Lithium Expulsion from the Solid-State Electrolyte Li6.4La3Zr1.4Ta0

Jan 18, 2018 - Because of the high mobility of lithium ions in the electrolyte, the expulsed substance may just ..... 2012, 15, A68 DOI: 10.1149/2.024...
0 downloads 4 Views 4MB Size
Download PDF
Research Article Cite This: ACS Appl. Mater. Interfaces 2018, 10, 5978−5983

www.acsami.org

Lithium Expulsion from the Solid-State Electrolyte Li6.4La3Zr1.4Ta0.6O12 by Controlled Electron Injection in a SEM Xiaowei Xie,† Juanjuan Xing,† Dongli Hu,† Hui Gu,*,† Cheng Chen,‡,§ and Xiangxin Guo*,‡,∥ †

School of Materials Science and Engineering, Materials Genome Institute, Shanghai University, Shanghai 200444, China Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, China § University of the Chinese Academy of Sciences, Beijing 100049, China ∥ College of Physics, Qingdao University, Qingdao 266071, China ‡

S Supporting Information *

ABSTRACT: The garnet ionic conductor is one of the promising candidate electrolytes for all-solid-state secondary lithium batteries, thanks to its high lithium ion conductivity and good thermal and chemical stability. However, its microstructure is difficult to approach because it is very sensitive to the inquisitive electron beam. In this study based on a scanning electron microscope (SEM), we found that the electron beam expulses the lithium out of Li6.4La3Zr1.4Ta0.6O12 (LLZTO), and the expulsed zone expands to where a stationary beam could extend and penetrate. The expulsion of metallic lithium was confirmed by its oxidation reaction after nitrogen inflow into the SEM. This phenomenon may provide us an effective probe to peer into the conductive nature of this electrolyte. A frame-scan scheme is employed to measure the expulsion rate by controllable and more uniform incidence of electrons. Lithium accumulation processes are continuously recorded and classified into four modes by fitting its growth behaviors into a dynamic equation that is mainly related to the initial ion concentration and ion migration rate in the electrolyte. These results open a novel possibility of using the SEM probe to gain dynamic information on ion migration and lithium metal growth in solid materials. KEYWORDS: lithium ion solid-state electrolyte, cubic garnet Li6.4La3Zr1.4Ta0.6O12, electron beam irradiation, lithium expulsion, ion migration rate

1. INTRODUCTION

Despite these advantages, the chemical stability of LLZO is susceptible toward moisture, air, and electron beam. Cheng et al. found that the lithium content was very high on the surface of LLZO exposed to air for 2 months as revealed by the laserinduced breakdown spectroscopic cross-sectional mapping,13 and the surface component was detected as Li2CO3 by X-ray photoelectron spectroscopy. The effect of beam irradiation on lithium battery materials has recently been reported by several theoretical and experimental studies,14−18 but this still remains absent for its connection to the highly ion-conductive LLZO. There was a high-resolution transmission electron microscopy study for LLZO, but little information on lithium distribution and behavior was obtained because of its quick amorphization, which is probably due to the easy decomposition induced by the electron beam.19 This limits the research on its structural nature and microstructure behaviors. It is recently reported that suitable Ta doping in LLZO, that is, Li7−xLa3Zr2−xTaxO12 (x = 0.6) or LLZTO, could significantly

The lithium ion conductor Li7La3Zr2O12 (LLZO) is a key solid electrolyte that attracts a lot of attention because of its high lithium ion conductivity and good thermal and chemical stability as well as the availability of dense materials, which are all beneficial factors for all-solid-state secondary lithium batteries.1 LLZO may exist in two structures subject to the sintering temperature,2,3 and its ionic conductivity could vary by 2 orders of magnitude from 1.63 × 10−6 to 3 × 10−4 S/cm, depending on whether it is in tetragonal or cubic forms.1,3 Although both structures are composed of dodecahedral LaO8 and octahedral ZrO6, the location of lithium ion and its transmission path are rather different.3−5 The cubic LLZO possesses isotropic three-dimensional ion transport channels and a high concentration of vacancies to favor the migration of lithium ions and hence a high ionic conductivity.6 To stabilize the cubic phase, substitution of Zr by Al, Si, Ta, W, or Nb has been extensively studied.7−12 In particular, Wang and Lai found that by partial substitution of Ta for Zr, a fully cubic phase was achieved via transformation from a tetragonal phase, which increased the total conductivity to 6.9 × 10−4 S/cm.10 © 2018 American Chemical Society

Received: November 13, 2017 Accepted: January 18, 2018 Published: January 18, 2018 5978

DOI: 10.1021/acsami.7b17276 ACS Appl. Mater. Interfaces 2018, 10, 5978−5983

Research Article

ACS Applied Materials & Interfaces improve the lithium discharge and cyclic performance to make it a better solid electrolyte and support for electrodes while its ion conductivity could reach 1.6 × 10−3 S/cm.20 To avoid or minimize the beam irradiation on this garnet-type lithium ion conductor, we attempt to investigate the structural behavior of LLZTO in a scanning electron microscope (SEM) and find that certain substances could be expulsed from the surface of the bulk electrolyte material, even under a relatively moderate electron beam, which is induced by the requirement of electrochemical neutrality. Elucidation of the relationship between the controllable electron probe and the rate of expulsion in this electrolyte material might help us to better understand the mechanism of ion conductivity and lithium metal growth in LLZTO as well as in other solid-electrolyte materials.

2. EXPERIMENTAL SECTION LLZTO ceramics were prepared via conventional solid-state reactions, from stoichiometric LiOH (Alfa Aesar, 99.995%), La(OH)3 (Alfa Aesar, 99.95%), ZrO2 (Aladdin Reagent, 99.99%), Ta2O5 (Aladdin Reagent, 99.99%), and a 15 wt % excess of LiOH to compensate volatile Li during synthesis. The powders were ball-milled for 12 h before being heated in air at 900 °C for another 12 h to ensure the formation of a cubic LLZTO phase and then ball-milled for 12 h again. Afterward, the resultant powders were pressed into pellets with a diameter of 12 mm at 100 MPa and sintered at 1140 °C for 9 h in oxygen. During sintering, the pellets were covered with mother powders and put in alumina crucibles in a tube furnace.21 To reach high surface flatness of the solid-electrolyte pellet for microstructural observation and analysis, 3, 1, and 0.5 μm corundum emery papers were used in succession during polishing, which leaves very few scratches on the smooth sample surfaces. Not only high-purity alcohol (99.7%) was poured intermittently onto the emery papers to lubricate and protect the polished surfaces, but also they were cleaned by alcohol-soaked cotton swabs under a roasting lamp to avoid moisture that could react with the naked surface, before being immediately placed onto the SEM stage to start venting. Microstructures of LLZTO were characterized by a SEM (model Supra55, G300, Carl Zeiss, Germany) in secondary electron (SE) and backscattered electron (BSE) imaging modes. Elemental analysis was carried out with an energy-dispersive spectrometer (model AZtec, Oxford Instrument, UK) attached to a SEM with an electron beam accelerating voltage of 8 kV, which covers all involved elements except for the undetectable lithium. The Monte Carlo simulation of electron trajectory was made by the commercial program Casino. Figure 2. Expulsion of a substance from the polished LLZTO surface induced by irradiation of a stationary electron beam at 10 kV; (a) SE images of expulsion sequence with the probe position marked as red; (b) increase of the accumulation area of the expulsed materials with time, which fits into a parabolic curve; (c) intensity distribution of the electron probe inside LLZTO bulk, simulated by Monte Carlo method.

3. RESULTS AND DISCUSSION 3.1. Expulsion of a Substance by a Stationary Electron Beam. Although X-ray diffraction analysis exhibited a monolithic cubic phase for this LLZTO material,21 SEM

imaging at 10 kV in the BSE mode reveals not only a rather dense ceramic body, which reveals it to be a good solid-state electrolyte, but also nonuniformity in local composition for grains with a size of 5 μm, as shown in Figure 1. The latter is shown as intragranular contrast that reflects the inhomogeneous distribution of relative mass within the LLZTO phase, and the white contrast should come from regions with a high concentration of heavy elements, that is, lanthanum and tantalum. Such a chemical inhomogeneity is most likely the elemental segregation during preparation. When irradiating by a stationary electron beam, white substances in the SE mode always expulse out from the surface of LLZTO; one case is

Figure 1. BSE imaging of LLZTO at 10 kV. 5979

DOI: 10.1021/acsami.7b17276 ACS Appl. Mater. Interfaces 2018, 10, 5978−5983

Research Article

ACS Applied Materials & Interfaces

Figure 3. EDS spectra taken from the expulsed substance (a) before and (b) after nitrogen inflow and their difference (c) exhibiting excessive oxygen due to the oxidation from this substance.

given in Figure 2 with a moderate probe current of 285.4 × 10−12 A at 10 kV. The expulsed substance starts to accumulate on the surface around the electron beam (the red spot) within one or few seconds of irradiation, and the accumulated substance grows while its visible area enlarges by continuing irradiation until pile-up, as shown in Figure 2a. Areas of the accumulated substance are measured against the irradiation time as plotted in Figure 2b, indicating that the expulsion volume increases in a decreasing rate to follow an exponential or power-law behavior. Monte Carlo simulation of this incident electron beam at 10 kV reveals that the peering probe penetrates into the bulk of LLZTO in a depth and size of about 500 and 700 nm, respectively, as shown in Figure 2c. This indicates that the injection of electrons into the solid-state electrolyte creates a driving force in an extended area to expulse a substance out from this LLZTO structure. 3.2. Confirmation of the Expulsed Substance as Metal Lithium. Because of the high mobility of lithium ions in the electrolyte, the expulsed substance may just be lithium ions reduced directly to metallic lithium by the incident electrons. Although the energy-dispersive spectrometry (EDS) method could not directly detect lithium, we employed an indirect method to measure the change of other elements, especially oxygen that may come out together with lithium, which would easily react with lithium, hence blocking nitration or further intake of oxygen. Nitrogen was introduced into the SEM chamber after the expulsion occurred; hence, the residual oxygen brought by flowing nitrogen is sufficient to oxidize the expulsed substance during the venting of the SEM chamber in a reasonably slow rate without affecting the native surface of LLZTO. EDS spectra were collected from the expulsed substances before and after nitrogen inflow, chosen at random over more than 20 cases, because their shapes have changed because of the reaction. As exemplified in Figure 3, the different spectra reveal always the intake of oxygen by the expulsed substance after nitrogen inflow that brings the dry air. Except for the increase of oxygen intensity, all detectable cations remain intact through the inflow process, indicating that the expulsed substance reacted with oxygen should be the undetectable metal lithium. There is also no detection of nitrogen in this analysis, indicating that the nitration reaction was either absent or occurred at a much slower rate than the oxidation for the expulsed lithium. It is interesting to find out the expulsed substance also collapsed after oxidation, indicating that the oxidation has collapsed the initial structure of expulsion. Further EDS analysis reveals no change in the LLZTO composition induced by the nitrogen inflow (see Figure S1); hence, the observed increase of oxygen originates only from its reaction with the expulsed substance, unaffected

for the nearby bulk. In other words, all remaining components in the electrolyte were not involved in the oxidation reaction, except for the expulsed metal lithium that could not be directly analyzed. Nonetheless, the expulsed substance may still be composed of low or trace amount of other elements besides the dominant lithium in the limit that their expulsion will not significantly affect the structure of LLZTO in any apparent way. 3.3. Delayed Lithium Expulsion at Different Rates by Frame Scan. The stationary probe interacts strongly with the LLZTO electrolyte to expulse lithium once the beam is placed on any position of the sample surface (Figure 2b). To understand the expulsion behavior as well as the underlining mechanism, it is necessary to lower the electron irradiation rate to expose the details in the expulsion process. We employ the fast frame-scan mode in the SEM to record continuous images every 10 s in the video mode without interruption, which also enlarges the irradiated area and makes the electron intensity more uniform to minimize the effect of the probe shape. Lithium expulsion from various locations at different rates and behaviors could be recorded in a same videoed series: the growth processes of many lithium expulsions from the flat LLZTO surface are presented in Figure 4a, with intervals at every 50 s between successive frames. It is revealed that the expulsion of lithium occurs from inside the grains rather than at grain boundaries or from triple points, which were often left with small holes in black contrast where no expulsion was found. In the meanwhile, each expulsion starts at a different time and grows into its own shape in a certain rate. As marked in Figure 4b, all of these individual lithium expulsion processes could be categorized into four groups according to their final sizes and shapes. The first group of lithium expulsion is encircled in dark lines: they all reach a micron size and grow rapidly, starting quite late after 5 min of irradiation. The second and third groups of expulsion are encircled in blue and green lines, respectively, and they grow correspondingly into submicron and nanoscale sizes in increasingly slower rates as compared to the first group, starting almost half times earlier. It is interesting to find a very fast expulsion of lithium into a strip shape, or effectively the ejection of lithium, which is marked in a red line and categorized as the fourth group. All of these four representative behaviors of lithium expulsion are plotted against time in Figure 4c, corresponding to the numbered processes in Figure 4b. These curves demonstrate not only that the accumulation rates of lithium expulsion are remarkably different but that the starting time of lithium expulsion is also variable among these groups: it is earlier for the slow accumulation rate, whereas it is delayed for the rapid growth, which is also true even for the last group where the 5980

DOI: 10.1021/acsami.7b17276 ACS Appl. Mater. Interfaces 2018, 10, 5978−5983

Research Article

ACS Applied Materials & Interfaces

Table 1. Lithium Expulsion Parameters Resulted by Fitting the Expulsion Curves in Figure 4c into Formula 6 position

a

b × 103

t0/s

fitting correlation factors

① ② ③ ④

0.9340 0.3292 0.0971 0.2331

10 5.36 3.72 13.81

302 256 138 380

0.9881 0.9931 0.9926 0.9796

the lithium expulsion, the rate of expulsed metal accumulation can be written as below to follow Saito et al.:22 dS = k1σE − k 3 dt

(1)

where S is the accumulated area of expulsion to measure the quantity of the expulsed substance, t is the time of irradiation, σ is the ionic conductivity, E is the strength of the electrical field in the electrolyte created by injected electrons from the beam, and k1 is the coefficient of expulsion, which is assumed as a constant. k3 is the evaporation rate of the accumulated metal, depending on its vapor pressure in the SEM, which can be ignored because of the good stability of lithium in vacuum. For lithium conductivity in LLZTO, we can use the general expression

σ = nZeu

(2)

where n, Ze, and u represent the mobile lithium ion concentration, the number of charges per ion which is 1, and the ionic mobility, respectively. The actual ion concentration during irradiation when it is affected by the accumulated metal:

n = n0 − k 2S

(3)

where n0 is the initial concentration of the mobile ion and k2 is a coefficient, taking again as a constant. Therefore, we have dS = k1n0uE − k1k 2uES = k1σ0E − k1k 2σES /n dt

(4)

where σ0 is the initial ionic conductivity. When the quantity of metal accumulation S is small as compared to the irradiated region within the electrolyte, the strength of the electrical field E could be considered as constant. Then, the logarithmic integration of formula 4 turns it into In(k1σ0E − k1k 2σES /n) = −k1k 2σEt /n + C

(5)

where C is a constant from integration. It is reasonable to set S = 0 at t = 0 when S is estimated from the accumulation area in the SEM images, which leads to

Figure 4. Lithium expulsion behaviors recorded under continuous frame scan every 10 s: (a) frames excerpt every 50 s; (b) four categories of expulsion behaviors (color-coded) distinguished by their shapes and sizes; and (c) representative growth curve of the four categories marked in (b).

S = a[1 − exp(−bt )]

(6)

a = n0 / k 2

(7)

b = k1k 2σE /n = k1k 2uE

(8)

Therefore, time-dependent accumulation of the lithium expulsion process could be quantitatively described, where the exponential constant b represents both the effects of lithium ion mobility and the expulsion filed created by the injected electrons. The phenomenon of lithium expulsion fundamentally reflects the ionic migration characteristics of LLZTO as a solid electrolyte. Table 1 summarizes the expulsion constants and correlation factors obtained by fitting the representative accumulation curves in Figure 4c into formula 6. The parameter

sudden ejection is effectively a very rapid expulsion to exceed even the frame recording rate. A higher rate of accumulation corresponds to a later starting time of expulsion; this indicates that the internal driving force for the expulsion process should be stronger. 3.4. Dynamic Equation To Describe Lithium Accumulation. To understand the underlying mechanism that drives 5981

DOI: 10.1021/acsami.7b17276 ACS Appl. Mater. Interfaces 2018, 10, 5978−5983

Research Article

ACS Applied Materials & Interfaces

Figure 5. Schematic of lithium expulsion from LLZTO after electron irradiation in (a) stationary beam and (b) frame-scan modes.

4. CONCLUSIONS To investigate the influence of beam irradiation on the structural behavior of LLZTO, we have found that the lithium metal could be expulsed from the surface of the bulk electrolyte material under a moderate electron beam in a SEM. The fast scanning beam reveals different behaviors of lithium expulsion, which could be fitted into a dynamic equation to gain parameters that are related to the initial ion concentration and ion migration rate in the electrolyte. The injection of electrons into the bulk of LLZTO forms a negative electric field to create lithium metal out of the materials surface; this might help us better understand the mechanism of lithium metal growth during polarization with great current densities in LLZTO as well as in other solid-electrolyte materials.

a is the destined area of accumulation at infinite time according to formula 6; hence, the larger the observed expulsion area, the bigger the a value , except for the fourth case where bending of the nanostrip onto the surface has overcounted the expulsion area. For the parameter b, the larger value corresponds to the faster rate of accumulation, which is even true for the fourth case, with the ejection rate exceeding the frame recording speed. In addition, the parameter b is correlated inversely with the starting time of expulsion, or t0, as also listed in Table 1. Such a correlation between t0 and b indicates that to initiate lithium expulsion, the driving electric field needs to be built up by the incident electron beam until certain strength: the stronger the expulsion with higher accumulation rate, the longer time (t0) it will require to build up stronger electric filed inside the LLZTO electrolyte. Differentiating the lithium expulsion into four modes under a fast-framing electron beam equivalent to uniform electron irradiation indicates that the different expulsion processes are associated with certain microstructural characters: the variation in grain orientation, the compositional inhomogeneity, unknown surface defects or invisible interfaces, or any combination of these. To further explore and identify the microstructural origin, future study is necessary, especially with elaborate modeling and simulation to verify and rationalize these experimentally discovered phenomena. The expulsions of lithium driven by internal electric fields created from both the stationary and fast-scan electron probes are schematically described in Figure 5a,b, respectively. Accelerated by a 10 kV electric field, the incident electrons could penetrate deep and wide into the bulk electrolyte to leave themselves inside the electronically insulating LLZTO lattice, and some of them may combine with the lithium ions to create neutral lithium atoms. After certain time of accumulation, the excessive electrons create an internal electron field that attracts nearby lithium ions to move toward the charged zones, which expels the concentrated lithium to channel out the electrolyte from certain surface defects or from grains with certain crystalline orientation or chemical inhomogeneity. This phenomenon of metal lithium expulsion from the electrolyte bears strong similarity to the creation of metal lithium dendrites in lithium batteries. The former case is initiated by a fast electron beam penetrating deep into the solid electrolyte to create an internal driving field, and the latter is driven by polarization with a large current at the interface between electrolyte and electrode during galvanostatic cycling; both are controlled by the unexpected electric fields. The effect of such internal electric fields and the way to control the lithium expulsion should be further studied in future work, which may bring this similarity even closer and truer.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b17276. Full footage video showing how each expulsion event, including the ejection of a strip, proceeded, from which all frames in Figure 4b were intercepted (ZIP) EDS spectra of the sample surface before and after nitrogen inflow and before and after the occurrence of accumulation (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (H.G.). *E-mail: [email protected] (X.G.). ORCID

Xiaowei Xie: 0000-0002-6288-8071 Cheng Chen: 0000-0001-7975-3686 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (grant nos. 51532002, 51532006, and 51771222) and by the Science and Technology Commission of Shanghai Municipality under grant no. 16DZ2260600. The authors would like to thank Prof. Bingkun Guo, Prof. Si-qi Shi, and Dr. Xian-hao Wang for beneficial discussion.

■

REFERENCES

(1) Murugan, R.; Thangadurai, V.; Weppner, W. Fast lithium ion conduction in garnet-type Li7La3Zr2O12. Angew. Chem. 2007, 46, 7778−7781.

5982

DOI: 10.1021/acsami.7b17276 ACS Appl. Mater. Interfaces 2018, 10, 5978−5983

Research Article

ACS Applied Materials & Interfaces (2) Toda, S.; Ishiguro, K.; Shimonishi, Y.; Hirano, A.; Takeda, Y.; Yamamoto, O.; Imanishi, N. Low temperature cubic garnet-type CO2doped Li7La3Zr2O12. Solid State Ionics 2013, 233, 102−106. (3) Awaka, J.; Kijima, N.; Hayakawa, H.; Akimoto, J. Synthesis and structure analysis of tetragonal Li7La3Zr2O12 with the garnet-related type structure. J. Solid State Chem. 2009, 182, 2046−2052. (4) Jalem, R.; Yamamoto, Y.; Shiiba, H.; Nakayama, M.; Munakata, H.; Kasuga, T.; Kanamura, K. Concerted Migration Mechanism in the Li Ion Dynamics of Garnet-Type Li7La3Zr2O12. Chem. Mater. 2013, 25, 425−430. (5) Awaka, J.; Takashima, A.; Kataoka, K.; Kijima, N.; Idemoto, Y.; Akimoto, J. Crystal Structure of Fast Lithium-ion-conducting Cubic Li7La3Zr2O12. Chem. Lett. 2011, 40, 60−62. (6) 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 fast lithiumion conductor. Inorg. Chem. 2011, 50, 1089−1097. (7) Kumazaki, S.; Iriyama, Y.; Kim, K.-H.; Murugan, R.; Tanabe, K.; Yamamoto, K.; Hirayama, T.; Ogumi, Z. High lithium ion conductive Li7La3Zr2O12 by inclusion of both Al and Si. Electrochem. Commun. 2011, 13, 509−512. (8) Jin, Y.; McGinn, P. J. Al-doped Li7La3Zr2O12 synthesized by a polymerized complex method. J. Power Sources 2011, 196, 8683−8687. (9) Rangasamy, E.; Wolfenstine, J.; Sakamoto, J. The role of Al and Li concentration on the formation of cubic garnet solid electrolyte of nominal composition Li7La3Zr2O12. Solid State Ionics 2012, 206, 28− 32. (10) Wang, Y.; Lai, W. High Ionic Conductivity Lithium Garnet Oxides of Li7−xLa3Zr2−xTaxO12 Compositions. Electrochem. Solid-State Lett. 2012, 15, A68. (11) Li, Y.; Wang, Z.; Cao, Y.; Du, F.; Chen, C.; Cui, Z.; Guo, X. WDoped Li7La3Zr2O12 Ceramic Electrolytes for Solid State Li-ion Batteries. Electrochim. Acta 2015, 180, 37−42. (12) Ohta, S.; Kobayashi, T.; Asaoka, T. High lithium ionic conductivity in the garnet-type oxide Li7−x La3(Zr2−x, Nbx)O12 (x=0−2). J. Power Sources 2011, 196, 3342−3345. (13) Cheng, L.; Crumlin, E. J.; Chen, W.; Qiao, R.; Hou, H.; Lux, S. F.; Zorba, V.; Russo, R.; Kostecki, R.; Liu, Z.; Persson, K.; Yang, W.; Cabana, J.; Richardson, T.; Chen, G.; Doeff, M. The origin of high electrolyte-electrode interfacial resistances in lithium cells containing garnet type solid electrolytes. Phys. Chem. Chem. Phys. 2014, 16, 18294−18300. (14) Egerton, R. F. Control of radiation damage in the TEM. Ultramicroscopy 2013, 127, 100−108. (15) Lin, F.; Markus, I. M.; Doeff, M. M.; Xin, H. L. Chemical and structural stability of lithium-ion battery electrode materials under electron beam. Sci. Rep. 2014, 4, 5694. (16) Phillips, P. J.; Iddir, H.; Abraham, D. P.; Klie, R. F. Direct observation of the structural and electronic changes of Li2MnO3 during electron irradiation. Appl. Phys. Lett. 2014, 105, 113905. (17) Grogan, J. M.; Schneider, N. M.; Ross, F. M.; Bau, H. H. Bubble and pattern formation in liquid induced by an electron beam. Nano Lett. 2014, 14, 359−364. (18) Zheng, H.; Liu, Y.; Mao, S. X.; Wang, J.; Huang, J. Y. Beamassisted large elongation of in situ formed Li2O nanowires. Sci. Rep. 2012, 2, 542. (19) Qian, D.; Ma, C.; More, K. L.; Meng, Y. S.; Chi, M. Advanced analytical electron microscopy for lithium-ion batteries. NPG Asia Mater. 2015, 7, No. e193. (20) Du, F.; Zhao, N.; Li, Y.; Chen, C.; Liu, Z.; Guo, X. All solid state lithium batteries based on lamellar garnet-type ceramic electrolytes. J. Power Sources 2015, 300, 24−28. (21) Li, Y.; Wang, Z.; Li, C.; Cao, Y.; Guo, X. Densification and ionic-conduction improvement of lithium garnet solid electrolytes by flowing oxygen sintering. J. Power Sources 2014, 248, 642−646. (22) Saito, Y.; Nakagawa, M.; Yamada, M.; Miyajima, Y.; Yamamoto, Y. Estimation of ionic mobility of sodium ion conductor by scanning electron microscopy observation of sodium metal deposition. J. Mater. Sci. Lett. 1996, 15, 898−901. 5983

DOI: 10.1021/acsami.7b17276 ACS Appl. Mater. Interfaces 2018, 10, 5978−5983