Comprehensive Insights into the Thermal Stability, Biodegradability, and Combustion Chemistry of Pyrrolidinium-Based Ionic Liquids

Article abstract of DOI:10.1002/cssc.201701006

The use of ionic liquids (ILs) as advanced electrolyte components in electrochemical energy-storage devices is one of the most appealing and emerging options. However, although ILs are hailed as safer and eco-friendly electrolytes, to overcome the limitations imposed by the highly volatile/combustible carbonate-based electrolytes, full-scale and precise appraisal of their overall safety levels under abuse conditions still needs to be fully addressed. With the aim of providing this level of information on the thermal and chemical stabilities, as well as actual fire hazards, herein, a detailed investigation of the short- and long-term thermal stabilities, biodegradability, and combustion behavior of various pyrrolidinium-based ILs, with different alkyl chain lengths, counteranions, and cations, as well as the effect of doping with lithium salts, is described.

Full text of DOI:10.1002/cssc.201701006

Accepted Article
Title: Comprehensive insights into the thermal stability, biodegradability
and combustion chemistry of Pyrrolidinium-based Ionic Liquids
Authors: Gebrekidan Gebreselassie Eshetu, Sangsik Jeong, Pascal
Pandard, Amandine Lecocq, Guy Marlair, and Stefano
Passerini
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201701006
Link to VoR: http://dx.doi.org/10.1002/cssc.201701006
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Comprehensive insights into the thermal stability, biodegradability and combustion
chemistry of Pyrrolidinium-based Ionic Liquids
[
a, b,c,d]
[a,b]
[e]
[e]
Gebrekidan Gebresilassie Eshetu,
Sangsik Jeong,
Pascal Pandard, Amandine Lecocq,
[
e]
[a,b]
Guy Marlair, Stefano Passerini*
Abstract
The use of ionic liquids (ILs) as advanced electrolytes components in
electrochemical energy storage devices is one of the most appealing and
emerging options. However, though ILs are hailed as safer and eco-friendly
electrolytes, overcoming the limitations imposed by the highly
volatile/combustible carbonate based electrolytes, the full scale and precise
appraisal of their overall safety levels under abuse conditions still need to be
fully addressed. With the aim of providing the entreated level of information on
the thermal and chemical stabilities as well as actual fire hazards, we
embarked on a detailed investigation of the short - and long - term thermal
stabilities, bio-degradability and combustion behaviour of various pyrrolidinium
([Pyr1A] +)-based ILs, enlisting different alkyl chain lengths, [Pyr1A] + (A=3-10),
counter-anions ([TFSI] /[FSI] /[BETI] ), cations (Pyr14+/Pyr12O1+) and the effect of
-
-
-
doping with Li salts (e.g. Li[TFSI]/[Pyr14] [TFSI]).
[
3,4,7][8–15]i
for such advanced applications.
Conventional lithium-
1
. Introduction
ion batteries utilizing solid electrodes and organic liquid
electrolytes suffer from potential thermal and chemical threats
associated to the thermal runaway hazard. This latter is
significantly related both in terms of occurrence and
The transition from fossil fuel-based to electrified means of
ground and even air transportation is a desirable future because
it provides innovative prospects to alleviate the overwhelming
challenges to the environment and its occupants (e.g., mankind),
such as emission of greenhouse gases and global warming but
consequences
to
electrolytes'
flammability/combustibility
[
16,17]
[
1–5]
properties.
Despite the remarkable progress in the research
also exhaustion of oil for chemical synthesis (e.g., polymers).
and development over the last decades, the stringent safety
requirements of large-format LIBs remain to be a major hurdle,
slowing down their expansion for the foretold stunning energy
demands in general and the electro-mobility sector in
However, such transformation requires the development of safer
and advanced energy storage technologies with higher energy
and power densities that meet the emerging large-scale
applications, such as in the electro-mobility (xEVs), for buffering
the intermittency nature of renewable energies sources (e.g.,
[
6,7]
particular.
One of the viable option could be to replace the
[
6]
existing state-of-the-art, organic carbonate-based with safer
electrolytes enabling the intrinsically safer and, possibly, eco-
friendly rechargeable Li-ion and beyond battery systems. Amid
available options, room temperature ionic liquids (RTILs) are the
most appealing candidates for overcoming such tough safety-
inspired demands since, being liquid, they do not require any
substantial change in the battery production process. Depending
on the chemistry and nature of the cation-anion combinations,
ILs are endowed with set of inimitable physical and chemical
features (cf. Table 1) such as nearly negligible vapour pressure,
high flash point (when measurable), and good thermal stability
solar, and wind), and utility/grid storage (e.g., load levelling).
Lithium-based rechargeable batteries are amid the viable choice
[
[
a]
b]
Dr. G. G. Eshetu, Dr. S. Jeong, Prof. Dr. S. Passerini
Helmholtz Institute of Ulm (HIU),
Helmholtz Str.11, 89081, Ulm, Germany
E-mail: geshetu@cicenergigune.com, stefano.passerini@kit.edu,
sangsik.jeong@kit.edu
Dr. G. G. Eshetu, Dr. S. Jeong, Prof. Dr. S. Passerini
Karlsruhe Institute of Technology (KIT)
P.O.Box 3640, 76021 Karlsruhe, Germany
E-mail: geshetu@cicenergigune.com, stefano.passerini@kit.edu,
sangsik.jeong@kit.edu
(
leading to significantly reduced flammability compared to
[
[
c]
d]
Dr. G. G. Eshetu
CIC Energigune, Albert Einstein 48, Alava Technology Park, 01510, Minano,
Vitoria-Gasteiz, (Spain),
conventional organic solvents, also attributed to their remarkable
flame retardancy potential), wide liquidus range (-100 to +200 °C,
avoiding possible evaporation or crystallization), tunable
E-mail: geshetu@cicenergigune.com
Dr. G. G. Eshetu
Department of Chemistry, College of Natural and Computational Sciences, Mekelle
University, P.O.Box – 231, Mekelle, (Ethiopia)
Email: geshetu@cicenergigune.com
solubility, wide electrochemical stability window (up to 5-6V)
[4,18–20]
etc.
These set of peculiar properties render them as very
promising alternative fluids for the development of safer large-
scale LIBs as well as good candidate components for innovative
[
e] Dr. P. Pandard, Dr. A. Lecocq, Dr. G. Marlair
iinstitut National de l’Environnement Industriel et des Risques (INERIS), Parc
Technologique Alata, BP2, 60550 Verneuil-en-Halatte (France)
E-mail : guy.marlair@ineris.fr, pascal.pandard@ineris.fr,
Amandine.LECOCQ@ineris.fr
[
21]
electrical energy storage devices.
Besides to their safety
ꢀ
features, ILs are reported to present comparable (even higher)
battery performances as their conventional carbonate-based
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
[
22,23]
electrolytes counter parts.
However, it will be such an over-
considered as the most promising electrolytes for the
development of highly performing, novel and sustainable energy
statement to advocate as if ILs are completely green chemicals
per se as often claimed in various commercial leaflets, scientific
[
24–26]
devices, such as LIBs.
[
16]
articles and conferences.
The possibility of building ILs via
1
8
infinite cation-anion combinations (~10 possibilities) leading to
a wide diversity requests a careful examination of their thermal,
actual fire hazards and potential persistence in the environment.
The lack of detailed knowledge on their thermal stability,
biodegradability and (eco-) toxicity data as well as actual fire
behaviour calls for more attention for their sustainable use in
general and a certain degree of caution should be thought when
ILs are used as electrolytes for applications in LIBs. Thus,
practical rules based on validated experimental data must be
used according to the intended applications and desired overall
performance of the battery system. Amid the existing classes of
ILs, those based on Pyrrolidinium (Pyr1A)-based cations
combined with imide-based anions and their analogues are
Hence, this work investigates the thermal stability, the
combustion behaviour and the biodegradation properties of
various Pyr1A-based ILs, and their corresponding electrolytes
(defined as solution of ILs with lithium salt in a 9 to 1 molar ratio),
appraising the effect of alkyl chain length, counter anion, cationic
centres, and addition of lithium salt. A number of experimental
tools including Thermogravimetric Analysis (TGA), oxygen bomb
calorimeter,
Fire
Propagation
Apparatus
(FPA)
have
and
biodegradability
testing methods/protocols
been
implemented aiming at characterizing the comprehensive and
full scale safety related profiles of the aforementioned ILs on a
case by case approach.
Table 1. Beneficial generic properties of ionic liquids (ILs) for electrochemical applications
§
§
§
§
§
Improved thermal and chemical stabilities
Negligible volatility
§
§
§
§
§
Low melting point
Moderate viscosity
Tunable solubility
High ionic conductivity
High polarity
High flash point
Flame retardancy
Negligible to high (eco-) toxicity
stability of the ILs is affected by the following four factors: 1) the
2
. Results and Discussion
+
alkyl chain length of the [Pyr1A
]
cation, 2) the nature of the
The thermal stability, actual fire behaviour and biodegradability
of neat ionic-liquids (ILs) and, in some cases, their parent
electrolytes (defined as a solution of ILs/ Li-salt in a 9 to 1 molar
ratio), were studied. Firstly, the thermal stability was evaluated
by both dynamic and isothermal TGA scans, next their energy
content and behaviour under fire conditions by making use of
oxygen bomb and Tewarson calorimeters, along with the
identification and quantification of both thermal and chemical
threats, and lastly, the biodegradability of representative ILs was
studied.
counter-anion, 3) the nature of the cationic core, and 4) the
addition of the Li salt, as detailed in the following sub-sections.
2
.1.1.
Short-Term Thermal Stability
Temperature-ramp measurements (typically associated to
-
1
-1
heating rates in the range 5°Cmin to 20°Cmin ) permit prompt
comparison of the thermal stability of materials in a series of
[
27,28]
samples.
temperatures which usually follow the order Tpeak >
the same IL) are used to define the thermal stability of ILs, which
The onset (Tonset), start (Tstart), and peak (Tpeak
)
T
onset >
T
start (for
[
17,27,29]
by no case shall be considered as an intrinsic property.
2
.1. Thermogravimetric analysis (TGA)
Fig.1 presents a typical dynamic TGA scan displaying the
method of determining the aforementioned parameters. The
onset temperature (Tonset), determined from the step - tangent
method, is one of the most commonly used parameter to define
the stability of ILs. However, ILs start to decompose at a lower
temperature than the Tonset, indicating that thermal stability would
be overestimated while using Tonset as the absolute reference for
TGA is among the most commonly employed technique for the
evaluation of the thermal stability of ionic liquid (ILs) materials.
Both dynamic (also called step-tangent or ramped) and
isothermal (also called static) temperatures analysis were
employed to access short- and long-term measurements
respectively,
enabling
real-scale
and
comprehensive
understanding of the stabilities of the ionic liquids targeted in this
study. Though the degree varies, it was found that the thermal
[
17]
the evaluation materials. This could be reinforced by the fact
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
that since the temperature increases rapidly at a fixed rate, it
makes easy to pass through the actual decomposition
temperature without detecting measurable mass loss.
Furthermore, TGA signal is only due to the formation of volatile
species, inferring degradation processes occurring without mass
loss are not discernible. Moreover, several factors including
heating rate, gas nature and its flow rate, pan material, sample
quantity, type of instrument and purity of the tested sample
temperature matching the highest peak is considered in this
experiment.
2
.1.1.1. Effect of alkyl chain-length
The TGA spectra and experimentally obtained thermal data of
the tested Pyr1A-based ILs, involving various alkyl chain-lengths
-
(
for a fixed counter anion, i.e., [TFSI] ), are provided in Fig.2a
and Table 4. For the sake of comparison, the TGA scan of the
state-of-the-art electrolyte, 1M Li[TFSI]/EC-DMC (1/1, wt.%), is
also provided in the same figure. Li[TFSI]/EC-DMC showed a
three step thermal history exhibiting maximum mass losses at
temperatures 115°C, 252°C, and 432°C, conforming to the
evaporation of DMC and EC and decomposition of Li[TFSI] salt,
respectively. As can be seen from Fig.2a and Table 4a, a mass
loss of > 5% is recorded at T < 87°C for the state-of-the-art
electrolyte, showing its lower thermal stability which could be
induced by the higher volatility of the solvent mixture. On the
other hand, all the ILs based on TFSI anion showed a one-step
decomposition behaviour and are found to be thermally stable
up to 340 °C, as indicated from the recorded onset temperatures,
inferring Pyr1A-based ILs of the specified alkyl chain lengths lies
in the moderate/more stable safety level according to the
[
30–32]
strongly affect the Tonset
.
Figure 1. Characteristic of dynamic TGA curve, indicating the
start, onset, peak and end temperatures.
[
29]
recently proposed classification.
In general, the effect of the
alkyl chain length of the Pyr1A cation does not present a huge
effect on the thermal decomposition of the ILs, in agreement
Accordingly, the scientific community is now favouring Tstart as a
more appropriate thermal stability parameter.
Tstart is the
[
32–36]
with experimental results reported for other classes of ILs.
temperature at which the decomposition of the IL begins. In this
study, the start (decomposition) temperature is defined as the
temperature for which the first derivative of the weight loss vs.
Elongation of the alkyl chains from 3 to 10 carbon atoms leads
to only a small decrease in the stability of the ILs between
successive groups, however, it does slightly more significant
-
4
-1 [17]
time curve (|dm/dT| is greater than 10 % s . The choice of
such a conservative reference is aimed to supply more realistic
and benchmarking values, emphasizing the need to keep the
electrolytes safe. However, the employed minimum weight loss
+
effect with drastic changes in the cation size (e.g., [Pyr13
]
→
+
[
Pyr110 ]). A diminution in thermal stability by ~ 45°C is observed
+
+
when passing from [Pyr13
]
to [Pyr110] (Figs.2a, d & Table 1).
The slightly higher thermal stability observed with shorter over
longer alkyl chain lengths might be attributed to longer alkyl
chain length cations may result in better leaving groups upon
heating and thereby more stable carbocations and carbon
radicals, favouring the decomposition phenomenon. This
observation is in line with what has been reported for
-
4
-1
(
10 % s ) could be still questionable and from a practical point
of view, thermal stability criteria need to be defined according to
a practical use and/or final end use. In some cases, monitoring
changes in colour or estimating an overall decomposition
fraction could be considered as sensible indicators. Peak
temperature (Tpeak) is still another very decisive parameter that
need to be considered during a given TGA analysis, and refers
to the temperature at which the sample undergoes a maximum
degradation. It can be obtained from the highest peak in the first
derivative thermogravimetry (DTG) curves (Fig.1). In case there
exist more than one peak in the DTG curves, only the
[
37]
[29]
piperidiunium, and imidazolium based ILs of different chain
lengths. However, with specific and unusual biobased ILs based
on quaternary ammonium moieties, the reverse trend was
[
38]ii
sometimes observed.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Besides to the targeted thermal stability study in this work, it is
also well documented in literature that a longer alkyl chain length
is accompanied by lower solubility, higher viscosity, higher
2.1.1.3. Effect of the nature of the cation backbone
Fig.2c, d and Table 1c describe the comparison in the thermal
stability of ether-functionalized Pyr1A-based ILs with their
corresponding alkyl-substituted analogues. Introducing oxygen
into the side chain of pyrrolidinium cations is well known to offer
unique benefits such as low viscosity and high conductivity.
However, there is lack of detailed information regarding their
thermal, and fire-induced hazards, and this work aims to fill
these existing gaps. Irrespective of the nature of the anion,
introduction of ether bonds (-C-O-) was found to worsen the
thermal stability of the cations, most likely linked to the cation-
anion electrostatic interactions. Similar observations for short-
chain-ether-functionalized ILs, but employing imidazolium
[
39–41]
density, slower diffusion, and thereby lower conductivity.
Thus, combining all these thermal, physico-chemical and
electrochemical properties, ILs based on shorter alkyl chain
lengths are more favoured for electrochemical applications as
compared to longer ones. Based on these facts, [Pyr13] and/or
[
Pyr14] could be regarded as the best choices among the Pyr-
based ionic liquids tested in this study for use as electrolytes for
lithium-ion batteries. However, since the thermal and
electrochemical behaviour of ILs in Li-ion cells is strongly
correlated with the nature of the active material, further tests in
the presence of active materials could underpin the above
conclusive remark. Furthermore, environmental consideration
might prevail for specific use over functional or safety aspects,
which might justify slightly other choice according to
biodegradability performance as discussed in section 2.3.
[
45–49]
cations have been reported before.
2.1.1.4. Effect of Li salt addition
The effect of adding a Li salt (i.e., LiTFSI) in the 1 to 9 molar
ratio to, e.g., [Pyr14] [TFSI] and [Pyr12O1] [TFSI], is shown in
Figure 2d and Table 4d. LiTFSI starts to decompose at 313°C,
and the neat ILs at 383°C and 377°C for [Pyr14] [TFSI] and
[Pyr12O1] [TFSI] respectively. While the Tstart of [Pyr14] [TFSI] is
almost unaffected, the thermal stability of [Pyr12O1] [TFSI] is
lowered by ~ 27°C by the salt addition.
2
.1.1.2. Effect of counter anion
The nature of the counter anion also plays a vital role in dictating
the thermal stability of ILs, and for this purpose, an investigation
on Pyr14-based ionic liquids with different fluorinated
-
-
-
sulfonylimides, namely [TFSI] , [FSI] , and [BETI] was carried
out. Unlike to the one-step decomposition of [TFSI]- and [BETI]-
based ILs, the FSI-based IL presented
a
two-stage
decomposition scheme (Fig.2b). Based on Figure 2b, d, and
Table 1b, the thermal stability of the investigated anions
decreases in the order: [TFSI] > [BETI] > [FSI]. The reason for
the lower stability of [Pyr14] [FSI] is the instability of the sulphur –
fluorine (S-F) bond in FSI compared to that carbon–sulphur (C-
-
-
S) and carbon fluorine (C-F) bonds in [TFSI] and [BETI] anions.
Moreover, it can be linked to the weaker interactions between
+
-
+
[
Pyr14
]
and [FSI] , compared to the stronger one in Pyr14 ---
-
+
-
TFSI and Pyr14 ---BETI . A difference in Tstart of ~100°C is
recorded between [Pyr14] [FSI] and [Pyr14] [TFSI], indicating the
effect of the anion on the thermal stability of Pyr1A-based ILs.
Figure 2. Dynamic TGA thermograms of the tested ionic liquids
-
-1
The higher thermal stability of [TFSI] and [BETI]-based ILs has
(at a heating ramp of 10°C min ), a) effect of alkyl chain length,
+
[42–44]
previously been reported with [EMI] cation,
corroborating
b) effect of identity of counter anion, c) effect of nature of
cationic core, and d) decomposition temperature of tested ILs.
+
the results observed with Pyr1A -based cation in this study.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Table 1. Start, Onset, Peak and End temperatures for the decomposition of ILs as determined from dynamic TGA
experiments
a) Effect of alkyl chain length
Thermal stability relating indicators
o
o
o
T
start/ C
T
onset/ C
469
451
443
440
429
443
447
421
T
peak/ C
470
469
478
466
462
457
456
+
[
[
[
[
[
[
[
[
Pyr13
Pyr14
Pyr15
Pyr16
Pyr17
Pyr18
Pyr19
]
]
]
]
]
]
]
387
383
381
380
377
360
352
342
6(1%)*
+
+
+
+
+
+
+
Pyr110
]
444
st
5
116 (1 )
nd
1
M LiTFSI/EC-DMC
88 (5%)**
n.a
251(2 )
rd
4
32(3 )
b) Effect of counter anion
-
[
[
[
TFSI]
BETI]
FSI]
383
342
287
451
420
321
469
426
327
-
-
c) Effect of ether-functionalized cations
[
[
[
[
Pyr14][TFSI]
Pyr12O1][TFSI]
Pyr14][BETI]
Pyr12O1][BETI]
383
377
342
332
451
457
420
414
469
457
426
419
d) Effect of Li salt addition
LiTFSI
313
383
388
377
351
382
451
459
457
459
417
469
471
457
459
[
Pyr14][TFSI]
LiTFSI/[Pyr14][TFSI]
Pyr12O1][TFSI]
LiTFSI/[Pyr12O1][TFSI]
[
*Tstart @ 1% mass loss, ** Tstart @ 5% mass loss, n.a= not available/cannot be determined
2
.1.2.
Long-term thermal stability
of ILs with Li-salt, the isothermal TGA measurements were
focused in that direction.
Though dynamic TGA scan can be exploited as a screening
method to roughly compare the thermal stability of ILs, long-term
stability measurements are required to determine the stability of
materials employed in industrial applications. Hence, to obtain
realistic data regarding the maximum operating temperature, a
long-term stability analysis of [Pyr14] [A] (where A=TFSI, FSI and
BETI) was conducted. Long-term thermal stability of the
aforesaid ILs at different temperatures under an isothermal
measurement for a prolonged time (i.e., 8 hrs) is presented in
Figures.3(a-c). Since the Tonset overrates the thermal stability of
the ILs, the chosen temperatures for the isothermal scans were
lower than the onset one obtained from the dynamic scan at
Figure 3a shows a deeper isothermal study of [Pyr14] [FSI] at five
different temperature scans, which are much lower than the
corresponding
Tonset (321°C) and/or Tstart (287°C). In the
temperature range of 100°C-140°C, the detected mass loss after
8 hrs was nearly null. However, a significant mass loss starting
at 150°C is detected reaching, for instance, at ~5.6% and 11.6%
after 4 hours of heating at 150°C and 200°C respectively. This
clearly indicates that the decomposition temperature for [Pyr14
]
[FSI] is somewhere in range of 140°C-150°C with a difference of
ca. 140°C and 170°C with the onset and start temperatures,
respectively. This is in turn consistent with the results obtained
-
1
[27]
1
0 °C min under N
scans were conducted in diminishing order of temperature, the
higher one being 200°C for [Pyr14] [FSI], and 275°C for [Pyr14
TFSI] and [Pyr14] [BETI]. Moreover, since the nature of the
2
atmosphere. Accordingly, the isothermal
by Maton et.al for imidazolium-based ILs. In the case of [Pyr14]
[TFSI] and [Pyr14] [BETI] (Figs. 3b, and c,), insignificant mass
losses were detected in the range of 150°C-250°C and 150°C-
225°C respectively over the isothermal scan for 8 hrs. However,
at temperatures higher than 225°C for [Pyr14] [BETI] and 250°C
for [Pyr14] [TFSI] presented a measurable weight loss during the
prolonged isothermal period (i.e., 8 hrs).
]
[
anion seems to influence the thermal stability of ILs more than
that of the alkyl chain length, ether-functionalization and doping
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
rate, energy conversion efficiency (%), emitted (toxic/pollutant)
product yields (mg of gas /g of IL sample), and elemental
conversion efficiencies (%), such as carbon (C), fluorine (F),
sulphur (S), and nitrogen (N) into related (and measured)
toxic/pollutant species.
2
.2.1. Ease of ignition (Ignitability) of ILs
Ignitability of ILs is one of the flammability hazard rating
indicators, and is mostly reflected by the time to ignition (TTI)
defined as the time needed to reach a sustainable flaming
combustion in given conditions (here piloted ignition under
calibrated heat flux stress). The shorter the TTI value, the more
ignitable the ionic liquid is. While considering liquids, the ease of
ignition or flammability is primarily indicated for regulatory
purposes by the flash point values of the material under
Figure 3. Isotherm thermogravimetric analyses of [Pyr14] [FSI]
between 100°C and 200°C as well as [Pyr14] [TFSI] and [PYR14
BETI] between 150°C and 275°C for 8 h.
]
[
[
52]
consideration. The flash point is defined as the experimentally
determined temperature at which the liquid material emits
sufficient vapour to form a combustible mixture with air in given
standard conditions, inferring the definition of a flammable liquid
as one capable of sustained burning with flames. In this regard,
because of their extremely low vapour pressures and
decomposition at elevated temperatures, ionic liquids are often
reported as “non-flammable” materials because of their high or
not measurable flash points, outscoring the upper limit of
As a general remark, though the long-term thermal stability
shows a similar trend as from temperature-ramped TGA, ILs
degrade at a considerably lower temperature than either the
T
onset and/or Tstart obtained from dynamic measurements using
scans at 10°C/min.
2
.2. Fire hazard analysis of ILs
Fire hazard analysis usually starts with the assessment of
ignitability of materials and substances, usually closely bound to
volatility or thermal stability aspects depending on how
combustion develops. In addition, behaviour in fire conditions
shall also address thermal and chemical threats evaluated from
rate of heat and toxic/pollutant product release to complete the
fire hazard analysis.
flammable liquid hazardous class of substances, from
a
regulatory viewpoint. However, non-flammability property of ILs
in a wider sense is not true as verified later by many researchers
and unlike to most common liquids whose ignitability is due to
evaporation, the flammability of ionic liquids is mainly due to
their thermal decomposition, generating flammable substances
instead of themselves vaporizing, similarly to flammability
Accordingly, to better assess the overall safety issues of ionic
[
53]
liquids,
a
detailed investigation on the actual fire-induced
characteristics of most polymers.
From Fig.4a and Table 2, the first observation is that Pyr1A
hazards of Pyr1A-based ILs was also carried out using a bench-
scale Fire Propagation Apparatus (FPA- ISO 12136:2011), also
known as the Tewarson Calorimeter. This multipurpose device
-
based ILs present higher resistance towards combustion
compared to the state-of-the-art alkyl carbonate-based
electrolyte (i.e.,1M LiTFSI/EC-DMC). The time to ignition of the
carbonate-based electrolyte differs (i.e., lower) from the most
combustible IL, [Pyr110] [TFSI], by ~ >240s (i.e., 4 min). As
[
1]
[16,17]
has been recently used by Ribiere et.al, A-O. Diallo et.al,
[
6,50,51]
and Eshetu et.al,
to inspect the combustibility and
behaviour in fire conditions of a wide range of materials
including solvents, electrolytes, and battery components. An
overview of the major results obtained from the combustion
analysis of the ILs in the Tewarson calorimeter is summarized in
Table 5, employing various parameters such as ignitability
defined by time to (piloted) ignition (TTI) under calibrated heat
flux stress, actual (effective) heat of combustion, heat release
[
21]
mentioned by Armand et al,
this clearly evidences the flame
retardancy ability of the ILs as a generic safety advantage,
which renders useless the addition of flame retardant additives
in IL-based battery electrolytes. However, it has to be underlined
that despite the delay in ignition time, ILs are able to burn and
thus cannot be hailed any longer as “non-flammable” materials.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Within ILs, those employing shorter alkyl chain lengths
demonstrated improved resistance towards burning, which is
linked to their slightly higher decomposition temperature
compared to those with longer chains. For instance, the TTI for
2.2.2. Heat Release Rate (HRR) profiles
The HRR refers to the rate at which the heat produced by
combustion is released with time, and is one of the key
parameter, which dictates the thermal and chemical impacts of a
fire involving ionic liquids. The Peak HRR (pHRR) is defined as
the recorded peak maximum obtained for each IL. Obviously,
the higher the pHRR value, the more the heat given off at that
event is. The HRR is also related to the rate of mass loss via
•
[
Pyr13] [TFSI] (~350s) is higher than that of [Pyr110] [TFSI] by
100s. From the data in Fig.4 and Table 2, the alkyl chain length
appears to have stronger influence on the combustion
~
a
behaviour than on the thermal stability of the ILs.
+
heat of combustion as in the following equation, where
q
,
For a fixed cation, [Pyr14
]
, the IL based on the FSI anion
showed the highest ignitability (< 76 s), compared to BETI (>
15s) and TFSI (> 326s)-based ones (Fig.4b and Table 2),
•
m
fuel
,
χ
and ΔΗc denotes the rate of heat release, rate of
3
which is linked to the different thermal stability of the ILs. In fact,
the lower time to ignition of FSI-based IL is due to the generation
of sufficiently enough flammable decomposition products at
much lower temperature, compared to that of TFSI- and BETI-
based ones.
-1
-2 -1
mass loss (kg s or gm s ), combustion efficiency (
≤
1) and
heat of combustion respectively.
•
•
q = χΔHcmfuel
(1)
The heat release profiles are presented in Fig.5 and further
detailed quantified results of the combustion experiments are
summarized in Table 2. Fig. 5a portrays the HRR profiles of
Pyr1A-based ILs of different alkyl chain lengths and a comparison
with the molecular solvent-based electrolyte, 1M LiTFSI/EC-
DMC. Undoubtedly, the neat ILs showed higher resistance
towards ignition and combustion compared to the alkyl
carbonate-based electrolyte, considering both the kinetics and
magnitude of the peak heat release rate. Within tested ILs, the
start of the HRR decreases with increasing the alkyl chain length
From Fig.4c and Table 2, ether-functionalized Pyr1A-based ILs
tend to slightly impact the fire hazard of parent cations, as
evidenced by their lower TTI, than their corresponding alkyl-
substituted equivalents. For instance, the TTI of [Pyr12O1] [TFSI]
and [Pyr12O1] [BETI] was found to be lower by >35s, compared to
their alkyl-based ILs, i.e. [Pyr14] [TFSI] or [Pyr14] [BETI]), and is
related to the same reasoning as above.
Data which is not presented here indicated that the addition of Li
salt does not change the ignitability of [Pyr14] [TFSI] markedly,
but acts as a flame retardant for [Pyr12O1] [BETI].
(
Fig. 5a), in agreement with the time to ignition (TTI) and
observed thermal stability of the ILs.
On the effect of anions, a fire hazard ranking of the pattern:
-
-
-
[
FSI] >> [BETI] > [TFSI] was observed from heat release and
associated kinetics viewpoints, once again a similar trend as
shown by their thermal stabilities (Fig.5b and Table 2). FSI-
based IL disclosed a much faster kinetics than BETI- and TFSI-
based ones, attributed to its decomposition at lower temperature
accompanied by the generation of flammable products.
Moreover, the highly exothermic decomposition of the FSI anion
may provide extra heat to the IL, resulting in an abrupt release of
[
2,5,51]
heat.
Moreover, [Pyr14] [FSI] presents a very high pHRR
-
2
-2
(
>1428 kW m ), compared to that of BETI (609 kW m ) and
Figure 4. Ignitability of: a) ILs of different alkyl chain lengths and
-2
TFSI (455 kW m ). On the other hand, the slightly higher heat
release commencing at a shortened time for BETI can be
explained due to the larger anion size resulting in a decrease
inter-molecular force and also due to an increase in the electron
conventional electrolyte, b) counter-anions, and c) ether-
2
functionalized Pyr1A-based ILs. *: under 50 kW/m of external
heat flow onto sample surface.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
withdrawing effect of –CF
2
CF
3
in BETI, compared to –CF
3
in
[Pyr12O1] [TFSI] and [Pyr14] [BETI] vs [Pyr12O1] [BETI], it can be
TFSI. These results confirm the important role of the IL anion in
influencing the kinetics of combustion and most likely
contributing as the essential component of the overall flame
retardancy property of test ILs.
easily noted that ether-functionalized ILs tend to be more easily
combustible
than
their
corresponding
alkyl-substituted
comparable ILs. An earlier ignition and enhanced peak heat
release rate is observed with both ILs containing ether bond. In
this case, the difference may be associated to the oxygen
already present within the structure, which may promote the
combustion.
ef
2
.2.3. Effective Heat of Combustion (ΔH
c
)
The overall heat dissipated, called effective heat of combustion,
can be experimentally derived making use of the Tewarson
calorimeter. For substantiating the results from Tewarson
calorimeter, the total energy of ILs with different alkyl chain
lengths was determined using oxygen bomb calorimeter. The
theoretical energy of combustion corresponding to the examined
ILs have been estimated by use of the Boie correlation using the
predictive equation (2), where C, H, O, and S are the mass
fractions of elemental carbon, hydrogen, oxygen, and sulphur in
Figure 5. Heat release rates (HRR) of: a) ILs of different alkyl
chain lengths and conventional electrolyte, b) counter-anions, c)
ether-functionalized Pyr1A-based ILs, and d) addition of Li-salts.
[
54,55]
the burning ILs.
Fig. 5c and Table 2 present the effect of introducing an ether
+
bond within the [Pyr1A
]
cation backbone, and clearly show its
negative attribution. From the comparison of [Pyr14] [TFSI] vs
−
1
ΔΗ (kJ Kg ) = 35,160C +116,225H −11,090O + 6,280N +10,465S
(2)
c
As it can be seen from Fig.6a and Table 2, the total and
predicted heat of combustions obtained by oxygen bomb
calorimeter and purpose-built Boie model, respectively, strongly
correlate with the experimentally obtained data. The measured
values lie a bit lower than the theoretical (complete) heats of
combustion. One of the most interesting result here is that the
Boie correlation works very well with Pyr1A–based ILs, which
however, much lower than of phosphonium-based ILs (29-42 kJ
-
1
[16,17]
g ).
Increasing the alkyl chain content in the Pyr1A-cations obviously
increases the fuel fraction (mainly combustible carbon) resulting
in increasing the overall heat of combustion (Fig.6a and Table 2).
The energy conversion efficiency (
χ
), determined as the ratio
[
55]
was not in the domain of validity of previous works. This result
ef
of the heat values obtained from Tewarson calorimeter (ΔH
c
) to
[
55]
also consolidates previous findings from Diallo et al where the
Boie
that calculated from Boie correlation (ΔH
c
), is found in the
Boie model was evaluated as being within the top 5 among the
range of 74.4% for [Pyr18][TFSI] to 80.3% for [Pyr19][TFSI].
1
7 predictive existing models of heats of combustion tested for
ef
ΔΗc
their applicability to ionic liquids. Moreover, the correlation
between the heat values obtained from Boie correlation and
integration of HRR is quite consistent. In general, the effective
heat of combustion of the reported Pyr1A-based ILs lies in the
χ =
(3)
Boie
ΔΗc
The effective heat of combustion of test Pyr1A-based ILs is also
affected by the counter-anion as shown in Fig.6b and Table 2,
and in practice, it is found to vary in the order of [FSI] > [TFSI] >
[BETI]. The relatively lower heat of combustion of [Pyr14] [BETI]
may be ascribed to the presence of more F atoms, inhibiting the
-
1
range of 14-20 kJ g of the burned IL materials, in the same
range as conventional organic
-
based solvents and
[
5,6]
-1
electrolytes
as well as imidazolium-based ILs (15-22 kJ g ),
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
formation of radicals during the gas-phase combustion.
Increasing the mole ratio of flame inhibiting elements (notably F
and P) to the mole ratio of combustible elements such as C, H,
N etc. reduces the cascading chain propagating gas-phase of
the combustion reaction via removing the highly reactive free
SO
2
etc. The product yields of major emitted gases are given in
Fig.7(a-c) and Table 2.
2
The amount of CO is higher for longer alkyl chain lengths, due
to higher combustion stoichiometry or fuel content (Fig.7a). This
finding is further corroborated by the decrease in the amount of
•
•
radical species (e.g., H and/or OH). This can be further
evidenced from the lower energy conversion efficiency (~
2
HF and SO gases, which is linked to the mass dilution of
6
2.5%) for [Pyr14] [BETI] compared to others (Table 2).
sulphur and fluorine with increasing alkyl chain length. In general,
though the tested ILs presented beneficial features in terms of
their thermal stability and thermal treats in fire conditions, their
fire-induced toxicity is higher than that of conventional
electrolytes (see Table 2), which can be considered as a major
issue calling for a detailed risk related examination.
Introducing oxygen into Pyr1A-based IL decreases the effective
heat of combustion as can be verified from the comparison of
the heat for [Pyr14] [TFSI] vs [Pyr12O1] [TFSI] and [Pyr14] [BETI] vs
[
Pyr12O1] [BETI] in Figure.6c.
From the combustibility point of view, the addition of the Li-salt
resulted in decreasing the overall effective heat of combustion of
[
Pyr14] [TFSI] by ~ 25%, however, not of [Pyr12O1] [TFSI] (Fig.6d).
Fig. 7d shows the conversion efficiencies of C, F, S and N atoms
of the ILs into their respected gases and traces of residual
species. Interestingly, the trend is found to be similar regardless
the size of the alkyl chain, and a high conversion efficiency is
recorded for C, F, and S atoms. This latter result is also
reflecting a very common trend observed with many chemicals
burning in oxidative pool fire conditions. The lower conversion
efficiency for N confirms the general behaviour of Fuel-N
conversion with N-containing organic liquids as shown in the
past by several researchers from the fire and combustion
[
56,57]
science communities.
Indeed, Fuel-Nitrogen combustion
chemistry studies revealed significant nitrogen recombination
into molecular nitrogen (N
conditions.
2
) as a main mechanism under fire
Figure 6. (a) Effective, total and theoretical heats of combustion
of ILs of different alkyl chain lengths, b) effective and theoretical
heats of combustion of ILs based on FSI, TFSI, and BETI-anions, The mass production of all detected gases emanating from ILs
c) effective and theoretical heats of combustion of ether-
functionalized ILs, and d) effective and theoretical heat of
combustion Li-salt containing ILs.
comprising of FSI-, TFSI-, and BETI- anions is provided in Fig.
7b and Table 2. High amounts of SO , and HF gases are
2
detected for all the tested ILs.
2
.2.4. Fire-induced toxicity examination of ILs
The potential fire gases toxicity driven by the release of
combustion products is another key safety issue that needs a
due consideration. Investigation in that field has dedicated to the
measurement of Soot, CO, CO
2
, unburned hydrocarbons (THC,
2 2
C H
), oxygenated compounds (such as aldehydes), HF, NO,
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Figure 7. Emission factors of ILs of: a) different alkyl chain lengths, b) counter-anion, c) ether-functionalized, and d) elemental
conversion efficiency.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Table 2. Combustion behaviour of pyrrolidinium-based ionic liquids in the Tewarson calorimeter under well
ventilated conditions
[
TFSI]
[Pyr14
]
[Pyr12O1
]
Li[TFSI]/
Li[TFSI]/
EC-DMC
[
Pyr14
[Pyr12O1
[
Pyr13
]
[Pyr14
]
[Pyr15
]
[Pyr16
]
[Pyr17
]
[Pyr18] [Pyr19
]
[Pyr110]
[FSI] [BETI] [TFSI] [BETI]
]
[TFSI] ][TFSI]
Initial mass (g)
Mass loss
Ease of ignition (s)
19.9
96.5
20
94
20
20.4
96.1
20
20
96
20
96
20.3
96.6
19.9
82.9
20.2
95.5
20.1
91.5
20
20
95
20.1
90.5
19.9
98
%
95.5
96.1
91.5
Time to ignition (TTI) 350
318
312
295
290
270
277
260
76
315
290
277
317
307
26
Heat of combustion
(
kJ)
Measured heat of
combustion
Predicted heat of
combustion
Energy conversion
efficiency (%)
1
1
9
4.2
4.9
5.3
14.7
15.8
93.0
15.6
16.7
93.4
16.3
18.2
89.5
17.4
18.1
19.1
94.8
18.7
19.8
94.4
19.5
20.5
95.1
14.5
18.1
80.1
9
8.9
6.9
11.1
15.6
71.2
8.8
10.3
13.7
75.2
18.75
92.8
14.4.
62.5
13.9
63.7
12.9
53.9
13.3
66.2
Product yields (mg
-
1
gas g IL.)
CO
2
784
23
837
21.7
72
858
24
857
24
972
27
979
28
1069
24
1172
25
889
14.7
2.9
663
28
639
24
524
27
811
23
584
22
1229
2.9
6.3
3.1
0
CO
Soot
THC
51
67
81
77
72
80
75
69
60
52
65
56
7.6
0.2
318
14.7
0.4
8.9
0.4
291
9.2
0.4
298
9.7
0.8
279
8.4
0.6
270
9.2
0.5
252
7.5
0.3
252
1.7
12.7
0.4
251
7.8
0.4
306
15.6
0.7
246
22.6
0.7
302
9.6
0.6
321
CH
SO
4
2
0.2
292
452
71.5
NO
6.1
1.8
4.4
5.7
2.4
6.8
5.5
1.7
4.8
7.1
1.5
5.4
6.5
1.4
4.6
7.2
2
6.6
1.9
3.4
7.0
1.8
3.2
12.5
0.6
-
4.6
1.5
4.3
4.8
2.0
5.8
4.7
1.4
5.8
5.3
1.5
5.5
5.8
2.1
6.1
0
0
0
2
N O
HCN
4.7
eq. HF (HF+SiF
4
)*
251
250
245
238
228
208
210
205
119
304
254
302
236
256
17.6
Elemental
conversion
efficiencies, (%) **
Carbon, %
92.5
82.4
97.9
9.0
92.9
70.6
90.5
11.3
92.6
82.3
94.8
9.3
99
96.5
84.8
97.4
11.0
92.7
79.4
96.7
11.8
96.3
82.3
93
98.9
83.6
96.1
11.4
82.3
72.1
54.7
2.2
87.9
75.7
97.9
9.8
81.5
82.2
92.9
9.1
74.1
72.4
92.2
11.1
-
-
-
-
-
-
-
-
-
-
Fluorine, %
Sulphur, %
Nitrogen, %
85.9
100
11.1
-
10.5
--
*
2
From HF and partly reacted with Quartz (SiO ). ** (C) into (COx + HCN and THCs); (S) into SO2; H into eq. HF; (N) into (NOx + HCN)
According to measured product releases, and considering high
conversion rates of hetero-atoms like fluorine and sulphur,
genuine fire induced toxicity studies could have to consider the
2.3. Biodegradation of pyrrolidinium-based ILs
In complement to electrochemical and thermal stability of ILs for
electrochemical applications, their chemical stability versus time
in various environments is also of interest from a sustainability
viewpoint. As a matter of fact, biodegradability is another key
parameter which governs the persistence of chemicals in the
environment and consequently influences their behavior in
environmental compartments and their toxicity towards aquatic
2
impact resulting from major irritants SO and HF in addition to
more conventional threat resulting from emissions of carbon
monoxide responsible of nearly all fire deaths by smoke
inhalation in conventional building fires. Data presented in table
4
have been presented in such a way that they can easily be
used to determine, in a given scenario of interest, how far fire
induced toxicity may be a real issue for the use of tested ILs, e.g. data are available for pyrrolidinium cations,
[
26,58,59]
and terrestrial organisms.
Although some biodegradation
[
60,61]
information
[
16,51]
by applying methodologies exemplified in recent papers.
regarding the biodegradation of the compounds selected in this
study was scarce.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Table 3. List of ionic liquids studied and results of their biodegradability tests (organic fraction parts only).
Test compound
Solubility
assessment
Ready biodegradability
Inherent biodegradability
OECD 302B
% biodegradation after 28
days (mean value)
Not biodegradable
0 %
OECD 301B
biodegradation after
8 days (mean value)
OECD 301F
%
2
% biodegradation after
28 days (mean value)
Not biodegradable
[
[
[
[
[
[
[
[
[
Pyr12O1][TFSI]
Pyr13][TFSI]
Pyr14][FSI]
Soluble
Soluble
Soluble
Soluble
Soluble
Soluble
Insoluble
Insoluble
Insoluble
N.T.
0
.9 %
Not biodegradable
Not biodegradable
0 %
N.T.
0
%
Not biodegradable
N.T.
N.T.
7
.7 %
Not biodegradable
Pyr14][TFSI]
Pyr16][TFSI]
Pyr17][TFSI]
Pyr18][TFSI]
Pyr19][TFSI]
Pyr110][TFSI]
N.T.
N.T
N.T
0
%
Not biodegradable
N.T.
N.T.
N.A.
N.A.
N.A.
0
%
Readily biodegradable
4.6 %
Readily biodegradable
0.4 %
Not biodegradable
9.8 %
Readily biodegradable
8 %
N.T.
N.T.
N.T.
N.T.
7
6
1
7
N.T.: Not Tested; N.A. Not Applicable
Our results summarized in table 3 showed that the alkyl chain
length had a significant influence on the biodegradation of
pyrrolidinium cations. No degradation was observed in the ready
biodegradation percentage obtained (i.e., 9.5%) confirmed that
this compound can be classified as not readily biodegradable.
+
biodegradability tests (i.e. OECD 301F and 301B) for [Pyr12O1] ,
3. CONCLUSIONS
+
+
+
[
Pyr13] , [Pyr14] , [Pyr16] . These results confirmed those of
In this work, a detailed experimental assessment of thermal
stability, biodegradability and fire-behaviour of pyrrolidinium
(Pyr1A)-based ILs has been performed. Aside from
demonstrating the higher thermal stability of IL-based
electrolytes vs the conventional ones, the above-mentioned
properties of such ILs are found to be highly dependent on the
structure of the ILs, i.e., alkyl chain length, cation modification
via introducing oxygen, cationic centre and counter-anion type.
Both short and long thermal stabilities have been investigated
making use of temperature-ramp and isothermal TGA scans,
respectively. The combustion analysis of the ILs was examined
in terms of fire hazard rating indicators, encompassing ease of
ignition indicators (i.e., time to ignition), and combustion-related
energetic variables such as effective heat of combustion, and
heat release rates, obtained from fire calorimetry experiments. In
general, tested ILs demonstrated superior thermal stabilities and
flame retardancy properties compared to conventional
solvents/electrolytes, but fire induced toxicity can be a significant
issue considering the high amount of toxic gases released. The
data obtained from both thermal and combustion investigations
are highly correlated, evidencing the actual fire hazard in the
case of ILs is initiated and sustained by the flammable
[
61]
[58]
+
Stolte et al, and Neuman et al, who concluded that [Pyr14
]
was not readily biodegradable. The most favorable test
conditions applied in OECD 302B (i.e. a low ratio of test
substance to biomass, which gives a better probability to obtain
a positive result compared to tests for ready biodegradability)
didn’t improve the degradation behavior of these compounds
leading to the conclusion that they might be persistent in the
+
environment. For [Pyr13
]
,
no enhancement of the
biodegradation level was observed by using an inoculum
adapted to the test compound during 15 days prior to the test
initiation.
+
Conversely, compounds with a longer alkyl chain: [Pyr17
]
,
+
+
[
Pyr18] , [Pyr110] , reached the pass level (i.e. 60% theoretical
oxygen demand in a 10-day window within the 28-day period of
the test) and have been accordingly classified as ‘readily
biodegradable’. Unexpected results have been obtained for
+
[
Pyr19
]
compared with the other compounds with long alkyl
chain. Biodegradation percentages after 28 days were
heterogeneous between the two duplicates performed: 0 % and
3
9.3 %. Therefore, the test was repeated. The low
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
decomposition products under strong external heat stress, not
by the evaporation of molecules as in the case of conventional
ones.
Overall, this work provides an in-depth analysis of the overall
safety related risks of pyrrolidinium-based ionic liquids, which
are crucial for their complete safety appraisal and their future
sustainable use. The further examination of potential eco-toxicity
issues relating to these materials is presently undergoing
The major results can be summarized as:
1
)
Isothermal TGA scans show that ILs exhibit decomposition
at temperatures much lower than the onset determined
from the dynamic measurement, indicating a due care
should be taken while attempting to evaluate the stability of
ILs for realistic and industrial applications according to the
context of production and use of such ILs.
[
62]
applying procedures described in literature.
4. MATERIALS and METHODS
4.1. Synthesis of Ionic Liquids (ILs)
2
3
)
)
The thermal stability is found to be inversely related to the
alkyl chain length and introduction of oxygen into the alkyl
side chain.
1
N-Methylpyrrolidine (Pyr , Sigma Aldrich), 2-bromoehtylmethyl
ether (TCI) and the series of bromine reagents from 1-
bromopropane to 1-bromodecane (Alfa Aesar) were purchased
and purified by distillation to remove residual impurities. After,
Counter anions play a major role in governing the thermal
stability of ILs. For a fixed cation, the thermal stability
decreases in the order of: TFSI > BETI > FSI. The highest
1
Pyr and one of the bromide reagents were mixed in a sealed
glass reactor using ethyl acetate as buffer solvent, which was
then dipped in an ice bath to avoid rapid chemical reactions. The
o
thermal instability of FSI-based IL (Tstart < 150 C) is
+
-
attributed to the weaker S-F bond and Pyr14 ---FSI
reaction led to the formation of the series of precursors ([Pyr1A]
interaction.
[Br]), which were then separated by filtration and washed with
twice the volume of ethyl acetate to remove unreacted reagents
and other impurities.
4
)
The ignitability, kinetics of heat release, and effective heat
of combustion of the IL that are major indicators of the
overall fire hazard were found to increase with increasing
the cation chain length. The first two variables are due to
the lower decomposition temperature, and the last one is
accredited to the increase in the fuel content of the ILs with
increasing the number of carbon atoms.
Room temperature ionic liquids (RTILs) were synthesized by
anion exchange dissolving in water the precursors and of the
following
anion
bearing
salts,
lithium
bis(trifluoromethanesulfonyl)imide (Li[TFSI], 3M), potassium
bis(fluorosulfonyl)imide
(fluorosulfonyl)(trifluoromethanesulfonyl)imide
Provisco), and lithium
(K[FSI],
Solvionic),
lithium
5
)
FSI-based IL exhibited much higher ignitability, faster heat
release rate, and large effective heat of combustion
compared to TFSI- and BETI-based ones. Moreover, the
conversion efficiencies of all elements (%) were found to be
relatively smaller for FSI, which could be related to the
speedy nature of the combustion.
(Li[FTFSI],
bis(pentafluoroethylsulfonyl)imide
(Li[BETI], Chameleon reagent). After anion exchange, the
obtained, water-insoluble ionic liquids were separated from the
aqueous phase (containing lithium or potassium and bromide)
and rinsed 5 times with de-ionized water to extract all water-
soluble impurities. Extra purification step was carried out using
activated charcoal and aluminium oxide. Ionic liquids were then
filtrated and dried to desired temperature, sequentially at range
6
)
The environmental persistence was also found to be related
to the alkyl chain length. Pyrrolidinium cations can be
classified as follows: readily biodegradable: Pyr17, Pyr18
Pyr110; potentially persistent compounds: Pyr12O1, Pyr13
,
,
-
7
of 1mbar to 10 mbar (≤ 2ppm H
2
O). The list of precursors and
Pyr14. As inherent biodegradation tests have not been
carried out for Pyr16 and Pyr19, these two compounds can
be classified as not readily biodegradable only. It can be
noticed that the trend regarding the effect of the alkyl chain
length reveals opposite with regard to trend towards
thermal stability. Therefore, in environment sensitive
applications, some compromise with the alkyl chain length
might have to be considered.
ILs synthesized for this study is given in table 4.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Table 4. Precursors and synthesized ILs used in this study
Precursors
Corresponding synthesized ILs
Structure
Structure
Acronym
Acronym
[
Pyr13][Br]
[Pyr13][TFSI]
[
Pyr14][Br]
[Pyr14] [TFSI]
[
[
Pyr14] [FSI]
Pyr14][BETI]
[
[
Pyr15][Br]
Pyr16][Br]
[Pyr15][TFSI]
[Pyr16][TFSI]
[
[
[
Pyr17][Br]
Pyr18][Br]
Pyr19][Br]
[Pyr17][TFSI]
[Pyr18][TFSI]
[Pyr19][TFSI]
[
[
Pyr110][Br]
[Pyr110][TFSI]
Pyr12O1][Br]
[Pyr12O1][TFSI]
[
Pyr12O1][BETI]
4
.2. Thermogravimetric Analysis (TGA)
30°C to the predetermined value, followed by a plateau for a
prolonged time (i.e., hours). For each experiment, the
8
The thermogravimetric analysis (TGA) scans were collected on
background was subtracted, and to ensure reproducibility and
reliability, two TGA measurements were conducted on each
ionic liquid.
®
a Netzch instrument and analysed using the Proteus software.
Experiments were carried out in 40 µL aluminium pans, filled
and sealed in an argon-filled glove box (O
2
and H
2
O < 0.1 ppm).
The measurements were conducted under N
2
(inert) atmosphere.
4
.3. Oxygen bomb calorimetry
Two types of TGA measurements, namely dynamic and
isothermal scans, were performed. The dynamic experiments
were carried out using a temperature rise from 30 to 580°C at a
Oxygen bomb calorimetry (Model 110P, Oxygen combustion
bomb, Parr Instrument Co., Moline, Illinois, following the ASTM
D240 protocol) experiments were used to determine the
theoretical energy combustion of the investigated ILs. Prior to
the sample testing, the bomb calorimeter was calibrated by
combusting approximately 1g of standard benzoic acid (NBS
Thermochemical standard 39g), which is translated into heat of
heating rate of 10°C/min. The start (Tstart), and onset (Tonset
)
temperatures are defined as the temperatures for which the first
-
4
derivative of the weight loss vs. time curve (|dm/dT| is > 10
%
-
1
s
and from the step tangent method, respectively. The
isothermal experiment entailed a temperature increase from
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
-
1
combustion of 26.454 MJ Kg . In order to achieve actual and
effective combustion of ILs (which is not 100% efficient via
standard ignition procedure) during the process of combustion in
oxygen, pharmaceutical paraffin oil (for which the heat of
combustion was firstly measured) was mixed in a predetermined
characterization of burning behaviour of materials and products
in fire conditions. It is known for its outstanding capability to
measure key material properties, encompassing ignitability, fire
propagation potential, thermal and chemical characteristics
under optimized fire conditions, including quantified heat and
combustion product release rates and related conversion
efficiencies. Repeatability and reproducibility of data count are
among the major advantages of the equipment. Preliminary tests
enabled to choose the most suitable operating conditions, and
under the adapted testing conditions, an external heat flux of 50
quantity with the IL samples.
A
weighed sample of
approximately 0.21 to 0.35 g of each IL was placed in the
platinum crucible and assembled in the bomb calorimeter. The
vessel was then pressurized with pure oxygen to 3.0 MPa and
placed inside a bath containing 2 litres of water in an insulated
jacket. A monitored stirrer was placed inside the water bath to
-
2
kW m was applied to each tested IL until ignition was observed.
-
2
circulate the water around the bomb creating
a
uniform
After ignition, the external heat flux was diminished to 25 kW m .
For the organic carbonate-based (i.e, 1M Li[TFSI] in EC-DMC)
electrolyte, the external heat flux was adjusted at all times to 25
temperature. The equilibrium rises of water temperature due to
combustion in the bomb calorimeter, reflecting the combustion
enthalpy of the test material, was recorded using a precision
thermistor (Omega Model 1417E). The amount of energy
released by paraffin oil was subtracted from the total energy
released in the bomb to compensate for the added charge.
−
2
kW m . The details of the instrument and operating conditions
have been described in our previous publications and readers
[
1,6,65,66]
are encouraged to refer to them.
Energy correction, e
1
, was made for the heat contribution
4.5. Determination of calorimetric and kinetic parameters
resulting from burning of nitrogen trapped in the bomb to form
nitric acid. It was assumed that a value of 10 calories is a good
Of the various fire characterising parameters, the heat released
upon burning a given combustible material, called (effective)
1 2
correction for e . A correction value e for the combustion of
c
heat of combustion (ΔH ), the kinetics at which the heat energy
sulphur leading to sulphur trioxide forming sulphuric acid instead
of sulphur dioxide was also applied. This adjustment was equal
to 13.7 calories per percentage of sulphur in the sample mass.
dissipated, also called heat release rate (HRR) are amid the
fundamental variables governing the fire hazard in a given
scenario. These parameters can experimentally be accessed
using fire calorimetry laws based on the assessment of oxygen
consumption (OC) and/or carbon dioxide generation (CDG) with
a high degree of repeatability and reproducibility while applied in
appropriate fire calorimeter tests. OC and CDG utilize direct
mass balance conversion relating to thermochemistry of
concerned combustion reaction rather than trying establishing
direct thermal balances making them more advantageous. In the
fire science community, indeed, it is well recognized that the
heat release is the most single important variable linked with the
flammability of the material, including all its extended
components such as ignition, fire growth, flame heights, mass
fluxes of toxic compounds etc., and can be estimated using
either OC and/or CDG as follows:
Finally, for the fuse wire an adjustment e
was applied. The gross heat of combustion or high heating value
HHV) is then calculated as:
3
equal to 15 calories
(
WΔT −e −e −e − HHVparaffinmparaffin
1
2
3
HHV =
(4)
IL
mIL
where W is the energy equivalent of the calorimeter obtained
from the calibration; ΔT is the temperature rise; HHVparaffin
paraffin and mIL are the gross heat of combustion and the weight
,
m
of paraffin and the weight of IL, respectively. Three replicates
are performed for each sample, and the typical relative error was
less than 0.2%.
4
.4. Fire Propagation Apparatus (FPA) (Tewarson
calorimeter)
The actual fire behaviour of PyriA-based ILs was determined by
means of the Tewarson calorimeter, also officially designated
under the name of Fire Propagation Apparatus (FPA) by
OC methodology: This is the most commonly employed
[
67]
standard technique, which relies on Thornton's principle.
OC
physical equations depend on the degree of combustion
efficiency.
[
63]
different standards, namely NFPA 287 ASTM E2058 and ISO
[
64]
1
2136
The instrument belongs to the family of bench-scale
For a combustion reaction that has reached completion, the
HRR can be estimated by the equation:
multipurpose testing fire calorimeters, focusing on the
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
o
organic solids or liquids or gaseous compounds (Thornton´s
principle).
•
•
•
q = E(m −mo2
)
(5)
OC
O2
However, the above equation overestimates the HRR in practice
because most of the combustion reactions, particularly those
involving liquids such as ILs, are incomplete leading to carbon
monoxide (CO) and soot particles (C). Thus, Hess's law is
applied to take into consideration the incomplete combustion to
CO and C, and is given by:
o
•
•
-
2
Where qOC is the heat release rate (kW or kW m ), mO2 and
•
m
o2 are the mass flow rates of oxygen from the incoming air (in
-
1
kg s ) and in the exhaust duct, respectively. The generic
constant,
E
, refers to the energy released (for a complete
consumed for a given fuel and
combustion) per mass unit of O
2
approximately constant (~13.1+0.7kJ/g O
2
) for a large number of
o
/
//
•
•
•
•
•
q = E(m −mo2 )−(E − E)ΔmO2 −(Es − E)Δm 2
O
(6)
OC
O2
CO
2
CO , CO, and soot (CDG calorimetry) so as to check any
possible inconsistency that might have resulted from sampling
issues (e.g., leaks). However, since the results obtained from
both methodologies were systematically in a good agreement,
Where ECO is the energy release per unit of oxygen consumed
-
1
for the oxidation of CO to CO (17.7MJ Kg ), ES is the energy
2
results are presented based on O
CO.
2
consumption, corrected for
release per mass unit oxygen consumed for the oxidation of soot
/
•
-
1
(
S) to CO2 (12.3 MJ Kg ), ΔmO2 is the oxygen mass flow
4
.6. Biodegradability test
/
/
•
In order to assess the degree of biodegradability of organic
-
1
required to oxidize CO to CO
2
(kg s ) and ΔmO2 is the mass
[69]
chemicals, several methods have been standardized in OECD
.
-
1
[68]
2
flow rate required to oxidize soot to CO (kg s ).
The testing strategy applied for assessing the biodegradability of
+
+
the organic fractions of the ILs (i.e., cations [Pyr12O1] , [Pyr13] ,
CDG methodology: the HRR for a complete combustion can be
+
+
+
+
+
+
[
Pyr14
]
, [Pyr16
]
, [Pyr17
]
, [Pyr18
]
, [Pyr19
]
and [Pyr110] ;
estimated using CDG as:
counter anion [TFSI] and [Pyr14][FSI]) is consistent with OECD
recommendations and combined ready biodegradability and
inherent biodegradability tests (see Table 5). Ready
biodegradability tests are stringent tests for which a positive
result will allow concluding that the chemical will degrade
ultimately and quickly in the environment; inherent
biodegradability tests have been designed to assess whether
the chemical has any potential for biodegradation under
optimized aerobic conditions. The OECD 301F test has been
performed for all the selected compounds, except for Pyr14FSI
for which the OECD 301B has been carried out. For soluble
compounds not classified as readily biodegradable according to
the OECD 301F test, the Zahn-Wellens test (OECD 302B) has
been carried out. For the purpose of the tests, stock solutions of
ILs were prepared in ultra-pure water in nominal concentrations
of 1g/liter. The water solubility of these stock solutions was
assessed visually and confirmed by total organic carbon (TOC)
analyses by comparing the TOC measurements with theoretical
o
•
•
•
qCDG = E(mCO2 −m )
(7)
CO2
o
•
•
2
Where qCDG is the heat release rate (kW or kW/m ), mCO2
•
and
m
CO2 are the mass flow rates of carbon dioxide carried by
-
1
the exhaust gases (Kg s ) when a combustion occurs or not,
respectively. To allow the HRR calculation in incomplete
combustion configuration, CO is added as a correction factor:
o
•
•
•
•
qCDG = E(mCO2 −m )+ E m
(8)
CO2
CO
CO
-
1
with ECO2 =13.3 MJ Kg
of CO ±11% and
2
-
1
ECO =11.1 MJ Kg of CO ±18%
.
In this work, the heat release rate values were calculated
making use of bothO consumption (OC calorimetry) as well as
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
organic content of the organic fractions of the related ILs. No
TOC measurement was performed for those compounds that
were qualified from observations as ‘non-soluble ‘.
Table 5. Biodegradation tests performed and their main characteristics
Test method
Concentration
Concentration of Applicability of
Biodegradability
test
Analytical
method
of
test inoculum in test testing
substance
test vessel
in vessel
method
reference
Poorly
30 mg/L of soluble;
CO
2
Evolution Test
Sturm
Ready
biodegradability
OECD 301 Respirometry: 10
–
20 mg
≤
(
Modified
B
CO evolution OC/L
2
suspended solids
adsorbing
compounds
Test)
Manometric
OECD 301 respirometry:
Poorly
30 mg/L of soluble;
1
5
00
mg/L;
Manometric
Respirometry Test
Ready
biodegradability
≤
0 – 100 mg
F
oxygen
suspended solids
adsorbing
compounds
ThOD/l
consumption
Dissolved
organic
carbon
Zahn-
Wellens/EMPA test
Inherent
biodegradability
50
–
400 200 – 1000 mg of Soluble
suspended solids compounds
OECD 302B
mg/DOC/L
removal
Acknowledgement
Key Words: Li-ion battery, [Pyr] based ionic liquids,
The authors would like to thank the following colleagues for their
outstanding contributions in operational testing: Jean-Pierre
Bertrand (fire calorimetry with the FPA), Peggy Gruez (oxygen
bomb calorimetry), Franck Gondelle and Estelle Michon
thermal stability, combustion, fire-induced toxicity,
biodegradability
(
biodegradability tests performance). Stefano Passerini, Sansik
Jeong and Gebrekidan Gebresilassie Eshetu acknowledge the
financial support of the Helmholtz Association.
References
Zhang, Angew. Chemie - Int. Ed. 2016, 55, 10662–10666.
[
11]
H.-G. Wang, S. Yuan, D.-L. Ma, X.-B. Zhang, J.-M. Yan,
Energy Environ. Sci. 2015, 8, 1660–1681.
[
[
1]
2]
P. Ribière, S. Grugeon, M. Morcrette, S. Boyanov, S.
Laruelle, G. Marlair, Energy Environ. Sci. 2012, 5, 5271–
[12]
[13]
[14]
[15]
[16]
[17]
H. Wang, S. Yuan, Z. Si, X. Zhang, Energy Environ. Sci.
2015, 8, 3160–3165.
H. G. Wang, S. Yuan, D. L. Ma, X. L. Huang, F. L. Meng, X.
B. Zhang, Adv. Energy Mater. 2014, 4, 1–7.
S. Yuan, Y. B. Liu, D. Xu, D. L. Ma, S. Wang, X. H. Yang, Z.
Y. Cao, X. B. Zhang, Adv. Sci. 2015, 2, 1–6.
X. L. Huang, D. Xu, S. Yuan, D. L. Ma, S. Wang, H. Y.
Zheng, X. B. Zhang, Adv. Mater. 2014, 26, 7264–7270.
A.-O. Diallo, A. B. Morgan, C. Len, G. Marlair, Energy
Environ. Sci. 2013, 6, 699.
5
280.
G. G. Eshetu, S. Grugeon, G. Gachot, D. Mathiron, M.
Armand, S. Laruelle, Electrochim. Acta 2013, 102, 133–141.
J.-M. Tarascon, M. Armand, Nature 2001, 414, 359–367.
J. Kalhoff, G. G. Eshetu, D. Bresser, S. Passerini,
ChemSusChem 2015, 8, 2154–2175.
[
[
3]
4]
[
[
5]
6]
G. G. Eshetu, J.-P. Bertrand, A. Lecocq, S. Grugeon, S.
Laruelle, M. Armand, G. Marlair, J. Power Sources 2014,
2
69, 804–811.
G. G. Eshetu, S. Grugeon, S. Laruelle, S. Boyanov, A.
Lecocq, J.-P. Bertrand, G. Marlair, Phys. Chem. Chem.
Phys. 2013, 15, 9145–55.
L. Chancelier, a O. Diallo, C. C. Santini, G. Marlair, T.
Gutel, S. Mailley, C. Len, Phys. Chem. Chem. Phys. 2014,
16, 1967–1976.
[
[
[
7]
8]
9]
Z. Chen, Y. Qin, Y. Ren, W. Lu, C. Orendorff, E. P. Roth, K.
Amine, Energy Environ. Sci. 2011, 4, 4023.
S. Yuan, X. L. Huang, D. L. Ma, H. G. Wang, F. Z. Meng, X.
B. Zhang, Adv. Mater. 2014, 26, 2273–2279.
X. L. Huang, R. Z. Wang, D. Xu, Z. L. Wang, H. G. Wang, J.
J. Xu, Z. Wu, Q. C. Liu, Y. Zhang, X. B. Zhang, Adv. Funct.
Mater. 2013, 23, 4345–4353.
[18]
[19]
S. S. Passerini, Energy Environ. Sci. 2015, 9, 49–61.
D. R. MacFarlane, N. Tachikawa, M. Forsyth, J. M. Pringle,
P. C. Howlett, G. D. Elliott, J. H. Davis, M. Watanabe, P.
Simon, C. a Angell, Energy Environ. Sci. 2014, 7, 232–250.
G. G. Eshetu, M. Armand, B. Scrosati, S. Passerini, Angew.
Chemie - Int. Ed. 2014, 53, 13342–13359.
[20]
[21]
M. Armand, F. Endres, D. R. MacFarlane, H. Ohno, B.
Scrosati, Nat. Mater. 2009, 8, 621–9.
[
10]
T. Sun, Z. J. Li, H. G. Wang, D. Bao, F. L. Meng, X. B.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
[
[
[
[
22]
23]
24]
25]
G. A. Elia, U. Ulissi, S. Jeong, S. Passerini, J. Hassoun,
Energy Environ. Sci. 2016, 9, 3210–3220.
L. G. Chagas, D. Buchholz, L. Wu, B. Vortmann, S.
Passerini, J. Power Sources 2014, 247, 377–383.
D. R. MacFarlane, S. A. Forsyth, J. Golding, G. B. Deacon,
Green Chem. 2002, 4, 444–448.
Huang, S. N. Liu, Green Chem. 2012, 14, 2001–2008.
S. A. Channiwala, P. P. Parikh, Fuel 2002, 81, 1051–1063.
A. O. Diallo, G. Fayet, C. Len, G. Marlair, Ind. Eng. Chem.
Res. 2012, 51, 3149–3156.
P. Glarborg, A. D. Jensen, J. E. Johnsson, Prog. Energy
Combust. Sci. 2003, 29, 89–113.
W. D. Woolley, P. J. Fardell, A. B. Atkinson, A. P. Verrall,
Fire Mater. 1978, 2, 122–131.
J. Neumann, S. Steudte, C.-W. Cho, J. Thöming, S. Stolte,
Green Chem. 2014, 16, 2174–2184.
S. Stolte, S. Steudte, A. Igartua, P. Stepnowski, Curr. Org.
Chem. 2011, 15, 1946–1973.
A. Jordan, N. Gathergood, Chem. Soc. Rev. 2015, 44,
8200–37.
S. Stolte, S. Steudte, O. Areitioaurtena, F. Pagano, J.
Thöming, P. Stepnowski, A. Igartua, Chemosphere 2012,
89, 1135–1141.
A. Bado-Nilles, A. O. Diallo, G. Marlair, P. Pandard, L.
Chabot, A. Geffard, C. Len, J. M. Porcher, W. Sanchez, J.
Hazard. Mater. 2015, 283, 202–210.
X. Lu, G. Burrell, F. Separovic, C. Zhao, J. Phys. Chem. B
2012, 116, 9160–9170.
M. A. Kiani, M. F. Mousavi, M. S. Rahmanifar, Int. J.
Electrochem. Sci. 2011, 6, 2581–2595.
S. Brohez, G. Marlair, C. Delvosalle, Fire Mater. 2006, 30,
131–149.
S. Brohez, G. Marlair, C. Delvosalle, Fire Mater. 2006, 30,
35–50.
W. Thornton, Philos. Mag. 1917, 3, 196–203.
S. Brohez, C. Delvosalle, G. Marlair, A. Tewarson, Fire Saf.
Sci. 2000, 265–276.
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
F. Castiglione, A. Famulari, G. Raos, S. V. Meille, A. Mele,
G. B. Appetecchi, S. Passerini, J. Phys. Chem. B 2014, 118,
1
3679–13688.
[
26]
C. Samorì, T. Campisi, M. Fagnoni, P. Galletti, A. Pasteris,
L. Pezzolesi, S. Protti, D. Ravelli, E. Tagliavini, ACS
Sustain. Chem. Eng. 2015, 3, 1860–1865.
[
[
27]
28]
C. Maton, N. De Vos, C. V Stevens, Chem. Soc. Rev. 2013,
4
2, 5963–77.
K. A. Kurnia, C. M. S. S. Neves, M. G. Freire, L. M. N. B. F.
Santos, J. A. P. Coutinho, J. Mol. Liq. 2015, 210, 264–271.
Y. Cao, T. Mu, Ind. Eng. Chem. Res. 2014, 53, asap.
M. Kosmulski, J. Gustafsson, J. B. Rosenholm,
Thermochim. Acta 2004, 412, 47–53.
M. T. Clough, K. Geyer, P. a Hunt, J. Mertes, T. Welton,
Phys. Chem. Chem. Phys. 2013, 15, 20480–95.
H. L. Ngo, K. LeCompte, L. Hargens, A. B. McEwen,
Thermochim. Acta 2000, 357–358, 97–102.
R. Ge, C. Hardacre, J. Jacquemin, D. W. Rooney, Ion. Liq.
From Knowl. to Appl. 2009, 43–60.
M. C. Kroon, W. Buijs, C. J. Peters, G. J. Witkamp,
Thermochim. Acta 2007, 465, 40–47.
[
[
29]
30]
[62]
[
[
[
[
[
[
31]
32]
33]
34]
35]
36]
[63]
[64]
[65]
[66]
H. Tokuda, K. Hayamizu, K. Ishii, A. Bin, H. Susan, M.
Watanabe, 2005, 6103–6110.
J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D.
Willauer, G. a. Broker, R. D. Rogers, Green Chem. 2001, 3,
[67]
[68]
1
56–164.
[69]
A. Tewarson, R. F. Pion, Combust. Flame 1976, 26, 85–
103.
[
[
37]
38]
M. Montanino, M. Carewska, F. Alessandrini, S. Passerini,
G. B. Appetecchi, Electrochim. Acta 2011, 57, 153–159.
A. Messadi, A. Mohamadou, S. Boudesocque, L. Dupont, P.
Fricoteaux, A. Nguyen-Van-Nhien, M. Courty, J. Mol. Liq.
2
013, 184, 68–72.
[
[
39]
40]
E. T. Fox, J. E. F. Weaver, W. A. Henderson, 2012.
R. Alcalde, G. García, M. Atilhan, S. Aparicio, Ind. Eng.
Chem. Res. 2015, 54, 10918–10924.
[
41]
A. Filippov, M. Taher, F. U. Shah, S. Glavatskih, O. N.
Antzutkin, Phys. Chem. Chem. Phys. 2014, 16, 26798–
2
6805.
[
[
[
42]
43]
44]
A. B. McEwen, H. L. Ngo, K. LeCompte, J. L. Goldman, J.
Electrochem. Soc. 1999, 146, 1687–1695.
M. Kerner, N. Plylahan, J. Scheers, P. Johansson, Phys.
Chem. Chem. Phys. 2015, 17, 19569–19581.
J. Reiter, S. Jeremias, E. Paillard, M. Winter, S. Passerini,
Phys. Chem. Chem. Phys. Phys. Chem. Chem. Phys 2013,
1
5, 2565–2571.
Z. Bin Zhou, H. Matsumoto, K. Tatsumi, Chem. - A Eur. J.
004, 10, 6581–6591.
[
[
45]
46]
2
H. S. Schrekker, D. O. Silva, M. a. Gelesky, M. P. Stracke,
C. M. L. Schrekker, R. S. Gonçalves, J. Dupont, J. Brazilian
Chem. Soc. 2008, 19, 426–433.
[
[
[
[
47]
48]
49]
50]
Z. Bin Zhou, H. Matsumoto, K. Tatsumi, Chem. - A Eur. J.
2
006, 12, 2196–2212.
Z. Bin Zhou, H. Matsumoto, K. Tatsumi, Chem. - A Eur. J.
005, 11, 752–766.
2
M. J. Monteiro, F. F. Camilo, M. C. C. Ribeiro, R. M. Torresi,
J. Phys. Chem. B 2010, 114, 12488–12494.
G. G. Eshetu, J. P. Bertrand, A. Lecocq, S. Grugeon, S.
Laruelle, M. Armand, G. Marlair, J. Power Sources 2014,
2
69, 804–811.
[
[
51]
52]
A. Lecocq, G. G. Eshetu, S. Grugeon, N. Martin, S. Laruelle,
G. Marlair, J. Power Sources 2016, 316, 197–206.
G. G. Eshetu, S. Grugeon, S. Laruelle, S. Boyanov, A.
Lecocq, J.-P. Bertrand, G. Marlair, Phys. Chem. Chem.
Phys. 2013, 15, 9145–9155.
[
53]
H. J. Liaw, C. C. Chen, Y. C. Chen, J. R. Chen, S. K.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701006
FULL PAPER
Entry for the Table of Contents
FULL PAPER
ILs, as electrolyte components
for high-energy batteries, offer a
series of advantages with
respect to molecular solvent-
based, conventional electrolytes.
Among others, the superior
thermal stabilities and flame
retardancy are certainly keys to
enable the realization of safer
Dr. Gebrekidan Gebresilassie Eshetu,
[
a,b,c,d]
Dr. Sangsik Jeong,[a,b] Dr. Pascal
Pandard, [e] Dr. Amandine Lecocq, [e] Dr.
Guy Marlai, [e] Prof. Dr. Stefano
Passerini*[a,b]
Page No. – Page No.
BeneficialꢀProper.esꢀofꢀILsꢀ
Comprehensive insights into the
thermal stability, biodegradability
and combustion chemistry of
Pyrrolidinium-based Ionic
Liquids
batteries. Having
a negligible
Wideꢀmoltenꢀrangeꢀꢀ
Improvedꢀthermalꢀ&ꢀchemicalꢀstabili.esꢀ
Negligibleꢀvola.lityꢀ
Highꢀionicꢀconduc.vityꢀ
Highꢀpolarityꢀ
Excellentꢀsolvent-abilityꢀ
Flameꢀretardancyꢀ
Highꢀflashꢀpointꢀ
volatility they are not easily
dispersed in the environment.
Additionally, they offer, in some
cases, biodegradability further
reducing the environemental
impact.
i
This article is protected by copyright. All rights reserved.