recent advancements in li-ion conductors for all-solid ...rsliu/publications/2017/16.pdf ·...

Recent Advancements in Li-Ion Conductors forAll-Solid-State Li-Ion BatteriesYedukondalu Meesala,† Anirudha Jena,‡ Ho Chang,*,‡ and Ru-Shi Liu*,†,‡

†Department of Chemistry, National Taiwan University, Taipei 106, Taiwan‡Department of Mechanical Engineering and Graduate Institute of Manufacturing Technology, National Taipei University ofTechnology, Taipei 106, Taiwan

ABSTRACT: Inorganic solid lithium ion conductors are potential candidates as replacement for conventional organicelectrolytes for safety concerns. However, achieving a Li-ion conductivity comparable to that in existing liquid electrolytes(>1 mS cm−1) remains a challenge in solid-state electrolytes. One of the approaches for achieving a desirable conductivityis doping of various elements into the lattice framework. Our discussion on the structure and conductivity of crystalline Li-ion conductors includes description of NAtrium Super Ionic CONductor (NASICON)-type conductors, garnet-typeconductors, perovskite-type conductors, and Lithium Super Ionic CONductor (LISICON)-type conductors. Moreover, wediscuss various strategies currently used to enhance ionic conductivity, including theoretical approaches, ultimatelyoptimizing the electrolyte/electrode interface and improving cell performance.

The continuous depletion of fossil fuels, increasing oilprices, and the need to mitigate CO2 emissions havestimulated intensive research on alternative energy

technologies based on renewable and clean sources. Among therenewable energy sources, solar and wind energies are theprominent energy options; however, the intensity of sunlightand wind constantly changes with locations. Thus, alternativeresources must be stored. Rechargeable Li-ion batteries (LIBs)are one of the promising candidates for storage of alternativeenergy resources. Moreover, the LIBs have received tremen-dous attention because of their significant role in today’sconsumer electronics market, such as mobile electronic devices,laptops, digital cameras, and electric vehicles.1 Within thebattery class, LIBs have outperformed Pb-acid, Ni−Cd, andNi−metal hydride batteries in terms of specific energy andpower as represented in the Ragone plot (Figure 1).2 However,commercial LIBs contain toxic and flammable organic liquid aselectrolyte, resulting in serious safety issues. Moreover, growthof Li dendrites within the electrode materials due to suchorganic electrolytes leads to short-circuit and finally death ofLIBs. Additionally, the limited electrochemical window oforganic liquid electrolytes limits the choice of electrodes in

LIBs.3−5 To address these problems, replacement of organicliquid with inorganic solid electrolytes (SEs) will improve safetyand will ensure powering a wide range of electronic devices,from portable devices to heavy locomotives. Also, use of SEs inplace of organic electrolytes will suppress formation and growthof Li dendrites during battery cycles.6 The desirable criterionfor SEs for their implementation in real application is the ionicconductivity, which should exceed 10−3 S cm−1 at roomtemperature. Thermal, chemical, and mechanical stability willensure the compatibility of the SEs with electrode materials,and this will improve the SE−electrode interface stability andhence minimize the interfacial resistance. Besides, for theoptimal ionic conductivity of SEs, the size of the mobile ionshould fit in to the size of the migration channel (bottleneck),and the bottleneck should be smooth without very wide ornarrow sections.7 The structural parameters can be finely tunedby dopants and substitutions with ions with different ionic radiiinto the lattice structure for the suitable sizing of the bottleneck

Received: September 7, 2017Accepted: October 30, 2017Published: October 30, 2017

Review

http://pubs.acs.org/journal/aelccp

© 2017 American Chemical Society 2734 DOI: 10.1021/acsenergylett.7b00849ACS Energy Lett. 2017, 2, 2734−2751

http://pubs.acs.org/journal/aelccp

http://dx.doi.org/10.1021/acsenergylett.7b00849

in the Li-ion conduction pathway (Figure 2a). The dopant ionswith larger size would cause lattice expansion, thus increasingthe bottleneck size, reducing the activation energy, andimproving the conductivity.Among the most studied fast ion conductors, NASICON

(NAtrium Super Ionic CONductor)-type, garnet-type, perov-skite-type, and LISICON (Lithium Super Ionic CONductor)-type inorganic ceramic SEs have displayed ionic conductivitiesin the order of 10−3 S cm−1 at room temperature (Figure2a).8−10 Furthermore, these conductors exhibit outstandingstability within the electrochemical window of 0−9 V vs Li/Li+

as revealed by cyclic voltammetry measurements.11 Figure 2bshows the Arrhenius plots of the ionic conductivities of theselected inorganic SEs.9 Knauth12 published a review on thestructures and conductivities for each class of SE. Bachman etal.13 published a review on the ion-transport mechanisms andfundamental properties of solid-state electrolytes.NASICON-Type Li-Ion Conductors. The general molecular

formula of NASICON-type conductors is AxBy(PO4)3, where Aand B are monovalent and multivalent cations, respectively.The crystallographic structure of the NASICON NaA2

IV(PO4)3(A = Ge, Ti, Sn, Hf, Zr) was first reported in 1968.14 The termNASICON was coined by Hong and Goodenough for the

framework Na1+xZr2P3−xSixO12 (0 ≤ x ≤ 3) in 1971.15,16 TheseNa superionic conductors display thermal and chemical stabilityand high ion diffusion properties. NASICON materials formthree-dimensional network structures, with [ByP3O12]

−x co-valent skeletons built via corner sharing of BO6 octahedra andPO4 tetrahedra along the c-axis direction. NASICONs generallycrystallize in thermally stable rhombohedral structures with R3cspace group, although triclinic phases have been reported incompositions NaA2(PO4)3 with A = Sn, Zr, Hf.17 Inrhombohedral phases, mobile A cations in the skeletons aredistributed in two types of interstitial sites (M′ and M″) forcharge compensation. M′ sites are surrounded by six oxygenatoms (octahedral symmetry) and situated between twoadjacent [ByP3O12]

−x units along the c-axis, whereas the M″sites are surrounded by 10 oxygen atoms and symmetricallydistributed around the 3-fold axis of the structure. The M″ sitesare located between two columns of [ByP3O12]

−x units alongthe c-axis. The mobile A cations migrate through bottlenecksfrom one site to another, and the size of the bottlenecksdepends on the nature of the skeleton ions and concentrationof mobile ions at interstitial sites.Consequently, the ionic conductivity of NASICON materials

depends on the concentration of Li ions and composition ofthe NASICON framework. Among the members of theNASICON family, LiM2(PO4)3 (M = Zr, Hf, Sn, Ti and Ge),LiZr2(PO4)3 (LZP), LiTi2(PO4)3 (LTP), and LiGe2(PO4)3(LGP) are the most promising materials being investigated inrecent years.18,19 LiZr2(PO4)3 displays a complex polymor-phism, and the phase of the LZP system depends on thesynthetic parameters. LiZr2(PO4)3 synthesized at high temper-ature (>1100 °C) adopts a rhombohedral α phase (R3c) atroom temperature and undergoes a phase transition to triclinicα′ phase at low temperature (<55 °C).20 The triclinic α′ phasedisplayed a very low ionic conductivity of <10−9 S cm−1 at 30°C, whereas the rhombohedral α phase showed a conductivityof ∼10−5 S cm−1 at the same temperature.21 The conductivitiesof Ti and Ge-based materials are greatly enhanced bysubstituting smaller Ti4+ and Ge4+ with Zr4+.22 TheNASICON-type Li1+xMxTi2−x(PO4)3 (M = Al, Sc, Y, La) Li+

conducting SEs were reported by Aono et al. in 1989.23 Thepolycrystalline pure LTP sample, namely, Li1+xMxTi2−x(PO4)3(x = 0), showed a very poor conductivity of ∼10−7 S cm−1 atroom temperature, and this was attributed to the high porosity

Figure 1. Ragone plot for various rechargeable batteries showingsuperiority of the Li-based battery systems as the desired energy-storing devices.

Figure 2. (a) Total ionic conductivities of selected solid-state lithium-ion conductors at room temperature. (b) Ionic conductivities of selectedsolid-state lithium-ion conductors with variation of temperature.

ACS Energy Letters Review

DOI: 10.1021/acsenergylett.7b00849ACS Energy Lett. 2017, 2, 2734−2751

2735


in the pellet. Solid-state conductors with higher porosity showlower conductivities because of the longer distances traversedby lithium ions during migration between grains.24 Kwatek andNowinski25 reported that a maximum conductivity of 7.3 ×10−6 S cm−1 at room temperature for LiTi2(PO4)3 sample wasachieved by adding 8 wt % lithium iodide (LiI). LiI producessmall crystallites around the intergrain region of the LTP-LiIcomposite and facilitates Li-ion transport between the grains. Liet al. studied the phase relationships and electrical con-ductivities of the systems Li1+xMxGe2−x(PO4)3 (M = Al, Cr).The host compound LiGe2(PO4)3 showed very low con-ductivity of 1.5 × 10−5 S cm−1 at 300 °C.26 However, ionicconductivity was greatly enhanced upon partial substitution ofTi4+/Ge4+ ions by trivalent and divalent cations, such as Al, Ga,Sc, In, Y, La, Fe, Cr, Zn, and Ca in LTP/LGP systems.23,26 Theincreased conductivity resulted from increased mobile Li-ionconcentration in the framework and increased density of thesintered pellets. Aono et al.23 reported that conductivity issignificantly enhanced by partially substituting Ti with Al bysolid-state reaction, and a maximum conductivity of 7 × 10−4 Scm−1 at 25 °C was achieved for Li1.3Al0.3Ti1.7(PO4)3. Moreover,they reported that addition of polycrystalline Li3PO4 or Li3BO3lithium compounds as binders increases the density of thepellets and improves overall Li+ conductivities of theNASICON-based materials. These binders form intermediateglassy phases during sintering and act as fluxes for the growth ofLTP grains.27 The conductivity of Li1.5Ge1.5Al0.5P3O12 rapidlyincreases from 1.5 × 10−5 S cm−1 to 1.4 × 10−2 S cm−1 at 300°C by partially substituting Ge4+ with Al3+ in LiGe2P3O12systems.26 Depending on the synthetic conditions, sinteringparameters, and value of x in Li1+xTi/Ge2−xAlx(PO4)3, the totalionic conductivity varies between 10−8 and 10−3 S cm−1 atroom temperature. By using the solid-state reaction, Arbi eta l . 2 8 prepared Li1+ xAl xTi2− x(PO4)3 (LATP) andLi1+xAlxGe2−x(PO4)3 (LAGP) (x = 0, 0.2, 0.4), which showedhigh bulk conduct iv i t ie s o f 3 .4 × 10− 3 S cm− 1

( L i 1 . 2 T i 1 . 8 A l 0 . 2 ( P O 4 ) 3 ) a n d ∼ 1 0 − 4 S c m − 1

(Li1.2Ge1.8Al0.2(PO4)3) at room temperature, respectively. Itwas observed that the amount of Al incorporated in NASICONframework was less than the expected value, and impurity phaseAlPO4 was observed for samples with x ≥ 0.2 in both LATPand LAGP series. The segregation of Al at the surface in theLAGP samples was higher than that in the LATP samples, andthe low conductivities of the LAGP samples were due to thedepletion of Al in the LAGP samples, which could be explainedby the core−shell model.28 According to this model, someheterogeneous distribution of Al at the surface and inside LATPand LAGP particles has been detected in samples with low Alcontent (LATP-02 and LAGP-02 have different core and shellcontributions) (Figure 3, top). This heterogeneity may causethe formation of less conducting phase (AlPO4), which has astrong influence on the total (bulk + grain-boundary) andgrain-boundary resistances. For the sample LAGP (x = 0.2), athicker poor Al/Li shell contribution (yellow) formed suggeststhat the low-frequency “bulk” contribution (the intrinsicconductivity of the material at low frequency) displays alower conductivity than the high-frequency “bulk” contribution(core contribution). In the case of LATP (x = 0.2), the thickerrich Al/Li core particle contribution (blue) formed suggests thehigh-frequency “bulk” contribution displaying high conductivitydominates, but this is reversed in LAGP samples (Figure 3).29

The results suggest that Al incorporation to the LAGP (x =0.2) samples is more difficult than that to the LATP (x = 0.2)

samples. With increase of the Al concentration, the segregationof AlPO4 phase is favored in both LAGP and LATP series,which produces the partial elimination of the core−shellstructure, leading to a homogeneous composition and a highLi+ conductivity.28 Shang et al.30 reported a water-stable Li+

conducting SE system of Li1+x−yAlxNbyTi2−x−y(PO4)3, in whichTi ions a re par t i a l l y subs t i tu ted by Nb ions .Li1.3Al0.5Nb0.2Ti1.3(PO4)3 showed a high total conductivity of7.5 × 10−4 S cm−1 at 25 °C, and this value is higher than that ofLi1.3Al0.3Ti1.7(PO4)3. The Nb-substituted LATP was stable in asaturated LiOH aqueous solution with saturated LiCl.30,31

Kothari and Kanchan32 investigated the gallium-doped LATPsystem Li1.3Al0.3−xGaxTi1.7(PO4)3 (x = 0.01, 0.03, 0.05, and0.07) by replacing Al3+ with Ga3+; this system showed amaximum Li+ conductivity of ∼4 × 10−3 S cm−1 at 140 °C. Leoet al.33 synthesized a series of NASICON-based LGP andLAGP materials through simultaneous addition of MgO andLi2O additives as binders. They demonstrated that theseadditives act as binders rather than being incorporated into theparent system. The connectivity between the grains wasimproved by the additives, resulting in drastic reduction inporosity. Moreover, the conductivities of LGP and LAGP serieswere enhanced by the addition of binders and reached ∼10−7 Scm−1, 1.2 × 10−4 S cm−1, and 1.03 × 10−3 S cm−1 at 100 °C forLGP0 (LiGe2(PO4)3), LGP5 (LiGe2(PO4)3 + 0.5 MgO + 0.5Li2O), and LAGP2 (Li1.4Al0.4Ge1.6(PO4)3 + 0.2 MgO + 0.2Li2O), respectively.33 Chung and Kang34 preparedLi1.5Al0.5Ge1.5(PO4)3 (LAGP) by adding an excess amount oflithium (10−20% Li) through a solid-state reaction method.The grain boundary and specific grain boundary conductivities[the specific grain boundary conductivity was calculated usingthe fitted grain capacitance (Cg) and grain boundarycapacitance (Cgb) values] increased linearly with the additionof an excess amount of lithium. They explained that theadditional lithium can segregate into the surface of the grain orinto the grain boundary, resulting in increased amount of Li+

charge carriers at the grain boundary region (Figure 4).However, the Li segregation in the grain boundary can affectboth the morphologies of the particles and contacts between

Figure 3. Core−shell model used to explain observed electricalheterogeneity in NASICON samples. Yellow regions indicate poor-Al/Li phases, and blue regions indicate rich-Al/Li phases. Greenregions stand for homogeneous materials. Reproduced withpermission from ref 28. Copyright 2015 Elsevier.



2736


the grains, possibly leading to the formation of small gaps; as aresult, relative density of the sample decreased. Though therelative density of the sample decreases, it still maintains highgrain boundary conductivity due to excess charge carriers in thesample. A maximum conductivity of 1.9 × 10−4 S cm−1 at roomtemperature was achieved for LAGP20%L (LAGP with 20%Li)with a low relative density of 78%.34 However, the solid-statereaction method requires a considerable amount of time formultistep synthetic processes and a high sintering temperature.Glass−ceramics present a number of key advantages over

sintered polycrystalline ceramics in terms of flexibility,uniformity, dense microstructures, and negligible grainboundary effects. Glass−ceramics can be obtained throughthe melt−quench method, wherein glass plates were madethrough fast cooling of liquid melt and then were crystallizedthrough annealing process. Fu35 reported Li2O−Al2O3−TiO2−P2O5 glass−ceramic prepared via melt−quench method, whichexhibits conductivity of 1.3 × 10−3 S cm−1 at room temperature.The glass−ceramic LAGP, namely, Li1+xAlxGe2−x(PO4)3,produced through melt−quench method, was reported in1997; Li1.5Al0.5Ge1.5(PO4)3 showed a maximum conductivity of4 × 10−4 S cm−1 at room temperature.36 Chowdari et al.37

synthesized Li2O−Al2O3−MO2−P2O5 (M = Ti and Ge) glass−ceramics, which showed high ionic conductivities of 6.53 ×10−4 and 3.99 × 10−4 S cm−1 at 30 °C with low activationenergies of 0.32 and 0.31 eV for Li1.3Al0.3Ti1.7(PO4)3 andLi1.4Al0.4Ge1.6(PO4)3 glass−ceramics, respectively. The slightlylower conductivity of LATP, when compared with Fu’sprevious result, was attributed to the presence of a significantamount of AlPO4 impurity in the system. Thokchom andKumar38 reported high conductivities of water-durablecomposite membranes prepared by tape-casting of a Li2O−Al2O3−TiO2−P2O5 glass−ceramic system. The tape-casting ofLi2O−Al2O3−TiO2−P2O5 glass−ceramic showed long grainsize along with pores in the specimen. The porosity of the tape-casting membrane was about 22%. As a result of this porosity,

they concluded that the Li1.3Ti1.7Al0.3(PO4)3 glass−ceramic wasslightly degraded in tap water and showed a slight buckling afterimmersion for 65 days. The 14Li2O·9Al2O3·38TiO2·39P2O5glass−ceramic displayed a bulk conductivity of 1.7 × 10−3 Scm−1 and a grain boundary conductivity of 6.25 × 10−4 S cm−1

at room temperature.38 Mohaghegh et al.39 reported amaximum bulk conductivity of 1.38 × 10−3 S cm−1 at 25 °Cfor the Fe2O3-added glass−ceramic Li2O−TiO2−P2O5−x(Fe2O3) (x = 5, GC5). Addition of Fe2O3 into glasses isspeculated to loosen the framework structure and promotecrystallinity of glasses to enhance ionic conductivity. Kumar etal.40 investigated the ionic properties of LATP glass−ceramicsby adding Al2O3 and Ba0.6Sr0.4TiO3 (0.6BST). Unfortunately,the additives reduced the conductivities of the LATP glass−ceramics. They found that space charge and blocking effectscoexist in the system and compete with each other. The spacecharge effect enhances conductivity, whereas the blocking effectdiminishes ionic mobility.40 Xu et al.41 studied the influence ofL i 2O on NAS ICON-b a s e d g l a s s− c e r am i c s o fLi1.5Al0.5Ge1.5(PO4)3−xLi2O, which were synthesized usingthe melt−quench method. The additive Li2O acts as anucleating agent and enhances the crystallization of the as-prepared glasses during heat treatment. The density of theceramic plate was improved and the ionic conductivity wase n h a n c e d w i t h e x c e s s L i 2 O a d d i t i o n . T h eLi1.5Al0.5Ge1.5(PO4)3−0.05Li2O glass−ceramic displayed abulk conductivity of 1.18 × 10−3 S cm−1 and a totalconductivity of 7.25 × 10−4 S cm−1 at room temperature.The electrochemical measurements revealed that the glass−ceramic LAGP Li1.5Al0.5Ge1.5(PO4)3−0.05Li2O displayed goodstability against Li metal, and no other reactions were observedin the potential range of up to 6 V vs Li/Li+.41 Jadhav et al.42

investigated the conductivities of 0.05 wt % B2O3-added LAGPglass−ceramics. Given its low melting temperature, B2O3liquefies and segregates along the grain boundary region andfacilitates the reorganization of grains during crystallizationprocess. The grains showed better connection betweenthemselves, and the densities of the samples were improvedthrough B2O3 addition (Figure 5). The Li1.5Al0.5Ge1.5(PO4)3−0.05B2O3 glass−ceramic sample crystallized at 825 °C for 5 hdisplayed a maximum ionic conductivity of 6.7 × 10−4 S cm−1

at room temperature. The B2O3-added LAGP glass−ceramicsplaced in tetraethylene glycol dimethyl ether containing 1 Mlithium bis(trifluoromethane) sulfonimide (LiTFSI-TEGDME)showed good chemical stability, and morphology of the latterwas not significantly influenced. The ionic conductivity ofLi1.5Al0.5Ge1.5(PO4)3−0.05B2O3 slightly decreased from 6.7 ×10−4 to 2.4 × 10−4 S cm−1 at room temperature after thespecimen was immersed for 3 weeks in LiTFSI-TEGDMEsolvent. The Li−O2 cell containing Li1.5Al0.5Ge1.5(PO4)3−0.05B2O3 glass−ceramic achieved higher discharge voltage

Figure 4. Schematic of the characteristics of grain boundaries in theLAGP sample and LAGP20%L (LAGP with 20%Li) sample.

Figure 5. Formation of LAGP glass−ceramic with B2O3 addition.



2737


compared to that with the pristine Li1.5Al0.5Ge1.5(PO4)3,indicating a decreased polarization in the cell due to theincreased ionic conductivity by the addition of 0.05 wt %B2O3.

42 Zhu et al.43 synthesized the LAGP Li1.5Al0.5Ge1.5(PO4)3glass−ceramic by using the melt−quench method, and theystudied the effects of crystallization temperature and sinteringtime. They crystallized the preannealed glass sheets at 750, 775,800, 825, and 850 °C for 8 h and found that as thecrystallization temperature increases, the grain size of thespecimen increased rapidly. The glass−ceramic specimencrystallized at 825 °C for 8 h showed the highest conductivityof 2.25 × 10−3 S cm−1 at room temperature with an activationenergy of 0.29 eV. Thokchom and Kumar44 studied the effectsof crystallization parameters on the glass−ceramic LAGPLi1+xAlxGe2−x(PO4)3 (x = 0.5) synthesized via the melt−quench method. The glass−ceramic material crystallized at 825°C for 8 h displayed the highest total conductivity of 4.22 ×10−3 S cm−1 at room temperature. The superionic conductivityof the LAGP specimen was illustrated using the ion transportmechanism. The presence of AlPO4 impurity in the specimenforms the space charge complex AlPO4:Li

+ through adsorptionof an electroactive species (Li+) onto a dielectric (AlPO4)surface in the grain boundary region. The formed space chargecomplex, AlPO4:Li

+, affects the path of lithium-ion conduction.However, with the increase in crystallization temperature andsintering time, a greater amount of the AlPO4 impurity isformed, resulting in a blocking effect.44 Santagneli et al.45

prepared a tin-doped LAGP, Li1+xAlxSnyGe2−(x+y)(PO4)3,through isovalent substitution of Ge with its larger homologueSn using the melt−quench method. The tin-doped LAGPdisplayed a grain ionic conductivity that was slightly higher thanthat of the tin-free LAGP because of lattice expansionassociated with the substitution of Ge by its larger homologueSn. The glass−ceramic LASGP Li1.25Al0.25Sn0.25Ge1.5(PO4)3showed a high conductivity of 1.73 × 10−4 S cm−1 at roomtemperature and a low activation energy of 0.356 eV.45

One of the major problems in the melt−quench method isthat it requires an extremely high temperature for heattreatment, leading to lithium loss and secondary phaseformation in final products. Moreover, formation of a glasssheet is restricted to a few oxides (B2O3, SiO2, P2O5, andGeO2), and they must be present in a molar ratio ofapproximately 50% to get a glass formation.46 The use ofsolution-based preparation methods generally can overcomesome of these problems because of low-temperature syntheticprocesses. Xu et al.47 reported a high conductivity of 1.12 ×10−3 S cm−1 at 25 °C for Li1.4Al0.4Ti1.6(PO4)3 glass−ceramicprepared using the citric acid-assisted sol−gel method. Thehigh conductivity of the LATP pellet is attributed to thereduced particle size and the extremely high density (100% ofthe theoretical density) of the sample, which was obtainedusing the spark plasma sintering technique. Zhang et al.48,49

prepared water-stable Fe3+ and Cr3+ doped Li+ conducting SEsLi1+x+yAlxFeyTi2−x−y(PO4)3 and Li1.4AlxCr0.4−xTi1.6(PO4)3 (x = 0to 0.4), respectively, using a sol−gel method. Partialsubstitution of Al3+ by Fe3+/Cr3+ increased the ionicconductivity of LATP, and the highest total conductivities of1.1 × 10−3 and 1.06 × 10−3 S cm−1 at room temperature werea c h i e v e d f o r L i 1 . 4 F e 0 . 2 5 A l 0 . 1 5 T i 1 . 6 ( PO 4 ) 3 a n dLi1.4Al0.3Cr0.1Ti1.6(PO4)3, respectively. The chromium-dopedLATP sample was stable in the saturated LiCl aqueous solutionbut unstable in distilled water.48,49

Zhang et al.50 prepared NASICON-type Li1+xAlxGe2−x(PO4)3samples using a citric acid- and ethylene glycol-based sol−gelmethod. The LAGP Li1.4Al0.4Ge1.6(PO4)3 sample sintered at900 °C for 11 h demonstrated the highest conductivity of 1.22× 10−3 S cm−1 at room temperature. However, the conductivityof Li1.4Al0.4Ge1.6(PO4)3 sample immersed in distilled waterdecreased, whereas the conductivity was enhanced byimmersion of the sample in saturated LiCl aqueous solutionat 50 °C for 1 week. The conductivity enhancement in LiClsolution is due to the formation of a LiCl−H2O conductingphase at the grain boundaries.50 Moreover, they prepared awater-impermeable Li1.4Al0.4Ge1.6(PO4)3−5 wt % TiO2-epoxyhybrid sheet through tape-casting method using fine LAGPpowder. The ionic conductivity of the tape-cast sheet wasenhanced by incorporating nanosize TiO2 as sintering additive.The highest conductivity of the tape-cast Li1.4Al0.4Ge1.6(PO4)3−5 wt % TiO2-epoxy hybrid sheet was 8.37 × 10−4 S cm−1 atroom temperature.51 Mariappan et al .52 preparedLi1.5Al0.5Ge1.5(PO4)3 ceramic powder with various grain sizesusing a flame spray technique, which is also called thr flashcreation method. They found that the relative densities of theLAGP samples were approximately 83% at a low sinteringtemperature range (600 °C−800 °C) and sharply decreased ata high sintering temperature because of the formation ofimpurity phases (mainly GeO2 and AlPO4). The authorsanalyzed the impedance results with the framework of theclassical brick layer model. According to the classical brick layermodel (BLM), the permittivity of grains and of grainboundaries was assumed to be identical, and the ratio ofgrain capacitance to grain boundary capacitance was given byCg/Cgb = d/D, where d is the thickness of a grain boundary andD is the diameter of a grain. The specific grain boundaryconductivities were calculated by σgb = (1/Rgb)(t/A)(d/D).Though the activation energies of grain and grain boundaryresistance are identical, the specific grain boundary conductiv-ities were 1−2 orders of magnitude lower than the grainconductivities, suggesting that ion-blocking effects at the grainboundaries are not caused by high activation barriers but bygeometrical constriction effects. The Li1.5Al0.5Ge1.5(PO4)3sample sintered at 750 °C for 2 h displayed a maximum totalconductivity of 2 × 10−4 S cm−1 at room temperature.Lang et al.53 investigated the influence of the ionic migration

barriers for an interstitial Li-ion and a Li vacancy inLiTi2(PO4)3 on the ionic conductivity by means of densityfunctional theory (DFT). They explained that the additionalLi+ are introduced to Li1.2Al0.2Ti1.8(PO4)3 (LATP) by partiallysubstituting Ti4+ with Al3+, leading to structural changes andinfluencing the Li+ mobility. The additional Li+ preferablyoccupy energetically favored M1/2 sites (Wyckoff position 36f),which are between M1 and M2 sites. This occupation inducesthe nearest-neighboring Li+ at M1 sites to displace towardneighboring M1/2 positions and therefore opens the fast-conducting migration pathways within the NASICONstructure.53 Kang et al.54 performed first-principles DFTcalculations and compared the calculated conductivities withthe experimental measurements for LiGe2(PO4)3 (LGP) andLi1.5Al0.5Ge1.5(PO4)3 (LAGP) systems. They demonstrated theenhancement in conductivity of LAGP relative to that of LGPby 4 orders of magnitude by employing the DFT-based nudgedelastic band method. The excess Li ions induced by doping Al3+

for Ge4+ thermodynamically prefer to occupy 36f positionsrather than 18e sites when two Li ions located in two 6b sites.The higher conductivity of LAGP may arise not only from



2738


increased Li concentration in the system but also from thenewly created conduction paths, which substantially reduce theactivation energy (Figure 6).54

Along with poor conductivity, issues like compatibility ofinorganic solid-state electrolytes with lithium metal wereovercome by employed approaches. Introduction of anadditional interfacial layer between the electrode and electrolytecan facilitate easier passage of Li-ion reversibly. Though LATP-based materials reached the conductivity of 10−3 S cm−1 atroom temperature, which is the target for the SEs, theirapplicability as Li+ conductors is limited because of the

chemical instability toward Li metal due to the reduction ofTi4+. For their application in solid-state batteries, there musthave a protective layer placed in the interface between theLiTi2(PO4)3-based SE and Li metal. The LATP glass−ceramicplate Li2O−Al2O3−SiO2−P2O5−TiO2−GeO2 is coated with aprotective layer of lithium phosphorus oxynitride (LiPON) filmthrough sputter deposition with thickness of 1 μm, and thereactivity of the LATP plate with metallic Li was investigated byWest et al. in 2004.55 They found that the unpassivated SEplate that was in contact with Li metal showed unstablereactivity with Li metal and forms a dark nonmetallic layer,which was electrically nonconductive. By contrast, addition of alayer of thin LiPON film on the SE plate increases the chemicalstability and reduced the reactivity with Li metal (Figure 7a,b).The composite LiPON-coated SE plate displayed a con-ductivity of 1.0 × 10−4 S cm−1 at room temperature with anactivation energy of 0.36 eV and an electrochemical stabilitywindow of at least 0−5.0 V vs Li/Li+.Jadhav et al.56 investigated the interface stability of LAGP SE

by employing LiPON thin films deposited on B2O3-addedLAGP (B-LAGP) by RF-sputtering technique. The B-LAGPsubstrate temperature varied as 25, 100, 200, 300, and 400 °Cduring thin-film deposition. They showed that the compact thinamorphous LiPON layer could act as a protective interlayer bysuppressing dendrite growth and by reducing the possibility ofvigorous reaction with air moisture. The ionic conductivities ofthe LiPON/B-LAGP composites increase linearly with B-LAGPsubstrate temperature. The increased lithium-ion conductivitywas attributed to the incorporation of nitrogen into the filmand increment of the particle size in the LiPON film. Theaprotic Li−O2 cell with lithium electrode-protected config-uration consisting of LiPON/B-LAGP composite demonstrateda reasonable cycling stability of more than 52 cycles with alimited capacity of 1000 mAh g−1.56 Jung et al.57 preparedhybrid solid electrolyte thin films composed of LAGP andPEO-based polymer electrolyte with thickness of 40−60 μm tofabricate all-solid-state lithium batteries (Figure 8a,b). The all-solid-state Li/LiFePO4 cell assembled with flexible hybrid SEdelivered an initial discharge capacity of 137.6 mAh g−1 withgood cycling stability at 55 °C and good interfacial contactswith electrodes.57 Safanama et al.58 prepared hybrid inorganic−organic membranes by using Li1+xAlxGe2−x(PO4)3 ceramic,along with the polymeric SE composite PEO:poly(vinylidenefluoride) (PVDF):LiBF4; they demonstrated that Li anodes

Figure 6. (a) Crystal structure of Li1.5Al0.5Ge1.5(PO4)3 and threetypes of diffusion paths of Li ions in the LGP and LAGP modelsystems, which are calculated by established ab initio moleculardynamic simulations. Solid spheres indicate the initial config-urations, and the dashed line circles represent the finalconfigurations. (b) Specific sites and directions of each diffusionprocess are indicated by arrows. Reproduced with permission fromref 54. Copyright 2015 Elsevier.

The additional Li+ are introduced toLi1.2Al0.2Ti1.8(PO4)3 (LATP) by partiallysubstituting Ti4+ with Al3+, leading tostructural changes and influencing theLi+ mobility.

Figure 7. Evaporated Li films on (a) uncoated and (b) LiPON-coated solid electrolyte plates. Reproduced with permission from ref 55.Copyright 2004 Elsevier.



2739


were protected by the membranes in Li−O2 batteries. The totalconductivity of composite membrane was approximately 1 ×10−4 S cm−1 at room temperature. The LAGP hybridmembranes showed superior stability compared with the purepolymer membranes in aqueous solutions.58 Very recently, theydemonstrated that LAGP ceramic pellets act as anode-protecting membranes for rechargeable aqueous and hybridLi−air cells (Figure 9). Rechargeable Li−air batteries were

fabricated by sandwiching LAGP ceramic pellets betweenlithium anodes and catholyte solutions with various pH values;the Li−air batteries displayed very low overpotentials. The Li/LAGP interfacial resistance gradually increased because of theevaporation of catholyte in the cell, limiting the cycle life.Filling the anode chamber with electrolyte restores the contactbetween Li and LAGP. The hybrid Li−air cell with a currentdensity of 0.03 mA cm−2 was cycled for more than 60 h (59cycles). The charge−discharge overpotential was lower than 0.1V for the first 59 cycles.59 Liu et al.60 fabricated an all-solid-stateLi−air battery consisting of Li metal as anode and LAGP SEwith single-walled carbon nanotubes (SWCNTs/LAGP nano-

particles composite) as air electrode (Figure 10); the batteryshowed a large capacity of 2800 mAh g−1 in the first-cycledischarge.Garnet-Type Li-Ion Conductors. SEs crystallizing in garnet

structure with general formula LixM3(XO4)3, where M = La andX = Zr, Nb, Ta, have increasingly received attention because oftheir high lithium-ion conductivities and wide electrochemicalstability windows. The garnet-based Li-ion conducting electro-lytes (Li5La3M2O12, M = Ta, Nb) were first synthesized byHyooma and Hayashi61 and then optimized by Thangaduraiand Weppner.62,63 The garnet-type Li-ion conducting oxidesLi5La3M2O12, where M = Ta, Nb, displayed total Li-ionconductivities of ∼10−6 S cm−1 at room temperature. The Ta-contained garnet Li5La3Ta2O12 displayed better electrochemicalstability against the reaction with molten Li metal with a widervoltage window of >6 V vs Li/Li+ than the Nb-contained garnetLi5La3Nb2O12.

63 The Li-ion conductivity can be enhancedsignificantly by partial substitution of La3+ or M5+ sites withaliovalent cations. The Ba-substituted garnet Li6BaLa2Ta2O12displayed a high ionic conductivity of 4 × 10−5 S cm−1 at roomtemperature and an activation energy of 0.4 e V. The garnetsystem shows that the bulk and total conductivities are nearly ofthe same magnitude, indicating a low grain boundaryresistance.63 The conductivity was further enhanced uponsubstitution with Y or In. The In-substituted garnetLi5.5La3Nb1.75In0.25O12 prepared at 950 °C displayed anenhanced bulk ionic conductivity of 1.8 × 10−4 S cm−1 at 50°C with a low activation energy of 0.51 eV.64 The solid-stategarnet doped with yttrium and lithium (Li6.5La3Nb1.25Y0.75O12)displayed the highest bulk conductivity of 2.7 × 10−4 S cm−1 at25 °C.65 The yttrium-doped samples were chemically stable upto 400−600 °C after water treatment and were chemicallycompatible in the presence of the high-voltage cathodematerials Li2CoMn3O8 and Li2FeMn3O8.

65 Li7La3Zr2O12(LLZO) with garnet-related structure has been widely studiedin recent years because of its good ionic conductivity at roomtemperature and stability against lithium metal. In 2007,Murugan et al.66 first reported the synthesis of the garnet-typeLLZO through substitution of Zr4+ for M5+ in Li5La3M2O12;this LLZO showed an ionic conductivity of 2.44 × 10−4 S cm−1

Figure 8. (a) Photo of hybrid solid electrolyte film LAGP-70 (70% LAGP to PEO-based polymer).57 (b) Schematic presentation of solid-stateLi/LiFePO4 cell assembled with hybrid solid electrolyte and composite positive electrode. Reproduced with permission from ref 57.Copyright 2015 Electrochemical Society Inc.

Figure 9. Schematic representation of rechargeable hybrid Li−aircell. Reproduced with permission from ref 59. Copyright 2017Elsevier.



2740


at room temperature with a relative density of 92%. However,LLZO was found in two crystalline polymorphs: a low-temperature stable tetragonal structure with a space group ofI41/acd and a high-temperature stable cubic structure with aspace group of Ia3d. Figure 11 shows the tetragonal phase ofthe garnet-type solid conductor LLZO with its connectivitypattern in the lattice structure. The garnet-type tetragonalLLZO framework structure consists of edge-sharing dodecahe-dral La(1)O8 and La(2)O8 and octahedral ZrO6 units. In thetetragonal framework structure, Li atoms occupy three types ofcrystallographic sites: Li(1) atoms occupied the tetrahedral-8asites, while the other tetrahedral-16e was unoccupied; Li(2)and Li(3) atoms fully occupied octahedral-16f and octahedral-32g sites, respectively.Awaka et al.67 synthesized a tetragonal Li7La3Zr2O12 garnet

by using the flux method; this garnet showed a bulkconductivity of 1.63 × 10−6 S cm−1 and a grain-boundaryconductivity of 5.59 × 10−7 S cm−1 at room temperature.Wolfenstine et al.68 suggested that the use of a hot-presstechnique could further enhance the conductivity of tetragonalLi7La3Zr2O12. The hot-pressed tetragonal LLZO showed a highrelative density of ∼98% and exhibited a total ionic conductivityof ∼2.3 × 10−5 S cm−1 at room temperature. Choi et al.69

synthesized composite membranes composed of organicpolymer matrixes and inorganic solid electrolytes. Theconductivity of as-synthesized tetragonal LLZO pellet-typemembrane showed 6.2 × 10−7 S cm−1 at room temperature.The ionic conductivity was synergistically enhanced by theincorporation of the inorganic SE (LLZO) into the organicpolymer matrix (PEO-LiClO4). The composite membranecontaining 52.5% LLZO displayed an ionic conductivity of 4.42× 10−4 S cm−1 at 55 °C.69

The Li-ion conductivity of the cubic phase is higher than thatof the tetragonal phase by 2 orders of magnitude for the garnet-type electrolytes. Therefore, to stabilize the cubic phase ofgarnet, researchers have developed many methods, and theconductivity was improved to the order of ∼10−4 to 10−3 Scm−1 at room temperature (Figure 12). Awaka et al.70 studiedthe detailed crystal structure of cubic Li7La3Zr2O12 by single-crystal X-ray structure analysis. In the cubic phase, the Li(1)atoms occupied the tetrahedral-24d sites and the Li(2) atoms

Figure 10. Schematic diagram of an all-solid-state Li−air battery using Li anode, LAGP ceramic electrolyte, and an air electrode composed ofSWCNTs and solid electrolyte particles. Reproduced from ref 60. Copyright 2015 American Chemical Society.

Figure 11. Crystal structure of tetragonal Li7La3Zr2O12 with connectivity pattern of Td (8a and 16e) and Oh (16f and 32g) cages projected ontwo dimensions.

Figure 12. Parameters in obtaining the cubic phase of the garnetstructure.



2741


occupied the distorted octahedral-96h sites. The disorderedcubic structure and partial occupation of Li atoms at the Li(2)sites play key roles for fast ion conduction. In the cubic garnetstructure, Li ions could conduct from 24d sites to 96h sitesthrough a three-dimensional pathway. The basic unit of thearrangement is drawn as a loop built by the Li(1) and Li(2)sites (Figure 13a,b). Geiger et al.71 found that LLZOsynthesized in Pt crucible led to a tetragonal phase, whereasthat synthesized in alumina crucible became a mixture oftetragonal and cubic phases. The accidental incorporation of Alfrom crucible to the garnet should contribute to stabilize thecubic phase. Rangasamy et al.72 examined the effect of dopingof Al into the Li7La3Zr2O12 garnet structure by intentionallyincorporating Al during synthesis, and the cubic phase wasstabilized by optimizing Al amount to 0.24 in molecularformula. The hot-pressed cubic garnet Li6.24La3Zr2Al0.24O11.98displayed a conductivity of 4.0 × 10−4 S cm−1 at roomtemperature with a relative density of 98%. Xu et al.73 proposeda multistep sintering process for the synthesis of the LLZOLi7La3Zr2O12 doped with 0.2 mol % Al2O3 by holding thesample at 900 °C for 6 h and at 1100 °C for 6 h and finally at1200 °C for 12 h. By holding the sample at 900 °C for 6 h, theprecursors undergo partial decomposition to yield the mixtureof tetragonal and cubic phase; by continuously holding at 1100°C for 6 h, the formed tetragonal LLZO completely turns intocubic garnet; finally by holding at 1200 °C for 12 h, the relativedensity of the sample increases without formation of impuritiesin cubic phase. They reported that multistep sintering enhancedthe relative density of garnet and the ionic conductivity. The Al-LLZO pellet sintered via a multistep process exhibited a purecubic phase with a relative density of 94.25% and displayed anionic conductivity of 4.5 × 10−4 S cm−1 at room temperature.73

Moreover, 0.25 mol % Al doped Li7La3Zr2O12 samples weresynthesized in Pt and alumina crucibles, and ionic conductivityand air stability of the samples were studied.74 The Al-dopedLLZO garnet sintered in Pt crucible exhibited higher relativedensity, conductivity, and air stability than that sintered inalumina crucible. The sample sintered in alumina cruciblefound with excess Al content coming from alumina cruciblemay undergo Li loss during high-temperature sintering. TheLLZO garnet sintered in Pt crucible displayed an ionicconductivity of 4.48 × 10−4 S cm−1 at room temperaturewith a relative density of ∼96%.74 El-Shinawi et al.75 developeda hybrid sol−gel solid-state approach to synthesize Al-doped

Li7La3Zr2O12 with low Al-incorporation level (∼0.12 mol %).Obtaining dense cubic Li7La3Zr2O12-type phases, however,often required the prolonged calcination at elevated temper-atures of up to 1230 °C.66 However, using the hybrid sol−gelsolid-state approach, dense LLZO ceramic pellets obtained bymixing Al2O3 nanosheets with sol−gel processed LLZO solidprecursor and followed by single calcination step at lowertemperature, 1100 °C for 3 h (Figure 14). The minor

secondary phases of LiAlO2 and Li2ZrO3 formed in situ duringthe calcination process could help to stabilize the cubic phaseand increase the density of the garnet. The Al-dopedLi7La3Zr2O12 exhibited total conductivity of ∼3 × 10−4 Scm−1 at room temperature with an activation energy of 0.27eV.75 Dumon et al.76 synthesized strontium-doped cubicLi7La3Zr2O12 by adding 0.9−8.4 wt % Sr. The additive SrCO3acts as a sintering aid to stabilize the cubic phase, and itincreased the grain size and enhanced the conductivity. Thetotal ionic conductivity of 1.7 wt % Sr-doped LLZO wasapproximately 5 × 10−4 S cm−1 at room temperature with anactivation energy of approximately 0.31 eV.76 Very recently, ourgroup77 demonstrated that the conductivity of Al-doped LLZOsolid electrolyte can be enhanced by employing cheap andrelatively simple voltammetric treatment in an all-solid-state Li |LLZO | Li cell configuration. The Li deposition−dissolutionsignal has been observed in the voltammograms, which was

Figure 13. (a) Loop structure constructed by arrangement of Li atoms. The occupancy value, g, for each site is noted in parentheses. (b)Three-dimensional network structure of the arrangement of Li atoms in cubic Li7La3Zr2O12. Reproduced with permission from ref 70.Copyright 2011 The Chemical Society of Japan.

Figure 14. Schematic representation of stabilization and densifica-tion of fast-ion conducting LLZO by interaction with (a) thealumina crucible at sintering temperature 1230 °C and (b)nanostructured alumina at 1100 °C.



2742


supported by neutron powder diffraction measurements, whichshowed that lithium content in the lattice increased aftervoltammetric treatment. Furthermore, a local rearrangement ofO atoms was detected by X-ray photoelectron spectroscopy,which indicated reduction of oxygen defects. The enhancementin conductivity of LLZO was attributed to both the reductionof oxygen defects and the increase of lithium content.The garnet-type Li7La3Zr2O12 shows remarkably high ionic

conductivities upon doping of the supervalent Ta, Nb, and Yions at the Zr sites.78 Ohta et al.11 synthesized the Nb-dopedLLZO Li7−xLa3Zr2−xNbxO12 (x = 0 to 2) by partially doping Nbat Zr site; this Li6.75La3Zr1.75Nb0.25O12 displayed a high ionicconductivity of 8 × 10−4 S cm−1 at room temperature. Theincrease in conductivity upon Nb substitution is due to thestructural modification around the lithium sites and increase inLi+ mobility. By doping alkaline earth metal element Ca at theZr sites, Song et al.79 enhanced the conductivity of LLZO. Thedoped Ca ions form octahedral geometry at Zr sites and expandthe lattice, thus facilitating the migration of Li ions throughbott lenecks wi th increased s ize . The Ca-dopedLi7.1La3Zr1.95Ca0.05O12 exhibited conductivity of 5.2 × 10−4 Scm−1 at room temperature with an activation energy of 0.27 eVand was electrochemically stable up to 9 V vs Li/Li+.79 Allen etal.78 investigated the cubic garnet Li6.75La3Zr1.75Ta0.25O12(LLZTO) by doping Ta into the LLZO framework; LLZTOdisplayed a relatively high total ionic conductivity of 8.7 × 10−4

S cm−1 at room temperature with a low activation energy of0.22 eV. Li et al.80 reported that the ionic conductivity of cubicLi6.75La3Zr1.75Ta0.25O12 (LLZTO-Al) SE prepared in an oxygensintering atmosphere was approximately 7.4 × 10−4 S cm−1 atroom temperature. Buannic et al.81 enhanced the conductivityof LLZO by doping Ga and Sc at the Li and Zr sites,respectively. The first dopant, Ga3+ doped at Li sites, stabilizesthe cubic garnet phase, and simultaneous incorporation of thesecond dopant Sc3+ at the Zr4+ site increases the amount ofmob i l e L i i o n s . The dua l - s ub s t i t u t e d LLZO ,Li6.65Ga0.15La3Zr1.90Sc0.10O12, exhibited the highest conductivityof 1.8 × 10−3 S cm−1 at room temperature with an activationenergy of 0.29 eV.Gu et al.82 studied the influence of pentavalent and trivalent

doping on LLZO garnet by combining experimental results

with molecular dynamics simulations. They showed that theconductivity of cubic garnet was enhanced upon pentavalentdoping (Ta5+ or Nb5+) which increased the disorder andvacancy concentrations, whereas the trivalent doping of Al3+ orGa3+ on the Li sites was showed to be slightly less effective. Theimmobile Al3+ or Ga3+ preferentially occupies 24d sites whichblocks conduction pathway of Li ions. The doping of Ga3+ onthe La sites stabilized the cubic phase and had no effect on theLi+ concentration.82 Parameters in obtaining the cubic phase ofthe garnet structure are shown in Figure 12, including thesintering temperature and time, additives, sintering atmosphere,element doping, Li content (occupation sites), crucibles, etc.Bernstein et al.83 studied the driving force behind the phase

transition from low-conducting tetragonal phase to high-conducting cubic phase by using density functional theoryand a variable cell shape version of molecular dynamics (MD).DFT calculations showed that the Li sublattice in tetragonalphase was ordered which was either fully occupied orunoccupied with Li ions, whereas that in the cubic phase wasdisordered, which was partially filled with Li ions. In the cubicstructure, the tetrahedral 24d sites (Li1c) were fully occupiedand surrounded by four pairs of octahedral 96h sites (Li2c),which were partially occupied. The Li2−Li2 pair wasenergetically prohibited to occupy both sites at the same timebecause of the short distance between the pair, which causesstrong Coulomb repulsion between two Li ions. In thetetragonal structure, cubic Li1c sites converted to fully occupied8a sites (Li1t) and unoccupied 16e sites (Li2ut). The partiallyfilled cubic Li2c sites adapted into fully occupied 16f (Li2t) and32g (Li3t) sites in tetragonal structure (Figure 15). For MDsimulations, 0.25 vacancies per formula unit were applied at 600K.The vacancies were formed randomly in the ordered,

tetragonal unit cell by removing Li ions from Li1t and Li3t

sites. The system spontaneously transformed into cubic within5−10 ps of simulation time. At 15 ps, the system fluctuatedbetween cubic and tetragonal states and finally settled into acubic phase after 30 ps. When the tetragonal phase transformsinto the cubic phase, the ratios of lattice constants, ax/az anday/az (the lattice constants along x and y to that along z), dropfrom 1.04 to 0.98 (Figure 16, top panel). When this drop

Figure 15. Li sublattice in the cubic (left) and tetragonal (right) phases of LLZO. All Li positions are included, although in the cubic phase notall are occupied. The Li(1) atoms (8at and 24dc) are large gray (gold), Li(2) atoms (16ft and 96hc) are white, and Li(3) atoms (32gt) are darkgray. The cubic Li(1) positions that become vacant upon transition to the ordered tetragonal structure (16et) are indicated by small (gold)spheres. Reproduced with permission from ref 83. Copyright 2012 American Physical Society.



2743


occurs, there is a redistribution of Li ions in tetrahedral sites.The occupancies of tetrahedral Li1t (red) sites and tetrahedralLi2ut (blue) sites are initially near 1 and 0, respectively,consistent with experimental tetragonal structure (Figure 16,bottom panel). When the system transforms from tetragonal tocubic, there is a sharp drop in the occupancies of Li1t (red,bottom panel) and Li2t and Li3t sites (red, middle panel), whileLi2ut sites (blue, bottom panel) observed sharp increment intheir occupancies. However, the cubic Li1c (black, bottompanel) and Li2c (black, middle panel) sites did not experienceincrease or decrease in their occupancies during phasetransition, indicating that Li-ion redistribution took placebetween occupied Li1t sites and unoccupied Li2ut sites duringsimulation studies. So the calculated disorder in cubic structureoriginated not only from the change in the lattice constants butalso from filling the empty Li2ut octahedral sites.Shin et al.84 studied the enhancement in conductivity and

lithium-ion transport phenomena in cubic Li7La3Zr2O12 uponmultidoping by combining the experimental and densityfunctional theory calculations. The blocking effect of immobileAl3+ at 24d sites and 96h sites in cubic structure is shown inFigure 17. When Al occupies 24d sites, the relative energysignificantly decreased by 11.9 meV per atom, compared withthe condition that Al occupies 96h sites, suggesting that Aloccupation at 24d sites is more favored (Figure 18). It is widelybelieved that the immobile Al dopant in the Li sites may reducethe Li-ion diffusion by blocking the conduction paths of Li ionsin the garnet.78 The blocking effect of Al at 24d sites is moreserious in decreasing the ionic conductivity compared with theblocking effect of Al at 96h sites. By multidoping on LLZOgarnet with Al and Ta atoms, the 96h sites were stabilized andthe preference of Al occupation at 24d sites was reduced. Athigher levels of Ta doping, the 96h sites were further stabilized,indicating that the blocking of Al at 24d junction sites wasreduced by multidoping on LLZO. The ionic conductivity ofmultidoped cubic Li6.2Al0.2La3Zr1.8Ta0.2O12 exhibited about 6.14× 10−4 S cm−1 at room temperature with activation energy of0.29 eV.

Kotobuki et al.85 successfully used the lithium-ion conductinggarnet Li7La3Zr2O12 (LLZO) as electrolytes to fabricate all-solid-state rechargeable lithium batteries with Li metal. Thecyclic voltammogram of the Li/LLZO/Li cell shows that thedissolution and deposition reactions of lithium occurredreversibly without any reaction with LLZO, indicating stabilityof the LLZO against Li metal. A full cell with a LiCoO2/LLZO/Li configuration was successfully operated and demonstrated a

Figure 16. Evolution over time of structure and site occupationquantities for a sample system with nvac = 0.25 (vacancy number) atT = 600 K. Top: unit cell shape (ax/az, blue; ay/az, red) and volume(black). Middle: 96hc (black) and 16ft + 32gt (red) lattice siteoccupations. Bottom: 24dc (black), 8at (red), and 16et (blue withsymbols) lattice site occupations. Reproduced with permissionfrom ref 83. Copyright 2012 American Physical Society.

Figure 17. Blocking effect of Al in cubic LLZO. Simulated BVMdata is shown as blue shaded area. (a) 3D illustration of undopedcubic LLZO. Projected view from along the [100] direction of (b)undoped cubic LLZO, (c) Al-doped LLZO with Al in 24 d sites, and(d) Al-doped LLZO with Al in 96 h sites. The paths blocked by Alare marked with red “X” in panels c and d. The green arrows depictdiffusive motion of Li. Reproduced with permission from ref 84.Copyright 2016 Macmillan Publishers Limited.

Figure 18. Effect of additional Ta doping on Al site preference inAl-doped LLZO. The calculated energies of LLZO with Al in 24dand 96h site are described as △ and □, respectively. All calculatedenergies are rescaled by making the most stable configuration ineach doping group correspond to zero. Inset image describes twodifferent doping sites of Al. Reproduced with permission from ref84. Copyright 2016 Macmillan Publishers Limited.



2744


discharge capacity of 15 μAh cm−2. However, an irreversiblebehavior was observed at the first charge and discharge cyclebecause of an interfacial issue between LiCoO2 and LLZO.85

Kim et al.86 reported that the interfacial layer La2CoO4 formedbetween the Li+ conducting SE LLZO and LiCoO2 cathodebecause of mutual diffusion, suppressing the lithium insertion−extraction at the LLZO/LiCoO2 interface.86 Ohta et al.87

fabricated an all-solid-state lithium-ion battery consisting ofLiCoO2/Li6.75La3Zr1.75Nb0.25O12/Li configuration to investigateits electrochemical performance and charge-transfer resistance.The cell showed good cycle performance and displayed acapacity retention of approximately 98% after 100 charge−discharge cycles. The discharge capacity at the first cycle was129 mAh g−1 and stabilized at 127 mAh g−1 after the 100thcycle (theoretical capacity of LCO is 137 mAh g−1, whichcorresponds to 0.5 Li per CoO2).

87 Moreover, they introducedLi3BO3 as buffer layer between LiCoO2 cathode and Nb-dopedLLZO SE to increase the interfacial contact between them. Theobtained battery showed a good charge−discharge capacitywith a low interfacial resistance between electrodes and SE. Thecharge and discharge capacities at the first cycle were 100 and85 mAh g−1, respectively (theoretical capacity of LCO is 115mAh g−1, which corresponds to 0.42 Li per CoO2).

88 Du et al.89

fabricated all-solid-state Li batteries using Ta-doped LLZOcoated with a composite cathode which is a mixture ofPVDF:LiTFSI, Ketjen Black, and carbon-coated LiFePO4 onone side of the pellet and attached to a Li anode on the otherside. The first discharge capacity of the battery was 150 mAhg−1 at 0.05 C and showed a 93% capacity retention after 100cycles at 60 °C. The performance of the battery can be furtherimproved by increasing the temperature to 100 °C.89 Jin et al.90

studied the stability of Al-doped Li7La3Zr2O12 in air byfabricating a Li/Li7La3Zr2O12/Cu0.1V2O5 solid-state battery.The Al-doped Li7La3Zr2O12 garnet was found to be unstable inhumid air because of the formation of impurity phases likeLa(OH)3 and LiOH·H2O, indicating the decomposition of Al-doped Li7La3Zr2O12. The increase in pH of the LLZO powder/

water suspension resulted from the H+/Li+ exchange reactionbetween water and the LLZO electrolyte. The conductivity ofas-synthesized LLZO declined from 2.4 × 10−4 to 1.6 × 10−4 Scm−1 at room temperature after 1 week of exposure to air. Ahnet al.91 also examined the effect of moisture on the Li-ionconduction of Li7La3Zr2O12 garnet by composing an all-solid-state battery with a LiFePO4 (LFP) film as cathode. The severedegradation of Li-ion conduction was observed when thespecimen was exposed to humid air because of the formation ofsecondary phase (La(OH)3) at the grain boundary. The chargeand discharge capacities of the battery were observed to beextremely low (6 × 10−4 μAh cm−2) at room temperaturebecause of the reduced Li-ion conduction, and the capacity wasimproved with increase of temperature. Yan et al.92 fabricatedan all-solid-state battery composed of Li/Li7La3Zr2O12/LiFe-PO4 using an ultrathin LLZO solid electrolyte film. The solidelectrolyte film with thickness of several micrometers wasprepared by mixing LLZO nanoparticle slurry with appropriatesolvent, dispersant, adhesives, and surfactant without cold or

hot-pressing. The Li/LLZO/LFPO cell exhibited the firstdischarge capacity of 160.4 mAh g−1 at room temperature andretained capacity of 136.8 mAh g−1 after 100 cycles.Thangadurai et al.93 published a critical review on the garnet-type solid-state fast Li-ion conductors for Li batteries.Perovskite-Type Li-Ion Conductors. The perovskite-type

Li3xLa(2/3)−xTiO3 (LLTO, 0 < x < 0.16) Li-ion conductorshave received considerable attention because of their high bulkconductivities (10−3 S cm−1) at room temperature and theirexcellent electrochemical stability windows (8 V vs Li/Li+).

Though perovskite-type conductors have high bulk conductiv-ities, the total conductivities were lower (∼2 × 10−5 S cm−1 atroom temperature) because of higher grain boundaryresistances.94 In 1987, Belous et al.95 conducted the pioneeringstudy on the conduction behavior and on the stoichiometricrange of stable perovskite-type Li3xLa(2/3)−xTiO3. The perov-skite structure with a general formula of ABO3 exists in differentstable phases at different temperatures, that is, a high-temperature stable phase with cubic Pm3m symmetry (α-LLTO) and a low-temperature stable phase with tetragonal P4/mmm symmetry (β-LLTO). In the cubic structure (α-LLTO),the A-site La3+, Li+ and vacancies are equally distributed; bycontrast, the tetragonal structure had a double-perovskitestructure with La-rich layers and Li-rich layers along the c-axis.In the cubic structure, conduction of the Li ions happensthrough the A-sites, which are partially occupied by Li and Laions. Because of high conductivity, good electrochemicalstability, negligible electronic conductivity, and good stabilityin a wide temperature range of 4−1600 K, the LLTO-basedmaterials attracted considerable attention for application inelectrochemical devices, such as high-energy lithium-ionbatteries, electrochromic systems, supercapacitors, and pHsensors.96 However, the relatively low grain boundaryconductivities (<10−5 S cm−1) of the LLTO systems limittheir application in batteries. Inaguma et al.94 studied theconductivity measurements on cubic perovskite structure,Li0.34(1)La0.51(1)TiO2.94(2) whose cell parameter was 3.8710(2)Å. The total conductivity of the cubic perovskite was 2−4 ×10−5 S cm−1 at room temperature, but the value reached to 1 ×10−2 S cm−1 after 3 days, and the color of the sample changedfrom ivory to blue-black. They found that the enhancements inconductivity and color changes are from electronic conductiondue to the reduction of Ti4+ by Li metal, which causes theincrement of electron carriers. Mei et al.97 introduced anamorphous silicate layer to improve the grain boundaryconductivity of ceramic electrolyte. By the introduction of asilicate layer, the total ionic conductivity exceeded 1 × 10−4 Scm−1 at room temperature. Morata-Orrantia et al.98 achieved ahigh ionic conductivity of 2.95 × 10−3 S cm−1 forLa0.56Li0.36□0.08Ti0.97Al0.03O3 at room temperature by dopingAl into the perovskite. Abreu-Sepulveda et al.99 synthesizedchromium-substituted La(2/3)−xLi3xTi1−yCryO3 perovskites usinga conventional solid-state reaction and the Pechini processmethod. Neutron diffraction refinements were performed on

The increase in pH of the LLZO powder/water suspension resulted from the H+/Li+ exchange reaction between waterand the LLZO electrolyte.

Though perovskite-type conductorshave high bulk conductivities, the totalconductivities were lower (∼2 × 10−5

S cm−1 at room temperature) becauseof higher grain boundary resistances.



2745


four different models of chromium-substituted perovskite todetermine the site occupation of Li ions. The rietveldrefinement analysis showed that lithium most likely occupiesat the B-site (Ti site) with a composition of La[Li1/3(Ti,Cr)2/3]-O3 with minor cubic phase of Li1−xTi1+xO2. The chromium-substituted LLTO, La(2/3)−xLi3xTi0.9Cr0.1O3 showed an en-hancement in the conductivity (1.2 × 10−4 S cm−1 at roomtemperature) as well as good mechanical strength due to thewell-connected grains without porosity. Kwon et al.100 preparedLi3xLa2/3−xTiO3 solid electrolytes by the Pechini method atvarious sintering temperatures to investigate the effect ofmicrostructural modifications and change of Li-ion conductingbehavior. The crystal structures of tetragonal and orthorhombicLLTO perovskites with a general formula of ABO3 are shown inFigure 19a. The sample sintered at high temperature (high-T =1400 °C) showed larger grain size than that sintered at the lowtemperature (low-T = 1200 °C), indicating that when thesintering temperature is increased, the domain growth of theLLTO can be facilitated (Figure 19b,c). The LLTO sintered atlow temperature showed the total conductivity of 4.4 × 10−5 Scm−1 at room temperature, whereas the conductivity wassignificantly improved for the LLTO sintered at hightemperature (2.0 × 10−4 S cm−1). Moreover, the 30 mol %Li-excess LLTO sintered at 1400 °C has fine microstructurewith well-developed grain/domain boundaries, exhibiting atotal Li+ conductivity of 4.8 × 10−4 S cm−1 at room temperature(Figure 19d). The activation energy (Ea) for Li migration

across the domain boundary was reduced in the order of low-TLLTO (0.453 eV) > high-T LLTO (0.432 eV) > Li-excessLLTO (0.391 eV) (Figure 19e). Chen et al.101 reported theimproved ionic conductivity of Li0.35La0.55TiO3 by mixingLi7La3Zr2O12 (LLZO) sol into its precursor powder. The Zrdoped into the LLTO mainly prefers the grain boundaryregion. The grain boundary conductivity of LLTO was largelyincreased by the introduction of LLZO. The LLTO-basedelectrolyte mixed with 5 wt % LLZO exhibited a totalconductivity of 1.2 × 10−4 S cm−1 and a grain boundaryconductivity of 1.5 × 10−4 S cm−1 at room temperature.The perovskite-type Li0.35La0.55TiO3 (LLTO) with high ionic

conductivities of 10−3−10−4 S cm−1 at room temperature couldbe used as the electrolytes for all-solid-state batteries (ASSBs).However, the reduction of Ti4+ by Li metal limits theirapplicability in ASSBs. Kotobuki et al.102 fabricated an all-solid-state battery with the composition of LiCoO2/LLT/Li4Mn5O12

using a honeycomb-structured Li0.35La0.55TiO3, which hasmicrosized holes on both sides of the membrane. Theimpregnation of active materials LiCoO2 and Li4Mn5O12

particles mixed with respective precursor sols into thehoneycomb holes forms good active materials/solid electrolyteinterfaces, which reduces the internal resistance of the cell andthus improves the discharge capacity. Li et al.103 showed thecharge−discharge performance of an all-solid-state Li/LiPON/LLTO/LiCoO2 cell using LiPON/LLTO thin film as solidelectrolyte. Amorphous lithium lanthanum titanate (LLTO)

Figure 19. (a) Schematic of crystal structures of LLTOs; a tetragonal structure (left) and an orthorhombic structure (right). FE-SEMmicrographs of LLTOs sintered at (b) low-T and (c) high-T. (d) Comparison of Li+ conductivities along with schematics of the domainmicrostructures. (e) Arrhenius plots of the boundary conductivities for low-T LLTO, high-T LLTO, and Li-excess LLTO. Reproduced withpermission from ref 100. Copyright 2017 Royal Society of Chemistry.



2746


solid electrolyte thin films have been fabricated using e-beamevaporation. LiPON thin film was used as coating layer to avoidundesirable reaction between LLTO and lithium metal. Thecell exhibited a discharge capacity of about 50 μAh cm−2, andthe capacity degradation was about 0.5% per cycle after 100charge−discharge cycles at a current of 7 μA cm−2. Le et al.104

examined the stability and durability of Al-substituted LLTOceramic by immersing it in aqueous solution for one month.The Al-substituted lithium lanthanum titanate (A-LLTO)sintered at 1350 °C for 6 h exhibited ionic conductivity ofabout 3.17 × 10−4 S cm−1 at room temperature. The Li-LiCoO2and Li−O2 cells with LiPON/A-LLTO as electrolytes exhibitedgood cyclability, stability, and superior electrochemicalperformance after the electrolytes were immersed in an alkalineaqueous solution. The charge−discharge capacity of the Li-LiCoO2 cell maintained 59.3% of the initial capacity with aCoulombic efficiency of 98.3% after 100 cycles at 1C rate. Twocomprehensive reviews on the structure and properties ofLLTO were published by Bohnke96 and Stramare et al.105

LISICON-Type Li-Ion Conductors. A polycrystalline Li+

conducting SE LISICON Li2+2xZn1−xGeO4 was first reportedby Bruce and West in 1983.106 The LISICON framework issimilar to the γ-Li3PO4 crystal structure. AlthoughLi3.5Zn0.25GeO4 stoichiometry (x = 0.75) showed the highestlithium-ion conductivity (0.125 S cm−1) at high temperature(300 °C), its conductivity at room temperature is only 1 × 10−7

S cm−1. For LISICON compounds, the ionic conductivity tendsto decrease with time at low temperature because of theformation of a defective complex, namely, Li4GeO4, which trapsthe mobile lithium ions through the immobile sublattice.107

Attempts to restore their conductivities to original values byreannealing the samples were unsuccessful in Li2+2xZn1−xGeO4compounds because of aging problems. Furthermore, LISI-CON, Li14ZnGe4O16 is highly reactive with lithium andatmospheric CO2.

108 In the LISICON family, various materialsdisplaying the γ-Li3PO4 framework have been synthesized. Thegermanium-doped framework (Li3.6Ge0.6V0.4O4) displayed thehighest conductivity of approximately 4 × 10−5 S cm−1 at roomtemperature.109 Deng et al.110 reported enhanced conductivitiesof LISICON SEs by doping Si in Li4SiO4 system with P, Al, orGe. They have synthesized a series of compounds, includingLi3.75Si0.75P0.25O4, Li4.25Si0.75Al0.25O4, Li4Al0.33Si0.33P0.33O4, andLi4Al0.33Si0.17Ge0.17P0.33O4 with cationic substitution of Si in theparent LISICON-like Li4SiO4 framework. The ionic con-ductivity of Li4SiO4 is very low (σ473 K = 4 × 10−6 S cm−1);upon substitution of P or Al at Si sites, the conductivity wasincreased by 2 orders of magnitude to 1 × 10−4 S cm−1 forLi3.75Si0.75P0.25O4 and 2 × 10−4 S cm−1 for Li4.25Si0.75Al0.25O4 at473 K. A further increase in the ionic conductivity was observedfor “ternary” composition Li4Al0.33Si0.33P0.33O4, with σ473 K = 1× 10−3 S cm−1. Molecular dynamics simulations were used toexplain the enhanced conductivities and Li-ion diffusionmechanisms for substituted compositions. Depending on thetemperature, they have proposed three Li-ion diffusionmechanisms: (i) local oscillation at low temperature, (ii)isolated hopping at intermediate temperature, and (iii)superionic motion at high temperature (Figure 20). MDsimulations reveal that the Li-ion diffusion mechanism in eachcomposition was temperature-dependent. The Li-ion mobilitybecame higher when the diffusion mechanism transforms fromlocal oscillation to isolated hopping or from isolated hopping tosuperionic flow. The mixed polyanion substitution lowers thetemperature at which the transition to the superionic state with

high Li-ion conductivity occurs.110 Song et al.111 reported afacile strategy of substituting Cl for O to adjust bottleneck sizeand binding energy to enhance the conductivities andelectrochemical stability of LISICON systems. The latticeparameters of Cl-substituted Li10.5−xSi1.5P1.5ClxO12−x solidsolutions are larger than those of the pure Li3PO4, indicatingexpansion of bottlenecks and decrease of energy barriers, whichenhanced the ionic conductivities. Li10.42Si1.5P1.5Cl0.08O11.92 andLi10.42Ge1.5P1.5Cl0.08O11.92 exhibited ionic conductivities of about1.03 × 10−5 and 3.7 × 10−5 S cm−1 at room temperature,respectively, with electrochemical stability windows of up to 9V vs Li/Li+. All-solid-state batteries were assembled with0.3Li2MnO3·0.7LiMn1.5Ni0.5O4 as cathode and lithium metal asanode. The battery based on the Li10.42Si1.5P1.5Cl0.08O11.92showed initial charge and discharge capacities of 114.5 and114.1 mAh g−1, respectively, with Coulombic efficiencies of100%. The battery based on the Li10.42Ge1.5P1.5Cl0.08O11.92showed initial charge and discharge capacities of 330.1 and133.2 mAh g−1, respectively. The battery based onLi10.42Si1.5P1.5Cl0.08O11.92 showed better cycle performancethan that based on Li10.42Ge1.5P1.5Cl0.08O11.92.

111

Interfacial Stability. The use of highly conductive solid-stateinorganic electrolytes in all-solid-state lithium-ion batteries hasbeen extensively researched in recent years because of the safetyand good performance of this state-of-the-art batterytechnology. However, incorporation of these materials in all-solid-state batteries remains very challenging because of theirreactivity with the electrode materials at interfaces. Introductionof a buffer layer between the electrode and electrolyte reducesinterfacial resistance. Kato et al.112 introduced a thin Nb layer(∼10 nm) to interface between SE Li7La3Zr2O12 and cathodeLiCoO2 to reduce interfacial resistance. The Nb layersuppresses the growth of mutual diffusion layer at the interfaceand produces an amorphous Li−Nb−O material through an insitu process, which is reported to be Li+ conductive.Introduction of the Nb layer reduces the interfacial resistanceand improves the cycle stability and rate capability of charge−discharge reactions of a Li/Li7La3Zr2O12/LiCoO2 solid-statebattery.112 Zhou et al.113 constructed an all-solid-state batteryusing a NASICON Li1.3Al0.3Ti1.7(PO4)3 (LATP) ceramicmembrane sandwiched with a cross-linked polymer, poly-(ethylene glycol) methyl ether acrylate (CPMEA), on bothsides. Insertion of a polymer layer between the SE and anode

Figure 20. Three types of Li-ion diffusion mechanisms forsubstituted Li4SiO4 NASICONs. Li ions are shown in yellow;XO4 groups are shown in red. Reproduced from ref 110. Copyright2017 American Chemical Society.

Insertion of a polymer layer betweenthe SE and anode suppresses dendriteformation because of uniform Li+ fluxacross the polymer/lithium interface.



2747


suppresses dendrite formation because of uniform Li+ fluxacross the polymer/lithium interface. At the same time, thepolymer layer showed a better ability of wetting toward lithiummetal, and the ceramic layer was protected from contact withlithium metal. The all-solid-state battery, Li/CPMEA/LATP/CPMEA/LiFePO4 cell, delivered a stable capacity of 130 mAhg−1 after 100 cycles with Coulombic efficiency of 99.7%−100%.113 Garnet-type LLZO SEs could react with carbondioxide and moisture in ambient air to form Li-ion-insulatingLi2CO3 layers on the surfaces, resulting in large interfacialresistances. Moreover, they reported that by addition of 2 wt %LIF to the garnet Li6.5La3Zr1.5Ta0.5O12, stability of the garnetelectrolyte against moist air was effectively increased. Addition-ally, they fabricated an all-solid-state battery, Li/polymer/LLZT-2LiF/LiFePO4, which demonstrated high Coulombicefficiency and long cycle life with reduced interfacial resistance.Because of the low interfacial resistance, the low overpotentialfurther decreased to 0.2 V at 80 μA cm−2.114

Summary and Future Outlook. In conclusion, LIBs with liquidelectrolytes are still providing a large part of the current energydemand. However, serious safety concerns arise because of theuse of liquid electrolytes. Leakage of electrolytes and growth ofLi dendrites caused by contact of anodes with liquidelectrolytes greatly hinder the performance of LIBs. Devicesoperated at elevated temperatures also suffer from electrolyteevaporation when liquid electrolytes are used. Hence, inorganicSEs with room-temperature conductivities >10−3 S cm−1 andwide electrochemical stability windows (6 V vs Li/Li+) havebeen intensively studied to develop all-solid-state batteries thatcan be used for applications with a wide temperature range.Stability of SE/electrode interface is a challenge for facilepassage of Li ions across the interface. From this perspective,we have discussed recent published research activities insynthetic methods and possible dopants and respective effectson the crystal lattices of NASICON-type, garnet-type, perov-skite-type, and LISICON-type SEs. Doping with isovalent oraliovalent elements and addition of additives into the parentstructural frameworks enhance the ionic conductivities of SEsbecause of the framework expansion and increase of mobile Liions. We have also discussed the interfacial reactions betweenSEs and electrodes. For example, the solid Li-ion conductorLiTi2(PO4)3-based material, which exhibited high ionicconductivity, was chemically unstable with Li metal becauseof the reduction of Ti4+. Introduction of a protective layer(LiPON film) between SE and Li metal improved the stability.Good interfacial contacts between the solid electrolytes and theelectrodes can be achieved given the nondeformable nature ofthe inorganic ceramic SEs. However, further studies are neededto better understand the interfacial processes between solidelectrolytes and electrodes.

■ AUTHOR INFORMATION

Corresponding Authors*E-mail: [email protected]. Fax: +886-2-33668671. Tel: +886-2-33661169.*E-mail: [email protected]. Fax: +886-2-27712171. Tel:+886-2-27317191.

ORCID

Ru-Shi Liu: 0000-0002-1291-9052NotesThe authors declare no competing financial interest.

Biographies

Yedukondalu Meesala obtained a Ph.D. in chemistry from IndianInstitute of Technology, Bombay, India (2011). He is currently apostdoctoral fellow in Prof. Ru-Shi Liu’s group at the Department ofChemistry, National Taiwan University, Taiwan. His current researchinterest focuses on transport properties of solid electrolyte materialsfor Li-ion batteries.

Anirudha Jena received a Ph.D. degree from Indian Institute ofScience, Bengaluru, India (2013). He is currently working as aResearch Assistant Professor in the Graduate Institute of Manufactur-ing Technology, National Taipei University of Technology. Hiscurrent research is focused on the development of materials for all-solid-state batteries.

Ho Chang received a Ph.D. degree from the National TaipeiUniversity of Technology (NTUT) (2004). He is currently workingas a professor at the Graduate Institute of Manufacturing Technology,NTUT. His current research is focused on dynamic frication behaviorsof pneumatic cylinders and fabrication of thermoelectric materials.

Ru-Shi Liu obtained two Ph.D. degrees in chemistry, one from theNational Tsing Hua University in 1990 and another from theUniversity of Cambridge in 1992. He is currently working as aprofessor at the Department of Chemistry, National TaiwanUniversity. His research interest is materials chemistry. He is anauthor and coauthor of more than 530 publications in internationalscientific journals.

■ ACKNOWLEDGMENTSThis study was financially supported by the Ministry of Scienceand Technology, Taiwan (Contract No. MOST 104-2113-M-002-012-MY3).

■ REFERENCES(1) Vetter, J.; Novak, P.; Wagner, M. R.; Veit, C.; Moller, K. C.;Besenhard, J. O.; Winter, M.; Wohlfahrt-Mehrens, M.; Vogler, C.;Hammouche, A. Ageing Mechanisms in Lithium-Ion Batteries. J. PowerSources 2005, 147, 269−281.(2) Tarascon, J.-M.; Armand, M. Issues and Challenges FacingRechargeable Lithium Batteries. Nature 2001, 414, 359−367.(3) Gachot, G.; Grugeon, S.; Armand, M.; Pilard, S.; Guenot, P.;Tarascon, J.-M.; Laruelle, S. Deciphering the Multi-Step DegradationMechanisms of Carbonate-Based Electrolyte in Li Batteries. J. PowerSources 2008, 178, 409−421.(4) Xu, W.; Wang, J.; Ding, F.; Chen, X.; Nasybulin, E.; Zhang, Y.;Zhang, J.-G. Lithium Metal Anodes for Rechargeable Batteries. EnergyEnviron. Sci. 2014, 7, 513−537.(5) Cheng, X. B.; Zhang, R.; Zhao, C. Z.; Wei, F.; Zhang, J. G.;Zhang, Q. A Review of Solid Electrolyte Interphases on Lithium MetalAnode. Adv. Sci. 2016, 3, 1500213.(6) Arico, A. S.; Bruce, P.; Scrosati, B.; Tarascon, J.-M.; VanSchalkwijk, W. Nanostructured Materials for Advanced EnergyConversion and Storage Devices. Nat. Mater. 2005, 4, 366−377.(7) Skaarup, S. Solid Electrolytes. J. Mater. Educ. 1984, 6, 667−730.(8) Thokchom, J. S.; Gupta, N.; Kumar, B. Superionic Conductivityin a Lithium Aluminum Germanium Phosphate Glass-Ceramic. J.Electrochem. Soc. 2008, 155, A915−A920.(9) Kamaya, N.; Homma, K.; Yamakawa, Y.; Hirayama, M.; Kanno,R.; Yonemura, M.; Kamiyama, T.; Kato, Y.; Hama, S.; Kawamoto, K.;et al. A Lithium Superionic Conductor. Nat. Mater. 2011, 10, 682.(10) Li, Y.; Han, J.-T.; Wang, C.-A.; Xie, H.; Goodenough, J. B.Optimizing Li+ Conductivity in a Garnet Framework. J. Mater. Chem.2012, 22, 15357−15361.(11) Ohta, S.; Kobayashi, T.; Asaoka, T. High Lithium IonicConductivity in the Garnet-Type Oxide Li7‑xLa3(Zr2‑x,Nbx)O12 (x = 0−2). J. Power Sources 2011, 196, 3342−3345.



2748

mailto:[email protected]

mailto:[email protected]

http://orcid.org/0000-0002-1291-9052


(12) Knauth, P. Inorganic Solid Li Ion Conductors: An Overview.Solid State Ionics 2009, 180, 911−916.(13) Bachman, J. C.; Muy, S.; Grimaud, A.; Chang, H. H.; Pour, N.;Lux, S. F.; Paschos, O.; Maglia, F.; Lupart, S.; Lamp, P.; et al. InorganicSolid-State Electrolytes for Lithium Batteries: Mechanisms andProperties Governing Ion Conduction. Chem. Rev. 2016, 116, 140−162.(14) Hagman, L.-O.; Kierkegaard, P.; Karvonen, P.; Virtanen, A. I.;Paasivirta, J. The Crystal Structure of NaMe2

IV(PO4)3; MeIV= Ge, Ti,Zr. Acta Chem. Scand. 1968, 22, 1822−1832.(15) Hong, H.-P. Crystal Structures and Crystal Chemistry in theSystem Na1+xZr2SixP3‑xO12. Mater. Res. Bull. 1976, 11, 173−182.(16) Goodenough, J.; Hong, H.-P.; Kafalas, J. Fast Na+-Ion Transportin Skeleton Structures. Mater. Res. Bull. 1976, 11, 203−220.(17) Catti, M.; Stramare, S.; Ibberson, R. Lithium Location inNASICON-Type Li+ Conductors by Neutron Diffraction. I. Triclinicα′-LiZr2(PO4)3. Solid State Ionics 1999, 123, 173−180.(18) Xu, H.; Wang, S.; Wilson, H.; Zhao, F.; Manthiram, A. Y-DopedNASICON-type LiZr2(PO4)3 Solid Electrolytes for Lithium-MetalBatteries. Chem. Mater. 2017, 29, 7206−7212.(19) Xiong, L.; Ren, Z.; Xu, Y.; Mao, S.; Lei, P.; Sun, M. LiF AssistedSynthesis of LiTi2(PO4)3 Solid Electrolyte with Enhanced IonicConductivity. Solid State Ionics 2017, 309, 22−26.(20) Sljukic, M.; Matkovic, B.; Prodic, B.; Scavnicar, S. Preparationand Crystallographic Data of Phosphates with Common FormulaMiM42(PO4)3 (Mi = Li Na K Rb Cs-M4 = Zr Hf). Croat. Chem. Acta.1967, 39, 145.(21) Sudreau, F.; Petit, D.; Boilot, J. Dimorphism, Phase Transitions,and Transport Properties in LiZr2(PO4)3. J. Solid State Chem. 1989, 83,78−90.(22) Subramanian, M.; Subramanian, R.; Clearfield, A. Lithium IonConductors in the System AB(IV)2(PO4)3 (B = Ti, Zr and Hf). SolidState Ionics 1986, 18, 562−569.(23) Aono, H.; Sugimoto, E.; Sadaaka, Y.; Imanaka, N.; Adachi, G.Ionic Conductivity of the Lithium Titanium Phosphate(Li1+xAlxTi2‑x(PO4)3, M = Al, Sc, Y, and La) Systems. J. Electrochem.Soc. 1989, 136, 590−591.(24) Francisco, B. E.; Stoldt, C. R.; M’Peko, J.-C. Lithium-IonTrapping from Local Structural Distortions in Sodium Super IonicConductor (NASICON) Electrolytes. Chem. Mater. 2014, 26, 4741−4749.(25) Kwatek, K.; Nowin ski, J. L. Electrical Properties of LiTi2(PO4)3and Li1.3Al0.3Ti1.7(PO4)3 Solid Electrolytes Containing Ionic Liquid.Solid State Ionics 2017, 302, 54−60.(26) Li, S.-c.; Cai, J.-y.; Lin, Z.-x. Phase Relationships and ElectricalConductivity of Li1+xGe2‑xAlxP3O12 and Li1+xGe2‑xCrxP3O12 Systems.Solid State Ionics 1988, 28, 1265−1270.(27) Aono, H.; Sugimoto, E.; Sadaoka, Y.; Imanaka, N.; Adachi, G.-y.Electrical Property and Sinterability of LiTi2(PO4)3 Mixed withLithium Salt (Li3PO4 or Li3BO3). Solid State Ionics 1991, 47, 257−264.(28) Arbi, K.; Bucheli, W.; Jimenez, R.; Sanz, J. High Lithium IonConduc t ing So l id E lec t ro l y t e s Based on NASICONLi1+xAlxM2‑x(PO4)3 Materials (M = Ti, Ge and 0 ≤ x ≤ 0.5). J. Eur.Ceram. Soc. 2015, 35, 1477−1484.(29) Bucheli, W.; Jimenez, R.; Sanz, J.; Varez, A. The Log (σ) Vs. Log(ω) Derivative Plot Used to Analyze the Ac Conductivity. Applicationto Fast Li+ Ion Conductors with Perovskite Structure. Solid State Ionics2012, 227, 113−118.(30) Shang, X.; Nemori, H.; Mitsuoka, S.; Mastuda, Y.; Takeda, Y.;Yamamoto, O.; Imanishi, N. High Lithium-Ion ConductingNASICON-Type Li1+x‑yAlxNbyTi2‑x‑y(PO4)3 Solid Electrolytes. SolidState Ionics 2016, 297, 43−48.(31) Shimonishi, Y.; Zhang, T.; Imanishi, N.; Im, D.; Lee, D. J.;Hirano, A.; Takeda, Y.; Yamamoto, O.; Sammes, N. A Study onLithium/Air Secondary BatteriesStability of the NASICON-TypeLithium Ion Conducting Solid Electrolyte in Alkaline AqueousSolutions. J. Power Sources 2011, 196, 5128−5132.

(32) Kothari, D. H.; Kanchan, D. K. Study of Study of ElectricalProperties of Gallium-Doped Lithium Titanium Aluminum PhosphateCompounds. Ionics 2015, 21, 1253−1259.(33) Leo, C.; Rao, G. S.; Chowdari, B. Effect of MgO addition on theionic conductivity of LiGe2(PO4)3 ceramics. Solid State Ionics 2003,159, 357−367.(34) Chung, H.; Kang, B. Increase in Grain Boundary IonicConductivity of Li1.5Al0.5Ge1.5(PO4)3 by Adding Excess Lithium.Solid State Ionics 2014, 263, 125−130.(35) Fu, J. Superionic Conductivity of Glass-Ceramics in the SystemLi2O-Al2O3-TiO2-P2O5. Solid State Ionics 1997, 96, 195−200.(36) Fu, J. Fast Li+ Ion Conducting Glass-Ceramics in the SystemLi2O-Al2O3-GeO2-P2O5. Solid State Ionics 1997, 104, 191−194.(37) Chowdari, B.; Rao, G. S.; Lee, G. XPS and Ionic ConductivityStudies on Li2O-Al2O3-(TiO2 or GeO2)-P2O5 Glass-Ceramics. SolidState Ionics 2000, 136, 1067−1075.(38) Thokchom, J. S.; Kumar, B. Water Durable Lithium IonConducting Composite Membranes from the Li2O-Al2O3-TiO2-P2O5

Glass-Ceramic. J. Electrochem. Soc. 2007, 154, A331−A336.(39) Mohaghegh, E.; Nemati, A.; Eftekhari Yekta, B.; Banijamali, S.Effects of Fe2O3 Content on Ionic Conductivity of Li2O-TiO2-P2O5

Glasses and Glass-Ceramics. Mater. Chem. Phys. 2017, 190, 8−16.(40) Kumar, B.; Nellutla, S.; Thokchom, J. S.; Chen, C. IonicConduction through Heterogeneous Solids: Delineation of theBlocking and Space Charge Effects. J. Power Sources 2006, 160,1329−1335.(41) Xu, X.; Wen, Z.; Wu, X.; Yang, X.; Gu, Z. Lithium Ion-Conducting Glass-Ceramics of Li1.5Al0.5Ge1.5(PO4)3-xLi2O (x = 0.0−0.20) with Good Electrical and Electrochemical Properties. J. Am.Ceram. Soc. 2007, 90, 2802−2806.(42) Jadhav, H. S.; Kalubarme, R. S.; Jang, S. Y.; Jung, K. N.; Shin, K.H.; Park, C. J. B2O3-Added Lithium Aluminium GermaniumPhosphate Solid Electrolyte for Li-O2 Rechargeable Batteries. DaltonTrans. 2014, 43, 11723−11727.(43) Zhu, Y.; Zhang, Y.; Lu, L. Influence of CrystallizationTemperature on Ionic Conductivity of Lithium Aluminum Germa-nium Phosphate Glass-Ceramic. J. Power Sources 2015, 290, 123−129.(44) Thokchom, J. S.; Kumar, B. The Effects of CrystallizationParameters on the Ionic Conductivity of a Lithium AluminumGermanium Phosphate Glass-Ceramic. J. Power Sources 2010, 195,2870−2876.(45) Santagneli, S. H.; Baldacim, H. V. A.; Ribeiro, S. J. L.; Kundu, S.;Rodrigues, A. C. M.; Doerenkamp, C.; Eckert, H. Preparation,Structural Characterization, and Electrical Conductivity of Highly Ion-Conducting Glasses and Glass Ceramics in the SystemLi1+xAlxSnyGe2‑(x+y)(PO4)3. J. Phys. Chem. C 2016, 120, 14556−14567.(46) McMillan, P. W. The Crystallisation of Glasses. J. Non-Cryst.Solids 1982, 52, 67−76.(47) Xu, X.; Wen, Z.; Yang, X.; Chen, L. Dense Nanostructured SolidElectrolyte with High Li-Ion Conductivity by Spark Plasma SinteringTechnique. Mater. Res. Bull. 2008, 43, 2334−2341.(48) Zhang, P.; Matsui, M.; Takeda, Y.; Yamamoto, O.; Imanishi, N.Water-Stable Lithium Ion Conducting Solid Electrolyte of Iron andAluminum Doped NASICON-Type LiTi2(PO4)3. Solid State Ionics2014, 263, 27−32.(49) Zhang, P.; Wang, H.; Si, Q.; Matsui, M.; Takeda, Y.; Yamamoto,O.; Imanishi, N. High Lithium Ion Conductivity Solid Electrolyte ofChromium and Aluminum Co-Doped NASICON-Type LiTi2(PO4)3.Solid State Ionics 2015, 272, 101−106.(50) Zhang, M.; Takahashi, K.; Imanishi, N.; Takeda, Y.; Yamamoto,O.; Chi, B.; Pu, J.; Li, J. Preparation and Electrochemical Properties ofLi1+xAlxGe2‑x(PO4)3 Synthesized by a Sol-Gel Method. J. Electrochem.Soc. 2012, 159, A1114−A1119.(51) Zhang, M.; Takahashi, K.; Uechi, I.; Takeda, Y.; Yamamoto, O.;Im, D.; Lee, D.-J.; Chi, B.; Pu, J.; Li, J.; et al. Water-Stable LithiumAnode with Li1.4Al0.4Ge1.6(PO4)3-TiO2 Sheet Prepared by TapeCasting Method for Lithium-Air Batteries. J. Power Sources 2013,235, 117−121.



2749


(52) Mariappan, C. R.; Yada, C.; Rosciano, F.; Roling, B. Correlationbetween Micro-Structural Properties and Ionic Conductivity ofLi1.5Al0.5Ge1.5(PO4)3 Ceramics. J. Power Sources 2011, 196, 6456−6464.(53) Lang, B.; Ziebarth, B.; Elsasser, C. Lithium Ion Conduction inLiTi2(PO4)3 and Related Compounds Based on the NASICONStructure: A First-Principles Study. Chem. Mater. 2015, 27, 5040−5048.(54) Kang, J.; Chung, H.; Doh, C.; Kang, B.; Han, B. IntegratedStudy of First Principles Calculations and Experimental Measurementsfor Li-Ionic Conductivity in Al-Doped Solid-State LiGe2(PO4)3Electrolyte. J. Power Sources 2015, 293, 11−16.(55) West, W. C.; Whitacre, J. F.; Lim, J. R. Chemical StabilityEnhancement of Lithium Conducting Solid Electrolyte Plates UsingSputtered LiPON Thin Films. J. Power Sources 2004, 126, 134−138.(56) Jadhav, H. S.; Kalubarme, R. S.; Jadhav, A. H.; Seo, J. G. HighlyStable Bilayer of LiPON and B2O3 added Li1.5Al0.5Ge1.5(PO4)3 SolidElectrolytes for Non-Aqueous Rechargeable Li-O2 Batteries. Electro-chim. Acta 2016, 199, 126−132.(57) Jung, Y.-C.; Lee, S.-M.; Choi, J.-H.; Jang, S. S.; Kim, D.-W. AllSolid-State Lithium Batteries Assembled with Hybrid Solid Electro-lytes. J. Electrochem. Soc. 2015, 162, A704−A710.(58) Safanama, D.; Damiano, D.; Rao, R. P.; Adams, S. LithiumConducting Solid Electrolyte Li1+xAlxGe2‑x(PO4)3 Membrane forAqueous Lithium Air Battery. Solid State Ionics 2014, 262, 211−215.(59) Safanama, D.; Adams, S. High Efficiency Aqueous and HybridLithium-Air Batteries Enabled by Li1.5Al0.5Ge1.5(PO4)3 CeramicAnode-Protecting Membranes. J. Power Sources 2017, 340, 294−301.(60) Liu, Y.; Li, B.; Kitaura, H.; Zhang, X.; Han, M.; He, P.; Zhou, H.Fabrication and Performance of All-Solid-State Li-Air Battery withSWCNTs/LAGP Cathode. ACS Appl. Mater. Interfaces 2015, 7,17307−17310.(61) Hyooma, H.; Hayashi, K. Crystal structures of La3Li5M2O12 (M= Nb, Ta). Mater. Res. Bull. 1988, 23, 1399−1407.(62) Thangadurai, V.; Weppner, W. Li6ALa2Ta2O12 (A = Sr, Ba):Novel Garnet-Like Oxides for Fast Lithium Ion Conduction. Adv.Funct. Mater. 2005, 15, 107−112.(63) Thangadurai, V.; Kaack, H.; Weppner, W. J. F. Novel FastLithium Ion Conduction in Garnet-Type Li5La3M2O12 (M = Nb, Ta).J. Am. Ceram. Soc. 2003, 86, 437−440.(64) Thangadurai, V.; Weppner, W. Effect of Sintering on the IonicConductivity of Garnet-Related Structure Li5La3Nb2O12 and In- andK-Doped Li5La3Nb2O12. J. Solid State Chem. 2006, 179, 974−984.(65) Narayanan, S.; Ramezanipour, F.; Thangadurai, V. Enhancing LiIon Conductivity of Garnet-Type Li5La3Nb2O12 by Y- and Li-Codoping: Synthesis, Structure, Chemical Stability, and TransportProperties. J. Phys. Chem. C 2012, 116, 20154−20162.(66) Murugan, R.; Thangadurai, V.; Weppner, W. Fast Lithium IonConduction in Garnet-Type Li7La3Zr2O12. Angew. Chem., Int. Ed.2007, 46, 7778−7781.(67) Awaka, J.; Kijima, N.; Hayakawa, H.; Akimoto, J. Synthesis andStructure Analysis of Tetragonal Li7La3Zr2O12 with the Garnet-RelatedType Structure. J. Solid State Chem. 2009, 182, 2046−2052.(68) Wolfenstine, J.; Rangasamy, E.; Allen, J. L.; Sakamoto, J. HighConductivity of Dense Tetragonal Li7La3Zr2O12. J. Power Sources 2012,208, 193−196.(69) Choi, J.-H.; Lee, C.-H.; Yu, J.-H.; Doh, C.-H.; Lee, S.-M.Enhancement of Ionic Conductivity of Composite Membranes for All-Solid-State Lithium Rechargeable Batteries Incorporating TetragonalLi7La3Zr2O12 into a Polyethylene Oxide Matrix. J. Power Sources 2015,274, 458−463.(70) Awaka, J.; Takashima, A.; Kataoka, K.; Kijima, N.; Idemoto, Y.;Akimoto, J. Crystal Structure of Fast Lithium-Ion-Conducting CubicLi7La3Zr2O12. Chem. Lett. 2011, 40, 60−62.(71) Geiger, C. A.; Alekseev, E.; Lazic, B.; Fisch, M.; Armbruster, T.;Langner, R.; Fechtelkord, M.; Kim, N.; Pettke, T.; Weppner, W.Crystal Chemistry and Stability of “Li7La3Zr2O12” Garnet: A FastLithium-Ion Conductor. Inorg. Chem. 2011, 50, 1089−1097.

(72) Rangasamy, E.; Wolfenstine, J.; Sakamoto, J. The Role of Al andLi Concentration on the Formation of Cubic Garnet Solid Electrolyteof Nominal Composition Li7La3Zr2O12. Solid State Ionics 2012, 206,28−32.(73) Xu, B.; Duan, H.; Xia, W.; Guo, Y.; Kang, H.; Li, H.; Liu, H.Multistep Sintering to Synthesize Fast Lithium Garnets. J. PowerSources 2016, 302, 291−297.(74) Xia, W.; Xu, B.; Duan, H.; Guo, Y.; Kang, H.; Li, H.; Liu, H.Ionic Conductivity and Air Stability of Al-Doped Li7La3Zr2O12

Sintered in Alumina and Pt Crucibles. ACS Appl. Mater. Interfaces2016, 8, 5335−5342.(75) El-Shinawi, H.; Paterson, G. W.; MacLaren, D. A.; Cussen, E. J.;Corr, S. A. Low-Temperature Densification of Al-Doped Li7La3Zr2O12:A Reliable and Controllable Synthesis of Fast-Ion Conducting Garnets.J. Mater. Chem. A 2017, 5, 319−329.(76) Dumon, A.; Huang, M.; Shen, Y.; Nan, C.-W. High Li IonConductivity in Strontium Doped Li7La3Zr2O12 Garnet. Solid StateIonics 2013, 243, 36−41.(77) Chen, Y. T.; Jena, A.; Pang, W. K.; Peterson, V. K.; Sheu, H.-S.;Chang, H.; Liu, R.-S. Voltammetric Enhancement of Li-IonConduction in Al-Doped Li7‑xLa3Zr2O12 Solid Electrolyte. J. Phys.Chem. C 2017, 121, 15565−15573.(78) Allen, J. L.; Wolfenstine, J.; Rangasamy, E.; Sakamoto, J. Effectof Substitution (Ta, Al, Ga) on the Conductivity of Li7La3Zr2O12. J.Power Sources 2012, 206, 315−319.(79) Song, S.; Sheptyakov, D.; Korsunsky, A. M.; Duong, H. M.; Lu,L. High Li Ion Conductivity in a Garnet-Type Solid Electrolyte ViaUnusual Site Occupation of the Doping Ca Ions. Mater. Des. 2016, 93,232−237.(80) Li, Y.; Wang, Z.; Li, C.; Cao, Y.; Guo, X. Densification andIonic-Conduction Improvement of Lithium Garnet Solid Electrolytesby Flowing Oxygen Sintering. J. Power Sources 2014, 248, 642−646.(81) Buannic, L.; Orayech, B.; Lopez Del Amo, J.-M.; Carrasco, J.;Katcho, N. A.; Aguesse, F.; Manalastas, W.; Zhang, W.; Kilner, J.;Llordes, A. Dual Substitution Strategy to Enhance Li+ IonicConductivity in Li7La3Zr2O12 Solid Electrolyte. Chem. Mater. 2017,29, 1769−1778.(82) Gu, W.; Ezbiri, M.; Rao, R. P.; Avdeev, M.; Adams, S. Effects ofPenta-and Trivalent Dopants on Structure and Conductivity ofLi7La3Zr2O12. Solid State Ionics 2015, 274, 100−105.(83) Bernstein, N.; Johannes, M.; Hoang, K. Origin of the StructuralPhase Transition in Li7La3Zr2O12. Phys. Rev. Lett. 2012, 109, 205702.(84) Shin, D. O.; Oh, K.; Kim, K. M.; Park, K.-Y.; Lee, B.; Lee, Y.-G.;Kang, K. Synergistic Multi-Doping Effects on the Li7La3Zr2O12 SolidElectrolyte for Fast Lithium Ion Conduction. Sci. Rep. 2015, 5, 18053.(85) Kotobuki, M.; Munakata, H.; Kanamura, K.; Sato, Y.; Yoshida,T. Compatibility of Li7La3Zr2O12 Solid Electrolyte to All-Solid-StateBattery Using Li Metal Anode. J. Electrochem. Soc. 2010, 157, A1076.(86) Kim, K. H.; Iriyama, Y.; Yamamoto, K.; Kumazaki, S.; Asaka, T.;Tanabe, K.; Fisher, C. A. J.; Hirayama, T.; Murugan, R.; Ogumi, Z.Characterization of the Interface between LiCoO2 and Li7La3Zr2O12 inan All-Solid-State Rechargeable Lithium Battery. J. Power Sources 2011,196, 764−767.(87) Ohta, S.; Kobayashi, T.; Seki, J.; Asaoka, T. ElectrochemicalPerformance of an All-Solid-State Lithium Ion Battery with Garnet-Type Oxide Electrolyte. J. Power Sources 2012, 202, 332−335.(88) Ohta, S.; Komagata, S.; Seki, J.; Saeki, T.; Morishita, S.; Asaoka,T. All-Solid-State Lithium Ion Battery Using Garnet-Type Oxide andLi3BO3 Solid Electrolytes Fabricated by Screen-Printing. J. PowerSources 2013, 238, 53−56.(89) Du, F.; Zhao, N.; Li, Y.; Chen, C.; Liu, Z.; Guo, X. All SolidState Lithium Batteries Based on Lamellar Garnet-Type CeramicElectrolytes. J. Power Sources 2015, 300, 24−28.(90) Jin, Y.; McGinn, P. J. Li7La3Zr2O12 Electrolyte Stability in Airand Fabrication of a Li/Li7La3Zr2O12/Cu0.1V2O5 Solid-State Battery. J.Power Sources 2013, 239, 326−331.(91) Ahn, C.-W.; Choi, J.-J.; Ryu, J.; Hahn, B.-D.; Kim, J.-W.; Yoon,W.-H.; Choi, J.-H.; Lee, J.-S.; Park, D.-S. Electrochemical Properties of



2750


Li7La3Zr2O12-Based Solid State Battery. J. Power Sources 2014, 272,554−558.(92) Yan, X.; Li, Z.; Wen, Z.; Han, W. Li/Li7La3Zr2O12/LiFePO4 All-Solid-State Battery with Ultrathin Nanoscale Solid Electrolyte. J. Phys.Chem. C 2017, 121, 1431−1435.(93) Thangadurai, V.; Narayanan, S.; Pinzaru, D. Garnet-Type Solid-State Fast Li Ion Conductors for Li Batteries: Critical Review. Chem.Soc. Rev. 2014, 43, 4714−4727.(94) Inaguma, Y.; Liquan, C.; Itoh, M.; Nakamura, T.; Uchida, T.;Ikuta, H.; Wakihara, M. High Ionic Conductivity in LithiumLanthanum Titanate. Solid State Commun. 1993, 86, 689−693.(95) Belous, A. G. N.; Polyanetskaya, S. V.; Gornikov, Y. I. Study ofComplex Oxides with the Composition La2/3−xLi3xTiO3. Inorg. Mater.1987, 23, 412.(96) Bohnke, O. The Fast Lithium-Ion Conducting OxidesLi3xLa2/3−xTiO3 from Fundamentals to Application. Solid State Ionics2008, 179, 9−15.(97) Mei, A.; Wang, X.-L.; Lan, J.-L.; Feng, Y.-C.; Geng, H.-X.; Lin,Y.-H.; Nan, C.-W. Role of Amorphous Boundary Layer in EnhancingIonic Conductivity of Lithium-Lanthanum-Titanate Electrolyte.Electrochim. Acta 2010, 55, 2958−2963.(98) Morata-Orrantia, A.; García-Martín, S.; Alario-Franco, M. A.Optimization of Lithium Conductivity in La/Li Titanates. Chem.Mater. 2003, 15, 3991−3995.(99) Abreu-Sepulveda, M.; Williams, D. E.; Huq, A.; Dhital, C.; Li, Y.;Paranthaman, M. P.; Zaghib, K.; Manivannan, A. Synthesis andCharacterization of Substituted Garnet and Perovskite-Based Lithium-Ion Conducting Solid Electrolytes. Ionics 2016, 22, 317−325.(100) Kwon, W. J.; Kim, H.; Jung, K.-N.; Cho, W.; Kim, S. H.; Lee,J.-W.; Park, M.-S. Enhanced Li+ Conduction in PerovskiteLi3xLa2/3−x□1/3−2xTiO3 Solid-Electrolytes Via Microstructural Engi-neering. J. Mater. Chem. A 2017, 5, 6257−6262.(101) Chen, K.; Huang, M.; Shen, Y.; Lin, Y.; Nan, C. ImprovingIonic Conductivity of Li0.35La0.55TiO3 Ceramics by IntroducingLi7La3Zr2O12 Sol into the Precursor Powder. Solid State Ionics 2013,235, 8−13.(102) Kotobuki, M.; Suzuki, Y.; Munakata, H.; Kanamura, K.; Sato,Y.; Yamamoto, K.; Yoshida, T. Fabrication of Three-DimensionalBattery Using Ceramic Electrolyte with Honeycomb Structure by Sol-Gel Process. J. Electrochem. Soc. 2010, 157, A493−A498.(103) Li, C.-L.; Zhang, B.; Fu, Z.-W. Physical and ElectrochemicalCharacterization of Amorphous Lithium Lanthanum Titanate SolidElectrolyte Thin-Film Fabricated by E-Beam Evaporation. Thin SolidFilms 2006, 515, 1886−1892.(104) Le, H. T.; Ngo, D. T.; Kim, Y.-J.; Park, C.-N.; Park, C.-J. APerovskite-Structured Aluminium-Substituted Lithium LanthanumTitanate as a Potential Artificial Solid-Electrolyte Interface forAqueous Rechargeable Lithium-Metal-Based Batteries. Electrochim.Acta 2017, 248, 232−242.(105) Stramare, S.; Thangadurai, V.; Weppner, W. LithiumLanthanum Titanates: A Review. Chem. Mater. 2003, 15, 3974−3990.(106) Bruce, P. G.; West, A. The A-C Conductivity of PolycrystallineLISICON, Li2+2xZn1‑xGeO4, and a Model for Intergranular Con-striction Resistances. J. Electrochem. Soc. 1983, 130, 662−669.(107) Robertson, A.; West, A.; Ritchie, A. Review of CrystallineLithium-Ion Conductors Suitable for High Temperature BatteryApplications. Solid State Ionics 1997, 104, 1−11.(108) Thangadurai, V.; Weppner, W. Recent Progress in Solid Oxideand Lithium Ion Conducting Electrolytes Research. Ionics 2006, 12,81−92.(109) Kuwano, J.; West, A. R. New Li+ Ion Conductors in theSystem, Li4GeO4-Li3VO4. Mater. Res. Bull. 1980, 15, 1661−1667.(110) Deng, Y.; Eames, C.; Fleutot, B.; David, R. n.; Chotard, J.-N. l.;Suard, E.; Masquelier, C.; Islam, M. S. Enhancing the Lithium IonConductivity in Lithium Superionic Conductor (LISICON) SolidElectrolytes through a Mixed Polyanion Effect. ACS Appl. Mater.Interfaces 2017, 9, 7050−7058.

(111) Song, S.; Lu, J.; Zheng, F.; Duong, H. M.; Lu, L. A FacileStrategy to Achieve High Conduction and Excellent Chemical Stabilityof Lithium Solid Electrolytes. RSC Adv. 2015, 5, 6588−6594.(112) Kato, T.; Hamanaka, T.; Yamamoto, K.; Hirayama, T.; Sagane,F.; Motoyama, M.; Iriyama, Y. In-Situ Li7La3Zr2O12/LiCoO2 InterfaceModification for Advanced All-Solid-State Battery. J. Power Sources2014, 260, 292−298.(113) Zhou, W.; Wang, S.; Li, Y.; Xin, S.; Manthiram, A.;Goodenough, J. B. Plating a Dendrite-Free Lithium Anode with aPolymer/Ceramic/Polymer Sandwich Electrolyte. J. Am. Chem. Soc.2016, 138, 9385−9388.(114) Li, Y.; Xu, B.; Xu, H.; Duan, H.; Lu, X.; Xin, S.; Zhou, W.; Xue,L.; Fu, G.; Manthiram, A.; et al. Hybrid Polymer/Garnet Electrolytewith a Small Interfacial Resistance for Lithium-Ion Batteries. Angew.Chem. 2017, 129, 771−774.



2751


recent advancements in li-ion conductors for all-solid ...rsliu/publications/2017/16.pdf ·...

Documents