Solid State Ionics 276 (2015) 107126

Contents lists available at ScienceDirect

Solid State Ionics

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

Electrolytes for Li-ion transport Review

M. Marcinek , J. Syzdek, M. Marczewski, M. Piszcz, L. Niedzicki, M. Kalita, A. Plewa-Marczewska,
A. Bitner, P. Wieczorek, T. Trzeciak, M. Kasprzyk, P.ak, Z. Zukowska, A. Zalewska, W. Wieczorek
Polymer Ionics Research Group, Faculty of Chemistry, Warsaw University of Technology, Noakowskiego 3, 00664 Warsaw, Poland

a r t i c l e i n f o a b s t r a c t

Article history: In the present paper an authoritative report based on opinion of the group working in the eld of electrolytes for
Received 20 December 2013 battery application are presented. In the following sections rst each group of the material will be introduced and
Received in revised form 23 December 2014 the recent ideas how to improve their performance in lithium and/or lithium-ion battery technology will be
Accepted 13 February 2015
discussed. This will be followed by the presentation of ideas which in our opinion open new elds of research
Available online xxxx
and possibly lead to an improvement of the battery performance.
Keywords:
2015 Published by Elsevier B.V.
Lithium-ion batteries
Liquid
Gel
Polymer electrolytes

This work was presented during the Polish Forum Smart Energy
Conversion and Storage, Krynica, Poland, and ERI Alistore White Paper
Presentation.
The energy density of an electrochemical cell is a linear function of
its useful voltage, and the power increases with its square, hence a
search for new electrode materials offering higher and higher cell volt-
ages goes on. One should keep in mind, that the electrolyte remains in
contact with both negative and positive electrode and by increasing
the cell voltage we increase the possibility of destroying the electrolyte
by reduction/oxidation at either of the electrodes. The introduction of
liquid nonaqueous, and then solid electrolytes, characterised by broader
electrochemical stability window (for water it is 1.2 V) was a real mile-
stone and enabled development of high power and high energy cells
through exploration of new electrode materials. Ideally, the electrolyte
should not undergo any net chemical changes in a voltaic cell, or
contribute to the Faradaic processes that are expected to take place
within/at the electrodes, or change its composition.
These processes can be avoided either by thermodynamic stability,
or kinetic restrictions. One example of a cell working out of equilibrium
is a well-known lead-acid battery. Its usual OCV (open circuit voltage) of
2.02 V exceeds the stability window of water (1.2 V), however the cell
works ne for years due to very slow kinetics of water reduction at
the lead electrode and oxidation at lead dioxide.
In the pursuit for high performance batteries, alkali metals have
been recognised as good candidates for electrode materials owing to
their low standard potentials and densities. Among them, lithium is
Corresponding author.
E-mail address: marekmar@ch.pw.edu.pl (M. Marcinek).
the most attractive. It has the lowest standard potential among all
elements, its the lightest metal in the periodic table and its cation is
very small giving it a chance to diffuse quickly in solids (hence fast inter-
calation). Alkali metals react violently with water, giving the metal
hydroxide and hydrogen gas. Therefore as early as in the 1950's, the
electrochemistry of lithium in nonaqueous solvents was established.
Its stability in certain liquids was attributed to the formation of protec-
tive passive lms that is also responsible for stability of stainless steel or
aluminium in oxidising conditions. In no time at all, the rst primary
lithium cells appeared on the market where they keep their well-
established position until the present days. But still there was a lot of
effort made to develop lithium technology as to introduce secondary
lithium cells. The problem was the plating of lithium during the re-
charge. Highly inhomogeneous nature of the lithium-electrolyte inter-
face provoked growth of lithium dendrites. Their erratic shape caused
their dissolution and detachment from the electrode on discharge, lead-
ing to capacity losses (that could be overcome by putting excess lithium
in a new cell). Even worse was the presence of nely divided metallic
lithium inside the electrolyte, that combined with this dendritic growth
could lead to short-circuits and thermal runaway. For years different
electrolyte compositions were examined in the hope of nding one
that could lead to uniform lithium deposition. An ether-based electro-
lyte was introduced into the market in 1980s, which operated satisfac-
torily for over 300 deep cycles, however these batteries were still not
perfect and several spectacular cells failures in 1989 nished the
commercial viability of lithium metal/liquid nonaqueous electrolyte
secondary cells.
Safety issues related to the lithium metal electrode stimulated re-
searchers to look for new negative electrodes for lithium technology.
Looking at the lithium cell one can see, that the positive electrode

http://dx.doi.org/10.1016/j.ssi.2015.02.006
0167-2738/ 2015 Published by Elsevier B.V.
108 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
chemistry is of the guest-host type, i.e. lithium cations are introduced
into/removed from the stable structure of the host. At the same time
electrons are injected into/removed from the highest energy bands of
the material. All that happens at a certain potential. This class of mate-
rials (TiS2, LiMn2O4, LiCoO2) was introduced by Whittingham, Armand
and Goodenough back in the 1970s [13]. These materials were cycled
vs. metallic lithium electrode where it is plated/stripped. Extending the
approach to the positive electrode into the negative one would lead to
something called a rocking-chair, shuttle-cock or swing battery in
which lithium exists only in ionic state. The battery operation results
from transfer of lithium between two materials of different Fermi levels
combined with concomitant transfer of electrons. First materials tested
were lithiated oxides: Li6Fe2O3, LiWO2 that were cycled vs. WO3, TiS2
and V2O5. Such cells, despite being very safe and offering long cycle life
were not able to enter the market because of low energy and power den-
sity as compared with other technologies. It was only at the beginning of
the 1990s when Li-ion cells were marketed thanks to the application of
petroleum coke as the negative electrode material. Cokes and graphitic
carbons offer relatively high gravimetric capacities, operating potential
very close to the one of lithium metal and most of all, low price.
It took over 10 years before non-carbonaceous materials were intro-
duced into the negative electrode of lithium ion cells. These were tin-
based composites in which lithium creates an alloy with tin upon charg-
ing the cell. However being quite satised with well-performing carbon
electrodes more effort was devoted to positive electrodes in the search
of cheaper and more available materials (Co is expensive and limited in
supply), characterised by higher capacity (140mAh g1 of LiCoO2 is
quite low compared with LiC6s 372mAh g1) and possibly higher
operating potential (however this is still under debate if very high
potential positive electrodes are really a good solution to enhance the
performance of the cell as a whole). The most explored materials are
LiNiO2 (and its intermediates with LiCoO2 and LiMnO2), LiMn2O4 and
LiFePO4 [4].
The general trend observed recently is that the search for new
electrode materials is ruled by the demand for safe, cheap and well-
performing cells for automotive industry. Use of rare and expensive el-
ements is minimized in order to lower the price and allow large-scale
production; operating potentials of the electrodes should lay within
the stability domain of the electrolyte in order to minimize the safety
hazards that in case of large cells/battery packs are an important issue.
The cycle life should be long enough to guarantee 10+ years operation.
Of course these cells should offer as good capacities and power capabil-
ities as possible, however these cannot be traded for the safety or unrea-
sonably high price. Also the ability of large charge and discharge rates is
not an issue since the power demand of a car is limited to short spikes,
and peak power can be increased with supercapacitors. Furthermore in
order to charge the cars battery pack in seconds a high power grid con-
nection would be necessary and the heat generated within the pack
would be difcult to dissipate (safety hazard).
Apart from optimisation of electrodes materials, that are not to be
discussed here, a lot of effort is devoted to the development of electro-
lytes, tailored to the specic electrochemical systems.
The role of electrolyte is two, or sometimes threefold:
It should provide ionic contact between electrodes allowing to close
the circuit when the cell is operational
It should assure electronic and spatial separation of the positive and
negative electrode in order to avoid short-circuit and as a result
self discharge of the cell, which in some cases can be very spectacular
(as those of failed high power Li-ion cells)
In case of electrochemical systems where electrode components are
not the only reactants appearing in the overall cell reaction, the elec-
trolyte is the source (storage) of the remaining ones:
neg: pos: el: neg: pos:
Pbs PbO2 s 2H2 SO4 PbSO4 s PbSO4 s 2H2 O
In most cases the 3rd function of the electrolyte is not obligatory,
therefore from the point of view of the cell weight, electrolyte is consid-
ered a waste which should be limitwed to the indispensable minimum.
In general the electronic and spatial separation can be obtained
easily with application of dielectric solids. On the other hand solutions
of proper electrolyte (acid, base or salt) in a suitable liquid solvent can
provide good ionic conductivity needed for operating the cell. For
many years these two functions have been separated and provided by
different phases i.e. liquid electrolytes soaked into a porous separator,
including primary and secondary lithium cells. The development of
the liquid electrolytes for lithium cells was described exhaustively in
the great review published a decade ago [5] (for completeness of the
work just a brief summary is given here). It is not commonly appreciat-
ed that most of the lithium cells work out of equilibrium and as such are
subject to runaway once the kinetic restrictions are lost for any reason. It
is just the choice of solvents and solutes that had to be made correctly to
make the system work.
In lithiumion technology generally three groups of electrolytes are
considered for ambient and moderate temperature application. These
are: liquid systems (solutions of lithium salt in aprotic solvents), PEs
(solid or gel systems) and solutions of lithium salts in ionic liquids
(ILs). The present report deals with pros and cons of each group of elec-
trolytes mentioned above and not a general overview of the electrolytes
studied for lithium and lithium-ion battery technologies. For these pur-
poses there are recent excellent reviews available dealing with liquid,
polymer and ionic liquids [511] systems. Readers are welcome to use
them for further extension of their knowledge regarding each group of
electrolytes. In this review we present an authoritative report based
on decades of our combined experience in the eld of electrolytes for
battery applications. In the following sections each category will be in-
troduced and then strategies of improving their performance in lithium
and/or lithium-ion battery technology will be discussed. This will be
followed by the presentation of ideas which in our opinion open new
elds of research and possibly lead to an improvement of the battery
performance.
1. Liquid electrolytes
In the introduction we described liquid electrolyte for a Li-ion cell as
a solution of an appropriate lithium salt in a polar, organic, aprotic sol-
vent. A list of solvents currently used, or tried in lithium-ion batteries
is included in Tables 1 and 2 [5]. Table 1 lists carbonates and esters,
Table 2 ethers.
Generally an ideal electrolyte solvent should meet the following
criteria:
be able to dissolve lithium salts to sufcient concentration
its viscosity should be low to enable fast ion transport and efcient
(quick) lling of a commercial cell on a production line
be inert to all cell components especially positive and negative elec-
trode materials and the current collectors
have sufcient wettability towards the electrodes and the separator
it should remain liquid in a wide temperature range (low melting and
high boiling temperature are desirable)
The choice of solvents for electrolyte dedicated for practical lithium
cell (consisting of very low potential negative and very high potential
positive electrode) is rather limited. Electrolyte should be characterised
by high dielectric constant and ability to complex at least one of the
ions coming from the dissociation of the electrolyte salt. In aprotic
solvents (because protic ones are unstable at low potentials) we
lose the ability to complex anions rst of all. Then in order to get
a high dielectric constant only a limited number of polar groups to
el: choose from remain (many of them can be easily oxidized/reduced at
the electrodes) i.e. carbonyl, nitrile, sulfonyl, ester and ether groups.
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126 109

Table 1
Organic carbonates and esters as electrolyte solvents. K. Xu Reprint with permission from Nonaqueous Liquid Electrolytes for Lithium Based Rechargeable Batteries, Chem. Rev., 104
(2004) 43034418 [5].

Solvent Structure M. Wt Tm/C Tb/C /cP /D Tf/C d/gcm3

25 C 25 C 25 C

EC 88 36.4 248 1.90 (40 C) 89.78 4.61 160 1.321

PC 102 48.8 242 2.53 64.92 4.81 132 1.200

BC 116 53 240 3.2 53

BL 86 43.5 204 1.73 39 4.23 97 1.199

VL 100 31 208 2.0 34 4.29 81 1.057

NMO 101 15 270 2.5 78 4.52 110 1.17

DMC 90 4.6 91 0.59 (20 C) 3.107 0.76 18 1.063

DEC 118 74.3 126 0.75 2.805 0.96 31 0.969

EMC 104 53 110 0.65 2.958 0.89 1.006

EA 88 84 77 0.45 6.02 3 0.902

MB 102 84 102 0.6 11 0.898

EB 116 93 120 0.71 19 0.878

Some heterocyclic compounds can be considered as well as long as they
lack acidic protons and are stable against oxidation and reduction.
Compounds that offer high dielectric constant and form low energy
complexes with lithium cations were found to be cyclic organic carbon-
ates. Many lithium salts can be dissolved in ethylene and/or propylene
carbonates (EC and PC). The problem is that these solvents are
characterised by high viscosity and fairly high melting point (EC is
solid at room temperature). In order to circumvent this issue, acyclic
alkyl carbonates are usually added because they offer low viscosities
and melting points. However, their dielectric constant is also low and
so are their boiling and ash points, thus limiting the safety margin of
the practical cell. Given the lack of better alternatives mixtures of
organic carbonates are the solvents of choice for the commercial cells.
It is essential to realize that the structure and properties of
electrode-electrolyte interfaces depend not only on the electrolyte sol-
vent but also vastly on the salt. This makes lithium salt the other critical
element of the electrolyte and one of key elements of the entire lithium-
ion cell. Surprisingly, while there is a long list of publications on the new

Table 2
Organic ethers as electrolyte solvents K. Xu Reprint with permission from Nonaqueous Liquid Electrolytes for Lithium Based Rechargeable Batteries, Chem. Rev., 104 (2004) 43034418
[5].

Solvent Structure M. Wt Tm/C Tb/C /cP /D Tf/C ss/gcm3

25 C 25 C 25 C

DMM 76 105 41 0.33 2.7 2.41 17 0.86

DME 90 58 84 0.46 7.2 1.15 0 0.86
DEE 118 74 121 20 0.84
THF 72 109 66 0.46 7.4 1.7 17 0.88

2-Me-THF 86 137 80 0.47 6.2 1.6 11 0.85

1,3-DL 74 95 78 0.59 7.1 1.25 1 1.06

4-Me-1,3-DL 88 125 85 0.60 6.8 1.43 2 0.983

2-Me-1,3-DL 88 0.54 4.39

110 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
lithium-ion cell electrode materials (or additions to those), and almost
as long on electrolytes additives, there has been very little attention
given to new salts. Ideally properties of the salt used for battery applica-
tions are as follows:
it should be able to completely dissolve in the applied solvent at
desired concentration and ions (particularly Li ions) should be able
to transfer through the solution
anion should be stable towards oxidative or reductive decomposition
at either electrode
anion should be inert to electrolyte solvent and other cell components
anion should be nontoxic and remain thermally stable at the battery
working conditions
promote formation of ionically conducting and stable interfacial layers
on both electrodes
cheap and easy to synthesize in large quantities
preferentially water-stable
As one can see, the choice of an appropriate electrolyte salt is not
easy. First of all, the cation is xed i.e. we need lithium cation as the
electroactive species in the electrolyte. Then the choice of anion is
restrained by several factors. Firstly, the dissociation free enthalpy of
its lithium salt should be as low as possible so that the dissociation con-
stant is high and so is the maximum concentration of lithium cations
available for conduction. Secondly, the anion should be stable at low
and high potentials, therefore not all the polar groups can be used to
make up the anion (for the anodic stability of anions and solvents see
Tables 4 and 5 in [5]). It should also be non-reactive towards carbonate
solvents. Small and simple anions such as O2 or F cannot be used
because their small ionic radii induce low dissociation constants. Use
of softer anions like S2 or I are prohibited because of their low oxida-
tion potentials.
Stability in water is of particular interest for the battery industry as it
removes many constraints from the production process, hence lead to
cost savings. That also allows for cheaper and more scalable anion syn-
thesis routes.
Before introduction of lithium-ion battery technology onto the
market, many lithium salts were known and used in experimental
cells. So it was already known for years that LiClO4 (lithium perchlorate)
can potentially produce unstable products [12] when used with elec-
trodes built from transition metal compounds.
In order to get an anion with well-distributed charge, simple uorine
or chlorine anions were combined with strong Lewis acids such as PF5,
AsF5, BF3 or AlCl3 in order to get PF
6 , AsF6 , BF4 and AlCl4 , respectively,
which are stable at low and high potentials and are highly soluble even
in low polarity media. Since these acids are so strong, their complexa-
tion with weak bases like uoride or chloride does not neutralize their
activity completely and they can react with cell components like elec-
trolyte solvent, current collectors and electrode active materials.
LiAsF6 (lithium hexauoroarsenate) is very toxic and LiBF4 (lithium
tetrauoroborate) produces second-rate solid electrolyte interphase
(SEI), which results in continuous degradation of the electrolyte, lithium
loss and increase of the cell impedance. On the other hand, LiPF6, still the
most popular salt nowadays, spontaneously decomposes to LiF and PF5
causing two issues: LiF layer on electrodes that can block lithium cations
movement through the SEI, while PF5 reacts with solvent and forms HF
in contact with moisture (always present in the battery). This in turn
can damage transition metal oxides. LiPF6 is also unstable above 50 C.
See Fig. 1 where exemplary salts are presented.
Another class of anions is designed around the Pearsons theory of
soft and hard acids/bases. According to it, soft acids are likely to form
stable (hence insoluble) salts with soft bases, as well as hard acids
with hard bases. LiF is a combination of hard acid (Li+) with a hard
base (F). Therefore its solubility is low in low dielectric and poorly
complexing solvents. If we combine a hard acid with a soft base (formal
charge localized at soft centre) in which charge is well-delocalized
(strong electron withdrawing groups attached to the centre by cova-
lent bonds) a well-soluble salt of high dissociation constant should
be obtained. After an acid based on imide with 2 electron withdraw-
ing groups (namely: triuoromethanesulfonate see Fig. 1) had been
reported in 1984, in 1989 Armand proposed using it in its lithiated
form as an electrolyte (Fig. 2). Despite excellent intrinsic properties
of electrolytes based on LiTFSI and its analogues, they have never
been applied in commercial cells because of corrosion problems.
The aluminium current collectors are stable at high potentials be-
cause of passive layer protecting the metal against corrosion. Many
uorinated anions like TFSI or Tf provoke aluminium pitting corro-
sion and inhibit their application.
A whole class of orthoborate chelating anions (chelatoborates)
was introduced by Barthel et al. [1317]] (Fig. 2), e.g. lithium bis[1,2-
benzenediolato(2-)-O,O]borate (LBBB). Then Xu et al. brought-in bo-
rates of lithium bis(oxalate)borate (LiBOB) type and other chelate
ones [1824]. While chelating anion-based systems suffered from high
viscosity (because of anion bulkiness) and ensuing poor conductivity,
LiBOB and similar salts electrolyte systems were found to form poorly
conducting solid electrolyte interphases, which led to poor power capa-
bility. These salts exhibit also poor solubility in commercial-type sol-
vents hence low conductivity. Syntheses of croconatoborate class was
also attempted, e.g. LiBCB (lithium bis(croconato)borate). Other asym-
metrical anion was found in search for the synergy advantage of LiBF4
and LiBOB LiDFOB (lithium diuoro(oxalato)borate) [25], which has
unfortunately very poor conductivity (order of magnitude lower than
LiPF6). The best of this class was lithium tetrakis(haloacyloxy)borates
family (LiTFAB) introduced by Yamaguchi et al. [26], also suffering
from low conductivity in the liquid electrolytes (much lower than
LiTFSI or LiPF6). LiBOB, most successful (in commercial terms) from
borate-class lithium conducting salts family, has an ionic conductiv-
ity of 3 10 3 S cm 1 in PC at 20 C (maximum at 0.5 mol dm- 3
concentration) and its commercial use is limited to small quantities
as an additive.
Even though there have been some previous achievements in the
eld since the introduction of LiPF6 in 1990s, in fact, no other salts
made it into mass production of rechargeable lithium-ion cells. It
gives the highest conductivity among salts known at the time of
lithium-ion cells commercialization: in organic carbonate solvents (op-
timized mixtures) around 11 mS cm1 and in plain PC 6 mS cm-1 at
25 C (maximum at 1 mol dm-3 concentration). Vulnerability of LiPF6
to water is not only a problem in its use but also in its synthesis.
Problematic preparation is one of the big disadvantages of this salt
and special methods were designed for obtaining battery grade LiPF6.
Application of LiPF6 started when moisture and HF free was possible
to obtain.
It should be mentioned that many novel salts for application in
lithium batteries still are to be discovered. However, in most cases
the same salt can be dissolved in all three types of media under con-
sideration in this manuscript (molecular liquids, polymers and ILs).
New possibilities offered by organic salts will be discussed in the fol-
lowing sections.
Modications of liquid electrolytes rely also on the application of
various additives which improve safety (overcharge or re protection)),
as well as, modifying the properties of electrode-electrolyte interfaces
by promoting passivation reactions before decomposition of the major
electrolyte components starts. They are discussed in a comprehensive
review devoted to liquid electrolytes [5] and are not a matter of discus-
sion here.
2. Polymer electrolytes
Among solid ionic conductors (not discussed in this paper as such e.
g glasses, ceramics) there is a class of materials called solid (or dry)
polymer electrolytes (PEs). They are "complexes" of electrodon or
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126 111

Fig. 1. Lithium salts known before the introduction of lithium-ion batteries. LiClO4 lithium perchlorate; LiAsF6 lithium hexauoroarsenate; LiTf LiSO3CF3 lithium triate lithium
triuoromethanesulfonate; LiBF4 lithium tetrauoroborate; LiPF6 lithium hexauorophosphate.

polymers and salts or acids [27]. The advantages of using solid PEs in of consecutive layers; moreover fabrication of such a cell would be
commercial lithium cells compared to liquid electrolytes described easier hence cheaper),
above could be numerous: strengthening of cells thanks to the all-solid-state construction,
shape exibility,
non volatility,
lowering the cell weight non-volatile, all-solid-state cells dont need
possibility to limit destructive decomposition at the electrodes,
heavy steel casing,
no possibility of leaks,
improved shock resistance,
viability of metallic lithium as a negative electrode in secondary cells
better overheat and overcharge allowance,
(lithium dendrites growing on the electrode surface would be stopped
improved safety.
by the non-porous and solid electrolyte),
lowering the cell price (PEO is cheaper than organic carbonates; it Various application possibilities and well-known methods for poly-
could be used as a binder for electrodes to improve the compatibility mer modication have stimulated investigations of solid polymer ionic

Fig. 2. New lithium conductive salts created for application in lithium-ion batteries. LiTFSI Li[N(SO2CF3)2] lithium bis(triuoromethane sulfone)imide; LiBETI Li[N(SO2CF2CF3)2]
lithium bis(pentauoroethane sulfone)imide; LiFAP Li[PF3(CF2CF3)3] lithium uoroalkylphosphate; LiTFAB Li[B(OCOCF3)4] lithium tetrakis(triuoroacetoxy)borate; LiBOB
Li[B(C2O4)2] lithium bis(oxalato)borate; LBBB Li[B(C6O2)2] lithium bis (1,2-benzenediolato(2-)-O,O) borate.
112 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
conductors over the last few decades. The main requirements for the
polymer to be used as a matrix in PE systems are:
the presence of a heteroatom (usually O, N, S) with lone electron pairs
of a donor power sufcient to complex cations,
appropriate distances between coordinating centres to insure the
hopping of charge carriers, and
sufcient exibility in polymer chain segments to facilitate move-
ments of ionic carriers.
In PEs ionic transport occurs mostly in a highly viscoelastic
(amorphous) state, however there are examples of engineered crys-
talline conductive phases [28]. The most intensively studied PEs are
based on poly(oxa alkanes), poly(aza alkanes) or poly(thia alkanes).
The present paper deals mainly with PEs based on poly(oxa alkanes)
polyethers and particularly on alkali metal salt complexes with
poly(ethylene oxide) (PEO). The most important and universal re-
quirements for PEs for application in lithium cells are as follows:
chemical and mechanical stabilities over a wide temperature range,
electrochemical stability of at least 34 V vs. Li/Li+ electrode; espe-
cially important for battery applications
low activation energies for conduction
high cationic transport numbers
good electrodeelectrolyte interphase characteristics
ease of preparation
appropriate ionic conductivity.1
As can be expected from the above list it is not easy to nd an elec-
trolyte fullling all the requirements. The low ambient and particularly
sub-ambient temperature conductivities together with lithium transfer-
ence number close to 0.10.2 are serious limitations of true solid PE
systems.
Despite extensive research there is still a considerable number of
questions about the fundamental understanding of design principles
(and related tradeoffs in critical properties, particularly mechanical sta-
bility and ionic conductivity), synthesis and applications of PEs. Their
complicated phase structure causes difculties in the interpretation of
ionic transport data and establishing conduction mechanism. Several
concepts have been proposed and clearly summarized in a review
paper [29] but none of them is generally valid for the wide range of ma-
terials under consideration at the present time. Berthier et al. [30] have
shown that fast ionic transport takes place in the amorphous electrolyte
phase in which ionic diffusion coefcients are about three orders of
magnitude higher than in the crystalline phases. Berthier's assumption
of the crucial role that an amorphous polymer phase plays in ionic con-
ductivity, forms the basis for some of the proposed conductivity mech-
anisms, such as free volume [31], congurational entropy [32,33] or
dynamic bond percolation [3438]. These models are mainly successful
in quantitatively describing conductivity mechanisms for single phase
amorphous electrolytes and fail when applied to multiphase systems.
The assumption of Berthier [30] requires high amorphous phase
content for fast ionic transport in PEs. Congurational entropy and dy-
namic bond percolation theories demonstrated the exibility of the
amorphous polymer phase to be crucial since ion and polymer segmen-
tal motions are coupled for good conductivity. Therefore a low glass
transition temperature (Tg) for the amorphous polymer phase is desir-
able. Unfortunately, the polyethersalt complexes which are the most
widely studied systems are highly crystalline at ambient temperatures.
Here their conductivities are in the range 10-7-10-8 S cm-1 which is too
low for most applications. The amount of exible amorphous phase in-
creases on approaching the melting point of the crystalline polymer
phase i.e. 6568 C. However, at temperatures exceeding the melting
1
The range of conductivities required depends on the kind of application and is equal to
10-1-10-2 S cm-1 for fuel cells and 10-3-10-4 S cm-1 for batteries; conductivities can be
lower for sensors and electrochromic applications.
point the mechanical stability of electrolytes is much lower and mem-
branes often creep under the pressure applied in electrochemical de-
vices leading to short-circuits.
In spite of all the limitations mentioned above, semi-crystalline PEO
still has attracted most attention as the best highly viscous polymeric
matrix. The mainstream work devoted to PEs focused on obtaining
stable amorphous systems containing a high amount of ethylene
oxide molecular repeat units in the main or side chain. The various
modications of PEO based electrolytes can be divided into three main
categories:
preparation of amorphous polymer
utilization of an appropriate ionic dopant
addition of substances which reduce the crystallizing ability of the
polyether matrix.
From the viewpoint of easy preparation as well as commercial
application the last idea is the simplest one. If done properly, it could
preserve all other benets of PEO. Below we discuss main methods of
modication of PEs.
2.1. Modications of polymer electrolytes.
2.1.1. Polymer networks
One of the methods to improve mechanical properties of polymer
electrolytes is crosslinking operation. It additionally involves another
prot of polymer properties: chemical bonding between polymer
chains limits or block tendency for organization of parts of repeated
units in polymer as in consequence reduces the volume of crystalline
phase than liner ones. Most of those that were tested as matrices for
PEs were prepared using crosslinking agents (typically isocyanates)
with polyether type diols or triols. The reactions with isocyanates lead
to the preparation of crosslinked polyurethane structures which were
mostly studied by Cheradane's group [3941]. As was suggested by
Cheradane and Le Nest [39] the optimization of the conductivity of a
polymer network is achieved for a system with PEG molecular weight
equal to 1000n, containing 2n-1 mol of lithium ions per mole of PEG
units. As can be expected crosslinking with isocyanate reduces the ex-
ibility of the polyether chains which is manifested by a Tg increase. It
was generalized by Cheradane] that for a variety of networks doped
with LiClO4, Tg changes as follows:
1 1 4
7:6  10 C 1
T g T g0
- for the salt free polymer,
1 1 4 0
2:7  10 C 2
T g T 0g
- for salt complexed network.
Here, C is the crosslink concentration, T g0 is the glass transition tem-
perature at C = 0, C' is the salt concentration, and Tg is the glass transi-
tion temperature of the salt free network.
The above equations are independent of the kind of crosslinking
agents used. This is explained by the fact that the free volume of the
network is only a function of the interactions between the salt and poly-
mer chains. In fact the addition of a salt provides additional transient
crosslinking in the polymer host, equivalent to one third of that of any
crosslinking agent. Increased Tg reduces conductivities in spite of the
fact that the system is completely amorphous. This effect is attributed
to the presence of hydrogen bonding and Van der Waals interactions
between urethane linkages and the surrounding polyether. Additionally
the presence of bulky groups introduced by the addition of isocyanates
sterically hindered ionic motion of the electrolyte. Therefore in spite of
an improvement of the mechanical stability of PEG-based systems
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
chemical crosslinking has a negative effect on electrolytes' conductivity.
Watanabe and co-workers [42] have studied polyurethane networks
based on crosslinked low molecular weight poly(propylene glycol) -
PPG. The effect of a crosslinking agent and salt concentration was
found to be similar to that described above for PEG based systems.
Ambient temperature conductivities of PPG networks were in the
range 10-7-10- 10 S cm-1. The same group [43] also utilized a block
PEO-poly(ethylene urethane urea) network. The amount of poly(ethyl-
ene urethane urea) was equal to about 30 weight % and provided excel-
lent mechanical stability for the electrolyte. Salt dissolved selectively in
the polyether part gave an enhanced conductivity after approaching the
percolation threshold which results in a continuous conduction path
[43]. Polyelectrolytes [44] with a cationic part bonded to the network
were also prepared but their conductivities were considerably lower.
Considering the above mentioned results polymer networks of an en-
hanced chain exibility were needed.
Triblock copolymers incorporating EO segments were subjected to a
crosslinking reaction with isocyanates by Cheradame [39,40] and Gray
[45,46]. Cheradame prepared triblock PEO-PPO-PEO copolymers to re-
duce the polyurethane crosslink density, hence keep Tg in the range typ-
ical for polyethers [47]. On the other hand Gray [45,46] has bonded PEO
segments as a central part B of a triblock [A-B-A] copolymer with a sty-
rene-butadiene-styrene (SBS) copolymer acting as part A. Elastomeric
host materials and the triblock system happen to be self-supporting
rubbers with conductivities of 10- 5 S cm-1 at room temperature.
Giles [48,49] has synthesized amorphous phosphate esters. After the
addition of LiCF3SO3 PE samples exhibit ambient temperature con-
ductivities of about 2 10- 5 S cm- 1. However, the most pronounced
effect of increasing network exibility was achieved by the incorpo-
ration of siloxane units into the polymer framework. Cheradame's
group [50] has studied systems based on poly(dimethylsiloxane).
For systems doped with LiClO 4 they reported conductivities of
about 10- 5-10- 4 S cm- 1 at room temperature and mechanical stabil-
ity up to 120 C, although no Young modulus values were included.
Similar results were obtained by Fang et al. [51] and Wnek [52] for
their siloxane networks. Ugumi and co-workers prepared
crosslinked siloxane based electrolytes via plasma polymerization
of tris(2-methoxyethoxy)vinyl silane [4356]. Such crosslinked sys-
tems exhibited ambient temperature conductivities of about
5 10- 5 S cm- 1. Another interesting approach to copolymer systems
was presented by Balsara [57] Synthesized diblock copolymer com-
posed of polystyrene-b-poly(ethylyneoxide) and LiTFSI salt with differ-
ent blocks length of PS and PEO were intensively studied. All the
materials exhibit a great mechanical stability. The electrolytes have
conductivity value in order 105 S cm-1 at temperature 30 C. An ex-
tensive work was done to analyze morphology of that systems [5860]
depending from applied chains length or salt content. Increasing salt
amount results in lamellar, bincontinous gyroids and cylinders mor-
phology respectively for salt content from 0,01 to 0,17 [Li+]:[EO]. Ten-
dency to cylindrical morphology was observed for long polymer chain
bonded in electrolyte. Normalized conductivity value of such electro-
lytes including dimensionality of conductive phase referred to ideal
electrolyte (here pure PEO with salt) is increasing at annealed temper-
ature for longer chains length. That proves a negligible resistance of ion
transport across PS boundaries phase. Nevertheless this detailed analy-
sis let for fabrication Li battery with promising cycle life.
It should be stressed that even for a crosslinked system with en-
hanced chain exibility it is rather difcult to approach 10-5 S cm-1
for ambient temperature conductivity in network based systems.
Therefore in spite of valuable mechanical properties chemically
crosslinked polymer networks have conductivities which are still
too low for battery applications. It should be mentioned that apart
from chemical crosslinking, physical processes involving gamma,
X-ray or UV irradiation have been widely used to produce PE net-
works. In high molecular weight PEs -irradiation was used to pre-
vent the recrystallization of the melted electrolytes and stabilize
113
their amorphous structure [6163]. The PEs from PEO-LiClO 4 and
PEO-LiCF3SO3 systems were melted at 78 C, and then subjected to
-irradiation from a 60 Co source. The amorphous structure was
only stable for the PEO-LiClO4 system, which in fact was completely
amorphous up to the melting temperature. The presence of residual
crystalline phases in the PEO-LiCF3SO3 structure allowed the system
to recrystallize after crosslinking. Detailed analysis of cross-linked
polymer electrolytes was presented by Ueno et al. [64]. The electro-
lytes was successfully fabricated from the mixture of poly(ethylyne
glycols) diacrylate/dimethylether with different molecular mass
and salt (LiTFSI) using electron beam irradiation. Final material
with lithium transport number close to 0.2 was successfully applied
in lithium cell.
In conclusion, crosslinking of polymer hosts does not lead to the
formation of electrolytes having sufcient conductivities for ambient
temperature applications. However, it results in an improvement in
the mechanical properties of the polymeric systems that enables
them to work even at temperatures higher than the melting temper-
ature of the crystalline PEO phase. Crosslinking is also very useful to
ensure the mechanical stability of gel PEs containing high amounts of
plasticizers.
2.1.2. Ethylene oxide copolymers
It is well known in polymer chemistry that copolymers usually ex-
hibit a much lower tendency to crystallization than homopolymers.
This is why a lot of effort has been made to obtain copolymers contain-
ing EO segments long enough to complex effectively cations but too
short to form a crystalline phase.
The rst totally amorphous copolymers of ethylene oxide (for elec-
trochemical applications) were obtained by Nicholas et al. [65,66] via
the condensation of poly(ethylene glycols) with methylene bromide ac-
cording to the classical Williamson type reaction. In these polymers the
ethylene oxide segments of chosen length were separated by single CH2
groups which was sufcient to stop crystallization. The matrices were
very exible as indicated by the relatively low Tg. For the optimum
number of EO units (5) the room temperature conductivity after doping
with lithium salts was found to be 1-5 10-5 S cm1. Since the struc-
ture of this copolymer is very close to that of PEO this conductivity value
is believed to be characteristic for the amorphous phase of PEO.
The linear ethylene oxide (EO) copolymers with regular lengths of
EO segments can also be prepared by ring opening polymerization of
appropriate cyclic monomers, like dioxolanes [67], cyclic phosphoric
acid esters or 12-crown-4 ethers. However, the number of EO units in
repeating units is too short to achieve very high conductivity: it is higher
than for the analogous system with PEO, but signicantly lower than
that expected for fully amorphous PEO. Thus these very elegant syn-
thetic methods are of no practical importance.
It was demonstrated that random copolymerization of EO with
propylene oxide, styrene oxide or epichlorohydrin in the presence of
partially hydrolysed organoaluminum compounds produced amor-
phous products which, at an appropriate composition and after doping
with inorganic salts, exhibited higher conductivity than analogous elec-
trolytes based on PEO [6871]. The best results were obtained for a co-
polymer with propylene oxide of an average length of EO segments
equal to 5. It is interesting that after doping with lithium salts this copol-
ymer exhibits a signicantly lower Tg (210 K) than systems based on
PEO (240-250 K). The maximum room temperature conductivity mea-
sured for the copolymer based electrolytes exceeded 10-4 S cm 1,
which is higher than expected for pure amorphous electrolytes having
PEO chains but still not enough for commercial applications. Further-
more, this copolymer electrolyte had a higher conductivity than pristine
PEO at temperatures exceeding the melting point of the PEO crystalline
phase. These observations lead to conclusion that the copolymerization
may lead not only to the elimination of crystallinity but also to an im-
provement in ionic mobility and thus the conducting properties of the
amorphous phase. This probably results from the internal plasticization
114 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
of the polymeric matrix due to the presence of a small amount of methyl
groups introduced by propylene oxide monomeric units.
Detailed studies of the role of the kind and concentration of alkali
metal salt on the conductivity of copolymer electrolytes were per-
formed for PEO-PPO system comprising 84 mol% of EO monomeric
units [71,72]. The highest conductivities, exceeding 10-4 S cm 1 at
ambient temperatures, were measured for copolymers doped with
LiBF4. However, these highly conductive uoroborate and perchlorate
doped polymers ow at temperatures slightly exceeding 40 C whereas
iodide and uoroacetate doped electrolytes are mechanically and ther-
mally stable up to 100 C.
Using the same type of organoaluminum catalysts Ballard et al. [73]
obtained a large number of EO copolymers with various co-monomers
and prepared many electrolytes with room temperature conductivity
above 10-5 S cm1. Therefore this relatively simple method of modi-
cation seems to be quite general. It is crucial that the differences in co-
monomer reactivities in the presence of aluminium based catalysts are
not signicant. Thus it is possible to prepare a statistical distribution of
co-monomers which provided these systems with appropriate lengths
of EO segments (4,5,6 monomeric units).
Conductivity of polyether based matrices modied with highly polar
co-monomers like CO2 or SO2 was also investigated [74,75]. However,
the polycarbonates obtained in the reaction with CO2 turned out to
be very poor conductors in spite of the fact that their low molecular
weight analogue cyclic carbonates are regarded as the most suitable
aprotic solvents for liquid electrolytes. Polysulphites and poly(ether
sulphites) obtained in copolymerisation of SO2 with EO exhibited
only slightly better conducting properties than PEO. DSC studies
showed that polysulphites exhibited stronger cation complexing
properties than PEO, thus the strong interaction between the chain
and cation might reduce the mobility of charge carriers.2 However
polysulphites interact very strongly with PEO chains and hinder
their crystallization. Thus mixtures of PEO and polysulphites exhibit
good conducting properties.
Comb-like copolymers form another group of amorphous matrices
for PEs. In these systems short-chain coordinating olygomers are at-
tached to inert polymer backbones in a comb-like fashion. The role of
a backbone is to maintain the mechanical stability of a PE whereas
short side chain polyglycols [n = 7-22] coordinate cations. At the begin-
ning high Tg metacrylates were used as main chain polymers [7679].
The highest conductivities achieved in the systems studied are in the
range 10- 5-10-6 S cm 1. Similar conductivities were obtained for
comb-like systems utilizing polyitaconates [80] or poly (-methyl-2
gluconate) as main chains. Conductivity depends strongly on the length
of the pedant chains as well as on the kind and concentration of the salt
added [81]. Synthesis of comb like polymer electrolyte with elastic main
chain was presented by Ikeda [82,83]. Polymer matrices with average
molecular mass in order 106 g mol1 consist ether units with bonded
tri(oxyethylene) segments as side chains. Increasing number of side
chains results in increasing exibility and ionic conductivity of material
proved by decreasing glass transition temperature and melting point
temperature. Measured conductivity of electrolyte with LiClO4 as
doped salt is 1.5 10-4 S cm1.
The dependence on the structure of the main chain was negligi-
ble. It was assumed that substitution of the rigid main chain by
exible polyphosphazene (T g = -70 C) or polydimethylsiloxane
(Tg = -120 C) backbones would provide better ambient tempera-
ture conductivities of the comb-like systems. A series of polysiloxane
electrolytes were synthesized by Nagaoka [84], Hall [85] and Smid
et al. [8689]]. High conductivity values exceeding 10- 4 S cm 1 at
room temperature were obtained for all of the samples studied. The
electrolytes were prepared in the form of a liquid or wax rather
than a self-standing solid. For some of the electrolytes recrystallization
2
A similar problem is observed in some matrices comprising polyamines.
of side chains was evident. The mechanical properties can be improved
and recrystallization suppressed by crosslinking which in turn lowers
the conductivity by an order of magnitude. Additionally Si-O-C bonds
are easy to hydrolyse and substantial degradation of the electrolyte
structure occurs in the presence of traces of moisture. The stability of
Si-C bonds is much better than Si-O-C and therefore the comb-like
polysiloxanes were also synthesized [88]. For such systems Skotheim
et al. [89] obtained conductivities of 10-5-10-7 S cm1. Nevertheless
the paste-like morphology of this group of electrolytes is still a limita-
tion for applications.
Blonsky and co-workers have synthesized a wide range of
polyphosphazene based electrolytes by polycondensation reaction
of poly (dichloro-phosphazene) with sodium 2-(2-methoxyethoxy-
ethoxy)ethanolate [90,91]. The polymers structure depended strongly
on the kind of dopant salt and its concentration. They ranged from a
exible plastic to a rigid rubber with an increase in salt concentration.
This was accompanied by a considerable increase in Tg. Studies of the
role of kinds of cations on ionic conductivity and cation transference
number were performed [91] for a series of triate salts and
showed that the highest conductivity values are obtained for sam-
ples doped with AgCF3SO3 ( = 10- 5-10- 4 S cm 1 at room tem-
perature, t+ = 0.32). Authors showed that with very short side
chains main chain nitrogen or phosphorus atoms can also be involved
in complexation of cations.
Polyphosphazene-based electrolytes usually exhibit a weak tem-
perature dependence of ionic conductivity. The main limitation to-
wards their application in electrochemical devices is related to
poor mechanical properties. Linear polyphosphazenes ow under
pressure, particularly at temperatures exceeding 70 C. To avoid me-
chanical instability of the comb-like PEs several methods of modi-
cation have been applied [75,76]. However it is now believed that
the limitations for polyphosphazene use are not only due to poor
mechanical stability but also due to a narrow electrochemical stabil-
ity window for these electrolytes. Cyclic voltammetry experiments
showed degradation of the electrolyte at potentials higher than
1.6 V versus lithium electrode which is considerably lower than for
PEO based systems [92].
To summarize: copolymer based solid electrolytes exhibit high am-
bient temperature conductivities but their electrochemical stability
usually was not studied. Preliminary studies [70] in a cell utilizing lithi-
um electrodes showed similar behaviour to that previously reported for
PEO complexes above the melting point, namely formation of resistive
layers at the electrodeelectrolyte interface. Poor mechanical stability
is a limitation in any application. Furthermore, copolymer matrices of
high conductivity are not commercially available.
Summarising the above section it has to be written that number of
publications in the area of cross-linked electrolytes is decreasing each
year. There is rather new trend observed, where all described methods
are mixed giving i.e. more complicated cross-linked electrolytes, where
inorganic blocks are incorporated in matrices or anions are chemically
bonded to that matrices. Materials like that are intermediate structures
between composite and cross-linked electrolytes. [5760]
2.1.3. Single ion conductors
Polyelectrolytes are systems in which one of the conducting species
(anion or cation) is chemically bonded to the polymer backbone. The
counter ion (either positive or negative), which is outside the back-
bone, balances the charge and is free to move. Usually the dielectric
constant of the polyelectrolyte is too low to ensure the dissociation
of the polyelectrolyte system into separated anions and cations.
Therefore a highly polar solvent is added in order to promote disso-
ciation. In polyelectrolytes ionic transport occurs in a quasi-liquid
phase i.e. counter ions move through polar solvent domains trapped
in a polymer membrane. A typical example of such a polyelectrolyte
is Naon proton conducting membrane in which anions are bound
to the polymer chain and solvated protons can move in the electric
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
eld. In polyelectrolytes either the cation or the anion transport
number is equal to unity. However, special interest is still devoted
to the preparation of polyelectrolytes in which ionic transport will
occur in a solid state (without addition of solvents).
In standard PEs both the cation and anion are mobile, however
anions are usually much more mobile due to weak interactions with
the polymer. The synthesis of single cation conductive exible PEs is
especially important for application in alkali metal batteries. During
the operation of a battery (in which both electrodes are reversible to-
wards cations but blocking against anions) bulky anions accumulate
near the anode and form a space charge layer. It interacts with the
anode leading to an increase of the overall electrode-electrolyte inter-
face impedance, obstructing the cation transport from the anode to
the bulk of the electrolyte.
For a while it was believed that in order to obtain a single cation
conducting system the counter ions should be sufciently large or
chemically attached to the backbone as in the case of typical poly-
electrolytes. However, it was demonstrated by M. Armand et al.
[93] that even relatively bulky anions are free to move in the PEO
matrix. Experiments on a series of salts of the general formula
LiCn F2n + 1SO3(n = 1-25) showed that the highest conductivities
were obtained for an electrolyte with n = 14 followed by only a
small decrease in conductivity for electrolytes with larger anions.
A large group of polyelectrolytes as single (cation or anion) PEs
have been studied [94]. Most of the systems studied showed moder-
ate conductivities at temperatures exceeding 100 C due to the
stiffness of the main polyanion chain as well as tight ion pairing
occurring in absence of polar liquid solvents. Many other systems
based on the nylon-1 backbone [94], polyphosphazenes [95,96] or
poly(olygo(oxyethylene methacrylate-co-alkali metal methacrylate))
[9799] or polybrene [100] were studied but their conductivities hardly
exceeded 10-7-10-8 S cm1 at 100 C, far too low for application in
electrochemical devices
Ohno and co-workers [101] have performed systematic studies
on the effect of the kind of alkali metal salt cations used on the con-
ductivity of polyelectrolytes based on the poly [(-carboxy)-olygo
(oxyelthylene) methacrylate] system. It was shown that conductivi-
ty increases with increasing size of the alkali metal cation from lith-
ium to caesium. Similar results were obtained by Bernabah et all.
[102] where tertpolymer composed of ethylene oxide, glycidoxy-2-
tetrauoro-ethanesulfonate and ally-glicydyl-ether with cations of
lithium, sodium and potassium. Opposite results were reported by
Florjanczyk and co-workers for comb-like polyelectrolyes based
on half esters of maleic anhydridestyrene copolymer with PEG
[68,103]. In this system an ion-pair is located close to the complexing
oxyethylene segments which enables relatively easy dissociation.
For the small lithium cation the optimum number of EO molecular
units was equal to 6 whereas for the larger potassium cation the opti-
mum number of EO molecular units was equal to 12. The conductivity
for some lithium salts is just on the border line of the 10-5 S cm1 re-
quired for practical applications. Easy sample preparation and commer-
cial availability of the substrates is also an advantage of this system.
2.1.4. Plasticized systems and polymeric gels
A common way to increase the exibility of host polymer is the
addition of a plasticizer. Same strategy can be applied to PEs. The plasti-
cizer should full the following requirements:
decrease the crystallinity and increase the mobility of the host poly-
mer chains below the melting temperature,
be miscible with the amorphous polymer phase to which a dopant salt
is added,
have a low vapour pressure in the projected application temperature
range,
be stable against electrodes (both low and high potential), dopant salt
and other cell components.
115
Low molecular weight polyglycols are often used as plasticizers for
PEO-alkali metal salt electrolytes. Kelly and co-workers [104,105] and
Tsuchida et al. [106] have examined the role of the polyether end
group on their plasticizing properties and the behaviour of the plasti-
cized electrolyte. It was shown that hydroxyl end capped PEG can
dissolve much more salt than end O-acetylated PEG, giving rise to
increased conductivity. On the other hand end-hydroxyl groups
can react with lithium leading to the formation of resistive inter-
face layers [104]. Kelly et al. [105] has suggested the utilization of
poly(dimethoxy ethylene glycols)(PEGDME) which are more sta-
ble against lithium than their hydroxyl end capped analogues.
The addition of PEGDME to a (PEO)8LiCF3SO3 electrolyte increased
the conductivity up to 5 10-5-10-4 S cm1 at ambient temperatures.
Wang et al. [107] studied electrolytes of the following general formula;
LiCF3SO3 (0.5PEO + 0.5Y),where Y = PEG,(Mw =600 g mol1), LiPEG
(Mw = 600 g mol1), PEGDME (Mw = 400 g mol1) or PEGDME
(Mw = 750 g mol 1). DSC experiments showed a decrease in Tg
from -21 C for the non-plasticized (PEO)9LiCF3SO3 electrolyte to
-48 C for plasticized systems independent of the plasticizer used.
NMR experiments conrmed an increase of the segmental mobility of
the polymer chains after the addition of a plasticizer. The highest
increase of conductivity was measured for electrolytes containing PEG
(rt = 10-5 S cm1) and the lowest for samples with the LiPEG addi-
tive (rt = 10-6 S cm1).
It should be stressed that electrolytes based on plasticized linear
polymers are often of poor mechanical stability. This disadvantage
greatly limits the possibility of their application. As was already men-
tioned crosslinked polymers provided electrolytes of a much better
mechanical stability but they suffer from limited ionic conductivity.
Another possible architecture that could offer a remedy is a three
phase system consisting of polymer network and a solution of inorganic
salt in low molecular weight polar solvents that forms a polymer gel.
Such gels have been intensively studied world-wide in the past and
gave birth to the so-called Li-polymer technology developed originally
at Bellcore in early 1990's.
Polymer gels are fundamentally different from standard PEs since
the charge transport occurs mainly in the plasticizer phase containing
the dissolved salt, while the polymer plays the role of a supporting
exible matrix. Nevertheless it should be noted that molecular model-
ling by G. Farrington [108,109] point out that the polymer phase may
contribute somewhat to ionic transport in gel electrolytes. The exact
nature of polymer-plasticizer-salt interactions is complicated and
not completely understood as highlighted in some contradictory obser-
vations. Some authors reported an increase in conductivity with an in-
crease in the amount of salt added [110,111]. Others stressed the
importance of salt to plasticizer ratio [106,112,113]. It was found [114]
that up to certain critical salt concentrations the conductivity increases
and then remains fairly constant. However the value of this critical
concentration depends strongly on the salt to plasticizer ratio. The
highest conductivities can be obtained for plasticizers characterized by
low viscosity and high dielectric constant. However, since an increase
of viscosity is generally related to an increase in the dielectric constant
the later requirement is not easily achieved.
Assuming the crucial role of the liquid phase, electrical properties of
the polymer host should be unimportant. Therefore it is not necessary to
utilize ionically conductive macromolecules. Many polymers, like poly-
styrene, poly(vinyl chloride), poly(vinyl acetate), polyacrylonitrile-
(PAN) and poly(vinyl diuoride) (PVdF) [110,112,113] were used as
matrices for polymer gels. It was found that only polymers possessing
high dipole moment like PAN or PVdF are suitable candidates. DMF,
PC, EC, H2O, PEG were used as plasticizers. For application in lithium
batteries only substances stable against lithium are of interest. There-
fore PC, EC and PEGDME are mostly used as plasticizing agents. The
range of conductivity achieved at ambient and sub-ambient tempera-
tures is 10-4-10-3 S cm1. It should be stressed however, that polymer
gels suffer from the drawbacks observed in lithium devices utilizing
116 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
classical liquid electrolytes (ammability, sensitivity to elevated tem-
peratures etc.) [115]. Scrosati [116] reported on the formation of
dendritic structures in lithium batteries utilizing gel electrolytes which
lead to shortcircuits in the devices.
The interfacial stability of electrolytes is therefore of particular
importance. Hong et al. [117] studied the electrochemical stability of
the PAN-(EC/PC)-LiClO4 system by impedance spectroscopy in a sym-
metric lithium cell. Interfacial resistance grew over the rst two days
of experiments and then stabilized. The increase in the interfacial resis-
tance was attributed to the formation of passivating layers which are
usually observed after the contact of an organic solvent with lithium.
The conductivities were about 10-4-10-3 S cm1 at room temperature.
Various groups have intensively studied the so-called MHB electro-
lyte introduced primarily by Lundsgar et al. [118,119]. The gel was
based on matrices obtained by radiation induced curing of ethylene gly-
col dimethacrylate or trimethylpropane trimethacrylate. PC, EC or their
mixtures were used as plasticizers and LiCF3SO3, LiClO4, or LiAsF6 were
applied as dopant salts. It was shown that mixing two plasticizers in-
creases the thermal stability of the gel electrolytes as well as extending
their stability window towards the lithium electrode [120]. The widest
stability window was obtained for a PC to EC ratio equal to 1:1. The gel
obtained was only slightly less conductive (rt = 8 10- 4 S cm 1)
than gels plasticized with pure PC (rt = 2.3 10- 3 S cm 1). Even
more importantly, the formation of lithium dendrites was not ob-
served when gel electrolytes based on this plasticizing mixture was
used in a microbattery. The mass loss of the described gel electrolytes
was found to be less than 35% by weight at temperatures up to 150 C.
The sub-ambient temperature conductivities are still within the
10-3 S cm1 range. Lithium batteries utilizing such an electrolyte were
showed capable of supporting current densities of up to 30 mA cm2
for over 1 minute.
Polymeric gels prepared by the polymerization or copolymerization
of glycidyl methacrylate with various vinyl monomers and carried out
in a plasticizer environment were also examined by our group [121,
122]. It was shown that various methods of crosslinking lead to different
exibilities of the polymer hosts and therefore to different ionic conduc-
tivities. The best transport properties were obtained for polymer matri-
ces utilizing copolymers of polar substances, such as acrylonitrile or
acrylamide crosslinked with SO2 [121]. The ambient temperature con-
ductivities of such derivatives exceeded 4x10-3 S cm1. It was also rec-
ognized that the utilization of lithium triate as a dopant salt leads to
conductivities of about one order of magnitude lower in comparison
to LiClO4. Additionally cationic conducting comb-like electrolytes previ-
ously studied by us were plasticized with propylene carbonate [122].
The addition of plasticizer at 30% w/w lead to an increase of conductivity
to 5 10-4 S cm1 at ambient temperatures.
Poly(vinylidene uoride-co-hexauoropropyene) (PVdF/HFP) is
another alternative for GPC matrix. It offers good mechanical and
electrochemical stability, as well as high capacity for liquid electrolyte
absorption. Physicochemical properties of resulting gels can be modi-
ed by:
- changing lithium salts
- changing the organic solvents
- replacing classical electrolytes with ionic liquids [123],
- addition of inorganic llers (described in more details later in the
manuscript):
- unmodied and surface modied silicas [124127],
- unmodied and surface modied aluminium and titanium oxide
[128130],
- anion or cation receptors [131,132]
Such PVdF/HFP-based electrolytes containing amorphous silica
ller has been commercialized as a part of the previously mentioned
Bellcore technology.
To summarize, despite the several advantages of polymer gel
technology, there are also severe intrinsic drawbacks. The main one
is the utilization of highly polar organic solvents which are generally
unstable in contact with alkali metal electrodes. Resulting chemical
and electrochemical processes at the electrode gel electrolyte inter-
face lead to the formation of passive layers with resistances exceeding
the bulk resistance of the electrolyte. Furthermore the formation of lith-
ium dendrites across the cell leading to short-circuiting effects cannot
be excluded.
2.1.5. Ceramic llers
One of the most successful approaches to modify the structure of
polyether based electrolytes was to synthesize composite PEs with ad-
dition of ceramic particles. The initial idea behind using llers composed
of ceramic fast ionic conductors was the expectation to get percolation
pathways composed of inorganic llers grains through a exible poly-
meric matrix. Such phenomenon could lead to an increase in ionic con-
ductivity followed, possibly, by an enhancement of the cation transport
number while preserving mechanical properties and exibility of the
composite electrolyte prepared in the thin lm conguration. Such ma-
terials also offer good adhesion to classically prepared electrodes. This
concept was explored by several research groups including the authors
of this review.It was shown that llers like NASICON, -alumina, or
glassy llers do not contriubute to ionic conductivity of mixed phase
system [133140]. Skaarup et al. [141,142] demonstrate decrease in
conductivity of composite in comparison with pristine ceramic electro-
lyte due to dilution effect. This phenomenon is observed above 85%v/v
of ller where polymer acts as binder of ceramic grains. Similar results
were given by Stevens and Mallander [143] for PEO and RbAg4I5 or
Kar4I5 conducting ceramic ller.
Mixed phase conducting system did not improve enough polymer
electrolyte properties. Parallel It was shown that addition of electro-
chemically inactive organic llers improve ionic conductivity [144]
and mechanical properties [145]. The ne inorganic ller (grain size
1-3 m) improve mechanical properties and conductivity at tempera-
ture below crystalline phase melting [146148]. This feature is strongly
connected with weak conducting crystalline phase volume what was
conrmed by XRD DSC or NMR analysis. Additional prot of inert inor-
ganic ller incorporation is connected with PE interface improvement
what was shown by Scrosati [140,149151]. All of the modications of
polymer solid electrolytes presented so far focused exclusively on the
enhancement of ionic conductivity. Below the concept of the use of
specially design additives to enhance lithium transference number is
briey presented.
The Lewis denition of acids and bases serves as a useful tool for
designing llers which might act as anionic receptors, thus, possibly
increasing the cation transport number. Scrosatis group [152,153]
demonstrate an increase of transference number in modied PE.
Registered values were much below unity. Lewis acid base theory in-
dicate another way for modication of ceramic additive. System
based on superacidic ller were investigated [154160]. Where acidity
of H0 about 15 on Hammett scale can be obtained [161].
Several studies using this concept have been published over the
years, however it wasnt until 2010 when it was realized that such
strong acids can catalyse scission of the PEO chain in presence of traces
of water [162]. Hence shall any water be present during (or after) the
preparation of such composites, the molecular weight of the polymer
can be signicantly decreased (as showed by GPC) and the short PEO
olygomers can plasticize the matrix in a fashion described earlier in
this paper. The side effect is deterioration of mechanical stability of
the lm.
Another issue is the uniformity of the ller dispersion. In order to
achieve it, very strict adherence to certain protocols needs to be follow-
ed, in terms of the sonication of the ller, viscosity of the PEO + salt so-
lution and the mixing of the two in order to avoid agglomeration of the
ller, its sedimentation in the casting slurry, as well as separation of the
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126 117

ller on the polymer-solvent interface while the two liquids are mixed C6P in coordination of anions BF
4 AsF6 N ClO4 N CF3SO3 N PF6 N I .

[163]. Sonication of the polymer itself should also be avoided as it was
shown that it leads to a decrease in its molecular weight. That in turns
affects the mechanical stability of the nal material. Last but not least
the thermal history of the composite affects signicantly its transport
properties. While comparing results from various groups it is essential
to take into account all experimental parameters because the tempera-
ture of the drying step (or the order of conductivity data acquisition)
can introduce a few orders of magnitude difference.
For well-designed and prepared materials, signicant enhancements
of conductivities can be achieved. Using surface-modied llers with
superacidic groups very high Li transference numbers and excellent sta-
bilities vs. Li metal electrodes upon extended periods of time were
found [163].
We would also like to draw attention to the newly developed con-
cept of using reversed phase systems, i.e. porous ceramic sponge lled
with liquid form of PE [164166]. It is yet another way of separating
the conicting requirements for the material to be rigid (and provide
mechanical stability) and good conductivity (which requires the
polymer to be exible, hence soft). These studies did not only show
the feasibility of such approach, but also provided a direct evidence
of polymer-ceramic interactions. When using ceramic pieces with
perfectly ordered pores, preferential orientation of polymer crystal-
lites was observed. This implies that polymer ordering at the ceramic
surface occurs. However this cannot be directly observed for random
pore structure, or particulate llers.
A further increase in the fraction of C6P results in only a small increase
in the lithium transference number.
An important parameter in the characterization of PEs is the window
of electrochemical stability, especially in view of applications for lithium
and lithium-ion all-solid-state batteries which is required to allow a se-
lection of redox couples as electrode materials. Given that ether-based
electrolytes are known to oxidize at relatively low voltages (~3.5 V vs.
Li/Li+) it was of interest to determine the electrochemical stability of
the LiCF3SO3:P(EO)n electrolytes in the presence of C6P and ceramic ll-
er. Cyclic voltammetry showed at 75 C, additionof the C6P additive
caused a positive effect on the cell's stability in compared to the pure
PE over a wide voltage range (0.0 4.0 V) [183]. Supplementary addi-
tion of silica caused an even greater electrochemical stability over a
wider potential range of 0.0 5.0 V. The same system was successfully
applied in all solid-solid-state lithium-ion battery. The capacity loss of
the batteries with the double modied PE is lower than those with
pure one and does not exceed 0.4%/cycle at 90 C. At 75 C the capacity
is very stable at the rst 20 cycles. It has been clearly proved that by
proper design of inorganic or organic ller properties affecting ion
transport phenomena can be modied in a variety of polyether based
electrolytes. The addition of llers results in an increase in ionic conduc-
tivity for amorphous and crystalline polymeric matrices, enhancement
in lithium transference number and stabilization of lithium electrode
PE interface [184].

2.1.6. Boron-organic additives 3. Ionic liquids

Anion receptors based on boron compounds were applied to the so-
lutions of lithium salts in aprotic (inert) electrolyte based on low molec- In the quest for safer batteries, the present choice of organic carbon-
ular weight solvents [167,168] as well as in gel polyelectrolytes [169]. ates and LiPF6 as solute is questioned, and new electrolyte concepts
Boron-based aza-ether compounds (borane and borate complexes) have to be considered. The main drawback of present technology is
have been studied by McBreen and co-workers [170173] using the runaway reaction from graphite exfoliation (negative electrode)
mainly Near Edge X-ray Absorption Fine Structure spectroscopy and/or O2 release (positive electrode) leading to the expulsion of am-
(NEXAFS). These studies showed that the degree of complexation mable vapours and subsequent self-ignition, and toxic chemicals (HF)
of Cl or I anions strongly depends on the structure of the boron release. Ionic liquids are considered to provide a solution to this prob-
compound. Also the dramatic enhancement in ionic conductivity lem [185].
upon their addition. Wieczoreks group already analyzed properties ILs are ionic compounds and by consensus, have a melting point
of triphenylborane (Ph3B) [174] in PE systems and we observed an below 100 C. Of course, the focus here is on room- or below room-
increase of lithium transference number. This benecial effect was temperature liquids. ILs are the focus of interest of a growing communi-
followed by changes in infrared and Raman spectra that conrmed ty for their unusual properties:
complex formation [173]. Some novel boron-organic compounds
have been also tested in order to verify their stability and application Negligible vapour pressure,
as anion trapping agents [175]. Non-ammability,
High thermal stability,
2.1.7. Supramolecular additives
Another approach leading to enhancement in lithium transference Table 3
number is to use anionic receptors: boron compounds [176,177], linear present values of lithium transference numbers measured for PEO-LiClO4 and PEO-LiBF4
or cyclic aza-ether compounds [178] or calix [4] arene derivatives with electrolytes and composite electrolytes based on these model systems with surface mod-
various types of active groups in the lower rim [179,180]. However, this ied supearcid Al2O3 and ZrO2 additives [137]. The lithium transference number increases
with the addition of alumina ller and a further increase is observed for electrolytes with
effect was quite frequently observed for relatively large fractions of the
surface modied additives. It should be noticed that the higher the acidic groups concen-
supramolecular additive which act as a steric hindrance, thus, lowering tration the higher is lithium transference number. Similar observations can be made for
electrolyte conductivity. Recently some of these limitations were over- composite electrolytes containing ZrO2 [137].
come when calix[6]pyrrole (C6P) was used as an anion trapping group
Type of the Molar fraction of Temperature/C Lithium transference
[181]. electrolyte calix-6-pyrrole number
Table 3 presents values of lithium transference numbers obtained for
(PEO)20LiI 0 70 0.25
PEO based electrolytes doped with various type of lithium salts and (PEO)20LiI 0.125 70 0.56
containing various amounts of C6P type supramolecular additive. The (PEO)20LiAsF6 0 75 0.44
addition of an even small molar fraction (~0.125) of C6P results in a con- (PEO)20LiAsF6 0.5 75 0.84
siderable increase of the lithium transference number (for this supra- (PEO)20LiBF4 0 70 0.32
(PEO)20LiBF4 0.125 70 0.78
molecular additive concentration all composite electrolytes seem to be
(PEO)20LiBF4 0.25 70 0.81
homogenous). The increase in lithium transference number is particu- (PEO)20LiBF4 0.5 70 0.85
larly clear for PEO-LiAsF6 and PEO-LiBF4 electrolytes. A smaller enhance- (PEO)100LiBF4 0.25 70 0.95
ment has been achieved for PEO-LiCF3SO3 system and the smallest one (PEO)100LiBF4 1 70 0.92
for PEO-LiI electrolytes. These observations are in good correlation with (PEO)20LiCF3SO3 0 75 0.45
(PEO)20LiCF3SO3 0.125 75 0.68
computational calculations [182] showing the following preference of
118 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
High conductivity up to 20 mS cm-1 at room temperature,
Very high resistance to oxidation.
The non-ammability has been the main attraction for battery engi-
neers, solving the most urgent electrolyte problem and it is hoped that
their resistance to oxidation would allow the use of high voltage elec-
trode materials (e.g. LiMn1.5Ni0.5O2 at 4.9 V).
The family of ILs, now with innumerable representatives, stems from
the study of chloroaluminates (AlCl-4, Al2Cl-7) of the delocalized-charge
cations based on imidazolium derivatives. However, the AlCl bond is
too fragile in reducing conditions (Al plating at +1.3 V vs. Li/Li+) and
hydrolyzes easily with traces of water (giving off HCl), to have any ap-
plication in the battery eld. They can also react with many common
solvents. The advent of neutral ILs, with the classical anions of Li
salts: BF-4, CF3SO-3, PF-6; and imidazolium cations has given large visibility
to the eld in the early 1990s.
While a large number of metals (Cu, Ag, Au Pd, Al, Fe, Ni, Co possibly
Nb and Ta) and non-metals like Si and Ge, important for solar ap-
plications, can be easily plated from ILs, the challenge of graphite in-
tercalation or lithium deposition in ILs is still largely unresolved. The
limitation of these new ILs was found not to come from the anion, but
the imidazolium cation. The acidity of the C2 proton is estimated to
pKa = 24, and this corresponds to a reduction potential of 1.5 V vs.
Li+/Li, which is observed in practice. The methylation of the C2 proton
still leaves possibility for reduction, and only ~ 300 mV are gained.
Thus, the imidazolium-based cations when in contact with electrodes
at b 1.2 V vs. Li/Li+ are always unstable and thus rely on the formation
of a protective SEI, similarly to classical, carbonate-based electrolytes.
The introduction of the TFSI [(CF3SO2)N]- in 1995 has considerably
enriched the population of ILs, as the soft, exible anion could now
give ILs with very low freezing point (or low Tgs) and not restricted to
the imidazolium derivatives. In particular, quaternary ammonium
(QA) salts, which are appreciably more resistant to reduction as com-
pared with azoles, were now giving low melting salts.
Despite higher viscosities and lower conductivities, QAs doped
with low-lattice energy lithium salts, either based on the compact
5-membred pyrrolidinium cycles (MPPy, BMPy) or those whose side
arm exibility is enhanced through an ether linkage (1,2,2,2O1), are
now the focus of most applications in the battery eld. However, oper-
ating in the 0800 mV domain still depends on the SEI.
With the same approach to soft anions, i.e. delocalization of
charge and the presence of N or C centres has even further broadened
the scope and temperature of operation of ILs. These anions tend to
give lower melting points and/or viscosities as compared with con-
ventional PF-6 or BF -4 . However, the very positive voltages N 4.5 V
can only be reached with the uorine containing anions: TFSI and
FSI. FSI is considered the most promising in terms of conductivity
and shows low rise in viscosity upon Li salt addition. In addition, FSI
has been shown to provide the ad hoc SEI layer to allow dendrite-free
cycling of lithium metal at the point of reconsidering this anode as a vi-
able alternative for high energy batteries [185]. ILs based on anions that
give corrosion of Al current collector (TFSI) in normal carbonate sol-
vents are far more tame in ILs, and 4+ volt operation has been demon-
strated. Here again FSI with its labile SF bond (similar to the PF bond
in LiPF6) passivates Al.
The non-uorinated anions on the other hand (like P(CN) 6 ) are stable
up to 4 V cut-off but have the advantage of being more recyclable and
possibly less expensive. It is not known if, and then under what condi-
tions, these anions under thermal stress can release HCN but studies are
ongoing. Obviously, such electrolytes are in the right potential range for
LiFePO4 and it has been shown that they are compatible with Li4Ti5O12.
3.1. Ionic liquids as additives
A non-obvious characteristic of ILs is to induce non-ammability to
organic solvent beyond a threshold composition. For instance
carbonates solvents and LiPF6 with 20-50% addition of an IL become
ame retardant. This is a mixed benet because in case of runaway
reaction the organic solvent vapours may distil-off and catch re out-
side the battery. However ILs can be added directly to high boiling cyclic
carbonates alone (EC and PC) omitting the more volatile dimethyl or
ethyl-methyl carbonates. Conductivities remain reasonable even with-
out the low-viscosity diluents. Graphite electrode operation has been
shown to be feasible in such solvent mixtures. Ragone plots show that
the performances (C rates) are still slightly below that of conventional
solvents, the difference is now marginal for FSI-based ILs.
ILs, as a whole, get a lot attention in numerous elds of science. It is
very clear that the battery community cannot ignore this option for safe
operation of batteries.
4. Other approaches
So far we have described incremental progress of well-established
electrolyte systems. Below the ideas leading to fundamental changes
in the designing of new electrolytes are presented.
4.1. Crystalline polymer electrolytes
Conductivity in PEs has long been viewed as conned to the amor-
phous phase above the glass transition temperature, Tg, where polymer
chain motion creates a dynamic, disordered environment that plays a
critical role in facilitating ion transport. Crystalline polymer-salt com-
plexes were considered to be insulators until the discovery of ionic con-
ductivity in PEO6:LiXF6 (X = P, As, Sb) [28,186]. It is the unique
structural features of these complexes that promote ionic conductivity
[187,188]. Pairs of PEO chains fold to form cylindrical tunnels within
which the Li+ ions are located and coordinated by the ether oxygen
atoms, providing pathways for the movement of cations. The anions
are located outside these tunnels in the interchain space and do not Co-
ordinate the cations [28]. It has been shown that the current is solely
carried by Li+ cations, hence the transport number t+ = 1.
Ionic conductivity of pristine crystalline PEs formed by LiXF6 salts
dissolved in PEO is too low for applications, however, it can be increased
by several orders of magnitude by isovalent and/or aliovalent anionic
doping, modifying the ends of the polymer chains or glymes, varying
the average molecular weight of the polymer and/or its dispersion
[189193].
Ionic conductivity in crystalline polymers is not unique to the 6:1
complexes formed with LiXF6 salts. Several new crystalline PEs contain-
ing different alkali metal salts (Na+, K+ and Rb+) were reported [194].
The structures of PEO8:YAsF6 (Y = Na, K, Rb) also contain tunnels but,
unlike the PEO6:LiXF6 structures, each tunnel is formed by a single
PEO chain folding into a helix. The anions are also located outside the
polymer chains and do not coordinate cations. The ionic conductivity
at and above room temperature of the best conductor PEO8:NaAsF6 dis-
covered so far, is 1.5 orders of magnitude higher than that PEO6:LiAsF6.
Rocking-chair batteries were constructed using Na0.44MnO2 electrodes
and PEO8:NaAsF6 electrolyte. Sustainable cycling demonstrated applica-
bility of such electrolytes.
4.2. Small-molecule electrolyte
Crystalline small-molecule electrolytes form a new class of ionic
conductors that are different from ceramic or PEs. They are soft solids,
unlike ceramic electrolytes, yet, unlike PEs, they are highly crystalline,
of low molecular weight, and have no polydispersity (distribution in
chain length) or chain entanglement. They are solid coordination com-
pounds in which the cations of a salt are coordinated by small mole-
cules, glymes CH3-O-(CH2CH2O)n-CH3 1 n b 15 (hereafter denoted
Gn). They do not exhibit plasticity and are therefore different from plas-
tic crystalline ionic conductors. Crystalline PEs form the same crystal
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
structures over a very wide range of molecular weights, from greater
than ~500 to several million g mol1.
On reducing the molecular weight below 500 g mol 1, a rich
variety of crystal structures is observed[195200]. A number of the
salt/small-molecule complexes, e.g. (G4)0.5:LiBF4, G3:LiAsF6, G4:LiAsF6,
show appreciable levels of ionic conductivity, greater than the undoped
PEs [199,201]. In addition, G3:LiAsF6 and (G4)0.5:LiBF4 exhibit high
values of the cation transport number, t+: 0.80 and 0.66, respectively.
As in the case of PEs, it is the specic structural features that determine
higher conductivity and greater t+ values. Higher conductivities and
cation transport values are found in those small-molecule electrolytes
whose structures feature convenient pathways for the cations to move.
4.3. Polymer-in-salt and condensed electrolytes
An effective approach to achieve PEs with pure cation conductivity
in a solvent free conguration appears to be the synthesis of so-called
polymer-in-salt systems, in which a large amount of salt is mixed with
a small fraction of polymer to induce mechanical integrity of the nal
membrane [202,203]. Amorphous lithium salts exhibit room conductiv-
ity in the range 102 103 S cm1, i.e. values which are suitable for
battery applications. However, the addition of the polymer component
usually results in a considerable decay in conductivity [204]. In previous
studies poly(acrylonitrile), poly(1-vinyl pyrrolidone) and poly(N,N-di-
methyl arylamide) were mostly used as polymer matrices [204206].
The highest ambient temperature conductivity exceeding
106 S cm1 was reported for poly(acrylonitrile)-lithium triate sys-
tem [206]. This unfavorable effect of reduction in conductivity upon the
addition of polymer can be suppressed by a proper selection of the func-
tionality of polymer chain. To conrm this expectation, we have pre-
pared and characterized novel polymer-in-salt electrolytes based on
poly(acrylamide). Turns out the polyacrylamide is very efcient in the
complexation of lithium salts. FT-IR and Maldi-ToF Mass Spectrometry
conrmed formation of salt-polymer associates and possibility of fast
ionic transport via exchange of lithium between neighboring associates
[207]. The interesting concept of polymer-in-salt systems has not been
further explored most probably due to unsatisfactory mechanical prop-
erties of resulting materials.
Condensed systems are also present in non polymer based electro-
lytes. Very recently a new class of Solvent-in-Salt and Ionic liquid-
in-Salt was reported. They can be used not only in lithium-ion but
also in lithium sulphur batteries.[208210]
4.4. Hybrid electrolytes
Hybrid organicinorganic materials can be dened as nanocom-
posites derived from organic and inorganic components of size
ranging from a few to several tens of nanometers [211]. Develop-
ment in the eld of hybrid organicinorganic materials is associated
with the research in the eld of inorganic chemistry, especially with
the reactions carried out under mild conditions, particular regard to
the solgel method, which enables the design of systems of various
combinations of domains of inorganic and organic compounds
[212]. Hybrids have a high potential for application due to such fea-
tures as light weight (hence high energy density for the nal cell),
compared with other classes of electrolytes. Solgel method allows
to obtain a fully homogeneous systems containing large amounts
of additives and having a uniform structure. [213] Popall and
Durand obtained organicinorganic copolymers based on inorganic
core (containing silicon and zirconium atoms) associated with the
organic matrix consisting of an ether moiety ( OCH2CH2-)n [214].
The organic part of the hybrid material is responsible for the disso-
ciation of lithium salts in this case lithium perchlorate. Described
electrolyte was fully amorphous and showed a conductivity at
room temperature of 10 5 S cm 1 . Subsequent work of Popall
and co-workers have resulted in the entire group of compounds
119
called ORMOCER mechanically and electrochemically stable electro-
lyte having a conductivity of 103 S cm- 1 at room temperature. [215]
Another group of electrolytes based on hybrid materials (based on TEOS,
PEG, PPG and lithium perchlorate) were obtained by Judeinstain and co-
workers [216,217]. In case of Archer and coworkers using of inorganic
core (SiO2) and organic chains enable to obtain nanoporous materials
characterized by high ionic conductivity and wide electrochemical win-
dow (up to 6 V) [218,219].
Another approach to the preparation of electrolytes based on hybrid
materials was presented by Di Noto et al. In this case, instead of the sol
-gel technique, direct reactions between organometallic precursors and
PEG having an average molecular weight of 400 g mol 1 was per-
formed [220,221].
In most cases the inorganic precursor is based on the silicon com-
pounds but also boron and aluminum compounds nd the application
in described systems. Alkoxy aluminum compounds are used in the
synthesis of electrolytes by solgel method. There is also the group of
electrolytes with the aluminum compounds which are the main build-
ing block of the inorganic portion. These systems are of interest because
of the possible link between different structures and a strong Lewis
acidity centers which leads to the possibility of the appearance of
the anion and aluminum atom strong interactions. The materials used
as precursors are aluminum compounds such as LiAlH4, alkoxy com-
pounds of aluminum, triethyl aluminum or aluminum carboxylates
[222226].
In the area of our interests was the characterization of a new group
of lithium PEs based on the hybrid organicinorganic matrix. The mate-
rial is obtained by the reaction between the organometallic precursor
methylalumoxane (MAO) and olygooxyethylene molecules. The
aluminum-oxide core of MAO molecule is built from three four-
member Al2O2 rings and two six-member Al3O3 rings. This molecule
contains in its outer shell twelve methyl groups bonded to aluminum
atoms [227].
The studied systems based on the MAO organometallic precursor
can be divided into three main groups [227,228]:
rst one, the product of the reaction of monocapped olygoether (of
various Mw being in range 350 to 1000 g mol1) with MAO was
used as the only constituent of the polymeric matrix
second group was obtained in a reaction of a branched system with
PEG molecules containing two OH moieties in this situation PEG
acted as a crosslinking agent.
third case a polymer blend was obtained by mixing PEGDME
(Mw = 500 g mol 1 ) with the product of a reaction of MAO
with PEGME (Mw = 350 g mol 1 In all cases three lithium
salts (LiClO 4 , LiCF 3SO 3 LiTf and LiN(SO2 CF3 )2 LiTFSI) were
used.Authors reported formation of a new type of hybrid organ-
icinorganic composite polymeric electrolyte. The novelty is
based on the fact that the organometallic precursor used formed
a nanosized Al-O cage itself. Thus, in consequence, we created
an intermediate between typical nano-composites with mechani-
cally dispersed ller grains and materials obtained previously by
the solgel type of reaction. This system was characterized by elec-
trochemical, thermal and mechanical properties which can be easily
tailored by an appropriate choice of branching and cross-linking
olygoglycols, molecular weight and molar ratio between nano-
cores and both reactants [229]. FTIR spectroscopy revealed en-
hanced salt dissociation in the electrolyte containing MAO and it
can be further correlated with improved cationic transference num-
bers measured for electrolytes containing LiTF and LiTFSI. The ther-
mal stability of the system was increased by over 40 C compared
to olygooxyethylene electrolytes (from 256 C up to 308 C). The
TG FTIR studies of the decomposition products of electrolytes con-
taining MAO lead to the conclusion that MAO additives prevent the
release of the extremely toxic gaseous product hydrouoric acid. It
is possible that MAO participates in the bonding of the gaseous
120 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
product. Hence the operational safety of the electrolyte in terms of
thermal decomposition is increased upon MAO addition.
The new concepts of electrolytes development are pursued via re-
search of synergistic properties of different kind of materials. One of
the good example are proposed by Pfaffenhuber [230] and references
therein] so called soggy-sand electrolytes (solidliquid composites, typ-
ically gel electrolytes, with enhanced electrical properties). The soggy-
sand electrolytes combine high Li conductivity, and good wettability
of electrode particles with the mechanical stability of semi-solids.
Common problems of all composities problems lay in the reproducibil-
ity and stability of the composites microstructure and morphology
which, however, can be optimized and manipulated. The technological
relevance cannot be underestimated.
4.5. New tailored anions
After almost 20 years of commercial dominance of lithium-ion cells
there is still a lack of potential substitutes for LiPF6, which is denitely
not awless. Among the most important drawbacks of the LiPF6 use in
electrolyte system is the formation of HF in the cell, which destroys
the cell from inside after certain time, but also LiPF6 and HF toxicity,
problematic and expensive synthesis of LiPF6 as well as its sensitivity
to moisture.
The constant seek for cheap, environment-friendly and easy to han-
dle materials created a gap to ll. List of parameters that lithium salts
have to fulll in order to become new predominant salt on market of
lithium-ion cells is not very long, but no existing salt satises them all.
Transference number above 0.5 (or at least better than LiPF6 that in
optimized carbonate solvent mixtures has transference number of
0.3-0.4), conductivity higher than 1 mS cm-1 (10-3 S cm1), no de-
composition in range of 0-5 V vs. Li and no aluminum corrosion in this
range, low price (at least lower than LiPF6, but the lower, the better),
non-toxicity, stability towards moisture and oxygen (easiness of han-
dling), thermal stability up to at least 100 C and low association rate
(lower than LiPF6 or very weakly associating LiClO4) is what is necessary
for a commercial success.
A promising concept for designing new anions is based on so called
Hckel anions presented in Fig. 3. The name came from the transposi-
tion of the Hckel rule predicting the stability of the aromatic systems.
One of the most common examples of this type of anions is 4,5-
dicyano-triazole (DCTA) (Fig. 4). This particular structure is completely
covalently bonded and shows very stable 6 (or 10 electron if CN
bonds are involved in calculations) conguration. It can be produced
from commercially available precursors and even more importantly
does not comprise uorine atoms. Li salts with this type of anion were
found to exhibit high (~300 C) thermal stability. LiDCTA was success-
fully tested in PEO matrices systems providing a promising, improved
electrolyte for lithium-ion batteries [231]. Unfortunately DCTA failed
as a component of the EC/DMC (1:1) battery electrolyte.
Basing on LiDCTA example, novel imidazole derivatives salts were
synthesized, using the procedure shown in Fig. 3. Molecular modelling
(diaminomaleonitrile) trifluoroacetic anhydride
Fig. 3. Synthesis scheme for LiTDI.
studies showed that benzimidazolide and imidazolide anions show a
typical behaviour of heterocyclic anion alternatives to PF 6 : if tailored
correctly they offer more dissociative lithium ion pairs compared to
LiPF 6 , but are not as electrochemically stable. The latter is, however,
not crucial as long as the anions exceed the stability window for the
intended application, for Li-ion batteries ~ 4.2 V [245].
The data obtained for the benzimidazolides, indicate that with very
small alterations, such as the positioning of CN substituents on the het-
erocycle, ion pairing can be changed drastically. With equal predicted
properties for analogous imidazolides and benzimidazolides, the differ-
ence in anion size is an important variable for choosing an appropriate
lithium salt.
Overall the potential of using cyano chemistry to create new lithium
salts with excellent electrochemical stabilities and ion pairing proper-
ties is evident. The previously recognized improvement of both the
electrochemical stability and the lithium ion pair dissociation ability,
by increasing the number of cyano groups, is apparent also for the
benzimidazole salts and further strengthened by the predictions made
for "P(CN)6. However, an increase in the number of CN groups leads
to a larger anion size and require increased synthesis efforts. The
tetracyanobenzimidazolides are the most promising candidates of the
explored families of anions.
Imidazole salts exhibit uncommon conductivity vs. concentration
curve in solvent mixtures, due to complicated solvation behaviour
[232]. The positive turnout is that a broad maximum of conductivity
for electrolytes based on LiTDI starts as low as 0.30.4 mol kg1.
When conductivity of more than 5 mS cm 1 is combined with high
transference number (higher than 0.5) at low concentrations, lithium
cation conductivity of LiTDI matches that of LiPF6, and so does it
match its performance using much lower concentrations, hence at
lower price.
Another approach to maximize lithium cation transference number
by weakening lithium-anion interactions was made by using multiva-
lent salts. Li3BTI, which is an trimer analogue of LiTDI, was synthesized
for this purpose [233]. Although not as high conducting as its monomer,
its transference number in plain electrolyte (no additives) reaches un-
paralleled value of 0.8 (Fig. 5). The salt is stable in ambient atmosphere,
its ionic conductivity is higher than 1 mS cm1, it is stable up to 160 C,
and 4.5 V vs. Li/Li+. Hence, new multivalent anion fullls general
requirements for battery electrolyte, providing extraordinary transfer-
ence number as well.
Another approach is to use heteroaromatic systems in which all the
ring-forming atoms in the anion are nitrogen or carbon and all the car-
bon atoms are substituted with cyano groups see Fig. 6. This results in
good distribution of the negative charge within the ion, furthermore
geometries of such systems are interesting and they avoid chelating
effect i.e. interaction between lithium cation and anion by more than
one functional group. Another good reason for seeking new salts is
their potentially low impact on the environment. The problem of
recycling and neutralizing lithium batteries is growing every day as
their sales increase. Most of the newer salts do not create toxic fumes
in the presence of water. The other important advantage of the systems
(lithium 4,5-dicyano-2-(trifluoromethyl)imidazolate)
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
Fig. 4. LiDCTA Hckel type salt: LiDCTA lithium 4,5-dicyano-1,2,3-triazole.
studied is that they do not contain uorine. This makes the salt degrada-
tion products less toxic and less bioacumulative. It is worth to note that
the most popular LiPF6-EC/DMC electrolyte, via reaction between sol-
vent molecules and "PF6 entity (catalyzed by cathode material), can
give rise to uoroethanol (CH2FCH2OH) and uoroacetic acetic acid
(CH2FCOOH). Both are highly toxic and bioaccumulative.
We would also like to stress out, that although new anions look more
complicated than PF 6 , they are much easier and cheaper to synthesize
(e.g. for lithium tricyanoimidazolate LiTIM, the substrates are: easy to
synthesize oligomer of HCN and cyanogen).
4.6. Sodium electrolytes
Currently lithium-ion batteries are the most commonly used chem-
ical power source in applications ranging from small portable electron-
ics to electric cars. The global market for lithium-ion batteries is
estimated at tens of billions of dollars a year and growing. The preva-
lence of use is due to many advantages, such as high energy density,
high single cell voltage and low self-discharge. On the other hand low
global lithium resources (currently estimated at 29 million tons [234])
pose a real limit for the application of Li-ion chemistry in large-scale
batteries for electric cars or grid storage of energy from renewable
sources. In our opinion this will accelerate the search for new types of
batteries which are expected not only to deliver high power and energy
density, and safety but also use readily available raw materials, allow for
easy disposal/recycling and low production cost.
A promising and natural alternative to lithium-ion seems to be
sodium-ion batteries. The relatively low molecular weight of sodium
(23 g mol1) and very low standard potential of the Na/Na + couple
(2.71 V vs. SHE), combined with low price and availability (it is the
sixth most abundant element on Earth) makes it an ideal alternative
0.8
0.7
0.6
t+
0.5
0.4 Li3BTI-PC
Li3BTI-EC:DMC
0.3
0.01 0.1
c / mol dm-3
Fig. 5. Li3BTI scheme and transference numbers of its solutions in propylene carbonate and
mixture of ethylene carbonate and dimethyl carbonate (1:1 ratio).
121
to lithium. Another advantage is the possibility to use aluminum current
collectors for both electrodes in sodium ion batteries (Na does not
alloy with Al) making them cheaper and lighter. In addition sodium
metal (rather than a sodium intercalation compound) can be used as a
negative electrode because it is softer than lithium, hence it is easier
to prevent dendritic sodium plating [235]. This is advantageous because
it provides high capacity (1190 mAh g1) and allows for higher cell
voltage (for example LiC6 decreases the useful cell voltage by about
150 mV compared to Li metal).
The rst successful attempt to use sodium in batteries took place
after the discovery of ionic conductivity in alumina (Na2O11Al2O3)
by scientists at the Ford Motor Company in 1967. Charge carriers were
determined to be sodium cations and the ionic conductivity was
approximately 10 S cm 1 at 300 C. Being an electronic insulator
alumina proved to be a promising ceramic solid electrolyte. This
has led to the construction of high temperature sodium-sulfur batteries:
negative electrode was liquid sodium and the positive was liquid sulfur
(mixed with carbonaceous material providing electronic conductivity
throughout the electrode) and both liquid electrodes were separated
the solid ceramic electrolyte made of alumina. Na-S batteries operate
at 300-400 C and have a voltage around 2.1 V (in the nal stage of
discharge dropping to 1.78 V). Despite good reversibility the technology
took up only in specic applications due to the high operating temper-
ature (and related safety measures), corrosion problems, and the prob-
lem of overloading (fully charged cells have extremely high internal
resistance) [236]. In parallel to Na-S battery an alternative chemistry
was developed: Na-NiCl2. The battery shared the liquid Na and solid
alumina separator, but the liquid sulphur was replaced by solid nick-
el chloride. In addition, an electrolyte intermediate NaAlCl4 was used in
order to improve the contact between the solid positive electrode and
solid electrolyte. A few advantages over Na-S are: lower operating
temperature (250-350 C), higher standard potential of the reaction
(2.58 V), and a new battery does not contain metallic sodium (safer
and easier to handle and make). It is produced at the negative electrode
during the rst charge. Corrosion is also much less of a problem [237].
Na-NaSICON-S built in 2000 exhibits a much higher energy density,
unfortunately the problem is still the high operating temperature that
makes them much less competitive in many applications.
LiTCP LiTIM LiTRIP
LiDCIT LiDCTA LiCP
1 LiPIT LiTCPI
Fig. 6. Examples of structures of heteroaromatic systems in which all the atoms in the
heteroaromatic anion are nitrogen or carbon and all the carbon atoms are substituted
with cyano group.
122 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126

Table 4
Ionic conductivities and transference numbers for chosen polymer electrolytes.

Type of matrix / temperature S cm1 t+/temperature

105 /100C 0.45 /90C 1LiTF:9(EO)

108 /room 0.2 /room [251]
1LiClO4:20(EO)

3.5 104 /room

1LiTF:20(EO) [252]

3.2 104 /room

1.9 103 /room
1LiClO4:16(EO) [253]

105 /room 0.2 /room

Cross-linked PEGDME-PEGDA (poly(ethylene glycol) 2.1 104/room 0.19/room [64]

diethyletherpoly(ethylene glycol) diacrylathe 1LiTFSI:10(EO)
1.2 104/room 0.06/50C
1LiBOB:32(EO) 1LiTFSI:44(EO)
0.07/50C
1LiBOB:32(EO) [254]

2 106/room Theoretically close to one [255]

103105 0.5/room
2.7 105/room (i.e. based on AlSi domens and bonded high dependence on the composition [256]
with PEGs Li TF (quantity unspecied)
high dependence on the composition

Composite polymer electrolyte 103105 0.20.6

high dependence on the composition high dependence on the composition [257, 258]

Ionic liquid P(EO)10LiTFSIPYR14TFSI 1.1 104/room 0.020.4/room for gel electrolytes based on organic
carbonates and PVDF-HFP [259,260,261]
5
Po 1.93 10 /50C 0.45/30C [262]

Gel electrolyte -PC: EC: PAN 103/room Higher than 0.5/ room [259,263]
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
Currently a lot of work is done on room temperature Na and Na-ion
cells. Here the main challenge lies in the development of electrode ma-
terials that would easily and quickly intercalate Na+ ions (that are larg-
er than Li+) hence the liquid electrolytes can be often adapted from Li-
ion technology by replacing Li salts with respective Na salts.
4.7. Impact of modeling
Various families or organic salts can be appled in lithium and
lithium-ion batteries. Molecular modelling can be very helpful in de-
signing new organic salts and reduce expensive and time consuming
synthetic work.
Two examples based on this concept were introduced in
2003, with the dicyanotriazolate C 5 N 4 (DCTA) [232,238] and
the bis(triuoroborane)imidazolide C 3 N 2 (BF 3 ) 2 (Id) [239]. The
bis(triuoroborane)imidazolide was dedicated for Li-ion batteries
and exhibited the highest ionic conductivity compared to the similar
idea. [240]. Tested in a fullcell, the lithium derivative of Id salt showed
results comparable to LiPF6. The benets included:temperature conduc-
tivity, solubility, and electrochemical stability reaching 4.8 V vs. Li/Li+),
cost and ease synthesis [239]. The system is not ideal since B-F bond
presence and potential risk of LiF and HF formation [241] plus the con-
ductivity drop resulting from BF 4 creation from the anion dispropor-
tionation [242]. Among the disadvantages Many of strategic
indications for experimental work came from theretical predictions
for imidazoles synthesis. [243]. Although many of these salts beats
DCTA, theory faced the practical synthesis problems and experimental
work was concentrated in heterocycled structures [244].
However based on the theoretical predictions ecently two anions
were synthetized in Warsaw group: 4,5-dicyano-2-triuoromethyl
imidazole (TDI) and its 2-pentauoroethyl analogue (PDI) as tailor
made anions and consequently tested as PEG based model electrolytes
[245247]. The rst report of these structures appeared nine years ago
[248], when it was prepared with good yield in connection with the
nding of a new, improved synthesis route for the protonated (un-
charged) form of TDI, known since the mid 70s [249].
The Chalmers group [250] used LiTDI and LiPDI as templates and
made a map for further synthesis candidates by ab initio methods.
New anions are proposed by rst extending the heterocyclic imidazole
ring to a benzimidazole ring, with two or four cyano groups symmetri-
cally positioned on the ring, and secondly by also looking at alternatives
where the uoroalkyl substituent is replaced by a smaller, less exible
group. However cyano substituted benzimidazoles even most prom-
ising are more difcult to synthesize than their imidazole analogues.
Successful nishing of the synthetic procedure will give us more in-
formation about the reaction yield and possibilities of scaling it up.
5. Summary
In the preceding sections the summary of the status quo of electro-
lytes used in lithium and lithium-ion battery technology has been
given for all three types of electrolytes utilizing organic liquid solvents,
polymers and ionic liquids as solvents. Currently most of lithium-ion
batteries utilize a solution of LiPF6 in carbonate mixture, either by itself
or entrapped in polymer matrices (PVdF, PAN) [204] forming gel-type
electrolytes. Environmental and safety concerns apart from technologi-
cal limitations are major drawbacks of liquid electrolyte technology. All-
solid-state lithium batteries remain one of the major goals for research
in the eld because they offer a step-change in safety and energy densi-
ty. Solid electrolytes are the necessary pre-requisite and true solid PEs
(solvent free systems) are one of the most attractive options. PEs,
including gels, solid amorphous and crystalline, along with some ILs
open the doors to using Li metal as an anode which would further in-
crease the energy density of batteries. The currently used liquid and
gel electrolytes exclude us of Li metal because of dendrite formation
on charging that lead to battery failure and compromise safety. PEs
123
and some ILs prevent dendrites formation and can be used as electro-
lytes on their own or as polymer/IL mixtures.
In the former sections we have tried to highlight various ways of
liquid and polymeric systems modication. For liquids that is mostly a
search for new solvents or solvent salt combinations. There have not
been any major breakthroughs in new solvents. The same conclusions
can be drawn when considering application of various types additives
used to enhance safety of batteries regulate the formation and stabiliza-
tion of SEI or modifying ionic transport. Very often the progress in one
area is associated with negative effect in another. On the contrary
searching for the new salt, which in fact seems to be a common issue
for all three types of solvent discussed in this report, seems to be suc-
cessful and offer even more possibilities in the future given the variety
of the salt synthesis strategies. Ideally the salt of choice should be inex-
pensive, easy to prepare and environmentally friendly. It should also ful-
l the electrochemical requirements.
In the preceding section example of performance of LiTDI and LiPDI
in all solid as well as in liquid electrolyte based batteries were shown.
Although, the preliminary results are promising further studies in vari-
ous electrode congurations need to be performed. It should be empha-
sized that the use of organic salts will extend the range of possible
solvent combinations which can lead to the improvement of safety as
well as reduction of environmental impact. It is also worth to mention
that organic salts can be used as components in the synthesis of ILs. In
such a case no additional lithium salt will be required as a component
of electrolyte. So far addition of the salt reduces the conductivity range
of IL-based electrolytes. As was shown before addition of TDI to PEO-
based PEs enhances the ambient temperature conductivity of these
systems. However, as it can be concluded on the basis of the review of
properties of modied PEs, it is hardly to achieve the level of conductiv-
ity exceeding 104 S/cm at ambient temperatures in truly solid electro-
lytes. Moreover, what also should be emphasized is the fact that most of
the method of electrolyte modication were developed in late 1980's or
early 1990's. Since that time small modications of particular proce-
dures are described in literature but it is difcult to nd a major break-
through in this area. It is quite common to see papers repeating data
published before the internet era, especially among American labs and
Universities. Currently the use of polysiloxane based electrolytes
seems to undergo a renaissance (papers 1O-04 and 2O-05 presented
at PBFC Conference held in 2009 in Yokohama). More detailed report
are however needed to see whether the performances obtained are
in fact much better than data presented by Japanese groups in the
1980's. An interesting concept of the use of crystalline electrolytes de-
veloped by Prof. Bruce group and described briey above is still in the
phase of laboratory tests.
Table 4 summarizes the properties of each class of electrolyte
discussed. In spite of the increase of the safety of electrolytes utilizing
ILs in terms of less ammability compared with liquid electrolytes re-
cent report highlights quite extensive thermal effect in the batteries uti-
lizing ILs (paper 2O-02 presented at PBFC 2009).
It should be quoted that for some PEs or polymer type ILs synthe-
sized in the form of polyelectrolytes or utilizing anion trapping additives
higher lithium transference numbers are reported usually connected
with considerable reduction in ambient temperature ionic conductivity.
To summarize the choice of particular electrolyte for lithium or
lithium-ion battery would depend on the application of the devices. It
will be difcult to quote the winner among the contenders however ap-
proach of introduction of new salts useful for all type of solvents might
benet the entire industry.
Acknowledgments
European Research Institute Alistore since this paper is a result of the
ERI Alistore white paper report preparation. Special thanks for profs.
M. Armand, P. Bruce and B. Scrosati.
124 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
References
[1] M.S. Whittingham, Science 192 (1976) 1126.
[2] M.S. Whittingham, Prog. Solid State Chem. 12 (1978) 1.
[3] M.B. Armand, Fast Ion Transp.Solids, Solid State Batteries Devices, Proc.NATO
Adv.Study Inst.1973. 665673.
[4] J.M. Tarascon, M. Armand, Nature 414 (2001) 359367.
[5] K. Xu, Chem. Rev. 104 (2004) 43034418.
[6] F.M. Gray, Polymer Electrolytes, RSC Materials Monographs1997.
[7] M. Armand, F. Endres, D.R. MacFarlane, H. Ohno, B. Scrosati, Nat. Mater. 8 (2009)
621629.
[8] D. Aurbach, O. Chusid, Ref. Mod in Chemistry, Encyclopedia fo Electrochemical
Power SourcesMol. Sci. and Chem Engeneering2009. 92111.
[9] G. Li, Z. Li, P. Zhang, H. Zhang, Y. Wu, Pure Appl. Chem. 11 (2008) 2553.
[10] M. Amereller, T. Schedlbauer, D. Moosbauer, C. Schreiner, C. Stock, F. Wudy, S.
Zugmann, H. Hammer, A. Maurer, R. Gschwind, H.D. Wiemhfer, M. Winter, H.J.
Gores, Progr. Sol. State Chem. 42 (2014) 3956.
[11] S. Passerini, W.A. Henderson, Encyclopedia of Electrochemical Power Sources,
2009. 8591.
[12] R. Jasinski, S. Carroll, J. Electrochem. Soc. 117 (1970) 218.
[13] J. Barthel, M. Wuhr, R. Buestrich, H.J. Gores, J. Electrochem. Soc. 142 (1995) 2527.
[14] J. Barthel, R. Buestrich, E. Carl, H.J. Gores, J. Electrochem. Soc. 143 (1996) 3572.
[15] J. Barthel, R. Buestrich, E. Carl, H.J. Gores, J. Electrochem. Soc. 143 (1996) 3565.
[16] J. Barthel, R. Buestrich, H.J. Gores, M. Schmidt, M. Wuhr, J. Electrochem. Soc. 144
(1997) 3866.
[17] J. Barthel, M. Schmidt, H.J. Gores, J. Electrochem. Soc. 147 (2000) 21.
[18] W. Xu, C.A. Angell, Electrochem. Solid-State Lett. 4 (2001) E1.
[19] K. Xu, S. Zhang, T.R. Jow, W. Xu, C.A. Angell, Electrochem. Solid-State Lett. 4 (2002)
A26.
[20] K. Xu, S. Zhang, B.A. Poese, T.R. Jow, Electrochem. Solid-State Lett. 5 (2002) A259.
[21] W. Xu, C.A. Angell, Electrochem. Solid-State Lett. 3 (2000) 366.
[22] W. Xu, C.A. Angell, Solid State Ionics 147 (2002) 295.
[23] M. Videa, W. Xu, B. Geil, R. Marzke, C.A. Angell, J. Electrochem. Soc. 148 (2001) A1352.
[24] W. Xu, A. Shusterman, M. Videa, V. Velikov, R.L. Marzke, C.A. Angell, J. Electrochem.
Soc. 150 (2003) E74.
[25] J.L. Allen, S.-D. Han, P.D. Boyle, W.A. Henderson, J. Power Sources 196 (22) (2011)
97379742 (and references therein).
[26] H. Yamaguchi, H. Takahashi, M. Kato, J. Arai, J. Electrochem. Soc. 150 (2003) A312.
[27] J.S. Syzdek, in: C.A.C. Sequeira, D. Santos (Eds.), Polymer Electrolytes: Fundamen-
tals and applications, Woodhead Publishing Ltd, Oxford, 2010.
[28] Z. Gadjourova, Y.G. Andreev, D.P. Tunstall, P.G. Bruce, Nature 412 (2001) 520523.
[29] B. Scrosati, R. Neat, Applications of Electroactive Polymers, 1993.
[30] C. Berthier, W. Gorecki, M. Minier, M.B. Armand, J.M. Chabagno, P. Rigaud, Solid
State Ionics 11 (1983) 9195.
[31] M.H. Cohen, D. Turnbull, J. Chem. Phys. 31 (1959) 11641169.
[32] G. Adam, J.H. Gibbs, J. Chem. Phys. 43 (1965) 139146.
[33] J.H. Gibbs, E.A. DiMarzio, J. Chem. Phys. 28 (1958) 373383.
[34] S.D. Druger, M.A. Ratner, A. Nitzan, Solid State Ionics 910 (1983) 11151120.
[35] S.D. Druger, A. Nitzan, M.A. Ratner, J. Chem. Phys. 79 (1983) 31333142.
[36] S.D. Druger, M.A. Ratner, A. Nitzan, Phys. Rev. B 31 (1985) 39393947.
[37] S.D. Druger, M.A. Ratner, A. Nitzan, Solid State Ionics 1819 (1986) 106111.
[38] M.A. Ratner, in: J.R. MacCallum, C.A. Vincent (Eds.), Polymer Electrolyte Reviews -
1, Elsevier, London, 1987, pp. 173236.
[39] H. Cheradame, J.F. Le Nest, in: J.R. MacCallum, C.A. Vincent (Eds.), Polymer Electro-
lyte Reviews-1, Elsevier, London, 1987 (Chapter 5).
[40] A. Bouridah, F. Dalard, D. Deroo, H. Cheradame, J.F. Le Nest, Solid State Ionics 15
(1985) 233240.
[41] H. Cheradame, A. Killis, L. Lestel, S. Boileau, Polym. Prepr. 30 (1989) 133.
[42] M. Watanabe, N. Ogata, in: J.R. MacCallum, C.A. Vincent (Eds.), Polymer Electrolyte
Reviews - 1, Elsevier, London, 1987, pp. 3968.
[43] M. Watanabe, S. Oohashi, K. Sanui, N. Ogata, T. Kobayashi, Z. Ohtaki, Macromole-
cules 18 (1985) 19451950.
[44] M. Watanabe, K. Nagaoka, M. Kanba, I. Shinohara, Polym. J. 14 (1982) 877.
[45] F.M. Gray, J.R. MacCallum, C.A. Vincent, J.R.M. Giles, Macromolecules 21 (1988)
392397.
[46] J.R.M. Giles, F.M. Gray, J.R. MacCallum, C.A. Vincent, Polymer 28 (1987)
19771981.
[47] J.F. Le Nest, S. Callens, A. Gandini, M. Armand, Electrochim. Acta 37 (1992)
15851588.
[48] J.R.S. Giles, M.R. Greenhall, Polym. Commun. 27 (1986) 360.
[49] J.R.M. Giles, Solid State Ionics 24 (1987) 155167.
[50] L. Lestel, S. Boileau, H. Cheradame, Polym. Prepr. 30 (1989) 420.
[51] S.-. Fang, P. Zhang, Y.-. Jiang, Polym. Bull. 19 (1988) 8187.
[52] G.E. Wnek, K. Gault, J. Serpico, C.Y. Yang, G. Venugopal, S. Krause, in: B. Scrosati
(Ed.)Proc. II International Symposium on Polymer Electrolytes (ISPE-2), 1990,
p. 73 (Siena June 1989).
[53] Y. Uchimoto, Z. Ogumi, Z. Takehara, Solid State Ionics 35 (1989) 417423.
[54] Z. Ogumi, Y. Uchimoto, Z. Takehara, F.R. Foulkes, J. Electrochem. Soc. 137 (1990)
2934.
[55] Y. Uchimoto, Z. Ogumi, Z. Takehara, Solid State Ionics 4041 (1990) 624627.
[56] Y. Uchimoto, Z. Ogumi, Z. Takehara, F.R. Foulkes, J. Electrochem. Soc. 137 (1990)
3540.
[57] S.A. Mullin, A.A. Teran, R. Yuan, N.P. Balsara, J. Polym. Sci. B Polym. Phys. 51 (12)
(2013) 927934.
[58] W.S. Young, T.H. Epps, Macromolecules 45 (11) (2012) 46894697.
[59] A.A. Teran, N.P. Balsara, J. Phys. Chem. B 118 (1) (2014) 417.
[60] T.J.L. helen, A.A. Teran, X. Wang, Z.-G. Wang, N.P. Balsara, Macromolecules 47 (8)
(2014) 26662673.
[61] F.M. Gray, in: J.R. MacCallum, C.A. Vincent (Eds.), Polymer Electrolyte Reviews - 1,
Elsevier, London, 1987 (Chapter 6).
[62] J.R. Maccallum, M.J. Smith, C.A. Vincent, Solid State Ionics 11 (1984) 307312.
[63] E. Kroni, K.V. Lovell, A. Hooper, R.J. Neat, Brit. Polym. J. 20 (1988) 275280.
[64] M. Ueno, N. Imanishi, K. Hanai, T. Kobayashi, A. Hirano, O. Yamamoto, Y. Takeda, J.
Power Sources 196 (2011) 47564761.
[65] J.R. Craven, R.H. Mobbs, C. Booth, J.R.M. Giles, Makromol. Chem. Rapid Commun. 7
(1986) 8184.
[66] C.V. Nicholas, D.J. Wilson, C. Booth, J.R.M. Giles, Brit. Polym. J. 20 (1988) 289292.
[67] G. Goulart, S. Sylla, J.Y. Sanchez, M. Armand, Int. Symp. Polym. Electrolyte 2nd
(1990) 99105.
[68] Z. Florjanczyk, W. Krawiec, W. Wieczorek, J. Przyluski, Angew. Makromol. Chem.
187 (1991) 1932.
[69] K. Such, Z. Florianczyk, W. Wieczorek, J. Przyluski, Int. Symp. Polym. Electrolyte
2nd (1990) 119127.
[70] J. Przyluski, W. Wieczorek, Solid State Ionics 5356 (1992) 10711076.
[71] Z. Florjanczyk, W. Krawiec, W. Wieczorek, M. Siekierski, J. Polym. Sci. B 33 (1995)
629636.
[72] J. Przyuski, W. Wieczorek, Solid State Ionics 5356 (1992) 10711076.
[73] D.G.H. Ballard, P. Cheshire, T.S. Mann, J.E. Przeworski, Macromolecules 23 (1990)
12561264.
[74] Z. Florianczyk, E. Zygadlo-Monikowska, K. Such, W. Wieczorek, Extended Abstract,
Proc. IIIrd Int. Symp. Polym. Electrolytes1991. 4344 (Annecy).
[75] Z. Florjanczyk, E. Zygadlo-Monikowska, D. Raducha, K. Such, W. Wieczorek,
Electrochim. Acta 37 (1992) 15551558.
[76] D.W. Xia, D. Soltz, J. Smid, Solid State Ionics 14 (1984) 221224.
[77] D.W. Xia, J. Smid, J. Polym. Sci. Polym. Lett. Ed. 22 (1984) 617621.
[78] N. Kobayashi, M. Uchiyama, K. Shigehara, E. Tsuchida, J. Phys. Chem. 89 (1985)
987991.
[79] N. Kobayashi, M. Uchiyama, E. Tsuchida, Solid State Ionics 17 (1985) 307311.
[80] J.M.G. Cowie, in: J.R. MacCallum, C.A. Vincent (Eds.), Polymer Electrolyte Reviews -
1, Elsevier, London, 1987 (Chapter 4).
[81] Y. Yamaguchi, S. Aoki, M. Watanabe, K. Sanui, N. Ogata, Solid State Ionics 4041
(1990) 628631.
[82] Y. Ikeda, Y. Wada, Y. Matoba, S. Murakami, S. Kohjiya, Electrochim. Acta 45 (2000)
11671174.
[83] Y. Ikeda, Y. Wada, Y. Matoba, S. Murakami, S. Kohjiya, Rubber Chem. Technol. 73
(4) (2000) 720730.
[84] K. Nagaoka, H. Naruse, I. Shinohara, M. Watanabe, J. Polym. Sci. Polym. Lett. Ed. 22
(1984) 659663.
[85] P.G. Hall, G.R. Davies, J.E. McIntyre, I.M. Ward, D.J. Bannister, K.M.F. Le Brocq,
Polym. Commun. 27 (1986) 98.
[86] D. Fish, I.M. Khan, J. Smid, Makromol. Chem. Rapid Commun. 7 (1986) 115120.
[87] R. Spindler, D.F. Shriver, J. Am. Chem. Soc. 110 (1988) 30363043.
[88] R. Spindler, D.F. Shriver, J. Am. Chem. Soc. 110 (1988) 6599-6599;
I.M. Khan, Y. Yuan, D. Fish, E. Wu, J. Smid, Macromolecules 21 (1988) 26842689.
[89] D. Fish, I.M. Khan, E. Wu, J. Smid, Brit. Polym. J. 20 (1988) 281288;
T. Yeh, H. Liu, Y. Okamoto, H.S. Lee, T.A. Skotheim, in: B. Scrosati (Ed.)Proc. II Int.
Symp. Polym. Electrolytes (ISPE-2), 1990, p. 83 (Siena June 1989).
[90] P.M. Blonsky, D.F. Shriver, P. Austin, H.R. Allcock, J. Am. Chem. Soc. 106 (1984)
68546855.
[91] P.M. Blonsky, D.F. Shriver, P. Austin, H.R. Allcock, Solid State Ionics 1819 (1986)
258264.
[92] R.A. Reed, T.T. Wooster, R.W. Murray, D.R. Yaniv, J.S. Tonge, D.F. Shriver, J.
Electrochem. Soc. 136 (1989) 25652570.
[93] W. Gorecki, P. Donoso, C. Berthier, M. Mali, J. Roos, D. Brinkmann, M.B. Armand,
Solid State Ionics 2830 (1988) 10181022.
[94] H. Liu, Y. Okamoto, T. Skotheim, Y.S. Pak, S.G. Greenbaum, K.J. Adamic, Solid State
Ionics (1989) 337.
[95] S. Ganapathiappan, K. Chen, D.F. Shriver, Macromolecules 21 (1988) 22992301.
[96] H.R. Allcock, D.T. Welna, A.E. Maher, Solid State Ionics 177 (2006) 741747.
[97] N. Kobayashi, T. Hamada, H. Ohno, E. Tsuchida, Polym. J. 18 (1986) 661666.
[98] E. Tsuchida, N. Kobayashi, H. Ohno, Macromolecules 21 (1988) 96.
[99] H.T. Kim, J.K. Park, Solid State Ionics 98 (1997) 237244.
[100] S. Toyota, T. Nogami, H. Mikawa, Solid State Ionics 13 (1984) 243247.
[101] H. Ohno, N. Kobayashi, S. Takeoka, H. Ishizaka, E. Tsuchida, Solid State Ionics 4041
(1990) 655658.
[102] D. Benrabah, S. Sylla, F. Alloin, J.-Y. Sanchez, M. Armand, Electrochim. Acta 40
(1314) (1995) 22592264.
[103] Z. Floriaczyk, W. Krawiec, T. Listo, W. Wieczorek, J. Przyuski, Synth. Met. 35
(1990) 249251.
[104] I.E. Kelly, J.R. Owen, B.C.H. Steele, J. Power Sources 14 (1985) 1321.
[105] I. Kelly, J.R. Owen, B.C.H. Steele, J. Electroanal. Chem. 168 (1984) 467478.
[106] E. Tsuchida, H. Ohno, K. Tsunemi, N. Kobayashi, Solid State Ionics 11 (1983)
227233.
[107] C. Wang, Q. Liu, Q. Cao, Q. Meng, L. Yang, Solid State Ionics 5356 (1992)
11061110.
[108] L. Xie, G.C. Farrington, Solid State Ionics 5356 (1992) 10541058.
[109] G.C. Farrington, Proc. VIth Int. Meeting on Lithium Batteries, Munster FRD, May
1992, p. 34.
[110] T. Nagamoto, H. Kakahati, C. Ichikawa, O. Omoto, Jpn. J. Appl. Phys. 24 (1985) 397.
[111] S. Reich, I. Michaeli, J. Polym. Sci. Polym. Phys. Ed. 13 (1975) 918.
[112] M. Watanabe, M. Kanba, H. Matsuda, K. Tsunemi, K. Mizoguchi, E. Tsuchida, I.
Shinohara, Makromol. Chem. Rapid Commun. 2 (1981) 741744.
 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
[113] M. Watanabe, M. Kanba, K. Nagaoka, I. Shinohara, J. Polym. Sci. Polym. Phys. Ed. 21
(1983) 939948.
[114] K. Tsunemi, H. Ohno, E. Tsuchida, Electrochim. Acta 28 (1983) 833.
[115] R. Herr, Electrochim. Acta 35 (1990) 12571265.
[116] B. Scrosati, F. Croce, Proc. VIth Int. Meeting on Lithium Batteries, Munster FRD, May
1992.
[117] H. Hong, C. Liquan, H. Xuejie, X. Rongjian, Electrochim. Acta 37 (1992) 16711673.
[118] J.S. Lundsgaard, S. Yde-Andersen, R. Koksbang, D.R. Shackle, R.A. Austin, D. Fauteux,
in: B. Scrosati (Ed.)Proc. II Int. Symp. Polym. Electrolytes (ISPE-2), 1990, p. 395
(Siena June 1989).
[119] R. Huq, R. Koksbang, P.E. Tonder, G.C. Farrington, in: B.V.R. Chowdari, Q.G. Liu, L.Q.
Chen (Eds.), Recent Advances in Fast Ion Conducting Materials and Devices, World
Scientic, Singapore, 1990, p. 63.
[120] R. Huq, R. Koksbang, P.E. Tonder, G.C. Farrington, Electrochim. Acta 37 (1992)
16811684.
[121] E. Zygadlo-Monikowska, Z. Florjanczyk, W. Wieczorek, J. Macromol. Sci. Pure Appl.
Chem. A31 (1994) 11211134.
[122] W. Krawiec, Z. Florjanczyk, W. Wieczorek, IXth Int. Conference on Solid State
Ionics, September 1993 (Haque).
[123] A. Zalewska, J. Duminska, N. Langwald, J.S. Syzdek, M. Zawadzki, Electrochim. Acta.,
submitted.
[124] M. Walkowiak, A. Zalewska, T. Jesionowski, D. Waszak, B. Czajka, J. Power Sources
159 (2006) 449453.
[125] M. Walkowiak, A. Zalewska, T. Jesionowski, M. Pokora, J. Power Sources 173 (2007)
721728.
[126] M. Osiska, M. Walkowiak, A. Zalewska, T. Jesionowski, J. Membr. Sci. 326 (2009)
582588.
[127] A. Zalewska, M. Walkowiak, L. Niedzicki, T. Jesionowski, N. Langwald, Electrochim.
Acta 55 (2010) 13081313.
[128] M. Stolarska, L. Niedzicki, R. Borkowska, A. Zalewska, W. Wieczorek, Electrochim.
Acta 53 (2007) 15121517.
[129] K. Piotrowska, A. Zalewska, J.S. Syzdek, L. Niedzicki, M. Marcinek, ECS Trans. 25
(2010) 221230.
[130] A. Sarnowska, I. Polska, L. Niedzicki, M. Marcinek, A. Zalewska, Electrochim. Acta 57
(2011) 180186.
[131] K. Perzyna, R. Borkowska, J. Syzdek, A. Zalewska, W. Wieczorek, Electrochim. Acta
57 (2011) 5865.
[132] A. Zalewska, A. Bernakiewicz, M. Bystrzycka, M. Marczewski, N. Langwald, Int. J.
Hydrogen Energy 39 (2014) 29772987.
[133] J. Plocharski, W. Wieczorek, Solid State Ionics 2830 (1988) 979982.
[134] W. Wieczorek, K. Such, J. Przyluski, Z. Florianczyk, Synth. Met. 45 (1991) 373383.
[135] W. Wieczorek, Mater. Sci. Eng. B 15 (1992) 108114.
[136] W. Wieczorek, K. Such, J. Przyuski, Z. Floriaczyk, Synth. Met. 45 (1991) 373383.
[137] F. Croce, F. Bonino, S. Panero, B. Scrosati, Philos. Mag. B Phys. Condens. Matter Elec-
tron. Opt. Magn. Prop. 59 (1989) 161168.
[138] B. Scrosati, J. Electrochem. Soc. 136 (1989) 27742782.
[139] F. Croce, S. Passerini, A. Selvaggi, B. Scrosati, Solid State Ionics 4041 (1990) 375379.
[140] F. Croce, F. Capuano, A. Selvaggi, B. Scrosati, G. Scibona, J. Power Sources 32 (1990)
381388.
[141] S. Skaarup, K. West, B. Zachau-Christiansen, Solid State Ionics 2830 (1988)
975978.
[142] S. Skaarup, K. West, P.M. Julian, D.M. Thomas, Solid State Ionics 4041 (1990)
10211024.
[143] J.R. Stevens, B.-. Mellander, Solid State Ionics 21 (1986) 203206.
[144] C.C. Liang, J. Electrochem. Soc. 120 (1973) 12891292.
[145] J.E. Weston, B.C.H. Steele, Solid State Ionics 7 (1982) 7579.
[146] H. Yuankang, C. Zhushang, Z. Zhigi, W. Chaoying, C. Liquan, Chem. J. Chin. Univ. 2
(1986) 97.
[147] C. Linquan, in: B.V.R. Chowdari, S. Radhakrishnan (Eds.), Materials for Solid State
Batteries, World Scientic, Singapore, 1986, p. 69.
[148] J. Plocharski, W. Wieczorek, J. Przyluski, K. Such, Appl. Phys. A A49 (1989) 5560.
[149] F. Capuano, F. Croce, B. Scrosati, J. Electrochem. Soc. 138 (1991) 19181922.
[150] W. Gang, J. Roos, D. Brinkmann, F. Capuano, F. Croce, B. Scrosati, Solid State Ionics
5356 (1992) 11021105.
[151] S. Panero, B. Scrosati, S.G. Greenbaum, Electrochim. Acta 37 (1992) 15331539.
[152] F. Croce, R. Curini, A. Martinelli, L. Persi, F. Ronci, B. Scrosati, R. Caminiti, J. Phys.
Chem. B 103 (1999) 1063210638.
[153] F. Croce, G.B. Appetecchi, L. Persi, B. Scrosati, Nature 394 (1998) 456458.
[154] L. Sannier, A. Zalewska, W. Wieczorek, M. Marczewski, H. Marczewska,
Electrochim. Acta 52 (2007) 56855689.
[155] J. Syzdek, R. Borkowska, K. Perzyna, J.M. Tarascon, W. Wieczorek, J. Power Sources
173 (2007) 712720.
[156] J. Syzdek, M. Armand, M. Marcinek, A. Zalewska, G. ukowska, W. Wieczorek,
Electrochim. Acta 55 (2010) 13141322.
[157] J. Xi, X. Tang, Chem. Phys. Lett. 393 (2004) 271276.
[158] M. Moskwiak, I. Giska, R. Borkowska, A. Zalewska, M. Marczewski, H. Marczewska,
W. Wieczorek, J. Power Sources 159 (2006) 443448.
[159] F. Croce, L. Settimi, B. Scrosati, Electrochem. Commun. 8 (2006) 364368.
[160] F. Croce, S. Sacchetti, B. Scrosati, J. Power Sources 162 (2006) 685689.
[161] M. Marczewski, A. Jakubiak, H. Marczewska, A. Frydrych, M. Gontarz, A. niegua,
Phys. Chem. Chem. Phys. 6 (2004) 25132522.
[162] J. Syzdek, M. Armand, M. Marcinek, A. Zalewska, G. ukowska, W. Wieczorek,
Electrochim. Acta 55 (2010) 13141322.
[163] J.S. Syzdek, PhD Thesis, Warsaw (2009).
[164] J. Syzdek, M. Armand, M. Gizowska, M. Marcinek, E. Sasim, M. Szafran, W.
Wieczorek, J. Power Sources 194 (2009) 6672.
125
[165] J.S. Syzdek, M.B. Armand, P. Falkowski, M. Gizowska, M. Karowicz, . ukaszuk,
M.. Marcinek, A. Zalewska, M. Szafran, C. Masquelier, J.M. Tarascon, W.
Wieczorek, Z.G. Zukowska, Chem. Mater. 23 (2011) 17851797.
[166] J.S. Syzdek, M. Armand, P. Falkowski, M. Gumienniczuk, M.L. Marcinek, C.
Masquelier, M. Szafran, J.M. Tarascon, W. Wieczorek, A. Zalewska, Electrochim.
Acta 55 (2010) 1314.
[167] T. Mizumo, K. Sakamoto, N. Matsumi, H. Ohno, Electrochim. Acta 50 (2005)
39283933.
[168] Z. Xue, K. Wu, B. Liu, C. Chen, J. Power Sources 171 (2007) 944947.
[169] S. Tabata, T. Hirakimoto, H. Tokuda, M.A.B.H. Susan, M. Watanabe, J. Phys. Chem. B
108 (2004) 1951819526.
[170] X. Yang, H.S. Lee, W. Yoon, J. McBreen, S. Hossain, Y. Kim, Meet. Abstr. Electrochem.
Soc. 702 (2007) 727-727.
[171] H.S. Lee, Z.F. Ma, X.Q. Yang, X. Sun, J. McBreen, J. Electrochem. Soc. 151 (2004)
A1429A1435.
[172] M.P. Ciosek, M.P. Ciosek, M. Marcinek, W. Wieczorek, G. Zukowska, Meet. Abstr.
Electrochem. Soc. 701 (2007) 30-30.
[173] M. Marcinek, G.Z. Zukowska, W. Wieczorek, Electrochim. Acta 50 (2005)
39343941.
[174] M. Ciosek, M. Marcinek, G. ukowska, W. Wieczorek, Electrochim. Acta 54 (2009)
44874493.
[175] G. Zukowska, M. Szczechura, M. Marcinek, A. Zubrowska, A. Adamczyk-Wozniak,
A. Sporzynski, W. Wieczorek, ECS Trans. 16 (2009) 105114.
[176] H.S. Lee, X.Q. Yang, C.L. Xiang, J. McBreen, L.S. Choi, J. Electrochem. Soc. 145 (1998)
28132818.
[177] X. Sun, H.S. Lee, S. Lee, X.Q. Yang, J. McBreen, Electrochem. Solid-State Lett. 1
(1998) 239240.
[178] F. Zhou, D.R. MacFarlane, M. Forsyth, Electrochim. Acta 48 (2003) 17491758.
[179] A. Blazejczyk, W. Wieczorek, R. Kovarsky, D. Golodnitsky, E. Peled, L.G. Scanlon, G.B.
Appetecchi, B. Scrosati, J. Electrochem. Soc. 151 (2004) A1762A1766.
[180] A. Blazejczyk, M. Szczupak, W. Wieczorek, P. Cmoch, G.B. Appetecchi, B. Scrosati, R.
Kovarsky, D. Golodnitsky, E. Peled, Chem. Mater. 17 (2005) 15351547.
[181] M. Kalita, M. Bukat, M. Ciosek, M. Siekierski, S.H. Chung, T. Rodrguez, S.G.
Greenbaum, R. Kovarsky, D. Golodnitsky, E. Peled, D. Zane, B. Scrosati, W.
Wieczorek, Electrochim. Acta 50 (2005) 39423948.
[182] J. Scheers, P. Zawadzki, M. Kalita, P. Johansson, G.Z. Zukowska, W. Wieczorek, P.
Jacobson, Communicate on 10th, Int. Symp. Polym. Electrolytes2006.
[183] H. Mazor, D. Golodnitsky, E. Peled, W. Wieczorek, B. Scrosati, J. Power Sources 178
(2008) 736743.
[184] W. Wieczorek, M. Siekierski, in: P. Knauth, J. Schoonman (Eds.), Nanocomposites,
Springer, 2008 (Chapter 1).
[185] A. Guer, M. Dontigny, P. Charest, M. Petitclerc, M. Lagac, A. Vijh, K. Zaghib, J.
Power Sources 195 (3) (2010) 845852.
[186] Z. Stoeva, I. Martin-Litas, E. Staunton, Y.G. Andreev, P.G. Bruce, J. Am. Chem. Soc.
125 (2003) 46194626.
[187] G.S. MacGlashan, Y.G. Andreev, P.G. Bruce, Nature 398 (1999) 792794.
[188] Z. Gadjourova, Y.M. Martin, K.H. Andersen, Y.G. Andreev, P.G. Bruce, Chem. Mater.
13 (2001) 12821285.
[189] A.M. Christie, S.J. Lilley, E. Staunton, Y.G. Andreev, P.G. Bruce, Nature 433 (2005)
5053.
[190] C. Zhang, E. Staunton, Y.G. Andreev, P.G. Bruce, J. Am. Chem. Soc. 127 (2005)
1830518308.
[191] S.J. Lilley, Y.G. Andreev, P.G. Bruce, J. Am. Chem. Soc. 128 (2006) 1203612037.
[192] E. Staunton, Y.G. Andreev, P.G. Bruce, Faraday Discuss. 134 (2007) 143.
[193] C. Zhang, E. Staunton, Y.G. Andreev, P.G. Bruce, J. Mater. Chem. 17 (2007) 3222.
[194] C. Zhang, S. Gamble, D. Ainsworth, A.M.Z. Slawin, Y.G. Andreev, P.G. Bruce, Nat.
Mater. 8 (2009) 580.
[195] W.A. Henderson, N.R. Brooks, W.W. Brennessel, V.G. Young, Chem. Mater. 15
(2003) 46794684.
[196] W.A. Henderson, N.R. Brooks, V.G. Young, Chem. Mater. 15 (2003) 46854690.
[197] V. Seneviratne, R. Frech, J.E. Furneaux, M. Khan, J. Phys. Chem. B 108 (2004)
81248128.
[198] Y.G. Andreev, V. Seneviratne, M. Khan, W.A. Henderson, R.E. Frech, P.G. Bruce,
Chem. Mater. 17 (2005) 767772.
[199] C. Zhang, Y.G. Andreev, P.G. Bruce, Angew. Chem. Int. Ed. 46 (2007) 28482850.
[200] C. Zhang, S.J. Lilley, D. Ainsworth, E. Staunton, Y.G. Andreev, A.M.Z. Slawin, P.G.
Bruce, Chem. Mater. 20 (2008) 40394044.
[201] C. Zhang, D. Ainsworth, Y.G. Andreev, P.G. Bruce, J. Am. Chem. Soc. 129 (2007)
87008701.
[202] C.A. Angell, C. Liu, E. Sanchez, Nature 362 (1993) 137139.
[203] M.G. McLin, C.A. Angell, Solid State Ionics 5356 (1992) 10271036.
[204] A. Ferry, L. Edman, M. Forsyth, D.R. MacFarlane, J. Sun, Electrochim. Acta 45 (2000)
12371242.
[205] M. Forsyth, D.R. MacFarlane, A.J. Hill, Electrochim. Acta 45 (2000) 12431247.
[206] M. Forsyth, S. Jiazeng, D.R. MacFarlane, Electrochim. Acta 45 (2000) 12491254.
[207] A. Zalewska, I. Pruszczyk, E. Sulek, W. Wieczorek, Solid State Ionics 157 (2003)
233239.
[208] L. Suo, Y.-S. Hu, H. Li, M. Armand, L. Chen, Nat. Commun. 4 (2013) 14811490.
[209] D.W. McOwen, D.M. Seo, O. Borodin, J. Vatamanu, P.D. Boyle, W.A. Henderson,
Energy Environ. Sci. 7 (2014) 416426.
[210] M.J. Marczewski, B. Stanje, I. Hanzu, M. Wilkening, P. Johansson, Phys. Chem. Chem.
Phys. 16 (24) (2014) 1234112349.
[211] F. Mammeri, E. Le Bourhis, L. Rozes, C. Sanchez, J. Mater. Chem. 15 (2005)
37873811.
[212] C. Sanchez, F. Ribot, New J. Chem. 18 (1994) 1007;
C. Sanchez, B. Julin, P. Belleville, M. Popall, J. Mater. Chem. 15 (2005) 35593592.
126 M. Marcinek et al. / Solid State Ionics 276 (2015) 107126
[213] O. Lev, Z. Wu, S. Bharathi, V. Glezer, A. Modestov, J. Gun, L. Rabinovich, S. Sampath,
Chem. Mater. 9 (1997) 23542375.
[214] M. Popall, H. Durand, Electrochim. Acta 37 (1992) 15931597.
[215] M. Popall, R. Buestrich, G. Semrau, G. Eichinger, M. Andrei, W.O. Parker, S. Skaarup,
K. West, Electrochim. Acta 46 (2001) 14991508.
[216] P. Judeinstein, J. Titman, M. Stamm, H. Schmidt, Chem. Mater. 6 (1994) 127134.
[217] K. Dahmouche, M. Atik, N.C. Mello, T.J. Bonagamba, H. Panepucci, M.A. Aegerter, P.
Judeinstein, J. Sol-Gel Sci. Technol. 8 (1997) 711715.
[218] J.L. Nugent, S.S. Moganty, L.A. Archer, Adv. Mater. 22 (2010) 36773680.
[219] J.L. Schaefer, S.S. Moganty, D.A. Yanga, L.A. Archer, J. Mater. Chem. 21 (2011)
1009410101.
[220] V. Mnchow, V. Di Noto, E. Tondello, Electrochim. Acta 45 (2000) 12111221.
[221] V. Di Noto, V. Zago, S. Biscazzo, M. Vittadello, Electrochim. Acta 48 (2003) 541554.
[222] K. Onishi, M. Matsumoto, Y. Nakacho, K. Shigehara, Chem. Mater. 9 (1996)
469472.
[223] K. Matsushita, Y. Shimazaki, M.A. Mehta, T. Fujinami, Solid State Ionics 133 (2000)
295301.
[224] Y. Masuda, M. Seki, M. Nakayama, M. Wakihara, H. Mita, Solid State Ionics 177
(2006) 843846.
[225] Y. Masuda, M. Nakayama, M. Wakihara, Solid State Ionics 178 (2007) 981986.
[226] V. Di Noto, V. Zago, J. Electrochem. Soc. 151 (2004) A216A223.
[227] M. Marczewski, M. Piszcz, A. Plewa-Marczewska, G.Z. ukowska, A. Pietrzykowski,
M. Siekierski, Electrochim. Acta 55 (2010) 13381346.
[228] M. Piszcz, M. Marczewski, A. Plewa-Marczewska, G.Z. ukowska, A. Zalewska, A.
Pietrzykowski, M. Siekierski, J. Power Sources 195 (2010) 74957505.
[229] M. Piszcz, M. Marczewski, G.Z. ukowska, J. Wjcik, W. Wieczorek, M. Siekierski,
Solid State Ionics 245246 (2013) 3342.
[230] C. Pfaffenhuber, M. Go bel, J. Popovic, J. Maier, Phys. Chem. Chem. Phys. 15 (2013)
1831818335.
[231] M. Egashira, B. Scrosati, M. Armand, S. Beranger, C. Michot, Electrochem.
Solid-State Lett. 6 (2003) A71A73.
[232] L. Niedzicki, E. Karpierz, A. Bitner, M. Kasprzyk, G.Z. Zukowska, M. Marcinek, W.
Wieczorek, Electrochim. Acta 117 (2014) 224229.
[233] T. Trzeciak, L. Niedzicki, G. Groszek, P. Wieczorek, M. Marcinek, W. Wieczorek, J.
Power Sources 252 (2014) 229234.
[234] US Department of the Interior, US Geological Survey, Mineral Commodity Summa-
ries (2011).
[235] A. Eftekhari, J. Power Sources 126 (2004) 221228.
[236] D. Linden, T.B. Reddy, Handbook of Batteries, McGraw-Hills, New York, 2003.
[237] R.M. Dell, Solid State Ionics 134 (2000) 139158.
[238] P. Johansson, S. Branger, M. Armand, H. Nilsson, P. Jacobsson, Solid State Ionics
156 (2003) 129139.
[239] T.J. Barbarich, P. Driscoll, Electrochem. Solid-State Lett. 6 (2003) A113.
[240] T.J. Barbarich, K.M. Abraham, J. DiCarlo, Proceedings of the 40th Power Sources
Conference, IEEE, Cherry Hill, NJ, 2002, p. 331.
[241] K. Tasaki, K. Kanda, S. Nakamura, M. Ue, J. Electrochem. Soc. 150 (2003)
A1628A1636.
[242] T.J. Barbarich, P.F. Driscoll, S. Izquierdo, L.N. Zakharov, C.D. Incarvito, A.L. Rheingold,
Inorg. Chem. 43 (2004) 77647773.
[243] P. Johansson, H. Nilsson, P. Jacobsson, M. Armand, Phys. Chem. Chem. Phys. 6
(2004) 895899.
[244] M. Armand, P. Johansson, J. Power Sources 178 (2008) 821825.
[245] L. Niedzicki, M. Kasprzyk, K. Kuziak, G.Z. ukowska, M. Armand, M. Bukowska, M.
Marcinek, P. Szczeciski, W. Wieczorek, J. Power Sources 192 (2009) 612617.
[246] L. Niedzicki, G.Z. ukowska, M. Bukowska, P. Szczeciski, S. Grugeon, S. Laruelle, M.
Armand, S. Panero, B. Scrosati, M. Marcinek, W. Wieczorek, Electrochim. Acta 55
(2010) 14501454.
[247] L. Niedzicki, S. Grugeon, S. Laruelle, P. Judeinstein, M. Bukowska, J. Prejzner, P.
Szczecinski, W. Wieczorek, M. Armand, J. Power Sources 196 (2011) 86968700.
[248] M. Bukowska, P. Prejzner, P. Szczecinski, Pol. J. Chem. 78 (2004) 417.
[249] R.W. Begland, D.R. Hartter, F.N. Jones, D.J. Sam, W.A. Sheppard, O.W. Webster, F.J.
Weigert, J. Org. Chem. 39 (1974) 23412350.
[250] J. Scheers, P. Johansson, P. Szczeciski, W. Wieczorek, M. Armand, P. Jacobsson, J.
Power Sources 195 (2010) 60816087.
[251] J. Evans, C.A. Vincent, P.G. Bruce, Polymer 28 (1987) 23242329.
[252] M.C. Wintersgill, J.J. Fontanella, P.E. Stallworth, S.A. Newman, S.H. Chung, S.G.
Greenbaum, Solid State Ionics 135 (2000) 155161.
[253] H. Zhang, S. Kulkarni, S.L. Wunder, J. Electrochem. Soc. 153 (2006) A239A248.
[254] S. Jankowsky, M.M. Hiller, H.-D. Wiemhfer, J. Power Sources 253 (2014) 256262.
[255] J.F. Snyder, M.A. Ratner, D.F. Shriver, J. Electrochem. Soc. 150 (2003) A1090A1094.
[256] L.M. Bronstein, E. Ashcraft, Peter DeSanto Jr., R.L. Karlinsey, J.W. Zwanziger, J. Phys.
Chem. B 108 (2004) 58515858.
[257] J. Zhou, P.S. Fedkiw, Solid State Ionics 166 (2004) 275293.
[258] A. Manuel Stephan, K.S. Nahm, Polymer 47 (2006) 59525964.
[259] G.B. Appetecchi, G.T. Kim, M. Montanino, F. Alessandrini, S. Passerini, J. Power
Sources 196 (2011) 67036709.
[260] A. Lewandowski, A. widerska-Mocek, J. Power Sources 194 (2009) 601609.
[261] H. Ye, J. uang, J.J. Xu, A. Khalfan, S.G. Greenbaum, J. Electrochem. Soc. 154 (11)
(2007) A1048A1057.
[262] N. Matsumi, K. Sugai, M. Miyake, H. Ohno, Macromolecules 39 (2006) 69246927.
[263] F.B. Dias, L. Plomp, Jakobert B.J. Veldhuis, J. Power Sources 88 (2000) 169191.