Metal Halide Superionic Conductors for All-Solid-State Batteries

Metal Halide Superionic Conductors for All-Solid-State BatteriesJianwen Liang Xiaona Li Keegan R Adair and Xueliang Sun

Cite This Acc Chem Res 2021 54 1023minus1033 Read Online

ACCESS Metrics amp More Article Recommendations

CONSPECTUS Rechargeable all-solid-state Li batteries (ASSLBs) areconsidered to be the next generation of electrochemical energy storagesystems The development of solid-state electrolytes (SSEs) which arekey materials for ASSLBs is therefore one of the most importantsubjects in modern energy storage chemistry Various types ofelectrolytes such as polymer- oxide- and sulfide-based SSEs havebeen developed to date and the discovery of new superionic conductorsis still ongoing Metal-halide SSEs (Li-M-X where M is a metal elementand X is a halogen) are emerging as new candidates with a number ofattractive properties and advantages such as wide electrochemicalstability windows (036minus671 V vs LiLi+) and better chemical stabilitytoward cathode materials compared to other SSEs Furthermore someof the metal-halide SSEs (such as the Li3InCl6 developed by our group)can be directly synthesized at large scales in a water solvent removing the need for special apparatus or handling in an inertatmosphere Based on the recent advances herein we focus on the topic of metal-halide SSEs aiming to provide a guidance towardfurther development of novel halide SSEs and push them forward to meet the multiple requirements of energy storage devicesIn this Account we describe our recent progress in developing metal halide SSEs and focus on some newly reported findings basedon state-of-the-art publications on this topic A discussion on the structure of metal-halide SSEs will be first explored Subsequentlywe will illustrate the effective approaches to enhance the ionic conductivities of metal halide SSEs including the effect of anionsublattice framework the regulation of site occupation and disorder and defect engineering Specifically we demonstrated thatproper structural framework balanced Li+vacancy concentration and reduced blocking effect can promote fast Li+ migration formetal halide SSEs Moreover humidity stability and degradation chemistry of metal halide SSEs have been summarized for the firsttime Some examples of the application of metal halide SSEs with stability toward humidity have been demonstrated Directsynthesis of halide SSEs on cathode materials by the water-mediated route has been used to eliminate the interfacial challenges ofASSLBs and has been shown to act as an interfacial modifier for high-performance all-solid-state LiminusO2 batteries Taken togetherthis Account on metal halide SSEs will provide an insightful perspective over the recent development and future research directionsthat can lead to advanced electrolytes

KEY REFERENCES

bull Liang J Li X Wang S Adair K R Li W Zhao YWang C Hu Y Zhang L Zhao S Lu S HuangH Li R Mo Y Sun X Site-Occupation-TunedSuperionic LixScCl3+x Halide Solid Electrolytes for All-Solid-State Batteries J Am Chem Soc 2020 142 157012minus70221 Optimizing conditions of the Li migrationmechanism to achieve the highest room-temperature ionicconductivities in metal halide-based electrolytes The designprinciples including the anion sublattice the local structurethe changing of site occupations and orderdisorderdistribution of elementalvacancies

bull Li X Liang J Adair K R Li J Li W Zhao F HuY Sham T Zhang L Zhao S Lu S Huang H LiR Chen N Sun X Origin of Superionic Li3Y1minusxInxCl6Halide Solid Electrolytes with High Humidity Toler-ance Nano Lett 2020 20 4384minus43922 The function of

the M atom in Li3MCl6 was clarif ied to reveal the structuralconversion as well as humidity tolerance mechanism

bull Li X Liang J Chen N Luo J Adair K R WangC Banis M N Sham T Zhang L Zhao S Lu SHuang H Li R Sun X Water-Mediated Synthesis ofa Superionic Halide Solid Electrolyte Angew Chem2019 131 16579minus165843 A halide Li+ superionicconductor Li3InCl6 which can be synthesized in waterMost importantly the as-synthesized Li3InCl6 shows a highionic conductivity of sim2 times 10minus3 S cmminus1 at 25 degC

Received November 15 2020Published January 29 2021

Articlepubsacsorgaccounts

copy 2021 American Chemical Society1023

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

Dow

nloa

ded

via

WE

STE

RN

UN

IV o

n M

ay 2

4 2

021

at 1

718

46

(UT

C)

See

http

sp

ubs

acs

org

shar

ingg

uide

lines

for

opt

ions

on

how

to le

gitim

atel

y sh

are

publ

ishe

d ar

ticle

s

1 INTRODUCTION

The search for solid-state electrolytes (SSEs) with high ionicconductivity and good environmentalelectrochemical stabilityis critical toward meeting the growing energy and safetydemands of all-solid-state lithium batteries (ASSLBs) Over thepast decades many different classes of SSEs have beendeveloped such as oxide- sulfide- and halide-based SSEs4minus8

Metal halides (fluoride chloride and bromide etc) whichpossess anion compounds that are more electronegative thanoxide and sulfide can achieve wide thermodynamic electro-chemical stability windows79 Recent findings such as room-temperature (RT) ionic conductivities over 1 times 10minus3 S cmminus1good stability toward oxide cathode airhumidity toleranceand even water-mediated synthesis routes have made the metalhalide SSEs quite attractiveIn fact metal halide SSEs have been explored for decades

Early halide SSEs (LiAlCl4 Li2MgCl4 etc)710minus13 which have

been ignored now due to their relatively low RT ionicconductivity played a crucial historical role in understandingthe Li+ diffusion in halide based anion frameworks Despite thegreat process recently metal halide SSEs with high ionicconductivity are still rare There are only a few examples withthe RT ionic conductivity higher than 10minus3 S cmminus1 includinghigh temperature phase Li3InBr6

14 Li3YBr615 Li3InCl6

216 Zrdoped Li3MCl6 (M = Y Er)17 Li3Y1minusxInxCl6

3 LixScCl3+x1 and

Li2Sc23Cl4 SSEs18 The ionic conductivities of the metal halide

SSEs as a function of temperature are shown in Figure 1 alongwith other solid or organic electrolytes Most of these metalhalide SSEs with high ionic conductivity are chlorides andbromides Fluoride-based SSEs have the widest electro-chemical windows however they still suffer from low RTionic conductivity Fluoride-based SSEs have only beenreported in a few forms including thin film LiAlF4 (1 times 10minus6

S cmminus1)19 bulk Li3AlF6 (44 times 10minus8 S cmminus1)20 thin-filmMIIF2M

IIIF3-doped LiF (MII = Mg Ca Ni Cu Zn Sr MIII =AI Ti V Cr Ga Y Ce)21 with the highest ionic conductivityreaching 5 times 10minus7 S cmminus1 Upon doping the RT ionicconductivity can be improved by 1 order of magnitude which

is still not high enough to achieve acceptable ASSLBs It stillneeds to pay more attention to develop the new fluorideelectrolyte with higher ionic conductivity In addition mostfluorides are toxic which also restricts their further develop-mentIn this Account we will focus on the Li-M-X component

where M is Sc Y Ln and In and X is a halogen We willdiscuss the work conducted by our group on the developmentof Li-M-X SSEs in the past few years (i) our work associatedwith an understanding of the structure Li+ transportmechanisms and effective approaches to enhance ionicconductivity (ii) our findings on revealing the complexphysical phenomenon of Li+ transport and the related factorsincluding anion sublattice framework cation orderminusdisordereffects Li+ carrier concentration and vacancy concentrationfor Li+ diffusion as well as cation blocking effects etc (iii) thedesign of metal-halide electrolytes with improved environ-mental stability and (iv) the humidity tolerance and water-mediated synthesis process of Li3InCl6 electrolyte Takentogether this Account on Li-M-X SSEs will provide aninsightful perspective over the recent development and futureresearch directions of advanced metal halide SSEs for the fieldof ASSLBs

2 RATIONAL DESIGN FOR SUPERIONICCONDUCTORS

21 Crystallography

The stacking structure of metal halide Li-M-X ionic crystalsrequires ions of different radius to be arranged based on theirelectrostatic forces Thus the structure of Li-M-X is highlydependent on the radius polarity and arrangement of ionsThe ionic radii of Fminus (122 pm) Clminus (167 pm) Brminus (182 pm)and Iminus (202 pm) is larger than that of Li+ (73 pm intetrahedron 90 pm in octahedron) and metal ions M3+ (88minus118 pm) Therefore the structural framework of these metalhalide electrolytes is built up by the anion stack sublattice andinfluenced by the volume and polarity of the cation species Itshould be noted that all of the ionic radii in this paper are

Figure 1 Evolution of ionic conductivity of the typical metal halide superionic conductors along with other lithium solid electrolytes and organicliquid electrolytes as a function of temperature1215171822minus30

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1024

Shannonrsquos radius31 Based on the law of ionic packing thestructure is stable only when the cation and anion are in closecontact The radius ratio of cation to anion (r+rminus) marked ast must meet certain conditions For example when t is rangefrom 0732 to 1 the cation will fill in the interstitial site of thecubic structure made up of 8 anions to form an MX8 cubewhich is similar to the CsCl structure When the radius of thecation is reduced and t is between 0414 and 0732 the cationwill occupy the octahedral site of six anions and then a MX6octahedron structure will be formed When a smaller cationcannot fully fill the octahedron site the structure will change toa more stable MX4 tetrahedron structure (t = 0255minus0414) ora MX3 triangle structure (t = 0155minus0255) Table 1 reveals theionic radius of cations and anions of typical Li-M-Xcomposition and the radius ratio of M cation to X anion(tMX) For fluoride the tMF value is higher than 0732 formost compositions Thus the Li-M-F structures tend to form aLiMF4 phase with M ion occupancy in the cubic site of the Fminus

framework (MF8 cube) All values of the tMX (X is Cl Br I) aswell as tScF are between 0414 and 0732 resulting in a stableLi3MX6 phase with the same local anion coordinationenvironment of MX6 octahedron However until recentlythere have only been a few examples of Li3MX6 phase found inthe inorganic crystal structure database7 There are many Li-M-X compounds especially for iodides that have not beensynthesized For example for chloride compounds there is noreport of any kind of Li-Ln-Cl phase when the Ln ion radius ishigher than 1087 pm (La3+ Ce3+ Pr3+ Nd3+ Pm3+ Sm3+Eu3+) For bromide compounds there are no Li-Ln-Br phasesthat have been reported when the Ln ion radius is higher than111 pm (La3+ Ce3+ Pr3+ Nd3+ Pm3+) At the same time onlytwo kinds of Li-M-I materials (Li3ScI6 Li3ErI6) have beenfound to date Whether the lack of Li-M-X components is anintrinsic thermodynamic result or due to difficult synthesisprocesses remains unclearIn the ternary chlorides of Li3MCl6 there are two kinds of

Clminus sublattice structures including cubic close packing (ccp)and hexagonal close packing (hcp)7 which can be furtherdivided into three phase structures The first is a trigonalstructure (P3m1) with hcp anion sublattice (hcp-T) Thesecond is an orthorhombic structure (Pnma) with hcp anionsublattice (hcp-O) The difference between the hcp-O andhcp-T crystal structures is the same ABAB stacking of close-packed Clminus with different symmetrical arrangements of Li+ andM3+ in the octahedral sites The last is monoclinic structurewith an ABCABC stacking of the ccp anion sublattice (ccp-M)Typically when the radius of M3+ ranges from 1063 pm(Tb3+) to 102 pm (Tm3+) the Li3MCl6 can maintain the hcp-T structure In this case tMCl ranges from 0611 to 0637Moreover when the radius of M3+ is smaller than 102 and

larger than 100 pm (Yb3+ and Lu3+) the structure of Li3MCl6converts to the hcp-O structure In this case tMCl ranges from0599 to 0611 When further reducing the radii of M3+ thestructure will convert from an hcp to ccp anion sublattice suchas Li3ScCl6 (tScCl = 0530) and Li3InCl6 (tInCl = 0563)Different to chloride-based ternary halides all of the Li3MBr6and Li3MI6 reveal the same ccp-M structure7 Although only afew systems with high ionic conductivity have been reported itcan be inferred that there will be a large number of potentialcomponents with high ionic conductivity Furthermore verylimited doping or substitution studies on the aforementionedmetal-halide SSEs can be found It is quite possible to furtherenhance the ionic conductivity and electrochemical perform-ance of the Li-M-X electrolytes

22 Li Ion Transport Kinetics

Li ion transport kinetics are the key factor in developing SSEsIn fact the hopping motion of Li+ is a universal feature of Li+

transport in SSEs The crystal lattice restricts the positions andthe conducting paths of Li+ For Li-M-X SSEs with a ccp-Mstructure915 the pathways of Li+ conduction are connected viatetrahedral interstitial sites between edge-sharing octahedralsites in all three directions forming a three-direction isotropicdiffusion network For the hcp anion arrangement (hcp-O andhcp-T)9 Li+ transports along the ab-plane are via tetrahedralinterstitial sites which is similar to the ccp-M structure Alongthe c-axis the diffusion paths are directly connected betweenneighboring octahedral sites Other than the long-range crystallattice structure local orderminusdisorder of Li+ sublattices anddistortion of the local structure also play a crucial role in theLi+ transport process It should be noted that the orderminusdisorder of the Li+ sublattice and distortion of the localstructure often appear together due to defects Othercrystallographic and electronic factors also affect the Li+

transport kinetics such as the polarity of ion induced latticedynamics and the proximity of metal ions to activate or blockLi+ hopping These transport phenomena will be discussed inthe following sections and have strong effects on the Li+

mobility in metal halide SSEs

23 Defect Chemistry

Different from the long-range structural characteristics defectsalways induce the local structure distortions or short-rangearrangement32 An attractive feature of solid-state structures isthe possibility of turning defect-chemistry to determine theeffectiveness of the Li+ hopping in the lattice The Li+ hoppingprocess is sensitive to the active Li+ and vacancy concentrationwhich can lead to orders of magnitude change in conductivityThe ion conduction through vacancies is the dominantmechanism for the diffusion of Li+ in many kinds of superionicconductors33 Figure 2 reveals the important crystal structures

Table 1 Polarizability and Ionic Radii of Cations and Anions of Metal Halide SSEs

ion Li+ In3+ Sc3+ Y3+ Er3+ La3+ sim Lu3+ Fminus Clminus Brminus Iminus O2minus S2minus

coordination number 6 6 6 6 6 6 6 6 6 6 6 6polarizability α (Aring3) 003 051 0286 055 069 sim114minus0606 104 366 477 710 388 102Pauling radius (pm) 76 80 745 90 89 sim1032minus861 136 181 195 216 140 184crystal radiusa (pm) 90 94 885 104 103 sim1172minus1001 122 167 181 202 124 170tMF 074 077 073 085 084 sim096minus082tMCl 054 056 053 062 062 sim062minus060tMBr 050 052 049 057 057 sim057minus055tMI 045 047 044 051 051 sim051minus050

aCrystal radius original from R D Shannonrsquos research31

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1025

and Li+ hopping mechanisms induced by vacancy clusters inLi-M-X systems Compared to the constriction against thedirect or ring exchange diffusion process the energy barrier ofLi+ diffusion based on the vacancy mechanism in the close-packed lattice is much smaller The structure of Li3MX6 isessentially a distorted rock-salt LiX structure which can beconsidered as M3+ doping in LiX16 Particularly one M3+

doping will remove three Li+ cations and introduce twovacancies in the octahedral interstitial sites As a result thevacancies occupy sim333 octahedral interstitial sites in thestructure of Li3MX6 and the ratio of Li+ M3+ and vacancies is312 Presumably these vacancies originate from aliovalentM3+ cation doping of Li+ and are essential to the high ionicconductivity of the crystalline Li3MX6 structures such asLi3InCl6 Li3ScCl6 and Li3YBr6 etc Vacancy clusters groupingof vacancies within the crystal lattice and defect chemistryserve as important parameters for Li+ transport in thesevacancy-rich environmentsFurther aliovalent substitution or doping of metal ions to

optimize vacancies as well as Li+ concentration in a mobile Li+

sublattice is an effective strategy to improve the ionicconductivity of inorganic materials Nazar et al reportedLi3minusxM1minusxZrxCl6 (M = Y Er) SSEs with RT ionic conductivityup to 14 times 10minus3 S cmminus1 by using the aliovalent substitutionstrategies17 The substitution of M3+ by Zr4+ can introduce alarge number of vacancies Meanwhile the lattice structure willbe affected by the radius of different metal cations in the metalhalide electrolyte The substitution of M3+ ion by a smallerradius Zr4+ ion is accompanied by an hcp-T to hcp-Otransition Moreover isovalent substitution is another strategyfor improving the ionic conductivity Although the isovalentsubstitution of metal ions cannot induce vacancy or Li contentit can distort the local structure or even affect the anionstacked sublattice based on different radii and polarizability ofthe introduced cation Our group reported a series ofLi3Y1minusxInxCl6 (0 le x lt 1) materials by using a smaller ion ofIn3+ to substitute the Y3+ cation in the Li3YCl6 SSE3 Whenincreasing the In3+ content the structure is gradually convertedfrom hcp-T to hcp-O and then to the ccp-M TheLi3Y1minusxInxCl6 (0 le x lt 1) SSEs with the ccp anion sublatticerevealing more than 10 times higher Li+ conductivity whencompared to those with an hcp anion sublattice Afterward Mo

et al performed a systematic study on the known Li-containingchlorides superionic conductor and their doped effect based onfirst-principles computation34 and they predicted manypotential compounds with high Li-ion conductivities Fur-thermore they found that the Li-ion migration is greatlyimpacted by the cation configuration and concentrations As aresult a low Li octahedral occupancy (asymp40minus60) low cationconcentration and sparse cation distribution can increase theLi-ion conduction in chlorides

24 Lattice Dynamics

Lattice dynamics describe the atomic vibrations in crystalstructures which can affect the diffusion of ions35minus38 In factgreater Li+ mobility is correlated with decreasing lithiumvibration frequency When analyzing the relationship betweenionic transport and the polarizability of the lattice the phononproperties etc it can be confirmed that the polarizability of thehost ion lattice affects the lattice vibration strength and the Li+

mobility kinetics by changing the activation barrier andprefactor of the Arrhenius equation39 Wakamura et al40

revealed that low activation energies of Li+ transport arecorrelated with ldquolow-energyrdquo phonon frequencies and the high-frequency dielectric constant Thus a softer more polarizableanion sublattice is beneficial for ionic transport Table 1 revealsthe polarizability of cation and anion of typical Li-M-Xelectrolytes The polarizability and ionic radius of O2minus and S2minus

are also shown for comparison It can be found that thehalogen anion with the lowest polarizability is Fminus which is lessthan one-third of Clminus Thus in all of the halide SSEs fluoridesreveal the lowest ionic conductivity So far no fluoride SSEswith RT ionic conductivity higher than 1 times 10minus5 S cmminus1 hasbeen discovered The polarizability of Clminus is similar to O2minus andmuch lower than S2minus Thus the lattice vibration strength of aClminus sublattice is similar to the O2minus sublattice and somechlorides exhibit high ionic conductivity At the same timebromide and iodide have higher polarizability than chloridewhich suggests that they can achieve a faster Li+ migration Forexample Li3YBr6 has a higher RT ionic conductivity thanLi3YCl6 (3 times 10minus3 S cmminus1 for Li3YBr6 vs 51 times 10minus4 S cmminus1 ofLi3YCl6)

1522 Li3ErI6 also has a higher RT ionic conductivitywhen compared to Li3ErCl6 (65 times 10minus4 S cmminus1 for Li3ErI6 vs33 times 10minus4 S cmminus1 of Li3ErCl6)

23 Although sulfides andiodides have softer lattices than chlorides and possess higher

Figure 2 Important crystal structures and Li hopping mechanisms in metal halide superionic conductors based on vacancy cluster (a) Layeredstructure (with hcp or ccp anion sublattice) and (b) spinel structure (cd) Diffusion in these structures by vacancy clusters (divacancies in thelayered form and triple and divacancies in the spinel form) Red is Li yellow is Cl Panels aminusd reproduced with permission from ref 33 Copyright2013 American Chemical Society

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1026

ionic conductivity their stability against electrochemicaloxidation is much lower than chlorides and fluorides941

Modifying the lattice dynamics will be an emerging strategy for

the discovery of new ionic conductors however a trade-offbetween enhanced conductivity and electrochemical stabilityshould be considered The structural framework built by

Figure 3 Metal-halide structures based on the anion sublattice framework (a) hcp anion sublattice framework (b) trigonal structure of Li3YCl6(c) orthorhombic structure of Li3YCl6 (d) orthorhombic structure of Li253Y053Zr047Cl6 (e) ccp anion sublattice framework (fminush) layer structureof Li3InCl6 Li3InCl6 based on the water-synthesized process and Li3ScCl6 Red is Cl white is vacancy and green is Li

Figure 4 Diffusion mechanism of the LixScCl3+x SSEs (a b) Li+ probability density marked by yellow isosurfaces (c) Li+ migration pathways in

ccp-anion stacking sublattice of Li3ScCl6 structure (dminusg) Li+ probability density of (d) Li5ScCl8 (x = 5) (e) Li3ScCl6 (x = 3) (f) Li18ScCl48 (x =18) and (g) LiScCl4 (x = 1) structures (h i) blocking effect of Sc3+ in Li+ diffusion and (j) Arrhenius plot of Li+ diffusivity in LixScCl3+x (x = 118 3 and 5) from simulations Panels aminusj are reproduced with permission from ref 1 Copyright 2020 American Chemical Society

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1027

multiple anions such as Li-M-Cl-F Li-M-Cl-O and Li-M-Cl-Br which can adjust the lattice dynamic by changing the ratioof different anions or by changing the chemicalcrystal fieldenvironment will open a new route toward the design of moreLi superionic conductors

25 Cation OrderminusDisorder and Blocking Effect

The same system and anion stack sublattice can include a largevariation or disorder in the sites of cations and vacancieswhich is referred to as the polymorphism of ionic crystalstructure Cations as well as anions are in continuous motionin the crystal structure at finite temperatures The crystallinestructures are statistically average based on the characterizationand refinement of XRD neutron diffraction etc Mostpolymorphic structures originate from the cation orderminusdisordered arrangement induced by different occupation andsite positions This can result in different crystal symmetryband structure and affect their physicalchemical propertiesThe most typical example is the cation-disordered rock-salttyped oxide cathode in lithium-ion batteries42 while fewreports consider these effects when it comes to SSEs Typically

specific cation and vacancy arrangements lead to unique Li+

transport properties Thus different cation sublattice structureswith the same anion stacking can be achieved by changing thecontent and arrangement of Li+ M cation and vacancies in Li-M-X SSEs which can help us to further understand andoptimize the Li+ transport pathway and hopping effectsAlthough the function of the cation orderminusdisorder effect inmetal halide electrolyte has not been discussed much there arestill many interesting phenomena Figure 3 reveals some of thereported Li3MCl6 structures and the corresponding cationoccupation and positions Wolfgang et al24 reported that insome Li3MCl6 (M = Y Er) SSEs the M3+vacancies disorderstrongly benefits the transport properties For example someM2 atoms in the Li3MCl6 (M = Y Er) structure (Figure 3b)which is occupied the Wyckoff 2d position can be swapped tothe M3 position This M2-M3 swapping leads to a significantchange of the distorted local structure and Li-ion migrationbarrier High M2-M3 disorder in the structure is beneficial forionic transport The RT ionic conductivity of Li3ErCl6compounds with different M2-M3 disorders can range from017 to 31 times 10minus4 S cmminus1

Figure 5 (a) Moisture stability versus reduction stability for Li-M-Cl (b) ionic conductivity evolution of Li3InCl6 and Li3YCl6 exposed tohumidity (c) ionic conductivities of the pristine Li3InCl6 and the reheated Li3InCl6 after dissolving in water (d) comparison of the ionicconductivity retention of Li3Y1minusxInxCl6 SSEs before and after exposure to air with 3minus5 humidity for 12 h followed by a reheating process and (e)schematic illustration of the humidity stabilities of Li3Y1minusxInxCl6 and Li3YCl6 Panel a is reproduced with permission from ref 43 Copyright 2020John Wiley amp Sons Inc Panels b and c are reproduced with permission from ref 2 Copyright 2019 John Wiley amp Sons Inc Panels d and e arereproduced with permission from ref 3 Copyright 2020 American Chemical Society

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1028

Moreover the occupation of M cations in the structure willalso result in a Coulombic repulsion to Li+ migration Based onthe theoretical calculation the Li+ migration pathways intetrahedral interstitial sites adjacent to the M cation areblocked This blocking effect not only reduces the migrationefficiency of Li+ but also distorts the diffusive channels115 Inthe ccp-M structure the Li+ conducting paths are connectedvia tetrahedral interstitial sites in all three directions (Figure4aminusc) Therefore the position and content of the M cationwill become a very important factor in the ionic conductionprocess Recently we synthesized a series of LixScCl3+x SSEs(25 lt x lt 4) with the fast Li+ conductivity up to 302 times 10minus3 Scmminus11 All LixScCl3+x SSEs exhibit a similar monoclinicstructure of Li3ScCl6 (Figure 4dminusg) The only difference istheir configurational variability and the occupation of Sc3+ Li+and vacancies in the octahedral interstice of Clminus Correspond-ingly the optimization of vacancies and cations content in thestructure can be adjusted by changing the x value By analyzingthe Li+ density probability of all LixScCl3+x no signal can befound at the tetrahedral site neighboring the Sc site at sim23 Aring(Figure 4i) Thus the tetrahedral interstitial sites adjacent toSc3+ are blocked which is shown as blue sites in Figure 4hThis blocking effect would distort the diffusion channels of Li+

(as shown in Figure 4dminusg) and result in different Li+ mobilitykinetics When increasing the x value in LixScCl3+x the Li+

carrier concentration will be increased while the oppositetrend is observed for the Sc blocking effect and the totalvacancy concentration for hopping within the structure Abalance is needed to achieve a structure with appropriate Li+

carrier concentration and vacancy concentration for Li+

diffusion as well as continuous diffusion channels Theconduction of ions in solids is a complex physical phenomenonand is affected by many factors Without a comprehensiveunderstanding of the mechanisms in these ion transportprocesses the rational design of new fast ionic conductors isnot possible

3 ENVIRONMENTAL STABILITY AND DEGRADATIONCHEMISTRY

Other than high ionic conductivity other attributes such asenvironmental stability and the airhumidity tolerance of metalhalide SSEs have recently gained significant interest2343 Herewe focus on the chemical properties of Li-M-Cl to reveal theenvironmental stability The oxidation potential of Clminus is muchhigher than O2minus which suggests that O2 from air cannotoxidize the Clminus anion in an ambient environment At the sametime most of the cations in the metal halide SSEs exist in ahigh valence state Thus it is reasonable to assume that mostmetal halide SSEs are stable in dry air Moreover based on thetheoretical calculations most ternary lithium chlorides Li-M-Cl except for Be2+ show positive hydrolysis reaction energieswhich means Li-M-Cl is generally stable against moisture(Figure 5a)43 The moisture stability for chlorides is much lessof an issue when compared to sulfide electrolytes Howeverexperimentally most of the Li-M-Cl SSEs suffer fromirreversible chemical degradation when exposed to a humidatmosphere Until now only Li3InCl6 can achieve a reversibleionic conductivity after being exposed to humidity andreheated216 It is believed that the chemical degradationprocess when exposed to humidity is very different from that ofsulfide electrolytes Thus deciphering the degradation processin halide SSEs is of paramount importance Figure 5b revealsthe degradation of ionic conductivity in Li3YCl6 and Li3InCl6

when exposed to humidity23 It can be seen that the reductionof ionic conductivity of Li3YCl6 is much higher than that ofLi3InCl6 After reheating Li3InCl6 can be recover over 92 ofits initial ionic conductivity while for Li3YCl6 a value of only08 of the pristine ionic conductivity can be retained (Figure5cd)To obtain a clearer picture of the degradation process of

Li3InCl6 exposed to air we tracked the chemistry and structureof Li3InCl6 during air exposure by using in situ and operandosynchrotron X-ray analytical techniques44 Li3InCl6 is hydro-philic leading to the absorption of moisture to form a hydrateLi3InCl6middotxH2O The Li3InCl6middotxH2O can be dehydrated toproducts Li3InCl6 and then the ionic conductivity can berecovered after a reheating at 200 degC under vacuumconditions The reversible interconversion between anhydrousand hydrated forms is the reason why Li3InCl6 has a hightolerance to water As the absorption of moisture continues toincrease Li3InCl6middotxH2O will be dissolved into the water toform a Li3InCl6 saturated solution The pH of Li3InCl6saturated solution is around 4 which suggests a slighthydrolysis process A small amount of white precipitate suchas In2O3 will form if the Li3InCl6 saturated solution is leftstanding in ambient air for too long (greater than 24 h) Thismight be the reason why therersquos a small irreversible loss of 8of ionic conductivity between pristine Li3InCl6 and theLi3InCl6 SSE with long air-exposure time followed by areheating process In the dry room environment the Li3InCl6SSE can be stored and used for more than 1 week withoutsignificant reduction in ionic conductivity However forLi3YCl6 the ionic conductivity cannot be retained even inlow humidity environments and will change to YCl3middot6H2O andLiClmiddotH2O after exposure to air with 3minus5 humidity for 12 h3

Moreover Li3YCl6 cannot be reformed after reheating thehumidity exposed sample in vacuum or inert atmosphere Theresulting product is LiCl and YOCl instead suggesting aserious hydrolysis reaction process (Figure 5e) Based on thedifferent degradation processes between Li3InCl6 and Li3YCl6we further studied the effect of M cations In3+ and Y3+ anddemonstrated the feasibility of increasing the humiditytolerance of Li3Y1minusxInxCl6 (0 le x lt 1) by optimizing thechemical properties via In3+ substitution of Y3+3 The functionof the M cation in Li3MX6 was clarified and the humiditytolerance is highly improved when the In3+ content is highenough to form hydrated intermediates as shown in Figure5de If we want to achieve metal halide SSEs with high airhumidity tolerance we need to understand the waterabsorption and hydrolysis process of the metal halide SSEsThe high humidity tolerance of metal halide SSEs originatesfrom the formation of hydrated intermediates rather thanseparated hydrated phases

4 ELECTRODE PROCESSING AND ALL-SOLID-STATEBATTERIES

An ideal SSE should exhibit high ionic conductivity andinterfacial compatibility with both cathode and anode541

However there is a trade-off between the ionic conductivityand oxidationreduction stability which deviates from theideal SSE45 Compared to sulfide and oxide SSEs switching theanion chemistry from O2minus and S2minus to halogens such as Fminus andClminus leads to lattice anions which are more difficult to oxidizeTherefore metal chloride and fluoride SSEs can achieve highstability toward oxide cathodes The stable cycling of LiCoO2and LiNi08Mn01Co01O2 etc oxide cathodes without any

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1029

protection layer can be achieved with the Li-M-Cl electrolyteswhich cannot be realized in sulfide electrolyte121518 More-over Li3InCl6 SSEs possess the intrinsic advantage of a water-mediated synthesis process which leads to several advancedpotential applications2 First the technique allows for large-scale and easy fabrication of Li3InCl6 Our group can achievemore than 100 g sample batches in the lab Currently incollaboration with an industrial partner the scale-up process ofLi3InCl6 up to the kilogram scale has been successfullyachieved Furthermore the synthesis is not limited to onlyanhydrous InCl3 precursors and cheaper hydrated indiumchlorides such as InCl3middot4H2O many other kinds of indiumcompounds such as In2(CO3)3 In2O3 and In(OH)3 may beused Second the water-mediated synthesis process of Li3InCl6enables the application of conformal coatings on cathodematerials with ease (Figure 6a) Our group reported an in situinterfacial growth of Li3InCl6 on LiCoO2 by the water-mediated synthesis process46 Owing to the strong interfacialinteractions excellent interfacial compatibility between

LiCoO2 and Li3InCl6 as well as high interfacial ionicconductivity LiCoO2 with 15 wt Li3InCl6 coating exhibitsa high initial capacity of 1317 mAh gminus1 at 01 C (1 C = 13 mAcmminus2) and can be operated up to 4 C at RT (Figure 6b) Thisinterfacial growth process resulted in a uniform coating whichnot only reduced the SSE content in the cathode layer but alsoimproved the kinetics of the Li ion transfer Moreover stablecycling is achieved with a capacity retention of 903 mAh gminus1

after 200 cycles (Figure 6c) In addition the water-mediatedsynthesis process can be further modified as an infiltrationprocess to improve the interfacial contact between twodifferent solid state particles such as Li15Al05Ge15(PO4)3(LAGP) electrolyte and nitrogen-doped carbon nanotube(NCNT) electrodes in LiminusO2 ASSLBs (Figure 6dminusf)47 Theresistance between LAGP and NCNT with the infiltrationprocess is comparable to the cells that rely on liquidelectrolytes wetting Moreover the cycling performance issuperior to that of the battery using liquid electrolyte Metalchloride SSEs are very effective considering their high ionic

Figure 6 (a) Illustration of the in situ synthesis of Li3InCl6 on LiCoO2 (LICLCO) (b) initial chargedischarge curves of LICLCO and (c)corresponding cycling stability (d) illustration of the synthesis process for the LAGP-NCNT-Li3InCl6 air electrode and the effect of Li3InCl6modifier on the decomposition of discharge products (e) Nyquist plots and (f) cycling performance of the LiminusO2 batteries with LAGP-NCNTLAGP-NCNT-05Li3InCl6 and LAGP-NCNT-liquid air electrodes respectively Panels aminusc are reproduced with permission from ref 46 Copyright2020 Elsevier Ltd Panels dminusf reproduced with permission from ref 47 Copyright 2020 Elsevier Ltd

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1030

conductivity low toxicity and ease of processabilityFurthermore they are stable in dry air and do not require anadditional protective coating layer when used in combinationwith certain oxide cathodes The main disadvantage of thesemetal halide SSEs is the low reduction stability due to theelectrochemical reduction of metal cations which prevents theuse of Li and graphite anodes

5 OUTLOOK AND PERSPECTIVE

Metal halide superionic conductors have generated significantinterest in the field of ASSLBs Our group has investigatedseveral novel halide SSEs with RT ionic conductivities higherthan 1 times 10minus3 S cmminus1 such as Li3InCl6 LixScCl3+xLi3Y1minusxInxCl6 and Li3YBr6 etc In terms of condensed matterphysics and chemistry via structural design and Li transportmechanism analysis there is a great opportunity to push theseexciting halide SSEs forward to meet the multiple requirementsfor energy storage in ASSLBs We believe a deeper under-standing of the Li+ transport is required to optimize thestructure and achieve higher conductivity Moreover weproposed and investigated the airhumidity tolerance anddegradation chemistry of Li3InCl6 electrolyte which is the onlymetal halide superionic conductors that can be synthesized inwater The water-mediated synthesis process can be furthermodified as an in situ interfacial growth or infiltration processto improve the interfacial contact between two different solid-state particles such as the cathode particles and SSEsCurrently metal halide SSEs with higher ionic conductivity(over 1 times 10minus2 S cmminus1) better environmentalelectrochemicalstability (especially stable with anode materials such as Limetal) and lower cost are still lacking and need to bedeveloped Furthermore factors such as control over thethickness of the metal halide SSEs layer as well as the largescale preparing process should be carried out to achievepractical application in pouch cells We expect that thedevelopment of metal halide SSEs with high ionic conductivityand chemical stability will lead to further understanding oftheir mechanism and guide new SSE design

AUTHOR INFORMATION

Corresponding Author

Xueliang Sun minus Department of Mechanical amp MaterialsEngineering University of Western Ontario London OntarioN6A 5B9 Canada orcidorg0000-0003-0374-1245Email xsun9uwoca

Authors

Jianwen Liang minus Department of Mechanical amp MaterialsEngineering University of Western Ontario London OntarioN6A 5B9 Canada

Xiaona Li minus Department of Mechanical amp MaterialsEngineering University of Western Ontario London OntarioN6A 5B9 Canada

Keegan R Adair minus Department of Mechanical amp MaterialsEngineering University of Western Ontario London OntarioN6A 5B9 Canada

Complete contact information is available athttpspubsacsorg101021acsaccounts0c00762

Notes

The authors declare no competing financial interest

Biographies

Jianwen Liang is a Mitacs Postdoc Fellow in Prof Xueliang (Andy)Sunrsquos Group at the University of Western Ontario Canada Hereceived his PhD degree in Inorganic Chemistry from the Universityof Science and Technology of China in 2015 He joined Prof Sunrsquosgroup in 2017 and his current research interests include sulfide andhalide solid electrolytes as well as all-solid-state LiLi-ion batteries

Xiaona Li is a Mitacs Postdoc Fellow in Prof Xueliang (Andy) SunrsquosGroup at the University of Western Ontario Canada She received herPhD degree in Inorganic Chemistry in 2015 from the University ofScience and Technology of China She joined Prof Sunrsquos group in2017 and her current research interests focus on the synthesis ofsulfide and halide solid electrolytes as well as all-solid-state lithiumbatteries

Keegan R Adair received his BSc in Chemistry from the Universityof British Columbia in 2016 He is currently a PhD candidate in ProfXueliang (Andy) Sunrsquos Group at the University of Western OntarioCanada Keegan has previous experience in the battery industrythrough internships at companies including E-One Moli Energy andGeneral Motors RampD His research interests include the design ofadvanced Li metal anodes and nanoscale interfacial coatings forbattery applications

Xueliang Sun is a Canada Research Chair in Development ofNanomaterials for Clean Energy Fellow of the Royal Society ofCanada and Canadian Academy of Engineering and Full Professor atthe University of Western Ontario Canada Dr Sun received hisPhD in Materials Chemistry in 1999 from the University ofManchester UK which he followed up by working as a postdoctoralfellow at the University of British Columbia Canada and as aResearch Associate at LrsquoInstitut National de la Recherche Scientifique(INRS) Canada His current research interests are focused onadvanced materials for electrochemical energy storage and conversionincluding solid-state batteries interface and solid state electrolytesand electrocatalysts

ACKNOWLEDGMENTS

This research was supported by the Natural Sciences andEngineering Research Council of Canada (NSERC) theCanada Research Chair Program (CRC) the CanadaFoundation for Innovation (CFI) the Ontario ResearchFund a Canada MITACS fellowship and the University ofWestern Ontario

REFERENCES(1) Liang J Li X Wang S Adair K R Li W Zhao Y WangC Hu Y Zhang L Zhao S Lu S Huang H Li R Mo Y SunX Site-occupation-tuned Superionic LixScCl3+x halide solid electro-lytes for all-solid-state batteries J Am Chem Soc 2020 142 7012minus7022(2) Li X Liang J Chen N Luo J Adair K R Wang C BanisM N Sham T K Zhang L Zhao S Lu S Huang H Li RSun X Water-mediated synthesis of a superionic halide solidelectrolyte Angew Chem Int Ed 2019 58 16427minus16432(3) Li X Liang J Adair K R Li J Li W Zhao F Hu YSham T K Zhang L Zhao S Lu S Huang H Li R Chen NSun X Origin of superionic Li3Y1minusxInxCl6 halide solid electrolyteswith high humidity tolerance Nano Lett 2020 20 4384minus4392(4) Bachman J C Muy S Grimaud A Chang H-H Pour NLux S F Paschos O Maglia F Lupart S Lamp P Giordano LShao-Horn Y Inorganic solid-state electrolytes for lithium batteriesmechanisms and properties governing ion conduction Chem Rev2016 116 140minus162

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1031

(5) Banerjee A Wang X Fang C Wu E A Meng Y SInterfaces and Interphases in All-Solid-State Batteries with InorganicSolid Electrolytes Chem Rev 2020 120 6878minus6933(6) Zhang Z Shao Y Lotsch B Hu Y-S Li H Janek J NazarL F Nan C-W Maier J Armand M Chen L New horizons forinorganic solid state ion conductors Energy Environ Sci 2018 111945minus1976(7) Li X liang J Yang X Adair K Wang C Zhao F Sun XProgress and perspectives of halide-based lithium conductors for all-solid-state batteries Energy Environ Sci 2020 13 1429minus1461(8) Xu L Li J Deng W Shuai H Li S Xu Z Li J Hou HPeng H Zou G Ji X garnet solid electrolyte for advanced all-solid-state Li batteries Adv Energy Mater 2021 11 2000648(9) Wang S Bai Q Nolan A M Liu Y Gong S Sun Q MoY Lithium chlorides and bromides as promising solid-statechemistries for fast ion conductors with good electrochemicalstability Angew Chem Int Ed 2019 58 8039minus8043(10) Plichta E Behl W Vujic D Chang W Schleich D TheRechargeable LixTiS2LiAlCl4Li1‑xCoO2 solid-state cell J Electro-chem Soc 1992 139 1509(11) Weppner W Huggins R A Thermodynamic and phaseequilibrium studies of the fast solid ionic conductor LiAlCl4 SolidState Ionics 1980 1 3minus14(12) Spector J Villeneuve G Hanebali L Cros C NMRInvestigations of the Li+ ion mobility in the double chloridesLi2MgCl4 and LiMgCl3 Mater Lett 1982 1 43minus48(13) Catlow C R A Wolf M A molecular-dynamics study of iontransport in lithium magnesium chloride solid electrolytes Proc RSoc London Math Phys Sci 1987 413 201minus224(14) Tomita Y Fuji-i A Ohki H Yamada K Okuda T Newlithium ion conductor Li3InBr6 studied by 7Li NMR Chem Lett1998 27 223minus224(15) Asano T Sakai A Ouchi S Sakaida M Miyazaki AHasegawa S Solid halide electrolytes with high lithium-ionconductivity for application in 4 V class bulk-type all-solid-statebatteries Adv Mater 2018 30 1803075(16) Li X Liang J Luo J Norouzi Banis M Wang C Li WDeng S Yu C Zhao F Hu Y Sham T K Zhang L Zhao SLu S Huang H Li R Adair K R Sun X Air-stable Li3InCl6electrolyte with high voltage compatibility for all-solid-state batteriesEnergy Environ Sci 2019 12 2665minus2671(17) Park K H Kaup K Assoud A Zhang Q Wu X Nazar LF High-voltage superionic halide solid electrolytes for all-solid-stateLi-ion batteries ACS Energy Lett 2020 5 533minus539(18) Zhou L Kwok C Y Shyamsunder A Zhang Q Wu XNazar L A new halospinel superionic conductor for high-voltage allsolid state lithium batteries Energy Environ Sci 2020 13 2056minus2063(19) Oi T Miyauchi K Amorphous thin film ionic conductors ofmLiF-nAlF3 Mater Res Bull 1981 16 1281minus1289(20) Feinauer M Euchner H Fichtner M Reddy M AUnlocking the Potential of Fluoride-based Solid Electrolytes for Solid-State Lithium Batteries ACS Appl Energy Mater 2019 2 7196minus7203(21) Oi T Ionic conductivity of LiF thin films containing Di-ortrivalent metal fluorides Mater Res Bull 1984 19 451minus457(22) Yu C Li Y Adair K R Li W Goubitz K Zhao YWillans M J Thijs M A Wang C Zhao F Tuning ionicconductivity and electrode compatibility of Li3YBr6 for high-performance all solid-state Li batteries Nano Energy 2020 77105097(23) Schlem R Bernges T Li C Kraft M A Minafra N ZeierW G Lattice Dynamical Approach for Finding the LithiumSuperionic Conductor Li3ErI6 ACS Appl Energy Mater 2020 33684minus3691(24) Schlem R Muy S Prinz N Banik A Shao-Horn Y ZobelM Zeier W G Mechanochemical Synthesis A tool to tune cationsite disorder and ionic transport properties of Li3MCl6 (M= Y Er)superionic conductors Adv Energy Mater 2020 10 1903719

(25) Steiner H J Lutz H Neue schnelle ionenleiter vom typMI

3MIIICl6 (M

I= Li Na Ag MIII= In Y) Z Anorg Allg Chem 1992613 26minus30(26) Stallworth P Fontanella J Wintersgill M Scheidler C DImmel J J Greenbaum S Gozdz A NMR DSC and high pressureelectrical conductivity studies of liquid and hybrid electrolytes JPower Sources 1999 81 739minus747(27) Kamaya N Homma K Yamakawa Y Hirayama M KannoR Yonemura M Kamiyama T Kato Y Hama S Kawamoto KMitsui A A lithium superionic conductor Nat Mater 2011 10682minus686(28) Kanno R Takeda Y Takada K Yamamoto O IonicConductivity and phase transition of the spinel system Li2minus2xM1+xCl4(M= Mg Mn Cd) J Electrochem Soc 1984 131 469(29) Yu X Bates J B Jellison G E Jr Hart F X A stable thin-film lithium electrolyte lithium phosphorus oxynitride J ElectrochemSoc 1997 144 524(30) Rettenwander D Redhammer G Preishuber-Pflugl FCheng L Miara L Wagner R Welzl A Suard E Doeff M MWilkening M Fleig J Amthauer G Structural and electrochemicalconsequences of Al and Ga cosubstitution in Li7La3Zr2O12 solidelectrolytes Chem Mater 2016 28 2384minus2392(31) Shannon R D Revised effective ionic radii and systematicstudies of interatomic distances in halides and chalcogenides ActaCrystallogr Sect A Cryst Phys Diffr Theor Gen Crystallogr 197632 751minus767(32) Schmitt R Nenning A Kraynis O Korobko R Frenkel AI Lubomirsky I Haile S M Rupp J L M A review of defectstructure and chemistry in ceria and its solid solutions Chem Soc Rev2020 49 554minus592(33) Van der Ven A Bhattacharya J Belak A A UnderstandingLi diffusion in Li-intercalation compounds Acc Chem Res 2013 461216minus1225(34) Liu Y Wang S Nolan A M Ling C Mo Y Tailoring theCation Lattice for Chloride Lithium-Ion Conductors Adv EnergyMater 2020 10 2002356(35) Rice M Roth W Ionic transport in super ionic conductors atheoretical model J Solid State Chem 1972 4 294minus310(36) Xu Z Chen X Liu K Chen R Zeng X Zhu H Influenceof anion charge on Li ion diffusion in a new solid-state electrolyteLi3LaI6 Chem Mater 2019 31 7425minus7433(37) Muy S Voss J Schlem R Koerver R Sedlmaier S JMaglia F Lamp P Zeier W G Shao-Horn Y High-throughputscreening of solid-state Li-ion conductors using lattice-dynamicsDescriptors iScience 2019 16 270minus282(38) Kraft M A Culver S P Calderon M Bocher F KrauskopfT Senyshyn A Dietrich C Zevalkink A Janek J R Zeier W GInfluence of lattice polarizability on the ionic conductivity in thelithium superionic argyrodites Li6PS5X (X= Cl Br I) J Am ChemSoc 2017 139 10909minus10918(39) Rice M J Roth W L Ionic transport in super ionicconductors a theoretical model J Solid State Chem 1972 4 294minus310(40) Wakamura K Roles of phonon amplitude and low-energyoptical phonons on superionic conduction Phys Rev B CondensMatter Mater Phys 1997 56 11593minus11599(41) Xiao Y Wang Y Bo S-H Kim J C Miara L J Ceder GUnderstanding interface stability in solid-state batteries Nat RevMater 2020 5 105(42) Cleacutement R J Lun Z Ceder G Cation-disordered rocksalttransition metal oxides and oxyfluorides for high energy lithium-ioncathodes Energy Environ Sci 2020 13 345minus373(43) Zhu Y Mo Y Materials design principles for air-stablelithiumsodium solid electrolytes Angew Chem Int Ed 2020 5917472minus17476(44) Li W Liang J Li M Adair K R Li X Hu Y Xiao QFeng R Li R Zhang L Lu S Huang H Zhao S Sham T KSun X Unraveling the origin of moisture stability of halide solid-state

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1032

electrolytes by in situ and operando synchrotron X-ray analyticaltechniques Chem Mater 2020 32 7019minus7027(45) Manthiram A Yu X Wang S Lithium battery chemistriesenabled by solid-state electrolytes Nat Rev Mater 2017 2 16103(46) Wang C Liang J Jiang M Li X Mukherjee S Adair KZheng M Zhao Y Zhao F Zhang S Li R Huang H Zhao SZhang L Lu S Singh C V Sun X Interface-assisted in-situgrowth of halide electrolytes eliminating interfacial challenges of all-inorganic solid-state batteries Nano Energy 2020 76 105015(47) Zhao C Liang J Li X Holmes N Wang C Wang JZhao F Li S Sun Q Yang X Liang J Lin X Li W Li RZhao S Huang H Zhang L Lu S Sun X Halide-based solid-state electrolyte as an interfacial modifier for high performance solid-state Li-O2 batteries Nano Energy 2020 75 105036

Accounts of Chemical Research pubsacsorgaccounts Article

httpsdxdoiorg101021acsaccounts0c00762Acc Chem Res 2021 54 1023minus1033

1033