Investigation of the N-butyl-N-methyl pyrrolidinium trifluoromethanesulfonyl-N-cyanoamide (PYR14TFSAM) ionic liquid as electrolyte for Li-ion battery

Article abstract of DOI:10.1016/j.electacta.2017.08.020

A new asymmetrical anion, trifluoromethanesulfonyl-N-cyanoamide (TFSAM^?), was paired with N-butyl-N-methyl pyrrolidinium (PYR₁₄⁺) to prepare PYR₁₄TFSAM. It has been investigated for Li-ion battery application and compared to its PYR₁₄⁺ analogs paired with either the dicyanamide anion (DCA^?) or other anions (i.e. bis(trifluoromethanesulfonyl) imide (TFSI^?), bis(fluorosulfonly)imide (FSI^?), trifluoromethanesulfonyl-fluorosulfonyl imide (FTFSI^?)). The conductivity of PYR₁₄TFSAM is not only higher than that of PYR₁₄TFSI, but also higher than that of PYR₁₄FTFSI with 3.8 mS cm^?1 at 20 °C and 12.6 mS cm^?1 at 60 °C. In addition, the ionic liquid does not crystallize and exhibits a viscosity similar to that of PYR₁₄FSI (and even lower above 30 °C, which also results in a higher conductivity at high temperature). Compared to PYR₁₄DCA, PYR₁₄TFSAM has a higher anodic stability, more compatible with state-of-the-art cathodes such as NCM, even though the PYR₁₄DCA electrolyte also allowed surprisingly good cycling results of NCM cathode considering its low anodic stability. PYR₁₄TFSAM also allows Li⁺ (de-)/insertion into graphite, using vinylene carbonate as additive. When used in conventional Li-ion electrolyte solvents, it leads to moderate conductivity (as compared with LiFSI or LiTFSI), although much higher than LiDCA. Additionally, it is shown that, even in EC/DMC-based electrolyte, LiTFSAM does not induce Al corrosion at 4.2 V.

Full text of DOI:10.1016/j.electacta.2017.08.020

Accepted Manuscript
Title: Investigation of the N-butyl-N-methyl pyrrolidinium
trifluoromethanesulfonyl-N-cyanoamide (PYR TFSAM)
1
4
ionic liquid as electrolyte for Li-ion battery
Authors: Jan-Philipp Hoffknecht, Mathias Drews, Xin He,
Elie Paillard
PII:
S0013-4686(17)31647-X
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.electacta.2017.08.020
EA 30024
To appear in:
Electrochimica Acta
Received date:
Revised date:
Accepted date:
9-5-2017
7-7-2017
5-8-2017
Please cite this article as: Jan-Philipp Hoffknecht, Mathias Drews, Xin He, Elie Paillard,
Investigation of the N-butyl-N-methyl pyrrolidinium trifluoromethanesulfonyl-N-
cyanoamide (PYR14TFSAM) ionic liquid as electrolyte for Li-ion battery,
Electrochimica Actahttp://dx.doi.org/10.1016/j.electacta.2017.08.020
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Investigation of the N-butyl-N-methyl pyrrolidinium
trifluoromethanesulfonyl-N-cyanoamide (PYR14TFSAM) ionic liquid as
electrolyte for Li-ion battery
a
a
a
a
Jan-Philipp Hoffknecht , Mathias Drews , Xin He , Elie Paillard *
a
Helmholtz Institute Münster – Forschungszentrum Jülich GmbH (IEK 12), Corrensstrasse 46, 48149
Münster, D-48149 Münster, Germany
e.paillard@fz-juelich.de
*
Corresponding author: e.paillard@fz-juelich.de
Abstract
-
A new asymmetrical anion, trifluoromethanesulfonyl-N-cyanoamide (TFSAM ), was paired with N-
butyl-N-methyl pyrrolidinium (PYR14+) to prepare PYR14TFSAM. It has been investigated for Li-ion
battery application and compared to its PYR14 analogs paired with either the dicyanamide anion (DCA-
+
-
-
)
or other anions (i.e. bis(trifluoromethanesulfonyl) imide (TFSI ), bis(fluorosulfonly)imide (FSI ),
-
trifluoromethanesulfonyl-fluorosulfonyl imide (FTFSI )). The conductivity of PYR14TFSAM is not only
-1
higher than that of PYR14TFSI, but also higher than that of PYR14FTFSI with 3.8 mS cm at 20°C and
-
1
1
2.6 mS cm at 60°C. In addition, the ionic liquid does not crystallize and exhibits a viscosity similar
to that of PYR14FSI (and even lower above 30°C, which also results in a higher conductivity at high
temperature). Compared to PYR14DCA, PYR14TFSAM has a higher anodic stability, more compatible
with state-of-the-art cathodes such as NCM, even though the PYR14DCA electrolyte also allowed
surprisingly good cycling results of NCM cathode considering its low anodic stability. PYR14TFSAM
+
also allows Li (de-)/insertion into graphite, using vinylene carbonate as additive. When used in
conventional Li-ion electrolyte solvents, it leads to moderate conductivity (as compared with LiFSI or
1

LiTFSI), although much higher than LiDCA. Additionally, it is shown that, even in EC/DMC-based
electrolyte, LiTFSAM does not induce Al corrosion at 4.2V.
Keywords: Pyrrolidinium; ionic liquid; electrolyte; Li-ion; Li metal; trifluoromethanesulfonyl-N-
cyanoamide; TFSAM; asymmetrical anion; lithium dicyanamide; LiDCA; LiTFSAM; ethylene
carbonate; dimethyl carbonate; vinylene carbonate; NCM; NMC; graphite.
1. Introduction
Li-ion batteries have become part of modern life and are now present in most portable electronic
devices. However, as their uses extend to larger applications such as electric vehicles (EVs) and
stationary storage for grid levelling, the question of their safety and sustainability must be addressed.
Indeed, state-of-the-art Li-ion electrolytes are mixtures of cyclic and linear alkyl carbonates and
contains LiPF
6
as lithium salt. While linear alkyl carbonates are highly volatile, flammable and have
decomposes at ca. 60°C in presence of water traces[2,3] or
very low flash points[1], LiPF
6
electrolyte components[4]. Moreover, it is readily hydrolyzed in ambient atmosphere to
hydrofluoric acid and thus cannot be recycled. Thus, alternatives have been sought for improving
the safety of Li-ion batteries and flame retardants[1,5,6], as well as higher flash point solvents such
as sulfones[7] and dinitriles[1,8] have been proposed. Lithium salts with higher stabilities toward
hydrolysis and temperature such as lithium bis(trifluoromethanesulfonyl)imide (LiTFSI)[9] or
lithium bis(fluorosulfonyl)imide (LiFSI)[10], shown in figure 1, have also been proposed. However,
the choice of lithium salt is, in fine, a compromise between many practical considerations such as
price, solubility, conductivity, anodic stability in the solvent mixture used, as well as the possibility
to form of a good solid electrolyte interphase (SEI) [11,12] at the anode (i.e. a passivation layer
needed for the operation of graphite electrodes). Moreover, the passivation of aluminum current
collectors is needed, for which LiTFSI is known to be detrimental, while LiFSI is
controversial[10,13].
2

One of the most promising approach for higher temperature applications of Li-ion batteries, such as EVs
and stationary storage (as the cooling of large batteries is more challenging than that of low power
batteries), is the use of ionic liquid (ILs)-based electrolytes. They have been proposed both for Li-metal
batteries, either as liquid[14,15] or as plasticizers for PEO-based electrolytes[16,17] and for Li-ion
batteries[18,19]. While their properties (protic, aprotic, hydrophobic hydrophilic, flammable or not)
-
depends on their constituents, those based one the TFSI anion and quaternary ammonium cations are
hydrophobic, possess high oxidative and thermal stability, negligible vapor pressure and flammability
-
and allow decent Li metal anode operation[14,20]. However, TFSI molecular weight induces high
viscosities at room temperature, which limits the conductivities of the resulting ionic liquids. ILs
-
-
incorporating smaller anions such as dicyanamide (DCA )[21–24] or bis(fluorosulfonly)imide (FSI )
18,25], shown on figure 1, allow obtaining lower viscosities and higher conductivities. In fact, the DCA-
[
anion leads to ILs with among the lowest viscosities and highest conductivities. However, replacing the
-
-SO
2
-CF
3
groups by -CN electron withdrawing substituents (as compared to the formula of TFSI , see
figure 1), if it allows decreasing the size of the anion, is still not enough for ensuring high oxidative
stability for a use in state-of-the-art Li-ion batteries (i.e. operating up to 4.2-4.3V). Moreover, the
-
stability of the DCA anion in reduction (and especially versus graphite anodes) is not well established,
although PYR14DCA has been proposed for Li metal batteries[22]. Nevertheless, even the low potential
LiFePO cathode does not lead to high efficiencies in PYR14DCA and it use seems, so far, limited to
4
electrochemical double layer capacitors operating at 2.6 V[24]. It could however be possible that Li-ion
batteries can operate upon a wider electrochemical window in this electrolyte, thanks to passivation
layers (SEI) formations that might form at the anode (and possibly cathode).
FIGURE 1
Several factors influence the performance of ILs. First of all, their liquid range depends on their ability
to crystallize which can lead to either low melting point or amorphous ILs. It depends on the strength of
the interactions between anion and cation, their symmetry, as well as their conformational flexibility
(
i.e. having different conformers of low energy)[26,27] as these impact the formation of crystalline
phases which require a regular space arrangement of their constituent with well-defined orientation and
3

conformation. If the use of asymmetrical cations is relatively common[28], ILs incorporating low
symmetry anions have also been proposed [29–31] for decreasing further the asymmetry within ILs and
suppressing their crystallization. This can also be achieved by mixing different ionic liquids [32],
especially with those which do not crystallize[33]. In particular, the fluorosulfonyl-
-
trifluoromethanesulfonyl imide anion (FTFSI ), shown in figure 1, is an asymmetrical hybrid between
-
-
-
the FSI and TFSI anion and displays a higher thermal stability than FSI . In addition, its ILs have
-
viscosities and conductivities closer to those of the FSI analogs with the main advantage of being
amorphous, thus liquid and conductive on a wide temperature window, thanks to its asymmetry[31].
In this work, we examine the physico-chemical properties of the trifluoromethanesulfonyl cyano amide
-
-
-
(
TFSAM ) anion, shown in figure 1, which is an asymmetrical hybrid between the TFSI and the DCA
anions. This anion was first synthesized and reported by Shaplov et al.[34] and various ILs were
prepared with this anion and showed an improved anodic stability as compared with DCA analogs, high
conductivities and high thermal stabilities. Thus, we compare here the electrochemical performance of
-
-
the DCA and TFSAM anions, in the IL form, combined with the N-methyl-N-butyl pyrrolidinium
cation (PYR14+) for a use in Li-ion battery and compare their properties with several PYR14-based ILs
incorporating other known promising anions.
2
. Experimental
.1 Chemicals used
Acetonitrile (99.9%, Merck), Potassium (2,2,2- trifluoromethylsulfonyl)-N-cyanoamide (KTFSAM)
97%, PROVISCO), N-methyl pryrrolidine (99.5%, Aldrich), , N-methyl pyrrolidone (99.5% Acros),
PYR14DCA (99.9%, Solvionic SA) and NaDCA (99%, Aldrich), Lithium 4,5 dicyano-2-
trifluoromethyl)imidazole (LiTDI) (Solvionic S.A.) and PYR14TDI (Solvionic S.A.) were used as
2
(
(
received. Bromobutane (99% Aldrich), N-methyl pyrrolidone (99.5% Acros) were distillated before use.
Dichloromethane (99.9%, Merck) was freshly purified by a PureSolv™ solvent purification system
(
INNOVATIVE TECHNOLOGY)
.2 Purity analysis
2
4

Ionic Chromatography (IC) was used to investigate the possible contamination of salts and ILs by
chloride or bromide impurities, using an 850 Professional IC (Metrohm) equipped with a ASUpp7-
2
50x4.0 mm column. A 3.6 mM Na
The samples were diluted (1:10 with water or ACN). Quantification was done via standard solutions (1,
, 5 and 10 ppm of chloride and bromide) and the measurement were run at 65°C.
2
CO solution in water with 45% acetonitrile was used as an eluent.
3
2
NMR: NMR experiments were performed on BRUKER AVANCE I (300 MHz) and BRUKER
AVANCE III (400 MHz) devices. All samples were prepared under argon atmosphere in deuterated
solvents. The solvent signals were used as reference. The data was analyzed with the TOPSPIN™
software (BRUKER). Characterization was done by means of chemical shift (in ppm), coupling constant
(
J in Hz), intensity and multiplicity (s singlet, d duplet, dd duplet in duplet, t triplet, q quartet).
.3 Preparation of lithium dicyanamide (LiDCA)
Following a procedure reported in the literature [35], the reaction of sodium dicyanamide (10.0 g,
12 mmol, 1.0 eq) and lithium chloride (4.75 g, 112 mmol, 1.0 eq) in THF (120 mL) at 90 °C for 3 d
2
1
yielded a colorless solid. The crude product was purified by multiple recrystallization from acetonitrile
to gain the title compound (5.90 g, 80.8 mmol, 72%). IC results: 0 ppm bromide, 9.7 ppm chloride.
2.4 Preparation of N-butyl-N-methyl pyrrolidinium (PYR14Br)
PYR14Br was prepared according to an already published procedure [28] using freshly distillated
bromobutane and N-methyl pyrrolidine in stoichiometric amounts in ethyl acetate. The reaction vessel
was maintained at 0°C for 2 hrs, left to reach room temperature and further stirred for 24h. The white
crystals were then filtrated and washed 3 times in fresh ethyl acetate. The residual solvent was
evaporated and the PYR14Br compound was then dried at 80°C for 24h under vacuum.
2.5 Preparation of N-butyl-N-methyl-pyrrolidinium (2,2,2- trifluoromethylsulfonyl)-N-cyanoamide
(
PYR14TFSAM)
PYR14TFSAM, was prepared according to the procedure published by Shaplov et al.[34], modified as
follows: N-butyl-N-methyl-pyrrolidium bromide (20.0 g, 90.0 mmol, 1.0 eq) was dissolved in deionized
water and cooled down to 0 °C. Potassium (2,2,2- trifluoromethylsulfonyl)-N-cyanoamide (21.0 g, 99.4
5

mmol, 1.1 eq) was added dropwise as aqueous solution. After stirring at 0 °C for 1 h, the reaction was
allowed to warm up and was stirred for additional 12 h at r.t. Thereafter, the reaction mixture was
extracted with CH
2
Cl
2
(3 x 50 mL). The combined organic layers were washed four times with
SO and the solvent was removed under reduced
equivolumes of deionized water, dried over Na
2
4
pressure. The obtained light yellow liquid was redissolved in acetonitrile and stirred with activated
carbon for 18 h. The suspension was filtrated and the solvent removed in vacuum. The product was
obtained as colorless liquid (25.0 g, 79.3 mmol, 88% yield).
1H NMR (400 MHz, CD
J=7.40 Hz, CH -CH -CH
.94 (3H, s, N-CH ), 3.23 (2H, t, J=8.52 Hz, CH
13C NMR (100 MHz, CD
CN) [ppm]: δ = 13.40 (1C, s), 19.92 (1C, t, J=1.50 Hz), 21.87 (1C, s), 25.80
1C, s), 48.67 (1C, d, J=3.87 Hz), 64.55 (1C, t, J=2.92 Hz), 64.80 (1C, d, J=2.38 Hz).
CN) [ppm]: δ = 0.93 (3H, t, J=7.36 Hz, CH
-CH ), 1.70 (2H, m, J=7.97 Hz, CH -CH -CH
-CH -CH -CH
-CH
-CH
-CH ), 1.34 (2H, q,
3
2
2
2
3
2
2
2
3
2
2
2
-CH ), 2.13 (4H, s, CH2-ring),
3
2
3
2
2
2
3
), 3.40 (4H, d, J=3.20 Hz, N-CH2-ring).
3
(
-
-
IC (0.7 mL / min, 65 °C): 2.9 ppm Br , 6.5 ppm Cl .
.6 Electrode preparation
2
NCM electrodes: Using a zirconia ball-milling jar and zirconia balls, 1.6 g of LiNi0.5Co0.2Mn0.3
NCM) active material (NCM523, Cabot), 200 mg of carbon black (Super C65, IMERYS) and 200 mg
of PVdF binder (Solef 6020, Solvay), i.e. 4 g of a 5 wt% solution in N-methyl pyrrolidone (NMP) and
O
2
(
2
mL additional NMP were mixed using a planetary ball milling machine (Fritsch Pulverisette 7)
balls/mAM = 10/1) to prepare the slurry (30 min rotation, 600 rpm, 10 min rest, 3 repetitions) with a
0/10/10 mass ratio of active material/carbon black/binder. The slurry was coated on an aluminum foil
(
m
8
previously washed with ethanol and dried. After drying the electrode sheets at 100 °C for 24 h, electrodes
12 mm Ø) were punched and dried in vacuum at 120 °C for 24 h. The mass active loading of the
(
electrodes was ca. 2 mg cm-2
Graphite electrodes: In a 40 mL vial, 3.6 g of graphite active material (SMG-A4, Hitachi Chemicals),
200 mg of carbon black (Super C65, IMERYS) and 200 mg of carboxymethyl cellulose (CMC) binder
(
WALOCEL CRT 2000 PPA 12, Dow Wolff Cellulosic), i.e. 10 g of a 2 wt% solution in water, were
6

weighed in. The mixture was stirred on a magnetic stirrer for 1 h then dispersed using a dispersion
system (Dispermat™ VL, VMA GmbH) (10000 rpm, 1 h). The slurry, with a 90/5/5 mass ratio of active
material/carbon black/binder, was coated on a copper foil previously washed with ethanol. After drying
the electrode sheets at 80 °C for 24 h, electrodes (12 mm Ø) were punched and dried in vacuum at 80 °C
for 48 h. The active mass loading of the electrodes was ca. 2 mg cm-2
2.7 Electrolyte preparation
All electrolytes were prepared under argon atmosphere, using dry lithium salts and ionic liquids (dried
-7
at 60°C for respectively 24h and 72h, using a turbo molecular pump (vacuum reaching 10 Bar)) and
solvents (as received and kept in a glove box with H O and O levels below 1 ppm). The desired amount
2
2
of salt was weighed in volumetric flask, pre-dissolved in half of the solvent, before completely filling
the flask. In the case of electrolytes based on ILs, a light heating at 50 °C was applied.
2
.8 Electrochemical testing
Cells preparation: For the electrochemical characterizations, three electrode SWAGELOK® cells were
used. The cell assembly was done in a glovebox under argon atmosphere with H O and O content below
ppm. Metallic lithium foil (99.99%, Rockwood Lithium) was used as counter and reference electrodes
CE and RE). Depending on the investigation, different working electrodes were used: nickel, platinum,
2
2
1
(
aluminum, NCM and graphite. The WE and CE were separated by either a FREUDENBERG FS2226
fleece, 13 mm Ø, six layers (for EC/DMC-based electrolytes) or a WHATMAN® GF/D glass, 13 mm
Ø, one layer fiber (for ionic liquid based electrolytes) soaked with electrolyte (120 μL). On the RE side,
a 10 mm Ø separator soaked with electrolyte (80 μL) was used. When using ionic liquid electrolytes,
vacuum was applied to the cells before sealing them, to ensure good electrode wetting of electrodes and
separator.
Linear voltamperaometry: To test the cathodic (resp. anodic) stability, a freshly polished Ni (resp. Pt)
-1
electrode was polarized from the open-circuit potential to -0.5V (resp. 6.0V) at 0.1 mV s .
Cyclic Voltammetry (CV): For lithium plating/stripping, the CV scans were done using a Ni electrode
-1
as WE. Starting at open circuit potential (OCP), the potential was scanned at 0.1 mV s down to -0.5 V
7

+
+
vs. Li /Li, then reversed to 0 V vs. OCP. Li insertion into graphite electrodes was done starting at OCP,
-1
+
the potential was scanned at 0.05 mV s down to 0 V vs. Li /Li, then reversed to 0 V vs. OCP.
Chronoamperometry: To investigate aluminum pitting corrosion, chronoamperometry was performed
using aluminum current collectors as WEs. After a linear sweep voltammetry from OCP to 4.2 V vs.
+
Li /Li, the potential was hold at 4.2 V for 12 h.
The conductivity of the electrolytes was measured using an impedance-based conductimeter (MCS10
Bio-logic). The conductivity cells were calibrated before use with a standard KCl solution. The cells
were sealed in a glovebox. The temperature was varied with 20 min ramps every 5°C (followed by
stabilization at each temperature, monitored by conductivity measurements), in decreasing steps. For
PYR14TDI, the temperature was ramped in both directions to acquire conductivity in the (metastable)
liquid state below the melting point.
2.9 Thermal analysis
Differential scanning calorimetry (DSC) was done on a Q2000 DSC (TA instruments) calibrated with
-1
indium, under He flow, at 10°C min in aluminum (TZero™) pans sealed in dry room with dew point
below -50°C. The glass transition temperatures (Tg) were taken at the inflexion points of the baselines.
Thermogravimetric analysis (TGA) was done on a Q5000 TGA (TA instruments) under N
2
and artificial)
air (i.e. a mixture of O
2
2
and N ). For TGA experiments, the samples were exposed to ambient air prior
to the measurements.
2.10
Viscosity and density
-1
Viscosity was measured using a rheometer (Anton Paar, MCR 301) at constant shear rate (1000 s 9
located in a dry room with dew point below -50°.
Density was measured using a Density Meter (METTLER TOLEDO DE40) located in a dry room with
dew point below -50°C.
2.11
Fitting of curves
8

Data analysis was done using a MATLAB R2016a version. Vogel Tammann Fulcher (VTF) fits for
conductivity and viscosity data and the corresponding error values were calculated using a least square
method.
3
Results and discussionThermal characterizations
The thermal properties measured for of PYR14TFSAM and PYR14DCA are summarized in table 1.
The TGA traces of PYR14DCA and PYR14TFSAM under N
2
and artificial air are shown in figure 2a.
in both cases. PYR14TFSAM exhibit a
The weight losses start at similar temperatures in air and N
2
higher decomposition temperature, at ca. 283°C, while that of PYR14DCA is lower, starting at ca. 238°C.
Both values are consistent with the reported values for PYR13TFSAM (285°C)[34] and PYR12DCA
(
250°C)[21]. The thermal stability is, in addition, higher than the reported values for PYR14FTFSI in
and O (256°C and 202°C respectively)[31]. The values are, however, lower than that of PYR14TFSI
above 300°C in N [36]). However, even for PYR14DCA, the stability is significantly higher than the
upper limit for any carbonate-based electrolytes.
N
2
2
(
2
FIGURE 2
The DSC thermograms of PYR14DCA and PYR14TFSAM are shown in Figure 2c. Both compounds
remain amorphous on the cooling ramp (not shown), exhibiting only their Tg. On the heating ramps,
they show similar glass transition temperatures. In fact, the Tg of PYR14TFSAM (-97.7°C) is only
slightly higher than that of PYR14DCA (-98.9°C), (i.e. slightly higher than reported previously[21]),
-
but significantly lower than that of PYR14TFSI (-85°C)[32]. For the TFSAM anion, the Tg contrasts
with that reported for PYR13TFSAM (-12°C), which is excessively high for ILs with high conductivity
and low viscosity and might have been an artifact (or a typo), PYR13+-based ionic liquids having
+
usually lower viscosity and Tg than their PYR14 counterparts. Moreover, for PYR14DCA, a clear
melting transition can be observed. On the heating ramp, a cold crystallization starts at ca. -55°C,
followed by an endothermic melting peak starting slightly before −15.5 °C with an corresponding
9

-
1
enthalpy ΔHm = 110.0 J g . This contrasts with previous report as only a weak transition was
reported around -55°C[21]. It could be due to the fact that ionic liquids often present different phase
behavior, depending on the thermal history, sample size or purity[37] and can remain amorphous
below their melting point for extended period of time. In fact, their tendency to stay in a metastable
liquid state (supercooling) explains the difference of temperature between (cold) crystallization and
melting observed for PYR14DCA. It is thus the first report concerning the crystallization and melting
of PYR14DCA. PYR14TFSI phase behavior has been extensively investigated and its melting point is -
3°C[37]. On the contrary, PYR14TFSAM does not exhibit any crystallization or melting, which
extends its liquid range down to its Tg, as for PYR14FTFSI[31]. In both cases, enhanced asymmetry
results in amorphous ILs on a wide temperature range.
3.2 Density, Viscosity and Conductivity
The variation of the densities for PYR14TFSAM and PYR14DCA with temperature have been measured
and are compared with those of PYR14TFSI in figure 2b. As can be seen, the density of PYR14TFSAM
stands, logically, in between those of the two other ILs. The variation with temperature is linear and can
be fitted with equation (1) and the parameters reported in table 2.
휌 = 휌 + 푎푇
(1)
0
The viscosities of PYR14DCA and PYR14TFSAM are compared with that of PYR14TFSI[28],
PYR14FSI[38] and PYR14FTFSI[31] in figure 2d. PYR14TFSI exhibits, by far, the highest viscosity,
while PYR14DCA has the lowest on the whole temperature range. PYR14TFSAM, however, performs
not only better than PYR14FTFSI, but similarly to PYR14FSI, especially above 30°C, as its viscosity
becomes lower. The data have been fitted with the Vogel-Tammann-Fulcher (VTF) equation (2), with
X = η (dynamic viscosity), X
∞
= η
∞
and the corresponding parameters are reported in table 2.
퐵
−ꢀ
(ꢀ
)
푋 = 푋 푒
ꢁ
(2)
∞
The conductivities of the pure ILs are reported on figure 3a. and compared to those incorporating other
-
-
-
-
-
anions (FSI , FTFSI , TFSI , and the 4,5 dicyano-2-(trifluoromethyl)imidazole anion (TDI )). TDI is an
1
0

aromatic anion, which incorporates two –CN substituents as well as one -CF group[39,40] and it has
3
been reported recently as high temperature alternative to PYR14TFSI, despites it high melting point[41].
PYR14TFSAM has a conductivity halfway between PYR14DCA and PYR14TFSI, whereas PYR14FTFSI
is closer to PYR14FSI than PYR14TFSI. However, given the much higher conductivity of PYR14DCA
-
1
-1
(
10.3 mS cm at 20°C) as compared with PYR14FSI, PYR14TFSAM still reaches 3.8 mS cm at 20°C,
which is higher than PYR14FTFSI. In fact, it can be seen that the conductivities of PYR14FSI are higher
than those of PYR14TFSAM in the temperature range investigated, but that the curves seem to cross at
higher temperature and the conductivities at 60°C are close, in accordance with the lower viscosity of
PYR14TFSAM at higher temperatures. The conductivities of PYR14DCA and PYR14TFSAM have been
fitted by equation (2) with X = σ, X
reported in table 2. The ‘ideal’ glassy transition temperature determined for viscosity and conductivity,
and T’ , are very close to each other (and from one IL to the other) and roughly 20°C below the
∞
= σ
∞
, B = B’ and T
0
= T ’ and the corresponding parameters are
0
T
0
0
experimental Tgs. For PYR14TFSI, the reported T’ is 182 K [30], which agrees with its higher viscosity
0
and higher Tg.
If we compare the curve of PYR14TDI, the conductivities are much lower and there is an obvious gap in
conductivity measurements as the IL crystallizes both during cooling ramp and heating ramp (on a
quenched sample). However, above its melting point (at 48°C), PYR14TDI provides higher
conductivities than PYR14TFSI. However, in this case, the slope of the conductivity curve is even more
marked than for PYR14TFSAM, which results in lower conductivity at low temperatures, probably
linked to the viscosity increase. In this case, not only the –CN groups, but also the rigid and planar
structure might affect the low temperature viscosity and conductivity.
1
1

FIGURE 3
For comparison, the conductivities of LiDCA and LiTFSAM in conventional 1M EC/DMC 1:1, wt
mixtures have been measured and are shown in figure 3b. As can be seen, the order of conductivity is
-1
inversed, with a strong advantage for LiTFSAM with 3.7 mS cm at 20°C, whereas LiDCA only leads
-
1
-
to 1.0 mS cm . In this case where the viscosity induced by the anion is not too severe, DCA does not
present any conductivity advantage, which can be linked, a priori, to its lower dissociation whereas the
small size of the anion bring significant advantage in the IL form..
On the other hand, the conductivity reached by LiTFSAM is higher, which can be attributed to the
enhanced charge delocalization brought by the SO
2
-CF group (also responsible for the increased anodic
3
stability) and the resulting increase of ionic dissociation. The conductivity of LiTFSAM is, however,
-1
lower than many other Li salts. In fact, LiTDI, at 20°C, has a conductivity of 6.7 mS cm , LiTFSI of
-
1
-1
9
.0 mS cm [40] in similar conditions and LiFSI, at 0.85 M and 25°C of 12 mS cm [13]. It shows that,
in ILs, the size of the anion and the viscosity induced has a stronger effect on conductivity than lowering
the coordination capability of the anion, which limits the possibility of improving conductivity by use
-
of large anions. The comparison with the TDI anion shows that, by use of aromatic charge
delocalization, assisted by two –CN groups and one -CF electron withdrawing groups it is possible to
3
reach a higher dissociation of the Li salt in EC/DMC, but that a high molecular weight, symmetry and
rigidity are not favorable for a use as IL at moderate temperature, given the high viscosity and melting
point.
The Walden plots of PYR14DCA, PYR14TFSI and PYR14TFSAM are shown in figure 3d., to get a better
idea of the relation between conductivity and viscosity[42,43]. It is often considered that the ILs that
fall well below the ‘ideal KCl line’ are more associated than those falling closer (thus a lower
conductivity at similar viscosity) and those falling above this line are sometimes qualified as ‘superionic
liquids’ [44], the position of the curve defining the ‘ionicity’ of an electrolyte. It is clear that PYR14DCA
falls closer to the line than PYR14TFSI. PYR14TFSAM seems to have an ‘ionicity’ comparable to that
of PYR14TFSI at low fluidity, which, however, gets higher as the fluidity increases (at higher
temperatures). In fact, its evolution is parallel to the ‘ideal KCl line’, while the two other salts get further
1
2

within increasing fluidity. It indicates that its ‘ionicity’ is rather constant and that its viscosity evolution
mostly explains its conductivity behavior.
It is worth noticing that the trend for ‘ionicity’ does follow the opposite to that of dissociation in
conventional EC/DMC solvent. It seems that, rather than the decrease of the PYR14+-anion interaction
-
-
going from DCA to TFSI , the size of the anion play a paramount role in ‘ionicity’, besides the effect
on viscosity. This was already noticed for some BF -based ionic liquids, which show a ‘superionic’
4
character[44][45], whereas LiBF
mixtures (as compared with LiPF
4
is not known for its good dissociation in conventional EC/DMC
or LiTFSI).
6
When the Li salt is introduced into the ILs, the conductivity drops, as the interaction between the anion
+
+
and Li is stronger than that between PYR14 and the anion. This is observed on figure 3c. for both
PYR14DCA and PYR14TFSAM. However, the conductivity drop, in the case of PYR14TFSAM is more
marked in the presence of LiTFSAM. In particular, the PYR14DCA electrolyte show 37% and 50%
conductivity drops at 0.4M and 0.6M respectively, while for PYR14TFSAM, the drops are 51% and 66%
at 20°C. These conductivity drops decrease with increasing temperature and the conductivities of the
-1
-1
0
3
.4M electrolytes, at 60°C are 7.4 mS cm for PYR14TFSAM and 19.9 mS cm for PYR14DCA.
.3 Electrochemical stability windows
FIGURE 4
The electrochemical stability window (ESW) of PYR14TFSAM and PYR14DCA electrolytes are shown
figure 4. As can be seen, the anodic stability of PYR14DCA is similar to that published for ethyl-methyl
-
+
imidazolium DCA [21], with a decomposition at ca. 3.5 V vs Li /Li, which is, a priori, too low for most
Li-ion battery application. The current do not increase significantly and decreases at high potential,
which probably indicates (partial) passivation of the electrode. The stability in reduction is also quite
low with reduction reaction starting at ca. 1.5V.
PYR14TFSAM, on the contrary, shows excellent stability in oxidation, with negligible current up to 4.5V
+
vs Li /Li, which shows that the replacement of one -CN group by a -SO
2
-CF allows significantly
3
lowering the HOMO of the anion and increase both the anodic stability and the dissociation in EC/DMC.
1
3

The cathodic stability of PYR14TFSAM is also rather low, with a decomposition starting even before
+
that of PYR14DCA, at ca. 1.9 V vs Li /Li.
The addition of the Li salt at 0.4M has a positive effect for both ILs concerning the cathodic stability,
as the extensive reduction currents convert into small peaks, which indicates that surface reactions take
place and that the passivation of the electrode, limits further electrolyte reactivity. This is known that
the Li salt plays a paramount role in SEI formation in ILs[46] and allow, for instance using some ILs/Li
salt mixtures for Li plating [47,48] with ILs which, alone, would decompose prior to Li deposition. In
the case of PYR14TFSAM, the current is then rather low until Li plating occurs, although a small
reduction wave can be seen at ca. 0.4V. In the case of PYR14DCA, further decomposition start slightly
+
below 1.0 V vs Li /Li. There is no marked effect on the stability in oxidation linked to the Li salt and
oxidation currents are overall increased, including, for the PYR14DCA electrolyte, a stronger
background current from OCP.
The use of vinylene carbonate (VC), a well-known SEI forming additive for conventional [49] and IL-
based [50] electrolytes has a further effect on decreasing reduction currents, for both ILs. In the case of
+
PYR14TFSAM, after a small peak starting at ca.1.6 V vs Li /Li, the background currents, prior to Li
-2
deposition are lower than 2 µA cm . They are higher for the VC-containing PYR14DCA electrolyte and,
+
in addition, a further decomposition step occurs at ca. 100 mV vs Li /Li.
Thus, it seems that the 0.4 M LiTFSAM PYR14TFSAM electrolyte (with or without VC) is a good
candidate for Li-ion or Li metal application combining at the same time a 4.5V anodic stability and good
passivation properties in reduction. However, as in Li-ion electrodes, surface groups, defects, current
collectors and higher surface areas can influence the practical ESW, further testing was conducted with
lithium metal, graphite and NCM cathodes.
Lithium plating /stripping
FIGURE 5
The plating/stripping behavior of Li metal has been tested in PYR14TFSAM with and without VC (the
curve obtained without LiTFSAM is shown for comparison). The voltammograms show plating and
1
4

stripping in both cases, with the reaching of the same steady-state (diffusion limited), current at ca. 150
-2
µA cm at the end of the plating step (which is above, for instance, the limiting current reported for
LiTDI/PYR14TDI at 60°C [[41]). On the reverse scan, in both cases, Li stripping occurs. However, it is
clear that, with the addition of VC, the Li stripping peaks is larger. By doing the ratio of the capacities,
coulombic efficiencies for Li plating/stripping of 37.8% without VC and of 55.7% with VC have been
calculated. The lithium plating/stripping conditions are not ideal, considering the very low reversing
+
potential of -500 mV vs Li /Li and the dendritic deposition of lithium occurring a priori when steady
state-current is reached (especially as Ni is not, in general, ideal to realize homogeneous Li plating[51]).
Nevertheless, it shows that the electrolytes are able to deposit Li metal even if the ptimization of the Li
metal performance is beyond the scope of the present study.
+
The insertion of Li into graphite has been probed by cyclic voltammetry at room temperature. Figure
5c. shows the four first voltammograms at 20°C. As can be seen, in the first cycle, a first reduction peak
+
starts slightly below 2.0V and another is visible at ca. 0.8V and overlaps with Li insertion into graphite.
+
On the reversed scan, significant amount of Li is de-inserted from the graphite electrode and it is
+
possible to see several peaks corresponding to the different stages of Li insertion into graphite. The
current density increase with cycling upon lithiation and the delithiation peaks also become more
defined. This could be due either to slow wetting of the electrode or to the establishment of a more
favorable SEI on the surface of graphite after the first cycle. The cycling of graphite seems possible
+
which confirms that a SEI allowing Li transport and preventing graphite exfoliation is formed.
Galvanostatic cycling of graphite was also performed at 40°C and in this case, the different insertion
stage of graphite are more marked and the curves of two consecutive cycles overlap, as illustrated in the
insert of figure 5c.
+
Li insertion into LiNi0.5 Co0.2Mn0.3
2
O (NCM523)
The 0.4M IL-based electrolytes have been used to cycle NCM cathodes at 40°C and the results are
shown in figures 5b. and 5d. Surprisingly, PYR14DCA allows cycling of NCM with a rather good
efficiency in the first cycle (87.4%), which contrasts with its low anodic stability. It is very likely that a
passivation layer forms and protect the cathode. In fact, the oxidation currents on Pt (see figure 4) are
1
5

rather low and in the form of peaks, which already suggested a passivation of the Pt electrode. On the
contrary, no feature is observed on the voltage profile that would correspond to extensive decomposition
of the electrolyte. While this relatively good cycling performance deserves further investigations, the
capacity decay with cycling is still rather marked, with a 14% decay in 30 cycles. Also, the efficiency
raises after the first cycle but it does not reach more than 97% before starting decaying after 15 cycles.
On the other hand, the PYR14TFSAM electrolyte exhibits a higher first cycle efficiency (92.6%) and the
efficiencies then raise to 98.4% after 15 cycle before slightly decaying. The capacity retention is much
improved, with a 90.5% capacity retention in the first 30 cycles. Several phenomena can influence the
cycling stability of NCM half-cells such as the loss of active NCM, increase of electrolyte (and/or SEIs)
+
resistance as well as slower Li diffusion in the SEIs. The voltage profiles, shown in inserts in figures
-1
5
b and d, overlap up to ca. 90 mAh g during charge in both cases, as the equivalent series resistance
ESR) stays constant over cycling (i.e. no marked increase of electrolyte and SEI resistance). Also, a
loss of active NCM would result in slope differences from the beginning of the charge. Thus, it is likely
(
-1
+
that the slope differences, above 90 mAh g , result from an increase of Li mass transport limitation in
some areas of the cells, either at the Li metal electrode or at the cathode. During discharge, differences
between voltage plateaus can be explained by the lower state of charge reached at the end of the charge,
in both cases. However, at the end of the plateau, the voltage curves stay parallel in the case of
PYR14TFSAM, which shows that the NMC electrode is fully lithiated at this point. On the other hand,
+
for PYR14DCA, the slope becomes less steep with cycling, which indicate Li mass transport limitation
during NCM lithiation as well. It is thus likely, that in both cases the Li metal interface evolves toward
+
slower Li transport (as seen in charge, corresponding to Li metal plating and NCM delithiation), but
during discharge, only the PYR14DCA cell evolve toward slower mass transport, which is a clue of
passivation layer formation at the cathode with this electrolyte.
such as Mn dissolution, especially at 40°C). Nevertheless, these results, although preliminary, do not
show extensive electrolyte decomposition in neither electrolytes. It is, moreover, clear that the higher
oxidative stability of PYR14TFSAM allows better cycling performance, as compared with PYR14DCA,
but the excellent results obtained with PYR14DCA contrast with its low anodic stability on Pt. This
1
6

promising result should be confirmed in full Li-ion cells as the Li metal counter electrode, especially at
moderate rate, might influence the performance due to the production of reduction products in the
electrolyte, which can diffuse to the cathode and be oxidized [52].
3.4 Al current collector corrosion
While TFSI-based ILs have been shown to prevent, to certain extend, Al current collector corrosion[53],
-
it is still an issue with LiTFSI in common alkyl carbonate mixture or in FSI containing ILs [54]. Thus,
LiTFSAM has been tested in EC/DMC for Al corrosion by chronoamperometry. Figure 6 shows the
comparison between LiTFSI and LiTFSAM in EC/DMC using a piece of Al current collector at 4.2V
+
vs (Li /Li). As can be seen, the current increase rather quickly in the case of LiTFSI, whereas it decreases
in the case of LiTFSAM (it increases after few hours, probably due to the Li metal counter electrode
used in the experiment). These result is of high practical interest as Li salts which do not induce Al
corrosion in EC/DMC, corrode even less in IL-based electrolyte including the same anion.
FIGURE 6
Conclusion
We presented an evaluation of PYR14TFSAM in view of use in Li-ion batteries with a focus on the
comparison with its parent ILs (PYR14DCA and PYR14TFSI), as well as other reported PYR14-based
ILs. It was shown that TFSAM possesses a higher thermal and anodic stability as compared with
PYR14DCA PYR14FSI and PYR14FTFSI and that its conductivity is higher than that of PYR14TFSI and
PYR14FTFSI at room temperature (and close to that of PYR14FSI at 60°C). PYR14TFSAM does not
crystallize, which extends its possible use to low temperature applications. The presence of LiTFSAM
in the electrolyte considerably extend its ESW in reduction and allows deposition of Li metal. The
addition of VC as SEI forming additive improves further the SEI formation, allowing an improvement
+
of the lithium plating/stripping efficiency as well as Li (de)-insertion into graphite at room temperature
or at 40°C. Finally, cycling of NCM Li-ion cathode was demonstrated at 40°C in PYR14TFSAM-based
electrolyte but also, surprisingly (although with lower performances) in PYR14DCA-based electrolyte.
Acknowledgements:
1
7

The research presented has been partly funded by the ‘SPICY’ project funded by the European Union’s
Horizon 2020 research and innovation program under grant agreement N° 653373. The research
presented here will be the base of further investigations which will be done within the GrEEn project,
funded by the Nordrhein-Westfalen region of Germany. Moreover, E.P. would like to thank Dr.
Sébastien Fantini and the company Solvionic S.A. for providing the PYR14TDI and LiTDI materials free
of charge. Dr. G. Appetecchi (ENEA, Italy) is also gratefully acknowledged for providing the data on
PYR14TFSI from ref [28].
References
[
[
[
1]
2]
3]
P. Isken, C. Dippel, R. Schmitz, R.W. Schmitz, M. Kunze, S. Passerini, M. Winter, A. Lex-
Balducci, High flash point electrolyte for use in lithium-ion batteries, Electrochim. Acta. 56
(
2011) 7530–7535. doi:10.1016/j.electacta.2011.06.095.
C.G. Barlow, Reaction of water with hexafluorophosphates and with Li
bis(perfluoroethylsulfonyl)imide salt, Electrochem. Solid-State Lett. 2 (1999) 362–364.
doi:10.1149/1.1390838.
H. Yang, G. V. Zhuang, P.N. Ross, Thermal stability of LiPF6 salt and Li-ion battery
electrolytes containing LiPF6, J. Power Sources. 161 (2006) 573–579.
doi:10.1016/j.jpowsour.2006.03.058.
[
[
4]
5]
C.L. Campion, W. Li, B.L. Lucht, Thermal Decomposition of LiPF[sub 6]-Based Electrolytes
for Lithium-Ion Batteries, J. Electrochem. Soc. 152 (2005) A2327. doi:10.1149/1.2083267.
K. Xu, S. Zhang, J.L. Allen, T.R. Jow, Evaluation of Fluorinated Alkyl Phosphates as Flame
Retardants in Electrolytes for Li-Ion Batteries: II. Performance in Cell, J. Electrochem. Soc.
150 (2003) A170. doi:10.1149/1.1533041.
[
[
6]
7]
X. Wang, E. Yasukawa, S. Kasuya, Nonflammable Trimethyl Phosphate Solvent-Containing
Electrolytes for Lithium-Ion Batteries: II. The Use of an Amorphous Carbon Anode, J.
Electrochem. Soc. 148 (2001) A1066. doi:10.1149/1.1397774.
C.A. Xu, K. Angell, High Anodic Stability of a New Electrolyte Solvent: Unsymmetric
Noncyclic Aliphatic Sulfone, J. Electrochem. Soc. 145 (1998) L70. doi:10.1149/1.1838419.
1
8

[
[
8]
9]
Y. Abu-Lebdeh, I. Davidson, High-Voltage Electrolytes Based on Adiponitrile for Li-Ion
Batteries, J. Electrochem. Soc. 156 (2009) A60. doi:10.1149/1.3023084.
M. Dahbi, F. Ghamouss, F. Tran-Van, D. Lemordant, M. Anouti, Comparative study of
EC/DMC LiTFSI and LiPF6 electrolytes for electrochemical storage, J. Power Sources. 196
(
2011) 9743–9750. doi:10.1016/j.jpowsour.2011.07.071.
[
10] H.B. Han, S.S. Zhou, D.J. Zhang, S.W. Feng, L.F. Li, K. Liu, W.F. Feng, J. Nie, H. Li, X.J.
Huang, M. Armand, Z. Bin Zhou, Lithium bis(fluorosulfonyl)imide (LiFSI) as conducting salt
for nonaqueous liquid electrolytes for lithium-ion batteries: Physicochemical and
electrochemical properties, J. Power Sources. 196 (2011) 3623–3632.
doi:10.1016/j.jpowsour.2010.12.040.
[
11] E. Peled, The Electrochemical Behavior of Alkali and Alkaline Earth Metals in Nonaqueous
Battery Systems—The Solid Electrolyte Interphase Model, J. Electrochem. Soc. 126 (1979)
2047. doi:10.1149/1.2128859.
[
[
12] J.R. Fong, R. von Sacken U. Dahn, Studies of Lithium Intercalation into Carbons Using
Nonaqueous Electrochemical Cells, J. Electrochem. Soc. 137 (1990) 2009–2013.
13] A. Abouimrane, J. Ding, I.J. Davidson, Liquid electrolyte based on lithium bis-fluorosulfonyl
imide salt: Aluminum corrosion studies and lithium ion battery investigations, J. Power
Sources. 189 (2009) 693–696. doi:10.1016/j.jpowsour.2008.08.077.
[
[
14] P.C. Howlett, D.R. MacFarlane, A.F. Hollenkamp, High Lithium Metal Cycling Efficiency in a
Room-Temperature Ionic Liquid, Electrochem. Solid-State Lett. 7 (2004) A97.
doi:10.1149/1.1664051.
15] E. Paillard, Q. Zhou, W.A. Henderson, G.B. Appetecchi, M. Montanino, S. Passerini,
Electrolytes with LiFSI FSI -Based 14 Electrochemical and Physicochemical Properties of PY
Electrochemical and Physicochemical Properties of PY 14 FSI-Based Electrolytes with LiFSI,
J. Electrochem. Soc. 156 (2009) 891–895. doi:10.1149/1.3208048͔.
[
16] J.H. Shin, W.A. Henderson, S. Passerini, Ionic liquids to the rescue? Overcoming the ionic
conductivity limitations of polymer electrolytes, Electrochem. Commun. 5 (2003) 1016–1020.
doi:10.1016/j.elecom.2003.09.017.
1
9

[
[
17] I. Osada, J. Von Zamory, E. Paillard, S. Passerini, Improved lithium-metal/vanadium pentoxide
polymer battery incorporating crosslinked ternary polymer electrolyte with N-butyl-N-
methylpyrrolidinium bis(perfluoromethanesulfonyl)imide, J. Power Sources. 271 (2014) 334–
341. doi:10.1016/j.jpowsour.2014.08.019.
18] M. Ishikawa, T. Sugimoto, M. Kikuta, E. Ishiko, M. Kono, Pure ionic liquid electrolytes
compatible with a graphitized carbon negative electrode in rechargeable lithium-ion batteries,
162 (2006) 658–662. doi:10.1016/j.jpowsour.2006.02.077.
[
[
[
19] M. Yamagata, K. Tanaka, Y. Tsuruda, Y. Sonem, S. Fukuda, S. Nakasuka, M. Kono, Ishikawa,
Electrochemistry, (2015).
20] L. Grande, E. Paillard, G.T. Kim, S. Monaco, S. Passerini, Ionic liquid electrolytes for Li-air
batteries: Lithium metal cycling, Int. J. Mol. Sci. 15 (2014) 8122–8137.
21] D.R. MacFarlane, S.A. Forsyth, J. Golding, G.B. Deacon, Ionic liquids based on imidazolium,
ammonium and pyrrolidinium salts of the dicyanamide anion, Green Chem. 4 (2002) 444–448.
doi:10.1039/b205641k.
[
[
22] H. Yoon, G.H. Lane, Y. Shekibi, P.C. Howlett, M. Forsyth, A.S. Best, D.R. MacFarlane,
Lithium electrochemistry and cycling behaviour of ionic liquids using cyano based anions,
Energy Environ. Sci. 6 (2013) 979–986. doi:10.1039/C3EE23753B.
23] S. Shen, S. Fang, L. Qu, D. Luo, L. Yang, S. Hirano, Low-viscosity ether-functionalized
pyrazolium ionic liquids based on dicyanamide anions: properties and application as
electrolytes for lithium metal batteries, RSC Adv. 5 (2015) 93888–93899.
doi:10.1039/C5RA17539A.
[
[
[
24] C. Wolff, S. Jeong, E. Paillard, A. Balducci, S. Passerini, High power, solvent-free
electrochemical double layer capacitors based on pyrrolidinium dicyanamide ionic liquids, J.
Power Sources. 293 (2015) 65–70. doi:10.1016/j.jpowsour.2015.05.065.
25] E. Paillard, Q. Zhou, W.A. Henderson, G.B. Appetecchi, M. Montanino, S. Passerini,
Electrochemical and Physicochemical Properties of PY[sub 14]FSI-Based Electrolytes with
LiFSI, J. Electrochem. Soc. 156 (2009) A891.
26] Q. Zhou, P.D. Boyle, L. Malpezzi, A. Mele, J.-H. Shin, S. Passerini, W.A. Henderson, Phase
2
0

Behavior of Ionic Liquid–LiX Mixtures: Pyrrolidinium Cations and TFSI – Anions – Linking
Structure to Transport Properties, Chem. Mater. 23 (2011) 4331–4337.
doi:10.1021/cm201427k.
[
[
27] I. Rey, P. Johansson, J. Lindgren, J.C. Lasse, Spectroscopic and Theoretical Study of ( CF 3
SO 2 ) 2 N - ( TFSI - ) and ( CF 3 SO 2 ) 2 NH ( HTFSI ), J. Phys. Chem. A. 102 (1998) 3249–
3258. doi:10.1021/jp980375v.
28] G.B. Appetecchi, M. Montanino, D. Zane, M. Carewska, F. Alessandrini, S. Passerini, Effect of
the alkyl group on the synthesis and the electrochemical properties of N-alkyl-N-methyl-
pyrrolidinium bis(trifluoromethanesulfonyl)imide ionic liquids, Electrochim. Acta. 54 (2009)
1325–1332. doi:10.1016/j.electacta.2008.09.011.
[
29] H. Matsumoto, N. Terasawa, T. Umecky, S. Tsuzuki, H. Sakaebe, K. Asaka, K. Tatsumi, Low
Melting and Electrochemically Stable Ionic Liquids Based on Asymmetric
Fluorosulfonyl(trifluoromethylsulfonyl)amide, Chem. Lett. 37 (2008) 1020–1021.
doi:10.1246/cl.2008.1020.
[
[
30] G.B. Appetecchi, M. Montanino, M. Carewska, M. Moreno, F. Alessandrini, S. Passerini,
Chemical-physical properties of bis(perfluoroalkylsulfonyl)imide-based ionic liquids,
Electrochim. Acta. 56 (2011) 1300–1307. doi:10.1016/j.electacta.2010.10.023.
31] J. Reiter, S. Jeremias, E. Paillard, M. Winter, S. Passerini, Fluorosulfonyl-
(
trifluoromethanesulfonyl)imide ionic liquids with enhanced asymmetry, Phys. Chem. Chem.
Phys. Phys. Chem. Chem. Phys. 15 (2013) 2565–2571. doi:10.1039/c2cp43066e.
32] M. Kunze, S. Jeong, E. Paillard, M. Winter, S. Passerini, Melting behavior of pyrrolidinium-
based ionic liquids and their binary mixtures, J. Phys. Chem. C. 114 (2010) 12364–12369.
33] M. Montanino, M. Moreno, F. Alessandrini, G.B. Appetecchi, S. Passerini, Q. Zhou, W.A.
Henderson, Physical and electrochemical properties of binary ionic liquid mixtures: (1-x)
PYR14TFSI-(x) PYR14IM14, Electrochim. Acta. 60 (2012) 163–169.
[
[
doi:10.1016/j.electacta.2011.11.030.
[
34] A.S. Shaplov, E.I. Lozinskaya, P.S. Vlasov, S.M. Morozova, D.Y. Antonov, P.-H. Aubert, M.
Armand, Y.S. Vygodskii, New family of highly conductive and low viscosity ionic liquids with
2
1

asymmetric 2,2,2-trifluoromethylsulfonyl-N-cyanoamide anion, Electrochim. Acta. (2015).
doi:10.1016/j.electacta.2015.02.228.
[
[
35] A.P. Purdy, E. Houser, C.F. George, Lithium dicyanamide, its reactions with cyanuric chloride,
and the crystal structures of LiN(CN)2(MeCN)2 and LiCN(C5H5N)2, Polyhedron. 16 (1997)
3671–3679. doi:10.1016/S0277-5387(97)00097-1.
36] J.H. Shin, E.J. Cairns, N-Methyl-(n-butyl)pyrrolidinium bis(trifluoromethanesulfonyl)imide-
LiTFSI-poly(ethylene glycol) dimethyl ether mixture as a Li/S cell electrolyte, J. Power
Sources. 177 (2008) 537–545. doi:10.1016/j.jpowsour.2007.11.043.
[
[
37] W. a Henderson, S. Passerini, Phase Behavior of Ionic Liquid LiX Mixtures: Pyrrolidinium
Cations and TFSI- Anions, Chem. Mater. 16 (2004) 2881–2885. doi:10.1021/cm049942j.
38] S. Passerini, E. Paillard, Q. Zhou, W.A. Henderson, G.B. Appetecchi, M. Montanino,
Electrolytes with LiFSI FSI -Based 14 Electrochemical and Physicochemical Properties of PY
Electrochemical and Physicochemical Properties of PY 14 FSI-Based Electrolytes with LiFSI,
J. Electrochem. Soc. 156 (2009) 891–895. doi:10.1149/1.3208048͔.
[
[
[
39] L. Niedzicki, S. Grugeon, S. Laruelle, P. Judeinstein, M. Bukowska, J. Prejzner, P.
Szczeciñski, W. Wieczorek, M. Armand, New covalent salts of the 4+ v class for Li batteries,
J. Power Sources. 196 (2011) 8696–8700. doi:10.1016/j.jpowsour.2011.06.030.
40] L. Niedzicki, E. Karpierz, M. Zawadzki, M. Dranka, M. Kasprzyk, A. Zalewska, M. Marcinek,
J. Zachara, U. Domańska, W. Wieczorek, Lithium cation conducting TDI anion-based ionic
liquids., Phys. Chem. Chem. Phys. 16 (2014) 11417–25. doi:10.1039/c3cp55354j.
41] A. Ochel, D. Di Lecce, C. Wolff, G.-T. Kim, D.V. Carvalho, S. Passerini, Physicochemical and
electrochemical investigations of the ionic liquid N-butyl -N-methyl-pyrrolidinium 4,5-
dicyano-2-(trifluoromethyl)imidazole, Electrochim. Acta. 232 (2017) 586–595.
doi:10.1016/j.electacta.2017.02.141.
[
[
42] E.I. Cooper, C. a Angell, W. Lafayette, E . L Cooper , CA . Angell / Ambient temperature
plastic crystal fast ion conductors, Science (80-. ). 19 (1986) 570–576.
43] C.A. Angell, (Invited) Ionic Liquids, Superionic Glasses, Quasi-Ionic Liquids, Quasi-Liquid
Ionics, All with High Conductivities but Some with Little Fluidity. Where does the Paradigm
2
2

End?, ECS Trans. 64 (2014) 9–20. doi:10.1149/06404.0009ecst.
[
[
44] C. Austen Angell, Y. Ansari, Z. Zhao, Ionic Liquids: Past, present and future, Faraday Discuss.
154 (2012) 9–27. doi:10.1039/C1FD00112D.
45] C. Schreiner, S. Zugmann, R. Hartl, H.J. Gores, Fractional walden rule for ionic liquids:
Examples from recent measurements and a critique of the so-called ideal KCl line for the
walden plot, J. Chem. Eng. Data. 55 (2010) 1784–1788. doi:10.1021/je900878j.
46] C. Liu, X. Ma, F. Xu, L. Zheng, H. Zhang, W. Feng, X. Huang, M. Armand, J. Nie, H. Chen,
Z. Zhou, Ionic liquid electrolyte of lithium bis(fluorosulfonyl)imide/N-methyl-N-
propylpiperidinium bis(fluorosulfonyl)imide for Li/natural graphite cells: Effect of
concentration of lithium salt on the physicochemical and electrochemical properties,
Electrochim. Acta. 149 (2014) 370–385. doi:10.1016/j.electacta.2014.10.048.
47] S. Ferrari, E. Quartarone, P. Mustarelli, A. Magistris, S. Protti, S. Lazzaroni, M. Fagnoni, A.
Albini, A binary ionic liquid system composed of N-methoxyethyl-N-methylpyrrolidinium
bis(trifluoromethanesulfonyl)-imide and lithium bis(trifluoromethanesulfonyl)imide: A new
promising electrolyte for lithium batteries, J. Power Sources. 194 (2009) 45–50.
doi:10.1016/j.jpowsour.2008.12.013.
[
[
[
48] J. Reiter, E. Paillard, L. Grande, M. Winter, S. Passerini, Physicochemical properties of N-
methoxyethyl-N-methylpyrrolidinum ionic liquids with perfluorinated anions, Electrochim.
Acta. 91 (2013) 101–107.
[
[
49] B. Simon, J.-P. Boeuve, US 5,626,981, 1997.
50] M. Holzapfel, C. Jost, P. Novák, Stable cycling of graphite in an ionic liquid based electrolyte.,
Chem. Commun. (Camb). (2004) 2098–2099. doi:10.1039/b407526a.
[
[
51] L. Grande, J. Von Zamory, S.L. Koch, J. Kalhoff, E. Paillard, S. Passerini, Homogeneous
Lithium Electrodeposition with Pyrrolidinium-Based Ionic Liquid Electrolytes, ACS Appl.
Mater. Interfaces. 7 (2015) 5950–5958. doi:10.1021/acsami.5b00209.
52] J. Von Zamory, M. Bedu, S. Fantini, S. Passerini, E. Paillard, Polymeric ionic liquid
nanoparticles as binder for composite Li-ion electrodes, J. Power Sources. 240 (2013) 745–752.
doi:10.1016/j.jpowsour.2013.04.127.
2
3

[
[
53] R.S. Kühnel, M. Lübke, M. Winter, S. Passerini, A. Balducci, Suppression of aluminum
current collector corrosion in ionic liquid containing electrolytes, J. Power Sources. 214 (2012)
178–184. doi:10.1016/j.jpowsour.2012.04.054.
54] E. Cho, J. Mun, O.B. Chae, O.M. Kwon, H.T. Kim, J.H. Ryu, Y.G. Kim, S.M. Oh,
Corrosion/passivation of aluminum current collector in bis(fluorosulfonyl) imide-based ionic
liquid for lithium-ion batteries, Electrochem. Commun. 22 (2012) 1–3.
doi:10.1016/j.elecom.2012.05.018.
2
4

Figure captions
-
Figure 1. Developed formulae of the bis(fluorosulfonyl) imide (FSI ), bis(trifluoromethanesulfonyl)
-
-
-
imide (TFSI ), dicyanamide (DCA ), fluorosulfonyl-trifluoromethanesulfonyl imide (FTFSI ) and
-
trifluoromethanesulfonyl cycanoamide (TFSAM ) anions.
Figure 2. (a) TGA traces and weight loss derivative curves of PYR14DCA and PYR14TFSAM at 10°C
-1
min under N and dry air. (b) Density of PYR14TFSI (taken from ref[30]), PYR14TFSAM and
2
-1
PYR14DCA. (c) DSC thermograms (heating ramp at 10°C min of PYR14TFSAM and PYR14DCA. (d)
Viscosities of PYR14DCA and PYR14TFSAM, compared with those of PYR14FSI[38], PYR14TFSI[38]
and PYR14FTFSI[31] (insert) VTF fit of the viscosities of PYR14DCA and PYR14TFSAM
Figure 3. (a) Conductivities of PYR14TDI, PYR14DCA, PYR14TFSAM compared with those of
PYR14TFSI [28], PYR14FSI [38] and PYR14FTFSI [31]. (insert) VTF fit of the conductivities of
PYR14DCA and PYR14TFSAM (b) Comparison of the conductivity in function of temperature for
LiDCA and LiTFSAM in 1M EC/DMC 1:1, wt electrolyte solutions. (c) Conductivity in function of
temperature for Li salts-containing electrolytes based on PYR14DCA and PYR14TFSAM (d) Walden
plot of PYR14DCA, PYR14TFSAM and PYR14TFSI (from ref. [28])
Figure 4. Linear sweep voltametries in PYR14DCA, PYR14TFSAM and their 0.4M electrolytes with
-1
corresponding Li salt. Scan rate: 0.1 mV s . WE (anodic scan): Pt, WE (cathodic scan): Ni, CE: Li, RE:
Li, 0.1 mV s-1
Figure 5. (a) Cyclic voltammograms of a Ni electrode in PYR14TFSAM, with and without addition of
-1
LiTFSAM and VC. Scan rate: 0.1 mV s . RE and CE: Li. (b) Cycling performance of NCM electrodes
in 0.4 M LiDCA in PYR14DCA. Inserts: Corresponding voltage profiles for selected cycles. (c) Initial
CV cycles of graphite electrode in 0.4 M LiTFSAM in PYR14TFSAM + 2 wt% VC. CE and RE: Li.
-1
Scan rate: 0.05 mV s . (Insert): Differential capacity plot of a similar cell cycled glavanostatically at
th
th
C/10 and 40°C. 5 and 6 cycles shown. (d) Cycling performance of NCM electrodes in 0.4 M
LiTFSAM in PYR14TFSAM.
2
5

Figure 6. Chronoamperometry at 4.2V of a Al electrode in 1M LiTFSI or 1M LiTFSAM in EC/DMC
:1, wt electrolytes. RE: Li: CE: Li.
1
2
6

Figure 1.
2
7

Figure 2.
2
8

Figure 3.
2
9