Research ArticleFormation of Stable Interphase of Polymer-in-Salt Electrolyte inAll-Solid-State Lithium Batteries

Hongcai Gao , Nicholas S. Grundish , Yongjie Zhao, Aijun Zhou ,and John B. Goodenough

Texas Materials Institute, The University of Texas at Austin, Austin Texas 78712, USA

Correspondence should be addressed to Hongcai Gao; gaohc@utexas.edu

Received 18 June 2020; Accepted 26 July 2020; Published 7 January 2021

Copyright © 2021 Hongcai Gao et al. Exclusive Licensee Beijing Institute of Technology Press. Distributed under a CreativeCommons Attribution License (CC BY 4.0).

The integration of solid-polymer electrolytes into all-solid-state lithium batteries is highly desirable to overcome the limitations ofcurrent battery configurations that have a low energy density and severe safety concerns. Polyacrylonitrile is an appealing matrix forsolid-polymer electrolytes; however, the practical utilization of such polymer electrolytes in all-solid-state cells is impeded byinferior ionic conductivity and instability against a lithium-metal anode. In this work, we show that a polymer-in-salt electrolytebased on polyacrylonitrile with a lithium salt as the major component exhibits a wide electrochemically stable window, a highionic conductivity, and an increased lithium-ion transference number. The growth of dendrites from the lithium-metal anodewas suppressed effectively by the polymer-in-salt electrolyte to increase the safety features of the batteries. In addition, we foundthat a stable interphase was formed between the lithium-metal anode and the polymer-in-salt electrolyte to restrain theuncontrolled parasitic reactions, and we demonstrated an all-solid-state battery configuration with a LiFePO4 cathode and thepolymer-in-salt electrolyte, which exhibited a superior cycling stability and rate capability.

1. Introduction

The ever-increasing demand for long-lasting, portable elec-tronic devices and long-range electric vehicles has stimulatedextensive efforts to increase the safety and energy density ofrechargeable batteries [1–3]. Methods to enable a safe lithium-metal anode are being explored actively to replace the graphiteanode in conventional lithium-ion batteries [4–9]. However,lithium-metal anodes with organic-liquid electrolytes areimpeded by the formation of an unstable solid-electrolyteinterphase (SEI), the extensive growth of lithium dendrites,and a lowplating/stripping efficiency during charge/dischargecycles [10–13]. The integration of solid-polymer electrolytesinto all-solid-state cell configurations represents an intriguingstrategy to suppress safety hazards associated with highlyflammable organic electrolytes; the affordable cost andmechanical flexibility also allow solid-polymer electrolytesto be processed into desirable sizes and shapes with a broadrange of fabrication techniques [14–17].

Departing from the numerous investigations on solid-polymer electrolytes with poly(ethylene oxide) and its deriva-tives, recent attention has shifted focus to polymers with

nitrile groups, such as polyacrylonitrile (PAN) [18]. Theappealing properties of PAN that make it a promising matrixfor solid-polymer electrolytes include a high dielectricconstant, high oxidation potential, and strong coordinationability with lithium salts [19–21]. However, the ionic conduc-tivity of PAN solid-polymer electrolytes is usually too low tomeet practical requirements even at elevated temperatures.For conventional salt-in-polymer electrolytes, the cation-coordinating macromolecule is the primary component. Inthese electrolytes, migration of mobile ions occurs in amor-phous regions and is assisted by segmental motions of themacromolecule host [22–24]. High ionic conductivities ofsuch electrolytes can be obtained only at temperatures abovetheir glass-transition temperature [25–27].

The related system of polymer-in-salt electrolyte withlithium salt as the major component exhibits much higherionic conductivities than salt-in-polymer scheme by creatinga rubbery version of a glass-like polymer electrolyte [28, 29].The polymer-in-salt electrolytes, therefore, provide an idealplatform to combine the desirable mechanical properties ofrubbery polymers with the fast-ionic conduction of glassyelectrolytes. The extremely high salt content in the polymer-

AAASEnergy Material AdvancesVolume 2021, Article ID 1932952, 10 pageshttps://doi.org/10.34133/2021/1932952

https://orcid.org/0000-0002-3671-8765https://orcid.org/0000-0003-1821-6734https://orcid.org/0000-0003-4142-8030https://orcid.org/0000-0001-9350-3034https://doi.org/10.34133/2021/1932952

in-salt electrolytes promotes the formation of an ionicconducting network consisting of aggregated cation/anionclusters that are interconnected to provide percolation path-ways for fast ion migration [30–32]. As a promising strategyto increase the ionic conductivity with respect to the salt-in-polymer scheme, the design of polymer-in-salt electrolyteshas been actively explored recently, but very limited attentionhas been focused on the PAN system [33–35].

The other critical challenge towards the application ofPAN solid-polymer electrolytes is the uncontrolled passiv-ation reactions against lithium-metal anodes at the interfacethat lead to the formation of poor quality SEI layers, andconsequently, inferior electrochemical performance of theall-solid-state batteries, including unsatisfactory coulombicefficiency and rapid capacity fading [36–39]. Nevertheless,the interfacial chemistry of the PAN polymer-in-salt electro-lyte against a lithium-metal anode remains an important, butunexplored topic. We herein report a PAN polymer-in-saltelectrolyte that facilitates the generation of a stable interphaselayer against a lithium-metal anode owing to the highconcentration of lithium salt. This interphase layer preventssuccessive parasitic reactions after formation, enablinglong-term, dendrite-free plating/stripping of a lithium-metal anode. By integrating the polymer-in-salt electrolytewith a lithium-metal anode and a LiFePO4 cathode, an all-solid-state lithium battery possessing stable cycling perfor-mance and high rate capability is demonstrated.

2. Results and Discussion

The polymer-in-salt electrolyte membrane was prepared witha solution-castingmethod. Amixture of PAN, lithium bis(tri-fluoromethanesulfonyl)imide (LiTFSI), and SiO2 nanoparti-cles in dimethylformamide was cast on a porous cellulosemembrane; the molar ratio of the acrylonitrile monomer inPAN and the lithium salt of LiTFSI was kept at 1 : 1, which sta-bilized the prepared polymer electrolyte in the polymer-in-salt state [40, 41]. The SiO2 nanoparticles accounted for 5%of the total weight of the solid-polymer electrolyte. The thick-ness of the resulting polymer-in-salt electrolyte membraneswas 100-120μm, having a relatively smooth surface whenanalyzed with scanning electron microscopy (SEM), and X-ray diffraction verified that the polymer-in-salt electrolytewas amorphous (Figure S1). With the resulting polymer-in-salt electrolyte membrane, we fabricated all-solid-statelithium cells with LiFePO4 as the cathodes and lithium metalas the anodes (Figure 1(a)), and we found that a stableinterphase can be formed between the electrolyte and thelithium-metal anode during repeated charge/dischargecycles to enable dendrite-free plating/striping of lithiumacross the polymer-in-salt electrolyte (Figure 1(c)).

Symmetric cells with stainless-steel blocking electrodeswere used for electrochemical impedance spectroscopy (EIS)measurements to obtain the ionic conductivities of the poly-mer electrolytes. The influence of lithium salt concentrationon ionic conductivity of the polymer electrolytes was firstinvestigated. The ionic conductivity of the PAN polymer elec-trolytes steadily increased at 60°C with increasing the lithiumsalt concentration (Figure S2). A polymer-in-salt electrolyte

with 84% wt.% LiTFSI and 16wt.% PAN exhibited an ionicconductivity that was about 5 orders of magnitude higherthan that of a salt-in-polymer electrolyte with 21% wt.%LiTFSI and 79wt.% PAN. Compared to the conventionalpolymer electrolyte with a low ionic conductivity arisingfrom the segmental motion of polymer chains, the increasedionic conductivity in the polymer-in-salt electrolyte can beattributed to the motion of Li+ ions through the ion-conducting network formed by aggregated ionic clustersarising from a high lithium salt concentration (Figure 1(b))[42, 43]. As an ionic conductivity in the range of 10-4 S cm-1

is needed for an electrolyte to power a battery [44], the salt-in-polymer electrolyte has an ionic conductivity too low toenable the operation of an all-solid-state cell. The polymer-in-salt electrolyte with 84% wt.% LiTFSI and 16wt.% PANwas, therefore, used in the following experiments. Thetemperature dependence of ionic conductivity for thepolymer-in-salt electrolyte was further investigated, whichfollows an Arrhenius behavior [45], exhibiting an ionicconductivity of 8:5 × 10−5 S cm-1 at 30°C and progressivelyincreasing to 1:8 × 10−4 S cm-1 at 60°C (Figure 2(a)).According to the Arrhenius equation, the activation energywas calculated to be 23.1kJmol-1, which is lower than that ofsolid-polymer electrolytes with poly(ethylene oxide)(~38kJmol-1) [46]. This lower activation energy indicates theformation of conduction network consisting of aggregatedionic clusters which can promote the transport of Li+-ions inthe polymer-in-salt electrolyte, whereas Li+-ion transport inconventional polymer electrolytes is assisted by the segmentalmotion of polymer chains in the amorphous regions [47, 48].

The electrochemical window of the polymer-in-salt elec-trolyte was evaluated with cyclic voltammetry and linearsweep voltammetry measurements of cells consisting of astainless-steel working electrode and a lithium-metal refer-ence electrode. The plating/stripping behavior of lithiummetal can be seen in the cyclic voltammogram (Figure 2(b)).The strong reduction peak at -0.47V (vs. Li+/Li) correspondsto lithium plating on the stainless-steel working electrode andthe oxidation peak centered at 0.32V (vs. Li+/Li) correspondsto lithium stripping from the working electrode. The compa-rable peak intensities of the reduction and oxidation currentssuggests that lithium can be plated/stripped reversibly acrossthe polymer-in-salt electrolyte onto/from the workingelectrode [49]. As indicated in the linear sweep voltammetryprofile, the polymer-in-salt electrolyte possesses an electro-chemical window stable up to 4.8V (vs. Li+/Li), only showingan oxidation current upon reaching that voltage. Thus, theelectrochemical stability window of the polymer-in-salt elec-trolyte is wide enough tomeet the requirements ofmost primecathode candidates for lithium-based batteries in conjunctionwith a lithium-metal anode [50, 51]. This oxidation stabilitycan be attributed to the high oxidation potential of PAN andits strong complexation interactions with the lithium saltswithin the polymer-in-salt electrolyte.

Most solid-polymer electrolytes used in lithium batteriesexhibit a low Li+-ion transference number (typically smallerthan 0.3) [52]. When polarizing a cell with a binary electro-lyte, the Sand’s time, which indicates the ionic concentrationof the electrolyte dropping to zero on the electrode surface to

2 Energy Material Advances

induce parasitic reactions of the electrolyte and nucleation oflithium dendrites by the formation of a local space charge, isinversely proportional to the anion transference numbersquared [53, 54]. The depletion of Li+ ions on the electrodesurface because of a low Li+-ion transference number willinduce decomposition reactions of the electrolyte rather thanthe plating/stripping reactions of the lithium-metal anode.Therefore, it is a promising strategy to reduce the decomposi-tion of electrolyte and suppress the nucleation of lithiumdendrites by using an electrolyte with a high Li+-ion transfer-ence number [55, 56]. The potentiostatic polarizationmethodwas used to obtain the Li+-ion transference number of thepolymer-in-salt electrolyte [57], with EIS measurements per-formed immediately before and after a small polarization of10mV was applied to a lithium-lithium symmetric cell. Theresults of this experiment are presented in Figure 2(c), andthe Li+-ion transference number of the polymer-in-saltelectrolyte was calculated to be 0.47, which is higher than thatof the typical solid-polymer electrolytes [58]. This highLi+-ion transference number may be attributed to the for-mation of aggregated ionic clusters within the polymer-in-salt electrolyte, significantly restricting the movement ofthe large [TFSI]- anions and promoting only Li+-ion trans-fer owing to its smaller size [59, 60]. With this high Li+-ion transference number, the polymer-in-salt electrolyte

is expected to reduce the polarization and increase the rateperformance in all-solid-state lithium batteries [61].

Interfacial stability of the polymer-in-salt electrolyteagainst the lithium-metal electrode was investigated withEIS of a lithium-lithium symmetric cell at different time inter-vals. The intercepts at the real-axis in the EIS spectra at highfrequencies represent the bulk resistances of the polymer-in-salt electrolyte, which remained constant throughout 30 daysof storage (Figure S3). The semicircles at low frequenciesrepresent the interfacial resistances between the polymer-in-salt electrolyte and the lithium-metal electrodes [62].Initially, the initial interfacial resistance was 257Ω cm2 for afresh lithium-lithium symmetric cell. After 10 days, itincreased to 293Ω cm2 before asymptotically stabilizing inthe following days (Figure 2(d)). The stabilization ofinterfacial resistance after 10 days of storage suggests theformation of a stable interphase layer on lithium-metalsurface in contact with the polymer-in-salt electrolyte.

Galvanostatic cycling of a lithium-lithium symmetric cellat varying current densities was performed to further probethe interfacial stability of the polymer-in-salt electrolyteagainst lithium-metal electrodes upon repeated plating/strip-ping. A considerable amount of lithium was transported inthe symmetric cell during each cycle to increase the possibil-ity of an internal short-circuit caused by the lithium dendrite

Curr

ent c

olle

ctor

Poly

mer

-in-s

alt e

lect

roly

te

Lith

ium

-met

al an

ode

Cathode: LiFePO4

(a)

Polymer-in-salt electrolyteSalt-in-polymer electrolyte

Ionic cluster

Mobile ion

Polymer matrix

(b)

LiF Li2O Li3N

CFx

LiF

LiF

LiFLi2O

Li2O

Li2O

Li2O

CFx

CFx

CFx

Li3N

Li3NLi3N

Li3N COx

COx

COx

LiF Li2O Li3N

CFx

LiF

LiF

LiFLi2O

Li2O

Li2O

Li2O

CFx

CFx

CFx

Li3N

Li3NLi3N

Li3N COx

COx

COx

Lithium-metal anode Lithium-metal anode

Lithium plating Lithium stripping

(c)

Figure 1: Schematic illustration of the all-solid-state lithium battery configuration and the polymer electrolytes. (a) Schematic illustration ofthe all-solid-state cell with the polymer-in-salt electrolyte sandwiched between a lithium-metal anode and a LiFePO4 cathode. (b) Schematicillustration of lithium-ion transports in a salt-in-polymer electrolyte and a polymer-in-salt electrolyte. (c) Schematic illustration of the lithiumplating and stripping across the interphase formed between the polymer-in-salt electrolyte and the lithium-metal anode.

3Energy Material Advances

proliferation. At a current density of 0.1mAcm-2, thesymmetric cell with the polymer-in-salt electrolyte exhibitsstable lithium plating/stripping behavior for 200 h without ashort circuit (Figures 3(a) and 3(b)). The slight decrease inpolarization voltage in the initial cycles indicates the in situformation of a Li+-conducting electrode/electrolyte inter-phase that allows a stable cycling voltage for the duration ofthe experiments [63–65]. When the current density increasedto 0.2mAcm-2, the symmetric cell retains stable lithiumplating/stripping behavior without a short circuit for 200 h(Figures 3(c) and 3(d)). These results demonstrate the effec-tiveness of the polymer-in-salt electrolyte in preventing lith-ium dendrite growth, thus providing a safe pathway forimplementation of lithium-metal anodes in all-solid-statecell configurations.

To evaluate full-cell performance of the polymer-in-saltelectrolyte, all-solid-state cell configurations with lithium-metal as the anodes and LiFePO4 as the cathodes werefabricated. Electrochemical performance of these cells was

evaluated at 60°C with galvanostatic charge/dischargecycling. The charge/discharge profiles of the resulting cellexhibit the typical voltage profile of the LiFePO4 cathode witha plateau centered at 3.4V (vs. Li+/Li) [66]. The high revers-ibility of the all-solid-state cell is indicated by the equivalenceof the charge and discharge capacities at each current rate(Figure 4(a)). The specific discharge capacity of the cell withrespect to the cathode was 142mAhg-1 at 0.05C with a cou-lombic efficiency of 98.6%. A coulombic efficiency below100% indicates the formation of a passivation layer on theelectrode surface in the initial cycles. This interfacial layercould prevent further parasitic reactions during the prolongedcycling of the cell but also adds an additional element to thecell impedance, causing a relatively lower coulombic efficiencyupon extended cycling than desired [67, 68]. After increasingthe rate to 1C, the all-solid-state cell possessed a specific dis-charge capacity of 83mAhg-1, keeping 58.5% of the capacityfrom 0.05C (Figure 4(b)). The specific discharge capacity ofthe all-solid-state cell was 56mAhg-1 after increasing the rate

2.8 3.0 3.2 3.4

–4.2

–4.0

–3.8

–3.6

–3.4lo

g (𝜎

/S cm

–1)

1000/T (K–1)

(a)

–1 0 1 2 3 4 5

–0.6

–0.4

–0.2

0.0

0.2

0.4

0.6

Curr

ent (

mA

cm–2

)

Potential (V vs. Li+/Li)

(b)

0 500 1000 1500 200010

11

12

13

14

15

0 100 200 300 4000

100

200

300

400

Curr

ent (𝜇

A cm

–2)

Time (s)

Before polarizationAfter polarization

(c)

0 5 10 15 20 25 30200

250

300

350

Resis

tanc

e (𝛺

cm2 )

Time (day)

(d)

Figure 2: Electrochemical characterization of the polymer-in-salt electrolyte. (a) The temperature dependence of ionic conductivity of thepolymer-in-salt electrolyte measured in a symmetric cell with stainless-steel blocking electrodes. (b) Cyclic voltammetry and linear sweepvoltammetry profiles of cells at 60°C consisting of the polymer-in-salt electrolyte, a stainless-steel working electrode, and a lithium-metalreference electrode. (c) Polarization profile of a lithium-lithium symmetric cell at 60°C with the polymer-in-salt electrolyte; inset shows theimpedance spectra of the cell before and after the polarization. (d) Variation of interfacial resistance of a lithium-lithium symmetric cell at60°C with the polymer-in-salt electrolyte as a function of time.

4 Energy Material Advances

to 2C, demonstrating a good rate capability compared to all-solid-state lithium cells with other solid-polymer electrolytes[69, 70]. Moreover, the all-solid-state battery showed stablelong-term cycling at 0.2C rate with 95.1% first-cycle capacityretention after 100 charge/discharge cycles (Figure 4(c)).

The integrity of the electrode/electrolyte interface uponrepeated cycling was investigated with EIS before and afterundergoing 100 charge/discharge cycles (Figure 4(d)). Anequivalent circuit was used to simulate the EIS spectra(Figure S4) [71]. Only a slight increase of interfacialresistance was observed after 100 charge/discharge cycles(Figure S5). This result corroborates the formation of amechanically robust and electrochemically stable interphasebetween the polymer-in-salt electrolyte and the electrodesthat does not deteriorate upon repeated cycling. The surfacemorphology of the lithium-metal anode after 100charge/discharge cycles in the all-solid-state cell wasinvestigated with SEM (Figure S6), showing a relativelysmooth surface without the formation of dendrites. Theformation of a stable interphase between the polymer-in-salt electrolyte and the lithium-metal electrode is proven toreduce the resistance of Li+-ion transfer across the interface

and enable a uniform distribution of Li+-ion flux to inhibitthe generation of a localized electric field and suppress thenucleation of lithium dendrites [72, 73].

The chemical composition of lithium-metal anode sur-face in contact with the polymer-in-salt electrolyte duringcycling was investigated after 100 charge/discharge cycleswith X-ray photoelectron spectroscopy (XPS). The all-solid-state cell was disassembled in an argon-filled glove box, andthe lithium-metal electrode was transferred to the XPS anal-ysis chamber through a sample-transfer device withoutexposing the sample to air [74]. The C 1s spectrum(Figure 5(a)) of the cycled lithium-metal electrode suggeststhe presence of C-F (288.9 eV), C-O (286.3 eV), and C-C(285.1 eV) species, which are generally derived from the deg-radation of PAN and the LiTFSI salt [75]. The Li 1s spectrum(Figure 5(b)) indicates the presence of LiF (55.4 eV) and Li2O(54.1 eV) species [76]. In accordance with the Li 1s and C 1sspectra, the F 1s spectrum (Figure 5(c)) confirms the pres-ence of LiF (684.9 eV) and C-F (688.2 eV) species [77]. TheN 1s spectrum (Figure 5(d)) suggests the formation of Li-N(398.7 eV) and N-S (396.8 eV) bonds, which are related tothe decomposition products of LiTFSI salt in the forms of

0 50 100 150 200

–150

–100

–50

0

50

100

150

Vol

tage

(mV

)

Time (h)

(a)

190 192 194 196 198 200

–150

–100

–50

0

50

100

150

Vol

tage

(mV

)

Time (h)

(b)

0 50 100 150 200

–300

–200

–100

0

100

200

300

Vol

tage

(mV

)

Time (h)

(c)

190 192 194 196 198 200

–300

–200

–100

0

100

200

300

Vol

tage

(mV

)

Time (h)

(d)

Figure 3: Electrochemical characterization of the lithium-lithium symmetric cell at 60°C with the polymer-in-salt electrolyte at differentcurrent densities. (a) Voltage profile of the symmetric cell at 0.1mA cm-1. (b) An enlarged portion of the voltage profile of the symmetriccell at 0.1mA cm-1. (c) Voltage profile of the symmetric cell at 0.2mA cm-1. (d) An enlarged portion of the voltage profile of thesymmetric cell at 0.2mA cm-1.

5Energy Material Advances

Li3N and Li2NSO2CF3 [78, 79]. The formation of LiF andLi3N in the interphase may play an important role inmodulating the plating/stripping behavior of lithium, tak-ing the advantages of the high surface energy and/or lowactivation barrier for Li+ ions to migrate across the inter-phase, enabling dendrite-free operation of the electrode[80–83]. In addition to preventing lithium dendrites, thisinterphase layer also inhibits successive deterioration ofthe polymer-in-salt electrolyte by the lithium-metal anodeonce it has formed. As a result, the all-solid-state batteryexhibits stable cycling even after an extended number ofcharge/discharge cycles.

3. Conclusions

In order to overcome the challenges of conventional salt-in-polymer electrolytes based on PAN that have low ionic

conductivity and experience severe parasitic reactions withlithium-metal electrode at the interface, we have investi-gated polymer-in-salt electrolytes consisting of PAN forall-solid-state battery configurations. The polymer-in-saltelectrolyte with the lithium salt of LiTFSI as the majorcomponent exhibits a high ionic conductivity and anincreased lithium-ion transference number compared totraditional solid-polymer electrolytes. Additionally, thepolymer-in-salt electrolyte suppresses effectively nucleationand proliferation of lithium dendrites through the forma-tion of a stable interphase layer. By integrating thepolymer-in-salt electrolyte with a lithium-metal anodeand LiFePO4 cathode, the resulting all-solid-state cell oper-ating at 60°C exhibited good rate capability up to 2C rateand excellent capacity retention after an extended numberof galvanostatic charge/discharge cycles at 0.2C rate. Theseperformance metrics make this PAN-based polymer-in-salt

0 50 100 150

2.5

3.0

3.5

4.0

2C 1C 0.5C

0.2C

0.1C

Capacity (mAh g–1)

Volta

ge (V

vs.

Li+ /

Li)

0.05C

(a)

0 10 20 30 40 50 60 700

50

100

150 0.05C

2C

1C

0.5C0.2C

0.1C

Cycle number

Disc

harg

e cap

acity

(mA

h g–

1 )

0.05C

0

20

40

60

80

100

Coul

ombi

c effi

cien

cy (%

)

Discharge capacityCoulombic efficiency

(b)

0

50

100

150

Cycle number

Disc

harg

e cap

acity

(mA

h g–

1 )

Coul

ombi

c effi

cien

cy (%

)

0 20 40 60 80 1000

20

40

60

80

100

Discharge capacityCoulombic efficiency

(c)

0 200 400 6000

200

400

600

Before cyclingAfter cycling

(d)

Figure 4: Electrochemical performance of the all-solid-state lithium battery at 60°C with the polymer-in-salt electrolyte, the lithium-metalanode, and the LiFePO4 cathode. (a) Charge/discharge profiles of the cell at different current rates. (b) Specific discharge capacity andcoulombic efficiency of the cell at different current rates. (c) Charge/discharge cycling stability of the cell at 0.2 C rate. (d) Electrochemicalimpedance spectra of the cell before and after the charge/discharge cycling test.

6 Energy Material Advances

electrolyte a promising strategy to enable the all-solid-stateconfiguration of lithium-metal batteries of increased safetyand energy density.

4. Materials and Methods

Experimental details including the preparation and charac-terization of the polymer electrolytes and all-solid-statelithium batteries can be found in Supplementary Materials.

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this article.

Authors’ Contributions

H.G. and J.B.G. conceived this project. H.G. designed exper-imental procedures. H.G., N.S.G., Y.Z., and A.Z. contributedto the materials preparation, experimental measurements,and data analysis. H.G., N.S.G., and J.B.G. cowrote themanuscript. J.B.G. supervised the project.

Acknowledgments

The preparation and characterization of the polymer electro-lytes and the all-solid-state lithium batteries was supportedby the United States Department of Energy, Office of BasicEnergy Sciences (DE-SC0005397). J.B.G. also acknowledgessupport from the Robert A. Welch Foundation (F-1066).

Supplementary Materials

Materials and methods. Figure S1: characterization of thepolymer-in-salt electrolyte membrane. Figure S2: ionic con-ductivities of the PAN-based polymer electrolytes as a func-tion of the concentration of lithium salt LiTFSI. Figure S3:electrochemical impedance spectra of the lithium-lithiumsymmetric cell with the polymer-in-salt electrolyte at differ-ent time of storage. Figure S4: equivalent electric circuit usedto fit the electrochemical impedance spectra of the all-solid-state battery. Figure S5: the interface resistance and charge-transfer resistance of the all-solid-state battery before andafter the charge/discharge cycling test. Figure S6: SEM imageof the surface morphology of lithium-metal anode from theall-solid-state cell after the charge/discharge cycling test.(Supplementary Materials)

294 292 290 288 286 284 282

C-F

C-O

Inte

nsity

(a.u

.)C-C

Binding energy (eV)

(a)

60 58 56 54 52 50

Li2O

LiF

Inte

nsity

(a.u

.)

Binding energy (eV)

(b)

694 692 690 688 686 684 682 680

LiF

C-F

Inte

nsity

(a.u

.)

Binding energy (eV)

(c)

404 402 400 398 396 394

Li-N

N-S

Inte

nsity

(a.u

.)

Binding energy (eV)

(d)

Figure 5: Characterization of the surface of the lithium-metal anode. XPS spectra of (a) C 1s, (b) Li 1s, (c) F 1s, and (d) N 1s on the surface ofthe lithium-metal electrode from the all-solid-state battery with the polymer-in-salt electrolyte after the charge/discharge cycling test.

7Energy Material Advances

http://downloads.spj.sciencemag.org/energymatadv/2021/1932952.f1.docx

References

[1] J. B. Goodenough, “Electrochemical energy storage in a sus-tainable modern society,” Energy & Environmental Science,vol. 7, no. 1, pp. 14–18, 2014.

[2] D. P. Dubal, O. Ayyad, V. Ruiz, and P. Gomez-Romero,“Hybrid energy storage: the merging of battery and supercapa-citor chemistries,” Chemical Society Reviews, vol. 44, no. 7,pp. 1777–1790, 2015.

[3] D. Kundu, E. Talaie, V. Duffort, and L. F. Nazar, “The emerg-ing chemistry of sodium ion batteries for electrochemicalenergy storage,” Angewandte Chemie International Edition,vol. 54, no. 11, pp. 3431–3448, 2015.

[4] W. Xu, J. Wang, F. Ding et al., “Lithium metal anodes forrechargeable batteries,” Energy & Environmental Science,vol. 7, no. 2, pp. 513–537, 2014.

[5] A. Mauger, M. Armand, C. M. Julien, and K. Zaghib,“Challenges and issues facing lithium metal for solid-staterechargeable batteries,” Journal of Power Sources, vol. 353,pp. 333–342, 2017.

[6] Y. Zhou, X. Wang, H. Zhu et al., “Ternary lithium-salt organicionic plastic crystal polymer composite electrolytes for highvoltage, all-solid-state batteries,” Energy Storage Materials,vol. 15, pp. 407–414, 2018.

[7] X. B. Cheng, C. Z. Zhao, Y. X. Yao, H. Liu, and Q. Zhang,“Recent advances in energy chemistry between solid-state elec-trolyte and safe lithium-metal anodes,” Chem, vol. 5, no. 1,pp. 74–96, 2019.

[8] J. B. Goodenough and H. Gao, “A perspective on the Li-ionbattery,” SCIENCE CHINA Chemistry, vol. 62, no. 12,pp. 1555-1556, 2019.

[9] E. Quartarone and P. Mustarelli, “Review- emerging trends inthe design of electrolytes for lithium and post-lithium batte-ries,” Journal of the Electrochemical Society, vol. 167, no. 5, arti-cle 050508, 2020.

[10] A. Varzi, R. Raccichini, S. Passerini, and B. Scrosati,“Challenges and prospects of the role of solid electrolytes inthe revitalization of lithium metal batteries,” Journal ofMaterials Chemistry A, vol. 4, no. 44, pp. 17251–17259, 2016.

[11] E. Peled and S. Menkin, “Review-SEI: past, present andfuture,” Journal of the Electrochemical Society, vol. 164, no. 7,pp. A1703–A1719, 2017.

[12] F. Wu, Y. X. Yuan, X. B. Cheng et al., “Perspectives forrestraining harsh lithium dendrite growth: towards robust lith-ium metal anodes,” Energy Storage Materials, vol. 15, pp. 148–170, 2018.

[13] Y. Zhang, T. T. Zuo, J. Popovic et al., “Towards better Li metalanodes: challenges and strategies,” Materials Today, vol. 33,pp. 56–74, 2020.

[14] V. Di Noto, S. Lavina, G. A. Giffin, E. Negro, and B. Scrosati,“Polymer electrolytes: present, past and future,” ElectrochimicaActa, vol. 57, pp. 4–13, 2011.

[15] H. Gao, L. Xue, S. Xin, and J. B. Goodenough, “A high-energy-density potassium battery with a polymer-gel electrolyte and apolyaniline cathode,” Angewandte Chemie InternationalEdition, vol. 57, no. 19, pp. 5449–5453, 2018.

[16] J. Y. Liang, X. X. Zeng, X. D. Zhang et al., “Engineering janusinterfaces of ceramic electrolyte via distinct functional poly-mers for stable high-voltage Li-metal batteries,” Journal ofthe American Chemical Society, vol. 141, no. 23, pp. 9165–9169, 2019.

[17] T. Jiang, P. He, G. Wang, Y. Shen, C. W. Nan, and L. Z. Fan,“Solvent-free synthesis of thin, flexible, nonflammablegarnet-based composite solid electrolyte for all-solid-statelithium batteries,” Advanced Energy Materials, vol. 10,no. 12, article 1903376, 2020.

[18] P. Hu, J. Chai, Y. Duan, Z. Liu, G. Cui, and L. Chen, “Progressin nitrile-based polymer electrolytes for high performance lith-ium batteries,” Journal of Materials Chemistry A, vol. 4, no. 26,pp. 10070–10083, 2016.

[19] K. M. Abraham, H. S. Choe, and D. M. Pasquariello, “Polyac-rylonitrile electrolyte-based Li ion batteries,” ElectrochimicaActa, vol. 43, no. 16-17, pp. 2399–2412, 1998.

[20] P. L. Kuo, C. A. Wu, C. Y. Lu, C. H. Tsao, C. H. Hsu, and S. S.Hou, “High performance of transferring lithium ion forpolyacrylonitrile-interpenetrating crosslinked polyoxyethy-lene network as gel polymer electrolyte,” ACS Applied Mate-rials & Interfaces, vol. 6, no. 5, pp. 3156–3162, 2014.

[21] D. Golodnitsky, E. Strauss, E. Peled, and S. Greenbaum,“Review- on order and disorder in polymer electrolytes,” Jour-nal of the Electrochemical Society, vol. 162, no. 14, pp. A2551–A2566, 2015.

[22] M. B. Armand, “Polymer electrolytes,” Annual Review ofMaterials Science, vol. 16, no. 1, pp. 245–261, 1986.

[23] L. Yue, J. Ma, J. Zhang et al., “All solid-state polymer electro-lytes for high-performance lithium ion batteries,” Energy Stor-age Materials, vol. 5, pp. 139–164, 2016.

[24] J. Mindemark, M. J. Lacey, T. Bowden, and D. Brandell,“Beyond PEO- alternative host materials for Li+-conductingsolid polymer electrolytes,” Progress in Polymer Science,vol. 81, pp. 114–143, 2018.

[25] D. E. Fenton, J. M. Parker, and P. V. Wright, “Complexes ofalkali metal ions with poly(ethylene oxide),” Polymer, vol. 14,no. 11, p. 589, 1973.

[26] P. V. Wright, “Ionic conductivity and organisation of macro-molecular polyether–alkali-metal salt complexes,” Journal ofMaterials Chemistry, vol. 5, no. 9, pp. 1275–1283, 1995.

[27] D. T. Hallinan and N. P. Balsara, “Polymer electrolytes,”Annual Review of Materials Research, vol. 43, no. 1, pp. 503–525, 2013.

[28] C. A. Angell, C. Liu, and E. Sanchez, “Rubbery solid elec-trolytes with dominant cationic transport and high ambientconductivity,” Nature, vol. 362, no. 6416, pp. 137–139,1993.

[29] L. Feng and H. Cui, “A new solid-state electrolyte: rubbery'polymer-in-salt' containing LiN(CF3SO2)2,” Journal of PowerSources, vol. 63, no. 1, pp. 145–148, 1996.

[30] C. A. Angell, “Mobile ions in amorphous solids,” AnnualReview of Physical Chemistry, vol. 43, no. 1, pp. 693–717, 1992.

[31] M. Forsyth, J. Sun, D. R. Macfarlane, and A. J. Hill, “Com-positional dependence of free volume in PAN/LiCF3SO3polymer-in-salt electrolytes and the effect on ionic conduc-tivity,” Journal of Polymer Science Part B: Polymer Physics,vol. 38, no. 2, pp. 341–350, 2000.

[32] M. M. Doeff, L. Edman, S. E. Sloop, J. Kerr, and L. C. DeJonghe, “Transport properties of binary salt polymer electro-lytes,” Journal of Power Sources, vol. 89, no. 2, pp. 227–231,2000.

[33] M. Forsyth, S. Jiazeng, and D. R. MacFarlane, “Novel highsalt content polymer electrolytes based on high Tg poly-mers,” Electrochimica Acta, vol. 45, no. 8-9, pp. 1249–1254, 2000.

8 Energy Material Advances

[34] B. Wu, L. Wang, Z. Li et al., “Performance of "polymer-in-salt"electrolyte PAN-LiTFSI enhanced by graphene oxide filler,”Journal of the Electrochemical Society, vol. 163, no. 10,pp. A2248–A2252, 2016.

[35] L. Chen and L.-Z. Fan, “Dendrite-free Li metal deposition inall-solid-state lithium sulfur batteries with polymer-in-saltpolysiloxane electrolyte,” Energy Storage Materials, vol. 15,pp. 37–45, 2018.

[36] K. M. Abraham and M. Alamgir, “Ambient-temperaturerechargeable polymer-electrolyte batteries,” Journal of PowerSources, vol. 43, no. 1-3, pp. 195–208, 1993.

[37] F. Groce, F. Gerace, G. Dautzemberg, S. Passerini, G. B.Appetecchi, and B. Scrosati, “Synthesis and characterizationof highly conducting gel electrolytes,” Electrochimica Acta,vol. 39, no. 14, pp. 2187–2194, 1994.

[38] H. H. Kuo, W. C. Chen, T. C. Wen, and A. Gopalan, “A novelcomposite gel polymer electrolyte for rechargeable lithiumbatteries,” Journal of Power Sources, vol. 110, no. 1, pp. 27–33, 2002.

[39] C. Sun, J. Liu, Y. Gong, D. P. Wilkinson, and J. Zhang, “Recentadvances in all-solid-state rechargeable lithium batteries,”Nano Energy, vol. 33, pp. 363–386, 2017.

[40] M. Forsyth, J. Z. Sun, and D. R. MacFarlane, “Novel polymer-in-salt electrolytes based on polyacrylonitrile (PAN) lithiumtriflate salt mixtures,” Solid State Ionics, vol. 112, no. 1-2,pp. 161–163, 1998.

[41] J. Li, X. Huang, and L. Chen, “X-ray diffraction and vibrationalspectroscopic studies on PAN-LiTFSI polymer electrolytes,”Journal of the Electrochemical Society, vol. 147, no. 7,pp. 2653–2657, 2000.

[42] D. Bresser, S. Lyonnard, C. Iojoiu, L. Picard, and S. Passerini,“Decoupling segmental relaxation and ionic conductivity forlithium-ion polymer electrolytes,” Molecular Systems Design& Engineering, vol. 4, no. 4, pp. 779–792, 2019.

[43] A. Ferry, L. Edman, M. Forsyth, D. R. MacFarlane, and J. Sun,“Connectivity, ionic interactions, and migration in a fast-ion-conducting polymer-in-salt electrolyte based on poly(acryloni-trile) and LiCF3SO3,” Journal of Applied Physics, vol. 86, no. 4,pp. 2346–2348, 1999.

[44] J. B. Goodenough and Y. Kim, “Challenges for rechargeable Libatteries,” Chemistry of Materials, vol. 22, no. 3, pp. 587–603,2010.

[45] H. Duan, Y. X. Yin, X. X. Zeng et al., “In-situ plasticized poly-mer electrolyte with double-network for flexible solid-statelithium-metal batteries,” Energy Storage Materials, vol. 10,pp. 85–91, 2018.

[46] W. Gorecki, M. Jeannin, E. Belorizky, C. Roux, andM. Armand, “Physical properties of solid polymer electrolytePEO(LiTFSI) complexes,” Journal of Physics: Condensed Mat-ter, vol. 7, no. 34, pp. 6823–6832, 1995.

[47] C. Berthier, W. Gorecki, M. Minier, M. B. Armand, J. M.Chabagno, and P. Rigaud, “Microscopic investigation of ionicconductivity in alkali metal salts-poly(ethylene oxide)adducts,” Solid State Ionics, vol. 11, no. 1, pp. 91–95, 1983.

[48] M. Forsyth, D. R. MacFarlane, and A. J. Hill, “Glass transitionand free volume behaviour of poly(acrylonitrile)/LiCF3SO3polymer-in-salt electrolytes compared to poly(ether uretha-ne)/LiClO4 solid polymer electrolytes,” Electrochimica Acta,vol. 45, no. 8-9, pp. 1243–1247, 2000.

[49] Z. Wei, Z. Zhang, S. Chen et al., “UV-cured polymer electro-lyte for LiNi0.85Co0.05Al0.1O2//Li solid state battery working

at ambient temperature,” Energy Storage Materials, vol. 22,pp. 337–345, 2019.

[50] M. S. Whittingham, “Lithium batteries and cathode materials,”Chemical Reviews, vol. 104, no. 10, pp. 4271–4302, 2004.

[51] H. K. Song, K. T. Lee, M. G. Kim, L. F. Nazar, and J. Cho,“Recent progress in nanostructured cathode materials for lith-ium secondary batteries,” Advanced Functional Materials,vol. 20, no. 22, pp. 3818–3834, 2010.

[52] F. Croce, L. Persi, F. Ronci, and B. Scrosati, “Nanocompositepolymer electrolytes and their impact on the lithium batterytechnology,” Solid State Ionics, vol. 135, no. 1-4, pp. 47–52,2000.

[53] C. Brissot, M. Rosso, J. N. Chazalviel, and S. Lascaud, “Den-dritic growth mechanisms in lithium/polymer cells,” Journalof Power Sources, vol. 81-82, pp. 925–929, 1999.

[54] M. Rosso, J. N. Chazalviel, and E. Chassaing, “Calculation ofthe space charge in electrodeposition from a binary electro-lyte,” Journal of Electroanalytical Chemistry, vol. 587, no. 2,pp. 323–328, 2006.

[55] K. E. Thomas, S. E. Sloop, J. B. Kerr, and J. Newman, “Com-parison of lithium-polymer cell performance with unity andnonunity transference numbers,” Journal of Power Sources,vol. 89, no. 2, pp. 132–138, 2000.

[56] M. D. Tikekar, S. Choudhury, Z. Tu, and L. A. Archer, “Designprinciples for electrolytes and interfaces for stable lithium-metal batteries,” Nature Energy, vol. 1, no. 9, article 16114,2016.

[57] S. Zugmann, M. Fleischmann, M. Amereller, R. M. Gschwind,H. D. Wiemhöfer, and H. J. Gores, “Measurement of transfer-ence numbers for lithium ion electrolytes via four differentmethods, a comparative study,” Electrochimica Acta, vol. 56,no. 11, pp. 3926–3933, 2011.

[58] D. Fauteux, A. Massucco, M. McLin, M. Van Buren, and J. Shi,“Lithium polymer electrolyte rechargeable battery,” Electrochi-mica Acta, vol. 40, no. 13-14, pp. 2185–2190, 1995.

[59] Z. Florjańczyk, E. Zygadło-Monikowska, W. Wieczorek et al.,“Polymer-in-salt electrolytes based on acrylonitrile/butyl acry-late copolymers and lithium salts,” Journal of Physical Chemis-try B, vol. 108, no. 39, pp. 14907–14914, 2004.

[60] C. A. Angell, Y. Ansari, and Z. Zhao, “Ionic liquids: past, pres-ent and future,” Faraday Discussions, vol. 154, pp. 9–27, 2012.

[61] M. Doyle, T. F. Fuller, and J. Newman, “The importance ofthe lithium ion transference number in lithium/polymercells,” Electrochimica Acta, vol. 39, no. 13, pp. 2073–2081,1994.

[62] Z. Qiu, C. Liu, J. Xin et al., “High conductive composite poly-mer electrolyte via in situ UV-curing for all-solid-state lithiumion batteries,” ACS Sustainable Chemistry & Engineering,vol. 7, no. 11, pp. 9875–9880, 2019.

[63] X. Han, Y. Gong, K. (. K.). Fu et al., “Negating interfacialimpedance in garnet-based solid-state Li metal batteries,”Nature Materials, vol. 16, no. 5, pp. 572–579, 2017.

[64] X. Judez, M. Piszcz, E. Coya et al., “Stable cycling of lithiummetal electrode in nanocomposite solid polymer electrolyteswith lithium bis(fluorosulfonyl)imide,” Solid State Ionics,vol. 318, pp. 95–101, 2018.

[65] W. Zhou, Y. Zhu, N. Grundish et al., “Polymer lithium-garnetinterphase for an all-solid-state rechargeable battery,” NanoEnergy, vol. 53, pp. 926–931, 2018.

[66] L. X. Yuan, Z. H. Wang, W. X. Zhang et al., “Development andchallenges of LiFePO4 cathode material for lithium-ion

9Energy Material Advances

batteries,” Energy & Environmental Science, vol. 4, no. 2,pp. 269–284, 2011.

[67] W. Zhou, Z.Wang, Y. Pu et al., “Double-layer polymer electro-lyte for high-voltage all-solid-state rechargeable batteries,”Advanced Materials, vol. 31, no. 4, article 1805574, 2019.

[68] L. Li, Y. Deng, and G. Chen, “Status and prospect of garnet/po-lymer solid composite electrolytes for all- solid-state lithiumbatteries,” Journal of Energy Chemistry, vol. 50, pp. 154–177,2020.

[69] L. Damen, J. Hassoun, M. Mastragostino, and B. Scrosati,“Solid-state, rechargeable Li/LiFePO4 polymer battery for elec-tric vehicle application,” Journal of Power Sources, vol. 195,no. 19, pp. 6902–6904, 2010.

[70] J. R. Nair, D. Cíntora-Juárez, C. Pérez-Vicente, J. L. Tirado,S. Ahmad, and C. Gerbaldi, “Truly quasi-solid-state lithiumcells utilizing carbonate free polymer electrolytes on engineeredLiFePO4,” Electrochimica Acta, vol. 199, pp. 172–179, 2016.

[71] D. Aurbach, B. Markovsky, I. Weissman, E. Levi, and Y. Ein-Eli, “On the correlation between surface chemistry and perfor-mance of graphite negative electrodes for Li ion batteries,”Electrochimica Acta, vol. 45, no. 1-2, pp. 67–86, 1999.

[72] R. Xu, X. B. Cheng, C. Yan et al., “Artificial interphases forhighly stable lithium metal anode,” Matter, vol. 1, no. 2,pp. 317–344, 2019.

[73] M. Yan, J. Y. Liang, T. T. Zuo et al., “Stabilizing polymer-lithium interface in a rechargeable solid battery,” AdvancedFunctional Materials, vol. 30, no. 6, article 1908047, 2019.

[74] H. Gao, S. Xin, L. Xue, and J. B. Goodenough, “Stabilizing ahigh-energy-density rechargeable sodium battery with a solidelectrolyte,” Chem, vol. 4, no. 4, pp. 833–844, 2018.

[75] V. Sharova, A. Moretti, T. Diemant, A. Varzi, R. J. Behm, andS. Passerini, “Comparative study of imide-based Li salts aselectrolyte additives for Li-ion batteries,” Journal of PowerSources, vol. 375, pp. 43–52, 2018.

[76] K. N. Wood and G. Teeter, “XPS on Li-battery-related com-pounds: analysis of inorganic SEI phases and a methodologyfor charge correction,” ACS Applied Energy Materials, vol. 1,no. 9, pp. 4493–4504, 2018.

[77] C. Fiedler, B. Luerssen, M. Rohnke, J. Sann, and J. Janek,“XPS and SIMS analysis of solid electrolyte interphases onlithium formed by ether-based electrolytes,” Journal of theElectrochemical Society, vol. 164, no. 14, pp. A3742–A3749,2017.

[78] S. Jiao, J. Zheng, Q. Li et al., “Behavior of lithium metalanodes under various capacity utilization and high currentdensity in lithium metal batteries,” Joule, vol. 2, no. 1,pp. 110–124, 2018.

[79] T. Jaumann, J. Balach, M. Klose, S. Oswald, J. Eckert, andL. Giebeler, “Role of 1,3-dioxolane and LiNO3 addition onthe long term stability of nanostructured silicon/carbonanodes for rechargeable lithium batteries,” Journal of theElectrochemical Society, vol. 163, no. 3, pp. A557–A564,2016.

[80] M.Wu, Z.Wen, Y. Liu, X.Wang, and L. Huang, “Electrochem-ical behaviors of a Li3Nmodified Li metal electrode in second-ary lithium batteries,” Journal of Power Sources, vol. 196,no. 19, pp. 8091–8097, 2011.

[81] Y. J. Zhang, W. Wang, H. Tang et al., “An ex-situ nitridationroute to synthesize Li 3 N-modified Li anodes for lithium sec-ondary batteries,” Journal of Power Sources, vol. 277, pp. 304–311, 2015.

[82] Y. Okuno, K. Ushirogata, K. Sodeyama, and Y. Tateyama,“Decomposition of the fluoroethylene carbonate additive andthe glue effect of lithium fluoride products for the solid electro-lyte interphase: an ab initio study,” Physical Chemistry Chem-ical Physics, vol. 18, no. 12, pp. 8643–8653, 2016.

[83] Z. L. Brown, S. Jurng, C. C. Nguyen, and B. L. Lucht, “Effect offluoroethylene carbonate electrolytes on the nanostructure ofthe solid electrolyte interphase and performance of lithiummetal anodes,” ACS Applied Energy Materials, vol. 1, no. 7,pp. 3057–3062, 2018.

10 Energy Material Advances

Formation of Stable Interphase of Polymer-in-Salt Electrolyte in All-Solid-State Lithium Batteries1. Introduction2. Results and Discussion3. Conclusions4. Materials and MethodsConflicts of InterestAuthors’ ContributionsAcknowledgmentsSupplementary Materials