Energy &

Environmental
Science
View Article Online
REVIEW View Journal | View Issue

Electrolyte additives for lithium ion battery

electrodes: progress and perspectives
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Cite this: Energy Environ. Sci.,

2016, 9, 1955
Atetegeb Meazah Haregewoin,a Aselefech Sorsa Wotangoa and Bing-Joe Hwang*ab

The need for lighter, thinner, and smaller products makes lithium ion batteries popular power sources for
applications such as mobile phones, laptop computers, digital cameras, electric vehicles, and hybrid electric
vehicles. For high power applications, the development of high capacity and high voltage electrode materials
is in progress. Battery performance and safety issues are also related to the properties of the electrolytes
used. To improve the properties of the electrolytes, small amounts of other components, known as
electrolyte additives, are incorporated. This paper reviews the recent progress in electrolyte additives
used to improve performance and other properties, such as safety. This review classifies the additives
Received 13th January 2016, based on their functions and their effects on specific electrode materials focusing on electrodes under
Accepted 4th May 2016 current development. From anodes: carbonaceous electrodes, silicon, tin and Li4Ti5O12; from layered
DOI: 10.1039/c6ee00123h cathodes: LiCoO2, Li-rich and LiNiyMnyCo12yO2 (NMC); from spinel: LiMn2O4, and from olivine: LiFePO4
are selected. We believe that this approach will help readers easily identify and understand the additives
[Link]/ees suitable for their target materials.

Broader context
In the current state-of-the-art battery technology, lithium-ion rechargeable batteries are commonly used as power sources for portable consumer products. The
performance and safety of a lithium ion battery can be related to the solid electrolyte interphase (SEI) layer formed on the electrode’s surface. The nature of this
film and its formation mechanism depend mainly on the electrolyte’s composition. The most economical approach to circumventing some of the problems
affecting lithium ion batteries is the use of electrolyte additives. During the past several years, various electrolyte additives have been reported as being able to
improve the properties of the SEI layer on the electrodes’ surfaces; the ionic conductivity of the electrolyte; the thermal stability of LiPF6; and the safety of the
battery, enable cathode material protection from dissolution and overcharging, as well as achieve reduction in irreversible capacity and gas generation. This
review summarizes various types of electrolyte additives by classifying them according to their functions and discusses the remaining issues that need to be
addressed so that lithium-ion rechargeable batteries can reach their full potential.

1. Introduction strength, lithium ion batteries have been widely used as power
sources in a broad range of applications, including mobile
Recently, secondary batteries have received renewed interest as phones, laptop computers, and digital cameras; they are also
power sources for electric and hybrid electric vehicles. Research and promising energy sources for electric vehicles, hybrid electric
development (R&D) has focused on improving the performance of vehicles and aerospace applications.2 Progress has been made
existing battery systems and developing new systems to achieve the on the development of flexible lithium ion batteries having
performance needed for specific applications. Among secondary nanostructured electrode materials, shape-conformable solid
batteries, lithium ion batteries have attracted the attention of many electrolytes, and soft current collectors.3,4 The lithium ion
researchers, due to the high demand for lighter, thinner and smaller battery is also known as the ‘‘rocking chair’’ battery because
products. Lithium has the lightest weight, with a density of lithium ions move back and forth between the cathode and the
0.534 g cm3, highest voltage (B3.6 V), and greatest energy anode through the electrolyte during electrochemical reactions.
density (B240 W h kg1) of all metals.1 On the basis of this Fig. 1 depicts the typical charging and discharging processes
in a lithium ion battery. During the initial charging, under an
a
applied voltage, some lithium ions migrate from the lithium
Department of Chemical Engineering, National Taiwan University of Science and
Technology, 43 Keelung Road, Section 4, Taipei 106, Taiwan, Republic of China.
compound of the cathode (such as LiCoO2) to the carbon anode
E-mail: bjh@[Link] material as shown in the forward reaction of eqn (1) and (2), with
b
National Synchrotron Radiation Research Center, Hsin-Chu 300, Taiwan the reverse reactions occurring during discharge (eqn (1) and (2)).

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1955
 View Article Online

Review Energy & Environmental Science

When the lithium ions are transported to and from the cathode
or anode during charging and discharging, the associated
transition metal is oxidized and reduced. For instance, in
LixCoO2 cobalt oxidizes from Co3+ to Co4+ during charging
and is reduced from Co4+ to Co3+ during discharge.

LiCoO2 $ Li1xCoO2 + xLi+ + xe (1)

xLi+ + xe + C $ LixC (2)

The main components of a typical lithium ion battery are the

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

cathode, anode, electrolyte (with lithium salts dissolved in non-

aqueous solvents), separator, and current collector (Al-foil for
cathode materials and Cu-foil for anode materials). Carbonaceous
materials, such as graphite, are commonly used as anode

Fig. 1 Schematic illustration of typical lithium ion battery processes.

Dr Atetegeb Meazah Haregewoin

earned her PhD in Chemical materials in rechargeable lithium ion batteries. They serve as
Engineering from the National host structures for lithium intercalation and when lithiated
Taiwan University of Science and they have stable phases up to LiC6 stoichiometry.4–6 Crystallinity,
Technology (NTUST), Department morphology, microstructure, texture, etc., have a significant
of Chemical Engineering, Taiwan effect on the electrochemical lithium intercalation performance
and her MSc in Inorganic of the carbonaceous materials. Among the wide variety of
Chemistry from Addis Ababa carbonaceous materials, because of its good cycling ability and
University, Ethiopia. She has small irreversible capacity, graphite (particularly MCMB) is now
worked as a postdoctoral fellow most widely used as an anode in lithium ion batteries.1,4,5
at the Nanoelectrochemistry Lab However, graphite anodes suffer from various problems such
of Department of Chemical as structural surface damage upon cycling,7 capacity fade,8,9 and
Atetegeb Meazah Engineering, NTUST, under the exfoliation due to co-intercalation of solvents e.g. propylene
Haregewoin supervision of Professor Bing-Joe carbonate (PC).10 To solve these problems, together with the
Hwang. She has also worked as an assistant professor at the need for much higher energy and power densities, efforts on
Chemistry Department of Bahir Dar University, Ethiopia. Her R&D have continued in the selection of materials such as Si,11–16
research focuses on identifying and understanding the surface SnO2/C,17 Sn and Sn/C composites,18–20 and on improving the
phenomenon in lithium ion batteries as well as designing, currently used graphite anode materials.21–23 Moreover, the fabrica-
synthesizing and characterizing potential electrolyte and tion of nanostructured materials with designed morphologies using
electrolyte additives that could form an effective surface film.

Aselefech Sorsa is currently a PhD Bing-Joe Hwang studied Chemical

student in Chemical Engineering Engineering and received his PhD in
department at the National 1987 at the National Cheng Kung
Taiwan University of Science and University, Taiwan. Since 2006, he
Technology, Taiwan. She received has served as a chair professor at
her bachelor’s degree in Chemistry the National Taiwan University of
from Jimma University in 2005 and Science and Technology. His
master’s degree in Chemistry research work has spanned a wide
(Organic) from Addis Ababa range of subjects from electro-
University, Ethiopia in 2007. chemistry to spectroscopy, inter-
Then, she joined Debre Markos facial phenomena, materials
University as a lecturer for science and theoretical chemistry.
Aselefech Sorsa Wotango undergraduate students in organic Bing-Joe Hwang He has established both experi-
chemistry. She changed her work mental and computational
place to Addis Ababa Science and Technology University in 2012. strategies for the development of new nanoscale materials. His work
Currently, she is studying PhD under Prof. Bing-Joe Hwang, focusing has led to a better understanding of reaction mechanisms on
on electrolyte additives for lithium ion batteries. nanoparticles and to an improved ability to predict the properties of
potential new materials for ion batteries, fuel cells and solar cells.

1956 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

the electrospinning method is another approach to solve the

problems related to powder materials.24
Based on their structures, lithium ion battery cathode
materials are classified into three groups, namely, layered
compounds LiMO2 (M = Co, Ni, Mn, etc.), spinel compounds
LiM2O4 (M = Mn, etc.), and olivine compounds LiMPO4 (M = Fe,
Mn, Ni, Co, etc.). Among the layered oxides, LiCoO2 is mainly
used as the positive electrode-active material of the current
commercial lithium ion batteries, due to its high theoretical
capacity (274 mA h g1), high working potential and excellent
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

cycling ability at room temperature. However, only half of the

theoretical value (B140 mA h g1) can be obtained in actual
cycling, due to structural instability when more than half of the
Li ions are extracted and cobalt undergoes dissolution in the
electrolyte.25,26 Intensive studies have been directed at finding
other transition metal ions to partially/fully substitute Co ions.
Ni and Mn, or a mixture of these, have been reported as
alternatives to LiCoO2.27–33 Spinel structured cathode materials
such as LiMn2O4, showing a flat operating voltage of 3.95–4.1 V
with a theoretical capacity of 148 mA h g1, have been developed
as possible alternatives to LiCoO2 to improve battery safety.
Another alternative cathode material reported to give the battery
a long cycle life is olivine LiFePO4. It has a flat potential of 3.5 V Fig. 3 The SEI/SPI film formation processes due to the decomposition of
vs. Li/Li+ and a theoretical specific capacity of 170 A h kg1.34 electrolytes on both the anode and cathode sides.
Most cathode materials are frequently exposed to voltages above
4.2 V vs. Li/Li+, i.e., the region where the oxidation of the
electrolyte occurs. The main issue in using these electrodes is carbonate (EC), PC, diethyl carbonate (DEC), and ethyl methyl
the significant oxidative decomposition of electrolytes on the carbonate (EMC). The use of polymer36 and ionic liquid37 based
surface of the electrodes that causes high capacity loss and poor electrolytes is also under investigation. The electrolytes in
cycling ability.35 lithium ion batteries transport lithium ions from the cathode
A fundamental understanding of lithium ion battery composite to the anode and vice versa during charging and discharging. In
cathode surface phenomena is essential for improving battery this process, the solvents must be anodically and cathodically
electrochemical performance and safety. Fig. 2 shows the current stable; however, this is not practical since they are thermo-
and proposed future technologies for the main components of a dynamically unstable with lithium or LixC6 in the operating
lithium-ion battery. potential range. Hence, a passivation film termed a solid
The electrolytes in a lithium ion battery consist of one, or electrolyte interphase (SEI) is formed during the lithium ion
more, conducting lithium salts, e.g., LiClO4, LiAsF6, LiBF4, battery’s first charge, due to the reduction of electrolytes on the
LiPF6, dissolved in a single or mixtures of non-aqueous solvents. surface of the electrode. Similarly, a surface film known as a
Such solvents include cyclic and acyclic carbonates such as ethylene solid permeable interface (SPI) is formed on the cathode’s
surface due to the oxidation of the electrolytes. A good SEI or
SPI layer should prevent further electrolyte degradation by
blocking the electron transport while allowing lithium ions to
pass through it during cycling.38 Fig. 3 shows the film formation
processes on both the anode and cathode sides due to the
reduction and oxidation of electrolytes, respectively.
For systems where a stable SEI cannot be formed, coating
the electrode’s surface with an artificial SEI is an alternative. Li
et al.39 have reported that coating a Si electrode with lithium
phosphorus oxynitride (Lipon) can prevent electrolyte decom-
position. Good results can be obtained by controlling the film’s
thickness and the potential range. A Si electrode coated with at
least 50 nm Lipon had a high columbic efficiency when cycled
between 2 V and 0.1 V. A recent report by Zhao et al.40 showed that
the high reactivity of prelithiated Si nanoparticles (LixSi NPs)
Fig. 2 Schematic diagram showing the current and future technologies under ambient conditions can be suppressed by exposing the
for lithium-ion batteries based on the main components. LixSi NPs to 1-fluorodecane. The highly reactive LixSi NPs reduce

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1957
 View Article Online

Review Energy & Environmental Science

1-fluorodecane, while the resulting LiF and lithium alkyl safety of the electrolyte. The additives should reduce irreversible
carbonate with long hydrophobic carbon chains cover the capacity and gas generation, improve the thermal stability of LiPF6
electrode’s surface. The artificial SEI effectively improves the against organic electrolyte solvents, and protect the cathode
stability of the electrode under ambient conditions. The LixSi material from dissolution and overcharging.42,43 In this review,
NPs coated with the artificial SEI showed B2100 mA h g1 in various electrolyte additives reported for the improvement of
dry air and 1600 mA h g1 in humid air. An artificial SEI formed different battery properties are discussed.
by a combination of two polymers, polyethylene glycol tert-
octylphenyl ether (PEGPE) and poly(allyl amine) (PAAm), has
been reported by Li et al.41 They used PEGPE to coordinate with 2. Anode additives
the Li+ ion (through the PEG oxygen lone pair) and for strong
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

adhesion with the electrode (thorough the aromatic moiety). Most commonly used anode materials in the current R&D of
PAAm was blended with PEGPE for a better coating integrity and lithium ion battery technology have limitations as summarized
mechanical strength. The coated electrodes (natural graphite in Scheme 1. To overcome these challenges, one strategy is the
(NG), artificial graphite (AG), and graphite with 5% Si composite use of electrolyte additives. Anode additives are expected to
(SG)) showed good cyclability and rate capability. NG exhibited reduce at higher potentials than the carbonate electrolytes to
336 mA h g1 at a 10C rate and AG exhibited 260 mA h g1 at a form a SEI layer on the electrode’s surface. The surface film
25C rate. formed should be thin enough to passivate the electrode and
The current state-of-the-art electrolyte systems for lithium prevent the decomposition of the electrolytes, while being at
ion batteries have their own drawbacks such as irreversible the same time conductive to lithium ions; however, most anode
capacity, temperature limits, and safety issues. One way of additives contain functional groups that make them vulnerable
minimizing these problems is to incorporate small amounts to reduction. In order to achieve reduction at a higher potential,
of other components into the system with the intention of the additive should have a lower lowest unoccupied molecular
changing the targeted properties of the electrolyte while main- orbital (LUMO) than the electrolyte solvents. However, having a
taining its bulk properties;42 thus the development of electro- lower LUMO does not guarantee the formation of a stable and
lyte additives is currently an important focus of research. Many compact SEI layer. Additives that have been tested for different
lithium ion battery problems are related to surface phenomena anode materials are discussed below.
that occur on the electrode’s surface. The use of electrolyte
additives is the most economical and efficient method to 2.1. Carbonaceous anodes
circumvent these problems and thus improve lithium ion Previously, Li was used as an anode material; however, due to
battery performance. The electrode’s surface morphology can dendrite formation, current lithium ion batteries use carbonaceous
be improved by the addition of small amounts of additives, as materials as the anodes. Even though state-of-the-art commercial
they are usually preferentially involved in the interfacial redox lithium ion batteries use carbon as anode material, this also has
process, i.e. prior to the electrolytes. Electrolyte additives in drawbacks such as (i) decomposition of electrolytes on the carbon
lithium ion systems improve not only the properties of the SEI electrode’s surface. Since the carbon electrode surface acts as a
on the electrodes’ surfaces, but also the ionic conductivity and catalyst, the electrolytes usually undergo reduction on the

Scheme 1 Key issues and challenges facing commonly used anode materials in lithium ion batteries.

1958 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

electrode’s surfaces at the potential below 1.5 V. Reduction of

the electrolytes leads to the formation of films such as
ROCO2Li, (ROCO2Li)2, and Li2CO3 and generation of gases.
The film formation process, which degrades the electrolytes
and consumes lithium ions, leads to capacity fading. On the
other hand, gas formation causes the build-up of pressure
inside the cell resulting in rupture. (ii) Capacity fading due to
the effect of electrolyte impurities. The presence of trace
amounts of impurities, such as H2O, O2, and CO2, in the
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

electrolyte leads to the formation of Li2CO3, Li2O2, LiO2, and

LiOH. Water has the ability to degrade LiPF6-based electrolytes
and generate HF, while reducing the electrolyte’s concentration,
and decomposing some of the SEI’s major components at the
carbon anode, and in doing so consumes Li ions leading Fig. 4 Structures of additives for carbon anodes containing maleimide,
to capacity fading. Moreover, the generated HF degrades vinylene, halogene and nitrogen functional groups.
the electrode’s surface. (iii) Exfoliation of graphite due to
co-intercalation of electrolyte solvents such as PC. When PC based
electrolytes are used for graphite anodes, PC co-intercalates with Degradation of electrolytes due to further reaction of the alkyl
lithium ions into the graphene layers and is subsequently carbonate solvents with their decomposition products, i.e. alkoxide
reduced, thereby exfoliating the graphite electrode. Moreover, anions, can be minimized by the addition of vinylene carbonate
PC reduction into the graphene layers causes the generation of (VC, 2), which can consume the alkoxide anions and improve
gases between the graphene planes and the build-up of pressure the battery’s cycling performance.70 VC also has the ability to
within the particles which leads to the cracking and splitting of form a very thin polymeric film on the electrode’s surface to
the layers. (iv) The carbon anode is also degraded by Mn dissolu- prevent solvent decomposition and minimize the resistance of
tion when used with spinel (LiMn2O4) cathodes. The dissolution the battery.71,72
of Mn into the electrolyte causes the deposition of Mn(II) on the The formation of alkoxides, due to the decomposition of
graphite electrode, which subsequently undergoes electro- electrolyte solvents, can also be reduced by the addition of vinyl
reduction and degrades the graphite surface. One solution to ethylene carbonate (VEC, 3).73 Reduction of VEC leads to the
these problems is the use of electrolyte additives. formation of Li2CO3, ROCO2Li and polymeric species. The
The problem of decomposition of electrolytes can be over- polymeric species formed by the reduction of VEC have the
come using SEI film forming additives which are reduced at ability to inhibit the attack of trace amounts of electrolyte
higher potentials than the electrolyte solvents and passivate the impurities such as H2O. VEC modifies the properties and
electrode’s surface to prevent further reduction of the electrolyte structures of surface films both on the anode and cathode
solvents. The SEI film formed by the electrolyte additives is sides by forming an effective SEI on both sides, thereby
dependent on the functional group incorporated. As the func- improving the performance of the battery. This approach
tional group in the additive is changed, its physicochemical enhances the high voltage stability of LiCoO2/graphite cells by
properties change, resulting in the formation of different SEI preventing LiCoO2 dissolution and also suppresses the reaction
components. Electrolyte additives with various functional between LiPF6 and solvents by stabilizing LiPF6.74
groups include inorganic compounds44–50 and organic compounds Maleimide (MI) additives, which are polymeric film forming
containing unsaturated carbon bonds,51–54 sulfur-containing agents, improve the reversible capacity and cycling perfor-
components,55–60 halogen-containing components,61–63 and mance of the battery. Wang et al., reported a series of phenyl-
other components.64–68 enedimaleimide compounds having the maleimide group in
Huang et al.69 reported that a low concentration (1 wt%) of the ortho (4), meta (5) and para (6) positions.75,76 They reported
ammonium perfluorocaprylate (APC, 1) electrolyte additive can that all of the MI additives were reduced prior to the solvents to
form a uniform and compact SEI film on the electrode’s surface form a stable SEI on the electrode’s surface, while showing a
at a higher potential than the electrolyte reduction potential better performance than the additive free electrolyte. However,
(Fig. 4). The SEI formed by APC can both prevent the reduction the additives with the para position substituents showed a
of electrolyte solvents and minimize the charge transfer resis- better performance than those with ortho and meta position
tance. When organic electrolyte solvents such as EC, DEC and substituents. Although their reduction mechanism is similar,
PC are reduced on the negative electrode surface, the reduction they show various SEI film distributions and densities. The
products include Li2CO3, ROCO2Li, and (ROCO2Li)2. Since additive at the para position forms a dense and highly uniform
Li2CO3 is a good passivating agent, one way of suppressing SEI; however, the ortho position shows loose and discontinuous
the reduction of the electrolyte is saturating the electrolyte with SEI formation.
Li2CO3.49 When a low concentration of Li2CO3 is added into the A series of methyl substituted glycolide derivatives, including
electrolyte, Li2CO3 will be deposited on the electrode’s surface, glycolide (GL00, 7), 3-methyl glycolide (GL01, 8), 3,6-dimethyl
thereby suppressing electrolyte reduction by surface passivation. glycolide (GL11, 9) and 3,3,6,6-tetramethyl glycolide (GL22, 10),

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1959
 View Article Online

Review Energy & Environmental Science

Fig. 5 Structures of additives for carbon anodes containing glycolide

derivatives and oxaspiro functional groups.
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

have been reported as SEI film forming additives and compared

with VC (Fig. 5).77 As the number of methyl substituents increased,
the important characteristics of additives, such as anodic stability,
cathodic reactivity, nucleophilic reactivity, and polymerizability, are
deteriorated due to the increased electron density and steric
hindrance. Hence, the use of glycolides with more methyl sub-
stituents leads to a reduced battery performance. Among all the
glycolides tested, GL01 is reported to be the best SEI forming
additive for practical applications because of its high anodic
stability, good cyclability, and excellent thermal stability. By varying
the functional group, it is possible to modify the applicability and
effectiveness of the additive. Hence, relating the functional groups
(nature and position) to the intended battery application is an
important electrolyte additive design consideration.
3,9-Divinyl-2,4,8,10-tetraoxaspiro[5,5]undecane (TOS, 11) also Fig. 7 Structures of additives for carbon anodes containing vinyl, allyl,
sulfur, oxalate functional groups, butyrolactone derivatives and ionic
formed a polymeric film on the MCMB electrode showing good
liquids.
capacity retention in the (MCMB)/Li1.1[Ni1/3Co1/3Mn1/3]0.9O2
cell.78 At high temperatures, TOS delays the onset temperature
of the SEI breakdown on the negative electrode and also reduces These additives suppress both the thermal decomposition of
the reactivity of the cathode by decreasing heat accumulation, the electrolyte solutions and their reactivity towards lithium.
thereby improving the thermal stability of the cell. However, it This is because of the presence of the poly-ether group, which
has the negative effect of increasing the cell’s impedance, hence has lower reactivity towards lithium. Siloxanes have also been
its use at concentrations below 1% has been suggested. The use of used as co-solvents but with limited success.80 Since LiPF6 is less
poly-ether modified siloxanes, such as diethylene glycol methyl- soluble in poly-ether modified siloxanes, carbonate-modified
(3-dimethyl(trimethylsiloxy)silyl propyl)ether (EGMSPE, 12), siloxanes have been tried as co-solvents, however, their effect
diethylene glycol methyl-(3-dinethyl(trimethylsiloxy)silyl propyl)-2- is insignificant.
methylpropyl ether (EGMSPMPE, 13), diethylene glycol methyl- The effects of water can be inhibited by the addition of a
(3-bis(trimethylsiloxy)silyl propyl)ether (EGMESPMP, 14) and Lewis base additive such as, dimethyl acetamide (DMAc, 17),81
diethylene glycol-(3-methyl-bis(trimethylsiloxy)silyl-2-methylpropyl)- which prevents the reaction of water with LiPF6 by capturing
ether (EGMSBPE, 15), has resulted in good cycling performance the Lewis acidic PF5 moiety to form a stable complex (Fig. 7).
for various anodes including graphite and silicon (Fig. 6).79 A LiNi0.8Co0.2O2/MCMB cell with the DMAc additive showed
good cycling performance in the presence of water contaminants.
Similarly, the cyclic performance of LiFePO4/graphite cells at
elevated temperatures has been improved by the presence of
DMAc,82 which forms a complex with PF5 generated when LiPF6
is thermally dissociated. 2,2-Dimethoxy-propane (DMP, 16), a well-
known H2O removal agent, has been reported to be a good
SEI film forming additive for graphite electrodes.83 At very low
concentrations (0.005 wt%), DMP can effectively minimize the
irreversible capacity. It also showed good cycling performance for
LiCoO2/graphite full cells after storage at high temperatures.
The other problem with the graphite anode is its incompatibility
with PC solvent. PC is among the commonly used solvents in
lithium ion batteries. It has a wide temperature range, high
Fig. 6 Structures of additives for carbon anodes containing siloxane with ionic conductivity, high dielectric constant, and static stability
various alkyl substituents. with lithium. However, PC cannot form an effective SEI layer on

1960 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

the graphite electrode surface and also co-intercalates with

lithium ions to cause graphite exfoliation.84 To solve this
problem and improve the performance of the battery in PC
based electrolytes, the use of co-solvents and/or additives was
suggested.42 Different people have reported the use of different
electrolyte additives to avoid the exfoliation of graphite in PC
based electrolytes.55,62,68,85–91 Formation of SEI films comprising
mainly organic polymers improves the elasticity of the films,
thereby helping them to withstand anode material volume
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

changes during the lithium ion insertion/extraction processes.

Moreover, polymeric species have the ability to passivate the
electrode’s surface and prevent solvent co-intercalation. For the
formation of such polymeric films, the presence of unsaturated
carbon–carbon bonds is essential. Abe et al.92 reported a number Fig. 8 Structures of additives for carbon anodes containing sultone,
of olefinic compounds as electrolyte additives such as vinyl sulfite, cyano, vinylene, halogen, and Si functional groups.
acetate (VA, 18), divinyl adipate (DVA, 19) and allyl methyl
carbonate (AMC, 20) as being able to improve the performance
of lithium ion batteries in PC based solvent systems. They tested reduction of BDMIm BFMB. They have also reported that the
the effect of these additives using natural graphite (NG), graphitized structure of the cation has a great influence on the performance
mesophase carbon microbeads (MCMB 6–28), and graphitized of the additive. Ionic liquids are good candidates not only to
mesophase carbon fiber (MCF), with different crystallinities. The improve the poor performance of PC based electrolytes for
effect of the additives on the performance of the battery depends graphite electrodes, but also to improve the safety of the organic
on the type of graphitic carbon used. This indicates that the degree electrolytes as they are non-volatile and non-flammable.
of graphitization has also a significant influence on the SEI Sulfur-containing compounds are believed to be reduced
formation process. Of all the additives they have tested, VA showed prior to the electrolytes to form a passivating film on the
the best performance in all the three graphitic carbon electrodes. electrode’s surface which inhibits PC co-intercalation.87 Some
Matsuo et al.64 used a series of butyrolactone derivatives of the additives containing sulfur are ethylene sulfite (ES, 33),57
(21–29) as additives to prevent the exfoliation of graphite and propylene sulfite (PyS, 34)58 1,3-propane sultone (PS, 35),96
PC co-intercalation. The suppression of PC co-intercalation and butyl sultone (BS, 36),97 vinyl ethylene sulfite (VES, 37),87 and
decomposition was due to a decrease in the number of PC prop-1-ene-1,3-sultone (PES, 38) (Fig. 8).98 Electrochemical
molecules coordinated to the lithium ion and an increase in the reduction of these sulfur containing additives leads to the
additive coordination. The reduction behavior of additives formation of Li2SO3 and (RSO3Li)2. Density functional theory
during the first discharge is dependent on the concentration (DFT) calculations by Leggesse et al.56 revealed the formation
of additives bound to lithium ions. Their results proved that the mechanism of these species through radical termination reactions
solvation behavior of the additive greatly affects the performance of by reacting with other species which are involved in the initial
the battery and can be considered as one parameter during reduction process, further reducing the radical by gaining an
electrolyte additive selection. Recently, Khasanov et al.93 reported electron from the polarized electrode or self-dimerization as shown
that graphite electrodes can be cycled in PC based electrolytes with in Scheme 2. Zhang et al.99 showed that the presence of these
the addition of 5% (4R,5R)-dimethyl 2-oxo-1,3-dioxolane-4,5- lithium alkyl sulphonate species in the SEI enhances the ionic
dicarboxylate (ODC, 30) with a high first cycle columbic efficiency conductivity of the SEI.
(93%). The additive also showed good cycling performance in Additives containing electron withdrawing groups such as
graphite/LiNi0.65Co0.2Mn0.15O2 (NMC) full cells. –CN, Cl, and F make the compound more electrophilic and
Similarly, PC decomposition and co-intercalation can also facilitate reduction. If the additive contains a vinyl group,
be prevented by adding ionic liquids that can compete with reduction of the additive may lead to the formation of an SEI
lithium ions in coordinating with PC. When 20% of PC is mixed with a polymeric structure in which polymerization is initiated
with the quaternary ammonium ionic liquid, trimethylhexyl- by the vinyl group. Yong et al.100 showed that low concentration
ammonium bis(trifluoromethane sulfonyl)imide (TMHA TFSI),94 of allyl cyanide (AC, 39) allows the use of a PC based electrolyte
there is a competition between lithium ions and TMHA cations for the graphite electrode by forming a polymeric SEI on the
which reduces the solvation of lithium ions by PC molecules and electrode’s surface. The polymerization is initiated by the
limits the amount of PC intercalation. This phenomenon also double bond of AC, followed by a chain reaction. Similarly, in
minimizes PC reduction on the electrode’s surface and leads to acrylic acid nitrile (AAN), which contains a mixture of vinyl groups
the formation of an effective SEI film. Sun et al., showed that and electron-withdrawing groups, the electro-polymerization
when 1-butyl-2,3-dimethylimidazolium bis(fluoromalonato)- reaction is initiated by radical anions formed by the reduction
borate (BDMIm BFMB, 5%)95 is added to a PC based electrolyte, of vinyl groups.90 When ethyl isocyanate (EtNCO, 40) is used as
PC co-intercalation and exfoliation of the graphite electrode are an additive, its reduction leads to the formation of a polymeric
prevented, due to the passivation of the graphite surface by the film which is initiated by the reaction of the isocyanate group.101

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1961
 View Article Online

Review Energy & Environmental Science

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Scheme 2 Mechanism of the reductive dissociation of Li + (PS) and different termination paths for the radical anion, adapted from ref. 56.

3-Fluoro-1,3-propane sultone (FPS, 41), which is a modification capacity of the cell can be increased by B10% in both cases,
of PS made by introducing an electron withdrawing fluorine and that the cyclability of graphite electrodes can also be
group, showed better performance compared with PS and VC in improved. Tetrachloroethylene (TCE, 46) is another electrolyte
a LiCoO2/graphite cell.102 The presence of the fluorine group additive reported by Hu et al.88 for PC based electrolytes, who
enhances the anodic stability and cathodic reactivity of the showed that 3% of the additive can suppress the co-intercalation
additive which helps to improve its SEI forming ability. of PC while improving cell performance. They suggested that the
Recently, compounds containing halogen functional groups decomposition products of TCE form an effective SEI on the
have attracted much attention in the field of rechargeable graphite and also allow the use of 4 V cathodes since the
batteries to improve their electrochemical performance. Surface oxidation potential of the electrolyte is as high as 4.5 V vs. Li/Li+.
modification (anode/cathode materials) with halogens, for The use of an additive with both an aromatic ring and
instance fluorine, has been suggested to improve the electro- fluorine leads to the formation of a stable SEI and prevents the
chemical characteristics of the electrode. It is reported that co-intercalation of PC and graphite exfoliation. 4-Fluorophenyl
introduction of fluorine into the surface of a cathode-active acetate (4-FPA, 47) showed good cycling performance in a
material increases the electronic conductivity of the surface, LiFePO4/graphite cell in a PC based electrolyte by forming a
lowers the oxidative degradation of the electrolytes, stabilizes stable SEI on the graphite surface.105 4-FPA decomposes to
the crystal structure during cycling, protects the surface from HF lithium acetate and a radical on the aromatic ring which leads
attack so that solvation/desolvation of the lithium ion proceeds to the formation of a dimer by combining with another radical
smoothly.103 Halogen containing compounds have also been group, or forms a polymer through radical termination by other
reported as electrolyte solvents and electrolyte additives for intermediates/solvents as shown in Scheme 3. The resulting
lithium ion batteries.62,63,88,104 McMillan et al.62,104 reported surface film is stable because of the incorporation of fluorine
the use of chloro-ethylene carbonate (Cl-EC, 42) and fluoro- together with the aromatic ring in the SEI.
ethylene carbonate (FEC, 43) as electrolyte solvents with other Additives containing the aromatic ring and silane functionality
co-solvents such as PC and/or EC. They showed that these also have the ability to suppress the co-intercalation of PC and
electrolyte solvents increase the cycle life of the battery and exfoliation of graphite. Methyl phenyl bis-methoxydiethoxysilane
decrease its irreversible capacity; however, FEC shows a better (MPBMDS, 48) improves the performance of the battery in PC
cell efficiency than Cl-EC. Naji et al.63 reported a-bromo-g- based electrolytes by forming a stable network of Si–O–Si, C–O–Si,
butyrolactone (Br-BL, 44) and methyl chloroformate (Me-ClF, 45) Si–C and Si–O on the graphite surface.106 Moreover, MPBMDS
as electrolyte additives for PC based electrolytes by directly effectively inhibits flame propagation at a loading level of
adding low concentrations of the additives to the electrolyte or 4–13 wt%. Fig. 9 summarizes some of the additives used for
by impregnation of the graphite electrode with halogenated PC based electrolytes on carbon electrodes and the resulting
solvents prior to cycling. They showed that the reversible first cycle efficiency obtained. However, direct comparison of

1962 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

on the carbon surface and blocking the electro-reduction of

dissolved Mn(II) on the graphite electrode.109 Felix et al.110
reported that 1,3-propanediolcyclic sulfate (PCS, 50) has the
ability to suppress the deposition of Mn on the anode side and
minimize capacity fading when a LiNi0.5Mn1.5O4/MCMB full
cell was used.

2.2. Silicon (Si) and tin (Sn) anodes

The high energy application of lithium ion batteries is limited
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Scheme 3 Reduction mechanism of 4-FPA to form a dimer with another by the capacity of the carbon anode. Si and Sn-based electrodes
radical or a polymer through radical termination by other intermediates/ are promising negative electrodes that are able to replace
solvents. graphite electrodes since they have high theoretical capacities
(nearly 4200 mA h g1 for Si and 996 mA h g1 for Sn) compared
to the widely used carbon-based electrodes (372 mA h g1).
However, Si and Sn have the problem of large volume changes
(upto 300%) during charging and discharging. This large
volume change leads to the disintegration of the active material
and the loss of electrical conduction paths, thereby reducing
the cycling life of the battery. Additionally, there is a problem of
continuous SEI-film formation because of repeated surface
pulverization upon cycling and the subsequent decomposition
of the electrolyte on the renewed fresh Si and Sn surfaces. To
avoid the stress-induced fracture of the SEI layer and the
resulting continuous SEI growth and formation, the mechanical
properties of the SEI (e.g. flexibility upon expansion and contraction
during cycling) are crucial. The other problem related to the Si
anode is the reaction of the electrolytes with the components of the
native layer (silicon oxide (Si–O–Si) and silanol (Si–OH)), which are
Fig. 9 First cycle columbic efficiency of selected additives, used for PC formed due to the reaction of Si with O2 or H2O.
based electrolytes, on carbon electrodes.
One approach to solve this problem is the use of electrolyte
additives. FEC111 and VC112 are the most commonly used
additives for both Si and Sn based electrodes. With Si electrodes,
the effect of the additives reported by different research groups
FEC is used as a co-solvent.113 When the Si anode is cycled with a
is difficult since the effect of the additives is also dependent on
FEC free electrolyte, a less dense SEI containing –Si–C and –Si–O
the working conditions. Although there have been many efforts
functional groups is formed. Since –Si–C and –Si–O bonding is
directed towards developing electrolyte additives for PC based
not too strong, further decomposition of the SEI occurs during
electrolytes, their practical application has not been achieved.
cycling which leads to continuous SEI formation. The addition of
Hence, research directions should target the practical applicability
low concentration of FEC as a co-solvent can suppress the
of the additives.
continuous SEI formation by forming –Si–F and LiF species
When carbon materials are used as anodes with spinel (LiMn2O4)
which have strong bonding energy.114 Moreover, FEC forms
cathodes, the carbon anode is degraded by manganese dissolution
polymeric species by generating HF to form VC and polymerizes
into the electrolyte. Sodium salts have been reported as the best
through the double bond as shown in Scheme 4.113 Borde
candidates to overcome this problem. Even though the contribution
et al.115 showed that the content of FEC has a great influence
of the sodium-containing additives has not been well studied,
separate studies using NaCl107 and NaClO4108 showed that the
presence of sodium salts helps to improve the kinetics rate and
characteristics of the film by forming a homogeneous, uniform
and smooth SEI layer on the graphite surface. Other inorganic
additives such as LiI, LiBr and NH4I have also been reported
as good additives to suppress the degradation of carbon anodes
by manganese dissolution.45 Manganese dissolution can be
eliminated by adsorption of iodide or bromide anions on the
Mn surface or by the formation of an amine complex with Mn(II)
in the case of NH4I. Moreover, organic compounds such as
2-vinylpyridine (VP, 49) also have the ability to prevent the Scheme 4 Reduction mechanism of FEC which leads to the formation of
degradation of the carbon anode by forming a polymeric film polymeric species adapted form ref. 113.

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1963
 View Article Online

Review Energy & Environmental Science

on the cycling performance of the battery. They used 0, 5, 10, and

20 wt% FEC in a 1 M LiPF6/EC : DEC (1 : 1 wt%) electrolyte and
tested the battery using a Si-graphene/NCA full cell. They obtained
good performance when 5% FEC was used; however, when the
amount of FEC was increased, the formation of LiF increased by
consuming lithium ions leading to a higher irreversible capacity.
In VC containing electrolytes, the presence of a double bond
makes VC more susceptible to reduction which occurs at a
higher potential than the carbonate solvents. VC reduction
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

leads to the formation of polycarbonates covering the Si surface

that prevent the continuous growth of the SEI film.116 Recently,
Hy et al.117 reported that a combination of atomic layer deposition Fig. 10 First cycle and end of cycle columbic efficiency of selected
coating (ALD) on a Si/rGO/CNT composite anode with VC as an additives used for the silicon electrode.
electrolyte additive dramatically enhanced the battery cycling
stability. Their results revealed a stable SEI having a flexible
Generally speaking, an effective electrolyte additive for anode
web like network formed on the electrode’s surface. The ALD-VC
materials should be able to be reduced prior to the electrolyte
combination helps to suppress the formation of SiOxFy species
and form a passivating film. In the case of Si and Sn electrodes,
by preventing the reaction between the electrode’s surface and
due to the problem of expansion of the electrode’s volume, the
the lithium salt. Another additive reported to improve the
SEI formed by the additives should have good flexibility and
performance of Si anodes is succinic anhydride (SA, 51).118 The
mechanical strength. Moreover, the additives should prevent the
presence of low concentrations of this additive helps to enhance
continuous growth of the SEI. Some of the major properties of
the formation of Li2CO3 and hydrocarbons which prevent the
anode additives are summarized in Scheme 5.
decomposition of LiPF6. The battery with SA shows better
capacity retention and cycling performance than a comparable
one made with a SA free electrolyte. 2.3. Li4Ti5O12 (LTO) anodes
A comparative study using VC and FEC for the Si anode The safety issues affecting state-of-the-art lithium ion batteries,
shows that 5% VC has better cycling performance than 3% due to the use of graphite anodes, can be addressed by the use
FEC.119 However, recent findings showed that when the of spinel lithium titanate (Li4Ti5O12) (LTO) anode materials.
concentration of FEC is increased to 10% and 15%, the battery LTO has an operating potential at about 1.55 V and a capacity of
has a lower impedance and better capacity than one containing 175 mA h g1. Its relatively high operating potential, structural
3–6%VC.120 Using a mixture of FEC and VC as an additive also stability upon cycling and good lithium ion reversibility make it
has a synergetic effect on the performance of the battery. Since an ideal candidate for ‘next generation’ lithium ion battery
the polycarbonates formed by VC and FEC reduction are stable technology. Since the operating potential is above the potential
at elevated temperatures, they can improve the battery’s per- of lithium deposition and reduction potential of the electrolytes,
formance at high temperatures. Moreover, when FEC and VC the electroplating of lithium and generation of the SEI can be
are thermally decomposed, they form additional SEIs on the avoided. This shows the excellent safety advantage of LTO over
electrode’s surface. These secondary SEIs together with the graphitic anodes for the large-scale and long-life battery applica-
primarily formed SEI, polycarbonates, prevents the battery tions. However, there are issues that still need to be solved.
from undergoing thermal runaway.121 Similarly, pulverization One of LTO’s problems is its poor rate capability due to the
of the Sn electrode during cycling can be prevented by the VC insulating character which arises because of the large band gap
reduction products forming a flexible SEI layer on the Sn of Ti 3d. This problem can be suppressed by surface modification,
surface. The Sn surface can be passivated by the formation of doping or modifying the synthesis method. Various methods have
poly(vinylene carbonate) species which suppresses additional been reported as being able to improve the rate capability of LTO,
electrolyte decomposition. This results in the formation of thin see the review by Yi et al.123 The second problem is the electrode/
SEI and reduces the polarization of the electrode caused by electrolyte interfacial reaction. Even though LTO has been con-
slow lithium ion diffusion through the thick SEI in the VC-free sidered as a SEI free anode material due to its high operating
electrolyte.122 The first cycle and end of cycle columbic potential, there is a possibility of SEI formation due to indirect
efficiency obtained with various additives on the Si electrode decomposition of the electrolytes.
is summarized in Fig. 10. Despite their great potential to One reason for the formation of SEI on LTO is the reaction
replace graphite electrodes, much work still needs to be done between LTO and the electrolytes, which occurs upon cycling
to address problems associated with using Sn and Si electrodes and leads to the generation of gases. The generated gases
with electrolyte additives. Most studies focus on the modifica- further react with the electrolytes resulting in the degradation
tion of the electrode material itself to suppress SEI growth and of the electrolyte. He et al. showed that the interfacial reaction
volume changes. A synergy between both additive and Si (Sn) is dependent on the morphology of the electrode.124 They
modification approaches is essential to realize Si (Sn) anode’s showed this difference by synthesizing LTO with various
applications in lithium ion batteries. morphologies using lithium acetate (LA) and lithium hydroxide

1964 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Scheme 5 Summary of the properties of the additive for anode materials.

(LH) as the lithium sources. LTO–LH has a large surface area

and reacts with the electrolyte to rapidly form a smooth SEI.
However, the small surface area of LTO–LA leads to the formation
of porous SEI film after prolonged cycling. Similarly, Song et al.
showed that the formation of a SEI on the LTO surface is due to the
reaction between the electrolyte and the electrode.125 They used
LTO with and without a carbon additive and cycled the cell both at
Fig. 11 Structures of additives for LTO anodes containing sulfonate,
high and room temperatures. They observed the formation of the
cyanate and anhydride groups.
SEI on both electrodes; however, the presence of carbon black
accelerated the reaction at high temperatures. Dedryvere et al.
reported that in the LiMn1.6Ni0.4O4/Li4Ti5O12 full cell configuration generation of gases in LTO and the subsequent decomposition
organic and inorganic species were deposited at the surface of of electrolytes.129 MMDS decomposes to Li2SO3 and ROSO2Li
LTO.126 They argued that the soluble species from the SEI on the and participates in the formation of the stable SEI. The
positive electrode diffuses or migrates towards LTO. However, some presence of the SO3 group in MMDS and its decomposition
studies have revealed that SEI formation may occur due to the products act as the weak base sites to deactivate the reaction of
reaction between the electrolyte and LTO. LTO was swollen upon PF5 with organic solvents.
storage due to the generation of gases.127 Bouayad et al. reported the effect of glutaric anhydride
Decomposition of electrolytes on the LTO surface can also (GA, 54) as an additive on the LiNi0.4Mn1.6O4 (LNM)/LTO cell.130
be accelerated by the presence of impurities, such as water, The cycling performance of the LNM/LTO cell was improved in
which lead to the generation of HF and H2. On the other hand, the presence of the GA additive. They showed that GA forms
thermal decomposition of LiPF6 leads to the formation of PF5 organic species on the surface of both electrodes and hinders
and LiF. The resulting PF5 deteriorates the SEI stability by the interfacial reaction between the electrodes and the electro-
further reacting with the electrolyte solvents and SEI compo- lytes. It also significantly prevents the self-discharge process.
nents. This results in the generation of CO and CO2 gases.127 In Kim et al. used a mixture of VC and PS additives for ionic liquid
the presence of water impurities, LiF generates HF which in based electrolytes in a LTO/LiCoO2 cell. Both VC and PS form a
turn degrades the electrode surface. When LTO is attacked by stable SEI on the surface of electrodes and suppress the
HF, catalytic active sites will be created and these sites initiate reductive decomposition of the ionic liquid on LTO. However, the
the decomposition of electrolytes. At room temperature gen- additives do not affect SEI formation on the cathode’s surface.131
eration of gases such as CO, CO2 and H2 could also occur in
LTO upon storage due to the reaction between LTO and the
electrolytes.127 The generated gases react with the organic 3. Cathode additives
solvents and degrade the electrolyte. Hence, using an electro-
lyte additive is crucial to prevent the interfacial reaction and the A key factor that limits the performance of lithium ion batteries
effect of impurities. Wang et al.128 used p-toluenesulfonyl is the degradation of cathode materials which can be caused by
isocyanate (PTSI, 52) as a SEI film forming additive to suppress transition metal dissolution, interfacial reaction, loss of contact
the decomposition of electrolytes on LTO (Fig. 11). The reduction with conductive particles, and so on. Although all the transition
of PTSI forms a less resistive SEI on the LTO surface and hinders metals in the cathode (such as Mn, Ni, Co, Fe, and Zn) are
the corrosion of the electrode by HF. Moreover, PTSI acts as weak susceptible to dissolution, manganese showed the largest
base and suppresses the reaction between PF5 and electrolyte amount of dissolution. This is a major reason for capacity fade
solvents. Methylene methanedisulfonate (MMDS, 53) has in spinel electrodes and other cathodes containing Mn. Metal
also been reported as an electrolyte additive to inhibit the dissolution leads to structural disordering of the cathode and

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1965
 View Article Online

Review Energy & Environmental Science

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Scheme 6 Key issues and challenges facing commonly used cathode materials in lithium ion batteries.

also increased the growth and breakdown of the SEI layer on Cobalt dissolution is accompanied by the release of lithium
the anode. The main challenges that commonly used cathode and oxygen, resulting in structural degradation.
materials face are summarized in Scheme 6. One way of solving Abe et al.133 reported organic additives having lower oxidation
these problems is the use of additives that can form an effective potential than the organic electrolytes to improve the cycle
surface film on the cathode. Film forming cathode additives are performance of LiCoO2 at high voltage by forming a conductive
expected to oxidize prior to the solvents and cover the electrode surface film and suppress electrolyte decomposition. These
surface to prevent the oxidative decomposition of the electrolytes. additives are biphenyl (BP (55), 4.44 V), o-terphenyl (OT (56),
Even though most common organic solvents such as EC and DEC 4.45 V), and m-terphenyl (MT (57), 4.25 V) (Fig. 12). All these
are stable up to 6 V, the presence of active sites on the cathode additives improve the cell performance; however, the amount of
surface enables them to be oxidized at a lower potential. Hence, the these additives should be kept to the minimum (below 0.2%) to
additives should decompose at a lower potential than the solvents form a less resistive and thinner SEI film. Thiophene derivatives
and cover the active sites to prevent the indirect oxidation of the (2,2 0 -bithiophene (2TH, 58) and 2,2 0 :5 0 ,200 -terthiophene (3TH,
electrolytes. Various additives used for film-forming for different 59)) have also been reported as potential additives to suppress
cathode materials are discussed below. the oxidative decomposition of the electrolytes when LiCoO2 is
used for high voltage applications.134 They screened these
additives by calculating their highest occupied molecular orbital
3.1. LiCoO2 cathodes (HOMO) and LUMO energies, and by experimentally testing their
Most commercial lithium ion batteries have used LiCoO2 as
the positive electrode since the first lithium ion battery was
commercialized by Sony in 1991. LiCoO2 is one of the positive
electrodes having a layered structure that offers a high energy
density, however, it has its own drawbacks. (i) One of the
problems of LiCoO2 is safety since it is thermally unsafe.
Layered cathode materials undergo exothermic reaction with
very large enthalpy (941 J g1) and the reaction is completed at
much lower temperature.132 For better utilization of the battery,
the positive electrodes should bechemically stable and cost
effective, and should have safe high capacity, good rate cap-
ability, and low toxicity. The other problem associated with the
performance of LiCoO2 is capacity fading caused by unwanted
side reactions between the electrode and electrolyte, especially
when overcharged, which leads to the consumption of lithium
ions. The third problem is the dissolution of Co into the solution Fig. 12 Structures of additives for the LiCoO2 cathode containing phenyl,
when LiCoO2 is exposed to a potential higher than 4.2 V. thiophene aniline and maleimide groups.

1966 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

oxidation potentials. These additives have lower oxidation

potentials than the electrolyte solvents (3.95 V for 2TH and
3.8 V for 3TH) and the additives reported by Abe et al.133 The
additives decompose and polymerize on the electrode’s surface
prior to the electrolyte to form a conducting protective layer and
prevent electrolyte decomposition. Hence, they can improve the
cycling stability and performance of LiCoO2 when cycled at
high voltage (4.4 V vs. Li/Li+). Increasing the thiophene unit
decreases the oxidation potential of the additive by increasing
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

the energy density on the additive and eases the electro-oxidative

polymerization of thiophene. Other compounds having good Scheme 7 Oxidation of OT to form a polymeric film by polymerizing
electron donating ability, such as N-(triphenylphosphoranylidene)- through the o-position of benzene rings.

aniline (TPPA, 60),135 decompose prior to electrolyte decom-

position to form a protective film. polymerizing through the o-position of benzene rings of mono-
A similar effect has been reported for a LiCoO2 electrode mers as shown in Scheme 7. These films prevent the decom-
using N,N 0 -4,4 0 -diphenylmethane-bismaleimide (BMI, 61)136 and position of the electrolytes by making the electrode active sites
2,2 0 -bis[4-(4-maleimidophenoxy)phenyl]propane (BMP, 62)137 inert, thereby improving the electrochemical performance of the
additives. These additives have enabled LiCoO2 to be cycled battery. Inorganic additives such as Li2CO3140 and nano-Al2O3141
up to 4.5 V (vs. Li/Li+) where BMP is more compatible with the
have also been reported as cathode additives that improve the
electrolyte than BMI. They form a protective film on the
high voltage performance of LiCoO2. These additives are soaked
cathode’s surface to prevent electrolyte decomposition and
into a commercial electrolyte and removed prior to battery
protect the crystal structure of LiCoO2 from phase transition assembling.
at high voltage. A full cell test (LiCoO2/MCMB cell) at high Even though the rate capability of the battery is related to
temperature (55 degree centigrade (1C)) using a maleimide- the lithium ion movement inside the electrode material, the
based self-polymerizing branched oligomer additive in a LiPF6 interface between the electrode and electrolyte has also its own
based electrolyte showed that the additive effectively improves contribution. An anion receptor tris(pentafluorophenyl)borane
the cycling performance with concentrations as low as 1%.138
(TPFPB, 75) has the ability to facilitate lithium ion transfer
Takeuchi et al.139 proposed 10 organic compounds as potential
through the electrode/electrolyte interface and improve the rate
additives having oxidation potential between 3.4 and 4.7 V vs.
capability of the battery.142 TPFPB can improve the lithium
Li/Li+ which is lower than the electrolyte solvents to improve salts’ dissociation ratio, and restrict anion migration by coupling
the performance of LiCoO2 at high voltages (Fig. 13). These with the anion. Polymerizable additives such as TH, 3,4-ethylene-
compounds are o-terphenyl (OT), Ph–X (CH3)n (n = 1 or 2, X = N, dioxythiophene (EDOTH, 76) and BP can also improve the rate
O or S, i.e. N 0 ,N-dimethyl-aniline (DMA, 63), anisole (An, 64), performance of LiCoO2 in gel polymer electrolytes by reducing
thioanisole (TA, 65), respectively), adamantyl toluene (AT i.e. the interfacial resistance in the cell during cycling.67
ortho, meta and para, 66–68), furans (Fu, 69), i.e. 2-methyl furan
(MFu, 70) and dimethyl furan (DMFu, 71) and thiophenes
(TH, 72), i.e. 2-methyl thiophene (MTH, 73) and dimethyl 3.2. Lithium nickel manganese cobalt oxide (NMC) cathodes
thiophene (DMTH, 74). They showed the effect of these additives Thermal instability and loss of oxygen from the lattice when
by taking OT as an example. OT forms a polymeric film by x o 0.5, causing limited capacity (140 mA h g1), are among the
problems in Li1xCoO2 systems. The layered LiNiyMnyCo12yO2
(NMC) systems have received great attention as 4 V-electrode
materials to replace LiCoO2 in lithium ion batteries. NMC
electrodes have better stability during cycling, better thermal
stability, higher reversible capacity (160 mA h g1) and are more
stable, i.e. up to 4.6 V than LiCoO2.143 However, the electrolyte
tends to oxidize on the surface of the cathode since it is
unstable in this potential range. The other problem related to
NMC is the dissolution of the transition metal together with the
generation of oxygen gas which causes structural instability.
Additives are often used to stabilize the cathode/electrolyte
interface and prevent metal dissolution into the electrolyte.
A series of polyfluoroalkyl (PFA) substituted ethylene carbonate
compounds have been reported as electrolyte additives to improve
the performance of high voltage lithium ion batteries.144 These
Fig. 13 Structures of additives for the LiCoO2 cathode containing aniline, compounds contain reactive head-groups which can be oxidized or
anisole, adamantly, furan, thiophenes, and borane groups. reduced and attach onto the electrode’s surfaces as an inner layer

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1967
 View Article Online

Review Energy & Environmental Science

on the effect of TMOBX on the various positive electrodes shows

that while TMOBX reduces the positive electrode impedance,
it increases the rate of electrolyte oxidation and/or shuttle-
type reactions at the positive electrode, especially at elevated
temperatures.147 Nitrile electrolyte additives, such as succino-
nitrile (SN, 82), improve the cycling performance of NMC/graphite
cells stored at 60  0.1 1C.148 SN suppresses the reduction of
transition metal ions and subsequent electrolyte decomposition
by forming complexes with the surface metal. However, its use
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

increases the cell impedance significantly. The authors suggested

that the large impedance can be suppressed by the addition of
ES and tris(trimethlysilyl)phosphite (TTSPi, 83) or trimethylene
sulfate (TMS, 84) and TTSPi. Hence, the use of a combination of
additives is an important approach aimed at overcoming pro-
Fig. 14 Structures of additives for the NMC cathode containing halogen, blems caused by other additives or providing multifunctions.
boron, nitrogen, phosphorous, sulfur, and silicone functional groups. However, when mixing additives, it is imperative to understand
clearly the properties of the additives and of the incorporated
with a polyfluoroalkyl chain that self-assembles on this inner functional groups.
layer as a solvophobic layer (outer layer) to prevent electrolyte Lithium bis(oxalate)borate (LiBOB, 85) is another electrolyte
decomposition. The substituents are 4-(trifluoromethyl) (TFM, additive that can be used to improve the electrochemical
77), 4-(perfluorobutyl) (PFB, 78), 4-(perfluorohexyl) (PFH, 79), performance of the battery (Fig. 15).82,149 LiBOB forms a more
and 4-(perfluorooctyl) (PFO, 80) (Fig. 14). Among these, 4-(per- stable SEI on graphite, inhibits the detrimental reactions of the
fluorooctyl) substituted ethylene carbonate (PFO-EC) shows electrolyte with high voltage positive electrodes, and also
best cell performance and reduces the cell impedance. This improves capacity retention; however, it has a problem of increasing
additive has a lower oxidation potential than the electrolytes the cell’s impedance. This problem of LiBOB can be overcome
and is sacrificially oxidized on the positive electrode to prevent by combining it with either triphenylamine (Ph3N, 86) or
electrolyte decomposition. Moreover, the presence of long PFA 1,4-benzodiozane-6,7-diol (BDOD, 87).150 Ph3N and BDOD have
chain creates a more solvophobic barrier and protects the a lower oxidation potential and protect the positive electrode
electrode surfaces. surface by forming electronically conducting polymers through
Glyme compounds, CH3O(CH2CH2O)nCH3 (Gn, with n = 2, electrochemically initiated radical polymerization processes.
3 or 4), are interesting additives that have the ability to complex Ph3N can generate three-dimensional polymers while BDOD
lithium salts in solution through strong solvation to the lithium can only form linear polymers. Mixing lithium difluoro(oxalato)-
ion.145 The presence of high donor numbers and relatively borate (LiDFOB, 88) with LiBOB also improves the cell performance
strong Lewis basicity make glymes good chelating agents and lowers the impedance at the positive electrode. Oxidation
towards lithium salts, e.g. LiF and Li2O, which precipitate products of LiDFOB passivate the electrode surface preventing
during SEI formation. Since they have high oxidation potentials, electrolyte decomposition and reducing transition metal disso-
they can be used in batteries with NMC electrodes. Glymes’ lution from the oxide.151
chain lengths greatly affect the performance of the battery. As Sulfur-containing and phosphorous containing electrolyte
the chain lengths increase the capacity decreases due to strong additives are also known to increase battery lifetimes, control
complexation of free lithium ions by glyme molecules which
decreases their availability for the insertion/de-insertion process.
The additive used to improve the specific property of the
battery may have an adverse influence on other properties.
Another way of enhancing the performance of the battery,
especially in the full cell configuration, is the use of a combi-
nation of additives. A mixture of additives can potentially have
a cooperative effect on the performance of the battery at room
or elevated temperatures. When ES and VC, known SEI film
forming additives on graphite, are used separately in a NMC/
graphite cell, ES increases the impedance on the positive
electrode and VC increases the impedance on the negative
electrode. However, when a mixture of ES and VC is used, the
impedance both on the negative and positive electrodes is
reduced due to the synergetic effect of the two additives. The
cell impedance caused by VC can also be reduced by adding low Fig. 15 Structures of additives for the NMC cathode containing fluorine,
concentration of trimethoxyboroxine (TMOBX, 81).146 A study boron, sulfur, nitrogen, phenyl and oxalate functional groups.

1968 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

impedance growth and suppress gas production, especially

when mixed with other additives. Using 1,3,2-dioxathiolane-
2,2-dioxide (DTD, 89), and TMS additives can give better
coulombic efficiency with a lower cell impedance and a reduced
voltage drop during storage.152 However, during SEI formation
these additives show significant gas generation. When these
additives are combined with VC, gas generation in DTD can be
reduced while in TMS can totally be suppressed and the
positive effects of the additives can be improved. PES improves
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

the performance of NMC/graphite cells with less gas produc-

tion. When it is blended with VC, a better cell performance can Fig. 16 Structures of additives for the Li-rich cathode containing phos-
be obtained than with the single additive but with an increase phorus derivatives, fluorine, boron, silicon, and phenyl functional groups.
in the cell’s impedance.153 However, a ternary mixture of PES,
MMDS (53), and TTSPi or VC, MMDS, and TTSPi can yield
calendar life improvements and excellent safety.105,154 MMDS Similarly, the boron and phosphorus functional group con-
and TTSPi are good additives for lowering the cell’s impedance taining additives, namely tris(trimethylsilyl)borate (TMSB, 91)
when used as a single additive and also improving coulombic and tris(trimethylsilyl)phosphate (TMSP, 92), have been found
efficiency by reducing electrolyte oxidation on the positive to improve the rate capability and cycling performance of
electrode.155,156 Mai et al.157 showed that TTSPi alone can Li-rich cathode materials by decomposing prior to the solvent
improve the cycling performance and rate capability of a NMC (at 3.76 and 4.1 V, respectively) to stabilize the electrode/
electrode by being preferentially oxidized on the cathode’s surface electrolyte interface.163,164 Other phosphorus derivatives such
to form a stable and less resistive surface film. It also prevents as triphenyl phosphine (TPP, 93), ethyl diphenylphosphinite
electrolyte decomposition and dissolution of the cathode. (EDP, 94), and triethyl phosphite (TEP, 95) can reduce the
internal pressure at the end of charging by reacting with O2
3.3. Li-rich cathodes generated during charging (Scheme 9).165 Their reaction with
Lithium rich cathodes are promising economical positive electrode O2 yields phosphine oxide compounds, by forming a double
materials as they have high potentials (44.5 V), large capacities bond between the phosphorus and the oxygen atom. The
(4250 mA h g1), good thermal stabilities. However, they have absorption of oxygen and reduction of pressure are dependent
poor cycle performance because of the instability of the electrode on the molarity, steric effect and anodic stability of the additive
with the electrolytes, metal dissolution and structural change with used. Among the three additives, the highest degree of pressure
the generation of O2 gas which leads to the formation of surface reduction can be obtained using TEP due to its low molecular
films such as Li2O.158 PS is a well-known additive that improves the weight and smaller steric hindrance. However, its ratio of
long term cycling performance of Li-rich cathode materials.159 pressure reduction to molarity is lower than EDP. This is
It forms a stable surface film on the electrode preventing metal because TEP has lower anodic stability (i.e., it decomposes
dissolution, from the bulk material, and electrode structural before 4.5 V) than EDP and this lowers its oxygen absorption ability.
changes.160 1,3,5-Trihydroxybenzene (THB, 96) exhibited a useful stability profile
Fluorine and phosphorus containing electrolyte additives by decomposing prior to the electrolyte (at 4.2 V) and passivat-
are promising agents able to improve the performance of high ing the over-lithiated cathode’s surface.166 The resulting film is
voltage cathode materials. Recently, fluorinated electrolytes both thermally and electrochemically stable and improves the
have been introduced to enable the battery to be cycled up to columbic efficiency and cycling performance or storage at
5 V.161 The presence of the fluorine functional group improves elevated temperature. It also prevents Mn deposition on the
the anodic stability of the electrolyte and yields better cycling graphite electrode.
performance than the conventional electrolytes. The introduc- The efficiency of the graphite/Li1.10Mn0.43Co0.23Ni0.23O2 cell
tion of low concentrations of fluorine containing compounds stored at 60 1C in the presence of THB remained 499.9% after
as additives rather than a solvent is a cost effective strategy. 200 cycles. Although additives with various functionalities have
Di-(2,2,2 trifluoroethyl)carbonate (DFDEC, 90) has the ability been reported, further efforts are needed to improve the long
to reduce voltage fade and maintain the structural stability of cycling performance of the lithium excess cathode materials for
Li-rich cathode materials (Fig. 16).162 It can also improve the practical applications.
cycling performance and rate capability of the cathode material
by forming a less resistive and stable SEI at the interface that 3.4. Spinel cathodes: LiMn2O4
contains –CF3 and –CF2– functionalities. These species are Spinel lithium manganese oxide (LiMn2O4) is a promising
formed when an electron is transferred from either oxygen or cathode material for lithium ion batteries because of its high
carbon of C–O of DFDEC to the metal cation. The resulting operating voltage, low cost and environmental friendliness.
product further reacts with EC to form a stable SEI layer and However, LiMn2O4 has problems related to capacity fading
effectively passivates the cathode’s surface as shown in and poor cycling, especially at elevated temperatures. Some of
Scheme 8. the reasons for capacity fading are: (i) its instability with the

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1969
 View Article Online

Review Energy & Environmental Science

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Scheme 8 Electrochemical oxidative decomposition of DFDEC and its further reaction with EC to form a stable SEI, adapted from ref. 162.

Scheme 10 Reaction mechanism of hexamethyldisilazane (10) and hep-

tamethyldisilazane (11) with H2O and HF, respectively, to protect the
cathode from acid impurities.

However, heptamethyldisilazane, which is insoluble in water,

Scheme 9 Reaction mechanism of phosphorus derivatives with oxygen
generated during charging. reduces the amount of HF in the electrolyte by reacting with the
acid to form salts as shown in Scheme 10, eqn (11).168 These
additives improve the cycling performance and anodic stability
carbonate electrolytes, since manganese tends to dissolve in of the electrolytes both at room temperature and after storage at
electrolytes and moves to the anode to deteriorate the SEI, (ii) 60 1C for 7 days. Since ethanolamine compounds can stabilize
the instability of the delithiated spinel structure by oxygen loss H2O and HF through hydrogen bonding, mixing this additive
in organic electrolyte solvents, (iii) disproportionation of Mn3+ with heptamethyldisilazane can improve the storage perfor-
into Mn2+ and Mn4+ by HF attack, which is formed from the mance of the battery.169
decomposition of the LiPF6 salt, and (iv) Jahn–Teller distortion Similarly, when additives with silane and phosphorus or
at the end of discharge. The addition of electrolyte additives boron functional groups are used, such as TMSB,170 TMSP,171
can help to circumvent these problems. Hexamethyldisilazane and tris(pentafluorophenyl)phosphine (TPFPP, 97),172 the high
((CH3)3SiNHSi(CH3)3) and heptamethyldisilazane ((CH3)3SiN- temperature storage performance and cycling ability of the
(CH3)Si(CH3)3) additives have been used to protect the LiMn2O4 battery can be enhanced (Fig. 17). These additives are film
cathode material from HF attack. Hexamethyldisilazane prevents forming additives and can help improve the high voltage
the formation of HF by dehydration and acid neutralization applicability of the battery. TMSB reduces the interfacial resistance
(as shown in Scheme 10, eqn (10)) and suppresses capacity by increasing the dissolution of LiF in the electrolyte. It
fading even after storage at high temperatures for 24 h.167 captures the electron rich fluorine ions from LiF through the

1970 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

Scheme 11 Reaction mechanism of the N,N-diethylamino trimethylsilane

additive with H2O and HF to protect metal dissolution from the cathode.
Fig. 17 Structures of additives for spinel and olivine cathodes containing
fluorine, phosphorus, silicone, nitrogen and boron functional groups.
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

when LiFePO4 is aged in air because of the formation of

electron deficient boron atom and form TMSB–F–. TPFPP is a LiFePO4(OH) which leads to a reaction between the surface
bifunctional additive which helps to improve the thermal OH and the LiPF6 based electrolyte. Moreover, the presence of
stability/flammability of the electrolyte and to suppress the HF and H2O can promote iron dissolution at elevated tempera-
decomposition of the electrolytes both on the anode and tures. (ii) The other problem is decomposition of electrolytes on
cathode side. It is sacrificially oxidized on the cathode and the electrode’s surface, especially at elevated temperatures. One
reduced on the anode prior to the electrolytes and forms a way to circumvent these problems is the use of electrolyte
stable and less resistive SEI. Similarly, TMSP is oxidized prior to additives. In order to avoid the dissolution of active material,
the electrolytes to form a conductive SEI which improves the resulting from the presence of HF and H2O impurities, addi-
kinetics of the cathode by preventing metal dissolution from tives that have the ability to react with HF and H2O are the best
the electrode bulk structure. It also prevents the migration of candidates. N,N-Diethylamino trimethylsilane (DEATMS, 99)
Mn to the anode side and improves the cycling stability of the can avoid problems associated with HF and H2O by reacting
electrode both at room and elevated temperatures. with the –OH group in LiFePO4(OH) which is formed when the
Silanes can improve the capacity and cycling performance of LiFePO4 cathode is exposed to moisture or to HF formed due to
LiMn2O4 not only at ambient or elevated temperature, but also the reaction of H2O with electrolyte salts. The reaction of the
at low temperature. Tetraethoxysilane (TEOS, 98), which has a additive with HF and H2O occurs by breaking the ‘‘N–Si’’ bond
low freezing point (85.2 1C), sacrificially oxidized on the as shown in Scheme 11.177 The resulting SEI is less resistive and
LiMn2O4 surface to form a conductive and stable SEI that stable which helps to improve the cycling performance even
protects the cathode’s surface from the electrolyte and minimizes after the cathode is air-aged.
manganese dissolution.173 The cycling performance and capacity Anion receptors are also good electrolyte additives since they
retention have been improved both at room and low temperatures. have the ability to form complexes with anions, thus preventing
Inorganic additives such as sodium phosphate compounds, the decomposition reaction of anions. Borane or borate based
Na4P2O7 and Na5P3O10, can enhance the high temperature cycling anion receptors can enhance the conductivity of the electrolyte
performance and capacity retention of LiMn2O4 by suppressing since they contain an electron-deficient B atom coordinated
acid formation resulting from the hydrolysis, which causes Mn with electron-withdrawing groups. TPFPB178 and trimethyl
dissolution.174 Sulfur containing additives, such as PES175 and borate (TMB, 100)179 additives are among the anion receptors
MMDS,176 are also promising additives for spinel cathodes to that can increase the dissociation of LiPF6 in electrolytes to Li+
enhance the performance of the battery at elevated temperature. and PF6. By coordinating with the PF6 anion, they prevent
PES forms a surface film both at the cathode and the anode and electrolyte decomposition. These additives can improve the
prevents the oxidation of the electrolytes on the cathode and thermal stability of LiPF6-based electrolytes by removing LiF
manganese deposition on the anode. Similarly, MMDS forms a from the SEI which is resistive and insoluble in most organic
conductive and protective layer on the cathode and inhibits the solvents. However, the concentration of the additives in the
dissolution of Mn at high temperatures. Additives for spinel electrolytes should be strictly controlled since they can release
cathode materials need to be bifunctional. They should improve highly reactive PF5 by capturing LiF from LiPF6, thereby
the performance of the cathode materials and at the same time accelerating the deterioration of the electrolyte solvents. More-
suppress deposition of Mn on the anode side. Hence, as well as over, the performance of the anion receptor is dependent on
modifying the electrode itself, research on spinel cathode materials the degree of localization of the LUMO at the boron center.180
should also focus on developing bifunctional electrolyte additives. Good electrochemical performance can be obtained when
anion receptors having a high degree of localization of the
3.5. Olivine cathodes, LiFePO4 LUMO at the boron center are used.
Due to its cost-effectiveness, nontoxicity, and good cycling The low electronic conductivity of LiFePO4 and its rate
performance, LiFePO4 is considered to be the most promising capability at low temperature can be improved with the addi-
cathode material for the large-scale application in the next- tion of low concentration of BS181 and FEC.182 Both additives
generation of lithium ion batteries. However, there are some are well-known film forming additives at room temperature.
issues that limit the practical applicablity of LiFePO4. (i) Olivine They modify the interfacial properties of the battery by forming
has a problem of iron dissolution and capacity fading when a conductive SEI film and decrease the electrochemical reaction
cycled at elevated temperatures. Iron dissolution also occurs polarization at low temperature. A similar effect can be obtained

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1971
 View Article Online

Review Energy & Environmental Science

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

Scheme 12 Summary of the properties of film forming additives for cathode materials.

using inorganic additives such as NaCl. Wu et al.183 showed that

1% NaCl can form a more uniform, less porous, conductive SEI
and enhance the fast transport of Li+ through the electrolyte and
the SEI film, which improves the lithiation/delithiation process
during cycling.
To summarize, the protective layer formed by cathode film-
forming additives should prevent the side reaction between the
electrolyte and electrode. It should also prevent the electrode
structural change thereby assisting in stabilizing the cathode
material. Scheme 12 summarizes some of the critical properties
Fig. 18 Schematic illustration of the redox shuttle overcharge protection
of film-forming additives for cathode materials. mechanism.

4. Additives for safety

the durability of the effect of the redox shuttles is the electro-
4.1. Redox shuttles chemical stability of the molecules. Even though, it is not
In addition to the film forming additives, there are other cathode straightforward to determine the stability and durability of a
additives that protect the cathode under abuse conditions such particular redox shuttle, the electrochemical stability of the
as overcharging. Overcharge is one type of abuse to lithium ion redox shuttles can be explained by examining the stability of
cells by forcing a current through it to charge the cell when the the oxidative species produced during overcharge protection
lithium ion cell is already fully charged. Overcharging the battery from the redox shuttle. Another factor that should be taken in
results in chemical and electrochemical reactions of the battery to account while evaluating redox shuttles is the system’s redox
components that can elevate the battery’s temperature resulting potential. Studies have shown that for better performance, the
in explosion. In order to prevent the battery from overcharging, redox potential of the additives should be about 0.3–0.4 V above
the additives should be electrochemically reversible at slightly the maximum operating potential of the cathode.184–187 Such a
higher potentials than the normal end-of-charge potential of the slight increase in potential will allow the cells to be normally
cathode. These additives are known as redox shuttle additives. charged before the shuttle molecule starts its intended function,
By reversibly oxidizing and reducing the cathode and anode side, hence minimizing the self-discharge effect. However, it should
respectively, the redox shuttle additives can divert the overcharge be noted that the potential should not exceed the operating
current and convert to heat to protect the reaction between electrochemical window of most commonly used electrolytes,
battery components at high voltage. Fig. 18 shows the mechanism thereby preventing the oxidation of the electrolyte, which could
of redox shuttle function during overcharge (black) and some lead to safety concerns.
possible decomposition pathways (red). The potential window of these additives is a key factor to
Based on previous studies of redox shuttle molecules, it can protect the battery from overcharge. It has been shown that
be inferred that electrochemical reversibility is an important aromatic compounds with two methoxy groups and halogen
requirement. As can be seen from Fig. 18, the oxidized form of a substituted directly on the benzene are stable up to 4 V and
typical redox shuttle molecule should be able to survive the undergo reversible redox reactions above 4 V, especially when
diffusion step, which is also an important factor in an electro- the two methoxy groups are located at the para and ortho
chemically reversible system. Another key property which dictates positions.188 The oxidized species are formed when electrons

1972 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

shows better cycling performance (50 cycles at 100% over-

charge) than DBDB but it is still lower than DDB.
Modifying DDB by introducing more-electron-withdrawing
trifluoroethoxy groups (i.e. 1,4-di-t-butyl-2,5-bis(2,2,2-trifluoro-
ethoxybenzene), DBTFB, 106) improves the potential (4.25 V vs.
Li/Li+) of the additive and its application for cathode materials
having potentials of up to 4.2 V such as, LiCoO2 or Li[Ni1/3Mn1/3-
Co1/3O2].193 Another modification of DDB is tetraethyl-2,5-di-tert-
butyl-1,4-phenylene diphosphate (TEDBPDP, 107) which can be
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

obtained by introducing diethylphosphate instead of the trifluoro-

ethoxy groups.194 This strong electron withdrawing group can
help to increase the potential to 4.8 V vs. Li/Li+, while avoiding the
cost and synthesis complexity problems due to the introduction of
fluorine containing groups. However, the radical cations of this
Fig. 19 Structures of redox shuttle additives containing derivatives of additive might cause side reactions and be susceptible to anode
dimethoxybenzene. reduction. Dimethoxydiphenylsilane (DDS, 108)195 which has a
comparable redox potential (4.8 V) with TEDBPDP also showed
are removed from the benzene where the methoxy group is effective overcharge protection by forming a polymeric film on the
attached: the lone pairs of the oxygen on the methoxy group can cathode and separator that consumes excess charging current.
stabilize the oxidized species by electron delocalization. Derivatives However, it has a negative effect on the performance of the battery
of dimethoxybenzene, by introducing various electron withdrawing when a high amount of the additive is used; hence, the amount
groups, are also used as overcharge protecting agents. 2,5-Di-tert- should be controlled.
butyl-1,4-dimethoxybenzene (DDB, 101) has been reported as a The structure of the electron withdrawing group is very
good redox shuttle additive for LiFePO4 cathode materials important to circumvent side reactions due to the formation
(Fig. 19).189 It shows excellent electrochemical stability and of radical cations. Introducing an electron withdrawing group
reversibility, due to its rigid molecular structure. However, its having a rigid tetrahedral structure can improve the potential
lower potential (3.9 V vs. Li/Li+) limits the application of this and also sterically protect the radical cation from further
additive for high voltage electrode materials. Moreover, it also reaction. The above DDB derivative redox shuttles focused
has a problem of low solubility in polar carbonate-based on the modification of the methoxy moiety while retaining
electrolytes because of its low polarity as a result of its symmetric the di-tert-butyl groups. On the other hand, derivatives of DDB
structure. by replacing di-tert-butyl groups and keeping the methoxy
To alleviate these problems, 3,5-di-tert-butyl-1,2-dimethoxy- group have been reported. Recently, Huang et al.196 reported
benzene (DBDB, 102), which is an isomer of DDB, has been 1,4-bis[bis(1-methylethyl)phosphinyl]-2,5-dimethoxybenzene
tested.190 It shows high solubility because of its asymmetric (BPDB, 109) as a functional redox shuttle additive. BPDB is a
structure and large dipole moment. Although the redox modification of DDB with the introduction of an electron
potential (4.20 V vs. Li/Li+) and solubility of DBDB are higher withdrawing organophospine oxide group instead of di-tert-
than DDB, it exhibits poor cycling performance (only 11 cycles butyl. Though the potential of BPDB (4.5 V vs. Li/Li+) is greater
at 100% overcharge) compared to DDB (4200 cycles). The than DDB (3.9 V vs. Li/Li+) and the radical cation can be
radical cation decomposition is easier for DBDB than for sterically protected at high potential, the performance of the
DDB due to the position of the two methoxy groups which battery is low. This is because of the presence of the alkyl diphos-
makes the hydrogen transfer possible to eliminate methane. In phine group which makes the additive vulnerable to electrons
a separate publication, the same group has reported 4,6-di-tert- and decomposes on the anode. In a separate work, this group
butyl-1,3-benzodioxole (DBBD1, 103) and 5,7-di-tert-butyl-1,4- has also reported 1,4-bis(trimethylsilyl)-2,5-dimethoxybenzene
benzodioxin (DBBD2, 104) as highly soluble redox shuttles.191 (BTMSDB, 110) as a potential overcharge protection additive
Both compounds contain a heterocyclic ring with two oxygen for the LiFePO4 electrode (Fig. 20).197 This additive has a redox
atoms fused to a tert-butyl-substituted benzene ring. Their potential at 4.1 V vs. Li/Li+, while a diffusion coefficient of
redox potentials (4.12 V for DBBD1 and 4.17 V for DBBD2 vs. 3.09  106 cm2 s1 was obtained for the radical cation which
Li/Li+) and diffusion coefficients are similar. However, DBBD2 is better than DDB. However, DDB showed 3 times longer
provided a better overcharge protection (42 cycles with 100% overcharge protection than BTMSDB because of the relatively
overcharge) than DBBD1 (only 3 cycles with 100% overcharge). poor stability of the C–Si bond and the subsequent side reaction
This is because of the presence of the extra CH2 group in of the BTMSDB radical cation.
DBBD2 which greatly enhances the electrochemical stability All the above redox shuttles with modification of DDB
compared to DBBD1. Another group reported 4-tert-butyl-1,2- contain a functional group for steric protection to minimize
dimethoxybenzene (TDB, 105) as a redox shuttle additive, which the formation of reactive sites, due to the delocalization of the
is highly soluble in organic carbonate based electrolytes.192 It has positive charge in the aromatic ring after oxidation. An alter-
the required redox potential for 4 V cathodes (4.10 V vs. Li/Li+) and native way to avoid the delocalization of positive charges and

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1973
 View Article Online

Review Energy & Environmental Science

MTP showed a greater number of cycle protection in a LiPF6

based electrolyte than LiBOB. However, they observed the reverse
phenomenon when a LiFePO4/Li4/3Ti5/3O4 cell was used.
Casselman et al.201 have studied the stability and reactivity
of various phenothiazine derivatives by varying the substituent
on the N position with H, alkyl, or aryl groups and the para
position with H, Cl, Br, or CF3 groups. They observed that even
though halogen substituents have the ability to improve the
redox potential, they cause a negative effect on the anode side
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

since the reduction potential of the additive also increases.

Their result proved that from the alkyl substituents on the
N atom, ethyl, isopropyl or phenyl groups offer the greatest
stability; and from the para substituents, the trifluoromethyl
group simultaneously provides higher redox potentials and
stable radical cations. Similarly, Ergun et al.74 showed the effect
of substituents by studying various derivatives of phenothiazine
by varying alkyl substituents on the N position and halogens at
the para position to N, changing the S hetroatom (phenothi-
azine) to the O hetero atom (phenoxazine, EPX, 116, II) or no
Fig. 20 Structures of redox shuttle additives containing derivatives of heteroatom (carbazole, ECA, 116, III) and no fused ring (diphenyl-
dimethoxybenzene, tetramethylpiperinyl-oxide, pyroxyl, phenothiazine, amine, DPAn, 121). Derivatives of ethyldiphenylamine (EDP,
phenoxazine, carbazole and diphenylamine. 116, IV) did not show any overcharge protection, however,
phenothiazine and phenoxazine derivatives exhibited over-
minimize the number of reactive sites is use of other shuttle charge protection with a varying degree of success.
classes such as 2,2,6,6-tetramethylpiperinyl-oxide (TEMPO, 111).198 In a separate study, the same group has reported the relation
In this structure, the positive charge of the oxidized TEMPO between the overcharge protection and stability of the neutral
between the nitrogen and oxygen of the nitroxide group is sterically and radical cation of N-substituted phenothiazine.202 They
protected by four neighboring methyl groups. Buhrmester et al.198 have compared the stability and performance of the neutral
have reported the derivatives of TEMPO with a five-membered ring and radical cation of MPT, EPT, N-isopropylphenothiazine
as redox shuttle additives for Li4/3Ti5/3O4/LiFePO4 cells. Among the (iPrPT, 122), PT, N-tert-butylphenothiazine (tBuPT, 123), and
various additives they have tested, 4-methoxy-TEMPO (4-cyano- N-phenylphenothiazine (PhPT, 124) (Fig. 21). EPT, iPrPT, and
TEMPO, 112) and 3-cyano-2,2,5,5-tetramethyl-1-pyrrolidinyloxy PhPT showed long overcharge cycling (4100 cycles at 100%
(3-cyano-PROXYL, 113) showed long cycle overcharge protection overcharge) and a shorter cycle life was observed for PT, MPT
(4100 cycles at 100% overcharge). Their low redox potentials and tBuPT. They have explained that in the case of PT, the N–H
can be increased by replacing the hydrogen atoms in the methyl bonds in the aromatic core can dimerize upon oxidation to
groups by fluorine atoms. In contrast, Taggougui et al.199 used form new N–N bonds or react with aromatic rings. In the case of
TEMPO as a redox shuttle for Li/Li4Ti5O12 half-cells and showed tBuPT, since the radical cation (tBuPT +) decomposes into PT +
that TEMPO cannot be used as a redox shuttle. This is because of
the instability of the TEMPO+ cation and/or TEMPO itself on the
electrodes, as it forms a passive layer on the positive electrode
that can impede lithium ion insertion–extraction.
Another group of redox shuttle additives having comparable
effects with that of DDB are phenothiazine (PT, 114) derivatives.
Although their redox potentials are low (near 3.5 V) for LiFePO4,
the redox potential can be increased by substitution of the alkyl
group with an electron withdrawing group. Buhrmester et al.200
reported the derivatives of phenothiazine such as 10-methyl-
phenothiazine (MPT, 115), 10-ethylphenothiazine (EPT, 116, I),
3-chloro-10-methylphenothiazine (3-chloro-EPT, 117), 10-isopropyl-
phenothiazine (IPT, 118), and 10-acetylphenothiazine (APT, 119) as
stable redox shuttle additives. They compared the effect of these
additives in LiPF6 and LiBOB electrolytes by taking MPT as an
example. LiFePO4/Li4/3Ti5/3O4 coin cells containing the additive
and LiBOB based electrolyte showed good overcharge and over- Fig. 21 Structures of redox shuttle additives containing fluorine, boron,
discharge protection for long cycles but LiFePO4/graphite coin phenyl, thiophene, aniline, phenylamine functional groups, ferrocene and
cells showed lower stability. At 30 1C, LiFePO4/graphite cells with derivatives of phenothiazine.

1974 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

when it is oxidized, PT is responsible for overcharge protection excess amounts of the additive leads to high-concentration of
instead of tBuPT itself. They have correlated the poor perfor- the redox shuttle which is designed to maximize its overcharge
mance of MPT with the low solubility of MPT +. However, their protection capability. However, the presence of high amounts
cycling performance results for MPT contradict the findings of of the anion receptor would accelerate the decomposition of
Buhrmester et al.200 LiPF6. Hence, to address this conflict, the authors suggested
The reversibility of the redox shuttles does not guarantee the the use of a combination of low-concentrations of PFPTFBB
presence of good overcharge performance. The stability of the and high-concentrations of PFPTFBB–F, which has the same
radical is correlated with the overcharge protection perfor- redox shuttle capability as PFPTFBB. In this combination, only
mance. Based on this concept, Odom et al.203 reported a fast PFPTFBB is an active anion receptor, and both PFPTFBB and
PFPTFBB–F are redox shuttles. Although redox shuttles with
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

and inexpensive approach to screening potential redox shuttles

for overcharge protection without the need for assembling a various functional groups have been reported, further work is
coin cell and cycling for a long time. They have suggested use of needed for their high voltage applications. Moreover, screening
a chemical oxidant whose oxidation potential occurs at a higher of the redox shuttles using computational methods is a good
potential than that of all the neutral compounds of the additive strategy to obtain new additives with desired properties.
to generate radical cations. They have used UV-vis spectra to Other bifunctional additives reported by Chen et al.206 are
confirm the generation of a radical cation and electron para- lithium borate cluster salts. They have investigated two lithium
magnetic resonance (EPR) to monitor the concentration of the borate cluster salts Li2B12H3F9 and Li2B12F12 which can act as
radical cation over time. They have proved their concept using both lithium salts and redox shuttles for overcharge protection.
N-ethylphenothiazine and similarly structured anthracene Both salts have close chemical and physical properties. However,
based heterocycles. Their results showed that in 5 h or less, it due to the variation in the degree of fluorination, the redox
is possible to predict which redox shuttles are the most stable potential of B12F122 (4.6 V vs. Li+/Li) is about 0.1 V higher than
and can give good overcharge protection. However, using this that of B12H3F92 (4.5 V vs. Li+/Li). This indicates the possibility
technique, it is impossible to predict the shuttle performance of designing various redox shuttles for different positive electrode
of a species that has a UV-vis spectrum that changes in shape materials by varying the degree of fluorination.
over time. In the full cell configuration, where the anode is covered by
Another class of redox shuttle additives for better overcharge the SEI film, the working mechanism of the redox shuttle
protection are thiphenylamines (TPA, 125).204 However, the additives should be well understood. The oxidized additive on
radical cation formed upon oxidation of TPA is not stable since the cathode during charging should be reduced on the anode
it dimerizes to form tetraphenylbenzidine (TPB). By introducing side during discharge. However, if the anode surface is covered
bromine substituents at the para position, the dimerization by the SEI film, which is electronically an insulator, how the
reaction can be prevented and the oxidation potential can be additive can gain electrons for back-reduction is a key question.
increased. As the number of Br substituents increased, the Tang and Newman207 showed that the reduction of the oxidized
oxidation potential can be increased from 3.76 V (for TPA) to redox shuttle on the SEI covered electrode is due to the
3.90, and 4.30 V (for tris(4-bromophenyl)amine, and tris- presence of a compact inner and porous outer SEI layer. The
(2,4-dibromophenyl)-amine, respectively). Even though the inner layer is resistant to the redox shuttle but the electron can
potential is lower than TPA, DPAn (3.58 V) showed effective tunnel through it. The electrons cannot tunnel through the
overcharge protection with no significant influence on the outer layer since it is beyond the electron tunneling range.
performances of the battery.205 The DPAn monomer forms a However, the pores are enough to allow the transport of the
polymer bridge (PDPAn) in-between the cathode and anode by redox shuttles. Kaymaksiz et al.208 addressed this matter in
electro-polymerization and prevents the battery from voltage their 2-layer SEI model, where back reduction is dependent on
runaway. the type of redox shuttle and the character of the SEI. They
Chen et al.185 reported 2-(pentafluorophenyl)-tetrafluoro- showed this phenomenon by taking ferrocene (127) and DDB
1,3,2-benzodioxaborole (PFPTFBB, 126) as a novel redox shuttle redox shuttles as examples. The SEI formed in the presence of
additive. Its high redox potential (4.43 V vs. Li/Li+) enables the ferrocene is too dense to allow the transport of oxidized
additive to be applicable for high voltage cathode materials. ferrocene and back reduction is blocked. However, for DDB
The other important feature of this additive is the presence the SEI did not block the oxidized DDB transport, thus back
of various functional groups which makes it a bifunctional reduction was possible. Moshurchak et al.209 have explained that
additive for overcharge protection and improving the life of reduction of the oxidized redox shuttle in the presence of the SEI
lithium-ion cells. The phenyl group directly attached to the occurs due to the difference in energy between the oxidized
oxygen atoms makes the additive electrochemically active, redox shuttle and lithium ions or solvent molecules. The SEI can
while the redox potential can be increased and the polymeriza- block electron tunneling to reduce lithium ions and solvent
tion reaction upon oxidation can be prevented by fluorine molecules, because tunneling of electrons to the oxidized redox
substitution. Moreover, PFPTFBB acts as an anion receptor, shuttle is possible due to the lower energy of the oxidized redox
due to the presence of the pentafluorophenyl borole functional shuttle than the lithium ions and solvent molecules.
group which improves the capacity retention of the battery The use of electroactive polymer separators such as poly-
by dissolving inorganic components, such as LiF. Addition of (3-butylthiophene) (P3BT, 128),210 p-polyphenyl (PPP, 129), the

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1975
 View Article Online

Review Energy & Environmental Science

In summary, when selecting additives for overcharge protec-

tion, it is important to consider the potential window of the
additive, which should be slightly higher than the normal end-
of-charge potential of the cathode. Moreover, the additive
should be electrochemically reversible and the radical cation
formed should be stable. Summary of some of the properties of
redox shuttle additives is shown in Scheme 13.

4.2. Flame retardants

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

One of the fundamental issue that restricts the widespread

application of lithium ion batteries is safety, which is an issue,
especially if the battery is abused. As discussed in the previous
section, overcharge, which is an electrochemical abuse condi-
Fig. 22 Potential windows of selected redox shuttle additives on various tion, is one safety issue. Other safety issues include mechanical
electrodes.
abuse, such as crush and penetration, and thermal abuse, such
as heat generation, combustion and flame attack.
polyaniline (PAn, 130) composite,211 and polytriphenylamine Though the properties of all the battery components should
(PTPAn, 131),212 is another method to achieve overcharge be considered when developing a safer battery, reducing the
protection. When the battery is overcharged, these electroactive highly flammable character of electrolyte solvents is vital, as it
polymers can reversibly switch from insulating states to conductive is one of the key factors able to cause thermal runaway.
states and generate an ‘electronic short’ across the cell to prevent Operation of the battery at elevated temperatures also affects
the cell’s voltage from exceeding the safety value. On the other the performance of the battery. Hence, the use of fire-retardant
hand, instead of using liquid electrolyte additives or electroactive (FR) additives to limit the flammability of electrolyte solvents
polymer separators as redox shuttles, the use of a solid-state anti- has become the main focus. When flame retardant additives are
overcharge additive for intrinsic overcharge protection has been used, their compatibility with the anode and other properties of
suggested.213 The authors demonstrated the concept of using the the battery such as ionic conductivity and performance of the
LiCoO2/CuO solid state composite cathode. Upon overcharging, battery should be considered. The cause for combustion is radical
CuO decomposition releases Cu2+ which is reversibly reduced at chain reactions in the gas phase; thus FR additives should have
the anode to Cu+. Thereafter, the Cu2+/Cu+ redox pair acts as a the ability to scavenge radicals and terminate reactions. Organic
shuttle maintaining a stable and safe low voltage. Moreover, the phosphorous compounds are promising candidates as FR
decomposition of CuO prevents electrolyte decomposition. Fig. 22 additives because of their high flame retarding ability in the
shows the potential window of some redox shuttle additives on vapor phase (by a radical mechanism) and they are environ-
various electrodes. mentally friendly.

Scheme 13 Summary of the properties of redox shuttle additives.

1976 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

large molecular volume and more stably conjugated P-electronic

structure compared to other organophosphate compounds. This
limits its co-intercalation into graphite. Even though the self-
extinguishing time (SET) can be lowered with increasing the
concentration of TMPP, only 10% of the additive can be used
since higher concentration also reduces the ionic conductivity
because of the high viscosity of TMPP.
Phenyl functional group containing phosphates, such as
cresyl diphenyl phosphate (CDP, 136)218 and diphenyloctyl
phosphate (DPOF, 137),219 have also been reported as effective
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

FR additives. CDP has the essential characteristics for FR

additives, such as a relatively high boiling point compared to
other phosphates, appropriate viscosity and melting point. It
showed a good FR ability (47 s SET) with low concentration
(only 5%) and only slight worsening of the performance of the
Fig. 23 Structures of flame retardant additives containing derivatives of battery. The SET can be lowered to 15 s by increasing the CDP
phosphonate, phosphate and phosphazene. concentration to 40%. However, increasing the concentration
of the additive can cause capacity decay. Another study using
CDP showed that increasing the concentration to 10 wt% can
Among the organic phosphorous compounds, dimethyl improve the cycling efficiency.220 Similarly, although only 5% of
methylphosphonate (DMMP, 132) has been reported as a good DPOF is enough for good FR,219 with optimum concentration (10%),
radical inhibitor by capturing the radicals H and HO in the it is also possible to improve the electrochemical performance of the
flame zone to weaken or terminate combustion reactions battery by forming a thin SEI on the anode.221
(Fig. 23).214 Decomposition of H3PO4 to PO and PO2 can also One method of improving the compatibility of organopho-
capture H and HO . 10% DMMP gives a nonflammable sphorus FR additives with the graphite electrode is use of
electrolyte for the Li/LiCoO2 cell, at the expense of some compounds having both film forming and flame retarding
capacity. In the full cell configuration, DMMP causes capacity abilities such as bis(2-methoxyethoxy)methylallylphosphonate
fading due to the formation of surface films on the anode. It (BMEMAP, 138)222 and allyl tris(2,2,2-trifluoroethyl)carbonate
has also a problem of exfoliating graphite if the surface is (ATFEC).223 Both additives effectively enhance the thermal stability
untreated. A comparison study for the compatibility of the FR of the electrolyte and also prevent the co-intercalation of PC by
additives with the graphitic anode using DMMP and diethyl forming a stable SEI film. The introduction of a fluorine group in
ethylphosphonate (DEEP, 133) showed that the effect is depen- the latter helps to facilitate the reduction of the additive and leads
dent on the solvation power of the additive.215 DMMP has high to the formation of a SEI layer on graphite. Another method to
lithium ion solvation ability and co-intercalates into the graphite avoid the trade-off between flammability and electrochemical
layer, which results in the deterioration of the electrode struc- performance is to replace organophosphorus FR additives by cyclic
ture. On the other hand, in the presence of DEEP, lithium ions phosphazenes. Even though cyclic phosphazenes have non-
reversibly intercalated due to the lower solvation ability of DEEP flammable characteristics, they have a relatively low efficiency for
compared with the electrolyte solvents such as EC. suppressing flammability compared to phosphates. Introduction of
Triphenyl phosphate (TPP, 134) showed a significant fluorine functional groups can result in a synergistic flame
reduction in the flammability of the electrolyte without altering retardant effect and good electrochemical compatibility.
the flash point.216 Though a considerable decrease in the A series of fluorinated phosphazenes with alkyl or aryl functional
cycling performance and capacity retention was not observed, groups showed good FR efficiency with no significant influence on
the presence of TPP causes a slight decrease in the electrolyte’s the electrochemical performance of the battery.224 Rollins et al.225
conductivity. This resistive nature results from the SEI film’s reported another series of chemically similar fluorinated cyclic
structure change due to TPP which hinders facile lithium phosphazene trimers with varying numbers of fluorinated
kinetics. However, with the addition of a reasonable amount pendant groups as a co-solvent. The fluorinated cyclic phos-
of TPP (10%), the negative effect can be minimized while phazenes were prepared by nucleophilic substitution of the
keeping the FR performance. Tri-(4-methoxythphenyl) phos- reactive chlorines on hexachlorocyclotriphosphazene with
phate (TMPP, 135) has been reported as a multi-functional ethoxy or 2,2,2-trifluoroethoxy. All the phosphazenes showed
additive for safety.217 It can not only lower the flammability of high thermal stability and higher discharge capacities with less
the organic carbonate electrolyte but also greatly enhance the capacity fade compared to the cell without them. Although
thermal stability of the battery. Moreover, it has the ability to good performance can be obtained with up to 30% concen-
protect the batteries from voltage runaway during overcharge tration of phosphazenes, the discharge capacity becomes lower
by polymerizing at 4.35 V (vs. Li/Li+) and form a conductive at high rates and high concentrations. This is because of the
polymer bypassing the reaction current. It is also found to be resulting high viscosity and lower conductivity of the electrolyte
compatible with the graphite anode, which is possibly due to its with the addition of phosphazenes.

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1977
 View Article Online

Review Energy & Environmental Science

A highly efficient FR additive with no effect on the electro- friendly and can serve as both FR and SEI forming additives.
chemical performance of the battery has been reported by Xia Other organophosphorus compounds reported as bi/multifunctional
et al.226 They used the (ethoxy)pentafluorocyclotriphosphazene additives or with low impact on the electrochemical performance
(PFPN, 139) additive which can make the electrolyte totally non- with good FR efficiency include bis(2,2,2-trifluoroethyl)methyl-
flammable at low concentrations (5 wt%). When decomposed, phosphonate (TFMP, 143),230 triphenyl phosphite (TPPi, 144),231
the additive produces fluorine radicals, which scavenge hydro- bis(2,2,2-trifluoroethyl)ethylphosphonate (TFEP, 145),232 tetra-
gen radicals from the flame zone and terminates the combus- bromobisphenol A (TBBA, 146),233 resorcinol bis(diphenyl phos-
tion reaction. Moreover, the presence of an alkoxy (–OCH2CH3) phate) (RDP, 147),234 bis(N,N-diethyl)(2-methoxyethoxy)methyl-
group in PFPN makes the formation of P radicals easier which phosphonamidate (DEMEMPA, 148),235 ethylene ethyl phosphate
(EEP, 149),236 and phosphaphenanthrene-based compounds.237 In
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

can also scavenge selectively the hydrogen and oxygen radicals

produced by the combustion reaction. The addition of 5% addition to the phosphorous based organic FR additives, boron
PFPN has no significant impact on the conductivity of the centered and phosphorus centered phosphoryl-rich lithium salts
electrolytes and the performance of the battery is not affected. have been reported as effective FR additives.238 The lithium salts
Its oxidation prior to the electrolytes and the formation of a SEI are Li[B(DPC)(oxalato)], Li[B(DPC)2], Li[B(DPC)F2], and Li[P(DPC)3],
film on the cathode surface make its application possible for where DPC is (3,6-diethoxyphosphoryl)-1,2-catecholato. During
high voltage cathode materials. Fluoride-rich triazines can also combustion these additives form high levels of char (polyphos-
give efficient FR effects and good battery electrochemical phoric acids and a variety of boron oxide enhanced chars) which
performance. Kim et al.227 reported 2,4,6-tris(trifluoromethyl)- lower the total heat release and reduce flammability.
1,3,5-triazine (TTFMT, 140) as a FR additive to improve both the Their nonvolatile and non-flammable nature makes ionic
thermal stability and performance of the battery (Fig. 24). This liquids ideal candidates as FR additives for lithium ion batteries.
synergetic effect is because it contains triazine, a well-known With an optimum concentration of ionic liquids, it is possible to
FR material, and the fluorine functional group for better obtain a balance between cycling performance and FR efficiency.
performance. More than half of the exothermic heat generation Choi et al.239 showed that 10 wt% of 1-butyl-1-methylpyrrolidinium
when oxygen is released from the cathode can be minimized hexafluorophosphate (BMP-PF6, 150) could improve the safety
with the addition of only 5 wt% TTFMT. without degrading the cell’s cycling performance (Fig. 25). A
To minimize the toxicity and high cost of fluorine containing comparison study using 1-ethyl-3-methylimidazolium bis(trifluoro-
compounds, combining alkoxy functional groups with nonflam- methanesulfonyl)imide (EMImTFSI, 151) and EMImPF6 (152)
mable phosphate and phosphazene instead of fluorine has been showed that the electrolytes with EMImTFSI has a stronger
suggested.228 10% triethoxyphosphazen-N-phosphoryldiethylester interaction with the carbonate solvents and higher flash point
(PNP, 141) showed a balance between the electrochemical (which is correlated to the chemical composition of the vapor
compatibility and FR efficiency. This is because of the presence phase above the liquid) than EMImPF6.240 Because of the
of the phosphazene group for electrochemical compatibility strong interaction, less carbonate solvents are ‘‘free’’ to evapo-
and the phospahate group for good FR. The presence of a high rate and EMImTFSI is proved to be an effective FR. N-Methyl-N-
phosphorous content (19.5%) in PNP helps to produce more propylpiperidinium bis(trifluoromethanesulfonyl)imide (PP13-
free P which can capture H and HO . Silane compounds, such TFSI, 153) also showed nonflammability and good discharge
as MPBMDS (48),106 as discussed in Section 2.1, and vinyl-tris- performance in the lithium ion cell with a carbon negative
(methoxydiethoxy)silane (VTMS, 142),229 are environmentally electrode at 40 wt%.241 When it was used as an electrolyte, the
addition of a mixture of triethylphosphate (TEP, 154) and EC
improves the ionic conductivity, viscosity, thermal stability,
electrochemical stability, and operational temperature range.242
This is because of the low viscosity, low melting point, high
boiling point, non-flammability and relatively high dielectric

Fig. 24 Structures of flame retardant additives containing derivatives

of triazine, phosphazene, silane, phosphite, phosphonate, phosphate and Fig. 25 Structures of flame retardant additives containing ionic liquids
phosphonamidate. and derivatives of phosphate.

1978 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

gases, leading to thermal runaway, which can potentially result in

explosion243 Moreover, thermal runaway causes reduction of large
amounts of the electrolyte because of the breakdown of the SEI on
the anode side. A thorough understanding of how to control the
degree of heat generation is essential to prevent thermal runaway,
which is due mainly to the violent exothermic reaction of the
flammable electrolyte with the positive electrode, resulting in
oxygen evolution from the cathode.244–246 In order to reduce
such reactions, previous studies have focused on reducing the
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

flammable nature of the electrolytes by adding additives and

co-solvents to the electrolytes.247–250
Fig. 26 The optimum amount used and the SET time obtained by As discussed in the previous sections, various types of
selected FR additives. electrolyte additives, such as redox shuttles187,188,251 and poly-
merizable species,252 have been explored as potential agents
able to prevent overcharging. In addition, nonflammable electrolytes
constant of TEP and the good SEI forming ability of EC. The
with added fire-retardants have been reported to reduce the
optimum amount used and the SET time obtained by some FR
electrolyte flammability. However, the fire-retardants also have
additives are shown in Fig. 26. For most FR additives, to minimize
negative effects on battery performance due to their reaction
the SET time, a large amount needs to be used. However, increasing
with the electrode active materials.253,254 Recently, several
the amount has a negative effect on the performance of the battery.
monomer additives that polymerize to form electrically con-
Only a few reports show low SET time with a low concentration of
ductive polymers at battery voltages greater than the maximum
additives. Hence, developing an FR additive that can minimize the
operating charging voltage have been reported in the patent
SET time at low concentrations with no effect on the performance
literature.255–257
of the battery should be addressed.
A self-terminated oligomer with hyper-branched architecture
Generally, FR additives should have the ability to scavenge
(STOBA, 156), developed in Taiwan at the Industrial Technology
the radicals generated in the flame zone and terminate the
Research Institute (ITRI), is a nanoscale polymeric cathode
combustion reaction. In addition, the additive should not affect
coating material which reportedly effectively suppresses thermal
the other properties of the electrolyte. Scheme 14 summarizes
runaway during abuse conditions in a lithium ion battery.258–260
some of the essential properties of FR additives.
It was reported that, when the battery heats up to 130 1C, STOBA
transforms from a porous material to a film and stops the
4.3. A self-terminated oligomer with hyperbranched reaction. 2% of STOBA can completely coat the electrode surface
architecture (STOBA) and successfully overwhelm thermal runaway with no significant
Thermal runaway, which is a rapid uncontrolled increase in effect on the electrochemical performance.261
temperature, can be triggered in lithium-ion cells under certain
abuse conditions such as internal short circuits, overheating of
the cell and nail-piercing. As the cell temperature increases, the
protective layer at the surface of the electrodes becomes unstable. If
the temperature keeps increasing, energetic reactions between the
oxygen which evolved from the cathode and the electrolyte may
occur. These reactions produce excessive heat and flammable Because of its hyperbranched structures, it has the ability to
form a stable complex with the electrode active material. In
addition, it is compatible with the electrolyte and increases the
adhesion between the electrode and the current-collector. At
room temperature, the STOBA coating does not affect the
lithium ion transfer upon cycling. As the temperature in the
cell increases, the surface deposited STOBA chains crosslink
and became nonporous and rigid which inhibits the decom-
position of electrolytes by blocking direct contact of the elec-
trolyte with oxygen. Moreover, this crosslinking and the change
from pores to a nonporous layer upon heating blocks both
lithium ion and electron transfer.260 The mechanism is shown
in Fig. 27. Lin et al. showed that since electrolyte decomposi-
tion is suppressed before the release of oxygen from the oxide
layer, the STOBA coated on the Li(Ni0.4Co0.2Mn0.4)O2 does not
affect the layer-to-spinel phase transformation.258 Despite a few
Scheme 14 Summary of the properties of flame retardant additives. reports on the mechanism, there is still a lack of detailed

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1979
 View Article Online

Review Energy & Environmental Science

information on how STOBA terminates thermal runaway. Moreover,

studies on the modification of STOBA with various functional
groups, or new additives having similar structure and properties,
such as additives that do not undergo reaction under storage or
normal operation conditions but trigger polymerization at a certain
temperature, should be a fruitful future research direction.

5. Salt type additives

Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

State-of-the-art commercial lithium ion batteries use LiPF6 as

the electrolyte salt because of its superior properties when
compared to other salts such as LiBF4 and LiClO4. However,
it has problems related to thermal stability and reaction with
moisture, which leads to the formation of LiF and PF5. In the
presence of trace H2O impurities, PF5 undergoes hydrolysis at Scheme 15 Reaction mechanism of PF5 with the electrolyte solvents and
elevated temperatures producing HF which leads to transition their decomposition products, adapted from ref. 262 and 263.
metal dissolution from the cathode material. In the absence of
H2O, PF5 can react with the SEI components on the graphite
and deteriorate the performance of the battery as shown in
Scheme 15, eqn (17)–(25).262,263
As described in the previous sections, although their con-
centrations need to be controlled, the introduction of anion
receptors has the ability to stabilize LiPF6 by complexing with
PF5. Another way of solving problems associated with LiPF6 is to
use salt additives such as lithium bis-(oxalato)borate (LiBOB),
lithium difluoro oxalate borate (LiDFOB), and lithium tetrafluoro
oxalato phosphate (LTFOP). As mentioned in Section 3.2, LiBOB
can form a stable SEI containing semicarbonate species both on
the anode’s and cathode’s surface, Scheme 16a. Oligomeric
borates are also formed where the polymerization reaction is
initiated through one of the two BOB anions, Scheme 16b.262
When the other BOB anion is electrochemically activated,
there is a cross-linking reaction with various BOB-based polymer
chains which leads to higher interfacial impedance. The interfacial
impedance can relatively be reduced by using additives with only one
oxalato group such as LiDFOB264 and LiTFOP.265 These additives
form a stable SEI similar to LiBOB, however, because of the absence Scheme 16 Mechanism of SEI formation by LiBOB, (a) semicarbonate
species formation, (b) oligomeric borate formation, adapted from ref. 262.
of the other oxalato ring, there is no cross-linking reaction which
results in lower interfacial impedance than with LiBOB. Both
additives improved the capacity retention and performance of a
MCMB/Li1.1[Ni1/3Co1/3Mn1/3]0.9O2 cell. Moreover, the thermal
stability of LiPF6-based electrolytes can be improved in the
presence of these additives by preventing the generation of PF5
during the thermal dissociation of LiPF6. At high temperatures,
LiDFOB undergoes disproportionation reactions with LiPF6 and
forms LiBF4 and LiPF4C2O4.266 Lee et al. showed that LiDFOB
resulted in a good discharge capacity retention of Si–C/LCO–
NMC full cells stored at high temperature by forming a stable SEI
both on the Si–C anode and the LCO–NMC cathode.267

6. Analytical tools for analysis of SEI

Fig. 27 Schematic diagram of the morphology change of the STOBA
The main role of electrolyte additives is modification of the SEI
layer in the STOBA-coated Li(Ni0.4Co0.2Mn0.4)O2 cathode after thermal to prevent interfacial reactions between the electrode and the
treatment, adapted from ref. 260. electrolyte. Hence, analysis of the SEI formation process is vital

1980 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

to understand how the additive tunes the SEI. The difficulty in circumvent this problem is to use in situ techniques even
demarcating the SEI/electrolyte interface, together with the though designing a special spectroelecrochemical cell is
formation of a complicated SEI layer, makes the analysis very required. Among the spectroscopic techniques, FTIR is an ideal
tedious. However, a variety of techniques for analyzing SEI tool for in situ analysis since it is a nondestructive technique.306
films have been reported. For identification of the surface films Other in situ techniques, such as macro-IR, Raman with point
and morphology studies, spectroscopic techniques such as by point mapping,302 TEM,307 topography AFM image,308 and
nuclear magnetic resonance (NMR),268–273 X-ray photoelectron ellipsometry309 have been reported to study the SEI growth in a
spectroscopy (XPS),71,114,118 extended X-ray absorption fine lithium ion battery. In situ AFM is also a powerful tool for
structure (EXAFS),274,275 X-ray absorption near edge structure analyzing interfacial phenomena in batteries because changes
(XANES),276 Raman,277,278 Auger electron spectroscopy (AES),279,280
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

in SEI layers as a function of applied potentials, or in the

Fourier transform infrared spectroscopy (FTIR),281,282–287 mass presence of additives, can be directly monitored.310,311 The
spectroscopy,288–292 and electrochemical impedance spectroscopy thickness and density of the SEI can be determined using
(EIS),120,293,294 and microscopy techniques such as scanning neutron reflectometry (NR). The main advantage of NR is its
electron microscope (SEM),295–298 transmission electron micro- high sensitivity for measuring lighter elements in the SEI such
scopy (TEM),298,299 and scanning tunneling microscopy (STM)300 as Li, H, C, O, and F.312 Veith et al. have used NR in combi-
have been used. nation with FTIR and XPS analysis to study the SEI composition
Among these, FTIR is the most widely used technique for SEI and thickness on silicon anodes.312
analysis as it is a surface sensitive and very powerful tool for Another critical tool for SEI characterization is in situ EIS, which is
differentiating functional groups based on their dipole moments. important for studying changes in the surface resistance as the
However, since most reduction products have similar FTIR peaks potential is varied. Wang et al. studied electrochemical reaction
to the electrolytes, other complementary techniques such as XPS resistance and intrinsic resistance changes during initial Li inser-
also need to be used. Andersson et al.301 showed the use of the tion into graphite by varying the interface terminal-to-electrode
depth profile and conventional XPS for both compositional and connections and the arrangement of the counter and reference
morphological studies of the SEI formed on the graphite surface. electrodes.313 Using this, they were able to study the formation
To understand SEI formation on high voltage cathode materials, and growth of the SEI film on graphite at room temperature and
they reported an XPS study of decomposition layer thickness and at 65 1C. Hence, special techniques should be developed to study the
thickness change, due to the aging of a commercial NMC material composition and distribution of the SEI components.
using a sputter depth profiling (SDP) method. Even though SDP-
XPS is a good tool that enables the composition of the SEI at
different depths to be understood, sputtering can damage the SEI 7. Conclusions and perspectives
(hence making the analytical results questionable) due to the
formation of additional products such as Li2O. It is also possible Currently commercialized lithium ion batteries are used for
to determine the SEI thickness using time-of-flight secondary ion various applications; however, work directed towards their use
mass spectrometry (ToF-SIMS), using dynamic analysis mode, and in higher power applications is still ongoing. Even though
focused ion beam (FIB) coupled with SEM by taking a cross-sectional success has been limited, there is much effort aimed at
image.291 Lee et al. have compared the power of TOF-SIMS, XPS, and upgrading battery performance by developing new materials
FIB/SEM for SEI characterization. They have found that even though or improving those in current use. Electrolyte additive devel-
all the three techniques are able to identify the SEI thickness change, opment is one of the current research directions aimed at
FIB/SEM was the most time consuming technique.291 Hy et al. improving lithium ion battery performance. This review has
have developed in situ shell-isolated nanoparticle-enhanced summarized various types of electrolyte additives used to
Raman spectroscopy (in situ SINERS) for studying SEI for- improve the performance and safety of lithium ion batteries.
mation on lithium rich layered oxide compounds.302,303 The additives are classified based on their functions for specific
When analyzing species on highly reactive electrode surfaces, electrode materials. Additives that modify the SEI should be
the selection of analysis method or the design of accessories that reduced (anode additives) or oxidized (cathode additives) prior
can avoid contamination should be considered. Li is sensitive to to the electrolyte and passivate the electrode’s surface to pre-
atmospheric components, except noble gases, thus when reac- vent further decomposition of the SEI. Additives for the safety
tive salts, such as LiBF4 and LiPF6 are used, their reduction of the battery are classified as redox shuttles, or flame retar-
products react with H2O, O2, and CO2 to form different inorganic dants. Redox shuttle additives should be electrochemically
compounds, such as LiF and LiO2 on the electrode’s surface. The reversible at slightly higher potentials than the normal end-
presence of such inorganic compounds degrades battery perfor- of-charge potential of the cathode. Moreover, the radical cation
mance by increasing electrode resistance.2,304,305 should be stable. Flame retardant additives should have the
The use of ex situ techniques leads to the interference of ability to scavenge radicals and terminate reactions. Additives
secondary reactions between the SEI layer and atmospheric for thermal runaway (STOBA) and those that improve the
contaminants, such as O2, N2, CO2, H2O, etc. Hence, transfer stability of the salt are also summarized. Although much effort
systems that can avoid the exposure of electrodes to reactive has been directed at improving the performance and safety of
atmospheric contaminants should be designed. The best way to the batteries using electrolyte additives, there are issues that

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1981
 View Article Online

Review Energy & Environmental Science

should be addressed for the high applicability of the battery screening techniques, it is essential to identify the descriptors
and some of which are discussed below. from previous experimental findings for a specific application
(i) Analytical tools for the analysis of the SEI: since the SEI is of the additive. After screening the additives, based on virtual
a determining factor with respect to the performance (affecting high throughput screening, it is also important to perform
the cycle life, power capability, shelf life, and safety) of the experimental validation for the practical applicability of the additive
battery, understanding and controlling it, to either enhance cell and the recipe developed. Moreover, selection of mixtures of addi-
performance or improve safety are vital. The formation of an tives using only the computational method is more complicated
ideal SEI layer is a fundamental challenge in lithium ion battery than selecting a single additive, hence, combining computational
design. An ideal SEI layer should prevent further electrolyte and experimental high throughput screening approaches is essential
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

degradation by blocking electron transport while allowing to develop a ‘recipe’ for a specific system.
Li-ions to pass through during cycling. The ideal SEI should
also have good thermal stability at elevated temperatures and
good mechanical properties that enable it to resist volume Acknowledgements
change during cycling. An ideal SEI, should among its other
The authors gratefully acknowledge financial support from the
functions prevent the dissolution of cathode materials. How-
Ministry of Science and Technology (MOST, Taiwan) (MOST
ever, due to the lack of analytical tools, whether the electrolyte
104-3113-E-011-001, MOST 104-2911-I-011-505-MY2, MOST 104-
additives lead to the formation of such an SEI is not clearly
3113-E-011-002), the Ministry of Economic Affairs (MOEA, Taiwan)
understood. The tools that have been used so far are not good
(101-EC-17-A-08-S1-183), the Top University Projects of Ministry of
enough to study the SEI formation process. Since the SEI film is
Education (MOE, Taiwan) (100H451401), and the National Taiwan
very thin and its composition is complicated, measuring some
University of Science and Technology (NTUST).
SEI properties, such as its mechanical properties, porosity,
lateral composition distribution, composition gradient, solubi-
lity, and stability is still very difficult. Developing powerful tools Notes and references
to characterize the physical and chemical properties of a SEI
film is crucial for a better understanding and engineering of 1 M. Endo, C. Kim, K. Nishimura, T. Fujino and K. Miyashita,
the SEI. Hence, developing in situ analytical tools, either singly Carbon, 2000, 38, 183–197.
or in combination, that enable us to understand the role of the 2 K.-I. Morigaki, J. Power Sources, 2002, 103, 253–264.
additives should be one of the main research directions. 3 L. Li, Z. Wu, S. Yuan and X.-B. Zhang, Energy Environ. Sci.,
(ii) Developing a mixture of additives: one additive may have 2014, 7, 2101–2122.
more than one function on the same electrode, or it may be 4 Z. Ogumi and M. Inaba, in Advances in Lithium-Ion Batteries,
appropriate for use with various electrodes. On the other hand, ed. W. A. V. Schalkwijk and B. Scrosati, Kluwer Academic,
the additive used to ameliorate one property of the battery may New York, 2002, ch. 2, pp. 79–102.
cause a negative impact on another property. E.g. additives that 5 M. Wakihara, Mater. Sci. Eng., R, 2001, 33, 109–134.
are used to improve the safety of the battery (such as FR, or 6 J. Chen, Materials, 2013, 6, 156–183.
overcharge protection) may have a negative effect on the 7 L. J. Hardwick, M. Marcinek, L. Beer, J. B. Kerr and
performance of the battery, or an additive may have a positive R. Kostecki, J. Electrochem. Soc., 2008, 155, A442–A447.
effect on the anode but it may have a negative effect on the 8 Q. Zhang and R. E. White, J. Power Sources, 2008, 179, 793–798.
cathode. Hence, using combinations of additives can lead to 9 D. Aurbach, B. Markovsky, A. Rodkin, M. Cojocaru, E. Levi
cooperative effects. The role of the additive is mainly dependent and H.-J. Kim, Electrochim. Acta, 2002, 47, 1899–1911.
on the functional group incorporated. To develop new multi- 10 D. Zane, A. Antonini and M. Pasquali, J. Power Sources,
functional additives, it is crucial to correlate the property of the 2001, 97–98, 146–150.
battery that we want to ‘tune’ to the effect of the functional 11 J. Yang and S. T. John, J. Mater. Chem. A, 2013, 1, 7782–7789.
group. Even though a lot of additives have been reported for 12 H. K. Liu, Z. P. Guo, J. Z. Wang and K. Konstantinov,
various electrode materials and diverse applications, only a few J. Mater. Chem., 2010, 20, 10055–10057.
have been commercialized. Hence, a lot of effort still needs to be 13 R. Teki, M. K. Datta, R. Krishnan, T. C. Parker, T.-M. Lu,
directed towards additive development for future applications. P. N. Kumta and N. Koratkar, Small, 2009, 5, 2236–2242.
(iii) Developing a recipe for selection of noble electrolyte 14 K. Peng, J. Jie, W. Zhang and S.-T. Lee, Appl. Phys. Lett.,
additives: in most previous studies, specific additives were 2008, 93, 033105.
selected on the basis of a trial and error approach, which is both 15 J. Cho, J. Mater. Chem., 2010, 20, 4009–4014.
time-consuming and inefficient. Hence, instead of developing an 16 Y.-S. Ye, X.-L. Xie, J. Rick, F.-C. Chang and B.-J. Hwang,
‘individual additive selection method’ for a particular purpose, it J. Power Sources, 2014, 247, 991–998.
is better to develop a screening method that is both cost effective 17 M.-Y. Cheng, C.-L. Hwang, C.-J. Pan, J.-H. Cheng, Y.-S. Ye,
and time saving. By using a high-throughput computational J. F. Rick and B.-J. Hwang, J. Mater. Chem., 2011, 21,
material design approach it is now relatively easy to generate, 10705–10710.
manage and analyze vast amounts of data in order to discover 18 G. Derrien, J. Hassoun, S. Panero and B. Scrosati, Adv.
novel materials. To select additives using high-throughput Mater., 2007, 19, 2336–2340.

1982 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

19 S. Liang, X. Zhu, P. Lian, W. Yang and H. Wang, J. Solid 45 S. Komaba, B. Kaplan, T. Ohtsuka, Y. Kataoka, N. Kumagai
State Chem., 2011, 184, 1400–1404. and H. Groult, J. Power Sources, 2003, 119–121, 378–382.
20 Y. Xu, Q. Liu, Y. Zhu, Y. Liu, A. Langrock, M. R. Zachariah 46 Y. Ein-Eli, J. Electroanal. Chem., 2002, 531, 95–99.
and C. Wang, Nano Lett., 2013, 13, 470–474. 47 J.-S. Shin, C.-H. Han, U.-H. Jung, S.-I. Lee, H.-J. Kim and
21 P. Lichty, M. Wirz, P. Kreider, O. Kilbury, D. Dinair, K. Kim, J. Power Sources, 2002, 109, 47–52.
D. King, A. Steinfeld and A. W. Weimer, Int. J. Appl. Ceram. 48 Y. Zhang, L. Wang, H. Feng, A. Zhang, C. Zhang and
Technol., 2013, 10, 257–265. P. Zhang, Ionics, 2011, 17, 677–682.
22 T. Achiha, S. Shibata, T. Nakajima, Y. Ohzawa, 49 Y.-K. Choi, K.-I. Chung, W.-S. Kim, Y.-E. Sung and
A. Tressaud and E. Durand, J. Power Sources, 2007, 171, S.-M. Park, J. Power Sources, 2002, 104, 132–139.
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

932–937. 50 Y. Ein-Eli, S. R. Thomas and V. R. Koch, J. Electrochem. Soc.,

23 H. Buqa, C. Grogger, M. V. Santis Alvarez, J. O. Besenhard 1997, 144, 1159–1165.
and M. Winter, J. Power Sources, 2001, 97–98, 126–128. 51 H. Ota, Y. Sakata, A. Inoue and S. Yamaguchi, J. Electrochem.
24 H.-G. Wang, S. Yuan, D.-L. Ma, X.-B. Zhang and J.-M. Yan, Soc., 2004, 151, A1659–A1669.
Energy Environ. Sci., 2015, 8, 1660–1681. 52 C. Wang, H. Nakamura, H. Komatsu, M. Yoshio and
25 G. G. Amatucci, J. M. Tarascon and L. C. Klein, Solid State H. Yoshitake, J. Power Sources, 1998, 74, 142–145.
Ionics, 1996, 83, 167–173. 53 K. C. Möller, H. J. Santner, W. Kern, S. Yamaguchi, J. O.
26 B. Xu, D. Qian, Z. Wang and Y. S. Meng, Mater. Sci. Eng., R, Besenhard and M. Winter, J. Power Sources, 2003, 119–121,
2012, 73, 51–65. 561–566.
27 H. S. Liu, Z. R. Zhang, Z. L. Gong and Y. Yang, Electrochem. 54 G. Chen, G. V. Zhuang, T. J. Richardson, G. Liu and J. P. N.
Solid-State Lett., 2004, 7, A190–A193. Ross, Electrochem. Solid-State Lett., 2005, 8, A344–A347.
28 Y. S. Lee, Y. K. Sun, K. Adachi and M. Yoshio, Electrochim. 55 R. Chen, F. Wu, L. Li, Y. Guan, X. Qiu, S. Chen, Y. Li and
Acta, 2003, 48, 1031–1039. S. Wu, J. Power Sources, 2007, 172, 395–403.
29 Z. Shi-xi, L. Han-xing, O. Shi-xi and L. Qiang, J. Wuhan Univ. 56 E. G. Leggesse and J.-C. Jiang, RSC Adv., 2012, 2, 5439–5446.
Technol., Mater. Sci. Ed., 2003, 18, 5–8. 57 G. H. Wrodnigg, J. O. Besenhard and M. Winter, J. Electrochem.
30 T. Ohzuku and Y. Makimura, Chem. Lett., 2001, 744–745. Soc., 1999, 146, 470–472.
31 X. Meng, S. Dou and W.-L. Wang, J. Power Sources, 2008, 58 G. H. Wrodnigg, T. M. Wrodnigg, J. O. Besenhard and
184, 489–493. M. Winter, Electrochem. Commun., 1999, 1, 148–150.
32 K. Kang, Y. S. Meng, J. Bréger, C. P. Grey and G. Ceder, 59 E. G. Leggesse and J.-C. Jiang, J. Phys. Chem. A, 2012, 116,
Science, 2006, 311, 977–980. 11025–11033.
33 B. J. Hwang, Y. W. Wu, M. Venkateswarlu, M. Y. Cheng and 60 G. H. Wrodnigg, J. O. Besenhard and M. Winter, J. Power
R. Santhanam, J. Power Sources, 2009, 193, 828–833. Sources, 2001, 97–98, 592–594.
34 P. Prosini, Iron Phosphate Materials as Cathodes for Lithium 61 Z. X. Shu, R. S. McMillan, J. J. Murray and I. J. Davidson,
Batteries, Springer, London, 2011, ch. 1, pp. 1–12, DOI: J. Electrochem. Soc., 1996, 143, 2230–2235.
10.1007/978-0-85729-745-7_1. 62 R. McMillan, H. Slegr, Z. X. Shu and W. Wang, J. Power
35 S.-W. Song, G. V. Zhuang and P. N. Ross, J. Electrochem. Sources, 1999, 81–82, 20–26.
Soc., 2004, 151, A1162–A1167. 63 A. Naji, J. Ghanbaja, P. Willmann and D. Billaud, Electro-
36 H. Cheng, C. B. Zhu, M. Lu and Y. Yang, J. Power Sources, chim. Acta, 2000, 45, 1893–1899.
2007, 174, 1027–1031. 64 Y. Matsuo, K. Fumita, T. Fukutsuka, Y. Sugie, H. Koyama
37 R. Sharabi, E. Markevich, V. Borgel, G. Salitra, D. Aurbach, and K. Inoue, J. Power Sources, 2003, 119–121, 373–377.
G. Semrau and M. A. Schmidt, Electrochem. Solid-State 65 H. Ota, A. Kominato, W.-J. Chun, E. Yasukawa and S. Kasuya,
Lett., 2010, 13, A32–A35. J. Power Sources, 2003, 119–121, 393–398.
38 G. M. Ehrlich, in Handbook of batteries, ed. D. Linden 66 Y. Ma, G. Yin, P. Zuo, X. Tan, Y. Gao and P. Shi, Electro-
and T. B. Reddy, McGraw-Hill, New York, 2002, ch. 35, chem. Solid-State Lett., 2008, 11, A129–A131.
pp. 35.31–35.94. 67 J.-A. Choi, S.-M. Eo, D. R. MacFarlane, M. Forsyth, E. Cha
39 J. Li, N. J. Dudney, J. Nanda and C. Liang, ACS Appl. Mater. and D.-W. Kim, J. Power Sources, 2008, 178, 832–836.
Interfaces, 2014, 6, 10083–10088. 68 M. Walkowiak, D. Waszak, G. Schroeder and B. Gierczyk,
40 J. Zhao, Z. Lu, H. Wang, W. Liu, H.-W. Lee, K. Yan, J. Solid State Electrochem., 2010, 14, 2213–2218.
D. Zhuo, D. Lin, N. Liu and Y. Cui, J. Am. Chem. Soc., 69 Q. Huang, T. Wang, M. Yan and Z. Jiang, J. Electrochem.
2015, 137, 8372–8375. Soc., 2006, 153, A2363–A2369.
41 F.-S. Li, Y.-S. Wu, J. Chou, M. Winter and N.-L. Wu, Adv. 70 T. Sasaki, T. Abe, Y. Iriyama, M. Inaba and Z. Ogumi,
Mater., 2015, 27, 130–137. J. Electrochem. Soc., 2005, 152, A2046–A2050.
42 K. Xu, Chem. Rev., 2004, 104, 4303–4418. 71 C.-C. Chang, S.-H. Hsu, Y.-F. Jung and C.-H. Yang, J. Power
43 S. S. Zhang, J. Power Sources, 2006, 162, 1379–1394. Sources, 2011, 196, 9605–9611.
44 J. O. Besenhard, M. W. Wagner, M. Winter, A. D. 72 O. Matsuoka, A. Hiwara, T. Omi, M. Toriida, T. Hayashi,
Jannakoudakis, P. D. Jannakoudakis and E. Theodoridou, C. Tanaka, Y. Saito, T. Ishida, H. Tan, S. S. Ono and
J. Power Sources, 1993, 44, 413–420. S. Yamamoto, J. Power Sources, 2002, 108, 128–138.

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1983
 View Article Online

Review Energy & Environmental Science

73 S.-D. Xu, Q.-C. Zhuang, J. Wang, Y.-Q. Xu and Y.-B. Zhu, Int. 99 B. Zhang, M. Metzger, S. Solchenbach, M. Payne, S. Meini,
J. Electrochem. Sci., 2013, 8, 8058–8076. H. A. Gasteiger, A. Garsuch and B. L. Lucht, J. Phys. Chem.
74 S. Ergun, C. F. Elliott, A. P. Kaur, S. R. Parkin and C, 2015, 119, 11337–11348.
S. A. Odom, J. Phys. Chem. C, 2014, 118, 14824–14832. 100 T. Yong, J. Wang, Y. Mai, D. Tang and L. Zhang, Ionics,
75 F.-M. Wang, M.-H. Yu, C.-S. Cheng, S. A. Pradanawati, 2013, 19, 1099–1103.
S.-C. Lo and J. Rick, J. Power Sources, 2013, 231, 18–22. 101 C. Korepp, W. Kern, E. A. Lanzer, P. R. Raimann,
76 F.-M. Wang, H.-M. Cheng, H.-C. Wu, S.-Y. Chu, C.-S. J. O. Besenhard, M. Yang, K. C. Moller, D. T. Shieh and
Cheng and C.-R. Yang, Electrochim. Acta, 2009, 54, M. Winter, J. Power Sources, 2007, 174, 628–631.
3344–3351. 102 H. M. Jung, S.-H. Park, J. Jeon, Y. Choi, S. Yoon, J.-J. Cho,
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

77 J. Jeon, S. Yoon, T. Park, J.-J. Cho, S. Kang, Y.-K. Han and S. Oh, S. Kang, Y.-K. Han and H. Lee, J. Mater. Chem. A,
H. Lee, J. Mater. Chem., 2012, 22, 21003–21008. 2013, 1, 11975–11981.
78 Y. Qin, Z. Chen, W. Lu and K. Amine, J. Power Sources, 103 S. Yonezawa and M. Takashima, in Fluorinated Materials
2010, 195, 6888–6892. for Energy Conversion, ed. T. Nakajima and H. Groult,
79 T. Inose, S. Tada, H. Morimoto and S.-I. Tobishima, Elsevier Ltd, Netherlands, 2005, ch. 6, pp. 125–136.
J. Power Sources, 2006, 161, 550–559. 104 Z. X. Shu, R. S. McMillan, J. J. Murray and I. J. Davidson,
80 T. Takeuchi, S. Noguchi, H. Morimoto and S.-I. Tobishima, J. Electrochem. Soc., 1995, 142, L161–L162.
J. Power Sources, 2010, 195, 580–587. 105 B. Li, Y. Wang, H. Lin, X. Wang, M. Xu, Y. Wang, L. Xing
81 W. Li and B. L. Lucht, Electrochem. Solid-State Lett., 2007, and W. Li, J. Power Sources, 2014, 267, 182–187.
10, A115–A117. 106 L. L. Li, L. Li, B. Wang, L. L. Liu, Y. P. Wu, T. van Ree and
82 S. Dalavi, M. Xu, B. Knight and B. L. Lucht, Electrochem. K. A. Thavhiwa, Electrochim. Acta, 2011, 56, 4858–4864.
Solid-State Lett., 2011, 15, A28–A31. 107 B. Wu, Y. Ren, D. Mu, C. Zhang, X. Liu, G. Yang and F. Wu,
83 C.-C. Chang, L.-J. Her, L.-C. Chen, Y.-Y. Lee, S.-J. Liu and Int. J. Electrochem. Sci., 2013, 8, 670–677.
H.-J. Tien, J. Power Sources, 2007, 163, 1059–1063. 108 S. Komaba, T. Itabashi, B. Kaplan, H. Groult and
84 M. Arakawa and J. I. Yamaki, J. Power Sources, 1995, 54, N. Kumagai, Electrochem. Commun., 2003, 5, 962–966.
250–254. 109 S. Komaba, T. Itabashi, T. Ohtsuka, H. Groult, N. Kumagai,
85 E. Krämer, R. Schmitz, S. Passerini, M. Winter and C. Schreiner, B. Kaplan and H. Yashiro, J. Electrochem. Soc., 2005, 152,
Electrochem. Commun., 2012, 16, 41–43. A937–A946.
86 G. Park, H. Nakamura, Y. Lee and M. Yoshio, J. Power 110 J.-H. Felix Cheng, S. Hy, J. Rick, F.-M. Wang and B.-J.
Sources, 2009, 189, 602–606. Hwang, J. Phys. Chem. C, 2013, 117, 22619–22626.
87 W. Yao, Z. Zhang, J. Gao, J. Li, J. Xu, Z. Wang and Y. Yang, 111 N.-S. Choi, K. H. Yew, K. Y. Lee, M. Sung, H. Kim and
Energy Environ. Sci., 2009, 2, 1102–1108. S.-S. Kim, J. Power Sources, 2006, 161, 1254–1259.
88 Y. Hu, W. Kong, Z. Wang, X. Huang and L. Chen, Solid 112 R. Lv, J. Yang, J. Wang and Y. NuLi, J. Power Sources, 2011,
State Ionics, 2005, 176, 53–56. 196, 3868–3873.
89 J.-T. Lee, M.-S. Wu, F.-M. Wang, Y.-W. Lin, M.-Y. Bai and 113 V. Etacheri, O. Haik, Y. Goffer, G. A. Roberts, I. C. Stefan,
P.-C. J. Chiang, J. Electrochem. Soc., 2005, 152, A1837–A1843. R. Fasching and D. Aurbach, Langmuir, 2012, 28, 965–976.
90 H. J. Santner, K. C. Möller, J. Ivančo, M. G. Ramsey, F. P. 114 J. S. Kim, D. Byun and J. K. Lee, Curr. Appl. Phys., 2014, 14,
Netzer, S. Yamaguchi, J. O. Besenhard and M. Winter, 596–602.
J. Power Sources, 2003, 119–121, 368–372. 115 A. Bordes, K. Eom and T. F. Fuller, J. Power Sources, 2014,
91 R. Mogi, M. Inaba, S.-K. Jeong, Y. Iriyama, T. Abe and 257, 163–169.
Z. Ogumi, J. Electrochem. Soc., 2002, 149, A1578–A1583. 116 L. Chen, K. Wang, X. Xie and J. Xie, J. Power Sources, 2007,
92 K. Abe, H. Yoshitake, T. Kitakura, T. Hattori, H. Wang and 174, 538–543.
M. Yoshio, Electrochim. Acta, 2004, 49, 4613–4622. 117 S. Hy, Y.-H. Chen, H.-M. Cheng, C.-J. Pan, J.-H. Cheng,
93 M. Khasanov, E. Pazhetnov and W. C. Shin, J. Electrochem. J. Rick and B.-J. Hwang, ACS Appl. Mater. Interfaces, 2015, 7,
Soc., 2015, 162, A1892–A1898. 13801–13807.
94 Z. Honghe, L. Gao and S. B. Vincent, J. Phys. Chem. C, 2010, 118 G.-B. Han, M.-H. Ryou, K. Y. Cho, Y. M. Lee and J.-K. Park,
114, 6182–6189. J. Power Sources, 2010, 195, 3709–3714.
95 X.-G. Sun, C. Liao, L. Baggetto, B. Guo, R. R. Unocic, 119 S. Dalavi, P. Guduru and B. L. Lucht, J. Electrochem. Soc.,
G. M. Veith and S. Dai, J. Mater. Chem. A, 2014, 2, 2012, 159, A642–A646.
7606–7614. 120 C. C. Nguyen and B. L. Lucht, J. Electrochem. Soc., 2014,
96 X. Zuo, M. Xu, W. Li, D. Su and J. Liu, Electrochem. Solid- 161, A1933–A1938.
State Lett., 2006, 9, A196–A199. 121 I. A. Profatilova, C. Stock, A. Schmitz, S. Passerini and
97 M. Q. Xu, W. S. Li, X. X. Zuo, J. S. Liu and X. Xu, J. Power M. Winter, J. Power Sources, 2013, 222, 140–149.
Sources, 2007, 174, 705–710. 122 S. Park, J. Heon Ryu and S. M. Oh, J. Electrochem. Soc.,
98 R. F. Lu, D. W. Rao, Z. S. Meng, X. B. Zhang, G. J. Xu, 2011, 158, A498–A503.
Y. Z. Liu, E. J. Kan, C. Y. Xiao and K. M. Deng, Phys. Chem. 123 T.-F. Yi, L.-J. Jiang, J. Shu, C.-B. Yue, R.-S. Zhu and
Chem. Phys., 2013, 15, 16120–16126. H.-B. Qiao, J. Phys. Chem. Solids, 2010, 71, 1236–1242.

1984 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

124 Y.-B. He, M. Liu, Z.-D. Huang, B. Zhang, Y. Yu, B. Li, 149 Y. An, P. Zuo, X. Cheng, L. Liao and G. Yin, Electrochim.
F. Kang and J.-K. Kim, J. Power Sources, 2013, 239, 269–276. Acta, 2011, 56, 4841–4848.
125 M.-S. Song, R.-H. Kim, S.-W. Baek, K.-S. Lee, K. Park and 150 Y. Zhu, Y. Li, M. Bettge and D. P. Abraham, Electrochim.
A. Benayad, J. Mater. Chem. A, 2014, 2, 631–636. Acta, 2013, 110, 191–199.
126 R. Dedryvere, D. Foix, S. Franger, S. Patoux, L. Daniel and 151 Y. Zhu, Y. Li, M. Bettge and D. P. Abraham, J. Electrochem.
D. Gonbeau, J. Phys. Chem. C, 2010, 114, 10999–11008. Soc., 2012, 159, A2109–A2117.
127 Y.-B. He, B. Li, M. Liu, C. Zhang, W. Lv, C. Yang, J. Li, 152 J. Xia, N. N. Sinha, L. P. Chen and J. R. Dahn,
H. Du, B. Zhang, Q.-H. Yang, J.-K. Kim and F. Kang, Sci. J. Electrochem. Soc., 2014, 161, A264–A274.
Rep., 2012, 2, 913. 153 J. Xia, L. Ma, C. P. Aiken, K. J. Nelson, L. P. Chen and
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

128 R.-H. Wang, X.-H. Li, Z.-X. Wang, H.-J. Guo and Z.-J. He, J. R. Dahn, J. Electrochem. Soc., 2014, 161, A1634–A1641.
ACS Appl. Mater. Interfaces, 2015, 7, 23605–23614. 154 L. Ma, J. Xia and J. R. Dahn, J. Electrochem. Soc., 2015, 162,
129 R. Wang, X. Li, B. Zhang, Z. Wang and H. Guo, J. Alloys A1170–A1174.
Compd., 2015, 648, 512–520. 155 N. N. Sinha, J. C. Burns and J. R. Dahn, J. Electrochem. Soc.,
130 H. Bouayad, Z. Wang, N. Dupre, R. Dedryvere, D. Foix, 2014, 161, A1084–A1089.
S. Franger, J.-F. Martin, L. Boutafa, S. Patoux and 156 J. Xia, N. N. Sinha, L. P. Chen, G. Y. Kim, D. J. Xiong and
D. Gonbeau, J. Phys. Chem. C, 2014, 118, 4634–4648. J. R. Dahn, J. Electrochem. Soc., 2014, 161, A84–A88.
131 J. H. Kim, S.-y. Bae, J.-H. Min, S.-W. Song and D.-W. Kim, 157 S. Mai, M. Xu, X. Liao, J. Hu, H. Lin, L. Xing, Y. Liao, X. Li
Electrochim. Acta, 2012, 78, 11–16. and W. Li, Electrochim. Acta, 2014, 147, 565–571.
132 C. Julien, A. Mauger, K. Zaghib and H. Groult, Inorganics, 158 S. Hy, F. Felix, J. Rick, W.-N. Su and B. J. Hwang, J. Am.
2014, 2, 132. Chem. Soc., 2014, 136, 999–1007.
133 K. Abe, T. Takaya, H. Yoshitake, Y. Ushigoe, M. Yoshio and 159 T. Yim, S. H. Kim, S.-G. Woo, K. Lee, J. H. Song, W. Cho,
H. Wang, Electrochem. Solid-State Lett., 2004, 7, A462–A465. K. J. Kim, J.-S. Kim and Y.-J. Kim, RSC Adv., 2014, 4,
134 L. Xia, Y. Xia and Z. Liu, Electrochim. Acta, 2015, 151, 19172–19176.
429–436. 160 J. Pires, L. Timperman, A. Castets, J. S. Pena, E. Dumont,
135 J.-N. Lee, G.-B. Han, M.-H. Ryou, D. J. Lee, S. Jongchan, S. Levasseur, R. Dedryvere, C. Tessier and M. Anouti, RSC
J. W. Choi and J.-K. Park, Electrochim. Acta, 2011, 56, Adv., 2015, 5, 42088–42094.
5195–5200. 161 Z. Zhang, L. Hu, H. Wu, W. Weng, M. Koh, P. C. Redfern,
136 J. Yang, P. Zhao, Y. Shang, L. Wang, X. He, M. Fang and L. A. Curtiss and K. Amine, Energy Environ. Sci., 2013, 6,
J. Wang, Electrochim. Acta, 2014, 121, 264–269. 1806–1810.
137 J. Yang, Y. Zhang, P. Zhao, Y. Shang, L. Wang, X. He and 162 H. Q. Pham, K.-M. Nam, E.-H. Hwang, Y.-G. Kwon,
J. Wang, Electrochim. Acta, 2015, 158, 202–208. H. M. Jung and S.-W. Song, J. Electrochem. Soc., 2014,
138 Y.-H. Li, M.-L. Lee, F.-M. Wang, C.-R. Yang, P. P. J. Chu and 161, A2002–A2011.
J.-P. Pan, Electrochim. Acta, 2012, 85, 72–77. 163 J. Zhang, J. Wang, J. Yang and Y. NuLi, Electrochim. Acta,
139 T. Takeuchi, T. Kyuna, H. Morimoto and S.-I. Tobishima, 2014, 117, 99–104.
J. Power Sources, 2011, 196, 2790–2801. 164 J. Li, L. Xing, R. Zhang, M. Chen, Z. Wang, M. Xu and
140 B. Wu, Y. Ren, D. Mu, X. Liu, G. Yang and F. Wu, RSC Adv., W. Li, J. Power Sources, 2015, 285, 360–366.
2014, 4, 10196–10203. 165 D. J. Lee, D. Im, Y.-G. Ryu, S. Lee, J. Yoon, J. Lee, W. Choi,
141 J. Liu, N. Liu, D. Liu, Y. Bai, L. Shi, Z. Wang, L. Chen, I. Jung, S. Lee and S.-G. Doo, J. Power Sources, 2013, 243,
V. Hennige and A. Schuch, J. Electrochem. Soc., 2007, 154, 831–835.
A55–A63. 166 Y.-S. Kang, T. Yoon, S. S. Lee, J. Mun, M. S. Park, J.-H. Park,
142 Y. M. Lee, Y.-G. Lee, Y.-M. Kang and K. Y. Cho, Electrochem. S.-G. Doo, I.-Y. Song and S. M. Oh, Electrochem. Commun.,
Solid-State Lett., 2010, 13, A55–A58. 2013, 27, 26–28.
143 C.-H. Chen, C.-J. Wang and B.-J. Hwang, J. Power Sources, 167 H. Yamane, T. Inoue, M. Fujita and M. Sano, J. Power
2005, 146, 626–629. Sources, 2001, 99, 60–65.
144 Y. Zhu, M. D. Casselman, Y. Li, A. Wei and D. P. Abraham, 168 Y. Li, R. Zhang, J. Liu and C. Yang, J. Power Sources, 2009,
J. Power Sources, 2014, 246, 184–191. 189, 685–688.
145 F. Chrétien, J. Jones, C. Damas, D. Lemordant, 169 X. Wu, X. Li, Z. Wang, H. Guo and L. Xiong, Ionics, 2012,
P. Willmann and M. Anouti, J. Power Sources, 2014, 248, 18, 907–911.
969–977. 170 Y. Liu, L. Tan and L. Li, J. Power Sources, 2013, 221,
146 J. C. Burns, N. N. Sinha, G. Jain, H. Ye, C. M. VanElzen, 90–96.
W. M. Lamanna, A. Xiao, E. Scott, J. Choi and J. R. Dahn, 171 H. Rong, M. Xu, L. Xing and W. Li, J. Power Sources, 2014,
J. Electrochem. Soc., 2012, 159, A1095–A1104. 261, 148–155.
147 P. Ping, X. Xia, Q. S. Wang, J. H. Sun and J. R. Dahn, 172 M. Xu, Y. Liang, B. Li, L. Xing, Y. Wang and W. Li, Mater.
J. Electrochem. Soc., 2013, 160, A426–A429. Chem. Phys., 2014, 143, 1048–1054.
148 G.-Y. Kim and J. R. Dahn, J. Electrochem. Soc., 2015, 162, 173 Z. Wang, Y. Huang, X. Wang, D. Jia, Z. Guo and M. Miao,
A437–A447. Solid State Ionics, 2013, 232, 19–23.

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1985
 View Article Online

Review Energy & Environmental Science

174 M. Sano, T. Hattori, T. Hibino and M. Fujita, Electrochem. 200 C. Buhrmester, L. Moshurchak, R. L. Wang and J. R. Dahn,
Solid-State Lett., 2007, 10, A270–A273. J. Electrochem. Soc., 2006, 153, A288–A294.
175 B. Li, Y. Wang, H. Rong, Y. Wang, J. Liu, L. Xing, M. Xu and 201 M. D. Casselman, A. P. Kaur, K. A. Narayana, C. F. Elliott,
W. Li, J. Mater. Chem. A, 2013, 1, 12954–12961. C. Risko and S. A. Odom, Phys. Chem. Chem. Phys., 2015,
176 X. Zuo, J. Wu, C. Fan, K. Lai, J. Liu and J. Nan, Electrochim. 17, 6905–6912.
Acta, 2014, 130, 778–784. 202 K. A. Narayana, M. D. Casselman, C. F. Elliott, S. Ergun, S. R.
177 M. Cuisinier, N. Dupré and D. Guyomard, Electrochem. Parkin, C. Risko and S. A. Odom, ChemPhysChem, 2015, 16,
Commun., 2014, 38, 138–141. 1179–1189.
178 C.-C. Chang and T.-K. Chen, J. Power Sources, 2009, 193, 203 S. A. Odom, S. Ergun, P. P. Poudel and S. R. Parkin, Energy
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

834–840. Environ. Sci., 2014, 7, 760–767.

179 C.-C. Chang, K.-Y. Lee, H.-Y. Lee, Y.-H. Su and L.-J. Her, 204 L. M. Moshurchak, C. Buhrmester and J. R. Dahn,
J. Power Sources, 2012, 217, 524–529. J. Electrochem. Soc., 2008, 155, A129–A131.
180 Y. Qin, Z. Chen, H. S. Lee, X. Q. Yang and K. Amine, J. Phys. 205 S. L. Li, X. P. Ai, J. K. Feng, Y. L. Cao and H. X. Yang,
Chem. C, 2010, 114, 15202–15206. J. Power Sources, 2008, 184, 553–556.
181 L. Liao, T. Fang, X. Zhou, Y. Gao, X. Cheng, L. Zhang and 206 Z. Chen, J. Liu, A. N. Jansen, G. GirishKumar, B. Casteel and
G. Yin, Solid State Ionics, 2014, 254, 27–31. K. Amine, Electrochem. Solid-State Lett., 2010, 13, A39–A42.
182 L. Liao, X. Cheng, Y. Ma, P. Zuo, W. Fang, G. Yin and 207 M. Tang and J. Newman, J. Electrochem. Soc., 2011, 158,
Y. Gao, Electrochim. Acta, 2013, 87, 466–472. A530–A536.
183 B. Wu, Y. Ren, D. Mu, X. Liu, G. Yang and Z. Sun, Solid 208 S. Kaymaksiz, M. Wachtler and M. Wohlfahrt-Mehrens,
State Ionics, 2014, 260, 8–14. J. Power Sources, 2015, 273, 123–127.
184 T. J. Richardson and P. N. Ross, J. Electrochem. Soc., 1996, 209 L. M. Moshurchak, C. Buhrmester, R. L. Wang and
143, 3992–3996. J. R. Dahn, Electrochim. Acta, 2007, 52, 3779–3784.
185 Z. Chen and K. Amine, Electrochem. Commun., 2007, 9, 210 G. Chen, K. E. Thomas-Alyea, J. Newman and T. J. Richardson,
703–707. Electrochim. Acta, 2005, 50, 4666–4673.
186 J. Chen, C. Buhrmester and J. Dahn, Electrochem. Solid- 211 L. F. Xiao, X. P. Ai, Y. L. Cao, Y. D. Wang and H. X. Yang,
State Lett., 2005, 8, A59–A62. Electrochem. Commun., 2005, 7, 589–592.
187 Z. Chen, Y. Qin and K. Amine, Electrochim. Acta, 2009, 54, 212 J. K. Feng, X. P. Ai, Y. L. Cao and H. X. Yang, J. Power
5605–5613. Sources, 2006, 161, 545–549.
188 M. Adachi, K. Tanaka and K. Sekai, J. Electrochem. Soc., 213 J.-W. Wen, D.-W. Zhang, C.-H. Chen, C.-X. Ding, Y. Yu and
1999, 146, 1256–1261. J. Maier, J. Power Sources, 2014, 264, 155–160.
189 J. R. Dahn, J. Jiang, L. M. Moshurchak, M. D. Fleischauer, 214 H. F. Xiang, H. Y. Xu, Z. Z. Wang and C. H. Chen, J. Power
C. Buhrmester and L. J. Krause, J. Electrochem. Soc., 2005, Sources, 2007, 173, 562–564.
152, A1283–A1289. 215 J. Feng, P. Ma, H. Yang and L. Lu, Electrochim. Acta, 2013,
190 Z. Zhang, L. Zhang, J. A. Schlueter, P. C. Redfern, L. Curtiss 114, 688–692.
and K. Amine, J. Power Sources, 2010, 195, 4957–4962. 216 R. P. Dunn, J. Kafle, F. C. Krause, C. Hwang, B. V. Ratnakumar,
191 W. Weng, Z. Zhang, P. C. Redfern, L. A. Curtiss and M. C. Smart and B. L. Lucht, J. Electrochem. Soc., 2012, 159,
K. Amine, J. Power Sources, 2011, 196, 1530–1536. A2100–A2108.
192 J. K. Feng, X. P. Ai, Y. L. Cao and H. X. Yang, Electrochem. 217 J. K. Feng, Y. L. Cao, X. P. Ai and H. X. Yang, Electrochim.
Commun., 2007, 9, 25–30. Acta, 2008, 53, 8265–8268.
193 L. M. Moshurchak, W. M. Lamanna, M. Bulinski, R. L. Wang, 218 D. Zhou, W. Li, C. Tan, X. Zuo and Y. Huang, J. Power
R. R. Garsuch, J. Jiang, D. Magnuson, M. Triemert and Sources, 2008, 184, 589–592.
J. R. Dahn, J. Electrochem. Soc., 2009, 156, A309–A312. 219 T.-H. Nam, E.-G. Shim, J.-G. Kim, H.-S. Kim and S.-I. Moon,
194 L. Zhang, Z. Zhang, H. Wu and K. Amine, Energy Environ. J. Power Sources, 2008, 180, 561–567.
Sci., 2011, 4, 2858–2862. 220 Q. Wang, P. Ping, J. Sun and C. Chen, J. Power Sources,
195 L. Chen, M. Xu, B. Li, L. Xing, Y. Wang and W. Li, J. Power 2010, 195, 7457–7461.
Sources, 2013, 244, 499–504. 221 E.-G. Shim, T.-H. Nam, J.-G. Kim, H.-S. Kim and S.-I. Moon,
196 J. Huang, N. Azimi, L. Cheng, I. A. Shkrob, Z. Xue, J. Zhang, Electrochim. Acta, 2009, 54, 2276–2283.
N. L. Dietz Rago, L. A. Curtiss, K. Amine, Z. Zhang and 222 Z. Jin, H. Gao, C. Kong, H. Zhan and Z. Li, ECS Electrochem.
L. Zhang, J. Mater. Chem. A, 2015, 3, 10710–10714. Lett., 2013, 2, A66–A68.
197 J. Huang, I. A. Shkrob, P. Wang, L. Cheng, B. Pan, M. He, 223 S. Chen, Z. Wang, H. Zhao, H. Qiao, H. Luan and L. Chen,
C. Liao, Z. Zhang, L. A. Curtiss and L. Zhang, J. Mater. J. Power Sources, 2009, 187, 229–232.
Chem. A, 2015, 3, 7332–7337. 224 T. Tsujikawa, K. Yabuta, T. Matsushita, T. Matsushima,
198 C. Buhrmester, L. M. Moshurchak, R. L. Wang and K. Hayashi and M. Arakawa, J. Power Sources, 2009, 189, 429–434.
J. R. Dahn, J. Electrochem. Soc., 2006, 153, A1800–A1804. 225 H. W. Rollins, M. K. Harrup, E. J. Dufek, D. K. Jamison,
199 M. Taggougui, B. Carré, P. Willmann and D. Lemordant, S. V. Sazhin, K. L. Gering and D. L. Daubaras, J. Power
J. Power Sources, 2007, 174, 643–647. Sources, 2014, 263, 66–74.

1986 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016
 View Article Online

Energy & Environmental Science Review

226 L. Xia, Y. Xia and Z. Liu, J. Power Sources, 2015, 278, 253 X. Wang, E. Yasukawa and S. Kasuya, J. Electrochem. Soc.,
190–196. 2001, 148, A1066–A1071.
227 K. Kim, S. Ahn, H.-S. Kim and H. K. Liu, Electrochim. Acta, 254 X. Wang, E. Yasukawa and S. Kasuya, J. Electrochem. Soc.,
2009, 54, 2259–2265. 2001, 148, A1058–A1065.
228 B. Wu, F. Pei, Y. Wu, R. Mao, X. Ai, H. Yang and Y. Cao, 255 H. Mao and U. Von Sacken, U.S. Pat., 5776627, 1998.
J. Power Sources, 2013, 227, 106–110. 256 H. Mao and U. Von Sacken, U.S. Pat., 6033797, 2000.
229 H. P. Zhang, Q. Xia, B. Wang, L. C. Yang, Y. P. Wu, 257 J. Reimers and B. Way, U.S. Pat., 6074777, 2000.
D. L. Sun, C. L. Gan, H. J. Luo, A. W. Bebeda and 258 C.-C. Lin, H.-C. Wu, J.-P. Pan, C.-Y. Su, T.-H. Wang, H.-S.
T. v. Ree, Electrochem. Commun., 2009, 11, 526–529. Sheu and N.-L. Wu, Electrochim. Acta, 2013, 101, 11–17.
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

230 Z. Zeng, X. Jiang, B. Wu, L. Xiao, X. Ai, H. Yang and Y. Cao, 259 C.-R. Yang, J.-P. Pan, C.-A. Chen and J.-M. Hsu, 2012.
Electrochim. Acta, 2014, 129, 300–304. 260 H.-M. Liu, D. Saikia, H.-C. Wu, C.-Y. Su, T.-H. Wang, Y.-H. Li,
231 H. Jia, J. Wang, F. Lin, C. W. Monroe, J. Yang and Y. NuLi, J.-P. Pan and H.-M. Kao, RSC Adv., 2014, 4, 56147–56155.
Chem. Commun., 2014, 50, 7011–7013. 261 C. R. Yang, J. P. Pan, C. A. Chen and J. M. Hsu, U.S. Pat.,
232 X. Zhu, X. Jiang, X. Ai, H. Yang and Y. Cao, Electrochim. 8137838, 2012.
Acta, 2015, 165, 67–71. 262 V. Aravindan, J. Gnanaraj, S. Madhavi and H.-K. Liu,
233 D. G. Belov and D. T. Shieh, J. Power Sources, 2014, 247, Chem. – Eur. J., 2011, 17, 14326–14346.
865–875. 263 P. Ping, Q. Wang, J. Sun, H. Xiang and C. Chen,
234 J. Feng and L. Lu, J. Power Sources, 2013, 243, 29–32. J. Electrochem. Soc., 2010, 157, A1170–A1176.
235 J. Hu, Z. Jin, H. Zhong, H. Zhan, Y. Zhou and Z. Li, J. Power 264 J. Liu, Z. Chen, S. Busking and K. Amine, Electrochem.
Sources, 2012, 197, 297–300. Commun., 2007, 9, 475–479.
236 D. Gao, J. B. Xu, M. Lin, Q. Xu, C. F. Ma and H. F. Xiang, 265 Y. Qin, Z. Chen, J. Liu and K. Amine, Electrochem. Solid-
RSC Adv., 2015, 5, 17566–17571. State Lett., 2010, 13, A11–A14.
237 J. Zheng, X. Li, Y. Yu, X. Feng and Y. Zhao, J. Therm. Anal. 266 M. Xu, L. Zhou, L. Hao, L. Xing, W. Li and B. L. Lucht,
Calorim., 2014, 117, 319–324. J. Power Sources, 2011, 196, 6794–6801.
238 M. F. Rectenwald, J. R. Gaffen, A. L. Rheingold, A. B. 267 S. J. Lee, J.-G. Han, Y. Lee, M.-H. Jeong, W. C. Shin, M. Ue
Morgan and J. D. Protasiewicz, Angew. Chem., Int. Ed., and N.-S. Choi, Electrochim. Acta, 2014, 137, 1–8.
2014, 53, 4173–4176. 268 N. Leifer, M. C. Smart, G. K. S. Prakash, L. Gonzalez,
239 J.-A. Choi, Y.-K. Sun, E.-G. Shim, B. Scrosati and D.-W. Kim, L. Sanchez, K. A. Smith, P. Bhalla, C. P. Grey and
Electrochim. Acta, 2011, 56, 10179–10184. S. G. Greenbaum, J. Electrochem. Soc., 2011, 158, A471–A480.
240 S. Wilken, S. Xiong, J. Scheers, P. Jacobsson and 269 N. Dupré, M. Cuisinier and D. Guyomard, Electrochem. Soc.
P. Johansson, J. Power Sources, 2015, 275, 935–942. Interface, 2011, 61–67.
241 H. Nakagawa, Y. Fujino, S. Kozono, Y. Katayama, 270 M. C. Smart, B. V. Ratnakumar, S. Surampudi, Y. Wang,
T. Nukuda, H. Sakaebe, H. Matsumoto and K. Tatsumi, X. Zhang, S. G. Greenbaum, A. Hightower, C. C. Ahn and
J. Power Sources, 2007, 174, 1021–1026. B. Fultz, J. Electrochem. Soc., 1999, 146, 3963–3969.
242 B. S. Lalia, N. Yoshimoto, M. Egashira and M. Morita, 271 L. Gireaud, S. Grugeon, S. Laruelle, S. Pilard and J.-M.
J. Power Sources, 2010, 195, 7426–7431. Tarascon, J. Electrochem. Soc., 2005, 152, A850–A857.
243 Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun and C. Chen, 272 Y. Dai, Y. Wang, V. Eshkenazi, E. Peled and S. G. Greenbaum,
J. Power Sources, 2012, 208, 210–224. J. Electrochem. Soc., 1998, 145, 1179–1183.
244 R. A. Leising, M. J. Palazzo, E. S. Takeuchi and 273 N. Leifer, M. Smart, G. Prakash, L. Gonzalez, L. Sanchez,
K. J. Takeuchi, J. Electrochem. Soc., 2001, 148, A838–A844. K. Smith, P. Bhalla, C. Grey and S. Greenbaum,
245 A. Du Pasquier, F. Disma, T. Bowmer, A. Gozdz, G. Amatucci J. Electrochem. Soc., 2011, 158, A471–A480.
and J. M. Tarascon, J. Electrochem. Soc., 1998, 145, 472–477. 274 H. Ota, T. Akai, H. Namita, S. Yamaguchi and M. Nomura,
246 J. Cho, B. Park and Y.-K. Sun, Lithium-Ion Batteries, J. Power Sources, 2003, 119, 567–571.
Springer, 2009, pp. 207–239. 275 M. Balasubramanian, H. S. Lee, X. Sun, X. Yang,
247 K. Xu, S. Zhang, J. L. Allen and T. R. Jow, J. Electrochem. A. Moodenbaugh, J. McBreen, D. Fischer and Z. Fu, Elec-
Soc., 2002, 149, A1079–A1082. trochem. Solid-State Lett., 2002, 5, A22–A25.
248 S. Zhang, K. Xu and T. Jow, J. Power Sources, 2003, 113, 276 T. Akai, H. Ota, H. Namita, S. Yamaguchi and M. Nomura,
166–172. Phys. Scr., 2005, 2005, 408.
249 X. Yao, S. Xie, C. Chen, Q. Wang, J. Sun, Y. Li and S. Lu, 277 S. Hy, Y.-H. Chen, J.-y. Liu, J. Rick and B.-J. Hwang, J. Power
J. Power Sources, 2005, 144, 170–175. Sources, 2014, 256, 324–328.
250 Z. Zeng, B. Wu, L. Xiao, X. Jiang, Y. Chen, X. Ai, H. Yang 278 J. Lei, L. Li, R. Kostecki, R. Muller and F. McLarnon,
and Y. Cao, J. Power Sources, 2015, 279, 6–12. J. Electrochem. Soc., 2005, 152, A774–A777.
251 S. El Khakani, J. C. Forgie, D. D. MacNeil and D. Rochefort, 279 K. Edström, M. Herstedt and D. P. Abraham, J. Power
J. Electrochem. Soc., 2015, 162, A1432–A1438. Sources, 2006, 153, 380–384.
252 L. Xiao, X. Ai, Y. Cao, Y. Wang and H. Yang, Electrochem. 280 K.-I. Morigaki and A. Ohta, J. Power Sources, 1998, 76,
Commun., 2005, 7, 589–592. 159–166.

This journal is © The Royal Society of Chemistry 2016 Energy Environ. Sci., 2016, 9, 1955--1988 | 1987
 View Article Online

Review Energy & Environmental Science

281 I. A. Profatilova, C. Stock, A. Schmitz, S. Passerini and M. Ue, Springer New York, New York, NY, 2014,
M. Winter, J. Power Sources, 2013, 222, 140–149. pp. 227–282, DOI: 10.1007/978-1-4939-0302-3_5.
282 A. Xiao, L. Yang, B. L. Lucht, S.-H. Kang and D. P. Abraham, 298 J. Cabana, in Electrolytes for Lithium and Lithium-Ion
J. Electrochem. Soc., 2009, 156, A318–A327. Batteries, ed. R. T. Jow, K. Xu, O. Borodin and M. Ue,
283 G. V. Zhuang and P. N. Ross, Electrochem. Solid-State Lett., Springer New York, New York, NY, 2014, pp. 323–369, DOI:
2003, 6, A136–A139. 10.1007/978-1-4939-0302-3_7.
284 A. M. Haregewoin, E. G. Leggesse, J. C. Jiang, F. M. Wang, 299 M. Dolle, S. Grugeon, B. Beaudoin, L. Dupont and
B. J. Hwang and S. D. Lin, J. Power Sources, 2013, 244, J. Tarascon, J. Power Sources, 2001, 97, 104–106.
318–327. 300 M. Inaba, Z. Siroma, Y. Kawatate, A. Funabiki and
Published on 06 May 2016. Downloaded by Freie Universitaet Berlin on 9/28/2018 [Link] AM.

285 A. M. Haregewoin, E. G. Leggesse, J. C. Jiang, F. M. Wang, Z. Ogumi, J. Power Sources, 1997, 68, 221–226.
B. J. Hwang and S. D. Lin, Electrochim. Acta, 2014, 136, 301 A. Andersson, A. Henningson, H. Siegbahn, U. Jansson and
274–285. K. Edström, J. Power Sources, 2003, 119, 522–527.
286 A. M. Haregewoin, T.-D. Shie, S. D. Lin, B.-J. Hwang and 302 S. Hy, F. Felix, J. Rick, W.-N. Su and B. J. Hwang, J. Am.
F.-M. Wang, ECS Trans., 2013, 53, 23–32. Chem. Soc., 2014, 136, 999–1007.
287 M. A. Teshager, S. D. Lin, B. J. Hwang, F. M. Wang, S. Hy 303 P. Niehoff and M. Winter, Langmuir, 2013, 29,
and A. M. Haregewoin, ChemElectroChem, 2016, 3, 337–345. 15813–15821.
288 C. Yang, Y. Wang and C. Wan, J. Power Sources, 1998, 72, 304 S. Amalraj and D. Aurbach, J. Solid State Electrochem., 2011,
66–70. 15, 877–890.
289 P. Verma, P. Maire and P. Novák, Electrochim. Acta, 2010, 305 S. F. Amalraj and D. Aurbach, J. Solid State Electrochem.,
55, 6332–6341. 2011, 15, 877–890.
290 Z. Ogumi, A. Sano, M. Inaba and T. Abe, J. Power Sources, 306 P. B. Balbuena and Y. Wang, Lithium-ion Batteries: Solid-
2001, 97, 156–158. electrolyte Interphase, Imperial College Press, 2004.
291 J. T. Lee, N. Nitta, J. Benson, A. Magasinski, T. F. Fuller and 307 Z. Zeng, W.-I. Liang, H.-G. Liao, H. L. Xin, Y.-H. Chu and
G. Yushin, Carbon, 2013, 52, 388–397. H. Zheng, Nano Lett., 2014, 14, 1745–1750.
292 Z. Zhu, Y. Zhou, P. Yan, R. S. Vemuri, W. Xu, R. Zhao, 308 C. R. Becker, K. E. Strawhecker, Q. P. McAllister and
X. Wang, S. Thevuthasan, D. R. Baer and C.-M. Wang, Nano C. A. Lundgren, ACS Nano, 2013, 7, 9173–9182.
Lett., 2015, 15, 6170–6176. 309 M. McArthur, S. Trussler and J. Dahn, J. Electrochem. Soc.,
293 J. Gnanaraj, R. W. Thompson, S. Iaconatti, J. DiCarlo and 2012, 159, A198–A207.
K. Abraham, Electrochem. Solid-State Lett., 2005, 8, A128–A132. 310 C. Shen, S. Wang, Y. Jin and W.-Q. Han, ACS Appl. Mater.
294 K. Xu, S. Zhang and R. Jow, J. Power Sources, 2005, 143, Interfaces, 2015, 7, 25441–25447.
197–202. 311 L. Wang, D. Deng, L. C. Lev and S. Ng, J. Power Sources,
295 B. Li, M. Xu, B. Li, Y. Liu, L. Yang, W. Li and S. Hu, 2014, 265, 140–148.
Electrochim. Acta, 2013, 105, 1–6. 312 G. M. Veith, M. Doucet, J. K. Baldwin, R. L. Sacci,
296 D. Saikia, T.-H. Wang, C.-J. Chou, J. Fang, L.-D. Tsai and T. M. Fears, Y. Wang and J. F. Browning, J. Phys. Chem.
H.-M. Kao, RSC Adv., 2015, 5, 42922–42930. C, 2015, 119, 20339–20349.
297 M. Xu, L. Xing and W. Li, in Electrolytes for Lithium and 313 C. Wang, A. J. Appleby and F. E. Little, Electrochim. Acta,
Lithium-Ion Batteries, ed. R. T. Jow, K. Xu, O. Borodin and 2001, 46, 1793–1813.

1988 | Energy Environ. Sci., 2016, 9, 1955--1988 This journal is © The Royal Society of Chemistry 2016