Journal of Power Sources xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Power Sources

journal homepage: www.elsevier.com/locate/jpowsour

An alternative composite polymer electrolyte for high performances

lithium battery
Vittorio Marangon a, Yoichi Tominaga b, c, Jusef Hassoun a, c, d, *
a
University of Ferrara, Department of Chemical and Pharmaceutical Sciences, Via Fossato di Mortara 17, 44121, Ferrara, Italy
b
Tokyo University of Agriculture and Technology, Graduate School of Bio-Applications and Systems Engineering (BASE), 2-24-16, Naka-cho, Koganei-shi, Tokyo, 184-
8588, Japan
c
Institute of Global Innovation Research (GIR), Tokyo University of Agriculture and Technology (TUAT), Tokyo, Japan
d
National Interuniversity Consortium of Materials Science and Technology (INSTM) University of Ferrara Research Unit, University of Ferrara, Via Fossato di Mortara,
17, 44121, Ferrara, Italy

H I G H L I G H T S G R A P H I C A L A B S T R A C T

� A new composite polymer electrolyte is

reported for high-performances lithium
battery.
� The polymer consists of polyethylene-
glycol dimethyl-ether (MW 2000 g
mol 1).
� LiTFSI salt, LiNO3 film-forming additive,
and nanometric SiO2 filler are used.
� The electrolyte has conductivity over
10 4 S cm 1 above 45 � C and high
stability.
� A polymer Li/LiFePO4 cell at 50 � C sta­
bly delivers a capacity of 150 mAh g 1.

A R T I C L E I N F O A B S T R A C T

Keywords: A composite electrolyte consisting of polyethylene-glycol dimethyl-ether (MW 2000 g mol 1), lithium bis(tri­
Polyethylene glycol dimethyl ether fluoromethanesulfonyl)imide (LiTFSI) conducting salt, lithium nitrate (LiNO3) film-forming additive, and
PEGDME MW2000 nanometric silica (SiO2) filler, is herein obtained by a scalable solvent casting pathway and thoroughly inves­
Polymer electrolyte
tigated for application in lithium metal polymer battery. Structure and morphology of the electrolyte are
Composite
Lithium battery
investigated by X-ray diffraction (XRD) and scanning electron microscopy (SEM), respectively, while its elec­
High-performances trochemical features are revealed by electrochemical impedance spectroscopy (EIS), cyclic voltammetry (CV),
chronoamperometry, and galvanostatic cycling tests in lithium cell. The room-temperature structure of the
composite electrolyte reflects the combination into a complex of the membrane components, while its
morphology appears smooth with uniform distribution of salts and ceramic. The electrolyte shows an ionic
conductivity over 10 4 S cm 1 above 40 � C promoted by repeated heating and cooling, lithium transference
number ranging from 0.22 at 45 � C to 0.27 at 70 � C, low interphase resistance and polarization in lithium cell,
and an electrochemical stability window extending above 4.4 V. These optimal features allow the membrane to
operate in a lithium cell with LiFePO4 cathode at 50 � C, with specific capacity exceeding 150 mAh g 1 and
coulombic efficiency approaching 100% over prolonged cycling.

* Corresponding author. University of Ferrara, Department of Chemical and Pharmaceutical Sciences, Via Fossato di Mortara 17, 44121, Ferrara, Italy.
E-mail address: jusef.hassoun@unife.it (J. Hassoun).

https://doi.org/10.1016/j.jpowsour.2019.227508
Received 1 August 2019; Received in revised form 11 November 2019; Accepted 26 November 2019
0378-7753/© 2019 Elsevier B.V. All rights reserved.

Please cite this article as: Vittorio Marangon, Journal of Power Sources, https://doi.org/10.1016/j.jpowsour.2019.227508
V. Marangon et al.
1. Introduction
Lithium ion battery is the most efficient and diffused energy storage
device for powering portable electronics, such as smartphones and
laptops, as well as hybrid-electric vehicles (HEVs), and electric vehicles
(EVs) [1]. This intriguing system has presently reached an energy den­
sity as high as 200 Wh kg 1 [2], high efficiency and long cycle life [3,4].
A further increase of demand for efficient energy storage has been
recently promoted by the expected large scale diffusion of renewable
power production plants and, in particular, of EVs due to possible
environmental issues ascribed to the excessive greenhouse gas emission
[5]. Particularly challenging targets, in terms of high-energy density and
safety content, have been posed by the EV market for extending to
commercially suitable level the driving-range, and to limit possible
hazards [6]. Battery configurations exploiting the lithium metal, which
has a theoretical capacity as high as 3860 mAh g 1 and the lowest redox
potential ( 3.04 V vs SHE), may actually satisfy the energy content
requested by EVs and reach a practical capacity higher than that
ascribed to the commercial Li-ion cell [7–10]. However, the use of
lithium is presently hindered by dendritic structures growth at the metal
surface during cell charge, which may lead to short circuit and possible
thermal runaways in presence of flammable electrolyte [6,11,12], such
as the solution typically used in Li-ion commercial battery which is
formed by highly volatile alkyl carbonates (e.g., dimethyl-carbonate,
DMC, and ethyl-carbonate, EC) [13]. Gel-type electrolytes based on a
polymer backbone, such as polyvinildene fluoride (PVDF), trapping low
fraction of carbonate solvents and lithium salt, appeared as a possible
solution for mitigating the safety issues [14,15]. Furthermore, solutions
using lowly flammable solvents, such as glymes of various chain length,
have shown very attracting characteristics in terms of thermal stability,
ion conductivity and lithium transport properties [16–21]. Beside gels
and liquid glymes, solid electrolytes exploiting dry amorphous polymers
capable of dissolving lithium salts for allowing the ion conduction, such
as poly (ethylene oxide) (PEO) [22–25], and polyethylene carbonate
(PEC) [26–28] appeared as the most suitable configuration for achieving
a safe use of the lithium metal in cell. In particular, PEO-based elec­
trolytes are lowly reactive due to the stability of their –(CH2CH2O)n–
polymer chains [29]. Furthermore, PEO can efficiently complex and
dissolve various lithium salts, such as lithium bis(tri­
fluoromethansulfonyl) imide (LiTFSI) [7], lithium tri­
fluoromethanesulfonate (LiCF3SO3) [23], and lithium bis(oxalato)
borate (LiBOB) [30], to achieve Liþ-ions conductivity values suitable for
battery application (i.e., higher than 10 4 Scm 1), however only at
temperatures above 65 � C [31]. Indeed, PEO is in large part crystalline at
the room temperature, while it becomes predominantly amorphous at a
temperature higher than 65 � C [32], thus allowing sufficient intra- and
inter-chain motion of Liþ ions [33,34]. Several strategies have been
adopted for improving the ionic conductivity of PEO-based electrolytes
at the lower temperatures, including the addition of liquid plasticizers
[35,36], ionic liquids [37–39], and the use of ceramic powders, such as
Al2O3 or SiO2, which may limit the formation of crystalline phases
[40–42]. End-capped polyethylene glycol dimethyl ether (i.e.,
CH3–(OCH2CH2)n–OCH3) with molecular weight higher than 1000 g
mol 1 is solid at the room temperature, and potentially represents a
suitable media for solid polymer electrolytes operating at temperatures
lower than that ascribed to PEO-based ones due to its lower melting
point [43]. Our previous studies demonstrated the applicability of low
molecular weight liquid PEGDMEs, e.g., MW 500 and 250 g mol 1, in
lithium ion [20] and Li/S cell [21], however, to the best of our knowl­
edge, a solid polymer exploiting PEGDME with high molecular weight
has not been previously reported in literature. Nevertheless, PEGDME
may actually suffer by issues ascribed to reactivity with lithium metal
which can lead to the increase of the cell resistance and the decrease of
the efficiency, particularly within the solid polymer configuration which
requires a temperature higher than 25 � C to reach the operating con­
dition [44]. The use of sacrificial additives, such as LiNO3, which form a
2
Journal of Power Sources xxx (xxxx) xxx
protective layer at the lithium surface for avoiding parasitic reactions,
has been demonstrated as a valid strategy, particularly suitable for Li/S
battery [45–47], and it may actually allow the efficient use of high
molecular weight PEGDME-based solid electrolyte in lithium cell.
Therefore, we propose herein an alternative composite electrolyte
exploiting PEGDME with an average molecular weight of 2000 g mol 1,
LiTFSI salt, LiNO3 film-forming additive, and fumed SiO2 ceramic filler,
for application in lithium battery operating at temperatures lower than
the ones ascribed to the common solid polymers (e.g., at 50 � C). The
solid electrolyte is fully characterized in terms of structure, morphology,
lithium ion transport properties and Li/electrolyte interphase stability,
while its applicability in lithium battery is demonstrated by exploiting a
LiFePO4 cathode with theoretical capacity of 170 mAh g 1 delivered at
3.5 V. The potentially scalable synthesis adopted in this work, the ex­
pected moderate cost, and the results reported below may actually
suggest the composite electrolyte as a suitable candidate for application
in lithium metal battery.
2. Experimental section
Lithium bis(trifluoromethansulfonyl)imide (LiTFSI, Sigma-Aldrich)
and lithium nitrate (LiNO3, Sigma-Aldrich) salts were mixed with
polyethylene-glycol dimethyl-ether solid powder (CH3O(C2H4O)nCH3,
average molecular weight 2000 g mol 1, Sigma-Aldrich) by a ratio of 1
mol of each salt in 1 kg of polymer. Fumed silica (SiO2, fumed, average
particle size: 0.007 μm, Sigma-Aldrich) was then added to the above
polymer/salts mixture by a 10% weight ratio. The components were
subsequently dissolved in acetonitrile (ACN, Sigma Aldrich) to obtain a
viscous slurry. Upon two days of magnetic stirring at the room tem­
perature (Fig. S1a in the Supplementary Material), the slurry was cast
onto plastic foil (Mylar) using a doctor blade with a thickness of 1.5 mm
(Figs. S1b and c). The slurry was dried at about 45 � C overnight,
transferred into a Büchi oven and heated under vacuum for 5 days at 45
�
C and for 2 days at 60 � C (Fig. S1d). The slurry was finally introduced
into an Ar-filled glove box (MBraun, O2 and H2O content below 1 ppm)
where it was left for about 20 days at 25 � C until complete crystallization
(Figs. S1e–i). The composite solid polymer electrolyte will be subse­
quently indicated by the acronym PEGDME_CPE, while the PEGDME
polymer with average molecular weight 2000 g mol 1 by the acronym
PEGDME2000. A membrane with the same composition except for
LiNO3 additive, indicated as blank, was prepared for comparison. It may
be remarked that the PEGDME_CPE membrane has shown a greater
mechanical stability with respect to the blank membrane.
The PEGDME_CPE structure was investigated by XRD at the room
temperature (25 � C) by using a Bruker D8 Advance diffractometer
equipped with a Cu Kα source, performing a scan in the 2θ range be­
tween 10� and 40� at a rate of 10 s per step with angle increments of
0.02� . The patterns of the single components of the PEGDME_CPE, that
is, PEGDME2000, LiTFSI, LiNO3 and fumed SiO2, were recorded using
the same conditions.
A Zeiss EVO 40 microscope equipped with a LaB6 thermoionic beam
was used to study the membrane morphology through SEM technique,
while energy dispersion X-ray analyses (EDX) were performed using a X-
ACT Cambridge Instruments analyzer.
The solid configuration of the PEGDME_CPE membrane was inves­
tigated at various temperatures by a heating-cooling treatment consist­
ing of 3 steps, that is, heating at 50 � C for 3 h, heating at 70 � C for 3 h
and, finally, cooling back to the room temperature (Fig. S2).
All the electrochemical measurements reported below were per­
formed using CR2032 coin-type cells (MTI) assembled in an Ar-filled
glove box (MBraun, O2 and H2O content below 1 ppm). The ionic con­
ductivity of the PEGDME_CPE was evaluated by performing EIS mea­
surements in a blocking electrode SS/PEGDME_CPE/SS cell, using an O-
ring (Mylar) with an internal diameter of 10 mm and thickness of 120
μm to fix the cell constant at 0.0153 cm 1. The impedance spectra were
recorded by a VersaSTAT MC Princeton Applied Research (PAR,
V. Marangon et al. Journal of Power Sources xxx (xxxx) xxx

AMETEK) instrument applying a signal of 10 mV in the 500 kHz-100 Hz
frequency range.
The Bruce-Vincent-Evans method [48] was applied to determine the
Liþ transference number (tþ) of the electrolyte in symmetrical
Li/PEGDME_CPE/Li cell at 45, 50 and 70 � C. The cells were assembled
by using three O-rings (CS Hyde, 23-5FEP-2-50) with an internal
diameter of 10 mm and thickness of 127 μm holding the polymer elec­
trolyte and 2 lithium disks. The chronoamperometric tests were per­
formed by a VersaSTAT MC Princeton Applied Research (PAR, AMETEK)
instrument using a voltage of 30 mV for 90 min, and the impedance
spectra were recorded before (initial state) and after (steady state) cell
polarization with the same instrument applying a signal of 10 mV in the
500 kHz-100 mHz frequency range. The values of the Liþ transference
number were calculated using equation (1) [48]:
iss ðΔV R0 i0 Þ
tþ ¼ � (1)
Rss iss Þ
i0 ðΔV
(Timcal), and PVDF 6020 (Solvay Solef) dissolved by a weight ratio of
80:10:10 in N-methyl pyrrolidone (NMP, Sigma–Aldrich). The slurry
was then dried at 70 � C for 3 h under air, cut into electrode disks with a
diameter of 10 mm, and heated at 110 � C under vacuum for 3 h to
remove traces of water and solvent. The lithium cells were assembled by
using two O-rings (CS Hyde, 23-5FEP-2-50) with an internal diameter of
10 mm and thickness of 127 μm holding the polymer electrolyte, a
lithium disk, and the LFP working electrode. EIS tests were performed
from the room temperature to 70 � C at the open circuit voltage (OCV) of
the polymer cell, recording the impedance spectra through a VersaSTAT
MC Princeton Applied Research (PAR, AMETEK) instrument applying a
signal of 10 mV in the 500 kHz-100 mHz frequency range. Galvanostatic
cycling measurements were carried out at 50 � C and 70 � C at various
C-rates (1C ¼ 170 mA g 1, referring to LFP mass) by using a MACCOR
Series 4000 battery test system. All the impedance spectra and related
Nyquist plots were analyzed by using a Boukamp tool [50,51] taking
into account only fits with a χ2 of the order of 10 4 or lower.

where i0 and iss are the current values at the initial and steady state,
3. Results and discussion
respectively, ΔV is the applied voltage, R0 and Rss are the interphase
resistance values before and after cell polarization, respectively, calcu­
Figure S1 in Supplementary Material reports the photographic im­
lated from the impedance spectra.
ages of the various steps adopted for the solvent casting procedure of the
The electrochemical stability of the PEGDME_CPE was investigated
PEGDME_CPE. The figure reveals that the membrane becomes self-
by performing lithium stripping-deposition tests on Li/Li cells. A con­
standing upon 20 days of aging in glove box at the room temperature,
stant current of 0.1 mA cm 2 was applied using a MACCOR Series 4000
which represents a key configuration suitable for adopting the electro­
battery test system at the temperature of 45, 50 and 70 � C. Afterwards, a
lyte in a solid-polymer lithium battery. Structure and morphology of the
constant temperature of 50 � C was selected for the subsequent mea­
aged solid electrolyte are detected by XRD, SEM and EDX in Fig. 1 in
surements, which were carried out by cooling back the cells at the above
order to investigate the dispersion of the various components into the
temperature after a first heating treatment at 70 � C for one day. The
solid polymer matrix, which is generally promoted by the solvent casting
PEGDME_CPE/Li interphase resistance was evaluated by EIS measure­
pathway [38]. Indeed, homogeneous dispersions of the components may
ments upon symmetrical Li/Li cell aging. The impedance spectra were
actually lead to the formation of salt/polymer/ceramic complexes with a
recorded by using a VersaSTAT MC Princeton Applied Research (PAR,
structure differing by the one of the single component, thus improving
AMETEK) instrument applying a signal of 10 mV in the 500 kHz-100
both the mechanical and electrochemical characteristics of the electro­
mHz frequency range. The electrochemical stability window of the
lyte [40,41]. The patterns of Fig. 1a show that the PEGDME_CPE (blue)
electrolyte was determined in Li/PEGDME_CPE/SP cells through linear
exhibits various peaks, particularly within the 17� –28.5� 2θ range, and
sweep voltammetry (LSV) from the open circuit voltage (OCV) to 5 V in
aside broad signal in pattern baseline. The latter may be likely attributed
the anodic region, and cyclic voltammetry (CV) within the 0.01–2.0 V
to the SiO2 ceramic filler (see the SiO2 cyan pattern in Fig. 1a for com­
range in the cathodic region. The anodic region was further investigated
parison), however the complex pattern of PEGDME_CPE appears only
through a chronoamperometry test at 50 � C in the potential windows
partially ascribed to the single species forming the electrolyte. Indeed,
ranging from 4.0 to 4.7 V. The cell potential was initially increased from
only few peaks for each component of the electrolyte, i.e., PEGDME2000
OCV to 3.9 V by linear scan voltammetry. The cell was held at 4 V and
the current flow measured for 1 h, then the potential was increased by
steps of 0.1 V every 1 h from 4 V until 4.7 V and the current flow
measured at the corresponding potentials by subsequent
chronoamperometry.
The working electrodes of the above cells (SP) were formed by doctor
blade casting of a slurry, containing Super P carbon (Timcal) and PVDF
6020 (Solvay Solef) dissolved by a weight ratio of 90:10 in N-methyl
pyrrolidone (NMP, Sigma–Aldrich), on an Al foil for anodic scan, and Cu
foil for cathodic scan. The slurry was then dried at 70 � C for 3 h under
air, cut into electrode disks with a diameter of 10 mm, and heated at 110
�
C under vacuum for 3 h to remove traces of water and solvent. The first
step at 70 � C is typically used in order to allow uniform drying by slow
solvent evaporation, thus avoiding the formation of bubbles or unde­
sired cracks on the wet electrode surface, while the second one at 110 � C
is performed to ensure proper performances of the electrode in lithium
battery and avoid side reactions. The cells were assembled by using two
O-rings (CS Hyde, 23-5FEP-2-50) with an internal diameter of 10 mm
and thickness of 127 μm holding the polymer electrolyte, a lithium disk, Fig. 1. (a) X-ray diffraction patterns of the PEGDME_CPE membrane (blue),
and the selected SP carbon working electrode. LSV and CV measure­ PEGDME2000 (orange), LiTFSI (yellow) and LiNO3 (purple) salts, and of the
ments were both performed by using a VersaSTAT MC Princeton Applied fumed SiO2 ceramic filler (cyan); (b) SEM image of the PEGDME_CPE mem­
Research (PAR, AMETEK) instrument with a scan rate of 0.1 mV s 1. brane; (c) EDX map of the PEGDME_CPE membrane (insets on the right show
LiFePO4 (LFP) was selected as the cathode material to evaluate the the single EDX elemental maps) and (d) its elemental composition. All mea­
electrochemical performances of PEGDME_CPE in lithium battery [49]. surements collected at the room temperature, where membrane and compo­
The LFP electrode was formed by doctor blade casting on aluminum foil nents are solid. See experimental section for sample’s acronym. (For
of a slurry containing the active material (LiFePO4), Super P carbon interpretation of the references to color in this figure legend, the reader is
referred to the Web version of this article.)

3
V. Marangon et al. Journal of Power Sources xxx (xxxx) xxx

(orange pattern), LiTFSI (yellow pattern) and LiNO3 (purple pattern),

may be identified, thus suggesting the formation of a new crystalline
structure by the rearrangement of the various components into a com­
plex during the synthesis procedure [40,41]. Several studies, in partic­
ular those focusing on PEO-based solid polymer electrolytes, have
indicated the role of the polymer/salt complexes for enhancing the ion
transport, and the multiple effect of efficiently dispersed nanometric
ceramics for favoring both the mechanical stability and the electro­
chemical characteristics of the electrolyte [22–24,40,41]. The uniform
dispersion of the various components into the PEGDME_CPE membrane
is further evidenced by EDX (Fig. 1c) performed on a SEM image of a
micrometric fractured portion of the surface, which appears uniform
without signs of grains or discontinuities (Fig. 1b). Furthermore, the
absence of impurities into the membrane is suggested by the weight
ratio of the elements reported in Fig. 1d, which shows weak signals of
fluorine, silicon, sulfur and nitrogen, attributed only to salts and ceramic
filler, beside the expected 87.6% of carbon and oxygen mainly related to
the polymeric chains of the PEGDME2000.
It is worth mentioning that the absence of contaminants and impu­
rities, promoted by the synthesis conditions adopted herein, may be
actually reflected into high efficiency of the PEGDME_CPE membrane in
lithium cell.
The electrolyte behavior in terms of conductivity change by repeated
cooling and heating, that is, the thermal history, plays a key role in
determining its applicability in lithium cell, and for setting up proper
operating protocols for achieving suitable performances and low po­
larization [52]. Fig. 2 reports the conductivity Arrhenius plots of the
PEGDME_CPE membrane upon six subsequent cooling/heating scans
from about 70 � C to room temperature. Relevantly, the plots of Fig. 2a
show a conductivity of the order or higher than 10 4 S cm 1 above 45
�
C, that is, a temperature much lower than the one ascribed to other
solid polymer electrolytes used in lithium battery such as PEO-based Fig. 2. (a) Arrhenius conductivity plots during subsequent thermal cycles of
ones that favors instead efficient and suitable ion conduction only at cooling and heating of the PEGDME_CPE from about 75 � C to room temperature
temperatures higher than 60 � C [42]. Furthermore, the figure reveals a (25 � C), and (b) corresponding individual cooling/heating cycles. Conductivity
very interesting and progressive change of the crystallization/melting determined by EIS using frequency ranging from 500 kHz to 100 Hz, with signal
temperature by the ongoing of cycles. Indeed, the single thermal cycle of amplitude of 10 mV. See experimental section for sample’s acronym and
Fig. 2b reveals a drop of the ionic conductivity during the first cooling experimental setup used for determining the conductivity. (c) XRD pattern of
the PEGDME_CPE membrane after heating at 70 � C and subsequent cooling to
scan (orange) below 32 � C due to membrane crystallization, while the
room temperature (25 � C), and (d) corresponding SEM image.
subsequent heating scan shows the two slope regions typically observed
in the Arrhenius plots of solid polymer electrolytes [23]. The above
heating trend may be reasonably represented by two different lines, investigated by performing XRD and SEM measurements on PEGD­
where the first, from room temperature to 45 � C, represents the con­ ME_CPE upon heating at 70 � C and subsequent cooling down to room
ductivity of an electrolyte mainly formed by crystalline part, instead the temperature. The XRD pattern (Fig. 2c) reveals only a broad signal be­
second from 45 � C to higher temperatures indicates the conductivity of tween 17� and 25� and the absence of defined diffraction, thus sug­
the electrolyte in its amorphous state [32,43]. The subsequent scans gesting membrane amorphization in line with the conductivity trends
evidence a continuous increase of the room temperature ionic conduc­ observed in Fig. 2a and b which indicate the gradual increase of the
tivity from about 4 � 10 6 S cm 1 during the first scan to 4 � 10 5 S conductivity at the lower temperatures by the thermal cycles. Further­
cm 1 during the last one, i.e., by a factor of 10, and at the same time a more, the SEM image (Fig. 2d) after the above heating/cooling protocol
progressive overlapping of heating and cooling trends into a displays a very flat, uniform electrolyte surface without significant signs
Vogel-Tammann-Fulcher (VTF) profile [38,53]. This behavior may be of cracks, therefore suggesting an actual improvement of the mechanical
attributed to initial kinetic limits avoiding fast crystallization and sub­ and interfacial properties of the membrane which can efficiently
sequent melting during cooling and heating scans, respectively, thus compensate the contact resistance between electrodes and the polymer
leading to the hysteresis loop observed in Fig. 2. electrolyte in the cell.
The mitigation and progressive vanishing of the above mentioned In order to study the retention of the solid configuration of PEGD­
loop may be most likely ascribed to the rearrangement of the polymer ME_CPE, a heating-cooling treatment was carried out on the membrane
chains into a thermally stable complex [33]. This favorable condition and the results are depicted in Fig. S2 in the Supplementary Material.
may be promoted both by heating/cooling cycles and by a plasticizing Fig. S2 shows that the membrane is characterized by a solid configura­
effect of the nanometric SiO2 particles dispersed into the electrolyte tion at room temperature (Fig. S2a, in agreement with Figs. S1h and S1i
which can actually prevent the segregation of the polymer from the in Supplementary Material) and during all the steps of the heating
electrolyte complexes, avoid excessive crystallization during cooling treatment, that is, after heating for 3 h at 50 � C (Fig. S2b) and, subse­
[40,42] and enhance the ion transport [33]. Remarkably, the last scan in quently, for 3 h at 70 � C (Fig. S2c). It is worth noting that after being
Fig. 2 (pink) shows that the PEGDME_CPE has an ionic conductivity heated at 70 � C, PEGDME_CPE presents a slight transparency, most
ranging from about 4 � 10 5 S cm 1 at the room temperature to about 4 likely due to amorphization processes discussed above. When cooled
� 10 4 Scm 1 at 73 � C, which are considered values well suitable for back to the room temperature (Fig. S2d), the membrane recovers the
application of a solid polymer electrolyte in lithium battery. initial self-standing condition. Therefore, Fig. S2 demonstrates that the
The effect of the thermal treatment on the electrolyte is further solid nature of PEGDME_CPE is retained from room temperature to 70

4
V. Marangon et al. Journal of Power Sources xxx (xxxx) xxx

�
C. However, partial membrane softening at the higher temperature
values is not excluded.
The ion transport characteristics of the PEGDME_CPE at various
temperatures are further investigated in Fig. 3 by determining the Liþ
transference number (tþ) by the Bruce-Vincent-Evans method [48] at
45, 50 and 70 � C (see the experimental section for details). Fig. 3a re­
ports the chronoamperometric curves for each temperature, while inset
shows the Nyquist plots recorded through EIS before and after cells
polarization (see Table 1 for the parameter values).
The tþ values of PEGDME_CPE (Table 1) calculated by Equation (1)
and the data of Fig. 3 range from 0.22 at 45 � C, to 0.23 at 50 � C, and to
0.27 at 70 � C, with a slight increase possibly due to a more efficient
anion solvation by polymer chains at the higher temperature which fa­
vors the lithium ion motion [23]. The tþ values of PEGDME_CPE are
comparable to those of solid polymer electrolytes such as PEO-based
ones at similar temperatures [54,55], however they are lower than the
ones ascribed to the liquid electrolytes, such as glyme-based solutions,
which generally approach and exceed 0.5 [18]. The relatively low tþ
expected by the polymer electrolyte design can actually limit the rate
capability of the PEGDME_CPE compared to liquid electrolytes, however
the solid state of the membrane likely ensures good safety content and
thermal stability, which represent key factors for allowing the use of the
high energy lithium metal in efficient and scalable battery. It is worth Fig. 4. Nyquist plots recorded by EIS measurements at the open circuit voltage
noting that the Nyquist plots of Fig. 3a are characterized by a (OCV) of the Li/PEGDME_CPE/LFP cell (a) performed every 2 h after cell as­
low-frequency linear Warburg element due to the Liþ ions diffusion, a sembly during heating from room temperature (25 � C) to 70 � C; (b) at the
middle-frequency semicircle ascribed to the interphase, and by a steady state upon lowering the temperature back to 50 � C. Frequency range:
high-frequency element (only at temperatures lower than 70 � C) most 500 kHz–100 mHz. Signal amplitude: 10 mV. See experimental section for
likely due to grain boundary or heterogeneity into the electrolyte with sample’s acronym.
resistances gradually decreasing by increasing the temperature. Grain
boundary and heterogeneity may be associated to the coexistence of demonstrated by the conductivity measurements discussed in Fig. 2.
crystalline and amorphous phases into the electrolyte structure, in This important aspect is further clarified and confirmed by the Nyquist
particular at the lower temperatures, as well as to possible agglomera­ plots of the EIS measurements carried out on a Li/PEGDME_CPE/Li cell
tion of the ceramic filler [23,42]. The electrolyte heterogeneity may be from the room temperature to 70 � C reported in Fig. S3 and corre­
strongly mitigated by adequate heat thermal cycles, performed in order sponding NLLSQ analysis in Table S1, which actually suggest gradual
to achieve the optimal characteristics of the membrane, as indeed crystallization during cooling and formation of homogeneous electrolyte

Fig. 3. (a) Chronoamperometric profiles of a sym­

metrical Li/PEGDME_CPE/Li cell used for deter­
mining the Liþ transference number of PEGDME_CPE
at 45 � C (cyan), 50 � C (yellow) and 70 � C (purple).
Chronoamperometric polarization voltage: 30 mV,
inset shows the corresponding Nyquist Plots ob­
tained by EIS at the initial and steady state for each
temperature. Frequency range: 500 kHz-100 mHz.
Signal amplitude: 10 mV. (b) Lithium stripping-
deposition galvanostatic test performed in symmet­
rical Li/PEGDME_CPE/Li cell at 45 � C (cyan), 50 � C
(yellow) and 70 � C (purple), inset shows a prolonged
lithium stripping-deposition test performed at 50 � C.
Applied Current: 0.1 mA cm 2. (c) Cyclic Voltam­
metry in the cathodic region (yellow) and linear
sweep voltammetry in the anodic region (red) per­
formed at 50 � C with a scan rate of 0.1 mV s 1 in
lithium cell using the PEGDME_CPE as the separator
and Super P carbon coated on copper or aluminum,
respectively, as the working electrode. Cyclic vol­
tammetry range: 0.01–2 V. Inset shows the chro­
noamperometry test at potentials from 4.0 V to 4.7 V
with increment of 0.1 V every 1 h at 50 � C (see
experimental section for further details). (d) Trend
of interphase resistance upon Li/PEGDME_CPE/Li
cell aging at 50 � C determined by NLLSQ fit of the
corresponding EIS Nyquist plots, reported in figure
inset. Frequency range: 500 kHz–100 mHz. Signal
amplitude: 10 mV. See experimental section for
sample’s acronym. (For interpretation of the refer­
ences to color in this figure legend, the reader is
referred to the Web version of this article.)

5
V. Marangon et al. Journal of Power Sources xxx (xxxx) xxx

Table 1 grain boundary element, a middle-frequency semicircle ascribed to the

Parameters used to determine lithium transference number of PEGDME_CPE at
different temperatures using the Bruce-Vincent-Evans equation (1). Resistance
values were obtained by NLLSQ analyses of the Nyquist plots displayed in
Fig. 3a, recorded using a frequency ranging from 500 kHz to 100 mHz with
signal amplitude of 10 mV, upon chronoamperometry performed for 90 min by
applying a voltage of 30 mV to a Li/PEGDME_CPE/Li symmetrical cell. See
experimental section for sample’s acronym.
Temperature Initial Steady Initial Steady Liþ
[� C] current state resistance state transference
(i0) [A] current (R0) [Ω] resistance number (tþ)
(iss) [A] (Rss) [Ω]
45 6.8 � 2.5 � 231 235 0.22
10 5 10 5
50 9.5 � 3.5 � 153 158 0.23
10 5 10 5
70 1.5 � 5.6 � 78 75 0.27
10 4 10 5
by heating. Fig. 3b reports the voltage profiles of the lithium strip­
ping/deposition tests performed at 45, 50 and 70 � C in a Li/PEGD­
ME_CPE/Li cell in order to evaluate the electrochemical stability of the
electrolyte. The low and constant overvoltage exhibited by the cell at
each temperature, that is 45 mV at 70 � C, 90 mV at 50 � C and 120 mV at
45 � C, and the absence of lithium dendrite formation reflect a relevant
stability of the Li/electrolyte interphase which may promote the safety
content of the polymer cell. Furthermore, inset of Fig. 3b reports the
voltage profiles of lithium stripping/deposition test prolonged to over
350 h at 50 � C, which is considered a suitable temperature for achieving
good electrochemical performance of the electrolyte and, at the same
time, a sufficiently moderate operating value, adequate for a wide range
of application, including electric vehicles (EVs) [1].
The figure reveals a pronounced stability, and an overvoltage limited
to about 100 mV during charge and discharge, thus suggesting a well
reversible lithium stripping-deposition upon cycling. The electro­
chemical stability window of the PEGDME_CPE is evaluated by the
voltammograms displayed in Fig. 3c in the anodic (red) and cathodic
(yellow) regions. The LSV anodic curve shows a flat profile without
significant current flow extending from the OCV of the cell until 4.5 V,
where a further increment of the voltage causes a raise of the current
over 10 μA due to the beginning of oxidative electrolyte decomposition.
The anodic stability was further investigated by measuring the current
by chronoamperometry from 4.0 to 4.7 V (see experimental section for
details). The results reported in inset of Fig. 3c reveal a significant
current flow only at 4.5 V where a value of 15 μA is observed, as well as
at higher potentials. Therefore, the chronoamperometry and voltam­
metry agree in indicating an anodic stability slightly below 4.5 V. This
remarkable anodic stability suggests a possible applicability of the
PEGDME_CPE extending from olivine cathodes, such as LiFePO4 which
operates at about 3.5 V [56], to electrodes working at higher potentials
such as layered metal oxides [57,58]. The cathodic CV scans show
irreversible peaks at about 1.4 V and below 1.0 V during the first cycle,
attributed to the reduction of the LiNO3 and of the PEGDME2000,
respectively, with concomitant formation of a solid electrolyte inter­
phase (SEI) layer which passivates the electrodes surface [21]. Indeed,
the subsequent CV cycles, performed in the cathodic region within 0.01
V and 2.0 V, are characterized by overlapping curves suggesting a
reversible and kinetically favorable Liþ ion insertion into the carbon
electrode and possibly lithium stripping/deposition around 0 V [59].
Thus, the electrochemical stability window of PEGDME_CPE extends
from 0.01 V to about 4.4–4.5 V at 50 � C that is well suitable for appli­
cation in lithium battery [31]. The chemical stability of Li/Electrolyte
interphase is investigated through EIS measurements upon aging of a
interphase and a low-frequency Warburg element. Furthermore, Fig. 3d
shows a raise of the electrode/electrolyte interphase resistance from
242 Ω up to 277 Ω by the initial 11 days of test due to SEI growth, a
decrease to about 225 Ω, and a stabilization during the subsequent days,
thus suggesting a partial dissolution and final consolidation of the
passivation layer with appropriate suitability for battery application
[60] in view of the limited resistance fluctuations (about 50 Ω).
The PEGDME_CPE is subsequently used in lithium metal battery,
exploiting a LiFePO4 (LFP) cathode already characterized in a previous
work [49]. Prior to cycling, the cell is held at 70 � C for 14 h and then
cooled down at the operating temperature of 50 � C, which was found to
be the most adequate value for the application of the PEGDME_CPE in
lithium battery (see Fig. 3 discussion). Fig. 4 shows the EIS Nyquist plots
recorded upon the above mentioned heat treatment. During the initial
stages at 70 � C (Fig. 4a) all the Nyquist plots may be reasonably repre­
sented by the equivalent circuit Re(R1Q1)(R2Q2)Q3 where the electrolyte
resistance (Re) is in series with a middle-high frequency element (R1Q1)
due to the electrode/electrolyte interphase, including SEI film at the
electrodes surface and charge transfer, and a low-frequency element
(R2Q2) due to a Warburg-type Liþ ion diffusion, and an almost vertically
tilted line (Q3) due to the cell geometrical capacity possibly accounting
for a blocking electrode configuration at the OCV of the cell (see Table 2
for EIS data analysis) [61]. The interphase resistance (R1) trend reported
in inset of Fig. 4a shows an initial value of about 40 Ω, slightly increasing
to about 45 Ω by 6 h of test and subsequently decreasing to about 33 Ω.
Additional EIS tests aimed at the study of the electrolyte/electrode
interphase resistance upon prolonged cell aging at 70 � C are performed
on a Li/PEGDME_CPE/LFP cell. The results, reported in Fig. S4 and
Table S2 in Supplementary Material, show stable interphase resistance
with only small changes from 26 to 30 Ω after 14 h of aging. Upon cell
cooling to 50 � C, the Nyquist plot reported in Fig. 4b reveals the addi­
tional element at high frequency due to possible grain boundary or
heterogeneity, already observed previously in Li/Li symmetrical cell,
due to partial crystallization of the electrolyte [23] (not considered by
NLLSQ in Table 2). Furthermore, the figure shows an increase of the
interphase resistance to about 374 Ω, while the final vertically tilted line
is not detected most likely due to the higher impedance value at 50 � C
compared to 70 � C [61]. Despite the increase of the interphase resistance
at 50 � C, which is however considered still within the applicability range
of the polymer membrane in efficient lithium cell, the lower operating
temperature is considered herein more suitable for a practical use.
Therefore, the electrolyte is subsequently studied in a Li/PEGD­
ME_CPE/LFP cell by using an operating temperature of 50 � C indicated
above (Fig. 5), while an additional test showing performance of the cell
at 70 � C is also performed and reported in Supplementary Material
(Fig. S5). Fig. 5 reports the voltage profiles (a) and the cycling trend of
the galvanostatic test performed at various current rates, i.e., C/10, C/8,
C/5, C/3, and 1C (1C ¼ 170 mA g 1) at 50 � C within 2.7 V and 3.9 V
which is considered the most suitable range for allowing proper cell
Table 2
NLLSQ analyses performed on the impedance Nyquist plots of Fig. 4 related to
the open circuit voltage (OCV) of the Li/PEGDME_CPE/LFP cell at 70 � C by
various time intervals and at 50 � C upon the steady state condition. Frequency
range: 500 kHz-100 mHz.
Equivalent circuit R1 [Ω] χ2
Cell condition
4
2 h at 70 � C Re (R1Q1) (R2Q2)Q3 40.6 � 2.1 1.0 � 10
5
4 h at 70 � C Re (R1Q1) (R2Q2)Q3 41.4 � 1.2 6.5 � 10
5
6 h at 70 � C Re (R1Q1) (R2Q2)Q3 45.4 � 0.9 3.6 � 10
8 h at 70 � C Re (R1Q1) (R2Q2)Q3 45.4 � 0.9 3.0 � 10 5

symmetrical Li/PEGDME_CPE/Li cell and reported in Fig. 3d along with 10 h at 70 � C Re (R1Q1) (R2Q2)Q3 43.7 � 0.7 1.3 � 10 5

the related Nyquist plots in inset. As already mentioned during discus­ 12 h at 70 � C Re (R1Q1) (R2Q2)Q3 37.6 � 0.9 1.3 � 10 5
5
sion of Fig. 2 and Fig. S3, the Nyquist plots exhibit a high-frequency 14 h at 70 � C Re (R1Q1) (R2Q2)Q3 33.9 � 1.0 2.0 � 10
4
Steady state at 50 � C Re (R1Q1)Q2 374 � 14 1.0 � 10

6
V. Marangon et al. Journal of Power Sources xxx (xxxx) xxx

Fig. 5. (a) Voltage profiles and (b) corresponding charge/discharge cycling trend of the galvanostatic test performed on a Li/PEGDME_CPE/LFP cell using various
current rates, i.e., C/10, C/8, C/5, C/3, and 1C (1C ¼ 170 mA g 1) at a temperature of 50 � C. Voltage limits 2.7–3.9 V. See experimental section for sam­
ple’s acronym.

operation without possible occurrence of side reactions [2,49,62]. The
figure evidences that the cell performs at C/10 by a voltage profile
reflecting the electrochemical process expected by the olivine cathode, i.
e., LiFePO4 ⇄ Li þ FePO4, centered at about 3.5 V with a flat profile
typical of a two phases process [62], with very limited polarization, and
a specific capacity approaching 150 mAh g 1, i.e., about 88% of the
theoretical capacity associated with the adopted cathode [49].
Remarkably, the cell operating at 50 � C reveals at C/10 almost the same
voltage profile, and specific capacity comparable to the one operating at
the same C-rate but at a temperature of 70 � C (compare Fig. 5a and
Fig. S5), thus suggesting the suitability of the PEGDME_CPE membrane
both at the higher and at the lower temperature. Furthermore, the cell
shows at 50 � C only a limited change of polarization and capacity at C/8
rate. The voltage shape appears more affected by the current raise to C/5
and C/3 (Fig. 5a), while the capacity holds a satisfactory value at both
c-rates, i.e., of about 140 and 125 mAh g 1, respectively (Fig. 5b).
However, a further raise of the current to 1C notably increases the cell
polarization and limits its capacity to negligible value, as indeed
expected by the solid configuration of the cell and the relatively low
lithium transference number of the polymer electrolyte which hinders
the charge transfer at high currents [23]. The cell stability is evidenced
by the recovery of the initial capacity by lowering back the current to the
pristine value of C/10 (Figure 6b), which is considered an optimal
characteristic of the PEGDME_CPE studied herein.
The stability of the PEGDME_CPE in lithium cell is further demon­
strated by the prolonged galvanostatic cycling test reported in Fig. 6.
The voltage profiles of the measurement performed at a current rate of
C/5 (Fig. 6a) well overlap upon the first cycle during which a passivating
SEI layer is formed over the electrodes [61], and the lithium polymer cell
reveals constantly low polarization, a working voltage centered at about
3.5 V, and a steady state capacity of about 150 mAh g 1. Remarkably,
the cell shows a coulombic efficiency exceeding 99% after the first cycle
which is reflected into a very stable cycling trend (Fig. 6c) and a capacity
retention as high as 99% with respect to the first cycle upon 140 cycles.
In spite, a different behavior is shown by cycling a lithium cell within the
same current and temperature conditions above reported using the

Figure 6. (a,b) Voltage profiles and (c,d) corresponding charge/discharge cycling trend in terms of delivered capacity (left y axis) and coulombic efficiency (right y
axis) of the galvanostatic tests performed on Li/PEGDME_CPE/LFP cells using current rates of (a,c) C/5 and (b,d) C/3 (1C ¼ 170 mA g 1) at 50 � C. Voltage limits
2.7–3.9 V. See experimental section for sample’s acronym.

7
V. Marangon et al.
membrane without LiNO3 (i.e., the blank electrolyte). Fig. S6 in Sup­
plementary Material shows the comparison of cycling responses of
Li/blank/LFP and Li/PEGDME_CPE/LFP cells using a current rate of C/5
at 50 � C. The voltage profile (Fig. S6a) related to the Li/blank/LFP cell
(black color) reveals a higher polarization with respect to Li/PEGD­
ME_CPE/LFP cell (orange color), a lower and a less stable capacity
characterized by an initial increase, followed by a more remarkable
decrease evidenced by the cycling trend comparison reported in Fig. S6a
inset. Furthermore, the voltage profile of Li/blank/LFP cell evidences
the presence of irregular spikes, likely ascribed to formation and
dissolution of dendritic structures on the lithium surface [63,64]. The
higher polarization of the cell using the blank membrane with respect to
PEGDME_CPE is reflected into a lower energy efficiency, which is further
lowered together with the coulombic efficiency by dendrite growth and
dissolution as evidenced by the trends of Fig. S6b. We can therefore
suppose that the decrease of the cell efficiency and the evolution of the
dendritic structures typical of the lithium metal anode is remarkably
mitigated by the use of the LiNO3 additive in the polymer electrolyte
studied in this work [21,46]. In order to further analyze the cycling life
of the and Li/PEGDME_CPE/LFP cell, we have performed a galvanostatic
test at a higher current (i.e., C/3 rather than C/5) prolonged over 300
cycles. The corresponding voltage profiles (Fig. 6b) exhibit a remarkable
overlapping upon the first cycle, and higher polarization compared the
test at C/5 (Fig. 6a), fully in line with the rate capability test (compare
with Fig. 5a), which decreases during cycles likely due to gradual
improvement of the electrode/electrolyte interphase. Furthermore, the
cell delivers a capacity exceeding 120 mAh g 1 (Fig. 6d), coulombic
efficiency higher than 99% and a remarkable stability over 300 cycles
with a capacity retention of about 99%.
The outstanding capacity retention herein obtained is not generally
observed for lithium metal cells using various polymer electrolytes
exploiting the solid configuration at temperature higher than 25 � C [42,
56,65]. It is worth noting that the voltage shape of the battery does not
modify upon repeated charge and discharge which occur without any
sign of increasing polarization, thus accounting for the absence of side
reactions both during oxidation and during reduction, in full agreement
with the electrochemical stability windows observed for the PEGD­
ME_CPE in Fig. 3. This behavior is notable, in particular considering the
relatively low temperature compared to the common solid polymer cells,
and suitable for application in high performance battery safely operating
using the high-energy lithium metal. Indeed, taking into account an
average operating voltage of 3.5 V and a reversible specific capacity of
150 mAh g 1 the polymer cell may deliver a theoretical energy density
of about 525 Wh kg 1, referred to the cathode weight, which may be
actually reflected into a practical capacity exceeding 200 Wh kg 1.
Beside the energy content and the high safety content, the cell herein
presented is expected to hold an additional bonus represented by the
modest environmental and economic impact due to the use of sustain­
able and relatively cheap materials, such as the LiFePO4 cathode and
polyethylene glycol dimethyl ether polymer, into the battery formula­
tion [2].
4. Conclusions
Several composite polymer electrolytes based on EO-bond have been
reported in literature according to a wide variety of configurations,
including ceramics, Li-salts and i) pure linear crystalline polymers, ii)
plasticized polymers, iii) co-polymer blends, iv) derivate polymers using
various species, e.g., ionic-liquids, organic or inorganic cores, phos­
phates, and v) hyperbranched and star-branched PEO derivatives [31].
Electrolytes based on pure linear PEO have the lowest ionic conductiv­
ity, i.e., with values below 10 4 S cm 1 at temperatures lower than 70
�
C, due to the high crystallinity of the ethylene oxide (EO) chains, which
can restrain the ionic transition due to the stiff structure [23]. Higher
conductivity has been achieved by using plasticizers such as liquid
PEGDME MW 500 (10 4 S cm 1 above 40 � C) [66,67], TEG and
8
Journal of Power Sources xxx (xxxx) xxx
TEGDME (7 � 10 5 S cm 1 at 30 � C) [68], or even EC/PC (10 3 S cm 1
at 30 � C) [69]. Furthermore, co-polymers such as PEO/PVDF [70],
PEO/PDMS [71] and PEG/PFPE [72] have actually led to conductivity
approaching 10 4 S cm 1 at about 30 � C. Among the other configura­
tions, the one including PEO and ionic liquids [38,65] has shown very
suitable characteristics for lithium battery, with conductivity of about
10 4 S cm 1 above 40 � C. Despite good performances ascribed to the
various proposed pathways for polymer electrolyte preparation, a sim­
ple approach such as the one including hot-pressing or casting of pure
linear crystalline PEO, salt and ceramics represents a very attracting
strategy due to a remarkable scalability [23]. Hence, in this work we
have originally reported a composite including a crystalline EO-based
substrate (PEGDME2000), LiTFSI, LiNO3 and SiO2, characterized at
the same time by the versatility of linear EO-polymer and by a high
conductivity at moderate temperature typically achieved in several ap­
plications such as electric vehicles [73,74], with the additional bonuses
of the relevant electrode/electrolyte interphase stability and the long
cycling performance in lithium-metal battery. The electrolyte has a
complex structure and smooth surface with uniform distribution of the
elements. The electrolyte, indicated by PEGDME_CPE, revealed an ionic
conductivity higher than 10 4 S cm 1 above 45 � C, with a maximum
value of about 4 � 10 4 S cm 1, according to a trend suggesting a
gradual change from Arrhenius-type to the VTF-type, promoted by
subsequent cooling and heating cycles from about 70 � C to the room
temperature. The above change relevantly increased the room temper­
ature conductivity of the membrane by a factor of 10, i.e., from about 4
� 10 6 S cm 1 to about 4 � 10 5 S cm 1. Study of the PEGDME_CPE in
symmetrical Li/Li cell at various temperatures, ranging from 45 � C to
about 70 � C, showed Liþ transference number ranging from 0.22 at the
lowest temperature to about 0.27 at the highest one, polarization always
lower than 150 mV, and the absence of lithium dendrite formation upon
prolonged lithium stripping/deposition tests. The Li-transference num­
ber obtained herein was lower than the typical one ascribed to linear
EO-membranes using ceramics, that is, about 0.4 [42], most likely due to
the presence of an additional sacrificial salt (LiNO3). However, this salt
has shown the important advantage of largely improving the electro­
de/electrolyte interphase, and allowing the cell a long cycling perfor­
mances. Indeed, the PEGDME_CPE membrane revealed an
electrochemical stability extending from 0.01 to about 4.4 V at 50 � C,
which was therefore selected as the preferred operating temperature for
application in lithium metal battery. Aging tests in symmetrical lithium
cell at the above temperature suggested a very stable Li/PEGDME_CPE
interphase, with resistance values ranging from 226 to 277 Ω. The
PEGDME_CPE has been employed with remarkable results in a solid
polymer cell employing the LiFePO4 cathode and operating at 50 � C. The
cell revealed low and stable interphase resistance, limited polarization
upon charge and discharge with operating voltage of about 3.5 V, rate
capability extending from C/10 to C/3, and a maximum specific ca­
pacity of about 150 mAh g 1, that is, a value approaching 90% of the
theoretical value ascribed to the cathode. Interestingly, the prolonged
galvanostatic cycling at a current rate of C/5 evolved with a specific
capacity of 150 mAh g 1 for over 140 cycles, while the one at a current
of C/3 has shown a capacity of about 125 mAh g 1 for 300 cycles, both
with coulombic efficiency approaching 100% upon the first cycle, and a
capacity retention approaching 99%. This outstanding behavior,
attributed herein to the optimized formulation and characteristics of the
PEGDME_CPE membrane, has led to a lithium solid polymer battery
with an expected energy density exceeding 200 Wh kg 1, using safe and
environmentally friendly materials. These results actually represent a
step forward in achieving a stable and performing Li-polymer battery at
moderate temperatures. In summary such an alternative self-standing
electrolyte, characterized by a relevantly improved performance in
lithium cell compared to literature, particularly in terms of long cycling
stability due to its suitable composition, is proposed as new medium for
application in high-energy storage systems safely employing the lithium
metal anode.
V. Marangon et al.
Declaration of competing interests
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgments
This work was funded by the grant “Fondo di Ateneo per la Ricerca
Locale (FAR) 2018”, University of Ferrara, and performed within the
collaboration project “Accordo di Collaborazione Quadro 2015” be­
tween the University of Ferrara (Department of Chemical and Pharma­
ceutical Sciences) and the Sapienza University of Rome (Department of
Chemistry). The authors acknowledge the Institute of Global Innovation
Research (GIR) in Tokyo University of Agriculture and Technology.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jpowsour.2019.227508.
References
[1] B. Scrosati, J. Garche, Lithium batteries: status, prospects and future, J. Power
Sources 195 (2010) 2419–2430, https://doi.org/10.1016/j.jpowsour.2009.11.048.
[2] D. Di Lecce, R. Verrelli, J. Hassoun, Lithium-ion batteries for sustainable energy
storage: recent advances towards new cell configurations, Green Chem. 19 (2017)
3442–3467, https://doi.org/10.1039/c7gc01328k.
[3] G.A. Elia, U. Ulissi, S. Jeong, S. Passerini, J. Hassoun, Exceptional long-life
performance of lithium-ion batteries using ionic liquid-based electrolytes, Energy
Environ. Sci. 9 (2016) 3210–3220, https://doi.org/10.1039/c6ee01295g.
[4] K.M. Abraham, Prospects and limits of energy storage in batteries, J. Phys. Chem.
Lett. 6 (2015) 830–844, https://doi.org/10.1021/jz5026273.
[5] C.D. Thomas, A. Cameron, R.E. Green, M. Bakkenes, L.J. Beaumont, Y.
C. Collingham, et al., Extinction risk from climate change, Nature 427 (2004)
145–148, https://doi.org/10.1038/nature02121.
[6] Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun, C. Chen, Thermal runaway caused fire
and explosion of lithium ion battery, J. Power Sources 208 (2012) 210–224,
https://doi.org/10.1016/j.jpowsour.2012.02.038.
[7] J.H. Shin, W.A. Henderson, G.B. Appetecchi, F. Alessandrini, S. Passerini, Recent
developments in the ENEA lithium metal battery project, Electrochim. Acta 50
(2005) 3859–3865, https://doi.org/10.1016/j.electacta.2005.02.049.
[8] S. Choudhury, R. Mangal, A. Agrawal, L.A. Archer, A highly reversible room-
temperature lithium metal battery based on crosslinked hairy nanoparticles, Nat.
Commun. 6 (2015) 1–9, https://doi.org/10.1038/ncomms10101.
[9] D. Di Lecce, V. Sharova, S. Jeong, A. Moretti, S. Passerini, A multiple electrolyte
concept for lithium-metal batteries, Solid State Ion. 316 (2018) 66–74, https://doi.
org/10.1016/j.ssi.2017.12.012.
[10] D. Di Lecce, J. Hassoun, Lithium metal battery using LiFe0.5Mn0.5PO4 olivine
cathode and pyrrolidinium-based ionic liquid electrolyte, ACS Omega 3 (2018)
8583–8588, https://doi.org/10.1021/acsomega.8b01328.
[11] K. Zhang, G.-H. Lee, M. Park, W. Li, Y.-M. Kang, Recent developments of the
lithium metal anode for rechargeable non-aqueous batteries, Adv. Energy Mater. 6
(2016), 1600811, https://doi.org/10.1002/aenm.201600811.
[12] W. Xu, J. Wang, F. Ding, X. Chen, E. Nasybulin, Y. Zhang, et al., Lithium metal
anodes for rechargeable batteries, Energy Environ. Sci. 7 (2014) 513–537, https://
doi.org/10.1039/C3EE40795K.
[13] P.G. Balakrishnan, R. Ramesh, T. Prem Kumar, Safety mechanisms in lithium-ion
batteries, J. Power Sources 155 (2006) 401–414, https://doi.org/10.1016/j.
jpowsour.2005.12.002.
[14] D. Di Lecce, C. Fasciani, B. Scrosati, J. Hassoun, A gel–polymer Sn–C/
LiMn0.5Fe0.5PO4 battery using a fluorine-free salt, ACS Appl. Mater. Interfaces 7
(2015) 21198–21207, https://doi.org/10.1021/acsami.5b05179.
[15] J. Hassoun, S. Panero, P. Reale, B. Scrosati, A new, safe, high-rate and high-energy
polymer lithium-ion battery, Adv. Mater. 21 (2009) 4807–4810, https://doi.org/
10.1002/adma.200900470.
[16] D. Di Lecce, L. Carbone, V. Gancitano, J. Hassoun, Rechargeable lithium battery
using non-flammable electrolyte based on tetraethylene glycol dimethyl ether and
olivine cathodes, J. Power Sources 334 (2016) 146–153, https://doi.org/10.1016/
j.jpowsour.2016.09.164.
[17] S. Wei, Z. Li, K. Kimura, S. Inoue, L. Pandini, D. Di Lecce, et al., Glyme-based
electrolytes for lithium metal batteries using insertion electrodes: an
electrochemical study, Electrochim. Acta 306 (2019) 85–95, https://doi.org/
10.1016/j.electacta.2019.03.051.
[18] L. Carbone, M. Gobet, J. Peng, M. Devany, B. Scrosati, S. Greenbaum, et al.,
Comparative study of ether-based electrolytes for application in lithium-sulfur
battery, ACS Appl. Mater. Interfaces 7 (2015) 13859–13865, https://doi.org/
10.1021/acsami.5b02160.
9
Journal of Power Sources xxx (xxxx) xxx
[19] D. Di Lecce, V. Marangon, A. Benítez, A. � Caballero, J. Morales, E. Rodríguez-
Castell�on, et al., High capacity semi-liquid lithium sulfur cells with enhanced
reversibility for application in new-generation energy storage systems, J. Power
Sources 412 (2019) 575–585, https://doi.org/10.1016/j.jpowsour.2018.11.068.
[20] R. Bernhard, A. Latini, S. Panero, B. Scrosati, J. Hassoun, Poly(ethylenglycol)
dimethylether-lithium bis(trifluoromethanesulfonyl) imide, PEG500DME-LiTFSI,
as high viscosity electrolyte for lithium ion batteries, J. Power Sources 226 (2013)
329–333, https://doi.org/10.1016/j.jpowsour.2012.10.059.
[21] L. Carbone, M. Gobet, J. Peng, M. Devany, B. Scrosati, S. Greenbaum, et al.,
Polyethylene glycol dimethyl ether (PEGDME)-based electrolyte for lithium metal
battery, J. Power Sources 299 (2015) 460–464, https://doi.org/10.1016/j.
jpowsour.2015.08.090.
[22] S. Lascaud, M. Perrier, A. Vall�ee, S. Besner, J. Prudʼhomme, M. Armand, Phase
diagrams and conductivity behavior of poly(ethylene oxide)-molten salt rubbery
electrolytes, Macromolecules 27 (1994) 7469–7477, https://doi.org/10.1021/
ma00103a034.
[23] G.B. Appetecchi, F. Croce, J. Hassoun, B. Scrosati, M. Salomon, F. Cassel, Hot-
pressed, dry, composite, PEO-based electrolyte membranes: I. Ionic conductivity
characterization, J. Power Sources 114 (2003) 105–112, https://doi.org/10.1016/
S0378-7753(02)00543-8.
[24] A. Manuel Stephan, K.S. Nahm, Review on composite polymer electrolytes for
lithium batteries, Polymer 47 (2006) 5952–5964, https://doi.org/10.1016/j.
polymer.2006.05.069.
[25] L. Carbone, J. Hassoun, A low-cost, high-energy polymer lithium-sulfur cell using a
composite electrode and polyethylene oxide (PEO) electrolyte, Ionics 22 (2016)
2341–2346, https://doi.org/10.1007/s11581-016-1755-5.
[26] K. Kimura, J. Hassoun, S. Panero, B. Scrosati, Y. Tominaga, Electrochemical
properties of a poly(ethylene carbonate)-LiTFSI electrolyte containing a
pyrrolidinium-based ionic liquid, Ionics 21 (2015) 895–900, https://doi.org/
10.1007/s11581-015-1370-x.
[27] K. Kimura, H. Matsumoto, J. Hassoun, S. Panero, B. Scrosati, Y. Tominaga,
A QuaternaryPoly(ethylene carbonate)-lithium bis(trifluoromethanesulfonyl)
imide-ionic liquid-silica fiber composite polymer electrolyte for lithium batteries,
Electrochim. Acta 175 (2015) 134–140, https://doi.org/10.1016/j.
electacta.2015.03.117.
[28] Z. Li, J. Mindemark, D. Brandell, Y. Tominaga, A concentrated poly(ethylene
carbonate)/poly(trimethylene carbonate) blend electrolyte for all-solid-state Li
battery, Polym. J. 51 (2019) 753–760, https://doi.org/10.1038/s41428-019-0184-
5.
[29] Q. Li, H.Y. Sun, Y. Takeda, N. Imanishi, J. Yang, O. Yamamoto, Interface properties
between a lithium metal electrode and a poly(ethylene oxide) based composite
polymer electrolyte, Spec. Issue Interfacial Phenom. Batter. 94 (2001) 201–205,
https://doi.org/10.1016/S0378-7753(00)00587-5.
[30] H.H. Sumathipala, J. Hassoun, S. Panero, B. Scrosati, High performance PEO-based
polymer electrolytes and their application in rechargeable lithium polymer
batteries, Ionics 13 (2007) 281–286, https://doi.org/10.1007/s11581-007-0137-4.
[31] Z. Xue, D. He, X. Xie, Poly(ethylene oxide)-based electrolytes for lithium-ion
batteries, J. Mater. Chem. A. 3 (2015) 19218–19253, https://doi.org/10.1039/
c5ta03471j.
[32] R. Frech, S. Chintapalli, P.G. Bruce, C. A Vincent, CrystalliCrystalline and
amorphous phases in the poly(ethylene oxide) LiCF3SO3 system, Macromolecules
32 (1999) 808–813, https://doi.org/10.1021/ma9812682.
[33] D. Devaux, R. Bouchet, D. Gl�e, R. Denoyel, Mechanism of ion transport in PEO/
LiTFSI complexes: effect of temperature, molecular weight and end groups, Solid
State Ion. 227 (2012) 119–127, https://doi.org/10.1016/j.ssi.2012.09.020.
[34] N.K. Karan, O.K. Pradhan, R. Thomas, B. Natesan, R.S. Katiyar, Solid polymer
electrolytes based on polyethylene oxide and lithium trifluoro- Methane sulfonate
(PEO-LiCF3SO3): ionic conductivity and dielectric relaxation, Solid State Ion. 179
(2008) 689–696, https://doi.org/10.1016/j.ssi.2008.04.034.
[35] L.R.A.K. Bandara, M.A.K.L. Dissanayake, B.-E. Mellander, Ionic conductivity of
plasticized(PEO)-LiCF3SO3 electrolytes, Electrochim. Acta 43 (1998) 1447–1451,
https://doi.org/10.1016/S0013-4686(97)10082-2.
[36] Y.T. Kim, E.S. Smotkin, The effect of plasticizers on transport and electrochemical
properties of PEO-based electrolytes for lithium rechargeable batteries, Solid State
Ion. 149 (2002) 29–37, https://doi.org/10.1016/S0167-2738(02)00130-3.
[37] J.-H. Shin, W.A. Henderson, C. Tizzani, S. Passerini, S.-S. Jeong, K.-W. Kim,
Characterization of solvent-free polymer electrolytes consisting of ternary
PEO–LiTFSI–PYR14TFSI, J. Electrochem. Soc. 153 (2006) A1649–A1654, https://
doi.org/10.1149/1.2211928.
[38] D. Di Girolamo, S. Panero, M.A. Navarra, J. Hassoun, Quaternary polyethylene
oxide electrolytes containing ionic liquid for lithium polymer battery,
J. Electrochem. Soc. 163 (2016) A1175–A1180, https://doi.org/10.1149/
2.0241607jes.
[39] E. Simonetti, M. Carewska, G. Maresca, M. De Francesco, G.B. Appetecchi, Highly
conductive, ionic liquid-based polymer electrolytes, J. Electrochem. Soc. 164
(2017) A6213–A6219, https://doi.org/10.1149/2.0331701jes.
[40] F. Croce, G.B. Appetecchi, L. Persi, B. Scrosati, Nanocomposite polymer
electrolytes for lithium batteries, Nature 394 (1998) 456–458, https://doi.org/
10.1038/28818.
[41] F. Croce, R. Curini, A. Martinelli, L. Persi, F. Ronci, B. Scrosati, et al., Physical and
chemical properties of nanocomposite polymer electrolytes, J. Phys. Chem. B 103
(1999) 10632–10638, https://doi.org/10.1021/jp992307u.
[42] B.W. Zewde, S. Admassie, J. Zimmermann, C.S. Isfort, B. Scrosati, J. Hassoun,
Enhanced lithium battery with polyethylene oxide-based electrolyte containing
silane-Al2O3 ceramic filler, ChemSusChem 6 (2013) 1400–1405, https://doi.org/
10.1002/cssc.201300296.
V. Marangon et al.
[43] J. Wang, Y. Wang, X. Zhu, X. Lu, Some insight into stability of amorphous poly
(ethylene glycol) dimethyl ether polymers based on molecular dynamics
simulations, J. Phys. Chem. Lett. 4 (2013) 1718–1722, https://doi.org/10.1021/
jz4008855.
[44] T. Sakakibara, M. Kitamura, T. Honma, H. Kohno, T. Uno, M. Kubo, et al., Cross-
linked polymer electrolyte and its application to lithium polymer battery,
Electrochim. Acta 296 (2019) 1018–1026, https://doi.org/10.1016/j.
electacta.2018.11.155.
[45] A. Rosenman, R. Elazari, G. Salitra, E. Markevich, D. Aurbach, A. Garsuch, The
effect of interactions and reduction products of LiNO3, the anti-shuttle agent, in Li-
S battery systems, J. Electrochem. Soc. 162 (2015) A470–A473, https://doi.org/
10.1149/2.0861503jes.
[46] W. Li, H. Yao, K. Yan, G. Zheng, Z. Liang, Y.M. Chiang, et al., The synergetic effect
of lithium polysulfide and lithium nitrate to prevent lithium dendrite growth, Nat.
Commun. 6 (2015) 7436, https://doi.org/10.1038/ncomms8436.
[47] S. Xiong, K. Xie, Y. Diao, X. Hong, Properties of surface film on lithium anode with
LiNO3 as lithium salt in electrolyte solution for lithium-sulfur batteries,
Electrochim. Acta 83 (2012) 78–86, https://doi.org/10.1016/j.
electacta.2012.07.118.
[48] J. Evans, C.a. Vincent, P.G. Bruce, Electrochemical measurement of transference
numbers in polymer electrolytes, Polymer 28 (1987) 2324–2328, https://doi.org/
10.1016/0032-3861(87)90394-6.
[49] S. Brutti, J. Hassoun, B. Scrosati, C.-Y. Lin, H. Wu, H.-W. Hsieh, A high power
Sn–C/C–LiFePO4 lithium ion battery, J. Power Sources 217 (2012) 72–76, https://
doi.org/10.1016/j.jpowsour.2012.05.102.
[50] B. Boukamp, A Nonlinear Least Squares Fit procedure for analysis of immittance
data of electrochemical systems, Solid State Ion. 20 (1986) 31–44, https://doi.org/
10.1016/0167-2738(86)90031-7.
[51] B. Boukamp, A package for impedance/admittance data analysis, Solid State Ion.
18–19 (1986) 136–140, https://doi.org/10.1016/0167-2738(86)90100-1.
[52] E. Staunton, Y.G. Andreev, P.G. Bruce, Factors influencing the conductivity of
crystalline polymer electrolytes, Faraday Discuss 134 (2007) 143–156, https://doi.
org/10.1039/b601945e.
[53] A.E. Visser, N.J. Bridges, R.D. Rogers (Eds.), Ionic Liquids: Science and
Applications, American Chemical Society, Washington, DC, 2012, https://doi.org/
10.1021/bk-2012-1117.
[54] M.M. Hiller, M. Joost, H.J. Gores, S. Passerini, H.D. Wiemh€ ofer, The influence of
interface polarization on the determination of lithium transference numbers of salt
in polyethylene oxide electrolytes, Electrochim. Acta 114 (2013) 21–29, https://
doi.org/10.1016/j.electacta.2013.09.138.
[55] J. Mindemark, B. Sun, E. T€ orm€ a, D. Brandell, High-performance solid polymer
electrolytes for lithium batteries operational at ambient temperature, J. Power
Sources 298 (2015) 166–170, https://doi.org/10.1016/j.jpowsour.2015.08.035.
[56] G.B. Appetecchi, J. Hassoun, B. Scrosati, F. Croce, F. Cassel, M. Salomon, Hot-
pressed, solvent-free, nanocomposite, PEO-based electrolyte membranes: II. All
solid-state Li/LiFePO4 polymer batteries, J. Power Sources 124 (2003) 246–253,
https://doi.org/10.1016/S0378-7753(03)00611-6.
[57] G.A. Elia, U. Ulissi, F. Mueller, J. Reiter, N. Tsiouvaras, Y.-K. Sun, et al., A long-life
lithium ion battery with enhanced electrode/electrolyte interface by using an ionic
liquid solution, Chem. Eur J. 22 (2016) 6808–6814, https://doi.org/10.1002/
chem.201505192.
[58] D. Di Lecce, P. Andreotti, M. Boni, G. Gasparro, G. Rizzati, J.-Y. Hwang, et al.,
Multiwalled carbon nanotubes anode in lithium-ion battery with LiCoO2, Li[Ni1/
3Co1/3Mn1/3]O2 , and LiFe1/4Mn1/2Co1/4PO4 cathodes, ACS Sustain. Chem. Eng. 6
(2018) 3225–3232, https://doi.org/10.1021/acssuschemeng.7b03395.
[59] L. Carbone, T. Coneglian, M. Gobet, S. Munoz, M. Devany, S. Greenbaum, et al.,
A simple approach for making a viable, safe, and high-performances lithium-sulfur
10
Journal of Power Sources xxx (xxxx) xxx
battery, J. Power Sources 377 (2018) 26–35, https://doi.org/10.1016/j.
jpowsour.2017.11.079.
[60] G.T. Kim, G.B. Appetecchi, F. Alessandrini, S. Passerini, Solvent-free, PYR1ATFSI
ionic liquid-based ternary polymer electrolyte systems. I. Electrochemical
characterization, J. Power Sources 171 (2007) 861–869, https://doi.org/10.1016/
j.jpowsour.2007.07.020.
[61] D. Aurbach, Review of selected electrode–solution interactions which determine
the performance of Li and Li ion batteries, J. Power Sources 89 (2000) 206–218,
https://doi.org/10.1016/S0378-7753(00)00431-6.
[62] F. Croce, A. D’Epifanio, J. Hassoun, A. Deptula, T. Olczac, B. Scrosati, A novel
concept for the synthesis of an improved LiFePO4 lithium battery cathode,
Electrochem. Solid State Lett. 5 (2002) A47–A50, https://doi.org/10.1149/
1.1449302.
[63] J.M. Tarascon, M. Armand, Issues and challenges facing rechargeable lithium
batteries, Nature 414 (2001) 359–367, https://doi.org/10.1038/35104644.
[64] L. Gireaud, S. Grugeon, S. Laruelle, B. Yrieix, J.M. Tarascon, Lithium metal
stripping/plating mechanisms studies: a metallurgical approach, Electrochem.
Commun. 8 (2006) 1639–1649, https://doi.org/10.1016/j.elecom.2006.07.037.
[65] M. Wetjen, M.A. Navarra, S. Panero, S. Passerini, B. Scrosati, J. Hassoun,
Composite poly(ethylene oxide)-electrolytes plasticized by N-alkyl-N-
butylpyrrolidinium bis(trifluoromethanesulfonyl)imide for lithium batteries,
ChemSusChem 6 (2013) 1037–1043, https://doi.org/10.1002/cssc.201300105.
[66] L. Niedzicki, M. Kasprzyk, K. Kuziak, G.Z. Zukowska,
_ M. Armand, M. Bukowska, et
al., Modern generation of polymer electrolytes based on lithium conductive
imidazole salts, J. Power Sources 192 (2009) 612–617, https://doi.org/10.1016/j.
jpowsour.2009.03.050.
[67] H. Wang, D. Im, D.J. Lee, M. Matsui, Y. Takeda, O. Yamamoto, et al., A composite
polymer electrolyte protect layer between lithium and water stable ceramics for
aqueous lithium-air batteries, J. Electrochem. Soc. 160 (2013) A728–A733,
https://doi.org/10.1149/2.020306jes.
[68] S. Chintapalli, R. Frech, Effect of plasticizers on ionic association and conductivity
in the (PEO)9LiCF3SO3 system, Macromolecules 29 (1996) 3499–3506, https://doi.
org/10.1021/ma9515644.
[69] H.M.J.C. Pitawala, M.A.K.L. Dissanayake, V.A. Seneviratne, B.-E. Mellander,
I. Albinson, Effect of plasticizers (EC or PC) on the ionic conductivity and thermal
properties of the (PEO)9LiTf: Al2O3 nanocomposite polymer electrolyte system,
J. Solid State Electrochem. 12 (2008) 783–789, https://doi.org/10.1007/s10008-
008-0505-7.
[70] R. Prasanth, N. Shubha, H.H. Hng, M. Srinivasan, Effect of poly(ethylene oxide) on
ionic conductivity and electrochemical properties of poly(vinylidenefluoride)-
based polymer gel electrolytes prepared by electrospinning for lithium ion
batteries, J. Power Sources 245 (2014) 283–291, https://doi.org/10.1016/j.
jpowsour.2013.05.178.
[71] A. Das, A.K. Thakur, K. Kumar, Exploring low temperature Liþ ion conducting
plastic battery electrolyte, Ionics 19 (2013) 1811–1823, https://doi.org/10.1007/
s11581-013-0898-x.
[72] D.H.C. Wong, A. Vitale, D. Devaux, A. Taylor, A.A. Pandya, D.T. Hallinan, et al.,
Phase behavior and electrochemical characterization of blends of
perfluoropolyether, poly(ethylene glycol), and a lithium salt, Chem. Mater. 27
(2015) 597–603, https://doi.org/10.1021/cm504228a.
[73] P. Baudry, S. Lascaud, H. Majastre, D. Bloch, Lithium polymer battery development
for electric vehicle application, J. Power Sources 68 (1997) 432–435, https://doi.
org/10.1016/S0378-7753(97)02646-3.
[74] H. Zhang, G.G. Eshetu, X. Judez, C. Li, L.M. Rodriguez-Martínez, M. Armand,
Electrolyte additives for lithium metal anodes and rechargeable lithium metal
batteries: progress and perspectives, Angew. Chem. Int. Ed. 57 (2018)
15002–15027, https://doi.org/10.1002/anie.201712702.