Spatial Distributions of Discharged Products of Lithium–Oxygen

Letter pubs.acs.org/NanoLett

Spatial Distributions of Discharged Products of Lithium−Oxygen Batteries Revealed by Synchrotron X‑ray Transmission Microscopy Mara Olivares-Marín,† Andrea Sorrentino,‡ Rung-Chuan Lee,§ Eva Pereiro,‡ Nae-Lih Wu,§ and Dino Tonti*,† †

Institut de Ciència de Materials de Barcelona, Consejo Superior de Investigaciones Científicas (ICMAB-CSIC), Campus UAB, ES 08193 Bellaterra, Barcelona, Spain ‡ ALBA Synchrotron Light Source, MISTRAL Beamline−Experiments Division, 08290 Cerdanyola del Vallès, Barcelona, Spain § Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan S Supporting Information *

ABSTRACT: The discharge products of ether-based Li−O2 cells were grown directly on common carbon-coated TEM grids and observed by oxidation-statesensitive full field transmission soft X-ray microscopy (TXM). The acquired data have permitted to quantify and localize with spatial resolution the distribution of the oxygen discharge products in these samples (i.e., lithium superoxide, peroxide, and carbonates) and appreciate several compositional, structural, and morphological aspects. Most of the peroxide particles had a toroidal shape, often with a central hole usually open on only one side, and which included significant amounts of superoxide-like phases (LiO2/Li2O2 ratio between 0.2 and 0.5). Smaller particles had smaller or no superoxide content, from which we infer that abundance of soluble LiO2 may have a role in toroid formation. Significant amount of carbonates were found irregularly distributed on the electrode surface, occasionally appearing as small particles and aggregates, and mostly coating lithium peroxide particles. This suggests the formation of a barrier that, similar to the solid electrolyte interface (SEI) critical in Li-ion batteries, requires an appropriate management for a reversible operation. KEYWORDS: Lithium−oxygen batteries, Li2O2, LiO2, soft X-ray microscopy, toroids

T

issues with lithium−oxygen batteries are their poor reversibility and cycle life. However, these aspects are also crucial for the cell capacity, which is limited by the surface passivation, as well as by the available space that can be filled by the discharge products. A precise understanding of the nature and the evolution of the discharge products is therefore crucial in designing and improving electrode materials, electrolytes, and operating conditions. For instance, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) studies have revealed that increasing discharge currents7 or using ionic liquid electrolytes8 leads to smaller electrodeposited particles, which in turn are easier to remove during charge. Recent work suggests that the solvent in the electrolyte affects the ORR by its ability to solubilize the lithium superoxide (LiO2) intermediate.9,10 The cathode ORR is in fact a multistep reaction, with two main different possible routes:

he huge interest in lithium−oxygen batteries has been frustrated by their poor reversibility in comparison to commercial lithium-ion batteries.1 In a conventional lithium battery the cathode interface is essentially preserved during operation, in fact the electrochemical reaction is a Li insertion into the active electrode material. Instead, in a Li/O2 cathode the reaction is rather an electrodeposition process, where molecular oxygen from the electrolyte is reduced and combines with lithium cations to form mainly peroxide on the electrode:1,2 O2 + 2Li+ + 2e− → Li 2O2 ↓

Unfortunately, the oxygen radicals involved in this oxygen reduction reaction (ORR), or, during charge, in its inverse oxygen evolution reaction (OER) are extremely reactive with the organic environment,2,3 and even with carbon.4 In particular, significant amounts of Li2CO3, HCO2Li, CH3CO2Li, polyethers/esters, CO2, and H2O have been found and described as common solvent decomposition products of ether-based electrolytes.5 In addition, lithium peroxide (Li2O2) is an insulator that passivates the electrode surface, and in many cases the reaction cannot be quantitatively reverted because of particles that become electrically disconnected from the electrode.6 Mainly for these two factors the most relevant © XXXX American Chemical Society

Li+ + O2 + e− → LiO2

(EC1)

Li+ + LiO2 + e− → Li 2O2

(EC2)

2LiO2 → Li 2O2 + O2

(C2)

Received: July 20, 2015

A

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters LiO2 is generated at a first electrochemical step EC1, while either a second electrochemical reduction EC2 or a chemical disproportionation C2 will produce Li2O2. LiO2 solubility seems to lead to solution peroxide precipitation,9,10 with formation of complex shapes, often as characteristic large toroids. In reality, there is now sufficient evidence that the precipitate often contains significant amounts of the superoxide intermediate, which is believed more conducting and easier to oxidize than peroxide.11,12 Superoxide has been reported as the main observed product when Na or K are used instead of Li.13 Significantly, Na/O2 and K/O2 cells have remarkably higher reversibility than Li/O2, even though their energy densities are indeed smaller, due to less reduced state of oxygen.14,15 The presence and distribution of insoluble electrolyte decomposition species and of LiO2 in the discharge products is therefore of high relevance for a clear understanding of the factors that can lead to enhanced reversibility in metal−oxygen cells. Given the light elements involved and the poor stability of incompletely reduced Li−O compounds many transmission imaging and other common spectroscopic or diffraction techniques are not suitable for their compositional analysis. While Li2O2 is often sufficiently crystalline to be detected by Xray16 and electron diffractions,17,18 this is usually not the case for Li superoxide. Some X-ray absorption near-edge spectroscopy (XANES),7 electron paramagnetic resonance (EPR) spectroscopy,19 and Raman11,20−22 evidence of “superoxide rich” phases have been given so far. Only very recently some diffraction evidence of stable and crystalline LiO2 has been provided by Zhai et al.23 However, none of these approaches is able to accurately spatially resolve superoxide and peroxide distributions within a single particle. In this work, the spatial distribution of various oxygencontaining species has for the first time been visualized by energy-resolved synchrotron full-field transmission X-ray microscopy realized at the Mistral beamline of the ALBA synchrotron.24,25 This allowed low-dose irradiation and, therefore, can avoid beam damage. A few earlier reports show that X-ray microscopy can be a remarkable support to understand how morphological processes affect the electrochemical behavior of battery materials. Yang et al. have recently applied this technique to study the evolution of chemical composition and morphology of high-voltage Li-ion cathode materials during electrochemical cycling.26 Also, Chueh at al. have combined scanning transmission X-ray microscopy (STXM) and X-ray absorption spectroscopy (XAS) to study intercalation pathway in LiFePO4-based electrodes.27 Wu et al. have followed the evolution of Sn-based anode materials for Liion,28,29 and the formation and decomposition of sulfur particles and different polysulfides while operating a Li−S cell.30 In our case, the spatially and chemically resolved oxygen distributions in combination with morphological information at nanoscale resolution provide valuable information to the understanding of electrolyte decomposition and formation mechanism of the deposited particles, which can help improving the cell capacity and reversibility. In particular, we observe a LiO2-like phase associated with toroidal particles, and a carbonate coating on all Li−O particles, which respectively suggest the implication of superoxide in the formation of large particles and the buildup of a barrier to their dissolution. A common carbon-coated Au TEM grid was used directly as a cathode and fully discharged in a lithium−oxygen battery (discharge profile and experimental conditions are available in the Supporting Information). Figure 1 shows SEM (a,b) and

Figure 1. SEM (a,b) and TEM (c,d) images of different representative regions of the carbon-coated TEM grid based electrode discharged at 100 mA per total gram of carbon. Inset of image d shows the electron diffraction pattern of the toroid selected area (yellow square).

TEM (c,d) images at different magnification of representative regions of the electrochemically discharged cathode. Different elements of interest appear well distributed on the conducting surface. Different morphologies and sizes, from a few nanometers to over 2 μm, can be observed, even on the separator glass fiber. The most common shape observed resembles the characteristic toroid or red blood cell-like particle shape previously reported by several authors for discharged cathodes in Li/O2 batteries.7,16−18,31−35 Electron diffraction pattern of a selected area of the toroid in Figure1d is consistent with the presence of Li2O2, as also found by several authors.7,17,18,31,32,35−38 Moreover, next to the toroidal objects images also reveal other kinds of products such as nanoparticles resembling thin platelets reported by other authors10,39,40 and irregular aggregates. TXM allowed to simultaneously map and discriminate between the principal possible reaction products involved (experiment and data processing details are available in the Supporting Information). Two examples are given in Figure 2, showing a representative region of an electrochemically discharged carbon coated TEM grid (a) and a detail of a toroid deposited on the separator glass fiber (b). The objects of different size, shape, and localizations observed by TEM and SEM (Figure 1) can be recognized in the TXM images. In this case, however, we directly detect oxygen and its chemical state since we obtain a full O K-edge absorption spectrum at each pixel. Figure 3 reports normalized O K-edge XANES spectra integrated over selected areas indicated by arrows in Figure 2. The variability of the local spectra proves that composition is significantly heterogeneous within the lateral resolution provided by the instrument. Furthermore, the large number of different spectra that can be obtained thanks to a large field of view allows to easily recognizing independent coexisting components in the same B

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 2. TXM images (a,b) of a carbon coated Au TEM grid after being fully discharged at 100 mA per total gram of carbon in a Li−O2 cell. The images are the result of overlapping three color maps with intensities proportional to the amounts of Li superoxide (cyan), Li peroxide (green), and carbonate (red). The single color maps are reported in Figure S7.

Li2O2 and 533.8 eV for Li2CO3. These are in excellent agreement with the transitions 1s → σ* (O−O) for peroxide and 1s → π* (C−O) for lithium carbonate.41 Furthermore, experimental XANES spectra depicted in Figure 3 present another component that appears at lower energy than peroxide. This species, highly associated with Li2O2 in toroidal shaped objects, shows a sharp feature at around 528.75 eV that can be assigned to the 1s → π* (O−O) transition of LiO2, in good agreement with measurements by Ruckman et al.42 The lower transition energy for superoxide can be attributed to the much lower energy of the respective final state,42,43 while the energy difference between the same states decreases to a lower extent by increasing the charge on molecular O2.42 As a matter of fact, although peroxide is expected as the main discharge product, several authors have evidenced by different techniques the presence of significant amounts of superoxide in the precipitate.11,20−22,19,42 The attribution of the 533.8 eV transition to carbonate is consistent with the expected partial TEGDME decomposition5 and with the intensity around 539 and 544−546 eV. We do not find evidence of other possible byproducts such as LiOH or Li2O, which should present relative maxima respectively at 542.5 eV (see Figure S3), or between 539 and 541 eV.41 The presence of LiO2, Li2O2, and carbonates, rather than Li2O, also supports that radiation damage can be considered negligible,41 as further discussed in the Supporting Information. Since the absorption edges of differently reduced oxygen atoms have sufficiently different energy, by simply subtracting the corresponding images it is possible to obtain the spatial distribution of each chemical species (see Figure S7). We then overlap the different distributions using different colors to obtain the image shown in Figure 2. The different color intensities in the image are directly proportional to the amounts of lithium superoxide (cyan), peroxide (green), and carbonate (red). Our experimental results indicate that, in general, differently shaped particles present different chemical composition. Remarkably, small platelets and irregular particle aggregates are mainly composed by carbonate and appear essentially red. Instead, Li superoxide and peroxide dominate in toroidal objects. The glass fiber is also red colored, as the

Figure 3. O K-edge XANES spectra at the selected points indicated by arrows in Figure 2a,b and reference spectra measured for Li2O2 and Li2CO3.

sample. A simple integrated spectrum for each sample would be indeed of much more difficult interpretation. We chemically prepared a reference Li peroxide and used commercial inorganic Li2CO3 powders as a reference for carbonate (see Supporting Information). The reference O Kedge spectra show a sharp feature at around 531.25 eV for C

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

blended, forming a Li−O phase composite, which is mainly localized in toroidal particles. The image indicates that this blend is not completely homogeneous, but there is not a typical distribution, for instance, between particle core and surface. This observation does not seem to support the hypothesis recently proposed of a superoxide-rich phase on the surface or edges of Li2O2 particles.7,11,33 Also, by integrating over the whole particle, we did not find a clear trend of the LiO2/Li2O2 ratio with particle size. Nevertheless, we calculate LiO2/Li2O2 ratios around 0.4−0.5, consistent with estimations from Zhai et al.,22,23 who determined even higher ratios of 0.7−1. Instead, we found a clearer tendency with the carbonate distribution. Figure 4c reveals that small toroids (1 and 2 in Figure 2) on the electrode surface are richer in carbonate, showing a fraction of around 0.5 with respect to the total amount of oxygen species (i.e., LiO2 + Li2O2 + carbonate). As a comparison, with small irregular particles and aggregates we found carbonate fractions over 0.9. In larger toroids, however, the amount of Li superoxide and Li2O2 increases, and the Li2CO3 fraction diminished down to 0.3. However, particles on the glass fiber seemed to show a systematically lower carbonate content than particles of the same size on the carbon substrate. Images as those shown in Figure 2 allowed the determination of composition profiles inside the toroidal-shaped electrochemical products. Figure 5 shows two examples of this analysis, corresponding to two different orientations of the toroidal plane with respect to the TEM grid (and the X-ray beam), i.e., perpendicular and parallel to carbon film surface, respectively. Panels a and b represent respectively Li−O phase (LiO2, Li2O2) and carbonate distributions within the same region, respectively. The intensity profiles along the dashed lines of panels a and b are reported in plots 1−4 in Figure 5, in green for the Li−O phase (LiO2, Li2O2) and in red for carbonate. They are consistent with profiles obtained taking full spectra at each point of the profile (see Figure S10 in SI). These experimental profiles were compared with the corresponding profiles calculated using a simulated object, where we assume a graded composition with carbonate-rich surface (see Figure S7 in SI). We take into account the shape determined by tomography (see below). In general, the intensity plots along the dashed lines in real samples are qualitatively consistent with those plotted for simulated objects. Basically, our results confirm that discharge products present a carbonate-rich surface and Li−O phase in the core. The fact that a better simulation is obtained with a graded composition may imply that there is a rough (su)peroxide/carbonate interface or formation of a composite. This suggests that carbonate may form either in parallel or at the expense of the Li−O compound, this latter possibility would imply a certain corrosion of the otherwise smoother surface. To summarize the carbonate distribution, the scheme of Figure 6 represents the different situations of deposits of different sizes and localizations, as we found in the analyzed sample. In all cases an external carbonate-rich shell is observed indicating the chemical decomposition of electrolyte. An approximately constant thickness can explain the increasing carbonate content in smaller objects. This distribution resembles that of the solid electrolyte interface (SEI) typically found on many Li-ion battery electrodes, in particular on carbon anodes.44 Nevertheless, a higher carbonate content is found associated with particles that are in direct contact with the conducting carbon surface (see Figure 4c), while no carbonates were detected directly on the carbon substrate. This

oxygen is fully reduced in silica as in carbonates, and it also presents a similar high energy absorption threshold (see SI for its local spectrum). The cyan color is not evident, given the generally low superoxide content with respect to the other components. Figure 4 illustrates the distribution of the LiO2/ Li2O2 ratio (a) and the carbonate fraction over the total O species (b), determined from the energy-dependent TXM images. Figure 4a illustrates that LiO2 and Li2O2 appear actually

Figure 4. (a) Map of the LiO2/Li2O2 ratio. The respective LiO2 and Li2O2 intensities have been obtained as for Figure 2. The gray noisy area results from regions with low LiO2 and Li2O2 values. (b) Carbonate fraction imaged by TXM. The intensities (see calibration bar) have been obtained from the absorbance subtraction between images at 533.8 and 531.25 eV. (c) Carbonate fractions integrated on different toroidal particles vs particle size. D

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 5. Li−O phase (a) and carbonate (b) distributions in a real discharge product imaged by TXM and oriented with the toroidal plane perpendicular (left) and parallel (right) to the carbon film plane. Plots 1−4 illustrate the Li−O phase (green) and carbonate (red) concentration profiles along the corresponding lines on panels a and b in both the experimental and simulated objects.

Figure 6. Scheme with the composition of toroidal particles of different sizes and localizations showing possible reactions that take place. The scheme is not intended to express a chronological growth sequence.

excludes a direct electrochemical reaction of TEGDME. Freunberger et al.5 have proposed that ether solvents decompose to carbonates mainly by reaction with reduced oxygen (O2−) generated at the cathode. We could explain the observed carbonate distribution by assuming that a higher concentration of electrogenerated O2− radicals is preferentially adsorbed at the surface of those particles located in direct contact with the conducting support, enhancing solvent decomposition at their surface. In contrast, particles that form by disproportionation of solubilized LiO2 more distant from the conducting support have a less reactive surface and will be covered by a thinner carbonate layer. The formation of a SEI-

like layer on the particles suggest that, in analogy with Li-ion batteries, the discharge conditions should favor the formation of Li−O deposits with small specific areas, to minimize electrolyte decomposition and the excessive buildup of the resulting carbonate-rich coating. In any case this barrier will present a serious challenge for the reversible formation and dissolution of the deposits, and its management is certainly a key issue for Li−O2 batteries, as it is for Li-ion batteries. For a morphological detail, tomography has also been carried out. As an example, four tomography reconstruction slices are reported in Figure 7a(1−4), from top (1) to bottom (4) in the sample volume (the full reconstruction video is available in the E

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 7. (a) Consecutive top view tomographic reconstruction images of a specific area with several objects, allowing visualizing inside them from top (1) to base (4). The field of view in each image is 10 × 10 μm2. (b) Three-dimensional morphology of a discharged particle showing remarkable deviations from the generally assumed toroidal shape. (c) Sliced view of the same particle shown in b, pointing out that the hole is open only on one side of the particle.

Supporting Information). Figure 7b,c shows the corresponding tridimensional reconstruction for the same deposit. Interestingly, we have found a remarkable distortion from an ideal toroidal shape in these electrochemical deposits. In effect, our products seem rather to have a shape like a mushroom cap, with a central hole actually open on only one side. In summary, we have combined the high chemical sensitivity and spatial resolution of the soft X-ray energy-resolved TXM technique to identify and spatially separate phases of cathodic discharge products (LiO2, Li2O2, carbonates) in an ether-based lithium−oxygen battery. We observed objects with different morphologies and sizes, with a dominant toroidal shape. Results confirm the presence of an important amount of carbonates due to the possible degradation of our ether electrolyte during discharge. On the one hand, small irregular particles and aggregates with main carbonate character appeared irregularly distributed on the electrode surface. On the other hand, toroidal shaped particles showed a predominating Li−O phase in the core, which is covered with an external carbonate-rich shell. This indicates that a discharge dominated by objects with smaller surface to volume ratio could benefit rechargeability, and in general suggests that strategies to control the thickness and mechanical and physical properties of this layer should be implemented. The significant LiO2-like phase amounts found in toroids strongly suggest that the formation of these characteristic particles can be related to the availability of large superoxide amounts in the solution phase. Also, we observed that in our case deposits have a remarkable distortion from an ideal toroidal shape, rather resembling a mushroom cap, with a central hole open on only one side. This information complements our knowledge on the material morphology with an insight on chemical information, which contributes to a better understanding of solvent decomposition, the formation mechanism, and the electrochemical behavior of the Li−O deposits. However, this work

also proves that full-field synchrotron soft X-ray transmission microscopy is a powerful tool for the characterization of discharge products in Li/O2 batteries and for any material involving oxygen redox processes.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.5b02862. Experimental procedures and data analysis (PDF) Full reconstruction TXM video (AVI)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +(34)935 801 853, ext. 362. Author Contributions

D.T. conceived and supervised the study. R.C.L. developed the electrode preparation and washing procedure. M.O.M. prepared the samples. A.S. developed the synchrotron measurements protocols and operated the instrument. E.P. built, maintained, and supervised the beamline experiment. M.O.M and A.S. analyzed the data. M.O.M., A.S., and D.T. discussed results. N.L.W. contributed significantly to the discussion and interpretation of the results. M.O.M. wrote the manuscript with contributions from A.S. and D.T. All authors revised the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The X-ray microscopy experiments were performed at MISTRAL beamline at ALBA Synchrotron with the collaboration of ALBA staff. Work was funded by the Spanish F

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

(24) Sorrentino, A.; Nicolás, J.; Valcárcel, R.; Chichón, F. J.; Rosanes, M.; Avila, J.; Tkachuk, A.; Irwin, J.; Ferrer, S.; Pereiro, E. J. Synchrotron Radiat. 2015, 22, 1112. (25) Pereiro, E.; Nicolas, J.; Ferrer, S.; Howells, M. R. J. Synchrotron Radiat. 2009, 16, 505−12. (26) Yang, F.; Liu, Y.; Martha, S. K.; Wu, Z.; Andrews, J. C.; Ice, G. E.; Pianetta, P.; Nanda, J. Nano Lett. 2014, 14, 4334−41. (27) Chueh, W. C.; El Gabaly, F.; Sugar, J. D.; Bartelt, N. C.; McDaniel, A. H.; Fenton, K. R.; Zavadil, K. R.; Tyliszczak, T.; Lai, W.; McCarty, K. F. Nano Lett. 2013, 13, 866−872. (28) Chao, S.-C.; Yen, Y.-C.; Song, Y.-F.; Sheu, H.-S.; Wu, H.-C.; Wu, N.-L. J. Electrochem. Soc. 2011, 158, A1335. (29) Chao, S.-C.; Yen, Y.-C.; Song, Y.-F.; Chen, Y.-M.; Wu, H.-C.; Wu, N.-L. Electrochem. Commun. 2010, 12, 234−237. (30) Lin, C.-N.; Chen, W.-C.; Song, Y.-F.; Wang, C.-C.; Tsai, L.-D.; Wu, N.-L. J. Power Sources 2014, 263, 98−103. (31) Mitchell, R. R.; Gallant, B. M.; Thompson, C. V.; Shao-Horn, Y. Energy Environ. Sci. 2011, 4, 2952−2958. (32) Mitchell, R. R.; Gallant, B. M.; Shao-Horn, Y.; Thompson, C. V. J. Phys. Chem. Lett. 2013, 4, 1060−1064. (33) Xia, C.; Waletzko, M.; Chen, L.; Peppler, K.; Klar, P. J.; Janek, J. ACS Appl. Mater. Interfaces 2014, 6, 12083−12092. (34) Adams, B. D.; Radtke, C.; Black, R.; Trudeau, M. L.; Zaghib, K.; Nazar, L. F. Energy Environ. Sci. 2013, 6, 1772−1778. (35) Zheng, H.; Xiao, D.; Li, X.; Liu, Y.; Wu, Y.; Wang, J.; Jiang, K.; Chen, C.; Gu, L.; Wei, X.; Hu, Y. S.; Chen, Q.; Li, H. Nano Lett. 2014, 14, 4245−4249. (36) Gallant, B. M.; Mitchell, R. R.; Kwabi, D. G.; Zhou, J.; Zuin, L.; Thompson, C. V.; Shao-Horn, Y. J. Phys. Chem. C 2012, 116, 20800− 20805. (37) Mizuno, F.; Takechi, K.; Higashi, S.; Shiga, T.; Shiotsuki, T.; Takazawa, N.; Sakurabayashi, Y.; Okazaki, S.; Nitta, I.; Kodama, T.; Nakamoto, H.; Nishikoori, H.; Nakanishi, S.; Kotani, Y.; Iba, H. J. Power Sources 2013, 228, 47−56. (38) Radin, M. D.; Rodriguez, J. F.; Tian, F.; Siegel, D. J. J. Am. Chem. Soc. 2012, 134, 1093−1103. (39) Li, Y.; Wang, J.; Li, X.; Geng, D.; Banis, M. N.; Tang, Y.; Wang, D.; Li, R.; Sham, T. K.; Sun, X. J. Mater. Chem. 2012, 22, 20170− 20174. (40) Zakharchenko, T. K.; Kozmenkova, A. Y.; Itkis, D. M.; Goodilin, E. A. Beilstein J. Nanotechnol. 2013, 4, 758−762. (41) Qiao, R.; Chuang, Y. D.; Yan, S.; Yang, W. PLoS One 2012, 7, e49182. (42) Ruckman, M. W.; Chen, J.; Qiu, S. L.; Kuiper, P.; Strongin, M.; Dunlap, B. I. Phys. Rev. Lett. 1991, 67, 2533−2536. (43) Ong, S. P.; Mo, Y.; Ceder, G. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85, 081105. (44) Goodenough, J. B.; Park, K. S. J. Am. Chem. Soc. 2013, 135, 1167−1176.

Government under contract MAT2012-39199-C02-01 and by the European Commission in the Seventh Framework Programme FP7-2010-GC-ELECTROCHEMICAL STORAGE, under contract no. 265971 “Lithium−Air Batteries with Split Oxygen Harvesting and Redox processes (LABOHR)”. M.O.M. acknowledges CSIC for a JAE-DOC research contract cofinanced by the European Social Fund. Authors thank Dr. Laura Simonelli for her useful discussion on the preliminary data analysis and Judith Oró for her help with TEM and the analysis of the electron diffraction pattern.

■

REFERENCES

(1) Bruce, P. G.; Freunberger, S. A.; Hardwick, L. J.; Tarascon, J. M. Nat. Mater. 2012, 11, 19−29. (2) Christensen, J.; Albertus, P.; Sanchez-Carrera, R. S.; Lohmann, T.; Kozinsky, B.; Liedtke, R.; Ahmed, J.; Kojic, A. J. Electrochem. Soc. 2012, 159, R1−R30. (3) Mizuno, F.; Nakanishi, S.; Kotani, Y.; Yokoishi, S.; Hideki, I. Electrochemistry 2010, 78, 403−405. (4) McCloskey, B. D.; Speidel, A.; Scheffler, R.; Miller, D. C.; Viswanathan, V.; Hummelshøj, J. S.; Nørskov, J. K.; Luntz, A. C. J. Phys. Chem. Lett. 2012, 3, 997−1001. (5) Freunberger, S. A.; Chen, Y.; Drewett, N. E.; Hardwick, L. J.; Bardé, F.; Bruce, P. G. Angew. Chem., Int. Ed. 2011, 50, 8609−8613. (6) Hase, Y.; Ito, E.; Shiga, T.; Mizuno, F.; Nishikoori, H.; Iba, H.; Takechi, K. Chem. Commun. 2013, 49, 8389−8391. (7) Gallant, B. M.; Kwabi, D. G.; Mitchell, R. R.; Zhou, J.; Thompson, C. V.; Shao-Horn, Y. Energy Environ. Sci. 2013, 6, 2518− 2528. (8) Elia, G. A.; Hassoun, J.; Kwak, W. J.; Sun, Y. K.; Scrosati, B.; Mueller, F.; Bresser, D.; Passerini, S.; Oberhumer, P.; Tsiouvaras, N.; Reiter, J. Nano Lett. 2014, 14, 6572−6577. (9) Johnson, L.; Li, C.; Liu, Z.; Chen, Y.; Freunberger, S. A.; Ashok, P. C.; Praveen, B. B.; Dholakia, K.; Tarascon, J. M.; Bruce, P. G. Nat. Chem. 2014, 6, 1091−1099. (10) Aetukuri, N. B.; McCloskey, B. D.; Garcia, J. M.; Krupp, L. E.; Viswanathan, V.; Luntz, A. C. Nat. Chem. 2015, 7, 50−56. (11) Zhai, D.; Wang, H. H.; Lau, K. C.; Gao, J.; Redfern, P. C.; Kang, F.; Li, B.; Indacochea, E.; Das, U.; Sun, H. H.; Sun, H. J.; Amine, K.; Curtiss, L. A. J. Phys. Chem. Lett. 2014, 5, 2705−2710. (12) Kang, S.; Mo, Y.; Ong, S. P.; Ceder, G. Nano Lett. 2014, 14, 1016−20. (13) Bender, C. L.; Hartmann, P.; Vračar, M.; Adelhelm, P.; Janek, J. Adv. Energy Mater. 2014, 4, 1301863. (14) Ren, X.; Wu, Y. J. Am. Chem. Soc. 2013, 135, 2923−2926. (15) Hartmann, P.; Bender, C. L.; Vračar, M.; Dürr, A. K.; Garsuch, A.; Janek, J.; Adelhelm, P. Nat. Mater. 2013, 12, 228−232. (16) Ganapathy, S.; Adams, B. D.; Stenou, G.; Anastasaki, M. S.; Goubitz, K.; Miao, X. F.; Nazar, L. F.; Wagemaker, M. J. Am. Chem. Soc. 2014, 136, 16335−16344. (17) Zhong, L.; Mitchell, R. R.; Liu, Y.; Gallant, B. M.; Thompson, C. V.; Huang, J. Y.; Mao, S. X.; Shao-Horn, Y. Nano Lett. 2013, 13, 2209−2214. (18) Jung, H. G.; Kim, H. S.; Park, J. B.; Oh, I. H.; Hassoun, J.; Yoon, C. S.; Scrosati, B.; Sun, Y. K. Nano Lett. 2012, 12, 4333−4335. (19) Cao, R.; Walter, E. D.; Xu, W.; Nasybulin, E. N.; Bhattacharya, P.; Bowden, M. E.; Engelhard, M. H.; Zhang, J. G. ChemSusChem 2014, 7, 2436−2440. (20) Frith, J. T.; Russell, A. E.; Garcia-Araez, N.; Owen, J. R. Electrochem. Commun. 2014, 46, 33−35. (21) Yang, J.; Zhai, D.; Wang, H. H.; Lau, K. C.; Schlueter, J. A.; Du, P.; Myers, D. J.; Sun, Y. K.; Curtiss, L. A.; Amine, K. Phys. Chem. Chem. Phys. 2013, 15, 3764−3771. (22) Zhai, D.; Wang, H. H.; Yang, J.; Lau, K. C.; Li, K.; Amine, K.; Curtiss, L. A. J. Am. Chem. Soc. 2013, 135, 15364−15372. (23) Zhai, D.; Lau, K. C.; Wang, H. H.; Wen, J.; Miller, D. J.; Lu, J.; Kang, F.; Li, B.; Yang, W.; Gao, J.; Indacochea, E.; Curtiss, L. A.; Amine, K. Nano Lett. 2015, 15, 1041−1046. G

DOI: 10.1021/acs.nanolett.5b02862 Nano Lett. XXXX, XXX, XXX−XXX