Diethyl(thiophen-2-ylmethyl)phosphonate: A novel multifunctional electrolyte additive for high voltage batteries

Article abstract of DOI:10.1039/c8ta01236a

Carbonate-based electrolytes used in Li-ion batteries encounter various challenges in extreme electrochemical environments, and hence their application requires various additives, especially when used with high voltage cathode materials. These additives are designed to form protective interphases that prevent parasitic carbonate oxidation, while in certain cases they stabilize electrolytes from reduction at anode surfaces or even serve as flame-retardants that postpone the thermal runaway during overcharge. However, most of these additives casts negative effects, lowering ionic conductivity of electrolyte or impairing the compatibility between cathode and electrolyte. An ideal solution of minimizing the presence of these inert molecules is to identify an additive that structurally integrates these multiple functions into a single compound. In this work, we report a novel additive, diethyl(thiophen-2-ylmethyl)phosphonate (DTYP). Its 0.5% presence in a base electrolyte dramatically improves the capacity retention of a high voltage Li-ion cell using LiNi_0.5Mn_1.5O₄ from 18% to 85% after 280 cycles at 1C at 60 °C, increases the endothermic reaction onset temperature from 193 °C to 223 °C, and reduces the self-extinguishing time of the electrolyte from 88 s to 77 s. Thus, such a multifunctional additive presents a cost-efficient solution to the issues often faced in high voltage lithium-ion batteries.

Full text of DOI:10.1039/c8ta01236a

View Article Online
View Journal
Journal of
M at erials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Y. Zhu, X. Luo, H.
Zhi, Y. Liao, L. Xing, M. Xu, X. Liu, K. Xu and W. Li, J. Mater. Chem. A, 2018, DOI: 10.1039/C8TA01236A.
Volume
4
Number
1
7
January 2016 Pages 1–330
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Accepted Manuscripts are published online shortly after
Materials for energy and sustainability
www.rsc.org/MaterialsA
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2050-7488
PAPER
Kun Chang, Zhaorong Chang et al.
Bubble-template-assisted synthesis of hollow fullerene-like
MoS nanocages as a lithium ion battery anode material
2
rsc.li/materials-a

Page 1 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
Diethyl (thiophen-2-ylmethyl) phosphonate: A Novel
Multifunctional Electrolyte Additive for High Voltage
Batteries
a
a
a
a,b
a,b
Yunmin Zhu , Xueyi Luo , Huozhen Zhi , Youhao Liao , Lidan Xing ,
a,b
a,b
c
a,b
Mengqing Xu , Xiang Liu , Kang Xu * and Weishan Li *
a. School of Chemistry and Environment, South China Normal University, Guangzhou 510006, China
b. Engineering Research Center of MTEES (Ministry of Education), Research Center of BMET (Guangdong
Province), Engineering Lab. of OFMHEB (Guangdong Province), Key Lab. of ETESPG (GHEI), and
Innovative Platform for ITBMD (Guangzhou Municipality), School of Chemistry and Environment, South
China Normal University, Guangzhou 510006, China
c. Electrochemistry Branch, Sensor and Electron Devices Directorate, Power and Energy Division, U.S. Army
Research Laboratory, Adelphi, MD 20783, USA.
*
Corresponding authors: k.xu.civ@mail.mil; liwsh@scnu.edu.cn
Carbonateꢀbased electrolytes used in Liꢀion batteries encounter various challenges at
extreme electrochemical environments, hence their application calls for the help of
various additives, especially when used with high voltage cathode materials. These
additives are designed to form protective interphases that prevent parasitic carbonate
oxidation, while in certain cases they stabilize electrolytes from reduction at anode
surfaces or even serve as flameꢀretardants that postpone the thermal runaway during
overcharge. However, most of these additives are nonꢀsolvent to Liꢀsalts, and their
excessive use often casts negative effects. An ideal solution of minimizing the
presence of these inert molecules is to identify an additive that structurally integrates
these multiple functions into a single compound. In this work, we report a novel
additive, diethyl (thiophenꢀ2ꢀylmethyl) phosphonate (DTYP). Its 0.5% presence in a
base electrolyte dramatically improves the capacity retention of a high voltage Liꢀion
o
cell using LiNi0.5Mn1.5
O
4
from 18% to 85% after 280 cycles at 1 C at 60 C, raises the
1

Journal of Materials Chemistry A
Page 2 of 40
View Article Online
DOI: 10.1039/C8TA01236A
o
o
endothermic reaction onset temperature from 193 C to 223 C, and reduces the
selfꢀextinguishing time of the electrolyte from 88 s to 77 s. Thus, such a
multifunctional additive presents a costꢀefficient solution to the issues often faced in
high voltage lithiumꢀion batteries.
Introduction
Lithiumꢀion battery (LIB) has been exclusively used as power source for
portable devices due to its high energy/power densities and excellent reversibility as
[
1ꢀ5]
compared to other secondary battery chemistries.
However, its performances,
especially energy density, cycleꢀlife, and cost, need further improvements when
[6ꢀ10]
considered for large scale applications, such as electric vehicles and gridꢀstorage.
Cathode materials with higher working voltage than conventional cathodes (LiCoO
2
,
,
LiMn
2
O
4
4 4 4 4
and LiFePO ), such as LiNiSO F, LiCoPO , LiNiPO and LiNi0.5Mn1.5O
4
[11ꢀ16]
have been developed,
among which LiNi0.5Mn1.5
O
4
is the most attractive, due to
[
17ꢀ19]
its steady discharge voltage, nonꢀtoxicity and low cost.
Unfortunately, its redox
potential of 4.6 V presents a severe challenges to the nonꢀaqueous electrolytes based
on organic carbonate esters.
6
Conventional electrolytes consist of lithium salt (LiPF ) dissolved in a mixed
solvent of cyclic and linear carbonates such as ethylene carbonate (EC), dimethyl
carbonate (DMC) and ethyl methyl carbonate (EMC). These electrolytes exhibit
excellent ionic conductivity and compatibility with battery materials including anode,
[
20]
cathode and membrane, and no substitute for them has been found up to date.
However, these organic carbonateꢀbased electrolytes are also susceptible to oxidative
2

Page 3 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
decomposition at potentials above 4.5 V, besides being thermal unstable and
flammable.
The interfacial instability between cathode and electrolytes results from the
electrochemical oxidation of the electrolyte at the high voltage cathode surface, which
is a complicated process that involves the complexes between solvent carbonate
ꢀ
6
molecules, salt anion PF
, transition metal ions from the cathode lattice, generation of
gaseous species such as oxygen, carbon dioxide, as well as acids such as HF and
[21,22]
polymers such as oligo ethylene oxides.
These products significantly raise the
cell impedance, the loss of transition metal ions or even the corrosion of aluminum
current collector, leading to early cell failure. Although the detailed underneath
mechanism for these processes remain to be understood, all of them accelerate at
o
[23ꢀ28]
elevated temperatures above 50 C.
with a layer of inert compounds such as Bi
graphemeꢀoxide could suppress the interfacial reactivity between LiNi Mn O and
Coating LiNi0.5Mn1.5
O
4
2 3
O and
0.5
1.5
4
[29ꢀ39]
electrolytes,
but coating process always introduces higher cost for cell
manufacturing. The coated layer also increases the cell interfacial impedance, which
[
40,41]
reduces the specific discharge capacity of the cathode and the rate capability.
Comparatively, forming a cathode interphase on LiNi0.5Mn1.5 by applying an
Several electrolyte
additives have been developed for this purpose, which include tris
O
4
[42,43]
electrolyte additive is a much more costꢀefficient approach.
[
44]
(hexafluoroꢀisopropyl) phosphate,
tris (trimethylsilyl) borate, tris (trimethylsilyl)
[
37,45ꢀ47]
phosphite, 4ꢀ(trifluoromethyl)ꢀbenzonitrile, and phenyl trifluoromethyl sulfide.
3

Journal of Materials Chemistry A
Page 4 of 40
View Article Online
DOI: 10.1039/C8TA01236A
The preferential oxidation of these additives are proposed to be responsible for the
cathode/electrolyte interfacial stability.
The thermal instability of organic carbonateꢀbased electrolytes results from the
dissociation of LiPF to LiF and PF under elevated temperature. The Lewis acidic
6
5
PF
electrolyte, such as ROH or H
The resulting OPF will initiate an auto catalytic decomposition of the electrolyte,
forming CO , alkyl ethers (R O), alkylfluorides (OPF ), and fluorophosphates
OPF OR and OPF(OR) ). These decomposition products will deteriorate the
5
directly attack carbonates, or reacts rapidly with trace protic impurities in the
2
O, forming phosphorus oxyfluoride (OPF ) and HF.
3
3
2
2
3
[48]
(
2
2
battery performance. To improve the thermal stability of the electrolyte, new salts
such as lithium bis (oxalato) borate (LiBOB) and lithium difluoro (oxalate) borate
[
49]
(
LiDFOB) have been developed to replace LiPF
6
.
However, these new salts cannot
provide sufficient ionic conductivity as LiPF does, nor are they stable against high
6
oxidation potential above 4.2 V. Lewis bases such as dimethyl acetamide are found to
be effective as electrolyte additives to improve the thermal stability of organic
carbonateꢀbased electrolyte, which is attributed to the coordination of additives with
ꢀ
[50]
PF
6
5
or PF .
The flammability of the organic carbonateꢀbased electrolytes results from
carbonate solvents, especially the linear carbonate molecules of low flash points,
which has been the main cause for safety concerns for LIB. Much effort has been
dedicated to this issue. Completely replacing carbonate solvents with ionic liquids (IL)
appears to be tempting due to their nonꢀflammability and excellent ionic
4

Page 5 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
[
51ꢀ58]
conductivity,
but there have been reports about the thermal runaway of LIBs
using IL, suggesting there is fundamental difference between nonꢀflammability and
safety in LIB environment, which is more related to electrochemical reactions
[59]
between atomic oxygen and organic species at high temperatures.
The high
viscosities of IL, especially after dissolving Liꢀsalt, further lower the expectation
because of compromises in ionic conductivity. Most importantly, ionic liquids tend to
[60]
insert graphite anode during charge process, leading to the poor cyclic stability. Gel
polymer electrolytes or even solid electrolytes provide promising solutions to the
[61ꢀ63]
flammability issue,
but their ionic conductivity is not sufficient to support battery
operations. On comparative basis, using electrolyte additives to alter the flammability
[
64]
of the organic electrolyte would be more feasible from practical viewpoint.
Because of these many multiple tasks expected from additives, a common
practice in formulating electrolytes is to employ multiple additives, with each
targeting one specific function. Unfortunately, excessive use of these additives
inevitably yields negative effects, including lowering ionic conductivity, altering
wettability toward electrode and membrane, and possible counterꢀeffecting each other
[
65ꢀ67]
in stabilizing electrode/electrolyte interfaces.
Additionally, the effectiveness of
any stateꢀofꢀart additive is by no means satisfactory. For example, A recently reported
electrolyte additive for interfacial stability improvement, 1ꢀpropylphosphonic acid
cyclic anhydride, yielded a capacity retention of only 80% for LiNi0.5Mn1.5
O
4
after
o
[66]
only 50 cycles when tested at a rate of 0.2 C and 55 C between 3 and 4.9 V. On the
6
other hand, the selfꢀextinguishing time of the electrolyte consisting of 1M LiPF in
5

Journal of Materials Chemistry A
Page 6 of 40
View Article Online
DOI: 10.1039/C8TA01236A
EC/DMC/EMC (1:1:1 in mass) was reduced from 60 s to 24 s, only after using a
copious amount (40%) of trimethyl phosphate (TMP), a commonꢀused flame
[
64]
retardant, which causes compromises in both ionꢀconduction and interphase quality
on both anode and cathode. It has remained a challenge to minimize the additive
presence while achieving these expected performances.
The molecular structure of electrolyte additives is vital for their effectiveness. As
an additive for interphase formation, it should be oxidized preferentially before bulk
electrolyte components on the one hand; elements in the additives should be
beneficial to the physical and chemical properties of the resultant interphase on the
other. As one of such additives, thiopheneꢀcontaining compounds were found to form
[
68ꢀ74]
electronically conductive interphase films,
beneficial to the ionic conductivity of the interphase.
consideration, a group that can stabilize the strong Lewis acid PF
and sulfurꢀcontaining compounds are
[44]
For the thermal stability
should be
while for flame retarding, phosphorus or
5
[50]
contained in the electrolyte additives,
fluorine species are preferred as subꢀstructures.
[
67,75ꢀ77]
With these knowledge, we
designed a novel electrolyte additive that integrate these multifunctional into a single
molecule, diethyl (thiophenꢀ2ꢀylmethyl) phosphonate (DTYP), for use in high voltage
lithiumꢀion battery. As shown in Scheme 1, DTYP is composed of thiophene and
phosphate, where the thiophene is able to form ionically conductive cathode
interphases through the polymerization of the radical resulting from its preferential
5
oxidation; the oxygen of phosphate have ability to neutralize PF through acidꢀbase
coordination, and phosphate is effective to terminate the radical propagation in the
6

Page 7 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
combustion process. Applying a low content of DTYP in organic carbonateꢀbased
electrolyte, the interfacial stability between LiNi0.5Mn1.5 and electrolyte, the
O
4
thermal stability and the flame retarding of electrolyte could be simultaneously
improved. These features were demonstrated by theoretical calculations,
electrochemical measurements and physical characterizations.
Calculation and experimental section
Synthesis of DTYP
To ensure its purity, DTYP was synthesized in lab. Thiophene (>99%) was
purchased from Aladin and triethyl phosphate (>99%) was purchased from Alfa. A
certain amount of concentrated hydrochloric acid, thiophene and methyl aldehyde
ꢀ1
ꢀ1
were mixed under an atmosphere of 10 ml h ~5 L h gaseous HCl at temperature
o
lower than 0 C. After reaction, 2ꢀ(chloromethyl) thiophene was extracted by using
methylene chloride, dried by using anhydrous sodium sulfate and concentrated under
o
4
0 C in vacuum. The resulting 2ꢀ(chloromethyl) thiophene was mixed with phosphite
ethyl ester at 120 °C for 7 h under stirring and refluxing to obtain raw product DTYP,
which was then purified by distillation to obtain pure DTYP. The reaction path is
presented in Scheme S1.
The resulting DTYP was characterized by NMR spectroscopy as shown in Figure
S1. The singlet at 1.259 ppm is characteristic of ꢀCH (a1 and a2 as the inset in Figure
3
S1), those at 3.374 and 3.322 ppm for the group ꢀCH
2
ꢀ (b1 and b2), that at 4.048 ppm
for ꢀCH ꢀ (c), those at 6.943 and 6.964 ppm for ꢀCHꢀ (d1 and d2), and those at 7.253
2
7

Journal of Materials Chemistry A
Page 8 of 40
View Article Online
DOI: 10.1039/C8TA01236A
and 7.153 ppm for ꢀCHꢀ (e), indicating that DTYP is highly pure. Based these peak
area, as shown in Table S1, the purity of DTYP is 99%, which is purer than the
reagent DTYP (98%, Adamas Reagent Co., Ltd.).
Calculations
All calculations were performed using the Gaussian 09 package. The equilibrium
structures were optimized with the B3LYP in conjunction with the 6ꢀ11++G (d) level
basis set. At the same level, frequency calculations were also performed to confirm all
optimized molecules with a consistent stationary. The calculated oxidation potential
was converted from the Freeꢀenergy cycle for the oxidation reaction. Atomic charge
distributions were analyzed by natural population analysis (NPA) theory.
Electrode preparation and cell assembly
Batteryꢀgrade carbonate solvents and lithium hexafluorophosphate (LiPF
6
) were
in
provided by Guangzhou Tinci Materials Technology Co. Ltd, China. 1.0 M LiPF
6
ethylene carbonate (EC)/ethyl methyl carbonate (EMC), (EC/EMC=3:7, in volume)
used as base electrolyte and 0.25, 0.5, 1, 2 and 4 wt. % DTYP as additive were used.
The electrolytes were prepared in an argonꢀfilled glove box (MBraun, Germany). The
HF and water content in these electrolytes were less than 20 ppm. The content of HF
in the electrolyte was determined by titration. 0.01 M NaOH in deionized water and
0
.001 wt. % bromothymol blue as indicator was used. LiNi Mn O electrode was
0.5 1.5 4
4
prepared with a mixture of 80 wt. % LiNi0.5Mn1.5O (Sichuan Xinneng Advanced
Material Co., Ltd. China), 10 wt. % acetylene carbon black (TIMCAL Ltd,
8

Page 9 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
Switzerland), 10 wt. % polyvinylidene fluoride (PVDF, Ofluorine Chemical Tech Co.,
Ltd. China) in Nꢀmethylꢀpyrrolidone (NMP). The mixture slurry was coated on Al foil
o
o
and dried at 80 C for 1 h and then at 120 C for 12 h under in vacuum. Graphite
electrode was composed of graphite (Dongguan Kaijin New Energy Technology Co.,
Ltd, China) (89 wt. %), Ks6 (4 wt. %), Superꢀp (2 wt. %) and poly (vinylidene
o
fluoride) (5 wt. %). The slurry was coated on a Cu foil and then dried at 80 C for 12h.
−
2
For half cells, the mass loading was about 2.5~2.8 mg cm on the LiNi0.5Mn1.5O
4
−2
electrode and 2.8~3.2 mg cm on graphite electrode, both of which are 12 mm in
diameter, while metallic lithium foil of 16 mm in diameter was used as the counter.
ꢀ2
For full cells, the mass loading was 25.2~25.4 mg cm for LiNi0.5Mn1.5
O
4
cathode (12
ꢀ2
mm in diameter) and 10.8~10.9 mg cm for graphite anode (12.5 mm in diameter),
ꢀ1
with anode capacity in excess of 25%, based on 120 mAh g for LiNi0.5Mn1.5
O
4
and
ꢀ1
3
50 mAh g for graphite. All the cells were assembled with Celgard 2400
microporous separator and various electrolytes in the argonꢀfilled glove box, in which
oxygen and water content were controlled to less than 0.1 ppm. The pouch cells
(capacity 1500 mAh) were provided by Dongguan Kaixin Materials Technology Co.
Ltd, China.
Electrochemical measurements
Cyclic performances of LiNi Mn O electrode were determined by cycling
0
.5
1.5
4
o
o
coin cells at room temperature (25 C) and elevated temperature (60 C), charged at
+
constantꢀcurrent to 4.9 V, followed by a constant voltage charge step (4.9 V, vs. Li/Li )
until the current decreased to 10% of the applied current, then discharged to 3 V with
9

Journal of Materials Chemistry A
Page 10 of 40
View Article Online
DOI: 10.1039/C8TA01236A
the same constant current on LAND system (CT2001A, Wuhan, China). Rate
performances of the cells were determined on LAND system (CT2001A, Wuhan,
China), with the currents of 0.5 C,1 C, 3 C, 5 C, 7 C, 10 C and 15 C. The cyclic
voltammetry (CV) was performed on Solartronꢀ1480 instrument (England) in
ꢀ1
Li/LiNi0.5Mn1.5
O
4
coin cells with a scan rate of 0.1 mV s . Electrochemical
impedance spectroscopy (EIS) was carried out at fully discharged state, the
5
ꢀ2
perturbation was ±5 mV, and the frequency range was from 10 ꢀ10 Hz on a
PGSTATꢀ30 electrochemical station (Autolab Metrohm, Switzerland).
Physical characterizations
For physical characterizations, the cycled cells were disassembled in the
argonꢀfilled glove box. Both the lithium foil and LiNi0.5Mn1.5
out, rinsed with anhydrous DMC for three times to remove the residual EC, EMC and
LiPF salt, and then dried overnight under vacuum at room temperature in glove box
4
O cathodes were taken
6
before exꢀsitu analysis. Surface morphology was observed by transmission electron
microscopy (TEM) performed on JEMꢀ2100HR (Japan) and scanning electron
microscopy (SEM) conducted on JSMꢀ6510 (Japan). The contents of Ni, Mn and Al
deposited on Li foils were analyzed by inductively coupled plasma atomic emission
spectrometer (ICPꢀAES, Optima 8300, USA). The organic functional groups in the
surface components were characterized by Fourier transform infrared spectroscopy
ꢀ1
(
FTIR, Bruker Tensor 27, Germany) within 500ꢀ2500 cm . Surface compositions on
the cycled cathode were determined by Xꢀray photoelectron spectroscopy (XPS,
Thermo Fisher Scientific, UK), using a focused monochromatized Al Kα Xꢀray source
1
0

Page 11 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
(excitation energy = 1468.6 eV) under ultra high vacuum with the spot size of 400 µm
and energy step size of 1.000 eV. The final adjustment of the energy scale was made
based on the C 1s peak of graphite at 284.8 eV. The XPS analyses were performed
without sputtering. Ionic conductivities of the electrolytes were measured with 856
conductivity module (Metrohm, Switzerland). The conductivity meter was calibrated
by 0.01 M KCl standard aqueous solution before measurements. The
selfꢀextinguishing time (SET) of the electrolyte with different amounts of additives
was measured in a cap with 0.5 mL electrolyte, which was ignited in a fume hood.
The burning time of the electrolytes was measured by a stopwatch. Each test was
repeated six times and the average value was reported. The puncture experiments of
pouch cells were performed by Shenzhen BAK Battery Co., Ltd. The decomposition
temperature of the different electrolytes was measured by using differential scanning
o
ꢀ1
calorimetry (DSC, Netzsch DSC 200 PC, Germany) at a heating rate of 10 C min in
o
o
the temperature range from 25 C to 300 C. The flameꢀretardant tests were performed
by igniting the membranes soaked with 0.8 mL base or 1% DTYPꢀcontaining
electrolyte.
Results and discussion
Interfacial stability
The effect of DTYP on stabilization of LiNi Mn O /electrolyte coupling is
0.5
1.5
4
6
shown in Figure 1 A, which was compared with the base electrolyte, 1.0 M LiPF in
11

Journal of Materials Chemistry A
Page 12 of 40
View Article Online
DOI: 10.1039/C8TA01236A
ethylene carbonate (EC)/ethyl methyl carbonate (EMC), (EC/EMC=3:7, in volume).
o
At 60 C and 1 C rate, the base electrolyte only retains 18% of the original capacity
after 280 cycles, indicating severe electrolyte breakdown, while only 0.5% DTYP
improved the capacity retention to 84%. Correspondingly, the coulombic efficiency
vibrated in the base electrolyte but remains stable in the additiveꢀcontaining
electrolytes. The improved cyclic stability of LiNi0.5Mn1.5
4
O by DTYP is far better
than those reported in literature. Examples include 0.5% thiophene (TH), with a
o
[68]
capacity retention of only 79% at 55 C after 100 cycles at 1 C,
or 0.5%
4ꢀ(Trifluoromethyl)ꢀbenzonitrile with only 91% at room temperature after 300 cycles
[46]
o
at 1 C, or 0.5% phenyl trifluoromethyl sulfide with 95% at 55 C after 100 cycles at
[47]
1
4
C. Apparently, DTYP is effective for stabilizing LiNi0.5Mn1.5O /electrolyte, and
the improvement relates to the interphase film formed from the preferential oxidation
of DTYP.
Figure 1B presents the highest occupied molecular orbital (HOMO)/lowest
unoccupied molecular orbital (LUMO) energies of DTYP, with a comparison with
solvents, EC and EMC, and TH. Apparently the HOMO energy of DTYP (ꢀ0.2406 au.)
is less negative than EC (ꢀ0.3209 au.), EMC (ꢀ0.3089 au.), and TH (ꢀ0.2475 au.),
suggesting that DTYP experiences preferential oxidation before the base electrolyte. It
appears that the exclusive decomposition products from DTYP oxidation leads to the
formation of a stable interphase, as DTYP has a lowest oxidation potential (4.88 V)
than EC (7.02 V), EMC (6.85 V) and TH (5.11 V), as shown in Figure 1C.
The oxidation of DTYP can be more clearly visualized via differential capacity
1
2

Page 13 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
curves of LiNi0.5Mn1.5
the initial charge process (Figure 1D). LiNi0.5Mn1.5
electrolytes. Two typical oxidation peaks occur at the potential range of 4.6~4.8 V,
O
4
electrodes in base and DTYPꢀcontaining electrolytes during
O
4
behaves similarly in these two
2+
corresponding to the lithium extraction processes related to the oxidations of Ni to
3
+
3+
4+
[39,78ꢀ80]
Ni and Ni to Ni , respectively.
An additional oxidation peak at about 4.2 V
appears in DTYPꢀcontaining electrolyte, as indicated by the inset of Figure 1D, where
the base electrolyte should be stable, indicating where DTYP is oxidized, before the
lithium extraction and the base electrolyte decomposition. The oxidation current of
DTYP can be recorded when the electrode is performed with cyclic voltammetry, as
shown in Figure 1E. Beside the redox pair between 4.5 and 4.9 V corresponding to the
[
81]
oxidation/reduction of nickel in LiNi0.5Mn1.5
O
4
,
an additional oxidation peak can
also be indentified at about 4.2 V for DTYP. This preferential oxidation leads to the
formation of an interphase originated from DTYP. It should be noted that in both
Figure 1E, at 4.77~5.0 V, a significantly higher background current was observed
following lithium extraction in the base electrolyte, which should be attributed to the
severe electrolyte decomposition. When DTYP is present, however, the background
current decreased dramatically, suggesting that the cathode interphase formed from
the preferential oxidation of DTYP results in a much better protection.
The improved interfacial stability between LiNi Mn O and electrolyte is also
0.5
1.5
4
o
reflected by the change of interfacial impedance after cycling under 60 C. Figure 1F
and G compares the impedance spectra of the electrodes in base and
DTYPꢀcontaining electrolytes before and after cycling test, in which characteristic
1
3

Journal of Materials Chemistry A
Page 14 of 40
View Article Online
DOI: 10.1039/C8TA01236A
depressed semicircle at high frequencies and a slope line at low frequencies were
+
observed, with the former representing the impedance of Li ꢀtransport through
f
interphase and the charge transfer on the interface, designated hereafter by R and Rct,
[
82, 83]
and the latter corresponding to the Warburg impedance.
results with the equivalent circuit (the inset in Figure 1F and G), it can be found that
there is a significant change in the interfacial resistance (defined as sum of R and Rct),
Based on the fitting
f
from 187 ꢁ to 3350 ꢁ for the base electrolyte, and from 292 ꢁ to only 378 ꢁ for
DTYPꢀcontaining electrolyte. Apparently, the interfacial stability between
LiNi Mn O and electrolyte has been improved by applying DTYP. It should be
0
.5
1.5
4
noted that the initial interfacial resistance of the electrode (292 ꢁ) in the
DTYPꢀcontaining electrolyte is slightly larger than that in the base electrolyte (187 ꢁ).
This difference can be attributed to the initially aggravated polarization of
4
LiNi0.5Mn1.5O electrode, due to deteriorated wettability of the electrolyte with
electrode material, the decreased ionic conductivity of the electrolyte and the resulting
resistance of the cathode interphase from the application of DTYP. This aggravated
polarization is responsible for the decreased oxidation peak current of LiNi0.5Mn1.5
electrode in the DTYPꢀcontaining electrolyte (Figure 1E).
O
4
4
LiNi0.5Mn1.5O in DTYPꢀcontaining electrolyte also displays better rate
capability than in base electrolyte, especially at high current rate, as shown in Figure
S2. At a high rate of 15 C (equivalent of fully charging a cell in 4 minutes), the
ꢀ1
discharge capacity of LiNi0.5Mn1.5
O
4
is 93 mAh g in DTYPꢀcontaining electrolyte,
ꢀ1
but only 57 mAh g can be achieved in base electrolyte. It should be noted that the
1
4

Page 15 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
discharge capacity of LiNi0.5Mn1.5
current is turned from 15 C back to 0.2 C, suggesting that the interface of
/electrolyte maintains stability during rate capability test. In fact, the
cyclic stability of LiNi Mn O in the base electrolyte under room temperature
4
O can be almost entirely recovered when the rate
LiNi0.5Mn1.5O
4
0.5
1.5
4
o
(
Figure S3) is much better than that under 60 C (Figure 1A). Therefore, the enhanced
rate capability of LiNi0.5Mn1.5 is related to the improved ionic conductivity of the
O
4
cathode interphase, which can be ascribed to the sulfur existing in DTYP. The
sulfurꢀcontaining oxidation decomposition compounds of DTYP are incorporated into
the cathode interphase, which are beneficial to the ionic conductivity of the
[14]
interphases.
Transition metal dissolution, especially at elevated temperatures, has always
been a failure mechanism responsible for high voltage Liꢀion cell failure. The metal
will deposit on anode and deteriorate the performance of graphite anode. Such
parasitic process seems to be suppressed by DTYP presence, as confirmed by
comparing the charge/discharge behaviors of LiNi0.5Mn1.5
base and DTYPꢀcontaining electrolytes. Figure S4A presents the cyclic stability of
/graphite full cells, with an indication of the energy density of the cells
. The cells cycled at 0.1 C for the initial three cycles and then
4
O /graphite full cells using
LiNi0.5Mn1.5
O
4
based on LiNi0.5Mn1.5
O
4
ꢀ1
0
.2 C for the subsequent cycles. Both cells deliver 520 Wh kg initially, however, the
ꢀ1
full cell using the base electrolyte suffers a fast capacity decay from 115 mAh g to
ꢀ1
7
7 mAh g at the 150th cycle, with a discharge capacity loss of 33%. This decay
involves the LiNi0.5Mn1.5 destructions and the negative effect of deposited
O
4
1
5

Journal of Materials Chemistry A
Page 16 of 40
View Article Online
DOI: 10.1039/C8TA01236A
transition metal on graphite anode. In sharp contrast, the cell using DTYP delivers its
ꢀ1
ꢀ1
initial capacity of 117 mAh g and retains 90 mAh g at the 150th cycle, with a
capacity retention of 77%. After 150 cycles, the energy density is reduced to 323 Wh
ꢀ1
ꢀ1
kg for the cell in the base electrolyte but remains 398 Wh kg with DTYP. Figure
S4BꢀD presents the charge/discharge curves and corresponding mean
4
charge/discharge voltages of LiNi0.5Mn1.5O /graphite full cells at the selected cycles
of Figure S4A, indicating that in the case of base electrolyte, the significant drop in
energy density comes from simultaneous capacity decay and cell polarization.
As indicated in Figure S3A, increasing the concentration of DTYP leads to an
ꢀ1
initial capacity loss. The initial capacity is 128 mAh g with a coulombic efficiency
ꢀ1
of 86% for the cell using the base electrolyte, but it is reduced to 120 mAh g , 119
ꢀ1
ꢀ1
ꢀ1
mAh g , 117 mAh g , and 109 mAh g with corresponding coulombic efficiency of
85%, 84%, 79%, and 75% for 0.25%, 0.5%, 1%, and 2% of DTYPꢀconcentrations,
respectively. This phenomenon is universal for the application of electrolyte additives
in carbonateꢀbased electrolytes whose ionic conductivity, compatibility with electrode
[84,85]
materials and interfacial resistance will be affected by nonꢀcarbonate components.
Fortunately, when 0.5% DTYP is used, the initial capacity loss is negligible and an
excellent capacity retention of 94% after 300 cycles is achieved. Therefore, 0.5%
DTYP was considered for further investigations.
The effect of DTYP on graphite anode was evaluated with Li/graphite halfꢀcell
as shown in Figure S5, which reveals that both electrolytes exhibit the same cyclic
stability, suggesting that DTYP does not affect graphite anode. The observed
1
6

Page 17 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
improvements in the cyclic stability of LiNi0.5Mn1.5
attributed to a cathode electrolyte interphase (CEI) originated from DTYP on
cathode.
4
O /graphite full cell should be
LiNi0.5Mn1.5O
4
The existence of such a CEI can be confirmed by physical characterizations from
SEM, TEM, ICP, FTIR, and XPS. Figure 2 presents the SEM and TEM images of
o
LiNi0.5Mn1.5
LiNi0.5Mn1.5
O
O
4
4
electrodes before and after cycling tests under 60 C, in which the
particles with a smooth and a clean surface surrounded by the
conductive carbon can be identified in the pristine electrode (Figure 2A and D). After
cycling in base electrolyte (Figure 2B and E), a thick deposit is observed (red arrows
in Figure 2B and E), which serves the instability of base electrolytes on
o
LiNi0.5Mn1.5
O
4
surface at 60 C. On the contrary, the LiNi0.5Mn1.5
O
4
particles cycled
in DTYPꢀcontaining electrolyte maintain the smooth surface of the fresh particles
Figure 2C), and a thin and uniform interphase of about 20 nm can be visually
identified (Figure 2F). This image confirms that DTYP can indeed form a protective
, which is the source for the improved cycling stability and low
(
CEI on LiNi0.5Mn1.5O
4
impedance observed. Figure S6 presents the SEM images from different position
angles under different resolutions from Figure 2A, confirming the different particle
morphology of LiNi0.5Mn1.5
that in base electrolyte.
4
O cathode cycled in DTYPꢀcontaining electrolyte from
The compositions of the CEI film on LiNi0.5Mn1.5
FTIR and XPS. Figure 3A presents the FTIR spectra of LiNi0.5Mn1.5
before and after cycling in base and DTYPꢀcontaining electrolytes. All electrodes
O
4
were further analyzed with
O
4
electrodes
1
7

Journal of Materials Chemistry A
Page 18 of 40
View Article Online
DOI: 10.1039/C8TA01236A
ꢀ1 [86ꢀ88]
show the typical absorptions of PVDF (1400, 1170, 1070 and 877 cm ).
Differing from the base electrolyte, the electrode cycled in DTYPꢀcontaining
ꢀ1
electrolyte shows addition peaks, including those at ~1683, ~1026, and 768 cm ,
which are characteristics of C=C, S=O and CꢀS species. These species should be
related to DTYPꢀoriginated products, indicating that the CEI bears clear chemical
signature of DTYP. In addition, the electrode cycled in base electrolyte shows much
ꢀ1
stronger peaks at ~1590, ~1490, ~1744 and 843 cm , which correspond to the
[
49,89]
electrolyte decomposition products including Li
2 3 x y
CO , polycarbonate and Li PF .
Apparently, CEI formed by DTYP suppresses the electrolyte decomposition.
Figure 3BꢀI presents the XPS spectra of the electrode before and after cycling
test. The elements from the pristine electrode, including nickel and manganese in
LiNi0.5Mn1.5
O
4
as Ni 2p and Mn 2p, oxygen in MꢀO at 529 eV of O 1s, carbon in
conductive carbon at 284.3 eV, carbon in PVDF at 290.4 eV and 285.7 eV of C 1s,
[90ꢀ92]
and fluorine in PVDF at 687.6 eV of F 1s,
can be identified in both electrodes
but with much lower abundances after being cycled in base electrolyte than in
DTYPꢀcontaining electrolyte. On the other hand, the electrolyte decomposition
products, including those at 286.5 and 290 eV of C 1s for lithium alkyl carbonates,
2 3
533.5 eV of O 1s for lithium alkyl carbonates and 531.6 eV of O 1s for Li CO , ~687
eV of F 1s for Li PO P , 56 eV of Li 1s for LiF, ~137 eV and ~134 eV of P 2p for
x
y
z
[
87,91,93,94]
Li
x
PF
z
and Li
x
PO
F
y z
,
are much stronger in the electrode cycled in base
electrolyte. This comparison confirms that there is more bulk electrolyte
decomposition on the electrode cycled in base electrolyte than that in
1
8

Page 19 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
DTYPꢀcontaining electrolyte. The peaks at ~169 eV and ~164 eV in S 2p spectrum,
corresponding to Li SO and C–S respectively, can only be identified on the electrode
2
3
cycled in DTYPꢀcontaining electrolytes but do not exist in the electrode cycled in
base electrolyte, confirming that the CEI is built from the oxidation decomposition
products of DTYP.
Metal elements on lithium anode taken from the cycled cells are also analyzed to
assess the effect of additive on transition metal dissolution. The cycled lithium anodes
3
were washed with DMC for three times, and dissolved in 25 ml 4% HNO solution
before being subject to ICP analyses, with the obtained results presented in Table S2.
There are high contents of Ni, Mn and Al deposited on the cycled lithium anode in the
4
base electrolyte, suggesting that the transition metal ions in LiNi0.5Mn1.5O and even
the aluminum in current collector has suffered from dissolution. These elements are
significantly reduced on lithium recovered from DTYPꢀcontaining cells, suggesting
that the CEI formed from DTYP not only prevents bulk electrolyte decomposition but
also suppresses the cathode active materials from dissolution, leading to the excellent
4
cyclic stability of LiNi0.5Mn1.5O under elevated temperature.
Based on the morphological observations from SEM and TEM and the
compositional analyses from FTIR and XPS, it can be concluded that the essential
differences between the deposit from electrolyte decomposition and the asꢀformed
CEI from DTYP are their compositions and the thicknesses. The deposit is thick,
mainly composed of polycarbonate and LiF, while the CEI is thin, mainly composed
of the DTYP polymer. The CEI formation mechanism involves the preferential
1
9

Journal of Materials Chemistry A
Page 20 of 40
View Article Online
DOI: 10.1039/C8TA01236A
oxidation of DTYP, which yields thiophene radical that is polymerized. The resulting
polymer enables such CEI with superior protection against parasitic reactions between
electrolytes and active cathode materials.
Thermal stability
Carbonateꢀbased electrolytes are known to be thermally labile, because the
6 5
lithium salt LiPF is easy to decompose under elevated temperature forming PF
followed by a series chain reactions that involve both the strong Lewis acid, carbonate
[48]
molecules and the trace moistures.
battery performance.
The resulting products are detrimental to the
o
The electrolytes were stored at 80 C for 150 hours and their color change is
closely monitored (Figure 4AꢀD). The base electrolyte becomes brown (Figure 4B),
indicative of the thermal decomposition of the electrolyte by copious HF presence.
Comparatively, the color is hardly affected with DTYP presence (Figure 4C and D).
5
This comparison suggests that the acidity of PF is effectively neutralized by
nucleiphilic nature of DTYP enabled by O or S species.
The thermal decomposition of base electrolyte brought the decreased ionic
conductivity and the increased HF concentration in the electrolyte. The ionic
ꢀ1
ꢀ1
conductivity of the base electrolyte decreased from 10.4 mS cm to 7.7 mS cm ,
ꢀ1
ꢀ1
while 0.5% DTYP maintains the conductivity at 10.1 mS cm to 8.3 mS cm . The HF
concentration in the base electrolyte, determined by NaOH titration, is increased by
0.31 wt. %, compared to the 0.06 wt. % for 0.5% DTYPꢀcontaining electrolyte. The
2
0

Page 21 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
thermal degradation of base electrolyte results in detrimental effect on the cycling
stability of LiNi0.5Mn1.5 (Figure 4E), when Li/LiNi0.5Mn1.5 halfꢀcells were
O
4
O
4
assembled using the electrolytes after thermal storage. Significant initial capacity loss
and a fast cyclic capacity decay are observed in the base electrolyte, delivering an
ꢀ1
initial capacity of 108 mAh g , yielding a capacity retention of 8% after 250 cycles
ꢀ1
and exhibiting a vibrating coulombic efficiency, compared to the 118 mAh g , the 80%
and the stable coulombic efficiency, respectively, for the electrode using 0.5%
DTYPꢀcontaining electrolyte.
To understand the mechanism of the improved thermal stability because of
DTYP, theoretical calculations and spectral analyses were performed. Figure 4F
presents the calculated relative energy (ꢂE) of the complexes of PF
solution based on the reaction PF + S = PF ꢀS, where S represents solvents, DTYP or
. P was considered because it might be generated from the phosphate in
DTYP. The relative energy of PF ꢀEMC, PF ꢀEC, PF ꢀDTYP and PF ꢀP O is ꢀ0.372,
5
in acetone
5
5
P
2
O
5
2 5
O
5
5
5
5
2
5
ꢀ1
ꢀ
0.428 (close to 0 as shown in Figure 4F), ꢀ7.55 and ꢀ38.13 kJ mol , respectively,
suggesting the PF ꢀDTYP and PF ꢀP are much more stable than PF ꢀEMC and
5
5
2
O
5
5
[
50]
PF
5
ꢀEC.
The generation of P
2
5
O from the phosphate in DTYP and its contribution to the
thermal stability can be confirmed by differential thermal analysis (DTA). Figure 4G
presents the DTA curves of the electrolytes with and without DTYP, with a
comparison of the electrolyte containing triethyl phosphate that has the same
phosphate group as DTYP. It can be found from Figure 4G that there is an
2
1

Journal of Materials Chemistry A
Page 22 of 40
View Article Online
DOI: 10.1039/C8TA01236A
o
endothermic peak at about 166 C, corresponding to the formation of P
2
O
5
from
phosphate, only for the electrolytes containing DTYP. As shown in Figure 4G, the
o
onset of the thermal decomposition of the base electrolyte appears at 193 C, which is
o
o
enhanced to 223 C and 252 C by applying 0.5% and 1% DTYP, respectively. This
postponement of thermal onset temperature indicates that DTYP can improve the
thermal stability of the base electrolyte. The effect of phosphate can be ascribed to the
coordination of the P
decomposition, with PF
DTYP itself also exhibits coordination with PF . In fact, DTYP behaves better than a
2
O
5
, which is formed from phosphate during thermal
5
as indicated by the calculation above. Beside this effect,
5
[67]
recently reported additive, tris (pentafluorophenyl) phosphine (TPFPP).
For
example, the thermal decomposition peak temperature of a base electrolyte is
o
o
increased by only 10 C by applying 2% TPFPP, but by 30 C with using only 0.5%
DTYP.
The electrolyte thermal decomposition products were analyzed with FTIR.
Figure 5 presents the FTIR spectra of electrolytes, with and without DTYP, before and
o
after storage at 80 C for 150 hours. The characteristic absorption peaks of DTYP are
ꢀ1
ꢀ1
located at 1643, 1533 cm for C=C bond, 1445 cm for ꢀCH
2
ꢀ bond, 1393 cmꢀ1 for
ꢀ1
ꢀ1
ꢀCH
3
bond, 1246, 1040 cm for P=O bond, 1165, 706 and 613 cm for CꢀS bond, 961,
ꢀ1
ꢀ1
ꢀ1
8
51 cm for PꢀOꢀC bond, 795 cm for PꢀC bond, and 528 cm for CꢀSꢀC bond
(Figure 5A). After storage at high temperature, the characteristic absorption peaks of
DTYP become weaker and there appears a blue shift for the peaks of P=O bond,
suggesting that DTYP has partially been decomposed. Lots of new stronger peaks
2
2

Page 23 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
appear in the base electrolyte after storage, including those at ~1805, ~1744, 1490,
ꢀ1
1
400, 1639, ~1280, 1015 and 843 cm (Figure 5B), indicative of the serious thermal
[49]
instability of base electrolyte under the catalysis of OPF
3
.
The peaks at ~1805,
ꢀ1
ꢀ1
~
1744 cm are characteristic of polycarbonate, peaks at 1490, 1400 cm are
[
89]
characteristic of Li
2
CO
species. When DTYP is present, however, the absorption peaks
are hardly changed before and after storage. There are a slight red shift from 1246
3
,
2 x y z
and the others are characteristic of ROCO Li, Li PO F ,
x y
alkoxides, and Li PF
ꢀ1
ꢀ1
ꢀ1
ꢀ1
cm to 1275 cm and a slight blue shift from 1040 cm to 1028 cm of P=O bond in
the DTYPꢀcontaining electrolyte after storage, suggesting that the complexes such as
PF
base electrolyte can be clearly found by comparing the related FTIR spectra (Figure
DꢀF). It is this interaction that is responsible for the thermal stability improvement.
5 5 2 5
ꢀDTYP and PF ꢀP O have been formed. The interaction between DTYP and the
5
Flame retardance
It is wellꢀestablished that flame propagation is initiated by the radicals, H· and
HO·, and the mechanism of flame retardance relies on how these radicals could be
[
76]
terminated or eliminated. Figure 6A and B present the relative energy (ꢂE) for the
reaction H·+S = H·ꢀS and HO·+S = HO·ꢀS, where S represents solvents, TEP and
ꢀ1
DTYP. The relative energy of ECꢀH, EMCꢀH and TEPꢀH is close to 0 kJ mol (Figure
6A). TEP was considered because it has the same phosphate group as DTYP and has
been used as an effective flameꢀretardant additive. The relative energy of H· or
HO· with TEP and DTYP is far lower than those with solvents (Figure 6A and B),
indicating that TEP and DTYP have ability to retard flame of carbonateꢀbased
2
3

Journal of Materials Chemistry A
Page 24 of 40
View Article Online
DOI: 10.1039/C8TA01236A
electrolyte. Comparatively, DTYP has lower relative energy than TEP when
combining H· or HO·, suggesting that DTYP as flameꢀretardant additive should
function better than TEP. The flame retardance of electrolytes is confirmed by full cell
puncture experiment and electrolyte burning experiment. The former is shown in
Figure S7, where the fully charged cells are vertically punctured from their central
ꢀ1
position by a steel needle of 3 mm in diameter with a rate of 80 mm s . During
puncturing, the cell surface temperatures were monitored. It can be seen from Figure
S7 that flame and smoke are observed for the cell using base electrolyte, but not for
o
o
the cell using DTYPꢀcontaining. The highest cell surface temperature is 112 C, 45 C
o
and 41 C for the cells using base, TEPꢀ and DTYPꢀcontaining electrolytes (As shown
in Figure 6C).
Figure 7A presents the burning of Celgard 2400 microporous separators soaked
with various electrolytes. The separator itself or soaked with base electrolyte is highly
flammable, but that soaked with DTYP is not. The FTIR spectra of the fresh separator
and those after burning (flamed) or those soaked with base (base) and
DTYPꢀcontaining (DTYP) electrolytes after burning are displayed in Figure 7B. The
ꢀ1
fresh separator shows the typical absorptions of polymers (1456, 1375 and 1631 cm
as shown at mark), which disappear after burning, indicating that the separator itself is
flammable. The separator soaked with base electrolyte after burning shows strong
ꢀ1
adsorption peaks at 1490, 1400, 1771, 1639, 1184, 1081, 843 and 777 cm , which are
characteristic of species from burnt electrolyte including Li CO , ROCO Li, CꢀO,
PFy, and C=O. When DTYP is applied, however, the adsorption peaks
2
3
2
[49]
alkoxides, Li
x
2
4

Page 25 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
for the separator itself are maintained, while those for the burnt electrolyte disappear.
ꢀ1
More importantly, new peaks for P=O (peak at 1240 and 1028 cm ) and PꢀOH
species in P appear. It is obvious that the burning of the base electrolyte or even
the separator can be retarded by DTYP, and P O should be responsible. P O captures
2
O
5
2
5
2
5
the radicals, H· and HO·, in the flame zone, so that the chain reactions for the
[
76]
burning of the base electrolyte is terminated.
Figure 7C shows the self extinguishing time (SET) and the ionic conductivity of
the electrolytes containing various DTYP concentrations. The base electrolyte, 1.0 M
LiPF in EC/EMC=3:7 by volume, is highly flammable with a SET of 88 s. When
6
DTYP is applied, the SET decreases. It can be noted that the significant drop in SET
occurs at DTYP concentration of 0.5%, after which the SET adopts an almost constant
value; meanwhile, the ionic conductivity of the electrolyte is reduced from 10.3 to 8.9
ꢀ1
mS cm in the range when 4% DTYP is applied (Figure 7C), but at 0.5% DTYP a
ꢀ1
high value of 10.1 mS cm is still maintained. This slight tradeꢀoff is advantageous
because it is highly desirable for the practical application that the electrolyte additive
concentrations be minimized. Compared with other additives, for example, diethyl
[95]
(2,6ꢀdiꢀtertꢀbutylꢀ4ꢀmethylphenyl) phosphate,
whose 2% presence in electrolyte
ꢀ1
would reduce ionic conductivity from 10.4 to 7.45 mS cm , the efficiency of DTYP is
remarkable.
Conclusions
In this work, we report a novel electrolyte additive, diethyl (thiophenꢀ2ꢀylmethyl)
phosphonate (DTYP), which integrate three functionalities in a single molecule. With
2
5

Journal of Materials Chemistry A
Page 26 of 40
View Article Online
DOI: 10.1039/C8TA01236A
only a small concentration in electrolytes, DTYP not only can effectively stabilize the
interface between carbonateꢀbased conventional electrolytes and a high voltage
cathode, prevent the dissolution of transition metal, but also can render the
electrolytes thermally stable and resistant against combustion. Comprehensive
characterization aided by computation reveals such stabilization to the preferential
oxidation of DTYP, which forms a unique cathode electrolyte interphase, with the
thiophene and phosphate subꢀstructures enabling such interphase with superior
protection against parasitic reactions between electrolytes and active cathode
materials.
Acknowledgments
This work is financially supported by the joint project of National Natural
Science Foundation of China and Natural Science Foundation of Guangdong Province
(Grant No. U1401248), the key project of Science and Technology in Guangdong
Province (Grant No. 2017A010106006), Guangzhou City Project for Cooperation
among Industries, Universities and Institutes (Grant No. 201604016011).
Notes and references
[
[
[
[
1] J.M. Tarascon, M. Armand, Nature 2001, 414, 359.
2] J.B. Goodenough, Y. Kim, Chemistry of Materials 2010, 22, 587.
3] B. Kang, G. Ceder, Nature 2009, 458, 190.
4] L.D. Xing, X.W. Zheng, M. Schroeder, J. Alvarado, A. von W. Cresce, K. Xu, Q.S. Li, W.S. Li,
Accounts of Chemical Research 2018, 5, 282.
[
5] B. Dunn, H. Kamath, J.M. Tarascon, Science 2011, 344, 928.
2
6

Page 27 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
[
[
6] C.F. Zhang, Z.X. Chen, Z.P. Guo, X.W.(David) Lou, Energy﹠Environmerital Science 2013, 6, 674.
7] S.P. Wu, R. Xu, M.J. Lu, R.Y. Ge, J. Iocozzia, C.P. Han, B.B. Jiang, Z.Q. Lin, Advanced Energy
Mater 2015, 5, 1.
[
[
8] M. Chen, Y. G. Zhang, L. D. Xing, Y. H. Liao, Y. C. Qui, S. H. Yang, W. S. Li, Advanced
Materials 2017, 1607015
.
9] Y.C. Yang, B.H. Qiao, X.M. Yang, L.B. Fang, C.C. Pan, W.X. Song, H.S. Hou, X.B. Ji, Advanced
Functional Materials 2014, 24, 4349.
[
[
[
[
[
10] Y Jin, B Zhu, Z.D. Lu, N. Liu, J. Zhu, Advanced Energy Materials 2017, 7, 1700715.
11] K. Kang, Y.S. Meng, J. Breger, C.P. Grey, G. Ceder, Science 2006, 311, 977.
12] L.H. Yu, D.D. Cai, H.H. Wang, M.ꢀM, Titirici, Nanomaterials 2013, 3, 443.
13] T. Mueller, G. Hautier, A. Jain, G. Ceder, Chemistry of Materials 2011, 23, 3854.
14] B. Li, H.J. Yan, J. Ma, P.R. Yu, D.G. Xia, W.F. Huang, W.S. Chu, Z.Y. Wu, Advanced Functional
Materials 2014, 24, 5112.
[
[
15] A. Kraystsbery, Y. EinꢀEli, Advanced Energy Materials 2012, 2, 922.
16] H. Arai, K. Sato, Y. Orikasa, H. Murayama, L. Takahashi, Y. Koyama, Y. Uchimoto, Z. Ogumi,
Journal of Materials Chemistry A 2013, 1, 10442.
[
[
17] L. Zhou, D.Y. Zhao, X.W.(David) Lou, Angewandlte Chemie 2012, 124, 243.
18] G.J. Xu, C.G. Pang, B.B. Chen, J. Ma, X. Wang, J.C. Chai, Q.F. Wang, W.Z. An, X.H. Zhou, G.L.
Cui, L.Q. Chen, Advanced Energy Materials 2017, 1701398.
[
19] P.V. Sushko, K.M. Rosso, J.ꢀG. Zhang, J. Liui, M.L. Sushko, Advanced Functional Materials 2013,
23, 5530.
[
[
[
20] A. Abouimrane, I. Belharouak, K. Amine, Electrochemistry Communications 2009, 11, 1073.
21] C.L. Campion, W.T. Li, B.L. Lucht, Journal of The Electrochemical Society 2005, 152, A2327.
22] L.D. Xing, O. Bordin, G.D. Smith, W.S. Li, The Journal of Physical Chemistry A 2011, 115,
1
3896.
23] M.Q. Xu, Y.L. Liu, B. Li, W.S. Li, X.P. Li, S.J. Hu, Electrochemistry Communications 2012, 18,
23.
[
1
[
[
24] M. Hu, X.L. Pang, Z. Zhou, Journal of Power Sources 2013, 237, 229.
25] Y.M. Zhu, X.Y. Luo, M.Q. Xu, L.P. Zhang, L. Yu, W.Z. Fan, W.S. Li, Journal of Power Sources
2016, 317, 65.
[
26] L.D. Xing, C.Y. Wang, W.S. Li, M.Q. Xu, X.L. Meng, S.F. Zhao, The Journal of Physical
Chemistry B 2009, 113, 5181.
[
[
27] J.W. Fergus, Journal of Power Sources 2010, 195, 939.
28] D.S. Lu, M.Q. Xu, L. Zhou, A. Garsuch, B.L. Lucht, Journal of The Electrochemical Society 2013,
160, A3138.
[
29] S.W. Kim, V.G. Kumar, D.H. Seo, Y.U. Park, J. Kim, H. Kim, J. Kim, J. Hong, K. Kang,
Electronic Materials Letters 2011, 7, 105.
[
[
[
[
30] J. Liu, A. Manthiram, Chemistry of Materials 2009, 21, 1695.
31] Y.ꢀK. Sun, C.S. Yoon, I.ꢀH. Oh, Electrochimica Acta 2003, 48, 503.
32] R. Alcantara, M. Jaraba, P. Lavela, J.L. Tirado, Electrochimica Acta 2002, 47, 1829.
33] H.M. Wu, I. Belharouak, A. Abouimrane, Y.ꢀK. Sun, K. Amine, Journal of Power Sources 2010,
195, 2909.
[
[
34] J.G. Li, Y.Y. Zhang, J.J. Li, L. Wang, X.M. He, J. Gao, Ionics 2011, 17, 671.
35] Y. Fan, J.M. Wang, Z. Tang, W.C. He, J.Q. Zhang, Electrochimica Acta 2007, 52, 3870.
2
7

Journal of Materials Chemistry A
Page 28 of 40
View Article Online
DOI: 10.1039/C8TA01236A
[
[
36] R. Santhanam, B. Rambabu, Journal of Power Sources 2010, 195, 5442.
37] H.B. Rong, M.Q. Xu, B.Y. Xie, X.L. Liao, W.Z. Huang, L.D. Xing, W.S. Li, Electrochimica Acta
014, 147, 31.
2
[
[
[
38] T.Y. Yang, N.Q. Zhang, Y. Lang, K.N. Sun, Electrochimica Acta 2011, 56, 4058.
39] X. Fang, M.Y. Ge, J.P. Rong, C.W. Zhou, Journal of Materials Chemistry A 2013, 1, 4083.
40] K. Dai, J. Mao, Z. Li, Y. Zhai, Z. Wang, X. Song, V. Battaglia, G. Liu, Journal of Power Sources
2014, 248, 22.
[
[
[
41] H.B. Lin, J.N. Hu, H.B. Rong, Y.M. Zhang, S.W. Mai, L.D. Xing, M.Q. Xu, X.P. Li, W.S. Li,
Journal of Materials Chemistry A 2014, 2, 9272.
42] J.W. Liu, X. Song, L. Zhou, S.Q. Wang, W. Song, W. Liu, H.L. Long, L.X. Zhou, H.M. Wu, C.Q.
Feng, Z.P. Guo, Nano Energy 2018, 46, 404.
43] B. Li, Y.Q. Wang, H.B. Rong, Y.T. Wang, J.S. Liu, L.D. Xing, M.Q. Xu, W.S. Li, Journal of
Materials Chemistry A 2013, 1, 12954.
[
[
44] A.V. Cresce, K. Xu, Journal of The Electrochemical Society 2011, 158, A337.
45] Y.ꢀM. Song, J.ꢀG. Han, S. Park, K.T. Lee, N.ꢀS. Choi, Journal of Materials Chemistry A 2014, 2,
9506.
[
[
[
[
[
46] W.N. Huang, L.D. Xing, T.T. Wang, M.Q. Xu, W.S. Li, F.C. Xie, S.G. Xia, Journal of Power
Sources 2014, 267, 560.
47] W.N. Huang, L.D. Xing, R.Q. Zhang, X.S. Wang, W.S. Li, Journal of Power Sources 2015, 293,
71.
48] B. Ravdel, K.M. Abraham, R. Gitzendanner, J. DiCarlo, B.L. Lucht, C. Campion, Journal of
Power Sources 2003, 119ꢀ121, 805.
49] M.Q. Xu, L. Zhou, L.S. Hao, L.D. Xing, W.S. Li, B.L. Lucht, Journal of Power Sources 2011,
1
96, 6794.
50] M.Q. Xu, L.S. Hao, Y.L. Liu, W.S. Li, L.D. Xing, B. Li, The Journal of Phyiscal Chemistry C
011, 115, 6085.
2
[
[
[
51] K. Xu, C.A. Angell, Journal of The Electrochemical Society 2002, 149, A920.
52] S. Li, Y. Zhao, X. Shi, B. Li, X. Xu, W. Zhao, X. Cui, Electrochimica Acta 2012, 65, 221.
53] K. Xu, B. Deveney, K. Nechev, Y. Lam, T.R. Jow, Journal of The Electrochemical Society 2008,
155, A959.
[
[
54] A.J. Gmitter, I. Plitz, G.G. Amatucci, Journal of The Electrochemical Society 2012, 159, A370.
55] Y.S. Zhu, S.Y. Xiao, Y. Shi, Y.Q. Yang, Y.Y. Hou, Y.P. Wu, Advanced Energy Materials 2014, 4,
1
300647.
56] P. Johansson, L.E. Fast, A. Matic, G.B. Appetecchi, S. Passerini, Journal of Power Sources 2010,
95, 2074.
[
[
1
57] S. Seki, Y. Ohno, H. Miyashiro, Y. Kobayashi, A. Usami, Y. Mita, N. Terada, K. Hayamizu, S.
Tsuzuki, M. Watanabe, Journal of The Electrochemical Society 2008, 155, A421.
[
[
58] H.H. Zheng, B. Li, Y.B. Fu, T. Abe, Z. Ogumi, Electrochimica Acta 2006, 52, 1556.
59] Y.D. Wang, K. Zaghib, A, Guerfi, F.F.C. Bazito, R.M. Torresi, J.R. Dahn, Electrochimica Acta
2007, 52, 6346.
[
[
60] K. Fic, G. Lota, M. Meller, E. Frackowiak, Energy
﹠ Environmerital Science 2012, 5, 5842.
61] D. Zhou, Y.ꢀB. He, R.L. Liu, M. Liu, H.D. Du, B.L. Li, Q. Cai, Q.ꢀH. Yang, F.Y. Kang, Advanced
Energy Materials 2015, 5, 1500353.
[
62] Y.H. Liao, T.T. Chen, X.Y. Luo, Z. Fu, X.P. Li, W.S. Li, Journal of Membrane Science 2016, 507,
2
8

Page 29 of 40
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA01236A
126.
[
[
[
63] X.Y. Luo, Y.H. Liao, H.L. Xie, Q.M. Huang, W.S. Li, Ionics 2016, 22, 1035.
64] D.Y. Zhou, W.S. Li, C.L. Tan, X.X. Zuo, Y.J. Huang, Journal of Power Sources 2008, 184, 589.
65] D. Gordon, M.Y. Wu, A. Ramanujapram, J. Benson, J.T. Lee, A. Magasinski, N. Nitta, C. Huang,
G. Yushin, Advanced Energy Materials 2016, 6, 1501805.
[
[
[
[
66] M.Q. Xu, Y. Liang, B. Li, L.D. Xing, Y. Wang, W.S. Li, Materials Chemistry and Physics 2014,
143, 1048.
67] M.N. He, C.ꢀC. Su, Z.X. Feng, L. Zeng, T.P. Wu, M.J. Bedzyk, P. Fenter, Y. Wang, Z.C. Zhang,
Advanced Energy Materials 2017, 7, 1700109.
68] H.B. Lin, W.Z. Huang, H.B. Rong, S.W. Mai, J.N. Hu, L.D. Xing, M.Q. Xu, W.S. Li, Journal of
Solid State Electrochemistry 2015, 19, 1123.
69] Ali. Abouimrane, S.A. Odom, H. Tavassol, M.V. Schulmerich, H.M. Wu, R. Bhargava, A.A.
Gewirth, J.S. Moore, K. Amine, Journal of The Electrochemical Society 2013, 160, A268.
70] L.Y. Xing, M. Hu, Q. Tang, J.P. Wei, X. Qin, Z. Zhou, Electrochimica Acta 2012, 59, 172.
71] S.L. Li, L. Xia, H.Y. Zhang, X.P. Ai, H.X. Yang, Y.L. Cao, Journal of Power Sources 2011, 196,
[
[
7021.
[
[
72] L. Xia, Y.G. Xia, Z.P. Liu, Electrochimica Acta 2015, 151, 429.
73] N. Oyama, T. Sarukawa, Y. Mochizuki, T. Shimomura, S. Yamauchi, Journal of Power Sources
2
009, 189, 230.
74] K.ꢀS. Lee, Y.ꢀK. Sun, J. Nob, K.S. Song, D.ꢀW. Kim, Electrochemistry Communications 2009,
1, 1900.
75] S. Dalavi, M.Q. Xu, B. Ravdel, L. Zhou, B.L. Lucht, Journal of The Electrochemical Society 2010,
57, A1113.
[
[
[
[
[
[
1
1
76] O.P. Koobeinichev, S.B. Ilyin, T.A. Bolshova, V.M. Shvartsberg, A.A. Chernov, Combustion and
Flame 2000, 121, 593.
77] H.F. Xiang, H.W. Lin, B. Yin, C.P. Zhang, X.W. Ge, C.H. Chen, Journal of Power Sources 2010,
195, 335.
78] M.Q. Xu, N. Tsiouvaras A. Garsuch, H.A. Gasteiger, B.L. Lucht, The Journal of Phyiscal
Chemistry C 2014, 14, 7363.
79] J.C. Arrebola, A. Caballero, M. Cruz, L. Hernan, J.L. Morales, E.R. Castellon, Advanced
Functional Materials 2006, 16, 1904.
[
[
[
80] Y.ꢀK. Sun, K.ꢀJ. Hong, P. Jai, K. Amine, Electrochemistry Communications 2002, 4, 344.
81] H.B. Rong, M.Q. Xu, L.D. Xing, W.S. Li, Journal of Power Sources 2014, 261, 148.
82] D.S. Lu, W.S. Li, X.X. Zuo, Z.Z. Yuan, Q.M. Huang, The Journal of Physical Chemistry C 2007,
111, 12067.
[
[
83] H. Huang, W.K. Zhang, X.P. Gan, C. Wang, L. Zhang, Material Letters 2007, 61, 296.
84] J. Kalhoff, D. Bresser, M. Bolloli, F. Alloin, J.ꢀY. Sanchez, S. Passerini, ChemSusChem 2014, 7,
2939.
[
[
[
85] N. Loeffler, J.V. Zamory, N. Laszczynski, I. Doberdo, G.ꢀT. Kim, S. Passerini, Journal of Power
Sources 2014, 248, 915.
86] L.D. Xing, W.Q. Tu, Jenel. Vatamanu, Q.F. Liu, W.N. Huang, Y.T. Wang, H.B. Zhou, R.H. Zeng,
W.S. Li, Electrochimica Acta 2014, 133, 117.
87] Y.M. Zhu, H.B. Rong, S.W. Mai, X.Y. Luo, X.P. Li, W.S. Li, Journal of Power Sources 2015, 299,
485.
2
9