Available online at www.sciencedirect.com

ScienceDirect

Journal of Materiomics 1 (2015) 22e32www.ceramsoc.com/en/ www.journals.elsevier.com/journal-of-materiomics/

Interfacial engineering of solid electrolytes

Jian Luo

Department of NanoEngineering, Program of Materials Science and Engineering, University of California, San Diego, CA 92093-0448, USA

Received 31 December 2014; revised 3 January 2015; accepted 30 January 2015

Available online 1 April 2015

Abstract

The roles of interfaces in either blocking or enhancing ionic conduction in various types of solid electrolytes, including lithium, sodium,oxygen and other types of ion conductors as well as proton conductors, are critically reviewed. Two important fundamental interfacial phe-nomena, namely the formation of space charges and two-dimensional interfacial phases (complexions), can markedly alter ion transport along oracross various types of interfaces, including grain and phase boundaries as well as free surfaces. Since the experiments and models of spacecharges have been well documented in literature, a new focus of this short review and viewpoint article is to propose and discuss an emergingopportunity of utilizing the formation and transition of interfacial phases to either alleviate the blocking effects or further enhance ionicconduction in solid electrolytes.© 2015 The Author. Production and hosting by Elsevier B.V. on behalf of The Chinese Ceramic Society. This is an open access article under theCC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Complexion; Grain boundary; Interface; Ionic conduction; Space charge; Solid electrolyte

1. Introduction

Solid electrolytes (ionic conductors) have important applica-tions in batteries, fuel cells, sensors and other types of devices.Interfacial engineering should be an essential component fordesigning novel solid electrolytes because interfaces can oftencontrol the final performance. Specifically, the ionic conductiv-ities of many solid electrolytes are often limited by the high grainboundary (GB) resistivity, which can be several orders ofmagnitude greater than the corresponding bulk resistivity [1e15].In some other instances, interfaces may provide fast ionic con-duction pathways [16e34]. In addition, electrode-electrolyteinterfaces can often control the performance of batteries [35e37].

In this short review and viewpoint article, two important andrelevant fundamental interfacial phenomena, namely the for-mation of space charges and two-dimensional (2-D) interfacialphases (also called complexions), are discussed. Subsequently,the roles of interfaces in either enhancing or blocking ionicconduction are briefly reviewed and discussed. Since interfacesin solid electrolytes, particularly the effects of space charges on

E-mail address: jluo@alum.mit.edu.

Peer review under responsibility of The Chinese Ceramic Society.

http://dx.doi.org/10.1016/j.jmat.2015.03.002

2352-8478/© 2015 The Author. Production and hosting by Elsevier B.V. on behalf

BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

ionic conduction, have been extensively discussed and reviewedpreviously [38e50], a specific new focus of this article is toexplore the previously-recognized roles of 2-D interfacialphases (complexions) in influencing ionic conduction with agoal of intentionally using them to engineer solid electrolytesvia either alleviating the detrimental blocking effects or utiliz-ing the beneficial conductivity enhancements.

2. Relevant fundamental interfacial phenomena

2.1. Space charges

An important character of interfaces (including both phaseand grain boundaries, as well as free surfaces) in solid elec-trolytes are represented by the formation of space charges, theorigin of which is similar to that of electrical double layers atinterfaces between metal electrodes and aqueous electrolytes;similar phenomena have also been discussed and modeled incolloidal theories (in the famous Derjaguin-Landau-Verwey-Overbeek or DLVO model) and semiconductor physics (formodeling Schottky barriers and p-n junctions).

The formation of a space-charge region is caused by theaccumulation of net charges at the interfacial core (typically

of The Chinese Ceramic Society. This is an open access article under the CC

23J. Luo / Journal of Materiomics 1 (2015) 22e32

being assumed to be the first 1e2 atomic layers, a width thatis usually smaller than that of the space-charge zone; Fig. 1),which form, e.g., as a result of the different adsorption (a.k.a.segregation) enthalpies of the positively and negativelycharged defects at the interfacial core. The charges in theinterfacial cores (jxj < d) are generally assumed to be adsor-bed or “bonded” at the interface, which may often havedifferent states, or even different effective charges, ascompared with their counterparts inside the bulk. In manycases, these charges at the interfacial cores (e.g., intrinsicoxygen vacancies at the GB cores in acceptor-doped ZrO2 andCeO2 [15]) are assumed be immobile, even if their counter-parts in the bulk and the space-charge zones are mobile car-riers (whereas, at least in principle, such charges may also bemobile along the interfaces to provide fast ion conduction insome specific cases).

Fig. 1 schematically illustrates two cases where the netcharge accumulations per unit area at the interfacial cores(Q0 ¼ Qþ

0 e Q�0 ) are positive (as examples; noting that the net

Q0 can also be negative); consequently, in the adjacent spacecharge zone(s), the positively-charged defects are depletedwhereas the negatively-charged defects are enriched due toelectrostatic interactions. This leads to the formation of aspatially-varying electrostatic potential profile V(x), where x isthe spatial variable perpendicular to the interface. Forexample, in acceptor-doped ZrO2 and CeO2, the positive Q0 isassumed to be resulted from the intrinsic oxygen vacancies atthe GB cores [15], which leads to a depletion of oxygen va-cancies in the space-charge zones, along with a possibleenrichment of acceptors (for the high-temperature cases,where/if the acceptors are sufficiently mobile to redistribute)[15] however, the space charge zones may be (negligibly)narrow for the high-temperature cases of ZrO2 and CeO2 with

Fig. 1. Schematic illustration of the formation of space charges at (a) a free surface

side) and (b) a symmetrical grain boundary (GB). In both cases, it is assumed that t

charged defects, both of which are sufficiently mobile to redistribute in the space-

high defect concentrations (as the effective Debye length,which is discussed subsequently, can be less than to the latticeconstant), though the space charge zones can be wider at lowtemperatures when the acceptors are frozen (see Fig. 2 andelaboration in section 3.1).

In equilibrium, the concentration profile of a (mobile)charged (defect) specie i obeys the Boltzmann equation:

ciðxÞcið∞Þ ¼ exp

�� zie½VðxÞ �Vð∞Þ�

kBT

�; ð1Þ

where ci(x) is the defect concentration (number of defects perunit volume), z is the number of the charges per defect, e is theelementary charge, kB is the Boltzmann constant and T is theabsolute temperature in Kelvin. The electrostatic potentialprofile is determined from the Poisson equation:

d2VðxÞdx2

¼�qðxÞεrε0

¼� 1

εrε0

Xi

ziciðxÞ; ð2Þ

where q(x) is the local charge density, ε0 is permittivity of freespace and εr is the relative dielectric constant. Eq. (1) and Eq.(2) can be combined to obtain the Poisson-Boltzmann equa-tion, as follows:

d2DVðxÞdx2

¼� 1

εrε0

Xi

�zicið∞Þexp

�� zieDVðxÞ

kBT

��; ð3Þ

for the range of jxj > d, where:

DVðxÞ≡VðxÞ �Vð∞Þ: ð4ÞThe detailed derivation and discussion of Eqs. (1)e(3) can

be found in literature, e.g., in Refs. [15,42]. Solving Eq. (3)requires two boundary conditions as well as knowing the

(or an interface with an inert phase, with negligible space charges on the other

he interfacial cores are positively charged and there are (only) two oppositely-

charge zone to obey the Poisson-Boltzmann equation.

Fig. 2. Schematic illustration of the space-charge model for YSZ at low

temperatures (<~1000 �C), where the acceptors are assumed to be constant

(immobile) and the Mott-Schottky approximation is presumably held [15].

This low-temperature case is of direct importance for understanding the high

GB resistivity.

24 J. Luo / Journal of Materiomics 1 (2015) 22e32

bulk concentrations of all charged defects (ci(∞)). Far awayfrom the interface, the following boundary condition is held:

dDVðxÞdx

����x/∞

¼ 0: ð5Þ

Thus, one needs to know either the interfacial potential(Schottky) barrier height (DV(d) ≡ V(d) e V(þ∞)) or (alter-natively) the total net charge accumulation at the interfacialcore to solve the Poisson-Boltzmann equation.

The Gouy-Chapman analysis is applicable when thecharged species in the space charge zone (for jxj > d) aresufficiently mobile to re-distribute. In such a case, the width ofspace charge zone is scaled by (and approximately on theorder of) the Debye length:

LD ¼ k�1D ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiεrε0kBT

e2Pi

z2i cið∞Þ

vuut : ð6Þ

For cases where there are only two oppositely-chargeddefects, Fig. 1(a) illustrates the space-charge (defect concen-trations) and electrostatic potential (DV(x)) profiles near a freesurface (or an interface with an inert phase, in which the spacecharges can be neglected), while Fig. 1(b) illustrates a case fora symmetrical GB.

For systems like acceptor-doped ZrO2 and CeO2, thecalculated Debye length (LD) can be as smaller as ~0.1 nm[15] because of the high concentrations of charged defects.This implies that the width of space-charge zone is negligiblynarrow at high temperatures (>1000 �C) where acceptors aresufficiently mobile to redistribute (see section 3.1 for further

discussion). However, at low temperatures (<1000 �C formany acceptor-doped ZrO2 and CeO2), the Mott-Schottkyapproximation (Fig. 2) might be correct and the spacecharge zones could be significantly wider (typically on theorder of 1 nm for Y2O3-doped ZrO2 and CeO2 at 500 �C[15]) to have a significant impact on GB resistivity. Forexample, in Y2O3-stabilized ZrO2 (YSZ), the acceptor (Y0

Zr)concentration was assumed to be a constant (up to the GBcore) and the electrostatic potential in the space charge zoneis dictated only by the depletion of oxygen vacancies (V��

O ),as shown in Fig. 2; adopting the Mott-Schottky approxima-tion (Fig. 2), the width of the space-charge zone can beestimated as [15]:

l* ¼ 2LD

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffieDVðdÞkBT

sð7Þ

for YSZ, which can be substantially greater than the Debyelength (LD); similar expressions can be derived for variousother solid electrolytes. Fig. 2 essentially represents the for-mation of a double Schottky barrier at the GB core. It shouldbe emphasized that this low-temperature (<1000 �C) case, asshown in Fig. 2, is of practical relevance and direct importancefor understanding the resistive GBs in acceptor-doped ZrO2

and CeO2 (see further discussion in section 3.1). Detailed andin-depth discussion of the space-charge models, variousapproximations and solutions for specific cases can be foundin several review and research articles [15,38e41] and are notrepeated here to avoid redundancy.

2.2. Two-dimensional interfacial phases or complexions

Yet another important interfacial phenomenon that candrastically impact on the ionic conductivity is represented bythe formation and transition of 2-D interfacial phases.Specifically, researchers have long recognized the widespreadexistence of a unique class of impurity-based, nanometer-thick, intergranular films (IGFs) in various oxide and non-oxide ceramic materials, including many solid electrolytes(see a critical review [51] and references therein). Clarkeoriginally proposed that these IGFs adopt an equilibriumthickness on the order of 1 nm in response to a balance amongseveral attractive and repulsive interfacial forces [52e54].Thermodynamically, these nanoscale IGFs can be alternativelyconsidered as a class of multilayer adsorbates or one specialtype of 2-D interfacial phases, which can exhibit atomicstructures and composition that are neither observed nornecessarily stable as bulk phases.

Fig. 3(a) shows a high-resolution transmission electronmicroscopy (HRTEM) image of a nanometer-thick IGFformed in LiFePO4; a series of recent studies suggested thatthe formation of such lithium phosphate-based IGFs, alongwith surface amorphous films (SAFs) of similar character (asthe free-surface counterparts to equilibrium IGFs) [55e63](Fig. 3(a)), may enhance the rate capabilities of battery cath-odes [35,36,55,56,64], which will be discussed furtherin section 5.

Fig. 3. (a) HRTEM image of a Li4P2O7-based, nanometer-thick IGF and an

analogous surface phase (SAFs) observed in LiFePO4, which may provide a

fast Liþ conduction pathway [35,56,64]. (b) Schematic of a series of 2-D

interfacial phases (DilloneHarmer complexions), which may be interpreted

as derivatives of IGFs with discrete (nominal) thicknesses of 0, 1, 2, 3, x and

þ∞ monolayers, respectively. These interfacial phases (complexions) are

thermodynamically 2-D (a.k.a. the thorough-thickness compositional and

structural profiles are thermodynamically-determined so there is no degree of

freedom in this perpendicular direction); their formation and transition can

drastically change both the space-charge profile (via changing Q0) and ionic

mobility along the interfaces (via inducing interfacial disordering). Panel (a) is

adapted from Ref. [56] and panel (b) is reprinted from Ref. [90] with

permission.

25J. Luo / Journal of Materiomics 1 (2015) 22e32

In a broader context, materials scientists have long recog-nized that GBs, as well as other types of interfaces, can beconsidered as “2-D interfacial phases” that may undergo phase-like transitions [52,53,65e85]. Tang, Carter and Cannon[71,72] named such 2-D interfacial phases as “complexions”based on an argument that they are not rigorously “phases”according to the Gibbs definition. Recently, a series of discrete2-D interfacial phases have been observed in doped Al2O3 byDillon and Harmer [86e89] and subsequently in various othermaterials [82,90e95]; this series of DilloneHarmer complex-ions can be considered as derivatives of IGFs with discretenominal thicknesses of 0, 1, 2, 3, x and þ∞ monolayers,respectively (Fig. 3(b)) [80,81,90,96]. The occurrences of first-order GB transitions were also evident [91,97]. These findingsof interfacial phase-like behaviors provide new insights todecipher several outstanding mysteries in materials science,including solid-state activated sintering [80,83,93,98e101],abnormal grain growth [86e88,102] and liquid metal embrit-tlement [82,92]. Moreover, interfacial engineering via utilizingsuch 2-D interfacial phases (complexions) to achieve superiorproperties that may not be attainable by bulk phase represents apotentially-transformative research direction [74,90]. Interestedreaders are referred to several recent review articles[74,88e90,96] for in-depth discussion of this subject.

Specific to solid electrolytes, the formation and transitionof 2-D interfacial phases (complexions), such as those shown

in Fig. 3, can markedly change: 1) the space charges in theabutting crystals by changing the net charge accumulation atthe interfacial core (Q0 in Fig. 1) and 2) the ion mobility alongthe interface by inducing interfacial structural or chemicaldisorder. Both can have significant (positive or negative)impacts on the ionic conduction, which have not been fullyrecognized and explored in prior studies.

3. Grain boundary blocking effects

It is well known that GBs are often “blocking” in manysolid electrolytes, such as ZrO2 and CeO2 based oxygen ionconductors [12e15], BaZrO3 and BaZr1�xYxO3�d based pro-ton conductors [1e3] and Li3xLa2/3�xTiO3 based lithium ionconductors [4e6]. The low GB conductivities were attributedto the space charges that deplete the carrier concentrations(Fig. 2) [7e9] and the formation of various types of resistiveinterfacial phases (Fig. 4(b) and 5) [15,103,104].

3.1. Acceptor-doped ZrO2 and CeO2

Using acceptor-doped ZrO2 and CeO2 as examples, Fig. 4schematically illustrates the GB configurations, as well aspossible formation of impurity (silicate)-based IGFs andassociated GB transitions in these systems. In the “intrinsic”case, GBs are atomically sharp in HRTEM images; yet, theseGBs are positively charged (Fig. 4(a)), which is presumablydue to the accumulation of intrinsic oxygen vacancies (Fig. 2)[12e15]. At low temperatures where the Mott-Schottkyapproximation is assumed, the positively-charged GB coresare offset by negative space charge zones of width of l* (Eq.(7)), where the (mobile) oxygen vacancies in the space-chargezones are severely depleted (Fig. 2), leading to high GBresistivity (which can be approximately 100 times higher thanthe bulk resistivity at the same temperature [12e15]). At hightemperatures (>~1000 �C), the acceptors are sufficientlymobile and the width of the space charge zones is reduced toapproximately the Debye length, which is on the order of0.1 nm because of the heavy doping; consequently, this GBblocking effect may disappear [15].

Practically, siliceous impurities are often present (due tounintentional contaminations), and the formation ofnanometer-thick, glass-like, impurity (mostly silicate)-basedIGFs (or derivative complexions), which are presumablyresistive (insulating to oxygen vacancies), can further increasethe GB resistivity substantially (Fig. 4(b)) [10,11,103]. In suchcases, two strategies can be employed to alleviate the detri-mental effects of siliceous (silicate-based) IGFs (Fig. 4) [103].First, post-sintering annealing can be used to “dewet” suchIGFs at a temperature lower than the typical sintering tem-perature (e.g., dewetting IGFs at ~1300 �C for YSZ vs. thetypical sintering temperatures of 1500e1600 �C where suchIGFs form) [103]. This represents a case of using temperature-induced GB phase-like (complexion) transition (from thenanoscale IGFs that formed at high temperatures to the“clean” GBs that are stable at low temperatures) to improveinterfacial ionic conductivity. Second, various scavengers,

Fig. 4. Schematic illustration of the GB configurations in acceptor-doped ZrO2 and CeO2. (a) In the “intrinsic” case, GBs are atomically sharp and positively

charged. Space-charge zones with severe depletion of oxygen vacancies can form at low temperatures where the Mott-Schottky approximation is assumed (Fig. 2),

leading to high GB resistivity; this GB blocking effect may disappear at T > ~1000 �C [15]. (b) Practically, the presence of nanometer-thick, siliceous IGFs can

further increase the GB resistivity (substantially) because such glass-like IGFs are considered highly insulating to oxygen vacancies/ions. Two remediating

strategies, namely (1) post-sintering annealing at a relatively low T (~1300 �C) to “dewet” and (2) reducing the m of SiO2 to scavenge theses siliceous IGFs, can be

employed [103]; both are examples of using GB transitions (via changing the complexion stability by changing the thermodynamic potential T or m) to alleviate the

detrimental effects.

Fig. 5. Schematic illustration of a possible thermodynamic origin of the resistive TiO2�x-like GB complexion in LLTO observed by Ma et al. [104], which may be

interpreted as a case of GB prewetting.

26 J. Luo / Journal of Materiomics 1 (2015) 22e32

which form either precipitated secondary phases (e.g., Al2O3

in YSZ and MgO in Gd2O3-doped CeO2 (GDC)) or solidsolutions with the electrolytes (e.g., MgO in YSZ and CaO inGDC), can be used to scavenge siliceous IGFs [103]; thisrepresents a case where the chemical potential (m) of SiO2 isreduced to destabilize a detrimental interfacial complexion(siliceous IGFs in the current case). The experimental obser-vations of the formation of resistive IGFs and remediatingstrategies were discussed in details in a recent review article[103]; here, the same observations are re-analyzed in theframework of GB complexion formation and transition.

3.2. Li3xLa2/3�xTiO3

Li3xLa2/3exTiO3 (LLTO) represents another classic case,where GBs are resistive. Recently, Ma et al. [104] used high-angle annular dark field scanning transmission electronmicroscopy (HAADF STEM), in conjunction with electronenergy loss spectroscopy (EELS), to investigate the GBstructure and chemistry in LLTO. They found significantdepletion of both La and Li at LLTO GBs, as well as partialreduction of Ti4þ to Ti3þ and deformation of TieO poly-hedral; thus, the GBs in LLTO are essentially TiO2�x-like; the

affected GB region are approximately 2e3 unit cell in thick-ness (a.k.a. on the order of 1 nm). Here, I propose that thethermodynamic stability of TiO2�x-like GB structures can beinterpreted as a case of GB prewetting (noting that a similarcase of solid-state GB prewetting can be found in Ref. [77];the prewetting theory was originally proposed in the Cahncritical-point-wetting model [68] and discussed in severalreview articles [51,59,74,90]), which may be thermodynami-cally driven by the lower GB energy of the TiO2-x-like phase(as shown in Fig. 5):

gLLTOGB >g

TiO2�xGB : ð8Þ

A more rigorous model for the formation of such a(presumably resistive) TiO2�x-like GB complexion can bedeveloped by adapting and revising the existing diffuse-interface models [71,105e107], which warrants a futurestudy. Presumably, this TiO2�x-like GB complexion is the rootcause of low GB conductivity in LLTO.

Ma et al.'s study of LLTO [104] pointed out the possibleexistence of GB complexions (a.k.a. specific GB structure,chemistry and thermodynamic states) that are essentiallyinvisible in conventional HRTEM (unlike the nanoscale glass-like IGFs that can be clearly detected by phase-contrast

27J. Luo / Journal of Materiomics 1 (2015) 22e32

HRTEM), which were therefore unrecognized in prior studies.In this regard, Ma et al.'s study [104] is important to suggestthat more studies have to be conducted to reveal the atomic-level GB structure and chemistry (GB complexions), whichcan often be markedly different from the bulk materials, aswell as their detrimental or beneficial roles in influencinginterfacial ionic conduction. LLTO is perhaps only one of themany unrecognized cases.

4. Interface-enhanced ionic conduction in multilayers andcomposites

Extensive studies have been conducted to utilize interfacesto enhance ionic conduction in multilayers and compositeelectrolytes. Two possible underlying mechanisms, namelyincreased carrier concentrations in the space-charge zones(mostly for electrolytes where the ionic mobilities are high butthe carrier concentrations are low) and improved ionic mo-bilities (particularly important and essential for systems likeYSZ and GDC, where the carrier concentrations are alreadyhigh and the Debye lengths are small). This area of researchhas been discussed in details in several research and reviewarticles [16,38e40,42e50,108e110]; thus, I only brieflyoutline some key results and ideas in the subsequent text.

4.1. Space-charge effects on increasing carrierconcentrations

Space charges have been used to enhance ionic conduc-tivities in composite electrolytes and multilayers, followingthe pioneering study of LiIeAl2O3 in 1970s [111,112]. Later,similar effects have been reported for various composite ormultilayer ionic conductors (AgCl/Al2O3 or SiO2 [16], LiI/Al2O3 [17], CuCl/Al2O3 [18] and KCl/Al2O3 [19]), mixedconductors (AgI/AgBr [20], CaF2/BaF2 [21] and LiI/Li3PO4

[22]) and glass-ceramic electrolytes (Li1þxTi2�xAlx(PO4)3[113] and LiAlSiO4 [114]). Interface-enhanced ionic conduc-tion has also been reported for various composite sodiumelectrolytes (Na2SO4/Al2O3 [115], Na2SO4/PbTiO3 [116],Na2SO4/MgO, Al2O3 or SiO2 [117], Na4Zr2Si3O12/SbF5eSiO2/Al2O3 [118] and Na4Zr2Si3O12/SO

2�4 -ZrO2 [119]).

In one instance, the conductivity of Na4Zr2Si3O12 wasimproved by two orders of magnitude by adding SO2�

4 /ZrO2

solid superacids [119]. In composites, whether the interfaceshave percolation paths is an important factor in determiningthe overall conductivity; relevant theories have been developedfor random composites [120,121].

Maier and co-workersmadegreat advancements indevelopingand validating (via well-designed and systematic experiments)quantitative space-charge models to explain the interface-enhanced ionic conductivity [16,42e45,47e50,108e110] andfurther proposed a provocative concept of “nanoionics” torepresent methods of using interfaces to engineer ion conductionand storage [43]. In most cases, it was believed that interface-enhanced ionic conductivity is caused by an increase of thecarrier concentration in the space-charge region. In particular, anelegantmodel experiment of (CaF2/BaF2)nmultilayers supported

this theory [122]; subsequent careful experiments of LiF/TiO2

and LiF/Al2O3 systems also confirmed the space-charge model[123,124]; but a recent study of LiF/SiO2 attributed the enhancedionic conductivity to the fast diffusion in the glassy matrix nearthe crystalline grains [125] (which is beyond the prediction of thespace-charge model and will be discussed further in section 5).Similarly, the enhanced sodium ionic conductivity in Na4Zr2-Si3O12/SbF5eSiO2/Al2O3 cannot be fully explainedby the space-charge formation [118].

4.2. Possible enhancements of ionic mobility

Interface-enhanced ionic conductivity may also stem froman increase of ion mobility; this was proposed for YSZ andother oxide thin films [27e34]. In YSZ, GDC and othersimilar oxygen ion conductors, the bulk carrier concentrationsare already high so that substantial increases in the carrierconcentrations by space-charge effects is unlikely (unless theaccumulated intrinsic oxygen vacancies at the interfacial cores[15], as schematically shown in Fig. 1, are mobile, which is aneffect that has not considered or investigated systematically);thus, any significant conductivity enhancement is likely due toa mobility effect (or proton/electron conduction as artifacts)[27,28,38,39]. In fact, many researchers investigated theinterface effects on ionic conductivity in ZrO2/YSZ basedmaterials and other similar oxides, and the results and expla-nations are somewhat controversial; see two review articles[38,39] and references therein. Moreover, interface-enhancedionic conductivity has also been observed in multilayers ofGd-doped (CeO2eZrO2)n [23e25], (ZrO2/YSZ-R2O3)n(R ¼ Y, Lu, Sc and Al) [126e132] and (YSZ-SiO2)n [26].Specifically, a colossal increase in conductivity was claimedfor epitaxial YSZ-SrTiO3 superlattice structures [133]; how-ever, whether the observed conductivity was ionic was ques-tioned [38,134].

The enhanced ionic conductivity via increasing ionicmobility was attributed to the effects of microstructural defects(including misfit dislocations, incoherent interfaces and otherextended defects) and/or strain effects (for more coherent in-terfaces); the exact mechanisms would likely to be differentfor different systems; in fact, the exact mechanisms remaincontroversial or unresolved in some/many cases. More com-plete discussions of the experimental results of different ma-terials, explanations and theories, as well as the controversiesand debates, can be found in several review articles [38e50]and are not repeated here.

5. Interfacial complexion engineering of solid electrolytes

An emerging opportunity of interfacial engineering of solidelectrolyte is represented by investigating the previously-recognized roles of 2-D interfacial phases (complexions) andsubsequently using them to tailor solid electrolytes by eitherremediating detrimental effects or utilizing beneficial effects.

As discussed in sections 2.2 and 3.1, one ubiquitous 2-Dinterfacial phase (complexion) is represented by theimpurity-based (particularly siliceous), nanoscale, equilibrium

28 J. Luo / Journal of Materiomics 1 (2015) 22e32

IGFs, which often block the conduction of oxygen vacancies.For such cases, useful strategies can be developed to desta-bilize these detrimental IGFs via complexion (interfacialphase-like) transitions by developing specific annealing ordoping (scavenging) recipes. In this regard, examples ofacceptor-doped ZrO2 and CeO2 have been discussed in section3.1 and illustrated in Fig. 4. Here, a major challenge is todevelop predictive thermodynamic models to forecast thestability of such IGFs as a function of temperature and bulkchemical potentials (doping concentrations) [83,90].

Although siliceous IGFs often block the conduction ofoxygen vacancies [12e15], a series of recent studies suggestedthat IGFs may enhance the conduction and transport of lithiumions and protons under certain conditions. In 2005, Li andGarofalini [135] first suggested that such nanoscale IGFs canact as rapid lithium ion transport pathway via moleculardynamics simulations of V2O5. In 2007, Harley, Yu and DeJonghe [136] reported that the formation of 1e4 nm thick(LaP3O9-based) IGFs in LaPO4 significantly increased theproton conductivity (Fig. 6). In 2009, Kang and Ceder reportedthat the formation of a glassy, Li4P2O7-like, “fastion-conducting surface phase” can help to achieve ultrafastdischarging of the LiFePO4 battery cathode [35]. In 2009, afollow-up study showed that this “fast ion-conducting surfacephase” is in fact equilibrium-thickness SAFs (Fig. 3(a)), whichare the free-surface counterparts to IGFs; furthermore,Li4P2O7-based IGFs of similar character were also observed insuch LiFePO4 cathode materials powders (that were typicallyannealed at ~600 �C so that the particles partially-sinteredwith a substantial amount of GBs to achieve best

Fig. 6. The formation of 1e4 nm thick IGFs in the LaPO4-based solid electrolytes

with permission.

performance). Fig. 3(a) shows a HRTEM image of both anIGF and two SAFs of similar character [56]. In 2012, a furtherstudy suggested that these IGFs also contributed to fast lithiumion diffusion in partially-sintered “secondary particles” in theKang and Ceder's material [64]. In 2013, phosphate-basedIGFs, along with SAFs of similar character, have also beenobserved in partially-sintered LiMn1.5Ni0.5O4 electrodes andpresumably enhanced the ionic conduction [36]. In a separatestudy, Mei et al. suggested that the formation of “siliceousamorphous boundary layers” (that are presumably IGFs) mayenhance the Liþ ionic conductivity of LLTO [137].

Here, I hypothesize that nanoscale glass-like IGFs, withappropriate chemistry, can serve as fast transport pathways forLiþ and Naþ ions (although they typically block the conduc-tion of oxygen vacancies). This is, in part, because these IGFsexhibit unique semi-crystalline and semi-glassy atomic struc-tures [51]; they can be viewed as nanometer-thick, glass-likefilms that are confined between two crystals, which imposepartial crystalline order into the films. Recent studies sug-gested that such partial structural order near the crystal-glassinterfaces can promote ion transport to achieve higher ionicconductivity than both the crystal and glass phases [114,125].

It is worthy reiterating that Ma et al.'s discovery [104] of aninsulating TiO2�x-like GB complexion in LLTO (Fig. 5) sentan important message. This TiO2�x-like GB complexion inLLTO, unlike those nanoscale glass-like IGFs, is essentiallyindistinguishable in conventional phase-contrast HRTEM. Yet,this “invisible” complexion can significantly block the lithiumion conduction. In fact, many types of complexions, such asthe monolayers, bilayers and trilayers shown in Fig. 3(b), are

can substantially increase the proton conductivity. Reprinted from Ref. [136]

29J. Luo / Journal of Materiomics 1 (2015) 22e32

nearly indistinguishable in conventional HRTEM. Thus, it ispossible that many different types of 2-D interfacial phases(complexions) have been unrecognized in prior studies; yettheir formation and transition may have significant detrimentalor beneficial effects on ionic conduction. In general, the for-mation and transition of 2-D interfacial phases (complexions),which can be induced by doping, temperature, and otherstimuli (e.g., overpotential during electrochemical cycling[106]), can substantially change the ionic conductivity viainfluencing the space charges and carrier mobilities. Hence, itis imperative to recognize their existence and important rolesin influencing ionic conduction (and many other transport andmaterials properties [74,83,90]). Further studies have to beconducted to develop quantitative models to forecast the sta-bility of various relevant 2-D interfacial phases (complexions)and how their formation and transition may change the ionicconductivity via changing the space charge profile and/orinterfacial ionic mobility.

6. Conclusions

Prior studies have clearly demonstrated that interfaces caneither block or enhance ionic conduction of various solidelectrolytes. Two important fundamental interfacial phenom-ena that can markedly affect ionic conduction are representedby the formation of space charges and 2-D interfacial phases(complexions). An emerging opportunity is represented byutilizing the formation and transition of 2-D interfacial phasesto control or alleviate the blocking effects and exploring thepossibility of using different types of 2-D interfacial phases toenhance ionic conductivity via either increasing the ionicmobility in the partially-disordered (and yet partially-ordered)interfacial region or tuning the space charges and associatedcarrier concentration profiles.

Here, a key research challenge is to establish the funda-mental understanding and quantitative models to forecast theformation and transition (a.k.a. the thermodynamic stability)of such 2-D interfacial phases (complexions) and how theyaffect ionic conduction along and across the interfaces. Inaddition to developing predictive interfacial thermodynamicmodels, we need to modify and refine the existing space-charge models to understand how the formation and transi-tion of different types of 2-D interfacial phases, such as shownin Figs. 3(b) and 5, can change the space-charge profiles(Fig. 1), as well as develop new models to understand andultimately predict the ionic mobility in different types of 2-Dinterfacial phases. The successful completion of these chal-lenging thrusts will eventually enable us to control and furtherdesign interfaces in solid electrolytes.

Acknowledgment

This review article is benefited from a series of recentstudies in several relevant research areas that were/are sup-ported by the National Science Foundation (under the grantno. CMMI-1436976 for interfacial engineering of sodiumsuperionic conductors, grant no. DMR-1320615 for studying

the surface phases/SAFs in lithium ion battery electrodes, andgrant no. CMMI-1436305 for investigating sintering andmaterials processing), the Air Force Office of ScientificResearch (under the grant no. FA9550-14-1-0174 for studyingthe interactions between an applied electric field and GBs,including a fundamental study of space charges in presence ofGB complexions) and the Office of Naval Research (via aMURI project under the grant no. N00014-11-1-0678 forfundamental studies of GB complexions in selected systems,including an earlier study of GB complexions in oxygen ionconductors in 2011e2012, and a National Security Scienceand Engineering Faculty Fellowship under the grant no.N00014-15-1-0030 for developing interfacial “phase” dia-grams, including LLTO based lithium solid electrolytes). Theauthor thanks Mojtaba Samiee for collecting and summari-zing some of the articles cited here, particularly on the topicsof sodium solid electrolytes and early studies of spacecharges.

References

[1] Shirpour M, Rahmati B, Sigle W, van Aken PA, Merkle R, Maier J.

Dopant segregation and space charge effects in proton-conducting

BaZrO3 perovskites. J Phys Chem C 2012;116:2453e61.

[2] Malavasi L, Fisher CAJ, Islam MS. Oxide-ion and proton conducting

electrolyte materials for clean energy applications: structural and

mechanistic features. Chem Soc Rev 2010;39:4370e87.

[3] Ryu KH, Haile SM. Chemical stability and proton conductivity of

doped BaCeO3eBaZrO3 solid solutions. Solid State Ionic

1999;125:355e67.[4] Ban CW, Choi GM. The effect of sintering on the grain boundary

conductivity of lithium lanthanum titanates. Solid State Ionic

2001;140:285e92.

[5] Yang K-Y, Wang J-W, Fung K-Z. Roles of lithium ions and la/li-site

vacancies in sinterability and total ionic conduction properties of

polycrystalline Li3xLa2/3�xTiO3 solid electrolytes. J Alloy Compd

2008;458:415e24.[6] Inaguma Y, Liquan C, Itoh M, Nakamura T, Uchida T, Ikuta H, et al.

High ionic conductivity in lithium lanthanum titanate. Solid State

Commun 1993;86:689e93.

[7] Shirpour M, Merkle R, Maier J. Space charge depletion in grain

boundaries of bazro3 proton conductors. Solid State Ionic

2012;225:304e7.

[8] Vollman M, Waser R. Grain boundary defect chemistry of acceptor-

doped titanates: space charge layer width. J Am Ceram Soc

1994;77:235e43.

[9] Denk I, Claus J, Maier J. Electrochemical investigations of SrTiO3

boundaries. J Electrochem Soc 1997;144:3526e36.[10] Aoki M, Chiang Y-M, Kosacki I, Lee LJ-R, Tuller H, Liu Y. Solute

segregation and grain-boundary impedance in high-purity stabilized

zirconia. J Am Ceram Soc 1996;79:1169e80.

[11] Lee J-H, Lee JH, Kim D-Y. Inhomogeneity of grain-boundary resistivity

in calcia-stabilized zirconia. J Am Ceram Soc 2002;85:1622e4.

[12] Guo X, Maier J. Grain boundary blocking effect in zirconia: a Schottky

barrier analysis. J Electrochem Soc 2001;148:E121e6.

[13] Guo X, Sigle W, Fleig J, Maier J. Role of space charge in the grain

boundary blocking effect in doped zirconia. Solid State Ionic

2002;154:555e61.

[14] Guo X, Sigle W, Maier J. Blocking grain boundaries in yttria-doped and

undoped ceria ceramics of high purity. J Am Ceram Soc

2003;86:77e87.

[15] Guo X, Waser R. Electrical properties of the grain boundaries of oxygen

ion conductors: acceptor-doped zirconia and ceria. Prog Mater Sci

2006;51:151e210.

30 J. Luo / Journal of Materiomics 1 (2015) 22e32

[16] Maier J. Space-charge regions in solid 2-phase systems and their con-

duction contribution .1. Conductance enhancement in the system ionic

conductor-inert phase and application on AgCl-Al2O3 and AgCl-SiO2. J

Phys Chem Solid 1985;46:309e20.

[17] Maekawa H, Tanaka R, Sato T, Fujimaki Y, Yamamura T. Size-

dependent ionic conductivity observed for ordered mesoporous

alumina-LiI composite. Solid State Ionic 2004;175:281e5.

[18] Jow T, Wagner JB. Effect of dispersed alumina particles on the

electrical-conductivity of cuprous chloride. J Electrochem Soc

1979;126:1963e72.

[19] Kumar A, Shahi K. Composition and particle-size effects on ionic-

conduction in KCl-Al2O3 composite solid electrolytes. J Phys Chem

Solid 1995;56:215e22.

[20] Shahi K, Wagner JB. Fast ion-transport in silver-halide solid-solutions

and multiphase systems. Appl Phys Lett 1980;37:757e9.

[21] Sata N, Jin-Phillipp NY, Eberl K, Maier J. Enhanced ionic conductivity

and mesoscopic size effects in structures of BaF2 and CaF2. Solid State

Ionic 2002;154:497e502.

[22] Kaus NH, Ahmad AH. Conductivity studies and ion transport mecha-

nism in Lii-Li(3)Po(4) solid electrolyte. Ionics 2009;15:197e201.

[23] Azad S, Marina OA, Wang CM, Saraf L, Shutthanandan V,

McCready DE, et al. Nanoscale effects on ion conductance of layer-by-

layer structures of gadolinia-doped ceria and zirconia. Appl Phys Lett

2005;86:131906.

[24] Sayle TXT, Parker SC, Sayle DC. Ionic conductivity in nano-scale

CeO2/YSZ heterolayers. J Mater Chem 2006;16:1067e81.

[25] Wang CM, Azad S, Shutthanandan V, McCready DE, Peden CHF,

Saraf L, et al. Microstructure of ZrO2-CeO2 hetero-multi-layer films

grown on ysz substrate. Acta Mater 2005;53:1921e9.

[26] Karthikeyan A, Ramanathan S. Temperature-dependent interfacial car-

rier transport in low-dimensional oxides using ionic conductor-insulator

(YDZ-SiO2) superlattices. J Appl Phys 2008;104:124314.

[27] Kosacki I, Rouleau CM, Becher PF, Bentley J, Lowndes DH. Surface

interface-related conductivity in nanometer thick YSZ films. Electro-

chem Solid State Lett 2004;7:A459e61.

[28] Kosacki I, Rouleau CM, Becher PF, Bentley J, Lowndes DH. Nanoscale

effects on the ionic conductivity in highly textured YSZ thin films. Solid

State Ionic 2005;176:1319e26.[29] Huang H, Gur TM, Saito Y, Prinz F. High ionic conductivity in ultrathin

nanocrystalline gadolinia-doped ceria films. Appl Phys Lett

2006;89:143107.

[30] Ishihara T, Yan J, Matsumoto H. Extraordinary fast oxide ion conduc-

tivity in la1.61geo5-delta thin film consisting of nano-size grain. Solid

State Ionic 2006;177:1733e6.

[31] Ishihara T, Tabuchi J, Ishikawa S, Yan J, Enoki M, Matsumoto H.

Recent progress in LaGaO3 based solid electrolyte for intermediate

temperature sofcs. Solid State Ionic 2006;177:1949e53.

[32] Yan J, Matsumoto H, Ishihara T. Nanosize effect on the oxide ionic

conductivity of lanthanum germanite thin films (vol 8, pg a607, 2005).

Electrochem Solid State Lett 2006;9. L1-L.

[33] Yan JW, Matsumoto H, Ishihara T. Nanosize effect on the oxide ionic

conductivity of lanthanum germanite thin films. Electrochem Solid

State Lett 2005;8:A607e10.

[34] Kulkarni A, Bourandas A, Dong JH, Fuierer PA, Xiao H. Synthesis and

characterization of nanocrystalline (Zr0.84y0.16)o-1.92-(Ce0.85sm0.15)

o-1.925 heterophase thin films. J Mater Res 2006;21:500e4.[35] Kang B, Ceder G. Battery materials for ultrafast charging and dis-

charging. Nature 2009;458:190e3.

[36] Chong J, Xun S, Song X, Liu G, Battaglia VS. Surface stabilized

LiNi0.5Mn1.5O4 cathode materials with high-rate capability and long

cycle life for lithium ion batteries. Nano Energy 2013;2:283e93.

[37] Sun K, Dillon SJ. A mechanism for the improved rate capability of

cathodes by lithium phosphate surficial films. Electrochem Commun

2011;13:200e2.

[38] Guo XX, Maier J. Ionically conducting two-dimensional hetero-

structures. Adv Mater 2009;21:2619e31.

[39] Ramanathan S. Interface-mediated ultrafast carrier conduction in oxide

thin films and superlattices for energy. JVacSci TechA2009;27:1126e34.

[40] Park HJ, Kwak C, Lee KH, Lee SM, Lee ES. Interfacial protonic

conduction in ceramics. J Eur Ceram Soc 2009;29:2429e37.

[41] Fabbri E, Pergolesi D, Traversa E. Ionic conductivity in oxide hetero-

structures: the role of interfaces. Sci Technol Adv Mater 2010;11.

054503.

[42] Maier J. Ionic-conduction in-space charge regions. Prog Solid State

Chem 1995;23:171e263.

[43] Maier J. Nanoionics: Ion transport and electrochemical storage in

confined systems. Nat Mater 2005;4:805e15.

[44] Maier J. Ionic and electronic carriers in solids - physical and chemical

views of the equilibrium situation. Solid State Ionic 2001;143:17e23.

[45] Maier J. Thermodynamic aspects and morphology of nano-structured

ion conductors - aspects of nano-ionics part i. Solid State Ionic

2002;154:291e301.

[46] Maier J. Nano-sized mixed conductors (aspects of nano-ionics. part III).

Solid State Ionic 2002;148:367e74.[47] Maier J. Complex oxides: high temperature defect chemistry vs. Low

temperature defect chemistry. Phys Chem Chem Phys 2003;5:2164e73.

[48] Maier J. Nano-ionics: trivial and non-trivial size effects on ion con-

duction in solids. Zeitschrift Fur Physikalische Chemie-Int J Res Phys

Chem Chem Phys 2003;217:415e36.

[49] Maier J. High temperature versus low temperature defect chemistry.

Solid State Ionic 2004;173:1e8.[50] Maier J. Ionic transport in nano-sized systems. Solid State Ionic

2004;175:7e12.

[51] Luo J. Stabilization of nanoscale quasi-liquid interfacial films in inor-

ganic materials: a review and critical assessment. Crit Rev Solid State

Mater Sci 2007;32:67e109.

[52] Clarke DR. On the equilibrium thickness of intergranular glass phases

in ceramic materials. J Am Ceram Soc 1987;70:15e22.[53] Clarke DR, Shaw TM, Philipse AP, Horn RG. Possible electrical

double-layer contribution to the equilibrium thickness of intergranular

glass films in polycrystalline ceramics. J Am Ceram Soc

1993;76:1201e4.[54] Bobeth M, Clarke DR, Pompe W. Diffuse interface description of

intergranular films in polycrystalline ceramics. J Am Ceram Soc

1999;82:1537e46.

[55] Huang J, Luo J. A facile and generic method to improve cathode ma-

terials for lithium-ion batteries via utilizing nanoscale surface amor-

phous films of self-regulating thickness. Phys Chem Chem Phys

2014;16:7786e98.

[56] Kayyar A, Qian HJ, Luo J. Surface adsorption and disordering in

LiFePO4 based battery cathodes. Appl Phys Lett 2009;95:221905.

[57] Qian HJ, Luo J, Chiang YM. Anisotropic wetting of ZnO by Bi2O3 with

and without nanometer-thick surficial amorphous films. Acta Mater

2008;56:862e73.

[58] Qian H, Luo J. Nanoscale surficial films and a surface transition in

V2O5-TiO2-based ternary oxide systems. Acta Mater 2008;56:4702e14.

[59] Luo J, Chiang Y-M. Wetting and prewetting on ceramic surfaces. Annu

Rev Mater Res 2008;38:227e49.

[60] Qian HJ, Luo J. Vanadia-based equilibrium-thickness amorphous films

on anatase (101) surfaces. Appl Phys Lett 2007;91. 061909.

[61] Luo J, Chiang Y-M, Cannon RM. Nanometer-thick surficial films in

oxides as a case of prewetting. Langmuir 2005;21:7358e65.

[62] Luo J, Chiang Y-M. Existence and stability of nanometer-thick disor-

dered films on oxide surfaces. Acta Mater 2000;48:4501e15.

[63] Luo J, Chiang Y-M. Equilibrium-thickness amorphous films on {11-20}

surfaces of Bi2O3-doped ZnO. J Eur Ceram Soc 1999;19:697e701.

[64] Chong J, Xun SD, Song XY, Ridgway P, Liu G, Battaglia VS. Towards

the understanding of coatings on rate performance of lifepo4. J Power

Sources 2012;200:67e76.

[65] Seah MP. Grain boundary segregation. J Phys F Metal Phys

1980;10:1043e64.

[66] Hondros ED, Seah MP. The theory of grain boundary segregation in

terms of surface adsorption analogues. Metall Trans 1977;8A:1363e71.

[67] Cahn JW. Transition and phase equilibria among grain boundary

structures. J de Physique 1982;43. C6.

[68] Cahn JW. Critical point wetting. J Chem Phys 1977;66:3667e72.

31J. Luo / Journal of Materiomics 1 (2015) 22e32

[69] Kikuchi R, Cahn JW. Grain boundary melting transition in a two-

dimensional lattice-gas model. Phys Rev B 1980;21:1893e7.

[70] Kikuchi R, Cahn JW. Grain boundaries with impurities in a two-

dimensional lattice-gas model. Phys Rev B 1987;36:418.

[71] Tang M, Carter WC, Cannon RM. Grain boundary transitions in binary

alloys. Phys Rev Lett 2006;97. 075502.

[72] Tang M, Carter WC, Cannon RM. Diffuse interface model for structural

transitions of grain boundaries. Phys Rev B 2006;73. 024102.

[73] Bishop CM, Cannon RM, Carter WC. A diffuse interface model of

interfaces: grain boundaries in silicon nitride. Acta Mater

2005;53:4755e64.

[74] Kaplan WD, Chatain D, Wynblatt P, Carter WC. A review of wetting

versus adsorption, complexions, and related phenomena: the Rosetta

stone of wetting. J Mater Sci 2013;48:5681e717.

[75] Tang M, Carter WC, Cannon RM. Grain boundary order-disorder

transitions. J Mater Sci 2006;41:7691e5.[76] Bishop CM, Tang M, Cannon RM, Carter WC. Continuum modelling

and representations of interfaces and their transitions in materials.

Mater Sci Eng A 2006;422:102e14.[77] Wynblatt P, Chatain D. Solid-state wetting transitions at grain bound-

aries. Mater Sci Eng A 2008;495:119e25.

[78] Wynblatt P, Chatain D. Anisotropy of segregation at grain boundaries

and surfaces (vol 37a, pg 2595, 2006). Metall Mater Trans A

2007;38A:438e9.

[79] Wynblatt P, Chatain D. Anisotropy of segregation at grain boundaries

and surfaces. Metall Mater Trans A 2006;37A:2595e620.

[80] Luo J, Shi XM. Grain boundary disordering in binary alloys. Appl Phys

Lett 2008;92:101901.

[81] Luo J. Grain boundary complexions: the interplay of premelting, pre-

wetting, and multilayer adsorption. Appl Phys Lett 2009;95. 071911.

[82] Luo J, Cheng H, Asl KM, Kiely CJ, Harmer MP. The role of a bilayer

interfacial phase on liquid metal embrittlement. Science

2011;333:1730e3.

[83] Luo J. Developing interfacial phase diagrams for applications in acti-

vated sintering and beyond: current status and future directions. J Am

Ceram Soc 2012;95:2358e71.

[84] Baram M, Chatain D, Kaplan WD. Nanometer-thick equilibrium films:

the interface between thermodynmaics and atomistics. Science

2011;332:206e9.

[85] Baram M, Kaplan WD. Intergranular films at au-sapphire interfaces. J

Mater Sci 2006;41:7775e84.

[86] Dillon SJ, Harmer MP. Multiple grain boundary transitions in ceramics:

a case study of alumnia. Acta Mater 2007;55:5247e54.

[87] Dillon SJ, Tang M, Carter WC, Harmer MP. Complexion: a new concept

for kinetic engineering in materials science. Acta Mater

2007;55:6208e18.

[88] Harmer MP. Interfacial kinetic engineering: how far have we come

since kingery's inaugural sosman address? J Am Ceram Soc

2010;93:301e17.[89] Harmer MP. The phase behavior of interfaces. Science

2011;332:182e3.

[90] Cantwell PR, Tang M, Dillon SJ, Luo J, Rohrer GS, Harmer MP.

Overview no. 152: grain boundary complexions. Acta Mater

2014;62:1e48.

[91] Ma S, Meshinchi Asl K, Tansarawiput C, Cantwella PR, Qi M,

Harmer MP, et al. A grain-boundary phase transition in si-au. Scr Mater

2012;66:203e6.

[92] Kundu A, Asl KM, Luo J, Harmer MP. Identification of a bilayer grain

boundary complexion in Bi-doped Cu. Scr Mater 2012;68:146e9.

[93] Shi X, Luo J. Developing grain boundary diagrams as a materials sci-

ence tool: a case study of nickel-doped molybdenum. Phys Rev B

2011;84. 014105.

[94] Shi X, Luo J. Decreasing the grain boundary diffusivity in binary alloys

with increasing temperature. Phys Rev Lett 2010;105:236102.

[95] Shi X, Luo J. Grain boundary wetting and prewetting in Ni-doped Mo.

Appl Phys Lett 2009;94:251908.

[96] Dillon SJ, Harmer MP, Luo J. Grain boundary complexions in ceramics

and metals: an overview. JOM 2009;61(12):38e41.

[97] Ma S, Cantwell PR, Pennycook TJ, Zhou N, Oxley MP, Leonard DN,

et al. Grain boundary complexion transitions in WO3- and CuO-doped

TiO2 bicrystals. Acta Mater 2013;61:1691e704.

[98] Luo J, Wang H, Chiang Y-M. Origin of solid state activated sintering in

Bi2O3-doped ZnO. J Am Ceram Soc 1999;82:916.

[99] Luo J, Gupta VK, Yoon DH, Meyer HM. Segregation-induced grain

boundary premelting in nickel-doped tungsten. Appl Phys Lett

2005;87:231902.

[100] Gupta VK, Yoon DH, Meyer III HM, Luo J. Thin intergranular films

and solid-state activated sintering in nickel-doped tungsten. Acta Mater

2007;55:3131e42.

[101] Luo J. Liquid-like interface complexion: from activated sintering to grain

boundary diagrams. Curr Opin Solid State Mater Sci 2008;12:81e8.

[102] Dillon SJ, Harmer MP. Demystifying the role of sintering additives with

“complexion”. J Eur Ceram Soc 2008;28:1485e93.

[103] Lee J-H. Highly resistive intergranular phases in solid electrolytes: an

overview. Monatsh Chem 2009;140:1081e94.

[104] Ma C, Chen K, Liang C, Nan C-W, Ishikawa R, More K, et al. Atomic-

scale origin of the large grain-boundary resistance in perovskite li-ion-

conducting solid electrolytes. Energy Environ Sci 2014;7:1638e42.

[105] Tang M, Carter WC, Chiang YM. Electrochemically driven phase tran-

sitions in insertion electrodes or lithium-ion batteries: examples in

lithiummetal phosphate olivines. Annu RevMater Res 2010;40:501e29.[106] Tang M, Huang HY, Meethong N, Kao YH, Carter WC, Chiang YM.

Model for the particle size, overpotential, and strain dependence of

phase transition pathways in storage electrodes: application to nano-

scale olivines. Chem Mater 2009;21:1557e71.[107] Li X, Gloter A, Gu H, Luo J, Cao X, Jin P, et al. Discovery of nanoscale

reduced surfaces and interfaces in vo2 thin films as a unique case of

prewetting. Scr Mater 2014;78e79:41e4.[108] Maier J. Ionic and mixed conductors for electrochemical devices.

Radiat Eff Defect Solid 2003;158:1e10.

[109] Maier J. Nano-ionics: more than just a fashionable slogan. J Electro-

ceram 2004;13:593e8.[110] Kim S, Fleig J, Maier J. Space charge conduction: Simple analytical

solutions for ionic and mixed conductors and application to nano-

crystalline ceria. Phys Chem Chem Phys 2003;5:2268e73.

[111] Liang CC. Conduction characteristics of the lithium iodide-aluminum

oxide solid electrolytes. J Electrochem Soc 1973;120:1289e92.

[112] Wagner C. The electrical conductivity of semi-conductors involving

inclusions of another phase. J Phys Chem Solid 1972;33:1051e9.

[113] Kumar B, Thomas D, Kumar J. Space-charge-mediated superionic

transport in lithium ion conducting glass-ceramics. J Electrochem Soc

2009;156:A506e13.

[114] Schirmeisen A, Taskiran A, Fuchs H, Bracht H, Murugavel S, Roling B.

Fast interfacial ionic conduction in nanostructured glass ceramics. Phys

Rev Lett 2007;98:225901.

[115] Jain A, Saha S, Gopalan P, Kulkarni A. Ionic conductivity of Na2SO4-

Al2O3 composite electrolytes: mechanism and the role of the prepara-

tory parameters. J Solid State Chem 2000;153:287e93.

[116] Bobade SM, Kulkarni AR, Gopalan P. Investigation of the conductivity

of Na2SO4-PbTiO3 composite electrolyte system: non linear electrical

behavior at curie temperature of the dispersoid. Solid State Ionic

2007;178:1585e9.

[117] Bobade SM, Kulkarni AR, Gopalan P. Electrical properties of Na2SO4-

based composite systems. Ionics 2007;13:257e62.[118] Saito Y, Nakamura O, Yamamoto Y. Conductivity enhancement inves-

tigation at the interface between sodium-ion conductive Na4Zr2Si3O12

and solid superacid particles. J Mater Sci 1996;31:2345e8.

[119] Saito Y, Asai T, Nakamura O, Yamamoto Y. Ionic-conductivity

enhancement of Na4Zr2Si3O12 by dispersed solid superacid SO-2(4)-/

ZrO2. Solid State Ionic 1989;35:241e5.

[120] Bunde A, Dieterich W, Roman E. Monte-carlo studies of ionic con-

ductors containing an insulating 2nd phase. Solid State Ionic

1986;18e9:147e50.

[121] Roman HE, Bunde A, Dieterich W. Conductivity of dispersed ionic

conductors e a percolation model with 2 critical-points. Phys Rev B

1986;34:3439e45.

32 J. Luo / Journal of Materiomics 1 (2015) 22e32

[122] Sata N, Eberman K, Eberl K, Maier J. Mesoscopic fast ion conduction

in nanometre-scale planar heterostructures. Nature 2000;408:946e9.

[123] Li C, Guo X, Gu L, Samuelis D, Maier J. Ionic space-charge depletion

in lithium fluoride thin films on sapphire (0001) substrates. Adv Funct

Mater 2011;21:2901e5.

[124] Li C, Gu L, Guo X, Samuelis D, Tang K, Maier J. Charge carrier

accumulation in lithium fluoride thin films due to li-ion absorption by

titania (100) subsurface. Nano Lett 2012;12:1241e6.[125] Li C, Gu L, Maier J. Enhancement of the li conductivity in lif by

introducing glass/crystal interfaces. Adv Funct Mater 2012;22:1145e9.

[126] Peters A, Korte C, Hesse D, Zakharov N, Janek J. Ionic conductivity

and activation energy for oxygen ion transport in superlattices - the

multilayer system csz (ZrO2þCaO)/Al2O3. Solid State Ionic

2007;178:67e76.

[127] Korte C, Peters A, Janek J, Hesse D, Zakharov N. Ionic conductivity

and activation energy for oxygen ion transport in superlattices e the

semicoherent multilayer system ysz (ZrO2þ9.5 mol% y(2)o(3))/y(2)

o(3). Phys Chem Chem Phys 2008;10:4623e35.

[128] Korte C, Schichtel N, Hesse D, Janek J. Influence of interface structure

on mass transport in phase boundaries between different ionic materials

experimental studies and formal considerations. Monatsh Fur Chem

2009;140:1069e80.

[129] Schichtel N, Korte C, Hesse D, Janek J. Elastic strain at interfaces and

its influence on ionic conductivity in nanoscaled solid electrolyte thin

films-theoretical considerations and experimental studies. Phys Chem

Chem Phys 2009;11:3043e8.

[130] Schichtel N, Korte C, Hesse D, Zakharov N, Butz B, Gerthsen D, et al.

On the influence of strain on ion transport: microstructure and ionic

conductivity of nanoscale ysz vertical bar Sc2O3 multilayers. Phys

Chem Chem Phys 2010;12:14596e608.[131] Aydin H, Korte C, Rohnke M, Janek J. Oxygen tracer diffusion along

interfaces of strained Y2O3/YSZ multilayers. Phys Chem Chem Phys

2013;15:1944e55.

[132] Korte C, Keppner J, Peters A, Schichtel N, Aydin H, Janek J. Coherency

strain and its effect on ionic conductivity and diffusion in solid elec-

trolytes e an improved model for nanocrystalline thin films and a re-

view of experimental data. Phys Chem Chem Phys 2014;16:24575e91.

[133] Garcia-Barriocanal J, Rivera-Calzada A, Varela M, Sefrioui Z, Iborra E,

Leon C, et al. Colossal ionic conductivity at interfaces of epitaxial

ZrO2: Y2O3/SrTiO3 heterostructures. Science 2008;321:676e80.

[134] Guo X. Comment on “colossal ionic conductivity at interfaces of

epitaxial ZrO2:Y2O3/SrTiO3 heterostructures”. Science 2009:324.

[135] Li WQ, Garofalini SH. Molecular dynamics simulations of li insertion

in a nanocrystalline V2O5 thin film cathode. J Electrochem Soc

2005;152:A364e9.[136] Harley G, Yu R, De Jonghe LC. Proton transport paths in lanthanum

phosphate electrolytes. Solid State Ionic 2007;178:769e73.

[137] Mei A, Wang XL, Lan JL, Feng YC, Geng HX, Lin YH, et al. Role of

amorphous boundary layer in enhancing ionic conductivity of lithium-

lanthanum-titanate electrolyte. Electrochim Acta 2010;55:2958e63.

Jian Luo graduated from Tsinghua University with

dual Bachelor's degrees. After receiving his M.S. and

Ph.D. degrees from MIT, he worked in the industry

for more than two years with Lucent Technologies

and OFS/Fitel. In 2003, Luo joined the Clemson

faculty, where he served as an Assistant/Associate/

Full Professor of Materials Science and Engineering.

In 2013, he moved to UCSD as a Professor of

NanoEngineering and Professor of Materials Science

and Engineering. Luo chaired the Basic Science Di-

vision of the American Ceramic Society for

2012e2013, and he was named as a National Security Science and Engi-

neering Faculty Fellow in 2014.