www.afm-journal.de www.advancedsciencenews.

com

Fluoroethylene Carbonate Additives to Render Uniform

full paper

Li Deposits in Lithium Metal Batteries

Xue-Qiang Zhang, Xin-Bing Cheng, Xiang Chen, Chong Yan, and Qiang Zhang*
Nevertheless, the practical application
Lithium (Li) metal has been considered as an important substitute for the of Li metal anode is severely hindered by
graphite anode to further boost the energy density of Li-ion batteries. How- the uncontrollable formation of dendritic
ever, Li dendrite growth during Li plating/stripping causes safety concern and and mossy Li on Li anode surface as well
poor lifespan of Li metal batteries (LMB). Herein, fluoroethylene carbonate as the low Coulombic efficiency during
repeated stripping/plating process.[5,6] Li
(FEC) additives are used to form a LiF-rich solid electrolyte interphase (SEI).
dendrites are generally induced by inho-
The FEC-induced SEI layer is compact and stable, and thus beneficial to mogeneous distributions of space charge,[7]
obtain a uniform morphology of Li deposits. This uniform and dendrite-free current density on anode surface,[8] and
morphology renders a significantly improved Coulombic efficiency of 98% the crack of solid electrolyte interphase
within 100 cycles in a Li | Cu half-cell. When the FEC-protected Li metal anode (SEI).[9] During the past several decades,
matches a high-loading LiNi0.5Co0.2Mn0.3O2 (NMC) cathode (12 mg cm−2), much effort has been devoted to prevent
the parasitic growth of Li dendrites in a
a high initial capacity of 154 mAh g−1 (1.9 mAh cm−2) at 180.0 mA g−1 is
working battery.[4,5,10,11] Among these strat-
obtained. This LMB with conversion-type Li metal anode and intercalation- egies, in situ formation of SEI with high
type NMC cathode affords an emerging energy storage system to probe uniformity and stability is one of the most
the energy chemistry of Li metal protection and demonstrates the material effective and convenient routes to suppress
engineering of batteries with very high energy density. Li dendrite growth due to its prominent
impact and low cost, particularly in indus-
trial manufacture of batteries.[12] Many
electrolyte additives are employed to facili-
tate the formation of stable SEI layer on Li metal anode, such as
1. Introduction
vinylene carbonate (VC),[13] fluoroethylene carbonate (FEC),[14,15]
The pursuit of an electrochemical energy storage system with LiNO3,[16] Cs+,[17] Li polysulfide,[18] lithium chloride,[19,20] copper
high energy density has never stopped since the first battery acetate,[21] and highly concentrated electrolyte.[22] Li metal anode
was invented by Count Volta in 1800. In the early 1990s, the with protective SEI layer exhibits a superior cycling performance
Sony Corporation commercially introduced a lithium (Li)- in high-energy-density Li metal battery (LMB).
ion battery (LIB) based on the use of Li-intercalation graphite Among the LMBs, Li–sulfur (LiS) and Li–oxygen (LiO2)
anode.[1] As LIB can deliver an energy density of at least 2.5 batteries with a high theoretical energy density of 2600 and
times higher than that of nickel–cadmium and lead–acid bat- 3500 Wh kg−1 have been strongly explored,[23] while LMB
teries, it realized a great success in portable devices, such pairing intercalation-type cathode (such as LiNi0.5Co0.2Mn0.3O2
as mobile phones and laptops.[2] However, either success or (NMC)) versus Li metal anode is little involved. This NMC-
failure of LIB all boils down to graphite anode. The relatively based LMB employs the same cathode with LIB, which is very
low specific capacity of graphite anode (372 mAh g−1) limits easy to integrate with the current LIB manufacturing process
the energy density of LIB to satisfy the demand of modern and further improve the energy density of the current LIB.[24]
electric vehicles and wearable devices. Among various anode However, Li deposition behavior has not been understood well
materials, metallic lithium anode with an extremely high theo- in this system. Lu et al. described that when charging LMB with
retical specific capacity (3860 mAh g−1) and the lowest reduc- lithium nickel cobalt aluminum oxide cathode versus Li anode
tion potential (−3.04 V vs standard hydrogen electrode) has at high rate, large dendrites and dead Li formed on the Li metal
been strongly reconsidered as a promising next-generation anode, leading to a large polarization in a working battery.[25]
anode.[3,4] The highly resistive porous dead Li layer dramatically increased
the cell impedance and resulted in cell degradation and final
failure. To suppress Li dendrite growth in these LMBs, Zheng
X.-Q. Zhang, X.-B. Cheng, X. Chen, et al. manipulated the cell operation model at low charge/rapid
C. Yan, Prof. Q. Zhang discharge rate to facilitate the formation of a stable and flex-
Beijing Key Laboratory of Green Chemical
Reaction Engineering and Technology ible SEI to suppress Li dendrite growth.[26] However, the mostly
Department of Chemical Engineering applied strategy of electrolyte additives to benefit the formation
Tsinghua University of robust SEI is rarely employed in the high-energy-density
Beijing 100084, China LMB.[27] It is of great significance to clearly demonstrate the Li
E-mail: [email protected]
deposition behavior and investigate the cycling performance of
DOI: 10.1002/adfm.201605989 these LMBs in organic electrolytes with additives.

1605989 (1 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Funct. Mater. 2017, 27, 1605989
www.advancedsciencenews.com www.afm-journal.de

full paper
Figure 1. Schematic illustration of the effect of FEC additives on a Li metal anode. The electrolyte is 1.0 m LiPF6 in EC/DEC (1:1 by volume) with/
without FEC additives.

Here, we report the use of FEC additives in high-energy- anode. The correspondingly proposed reaction mechanism for
density LMB to protect a Li metal anode (Figure 1). The NMC generation of LiF is shown in Scheme S1 (Supporting Infor-
cathode was chosen to match the Li metal anode, which can mation). The as-obtained LiF is considered as an important
effectively improve the cathode capacity relative to the conven- SEI component to form the uniform and compact lithium
tional oxide cathodes (LiFePO4, LiCoO2, etc.). We primarily veri- deposits.[15,20] EC solvent can also decompose on the Li sur-
fied the hypothesis of LiF-rich SEI induced by FEC additives face. However, there is no/less EC decomposition reaction
through first-principles calculations and X-ray photoelectron observed on the LiF surface during the same ab initio molec-
spectroscopy (XPS). In the Li | Cu half-cell test, we demon- ular dynamics (AIMD) (Figure S1, Supporting Information).
strated a uniform and dense morphology of Li deposits as well Therefore, it can be inferred that the LiF products induced by
as improved cell cycling performance with the presence of FEC. FEC sacrifice act as the protective SEI layer on Li anode. With
Voltage polarization was greatly reduced and Coulombic effi- the FEC-induced SEI protecting Li metal, parasitic reactions
ciency was effectually improved due to the thinner and denser between Li and electrolyte are effectively suppressed.
interphase on Li anode. In the Li | NMC cell, a high capacity XPS of the anode was conducted to further probe the
and long lifespan are achieved for the FEC-protected Li metal surface chemistry of FEC-induced SEI layer (Figure 3 and
anode. Figure S2, Supporting Information). A high F content of
10.8 at% is achieved in the FEC electrolyte, which is apparently

2. Results and Discussion

2.1. FEC-Induced SEI Formation

FEC is beneficial for robust SEI formation

on graphite,[28] silicon,[29] and high-voltage
cathode.[30,31] To gain the molecular insights
into the SEI formation, we first investigated
the SEI formation induced by FEC addi-
tives in ethylene carbonate (EC)/diethyl car-
bonate (DEC) by first-principles calculation
(Figure 2a,b). During cell charging, FEC with
lower level of the lowest unoccupied molec-
ular orbital (LUMO) (−0.87 eV) is easier to be
reduced than EC (−0.38 eV) and DEC (0 eV)
on the surface of Li metal anode (Figure 2b),
significantly decreasing the consumption of
solvent.
During the FEC-induced SEI formation,
the carbon–fluorine bond is first broken
Figure 2. First-principles study of the role of FEC. a) Molecular structure of EC, FEC, and
around 310 fs (Figure 2c) and the fluorine–
DEC. b) Visual LUMO and corresponding relative energy of EC, FEC, and DEC. c) Ab initio
lithium distance is near to that in lithium molecular dynamics model. d) Complete sequence of FEC molecule decomposed on Li anode.
fluoride crystal (0.2032 nm), indicating the The hydrogen, lithium, carbon, oxygen, and fluorine atom were marked with white, purple, gray,
generation of LiF on the surface of Li metal red, and blue, respectively.

Adv. Funct. Mater. 2017, 27, 1605989 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (2 of 8) 1605989
 www.afm-journal.de www.advancedsciencenews.com
full paper

electrolyte, an LiF-rich SEI layer is achieved,

which is expected to suppress Li dendrite
growth and render a long lifespan of Li metal
anode.

2.2. Li | Cu Half-Cell Performance

To evaluate the long-term stability of Li

plating/stripping behavior in a working bat-
tery, the coin-type Li | Cu half-cells with and
without FEC additives in electrolyte were
assembled. In order to obtain a stable and
low-viscosity electrolyte, the amount of FEC
additives in Li hexafluorophosphate (LiPF6)-
EC/DEC electrolyte is set as 5% (by volume)
in this work. The LiPF6-EC/DEC electrolyte
without FEC additive (0% FEC) was also pre-
pared as a control sample.
Coulombic efficiency was employed to
indicate Li utilization rate during one Li
plating/stripping cycle and calculated by the
ratio of the amount of Li stripped and plated
on Cu substrate. The Coulombic efficien-
cies of Li | Cu cells at 0.1 and 0.5 mA cm−2
are shown in Figure 4a,b, respectively. The
initial Coulombic efficiency of cells in 0%
and 5% FEC electrolyte is 78% and 90% at
0.1 mA cm−2, respectively, demonstrating a
superior role of FEC in constructing a highly
efficient SEI (Figure 4a). During 100 cycles
(>1000 h), the Coulombic efficiency of 0%
FEC electrolyte is evidently lower than 92%,
while a remarkably high Coulombic effi-
Figure 3. XPS characterization of the SEI layer. a) XPS spectra of the SEI layer, the inset table: ciency of 98% is achieved in 5% FEC elec-
atomic ratio of elements in the SEI layer. b,c) F 1s and Li 1s spectra of the SEI layer induced by trolyte. The superior cycling performance of
0% and 5% FEC after lithium stripping on Cu substrate after ten cycles. 5% FEC electrolyte is attributed to the uni-
form and dendrite-free Li metal protected
larger than that of 9.3 at% in FEC-free routine electrolyte by the LiF-rich SEI. When the current density is increased
(Figure 3a). There are two peaks in the F 1s spectra: LiF at to 0.5 mA cm−2, Coulombic efficiencies of 88% and 95% are
684.8 eV and CF bonds at 686.6 eV (Figure 3b). The propor- obtained in 0% and 5% FEC electrolyte (Figure 4b), respec-
tion of LiF in the FEC-containing electrolyte induced SEI layer tively, further confirming the effective role of FEC-induced LiF-
(48.2%) accounts more than that of in the FEC-free induced rich SEI at high current density. A 2% FEC electrolyte was also
SEI layer (29.5%) according to the ratio of peak area. There- prepared. The Coulombic efficiency of 2% FEC electrolyte is
fore, more LiF forms in FEC-induced SEI layer than FEC-free inferior to 5% FEC electrolyte but superior to 0% FEC electro-
induced SEI layer. lyte (Figure S3, Supporting Information), which demonstrates
The LiF-rich SEI is the product of spontaneous reactions that 5% FEC is a better choice for further investigation.
between fresh Li and FEC, instead of the adverse reaction The polarization profiles in different cycles were recorded to
between Li and EC or DEC. LiF is critically important in the gain the kinetic behavior during Li stripping/plating (Figure 4c
formation of stable and uniform SEI to suppress Li dendrite and Figure S4, Supporting Information). The polarization of
growth.[15,20] On one hand, LiF is a robust electrical insulator 5% FEC electrolyte at 0.5 mA cm−2 from the charge/discharge
(≈10 −31 −1
S cm ) to prevent electrons from crossing the SEI curves is nearly the same, 50 mV at both 10th and 50th cycles,
layer.[32] On the other hand, LiF with a low diffusing energy while it is 58 mV at 10th and 72 mV at 50th cycles for 0% FEC
and high surface energy for Li ions exhibits an overwhelming electrolyte. The large polarization of 0% FEC electrolyte is
advantage in enhancing surface diffusion of Li+ during elec- caused by the continuous Li dendrite growth. These dendrites
trodeposition and directing a uniform and dendrite-free mor- result in a highly resistive layer with SEI entangled with dead Li
phology.[20] In Li 1s spectra, a similar result can be reached metal, which increases the ion and electron diffusion resistance
(Figure 3c). The peak of LiF (55.7 eV) is also enhanced in FEC- within the anode matrix. Therefore, the polarization is rising
induced SEI layer. Consequently, by introducing FEC into the gradually. In contrast, Li anode protected by the LiF-rich SEI

1605989 (3 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Funct. Mater. 2017, 27, 1605989
www.advancedsciencenews.com www.afm-journal.de

full paper
Figure 4. Electrochemical performance and SEM images of Li | Cu cells. Coulombic efficiency (CE) at a current density of a) 0.10 and b) 0.50 mA cm−2
with a capacity of 0.5 mAh cm−2. c) Polarization curves of plating/stripping process in (b). d) EIS of Li | Cu cells at frequency ranging from 105 to
10−1 Hz under amplitude of 10 mV. SEM images of Li depositing morphology on Cu foils after 50 cycles with e) 0% and f) 5% FEC.

exhibits a stable interface and dendrite-free Li depositing mor- Information). 0% FEC electrolyte results in an uneven sur-
phology, and therefore small and stable overpotential during Li face and optically visible dead and porous Li on the substrate.
stripping/plating. When FEC was introduced into the electrolyte, a uniform and
Electrochemical impedance spectroscopy (EIS) measure- smooth surface without any dendrite and porous structure was
ments of Li | Cu half-cells at 0.5 mA cm−2 were carried out to obtained. The porous and loose layer is mainly composed of
reveal the conductivity of SEI on Li metal anode. Nyquist plots deposited Li losing contact with electrons (dead Li).[25,34] For
of the cells with 0% and 5% FEC are with one semicircle in dendritic and nonuniform Li due to the lack of protective SEI,
high frequency (Figure 4d), which is always ascribed to Li+ dead Li forms easily during the stripping process. The accumu-
migration through the SEI on the electrode surface.[33] 5% FEC lation of dead Li in repeated plating/stripping process leads to
electrolyte can effectively decrease the SEI resistance at second porous and resistive Li films in 0% FEC electrolyte. However,
cycle (60 Ω for 5% FEC electrolyte, 110 Ω for 0% FEC electro- the LiF-rich SEI induced by FEC additives significantly benefits
lyte), implying high conductivity of Li+ through SEI layer. After the uniform Li deposits and reduces the formation of dead Li,
50 cycles, the cell with 5% FEC exhibits evidently lower SEI leading to dense and less resistive deposited Li films.
resistance than that with 0% FEC electrolyte (110 Ω for 5% FEC
electrolyte, 250 Ω for 0% FEC electrolyte). Consequently, SEI
formed in the FEC electrolyte is more stable and conductive, 2.3. Li | NMC Full-Cell Performance
rendering a high Li-ion conductivity.
After 50 cycles in coin-type Li | Cu half cells, morphologies of We verified the concept of LMB with intercalation-type NMC
Li deposits were recorded (Figure 4e,f and Figure S5, Supporting cathode and conversion-type Li metal anode. Commercial

Adv. Funct. Mater. 2017, 27, 1605989 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (4 of 8) 1605989
 www.afm-journal.de www.advancedsciencenews.com
full paper

Figure 5. Electrochemical performance and SEM images of Li | NMC cells. a) Cycling performance and b) voltage profiles of cells at 1.0 C, 3.0–4.3 V,
with one formation cycle initially performed at 0.1 C prior to 1.0 C cycling. c) EIS of Li | NMC cells after 2 and 50 cycles. SEM images of the Li anode
surfaces after Li plating obtained from the Li | NMC cells at 1 C after cycling for 50 cycles with d) 0% and e) 5.0% FEC.

LiNi0.5Co0.2Mn0.3O2 cathode with a very high areal loading of decay for 5% FEC electrolyte (65% retention after 100 cycles).
12–14 mg cm−2 was used in this work. Prior to investigating The fading of the capacity in the Li | NMC cells is synergy
the FEC-protected Li metal, the effect of FEC on high-voltage effects among Li anode, NMC cathode, and organic electro-
cathode materials as film-forming additives must be excluded. lyte. The charge–discharge voltage profiles at different cycles
Cyclic voltammetry (CV) confirms that there are no significant (Figure 5b) clearly demonstrate the higher capacity retention
decompositions of FEC in the range of 3.0–4.3 V (Figure S6, in 5% FEC electrolyte (65%) than that in 0% FEC electrolyte
Supporting Information). (14%), indicating FEC is apparently beneficial for the capacity
Figure 5 exhibits the cycling performance of Li | NMC cells stability of Li | NMC cells.
with 0% and 5% FEC electrolyte at 1.0 C (1.0 C = 180 mA g−1) Figure 5c displays the Nyquist plots of Li | NMC cells. The
after 0.1 C activation for one cycle. All cells deliver nearly the cells with 0% and 5% FEC electrolyte are with nearly the same
same reversible capacity (154 mAh g−1) and Coulombic effi- resistance (40 Ω) at second cycle. However, with the depth of dis-
ciency of 99.5% in the initial cycles (Figure 5a). The effect of charge, the SEI resistance of 0% FEC electrolyte rises to 100 Ω,
FEC is not conspicuous in the initial 40 cycles. However, a while that of 5% FEC electrolyte is only 60 Ω after 50 cycles.
sharp capacity decay at 40th cycle for 0% FEC is recorded. The The ever-increasing resistance indicates the unstable interface
precipitous drop in capacity is attributed to the failure of the and crackly SEI of 0% FEC electrolyte.
conductive framework in the anode, which is induced by the To obtain the depositing morphology of Li metal anode in
highly resistive layer with SEI entangled with dead Li metal. Li | NMC cells, cycled Li anodes were characterized by scanning
There is no sharp capacity decay in 5% FEC electrolyte, due to electron microscopy (SEM) images. As shown in Figure 5d,e,
the stable LiF-rich SEI and uniform Li deposition contributed the surface of Li anode (in charged state) in 0% electrolyte is
by FEC additives. However, there is also a continuous capacity porous and loose, exhibiting the existence of dendrite growth

1605989 (5 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Funct. Mater. 2017, 27, 1605989
www.advancedsciencenews.com www.afm-journal.de

and dead Li, while the surface of cycled Li anode obtained from LiS and Li–air battery. Commercialized NMC with high spe-

full paper
5% FEC electrolyte is dense and uniform. The morphology of cific capacity was employed as high-voltage cathode materials
Li anode in discharged state was also recorded in Figure S7 and paired with Li metal anode to construct Li | NMC cells.[26,49]
(Supporting Information). Uniform stripping of Li is observed FEC additive can effectively prolong the cycling life, reduce the
in 5% FEC electrolyte compared to 0% FEC electrolyte. From interphase resistance and the amounts of dead Li. This contri-
the cross-section view of deposited Li, there is a layer consisting bution provides new perspective for the effect of LiF-rich SEI
of Li participating in cycling (reactive Li) on the top of bulk Li induced by FEC on uniform Li deposits and high Coulombic
(Figure S8, Supporting Information). A porous and loose dead efficiency, and simultaneously exhibits the potential application
Li layer induced by dendrite growth results in the formation of of LMB with conversion-type Li metal anode and intercalation-
a thick reactive Li layer, affording a high Li-ion diffusion resist- type NMC cathode. However, the mechanism of reductive
ance in a working cell. The thickness of reactive Li in 5% FEC decomposition of FEC should be further probed to the dynamic
electrolyte (≈40 µm) is much less than that in 0% FEC electro- role of FEC in forming stable SEI profoundly.
lyte (≈100 µm). The thinner reactive Li layer in 5% FEC electro-
lyte indicates the reduction of the formation of dead Li and Li
dendrites. 3. Conclusions
FEC additives were used in a working battery to achieve an
2.4. The Role of FEC on Li Metal Protection LiF-rich SEI and therefore effectively protect Li metal. The first-
principles calculations indicate that FEC with a very low LUMO
Based on the aforementioned results, it is clearly implied that level of −0.87 eV is very easy to reduce on the surface of Li
the FEC-induced SEI is more stable and conductive, promoting metal anode to form LiF. The LiF-rich SEI is efficient and effec-
the uniform Li stripping and plating and superior long-term tive to suppress Li dendrite growth. The Coulombic efficiency
cycling performance of Li | NMC cells. Recently, LiF is verified of Li | Cu cells is 98% and a reduced and stable polarization of
to significantly promote Li-ion diffusion on Li anode/electrolyte 50 mV is achieved with 5% FEC electrolyte. This affords uni-
interface, which renders the possibility of uniform Li deposition form and dense morphology of deposited Li in a working cell.
in organic electrolyte.[35] FEC additive, a frequently used electro- When the FEC-protected Li metal anode matches a high loading
lyte additive in LIB, is reported to lead to LiF-containing com- NMC cathode (12 mg cm−2) to assemble an LMB, a high initial
ponents on graphite,[28,36] silicon anode,[29,37] hard carbon,[38] capacity of 154 mAh g−1 (1.9 mAh cm−2) at 180.0 mA g−1 was
TiO2@C anode,[39] SnP composite,[40] Sb-based alloys,[41] obtained. A sharp capacity drop of cells in FEC-free electrolyte
fluorine-doped SnNi film electrode,[42] LixSiy alloy,[43] Sn,[44] is induced by unstable interface with large dead Li covered on
MoS2-C anode,[45] Mo6S8,[46] and so on. FEC is found to effec- Li anode. The FEC-protected Li metal anode guarantees the
tively enhance the cycling performance of high-voltage Li-ion efficient operation of LMB with long cycle life. The strategy of
cells.[30] However, the systematical insight into the effect of FEC Li metal battery with conversion-type Li metal anode and inter­
additives on in situ formed SEI on the surface of Li metal in calation-type NMC cathode affords an emerging energy storage
LMB (such as Li | NMC cell) is lacking. Moreover, FEC additive system to probe the energy chemistry of Li metal protection
is soluble in carbonate electrolyte and easy to process, which and demonstrate the materials engineering of batteries with
circumvents the trouble such as toxicity, corrosion, and insolu- very high energy density. The in situ generated stable LiF-rich
bility led by HF[47] and solid LiF[20] in practical applications. SEI strategy is expected to be applied in other metal batteries
Carbonate electrolytes are compatible with high-voltage (e.g., LiS, Li–air, and Zn–air batteries) with dendrite-free
cathode (such as NMC), however, carbonate electrolytes (such growth and very high Coulombic efficiency.
as EC, PC, DMC, etc.) generally exhibit low Coulombic effi-
ciency (≈85%) compared to ether electrolyte.[10,48] The addition
of FEC enhances the Coulombic efficiency of carbonate electro-
lyte. We first verified the hypothesis of the formation of LiF-rich 4. Experimental Section
SEI in the presence of FEC additive by first-principles calcula- Materials: Li metal foil was commercially available from China Energy
tion and XPS. FEC with lower LUMO energy is reduced prior Lithium Co., Ltd. The electrolyte composed of LiPF6 (1.0 m) and EC/DEC
to EC and DEC, forming LiF-rich SEI to prevent the reactions (1:1 by volume) was purchased from the Beijing Institute of Chemical
between electrolyte and Li metal. Then further researches in Reagents. FEC was purchased from Alfa Aesar. NMC cathode materials
(NMC coated on Al foil) were purchased from Henan Huanyu Power
Li | Cu cells show that high Coulombic efficiency (≈98%), uniform
Source Co., Ltd., containing 80% active material, 10% super P, and 10%
and dense Li deposits after long cycles in FEC electrolyte are polyvinylidene fluoride binder. The typical active materials loadings of
achieved. EIS measurement also demonstrates that the resist- cathode were about 12–14 mg cm−2. All the NMC electrodes were dried
ance is largely reduced in FEC electrolyte compared to FEC-free under vacuum for 6 h before use.
electrolyte. These results confirm that the LiF-rich SEI induced Structure Characterizations: The morphologies of the Li metal
by FEC is uniform and conductive for Li-ion diffusion, realizing anodes and corresponding Li deposits were characterized by SEM
smooth and dense Li deposits, high Coulombic efficiency, and (JSM 7401F, JEOL Ltd., Japan) operated at 3.0 kV. An AlKα radiation
(72 W, 12 kV) at a pressure of 10−9 Torr was applied to acquire XPS
reduced formation of dead Li. spectra on ESCALAB 250Xi (Thermo Fisher Scientific Inc., USA). The
LMB employing Li metal anode and intercalated cathode diameter of the analyzed area was 400 µm. The Li metal anodes and Li
(such as NMC) provides more promising opportunities to deposits on Cu foil (stripping state) sample obtained from disassembled
achieve a high-energy-density secondary battery compared to cells were first cleaned by DEC solvent three times, and then dried until

Adv. Funct. Mater. 2017, 27, 1605989 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (6 of 8) 1605989
 www.afm-journal.de www.advancedsciencenews.com
full paper

the solvent volatilized thoroughly in the glove box. During transferring [3] a) M. D. Tikekar, S. Choudhury, Z. Tu, L. A. Archer, Nat. Energy
process before characterizations, all samples were protected with Ar in 2016, 1, 16114; b) P. He, T. Zhang, J. Jiang, H. Zhou, J. Phys. Chem.
homemade container to avoid contact with air. Lett. 2016, 7, 1267.
Computational Details: The geometry optimization of EC, FEC, and [4] Y. Sun, N. Liu, Y. Cui, Nat. Energy 2016, 1, 16071.
DEC molecule was applied in DMol3 module in Materials Studio[50] [5] a) W. Xu, J. Wang, F. Ding, X. Chen, E. Nasybulin, Y. Zhang,
(version 5.5) of Accelrys Inc. with B3LYP[51] functional. An all-electron J.-G. Zhang, Energy Environ. Sci. 2014, 7, 513; b) X. B. Cheng,
numerical basis set with polarization functions (DNP basis set) and R. Zhang, C. Z. Zhao, F. Wei, J. G. Zhang, Q. Zhang, Adv. Sci. 2016,
a basis file of 4.4 were used during the calculation. The convergence 3, 1500213.
tolerance was set to be fine: 1.0 × 10−5 Ha, 2.0 × 10−3 Ha Å−1, and [6] M. Rosso, T. Gobron, C. Brissot, J. N. Chazalviel, S. Lascaud,
5.0 × 10−3 Å for energy, maximum force, and maximum displacement,
J. Power Sources 2001, 97, 804.
respectively. The AIMD was conducted in CASTEP[52] module with
[7] a) Z. Tu, P. Nath, Y. Lu, M. D. Tikekar, L. A. Archer, Acc. Chem. Res.
general gradient approximation[53] (GGA) and Perdew–Burke–
2015, 48, 2947; b) X. B. Cheng, T. Z. Hou, R. Zhang, H. J. Peng,
Ernzerhof[54] (PBE) exchange-correlation functional. Three models were
considered here: Li-110+10EC+5DEC, Li-110+9EC+5DEC+FEC, and LiF- C. Z. Zhao, J. Q. Huang, Q. Zhang, Adv. Mater. 2016, 28, 2888.
100+5EC+3DEC.[55] Take the model, Li-110+10EC+5DEC, as an example: [8] a) R. Zhang, X. B. Cheng, C. Z. Zhao, H. J. Peng, J. L. Shi,
Li-110 meant a seven-layer 2 × 2 supercell of lithium (110) surface with J. Q. Huang, J. Wang, F. Wei, Q. Zhang, Adv. Mater. 2016, 28,
a vacuum of 1.5 nm; 10EC+5DEC meant ten EC molecules and five DEC 2155; b) C. P. Yang, Y. X. Yin, S. F. Zhang, N. W. Li, Y. G. Guo, Nat.
molecules were put in the 1.5 nm vacuum simulating the electrolyte. Commun. 2015, 6, 8058.
A minimization calculation was conducted first in Discover module in [9] a) N. W. Li, Y. X. Yin, C. P. Yang, Y. G. Guo, Adv. Mater. 2016, 28,
Material Studio and consistent valence force field (cvff) was adopted 1853; b) Z. Liu, Y. Qi, Y. X. Lin, L. Chen, P. Lu, L. Q. Chen, J. Electro-
(compass force field was adopted in LiF-100+5EC+3DEC). Then a 4.0 ps chem. Soc. 2016, 163, A592; c) E. Peled, J. Electrochem. Soc. 1997,
AIMD was implemented in CASTEP3 with NVT ensemble. The time step 144, L208.
was set to 1.0 fs and the temperature was set to 300 K. Besides, Nose [10] K. Xu, Chem. Rev. 2004, 104, 4303.
thermostat with a Nose Q ratio of 0.5[56] was chosen. [11] a) V. Thangadurai, S. Narayanan, D. Pinzaru, Chem. Soc. Rev. 2014,
Electrochemical Measurement: Two-electrode cells configuration using 43, 4714; b) Y. Yamada, A. Yamada, J. Electrochem. Soc. 2015, 162,
standard 2025 coin-type cells was employed and assembled in an Ar-filled A2406; c) D. Aurbach, Solid State Ionics 2002, 148, 405; d) D. Wang,
glove box with O2 and H2O content below 5.0 ppm. The electrolytes W. Zhang, W. Zheng, X. Cui, T. Rojo, Q. Zhang, Adv. Sci. 2016, 3,
were prepared by adding 5% FEC additives (by volume) into 1.0 m LiPF6 1600168.
in EC/DEC (1:1 by volume), and then stirred for 5 h. The Cu foil and [12] K. Xu, Chem. Rev. 2014, 114, 11503.
NMC electrode foil were punched into 13.0 mm disks as the working
[13] a) H. Ota, K. Shima, M. Ue, J.-I. Yamaki, Electrochim. Acta 2004, 49,
electrodes. The Li | Cu cells were cycled within a voltage range of −0.5 to
565; b) J. Guo, Z. Wen, M. Wu, J. Jin, Y. Liu, Electrochem. Commun.
1.0 V at a current density of 0.1 and 0.5 mA cm−2. In each cycle, the cells
2015, 51, 59.
were first discharged for 1.0 h (except for 0.1 mA cm−2) and then charged
to the cutoff voltage of 1.0 V. The Li | NMC coin cells were monitored [14] J. Heine, P. Hilbig, X. Qi, P. Niehoff, M. Winter, P. Bieker, J. Electro-
in galvanostatic mode within a voltage range of 3.0–4.3 V versus Li+/ chem. Soc. 2015, 162, A1094.
Li using LAND multichannel battery cycler (Wuhan LAND Electronics [15] Q. C. Liu, J. J. Xu, S. Yuan, Z. W. Chang, D. Xu, Y. B. Yin, L. Li,
Co., Ltd.). The Li | NMC cells were first cycled at 0.1 C for one cycle H. X. Zhong, Y. S. Jiang, J. M. Yan, X. B. Zhang, Adv. Mater. 2015,
and then cycled at 1 C (≈180 mA g−1 based on NMC cathode materials). 27, 5241.
The EIS measurement was performed in frequency ranging from 105 to [16] a) D. Aurbach, E. Pollak, R. Elazari, G. Salitra, C. S. Kelley, J. Affinito,
10−1 Hz under amplitude of 10 mV using Solartron 1470E electrochemical J. Electrochem. Soc. 2009, 156, A694; b) S. S. Zhang, J. Power Sources
workstation (Solartron Analytical, UK). The equivalent circuit fitted for 2016, 322, 99.
the EIS spectra is from a previous publication.[34] [17] F. Ding, W. Xu, G. L. Graff, J. Zhang, M. L. Sushko, X. Chen,
Y. Shao, M. H. Engelhard, Z. Nie, J. Xiao, X. Liu, P. V. Sushko, J. Liu,
J. G. Zhang, J. Am. Chem. Soc. 2013, 135, 4450.
[18] a) C. Yan, X.-B. Cheng, C.-Z. Zhao, J.-Q. Huang, S.-T. Yang,
Supporting Information Q. Zhang, J. Power Sources 2016, 327, 212; b) W. Li, H. Yao, K. Yan,
Supporting Information is available from the Wiley Online Library or G. Zheng, Z. Liang, Y. M. Chiang, Y. Cui, Nat. Commun. 2015, 6,
from the author. 7436; c) C.-Z. Zhao, X.-B. Cheng, R. Zhang, H.-J. Peng, J.-Q. Huang,
R. Ran, Z.-H. Huang, F. Wei, Q. Zhang, Energy Storage Mater. 2016,
3, 77.
[19] Y. Lu, Z. Tu, J. Shu, L. A. Archer, J. Power Sources 2015, 279, 413.
Acknowledgements [20] Y. Lu, Z. Tu, L. A. Archer, Nat. Mater. 2014, 13, 961.
[21] C. Zu, A. Dolocan, P. Xiao, S. Stauffer, G. Henkelman,
X.Q.Z., X.B.C., and X.C. contributed equally to this work. This work was
supported by National Key Research and Development Program (No. A. Manthiram, Adv. Energy Mater. 2016, 6, 1501933.
2016YFA0202500) and Natural Scientific Foundation of China (Nos. [22] a) J. Qian, W. A. Henderson, W. Xu, P. Bhattacharya, M. Engelhard,
21422604, 21676160, and 21561130151). The authors thank Jia-Qi O. Borodin, J. G. Zhang, Nat. Commun. 2015, 6, 6362; b) L. Suo,
Huang, Rui Zhang, Chen-Zi Zhao, and Hong-Jie Peng at Tsinghua Y. S. Hu, H. Li, M. Armand, L. Chen, Nat. Commun. 2013, 4, 1481.
University for helpful discussion. [23] a) A. Manthiram, S. H. Chung, C. X. Zu, Adv. Mater. 2015, 27, 1980;
Received: November 14, 2016 b) J.-Q. Huang, Q. Zhang, F. Wei, Energy Storage Mater. 2015, 1,
Revised: December 19, 2016 127; c) R. Cao, W. Xu, D. Lv, J. Xiao, J.-G. Zhang, Adv. Energy Mater.
Published online: January 20, 2017 2015, 5, 1402273; d) S. Zhang, K. Ueno, K. Dokko, M. Watanabe,
Adv. Energy Mater. 2015, 5, 1500117; e) Q.-C. Liu, J.-J. Xu, S. Yuan,
Z.-W. Chang, D. Xu, Y.-B. Yin, L. Li, H.-X. Zhong, Y.-S. Jiang,
J.-M. Yan, X.-B. Zhang, Adv. Mater. 2015, 27, 5241.
[1] J. M. Tarascon, M. Armand, Nature 2001, 414, 359. [24] A. Manthiram, J. C. Knight, S.-T. Myung, S.-M. Oh, Y.-K. Sun,
[2] B. Dunn, H. Kamath, J. M. Tarascon, Science 2011, 334, 928. Adv. Energy Mater. 2016, 6, 1501010.

1605989 (7 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Funct. Mater. 2017, 27, 1605989
www.advancedsciencenews.com www.afm-journal.de

[25] D. Lu, Y. Shao, T. Lozano, W. D. Bennett, G. L. Graff, B. Polzin, [38] a) S. Komaba, T. Ishikawa, N. Yabuuchi, W. Murata, A. Ito,

full paper
J. Zhang, M. H. Engelhard, N. T. Saenz, W. A. Henderson, Y. Ohsawa, ACS Appl. Mater. Interfaces 2011, 3, 4165; b) M. Kohl,
P. Bhattacharya, J. Liu, J. Xiao, Adv. Energy Mater. 2015, 5, 1400993. F. Borrmann, H. Althues, S. Kaskel, Adv. Energy Mater. 2016, 6,
[26] J. Zheng, P. Yan, D. Mei, M. H. Engelhard, S. S. Cartmell, 1502185.
B. J. Polzin, C. Wang, J.-G. Zhang, W. Xu, Adv. Energy Mater. 2016, [39] N. Wang, Y. Gao, Y.-X. Wang, K. Liu, W. Lai, Y. Hu, Y. Zhao,
6, 1502151. S.-L. Chou, L. Jiang, Adv. Sci. 2016, 3, 1600013.
[27] A. M. Haregewoin, A. S. Wotango, B.-J. Hwang, Energy Environ. Sci. [40] W. Li, S.-L. Chou, J.-Z. Wang, J. H. Kim, H.-K. Liu, S.-X. Dou, Adv.
2016, 9, 1955. Mater. 2014, 26, 4037.
[28] D. Y. Wang, N. N. Sinha, J. C. Burns, C. P. Aiken, R. Petibon, [41] a) Z. Li, J. Huang, B. Y. Liaw, V. Metzler, J. Zhang, J. Power Sources
J. R. Dahn, J. Electrochem. Soc. 2014, 161, A467. 2014, 254, 168; b) Z. Zeng, X. Jiang, R. Li, D. Yuan, X. Ai, H. Yang,
[29] a) V. Etacheri, O. Haik, Y. Goffer, G. A. Roberts, I. C. Stefan, Y. Cao, Adv. Sci. 2016, 3, 1600066.
R. Fasching, D. Aurbach, Langmuir 2012, 28, 965; b) C. Xu, [42] S. Hong, M.-H. Choo, Y. H. Kwon, J. Y. Kim, S.-W. Song, Adv. Mater.
F. Lindgren, B. Philippe, M. Gorgoi, F. Björefors, K. Edström, Interfaces 2016, 3, 1600172.
T. Gustafsson, Chem. Mater. 2015, 27, 2591. [43] I. A. Shkrob, J. F. Wishart, D. P. Abraham, J. Phys. Chem. C 2015,
[30] E. Markevich, G. Salitra, K. Fridman, R. Sharabi, G. Gershinsky, 119, 14954.
A. Garsuch, G. Semrau, M. A. Schmidt, D. Aurbach, Langmuir 2014, [44] Z. Yang, A. A. Gewirth, L. Trahey, ACS Appl. Mater. Interfaces 2015,
30, 7414. 7, 6557.
[31] Y. Li, F. Lian, L. L. Ma, C. L. Liu, L. Yang, X. M. Sun, K. C. Chou, [45] J. Wang, C. Luo, T. Gao, A. Langrock, A. C. Mignerey, C. Wang,
Electrochim. Acta 2015, 168, 261. Small 2015, 11, 473.
[32] J. Pan, Y.-T. Cheng, Y. Qi, Phys. Rev. B 2015, 91, 134116. [46] L. Suo, O. Borodin, T. Gao, M. Olguin, J. Ho, X. Fan, C. Luo,
[33] a) L. Chen, H. W. Zhang, L. Y. Liang, Z. Liu, Y. Qi, P. Lu, J. Chen, C. Wang, K. Xu, Science 2015, 350, 938.
L.-Q. Chen, J. Power Sources 2015, 300, 376; b) H. Xiang, P. Shi, [47] K. Kanamura, S. Shiraishi, Z. i. Takehara, J. Electrochem. Soc. 1996,
P. Bhattacharya, X. Chen, D. Mei, M. E. Bowden, J. Zheng, 143, 2187.
J.-G. Zhang, W. Xu, J. Power Sources 2016, 318, 170. [48] a) S. S. Zhang, J. Power Sources 2006, 162, 1379; b) F. Ding, W. Xu,
[34] X. B. Cheng, H. J. Peng, J. Q. Huang, R. Zhang, C. Z. Zhao, X. Chen, J. Zhang, M. H. Engelhard, Y. Zhang, B. R. Johnson,
Q. Zhang, ACS Nano 2015, 9, 6373. J. V. Crum, T. A. Blake, X. Liu, J. G. Zhang, J. Electrochem. Soc. 2013,
[35] a) G. Deniz, L.-W. Kendra, S. Ravishankar, A. S. Kathleen, T. A. Arias, 160, A1894.
Model. Simul. Mater. Sci. 2013, 21, 074005; b) A. L. Michan, [49] S. Choudhury, L. A. Archer, Adv. Electron. Mater. 2016, 2, 1500246.
B. S. Parimalam, M. Leskes, R. N. Kerber, T. Yoon, C. P. Grey, [50] a) B. Delley, J. Chem. Phys. 1990, 92, 508; b) B. Delley, J. Chem. Phys.
B. L. Lucht, Chem. Mater. 2016, 28, 8149. 2000, 113, 7756.
[36] M. Y. Nie, J. Demeaux, B. T. Young, D. R. Heskett, Y. J. Chen, [51] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
A. Bose, J. C. Woicik, B. L. Lucht, J. Electrochem. Soc. 2015, 162, [52] S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. I. J. Probert,
A7008. K. Refson, M. C. Payne, Z. Kristallogr. - Cryst. Mater. 2005, 220, 567.
[37] a) M. Sina, J. Alvarado, H. Shobukawa, C. Alexander, V. Manichev, [53] Y. Zhou, C. Zhou, Q. Li, C. Yan, B. Han, K. Xia, Q. Gao, J. Wu, Adv.
L. Feldman, T. Gustafsson, K. J. Stevenson, Y. S. Meng, Adv. Mater. Mater. 2015, 27, 3774.
Interfaces 2016, 3, 1600438; b) X. Chen, X. Li, D. Mei, J. Feng, [54] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
M. Y. Hu, J. Hu, M. Engelhard, J. Zheng, W. Xu, J. Xiao, J. Liu, [55] K. Gaedt, H. D. Höltje, J. Comput. Chem. 1998, 19, 935.
J.-G. Zhang, ChemSusChem 2014, 7, 549. [56] S. I. Nosé, Mol. Phys. 2002, 100, 191.

Adv. Funct. Mater. 2017, 27, 1605989 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (8 of 8) 1605989