I. Krossing Institut fr Anorganische und Analytische Chemie and Freiburger Materialforschungszentrum (FMF) Universitt Freiburg, Albertstr. 21, 79104 Freiburg (Germany) E-mail: [email protected] [b] Dr. M. Schmidt, Dr. A. Garsuch, Dr. G. Semrau BASF SE, Ludwigshafen (Germany) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cphc.201402680.

ChemPhysChem 0000, 00, 0 – 0

These are not the final page numbers! ÞÞ

1

0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

Articles to-O,O’]borate,[27] lithium bis[2,2’-biphenyldiolato)-O,O’]borate, lithium bis[salicylato]borate,[28] lithium bis[5-fluoro-2-olato-1benzenesulfonato-O,O’]borate,[29] and lithium bis[2,3-pyridinediolato-O,O’]borate.[30] Other non-aromatic chelate borates such as lithium bis(malonato)borate[31] and lithium (perfluoropinacolato)borate[32] should also be mentioned. In this manuscript, the first synthesis, characterization and electrochemical performance of Li[B(OTfe)4] as a electrolyte salt for LiSB up to the pouch-cell level is described.

2. Results and Discussion

ponent. A glance at the electrostatic potential map (Figure 1) confirms a rather uniform distribution of the negative charge, with the oxygen atoms being the most negatively charged entities. This indicated that [B(OTfe)4] should be a weakly coordinating anion (WCA). In addition, the calculated molecular orbital energies (Table 1) gave an idea of the stability against reduction and oxidation in comparison to related, as well as typical electrolyte anions.

Table 1. Comparison of molecular orbital energies of tetrakis(trifluoroethoxy)borate with other weakly coordination anions (RI-BP86/def2-TZVPP). Tf = SO2CF3 ; [BOB] = bis(oxalato)borate.

The lithium salt of the homoleptic anion [B(OTfe)4] , (OTfe = OCH2CF3, Figure 1) was considered a promising candidate for the use as conducting salt in lithium-ion batteries. The four chemically robust OTfe groups

MO

[B(OTfe)4]

[PF6]

[B(Ohfip)4]

[Al(Ohfip)4]

[NTf2]

[BOB]

HOMO LUMO Gap

2.763 eV + 3.541 eV + 6.304 eV

3.360 eV + 5.525 eV + 8.885 eV

3.849 eV + 2.981 eV + 6.831 eV

4.102 eV + 2.346 eV + 6.447 eV

2.853 eV + 3.172 eV + 6.025 eV

2.958 eV + 0.604 eV + 3.562 eV

The low HOMO energy of 2.76 eV is similar to the HOMO energy of [NTf2] (HOMO = 2.85 eV), which suggests a high stability against oxidation on the cathode. The high LUMO energy of + 3.54 eV and large HOMO–LUMO gap of + 6.30 eV presume a high resistance against reduction. These calculations suggest that the [B(OTfe)4] WCA has the potential to favorably act as part of a novel conducting salt for lithium-ion batteries. 2.2. Synthesis and NMR Spectroscopic Characterization The reaction between lithium borohydride and four equivalents HOTfe led to Li[B(OTfe)4] with formation of hydrogen gas [Eq. (1)]. This reaction works most simple and with quantitative yield in 1,2-dimethoxyethane (DME).

Figure 1. Left: Lewis and ball and stick formulas of [B(OTfe)4] ; right: electrostatic potential of [B(OTfe)4] projected onto an isodensity surface (0.01 e 3 ; calculated at the RI-BP86/def2-TZVP level).

LiBH4 þ 4 HOCH2 CF3solvent !Li½BðOCH2 CF3 Þ4 þ 4 H2

However, due to the cost aspect of the rather expensive DME, it was also carried out in several more affordable solvents like toluene or mixtures of those with DME. Based on our experience with MI[B(Ohfip)4],[34, 35] attention should be paid to the complete conversion with respect to the formation of the intermediate Li[HB(OTfe)3]. To avoid this, the reaction mixture in all of the considered solvents/solvent mixtures had to be refluxed for more than four hours with the use of a reflux condenser, which was cooled to 208C to avoid loss of the volatile alcohol.

connected to the central boron atom are strongly electronwithdrawing and lead to a delocalization of the negative charge over the entire anion surface. The synthesis of the respective sodium salt Na[B(OTfe)4] from sodium borohydride and the alcohol as a product of the disproportionation of Na[HB(OTfe)3], was described first by Golden et al. in 1992.[33] The related borate anion [B(Ohfip)4] (Ohfip = C(H)(CF3)2) was first synthesized and characterized 2011 by the Krossing group.[34] In contrast to the more bulky fluorinated alkoxy groups, like Ohfip, the OTfe group is lighter and the molecular weight of Li[B(OTfe)4] is with 413.88 g mol1 at the limit of being acceptable for the use in LIB/LiSB. In addition, HOTfe is much cheaper than HOhfip (Sigma Aldrich: 500 g HOTfe = 213 E, HOhfip = 816 E).

2.2.1. Reaction in DME In the 11B NMR of the reaction performed in DME solvent, only one singlet at 2.3 ppm was observed; the absence of other signals indicate a complete conversion to the desired Li[B(Ohfip)4]. The 1H NMR spectrum showed the signal of the OCH2 unit at 3.86 ppm (quartet, 3J(1H,19F) = 9.62 Hz). In the 19F NMR spectrum the main product signal occurred at 76.53 ppm (triplet, 3J(1H,19F) = 9.62 Hz), and two other, very low intensity

2.1. Initial Quantum Chemical Investigations Orienting quantum chemical calculations were performed to judge the quality of the [B(OTfe)4] anion as a electrolyte com-

&


www.chemphyschem.org

ð1Þ

2


ÝÝ These are not the final page numbers!

Articles signals at 76.30 ppm and 78.20 ppm (See the Supporting Information)—all in the range of chemically equivalent CF3 groups. The low-intensity signal high field at 78.20 ppm (0.5 % of the main signal) is also split into a triplet, but is slightly broadened. With a blind sample, the signal was assigned to HOTfe, indicating that the product was insufficiently dried in vacuum. The third, least intense signal at 76.30 ppm (0.2 % of the main signal), was observed between the 13C satellite and the main signal and remained unassigned. Only one sharp signal at 1.04 ppm (D1/2 = 1.7 Hz) was detected in the 7 Li NMR spectrum.

blind spectrum of Li[OTfe] or, more importantly, 2,2,2-trifluoroethanol at 76.80 ppm (NMR of a blind probe).

2.3. Vibrational Spectroscopy of Li[B(Ohfip)4] The colorless powder synthesized in DME was dried overnight in vacuum (1 103 mbar) to remove all traces of HO-Tfe or DME. Then an ATR-IR and Raman spectrum was measured. For assignment of the vibrational bands, an IR and Raman spectrum was simulated from a quantum chemical calculation of isolated [B(OTfe)4] at the PBE0/def2-TZVPP level and is superimposed in Figure S1. The bands of the simulated spectra are good accordance to the experimentally measured vibrational bands (Table 2). In the IR spectrum BO2 bending vibrations are visible at 514, 568 and 602 cm1. The corresponding stretching vibrations are at 1006 and 1053 cm1. The strong CF bands lie around 1159 cm1. In the Raman spectrum, from 699 down to 470 cm1 the bending vibrations of the borate anion as well same OTfe deformations occur. The intense Raman active B(OC)4 breathing mode of the anion appears at 853 cm1. An interesting difference of the experimental to the calculated spectrum in the gas phase is found for the CH stretching vibrations and deformation vibrations of the OCH2CF3 group. In the gas phase only one d(CH2CF3) vibration band was calculated at 1243 cm1 (Raman), due to the full S4 symmetry. In the solid state this vibration splits into two bands at 1278 and 1312 cm1, implying that the local symmetry is reduced to at least C2v. This indicates that the lithium cation probably coordinates to the oxygen atoms and influences the vibrations. This assignment was supported by the similar behavior of the corresponding CH stretching vibrations: The calculated gas phase Raman spectrum predicts only two bands at 2905 and 2936 cm1. Due to symmetry lowering, four bands (2757, 2819, 2906, 2966 cm1) were observed in the Raman experiment.

2.2.2. Reaction in Toluene By contrast, the reaction in pure toluene was less clean. In a control NMR of the solution taken after one hour reflux the signal of the intermediate Li[HB(OTfe)3] was visible, but after three additional hours reflux this signal was absent in the reaction solution. However, dissolving the colorless solid obtained after work-up in DME showed that 6 % of the Li[HB(OTfe)3] was still present [dublet, d11B = 6.4, 1J(1H, 11B) = 133 Hz]. Note that these signals were not observed in NMR samples taken in DMC and THF, perhaps due to the mediocre solubility of the intermediate product in these solvents.

2.2.3. Reaction in Toluene with Two and Four Equivalents of DME To overcome the potential solubility issues in pure toluene, stoichiometric amounts of chelating DME were added to form the Li(DME)x + (x 2) complex and thereby increase dissociation of the Li[HB(Ohfip)3], probably existing as a tight ion pair in toluene to facilitate the last substitution reaction. To obtain a homogeneous reaction mixture it was necessary to dissolve the Li[BH4] with at least four equivalents of the chelating DME. However, regardless whether two or four equivalents of chelating DME as solvent additive to toluene were used, both reactions led after 4.5 h of reflux to complete conversions with 83 % (4 equiv DME) and 88 % (2 equiv) yield after standard work-up (Supporting Information). Even when using high concentrations of the crude solid products in DME as NMR solvent, no signal(s) of Li[HB(OTfe)3] were detected. Thus, the reaction in toluene with as little as a stoichiometric amount of 2 equiv of DME added, led to a convenient and cheap route to multigram quantities of spectroscopically pure Li[B(Ohfip)4] (e.g. 16.7 g product in one reaction).

2.4. Thermal Stability The thermal stability of the conducting salt is of great importance for the safety of the battery, taking into account that a lithium batteries use organic electrolytes, which have high volatility and flammability (danger of thermal runaway[36] and release of heat and gas). The main commercial lithium salt Li[PF6] is known to begin to decompose at about 508C into LiF and PF5.[37–40] PF5 is a strong Lewis acid[41] and can attack the ethylene carbonate with formation of poly(ethyleneoxide) and CO2.[42, 43] Furthermore, the formed PF5 reacts readily with water traces to toxic HF, leading to further solvent decomposition and gas generation.[44–46] DSC measurements of Li[B(OTfe)4] can be seen in Figure S2. The onset of decomposition of the lithium borate begins at 2718C, the maximum as an endothermic peak at 2818C. This high thermal stability of the lithium salt is similar to the decomposition temperature of Li[BOB] (2868C);[47] Li[Tf2N] melts at 2348C.[48] With a decomposition temperature of 3108C, only Li[BF4] decomposes at higher temperatures.[49, 50]

2.2.4. Long-term Stability of Li[B(OTfe)4] in Water The borate is stable in water for at least eleven days, as shown by a long-term NMR measurement (Table S1). The 11B NMR shift of Li[B(OTfe)4] remains at about 3 ppm. The likely decomposition product of a borate in water is boric acid, which was not observed in the NMR, not even after eleven days. Also the more reliable 19F NMR shift of the -CF3 group of Li[B(OTfe)4] in D2O at 76.69 ppm differs significantly in all spectra from the ChemPhysChem 0000, 00, 0 – 0



3


&

Articles Table 2. Comparison of the experimental ATR-IR and Raman bands of Li[B(OTfe)4] powder, the experimental IR bands of HOTfe and the calculated bands of the S4-symmetric [B(OTfe)4] anion (PBE0/def2-TZVPP level).[a] Calc. IR [cm1]

Exp. IR [cm1]

Exp. Raman [cm1]

Calc. Raman [cm1]

Assignment[b]

Exp. IR HOTfe [cm1]

– – 526 (vw) 551 (vw) 558 (w) 577 (w) – 679 (w) 695 (w) – – 836 (w) 845 (w) 954 (w) – 1007 (s) 1038 (vs) – 1141 (vs) 1166 (sh) 1171 (vs) 1181 (vs) 1282 (s) – – 1413 (w) 1467 (vw) – – 2942 (m) 2979 (w) –

– – 514 (w) 534 (w) 551 (w) 568 (w) 602 (w) 679 (m) 694 (sh) – – 839 (m) 867 (sh) 962 (s) 973 (sh) 1006 (m) 1053 (vs) 1101 (vs) 1159 (vs) – – 1195 (sh) 1282 (s) 1303 (sh) 1379 (vw) 1431 (w) 1464 (vw) 2825 (vw) 2851 (vw) 2956 (vw) – –

470 (vw) – 535 (vw) 549 (vw) 572 (vw) 597 (vw) 633 (w) 680 (vw) 699 (vw) 799 (vw) 853 (s) – – – 966 (w) 1004 (vw) – – 1164 (w) 1185 (sh) – 1210 (vw) 1278 (w) 1312 (w) – – 1467 (w) 2757 (vw) 2819 (vw) 2906 (m) 2966 (m) –

495 (vw) 506 (vw) 523 (vw) 541 (vw) 561 (w) – 605 (vw) 655 (vw) 670 (vw) 787 (vw) 815 (s) – – 920 (m) 966 (w) – – 1112 (w) 1150 (w) – – – 1243 (m) – 1360, 1369 (vw) – 1149 (w) – – 2905 (s) 2936 (s) –

d(BO2) d(BO2) d(BO2) d(OTfe) d(OTfe) d(BO4) d(BO4) d(BOC) d(BOC) dbreathing(BO4) dbreathing(B(OC)4) d(CC) d(CC) drocking(CH) drocking(CH) n(BO) n(BO) n(CO) n(CF) d(CH) n(CO) n(CC) d(CH2CF3) – d(CH) dwagging(CH) dscissoring(CH) n(CH) n(CH) nsymm(CH) nasymm(CH) –

– – – 534 (vw) 549 (w) –

2.6. Conductivity Measurements of Li[B(Ohfip)4] Solutions

663 (w) – – – 828 (w) – 946 (m) – – – 1080, 1088 (m) 1141 (vs) – – 1208 (vw) 1275 (vs) – 1374 (w) 1415 (w) 1455 (w) – – 2890 (w) 2964 (w) 3387 (vs)

[a] w: weak, m: medium, s: strong, sh: shoulder, v: very. [b] From a visualization of the calculated spectra.

2.5. Qualitative Solubility Measurements of Li[B(OTfe)4] in Several Solvents/Solvent Mixtures The commonly used solvent in liquid non-aqueous electrolytes for LIB are carbonate-based solvents, like dimethyl carbonate, ethylene carbonate, propylene carbonate or ethyl methyl carbonate.[51] For LiSB mainly ethers are used.[5] For that reason the solubility of the Li[B(OTfe)4] in diverse solvents was tested. Attempts to dissolve Li[B(OTfe)4] in several solvents at mainly higher concentrations were accompanied by formation of two phases (see Section 2.7). In Table 3 the results are listed. In the commonly used carbonate mixtures, Li[B(OTfe)4] was only soluble in the concentration range from 0.2–1.0 mol L1, with formation of the two phases. Solutions in dimethyl carbonate were not even liquid, but rather gel-like or highly viscous. Only at low concentrations (0.1 mol L1) in a 1:1 wt % mixture of ethylene carbonate and dimethyl carbonate it was completely soluble. By addition of the chelating ether DME to the ethylene carbonate/dimethyl carbonate (1:1) mixture the solubility could be increased. In other solvents like o-difluorobenzene, propylene carbonate, ethylene carbonate, methylene chloride, diethyl ether a 0.5 mol L1 solution was not clear, but showed phase separation. Good solvents were shown to be

&



ethyl acetate and 1,2-dimethoxyethane. Even better solubility was achieved in acetone, THF and water. In these solvents, no phase separation was observed.

The electrical conductivities were measured for some electrolyte solutions. In Figure 2 the electrical conductivities for several concentrations are shown and the values of all measurements are listed in the Table 4. The conductivity of Li[B(OTfe)4] in THF passes through a maximum at a concentration of 0.9 mol L1 and is 276 mS cm1, which is rather low, considering that the requirements of LiSB are around 4 mS cm1. Even in ethyl acetate (197 mS cm1) or ethylene carbonate/dimethyl carbonate (1:1) (411 mS cm1) a conductivity beyond the 1 mS cm1 threshold was not observed. In the gel-like dimethyl carbonate solution no electrical conductivity was measurable. A solution in DME had a maximum conductivity of 3.91 mS cm1 at a concentration of 0.8 mol L1. A mixture of ethylene carbonate/dimethyl

Table 3. Solublility behaviour of lithium tetrakis(2,2,2-trifluoroethoxy)borate in several solvents.

4

Solvent

Permittivity er[6]

Concentration [mol L1]

Solution…?

acetone dichloromethane dimethyl carbonate ethylene carbonate diethyl ether ethyl acetate ethyl acetate propylene carbonate tetrahydrofurane 1,2-dimethoxyethane 1,2-dimethoxyethane o-difluorobenzene water EC:DMC (1:1) EC:DMC (1:1) EC:EE (1:1) EC:DMC:DME (1:1:1) EC:DMC:DME (1:1:1)

21.01 8.93 3.09 89.78 4.27 6.08 6.08 64.95 7.52 7.30 7.30 13.38 80.10 – – – – –

1.0 0.5 1.0–0.1 0.5 0.5 1.0–0.6 0.5 0.5 1.0 1.0–0.6 0.5 0.5 1.0 1.0–0.2 0.1 0.5 1.0–0.3 0.2

clear phase phase phase phase phase clear phase clear phase clear phase clear phase clear phase phase clear

formation formation formation formation formation formation formation formation formation formation formation



Articles

Figure 3. Proposed structures for lithium cation coordination. a) Dimeric core of the anion. b) Solvation by one DME molecule and one anion to the lithium cation.

dicate an asymmetric environment and recorded from a 0.9 mol L1 solution of Li[B(OTfe)4] in DME (0.9 mol L1 corresponds to 10.7 DME solvent molecules per one dissolved lithiFigure 2. Electrical conductivity of Li[B(OTfe)4] in several solvents at 258C. um cation). At this concentration, the solution separated in Arrows indicate the occurrence of phases; below the indicated threshold two phases, an upper clear and a lower slightly turbid phase. concentration, clear and colorless solutions were formed. The 7Li NMR spectrum of the clear solution (Figure S3 a) is sharp and symmetric (D1/2 = 1.7 Hz). This implies that the lithium ion resides in a symmetric Table 4. Electrical conductivity of solutions of Li[B(OTfe)4] in several solvents at 258C. environment. By contrast, the 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 c [molL1] 7 Li NMR resonance of the turbid 1 Solution Electrical conductivity of the solution [mS cm ] lower phase (Figure S3 b) is obviwater 11.50 12.82 13.17 13.01 12.52 12.20 11.68 – – – ously deformed, due to the presacetone 4.16 4.51 4.86 4.68 4.79 4.58 4.20 3.82 – – ence of at least two independDME 3.36 3.76 3.91 3.63 3.41 – – – – – DME:EC:DMC – – – – – 1.56 2.22 2.38 2.15 1.38 ent lithium signals. This implies EC:DMC 0.289 0.322 0.341 0.365 0.382 – 0.383 0.406 0.411 0.405 that the lithium is in another coTHF 0.271 0.276 0.257 0.234 0.205 – – – – – ordination environment. Because EE 0.179 0.189 0.197 0.184 0.163 0.100 0.077 0.050 0.022 – there are no other impurities, it DMC 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 is likely to assume that the lithium is coordinated by the [B(OTfe)4] anion, if there is only little donor solvent present. Otherwise, the lithium cation may carbonate/DME (1:1:1 wt %) exhibits a maximum conductivity be coordinated by one anion and one DME molecule (Figat 0.3 mol L1 of 2.38 mS cm1. Very good conductivities were ure 3 b). Such a proposed structure is close to the published observed in acetone (4.86 mS cm1 at 0.8 mol L1; water consolid state structure of (DME)Na(m-Ohfip)2B(Ohfip)2.[34] tent acetone < 75 ppm) and water, where the conductivities 1 1 reached 13.17 mS cm (at 0.8 mol L ). 2.7.2. Vibrational Spectra 2.7. Investigation of the Phase Behavior in DME Solution

To verify the assumption that the lithium cation remains coordinated to the oxygen atoms, further calculations and measurements were done. Therefore, a quantum chemical calculation at the simpler RI-BP86/def-SV(P) level was used, to simulate a vibrational spectrum of the anion, which coordinates a lithium cation. For the saturation of the coordination sphere of the lithium cation two formaldehyde molecules were used as additional donor molecules. This coordination compound simulates the effect of the lithium cation on the stretching bands well, especially the influence on the BO bands. On the other side, gas-phase calculations of the isolated anion simulate the naked status of the anion. Experimentally, the lithium should be detached from the anion in the presence of a larger excess of the strong donor solvent DME and a naked borate anion should be available in a dilute solution. To verify these assumptions, a more dilute solution (0.1 mol L1) of Li[B(OTfe)4] in DME was prepared to obtain the solvent-separated ions [Li(DME)x] + and [B(OTfe)4] . Figure 4 shows the ATR-IR spectra

2.7.1. NMR Spectra Upon closer inspection of the 11B NMR spectra of highly concentrated solutions in DME, a small shoulder next to the product signal was visible. Also in the 19F NMR, a slight increase in the previously unknown signal was noticed, downfield to the main signal. The intensity of this resonance increases in both the 11B NMR and 19F NMR with the reduction of DME content during several syntheses. This suggested that the resonances come from the same compound. It may be that the unknown compound is a complex of two borate anions that chelate one lithium cation (Figure 3 a). Such a dimeric anion would account for the similar chemical shifts. This structure complies to a related known solid-state structure, which includes such a complex: Li{[Al(Ohfip)4]2} with [CPh3] + or [EMIm] + cations.[52] Moreover, it is in accordance with the broader 7Li NMR signals (up to D1/2 = 3.7 Hz), which inChemPhysChem 0000, 00, 0 – 0



5


&

Articles fore, each phase was investigated separately by IR spectroscopy. In Figure 5 the ATR-IR spectra are shown. Additionally the spectra of DME and the ATR-IR spectra of the solid product are superimposed.

Figure 4. Comparison of the naked borate anion versus the anion, which is coordinated to a lithium cation. Detailed part of the ATR-IR spectra of Li[B(OTfe)4] powder (red line), a 0.1 m DME solution of Li[B(OTfe)4] (blue line), the calculated vibrational spectrum of isolated S4-symmetric [B(OTfe)4] (blue shaded spectrum) and a calculated vibrational spectrum of (formaldehyde)2Li(m-OTfe)2B(OTfe)2 at the BP86/def2-SV(P) level (red shaded spectrum). The upper right inset visualizes the calculated structure of (formaldehyde)2Li(m-OTfe)2B(OTfe)2.

Figure 5. ATR-IR spectra of the upper clear and lower turbid phase of a 0.8 mol L1 solution of Li[B(OTfe)4] in DME. Additionally the ATR-IR spectra of the solvent DME and the solid Li[B(OTfe)4] are included.

of solid Li[B(OTfe)4], Li[B(OTfe)4] dissolved in DME (0.1 mol L1), the calculated vibrational spectrum of naked [B(OTfe)4] and the calculated spectrum of (formaldehyde)2Li(m-OTfe)2B(OTfe)2. By comparing the experimental spectra the range at 1000– 1050 cm1 stands out because of the significant deviations. In this range, stretching vibrations of the BO bands occur, but no vibrations from the solvent DME. The BO bands of the solid product were assigned as 1006 and 1053 cm1. In the IR spectrum of the dilute dissolved Li[B(OTfe)4] these bands disappear and new bands at 1042 and 1032 cm1 occur. This is in agreement with the simulated spectrum. The calculated BO stretch of the coordinated anion should vibrate at 1067 and 1010 cm1. For the naked anion, the BO bands were calculated to be more similar, but markedly shifted to 1025 and 1012 cm1. These facts demonstrate the significant effect of lithium coordination to the BO vibration and on the corresponding connected OCH2CF3 group. The degeneration of the CH stretching vibrations due to the coordination was also confirmed by the calculation. For completion, at a closer look at the spectrum of Li[B(OTfe)4] in DME, a shoulder at 1085 cm1 is recognizable. This is probably due to DME, which coordinates the lithium cation. The coordination to the lithium ion weakens the CO bond of the ether, which leads to a red shift in the IR spectrum. One consequence of the lithium coordination at the anion is the limited solubility: Attempts to make a 1.0 or 0.8 mol L1 solution of Li[B(OTfe)4] in DME were always accompanied by the occurrence of two phases, namely an upper clear phase and a turbid lower phase. After stirring, a separation of the phases was observed. The lower phase was not a solid residue, it was a liquid and similarly viscous to the upper clear phase. There-

&



The difference between the two phases can again be seen in the range between 1000–1050 cm1. The spectrum of the turbid phase (red line) resembles the spectrum of the powder (black line), with BO stretches at 1050 and 1009 cm1. This is close to the BO bands in the solid state (1053 and 1006 cm1). By contrast, the BO bands of the clear phase (cyan) appear at 1042 and 1032 cm1. Therefore, the BO bands differ significantly from the BO band in the turbid phase and from the powder. Instead, these stretches resemble rather the BO band of the naked anion. This observation is similar to the behavior described above. In the lower turbid phase, the lithium ion appears to still be coordinated to the anion. This suggests that at higher concentrations, the bonding of the borate anion [B(OTfe)4] to the lithium cation is competitive with the chelating solvent DME. We conclude that in the upper clear phase solvent separated ions like [Li(DME)2] + (solv)/[B(OTfe)4] (solv) probably are present, and in the lower phase the ion pairs are not dissociated, for example, as in in (DME)Li(m-OTfe)2B(OTfe)2 shown in Figure 3 b. It appears that this compound is liquid at room temperature. For all further analyses, it was verified that only clear solutions with solvent-separated ions were used for the investigations. 2.8. Cyclic Voltammetry (CV) The stability of electrolytes containing Li[B(OTfe)4] was investigated over a wide potential range of 0 V to 5 V. A half molar solution of Li[B(OTfe)4] in DME and a 0.1 mol L1 solution in ethylene carbonate/dimethyl carbonate (1:1) were surveyed in two different potential ranges. A platinum working electrode, 6



Articles a lithium metal counter electrode and a lithium metal reference electrode were used. Cycling in a potential range of 3– 5 V against Li/Li + simulates the charging of a battery cell. The results of this anodic cycling reveal the oxidation stability of the electrolyte. On the other side, the cathodic cycles, at a range of 3–0 V versus Li/Li + , investigate the stability against reduction. In Figure 6 the CV measurements of Li[B(OTfe)4] solutions are shown.

is probably stable down to 0 V. The first five scans in carbonate solvents in the anodic scan (Figure 6 d) hardly differ from each other. The current at the working electrode remains low up to 4.2 V. The current at this plateau is low, but indicates an oxidation process, but likely not from the lithium salt, because the current is too small. At potentials higher than 4.8 V, the current increases to 6 mA. It is likely an irreversible oxidation process of the anion. Together, we assume that the Li[B(OTfe)4] salt is stable in the voltammetric range between 0 to 4.8 V. 2.9. Cycle Behavior as Electrolyte in a LiSB Battery

Because of the good solubility of the Li[B(OTfe)4] salt in ethers, such as 1,2-dimethoxyethane, and electrochemical stability in the potential range of LiSB ( 2.2 V),[53a] it is a candidate for the use as conducting salt in this battery technology. Therefore, the salt was tested in pouch bag cells with a lithium metal anode. The cathode consisted of 60 wt % elemental sulfur as active material and 35 wt % carbon black (Vulcan XC72) as conductive material with polyvinylidene difluoride binder (5 wt %). The electrolyte, a soluFigure 6. Cyclic voltammograms of several solutions of Li[B(OTfe)4]. All potentials are given against Li/Li + and tion of 8 wt % Li[B(OTfe)4] salt in were measured with a scan rate of 0.01 V cm1: a) cathodic scans of Li[B(OTfe)4] in DME (0.5 mol L1), b) anodic scans of Li[B(OTfe)4] in DME (0.5 mol L1), c) cathodic scans of Li[B(OTfe)4] in ethylene carbonate/dimethyl carbona 1,3-dioxolane/1,2-dimethoxyate (1:1) (0.1 mol L1), d) anodic scans of Li[B(OTfe)4] in ethylene carbonate/dimethyl carbonate (1:1) (0.1 mol L1). ethane mixture with 1:1 weight ratio (0.2 mol L1), was prepared and cycled in the test cell 2.8.1. Measurement in DME system. Polypropylene separators were used in all cells. For The cyclic voltammetry measurement of the solution of preparation, the cells were discharged in the first cycle with Li[B(OTfe)4] in DME differs from the measurement in ethylene a 0.02 C rate. The cycling tests were performed with a Maccor battery test system at room temperature by using a charge carbonate/dimethyl carbonate (1:1). The same oxidation peak rate of 0.1 C and discharge rate of 0.15 C within the potential of a small impurity occur at 4.2 V. Up to 4.4 V the currents are range of 1.7 to 2.5 V. The resulting discharge and charge calow in the anodic scan (Figure 6 b), but at higher potentials the pacities of each cycle are shown in Figure 7, also the coulomcurrents increases up to 70 mA. This is caused by the known bic efficiency during the cycles. oxidation of the solvent DME, since the ether is easier to oxiThe discharge capacity after the cycle 40 is 822 mAh, with dize than the carbonates. In the cathodic scans (Figure 6 a) the capacity of the 5th cycle set to 100 %. Therefore, the caa similar oxidation process of a slight impurity can be seen at pacity left after the 40th cycle is 86 %. Furthermore, the cou2.05 V. The current of this oxidation increases with each cycle, lombic efficiency during each cycle is constantly above 97 %. but remains very low (< 2 mA) and probably does not originate The results of cycling an analogously built comparative lithium from the anion. sulfur battery with an electrolyte composition containing a mixture of DME/DOL (1:1 by weight) and 8 wt % Li[NTf2] showed 2.8.2 Measurement in EC/DMC a remaining capacity after 40 cycles of only 70 % (the capacity The cathodic scan (Figure 6 c) shows a similar reversible redox is referred to the capacity after 5 cycles set to 100 %). reaction at 2.03 V, but the currents are low, which again sugThe discharge profiles[53b,c] shown in Figure 8 a reveal the gests a reaction of an impurity. At around 0.5 V the typical eththree-stage process during discharge of a lithium-sulfur cell ylene carbonate decomposition was observed. At lower potencontaining 8 wt % (0.2 mol L1) Li[B(OTfe)4] salt solution in DOL/ tials down to 0 V, the currents are low and the Li[B(OTfe)4] salt DME (1:1). During the first plateau (from 2.4 to 2.1 V) sulfur is ChemPhysChem 0000, 00, 0 – 0



7


&

Articles 3. Conclusions We demonstrated the straightforward preparation of the Li[B(OTfe)4] salt in toluene from the reaction of lithium borohydride with a slight excess of HOTfe. For the complete conversion of the starting material and to avoid the intermediate Li[HB(OTfe)3], it is necessary to saturate the lithium cation with at least two equivalents of 1,2-dimethoxyethane. The preferred route uses toluene and two to four equivalents DME as solvent, this ensures the complete solvation of Li[BH4] combined with a minimal amount of DME. The donor-free Li[B(OTfe)4] salt was obtained in multigram scale and characterized with heteronuclear magnetic resonance and vibrational spectroscopy. The [B(OTfe)4] anion performs intermediate as weakly coordiFigure 7. Capacity retention of a lithium sulfur test cell during cycling (charge rate: 0.1 C discharge rate: 0.15 C within a voltage range of 1.7 to nating anion, because apparently the oxygen atoms tend to 2.5 V). The Li[B(OTfe)4] salt was used as conducting salt in a 8 wt % coordinate the lithium cation (cf. influence on the stretching 1 (0.2 mol L ) solution in DOL/DME (1:1). The labels shown are the values vibrations of the OCH2CF3 groups and the corresponding from the 5th and 40th cycle. BO stretches). In the condensed phase, the lithium ion is connected to the oxygen atoms. This fact has consequences on the solubility, so the lithium coordination is only broken with strong donor solvents like DME below 0.8 mol L1. All solutions in the tested organic solvents, with the exception of acetone and THF, showed separation into two phases. It was shown for DME that lithium coordination of the anion still exists in the lower, slightly turbid phase. Probably it mainly consists from (DME)Li(mOTfe)2B(OTfe)2 as shown in FigFigure 8. a) Discharge profiles after the 1st, 10th and 20th cycles. The profiles indicate a high overpotential during ure 3 b. In the clear upper phase, the first charging step. b) Charge profiles after the 1st, 10th and 20th cycles. During cycling, the charging overposolvent separated ions prevail. tential is significantly reduced at 10th and 20th cycle. Nevertheless, DME at concentrations below 0.8 mol L1 is a suitable solvent. The maximum ionic conductivity of Li[B(OTfe)4] in reduced to soluble polysulfides (Li2S8 to Li2S3). The voltage plateau at 2.1 V corresponds to the reduction to insoluble Li2S2 DME (0.8 mol L1) is around 4 mS cm1, which qualified for use as electrolyte in lithium-sulfur batteries. Due to the formation and the discharge region between 2.1 and 1.7 V is due to the of gel-like solutions in dimethyl carbonate, an important solconversion of Li2S2 to Li2S. This solid-state diffusion in the bulk is very sluggish. After 10 and 20 cycles, these plateau regions vent in LIB, no electrical conductivity could be measured. Howcan still be observed, but the cell capacity is somewhat reever, the formation of gel-like electrolytes could be an advantduced. age for safety issues of high-voltage batteries. Here the lithium The charge profiles[53b,c] indicate a high overpotential during borate might be used as an additive to adjust the viscosity of the electrolyte. Additionally, the new lithium salt has shown the first charging step. During cycling the charging overpotena remarkable thermal stability. A decomposition of the pure tial is significantly reduced at the 10th and 20th cycle as lithium salt (DSC) slowly begins at a temperature above 2718C, shown in Figure 8 b. and ranks similar to other highly stable, but less conducting Overall, the use of this novel lithium borate in the electrolyte salts, such as Li[BOB] or Li[BF4]. However, it clearly outperforms showed a significantly improved capacity retention. Since it improved the main problem of LiSB, the capacity fading during Li[PF6]. Also the long-term stability against water is a great adcyclization,[10, 5, 53a] it represented a step forward in LiSB research vantage over other fluorine-containing conducting salts, like Li[PF6] or Li[BF4], which react with water to form toxic and corand this invention was registered as a patent.[54, 55] rosive HF.[37, 44–46] The results from cyclic voltammetry indicate an electrochemical stability in the potential range from 0 V to 4.8 V. This meets the requirements for the use in LIB and largely exceeds the demands of LiSB. The novel lithium salt is prom-

&



8



Articles [6] C. Daniel, J. O. Besenhard, Handbook of Battery Materials, 2nd ed., WileyVCH, Weinheim, 2011, p. 822. [7] S. S. Zhang, J. Power Sources 2013, 231, 153 – 162. [8] V. Etacheri, R. Marom, R. Elazari, G. Salitra, D. Aurbach, Energy Environ. Sci. 2011, 4, 3243 – 3262. [9] Y. Mikhaylik, I. Kovalev, R. Schock, K. Kumaresan, J. Xu, J. Affinito, ECS Trans. 2010, 25, 23 – 34. [10] X. Ji, K. T. Lee, L. F. Nazar, Nat. Mater. 2009, 8, 500 – 506. [11] X. Ji, L. F. Nazar, J. Mater. Chem. 2010, 20, 9821 – 9826. [12] Y. Yang, G. Yu, J. J. Cha, H. Wu, M. Vosgueritchian, Y. Yao, Z. Bao, Y. Cui, ACS Nano 2011, 5, 9187 – 9193. [13] J. Y. Kim, Y. G. Ryu, US Patent Application 20050042503, 2005. [14] Y. Yang, M. T. McDowell, A. Jackson, J. J. Cha, S. S. Hong, Y. Cui, Nano Lett. 2010, 10, 1486 – 1491. [15] J. Hassoun, J. Kim, D.-J. Lee, H.-G. Jung, S.-M. Lee, Y.-K. Sun, B. Scrosati, J. Power Sources 2012, 202, 308 – 313. [16] Y. V. Mikhaylik, US Patent 7354680, 2008. [17] D. Aurbach, E. Pollak, R. Elazari, G. Salitra, C. S. Kelley, J. Affinito, J. Electrochem. Soc. 2009, 156, A694 – A702. [18] R. Elazari, G. Salitra, Y. Talyosef, J. Grinblat, C. Scordilis-Kelley, A. Xiao, J. Affinito, D. Aurbach, J. Electrochem. Soc. 2010, 157, A1131 – A1138. [19] a) I. Krossing in Comprehensive Inorganic Chemistry II, Vol 1. (Eds.: J. Reedijk, K. Poeppelmeier), Elsevier, Oxford, 2013, pp. 681 – 705; b) I. Krossing, A. Reisinger, Coord. Chem. Rev. 2006, 250, 2721 – 2744; c) I. Krossing, I. Raabe, Angew. Chem. Int. Ed. 2004, 43, 2066 – 2090; Angew. Chem. 2004, 116, 2116 – 2142. [20] I. Krossing, Chem. Eur. J. 2001, 7, 490 – 502. [21] U. P. Preiss, G. Steinfeld, H. Scherer, A. M. T. Erle, B. Benkmil, A. Kraft, I. Krossing, Z. Anorg. Allg. Chem. 2013, 639, 714 – 721. [22] a) T. Kçchner, T. A. Engesser, H. Scherer, D. A. Plattner, A. Steffani, I. Krossing, Angew. Chem. Int. Ed. 2012, 51, 6529 – 6531; Angew. Chem. 2012, 124, 6635 – 6637; b) A. J. Lehner, N. Trapp, H. Scherer, I. Krossing, Dalton Trans. 2011, 40, 1448 – 1452; c) T. Kçchner, S. Riedel, A. J. Lehner, H. Scherer, I. Raabe, T. A. Engesser, F. W. Scholz, U. Gellrich, P. Eiden, R. A. P. Schmidt, D. A. Plattner, I. Krossing, Angew. Chem. Int. Ed. 2010, 49, 8139 – 8143; Angew. Chem. 2010, 122, 8316 – 8320; d) I. Raabe, D. Himmel, S. Mueller, N. Trapp, M. Kaupp, I. Krossing, Dalton Trans. 2008, 946 – 956; e) M. Gonsior, I. Krossing, E. Matern, Chem. Eur. J. 2006, 12, 1986 – 1996; f) M. Gonsior, I. Krossing, E. Matern, Chem. Eur. J. 2006, 12, 1703 – 1714; g) I. Krossing, A. Bihlmeier, I. Raabe, N. Trapp, Angew. Chem. Int. Ed. 2003, 42, 1531 – 1534; Angew. Chem. 2003, 115, 1569 – 1572. [23] a) J. Schaefer, D. Himmel, I. Krossing, Eur. J. Inorg. Chem. 2013, 2712 – 2717; b) J. Schaefer, A. Kraft, S. Reininger, G. Santiso-Quinones, D. Himmel, N. Trapp, U. Gellrich, B. Breit, I. Krossing, Chem. Eur. J. 2013, 19, 12468 – 12485; c) A. Higelin, S. Keller, C. Gohringer, C. Jones, I. Krossing, Angew. Chem. Int. Ed. 2013, 52, 4941 – 4944; Angew. Chem. 2013, 125, 5041 – 5044; d) J. Schaefer, A. Steffani, D. A. Plattner, I. Krossing, Angew. Chem. Int. Ed. 2012, 51, 6009 – 6012; Angew. Chem. 2012, 124, 6112 – 6115; e) A. Reisinger, N. Trapp, C. Knapp, D. Himmel, F. Breher, H. Ruegger, I. Krossing, Chem. Eur. J. 2009, 15, 9505 – 9520; G. Santiso-QuiÇones, R. Bruckner, C. Knapp, I. Dionne, J. Passmore, I. Krossing, Angew. Chem. Int. Ed. 2009, 48, 1133 – 1137; Angew. Chem. 2009, 121, 1153 – 1157. [24] I. Raabe, K. Wagner, K. Guttsche, M. K. Wang, M. Gratzel, G. Santiso-Quinones, I. Krossing, Chem. Eur. J. 2009, 15, 1966 – 1976. [25] S. Tsujioka, B. G. Nolan, H. Takase, B. P. Fauber, S. H. Strauss, J. Electrochem. Soc. 2004, 151, A1418 – A1423. [26] H. Yamaguchi, H. Takahashi, M. Kato, J. Arai, J. Electrochem. Soc. 2003, 150, A312 – A315. [27] J. Barthel, R. Buestrich, E. Carl, H. J. Gores, J. Electrochem. Soc. 1996, 143, 3565 – 3571. [28] J. Barthel, R. Buestrich, H. J. Gores, M. Schmidt, M. Wuhr, J. Electrochem. Soc. 1997, 144, 3866 – 3870. [29] J. Barthel, M. Schmidt, H. J. Gores, J. Electrochem. Soc. 1998, 145, L17 – L20. [30] J. Barthel, A. Schmid, H. J. Gores, J. Electrochem. Soc. 2000, 147, 21 – 24. [31] W. Xu, L.-M. Wang, R. A. Nieman, C. A. Angell, J. Phys. Chem. B 2003, 107, 11749 – 11756. [32] M. Videa, W. Xu, B. Geil, R. Marzke, C. A. Angell, J. Electrochem. Soc. 2001, 148, A1352 – A1356. [33] J. H. Golden, C. Schreier, B. Singaram, S. M. Williamson, Inorg. Chem. 1992, 31, 1533 – 1535.

ising for the use in LiSB: its most notable feature is the good cycling performance. The use of an electrolyte consisting of a 0.2 mol L1 solution of Li[B(OTfe)4] in a 1:1 weight ratio mixture of 1,3-di-oxolane/1,2-dimethoxyethane revealed a capacity of the LiSB of 86 % after the 40th cycle. This increased capacity retention due to the new lithium salt is an interesting progress in the lithium-sulfur battery technology, and may help to resolve the main problem of this battery type.

Experimental Section Full details are presented in the Supporting Information; here we only present the optimized synthesis procedure. The Supporting Information contains full experimental details of all reactions leading to the optimized procedure, weights, NMR spectra of the reactions and intermediates, analytical techniques and equipment, and details to the quantum chemical calculations. Synthesis in toluene with two equivalents of 1,2-dimethoxyethane: Lithium borohydride (1.00 g, 46.1 mmol, 1 eq.) was dissolved in 1,2-dimethoxyethane (9.51 mL, 8.27 g, 91.8 mmol, 2 eq.) and then toluene (250 mL) was added. Into this mixture 2,2,2-trifluorethanol (16.64 mL, 22.96 g, 230 mmol, 5 eq.) was dropped at RT within 1 h under stirring. Gas evolution was observed. After the addition, the reaction mixture was refluxed for 4.5 h. The mixture was allowed to reach room temperature. Next, the resulting suspension was filtered, and the colorless powder dried under vacuum (overnight, 103 mbar). The product Li[B(OTfe)4] was obtained as a very fine and fluffy colorless powder (16.7 g, yield: 88 %). Losses in yield are attributed to the fine nature of the powder. NMR spectra of the powder dissolved in 1,2-dimethoxyethane: 1 H NMR (400.17 MHz, DME = 3.30 ppm, toluene-D8 external lock, RT): d = 3.82 ppm (q, 2 H, 3J(1H,19F) = 9.62 Hz, [B(OCH2CF3)4]); 19 F NMR (376.54 MHz, toluene-D8 external lock, RT): d = 76.48 ppm (t, 3F, 3J(1H,19F) = 9.61 Hz, [B(OCH2CF3)4]); 7Li NMR (155.52 MHz, toluene-D8 external lock, RT): d = 1.0 ppm (s, D1/2 = 2.1 Hz, Li[B(OTfe)4]); 11B NMR (128.39 MHz, toluene-D8 external lock, RT): d = 2.3 ppm (s, D1/2 = 9.45 Hz, [B(OCH2CF3)4]). Vibrational spectra are in Table 2.

Acknowledgements This work was supported by the Albert-Ludwigs-Universitt Freiburg and by the BASF SE as part of the Battery Network. We would like to thank Dr. Harald Scherer and Fadime Bitgl for the measurement of the NMR spectra, and Dr. Daniel Himmel for discussions of DFT calculations. Keywords: electrolytes · lithium-ion batteries · lithium-sulfur batteries · vibrational spectroscopy · weakly coordinating anions [1] [2] [3] [4] [5]

H. Li, Z. Wang, L. Chen, X. Huang, Adv. Mater. 2009, 21, 4593 – 4607. M. Armand, J. M. Tarascon, Nature 2008, 451, 652 – 657. M. Armand, J. M. Tarascon, Nature 2001, 414, 359 – 367. J. B. Goodenough, K.-S. Park, J. Am. Chem. Soc. 2013, 135, 1167 – 1176. K. Xu, Chem. Rev. 2004, 104, 4303 – 4418.




9


&

Articles [34] S. Bulut, P. Klose, I. Krossing, Dalton Trans. 2011, 40, 8114 – 8124. [35] L. lvarez Hernndez, Dissertation thesis, Albert-Ludwigs-Universitt (Freiburg), 2013. [36] P. G. Balakrishnan, R. Ramesh, T. Prem Kumar, J. Power Sources 2006, 155, 401 – 414. [37] S. E. Sloop, J. K. Pugh, S. Wang, J. B. Kerr, K. Kinoshita, Electrochem. Solid-State Lett. 2001, 4, A42 – A44. [38] M. Salomon, B. Scrosati, Gazz. Chim. Ital. 1996, 126, 415 – 427. [39] B.-T. Yu, W.-H. Qiu, F.-S. Li, L. Cheng, J. Power Sources 2007, 166, 499 – 502. [40] K. Amine, J. Liu, S. Kang, I. Belharouak, Y. Hyung, D. Vissers, G. Henriksen, J. Power Sources 2004, 129, 14 – 19. [41] K. O. Christe, D. A. Dixon, D. McLemore, W. W. Wilson, J. A. Sheehy, J. A. Boatz, J. Fluorine Chem. 2000, 101, 151 – 153. [42] L. Vogdanis, W. Heitz, Macromol. Chem. Rapid Commun. 1986, 7, 543 – 547. [43] L. Vogdanis, B. Martens, H. Uchtmann, F. Henseland, W. Heitz, Makromol. Chem. 1990, 191, 465 – 472. [44] U. Heider, R. Oesten, M. Jungnitz, J. Power Sources 1999, 81 – 82, 119 – 122. [45] D. Aurbach, G. E. Blomgren, L. A. Dominey, I. Galasiu, R. Galasiu, Y. Gofer, S. Komada, X. Liu, T. Osaka, D. A. Sherson, J. Thonstad, I. Weissman, A. Zaban in Nonaqueous Electrochemistry (Ed.: D. Aurbach), Marcel Dekker, New York, 1999. [46] K. M. Abraham, D. M. Pasquariello, F. J. Martin, J. Electrochem. Soc. 1986, 133, 661 – 666.

&



[47] E. Zinigrad, L. Larush-Asraf, G. Salitra, M. Sprecher, D. Aurbach, Thermochim. Acta 2007, 457, 64 – 69. [48] Y. Hu, H. Li, X. Huang, L. Chen, Electrochem. Commun. 2004, 6, 28 – 32. [49] P. Ping, Q. Wang, J. Sun, H. Xiang, C. Chen, J. Electrochem. Soc. 2010, 157, A1170 – A1176. [50] K. S. Gavrichev, G. A. Sharpataya, V. E. Gorbunov, Thermochim. Acta 1996, 282, 225 – 238. [51] D. Bresser, S. Passerini, B. Scrosati, Chem. Commun. 2013, 49, 10545 – 10562. [52] I. Krossing, H. Brands, R. Feuerhake, S. Koenig, J. Fluorine Chem. 2001, 112, 83 – 90. [53] a) Y.-X. Yin, S. Xin, Y.-G. Guo, L.-J. Wan, Angew. Chem. Int. Ed. 2013, 52, 13186 – 13200; Angew. Chem. 2013, 125, 13426 – 13441; b) H. Schneider, A. Garsuch, A. Panchenko, O. Gronwald, N. Janssen, P. Novk, J. Power Sources 2012, 205, 420 – 425; c) O. Gronwald, A. Garsuch, A. Panchenko, Chimia 2013, 67, 719 – 723. [54] Patent pending Pr.: EP 13177155 (19.07.2013), BASF SE—Use of lithium alkoxy- borates and lithium alkoxyaluminates as conducting salts in electrolytes of lithium sulfur batteries. [55] Patent pending Pr.: EP 13177159 (19.07.2013), BASF SE—Use of lithium alkoxy- borates and lithium alkoxyaluminates as conducting salts in electrolytes of lithium ion batteries.

Received: September 29, 2014 Published online on && &&, 2014

10



ARTICLES Conducting salt for Li batteries? Li[B(OTfe)4] (OTfe = OCH2CF3) was synthesized as a water- and thermally stable lithium-ion electrolyte salt. The electrochemical stability in EC/DMC and in DME lies outside the limits of the solvents. As electrolyte salt, it exhibits good conductivities in many solutions. It was tested in LiS Batteries and shows good performance.




M. Rohde, P. Eiden, V. Leppert, M. Schmidt, A. Garsuch, G. Semrau, I. Krossing* && – && Li[B(OCH2CF3)4]: Synthesis, Characterization and Electrochemical Application as a Conducting Salt for LiSB Batteries

11


&

Facile synthesis and electrochemical performances of hollow graphene spheres as anode material for lithium-ion batteries.

Synthesis and Electrochemical Property of LiMn2O4 Porous Hollow Nanofiber as Cathode for Lithium-Ion Batteries.

reduced graphene oxide films: synthesis, characterization, and electrochemical performance.

Synthesis and characterization of new light emitter symmetrical phenoxazinium salt and its potential application as sensor for assessment of Hg2+.

Hollow Cobalt Selenide Microspheres: Synthesis and Application as Anode Materials for Na-Ion Batteries.

Sonochemical synthesis of LiNi0.5Mn1.5O4 and its electrochemical performance as a cathode material for 5 V Li-ion batteries.

Synthesis and characterization of a helicene-based imidazolium salt and its application in organic molecular electronics.

Synthesis of conducting polymer nanospheres of high electrochemical activity.

Facile synthesis of porous Mn2O3 nanoplates and their electrochemical behavior as anode materials for lithium ion batteries.

Conducting polypyrrole films as a potential tool for electrochemical treatment of azo dyes in textile wastewaters.

Electrochemical synthesis, characterization of Ir-Zn containing coordination polymer, and application in oxygen and glucose sensing.

Controllable synthesis, morphology evolution and electrochemical properties of LiFePO4 cathode materials for Li-ion batteries.

Nitro-substituted bishomocubanes: synthesis, characterization, and application as energetic materials.

Synthesis and electrochemical properties of silicon nanosheets by DC arc discharge for lithium-ion batteries.

Noble metal aerogels-synthesis, characterization, and application as electrocatalysts.

WOx nanotubes for photocatalytic application.

One-Pot Hydrothermal Synthesis of FeMoO₄ Nanocubes as an Anode Material for Lithium-Ion Batteries with Excellent Electrochemical Performance.

Electrochemical Synthesis of Magnesium Hexaboride by Molten Salt Technique.

Electrochemical properties of cobalt hydroxychloride microspheres as a new anode material for Li-ion batteries.

C nanosheets and microspheres as anode materials for lithium-ion batteries.

Luminescent terbium-containing metal-organic framework films: new approaches for the electrochemical synthesis and application as detectors for explosives.

Facile molten salt synthesis of Li2NiTiO4 cathode material for Li-ion batteries.

An inter-tangled network of redox-active and conducting polymers as a cathode for ultrafast rechargeable batteries.

C composite via a self-expanding process and its electrochemical performance as anode material for lithium-ion batteries.