mass of both reactants and a favorable 3.0 V potential. [3][4][5][6] The theoretical specifi c energy of 3505 Wh kg -1 , and energy density of 3435 Wh L -1 (based on Li 2 O 2 ), is very attractive compared to contemporary Li-ion couples that are calculated to have theoretical values near 400 Wh kg -1 and 1400 Wh L -1 . Practical values for Li-air batteries are estimated near 1000 Wh kg -1 , [ 7 ] which compare favorably to current Li-ion commercial cells (210 Wh kg -1 ), but realizing this specifi c energy over continuous operation requires surmounting many challenges in the underlying chemistry.The most prominent of these is creating an electrolyte for the non-aqueous cell which is inert to nucleophilic attack on discharge and charge of the battery, is stable to metallic lithium and solvates Li salts. Although the search for more stable systems has resulted in many investigations of different solvent/ salt combinations, it is generally agreed that there is presently no electrolyte that fi ts these requirements. Dimethylacetamide (DMA) [ 8 ] and dimethylformamide (DMF) [ 9 ] have recently been shown to be quasi-stable in combination with the LiTFSI salt. Nonetheless, both solvents react to form Li-X salts on cycling (X = formate, acetate and carbonate). [ 10 ] These decomposition products, particularly Li 2 CO 3, precipitate on the cathode where they increase impedance and create high cell polarization on charge owing to their high oxidation potentials. [11][12][13] Similar problems are created by the carbon support typically used for the gas diffusion membrane cathode which has been shown to react with the peroxide discharge product and produce Li 2 CO 3 interfacial impedance layers. [ 10 ] Two promising solutions to this dilemma have been presented by Peng et al., [ 14 ] who employed dimethyl sulfoxide (DMSO) as a solvent in combination with a nanoporous gold foil as a gas diffusion membrane, or Thotiyl et al., [ 15 ] utilizing TiC as a stable cathode material. Decomposition of DMSO leads to soluble products such as dimethyl sulfone and lithium sulfate, [16][17][18] which do not passivate the cathode surface to the same extent as do the carbonates. The high reactivity of DMSO with the lithium metal anode, and the eventual precipitation of the decomposition products renders this a fi rst-step solution. The higher rate capability of the noncarbonaceous cathode materials is also highly benefi cial because it favors formation of quasi-amorphous Li 2 O 2 thin-fi lms, which exhibit lower charging potentials. [ 19,20 ] TiC exhibits limited cycling behavior at fast rates even with tetraglyme (TEGDME).A new lithium-ether-derived chelate ionic liquid is synthesized to serve as an electrolyte for the Li-O 2 battery that is stable to metallic lithium, and whose ethereal framework is much more inherently stable to superoxide-initiated hydrogen abstraction than the simple glyme, dimethoxyethane (DME). Reactions of chemically generated superoxide with this electrolyte show that virtually no decomposition products such as lith...

Recently, there has been a transition from fully carbonaceous positive electrodes for the aprotic lithium oxygen battery to alternative materials and the use of redox mediator additives, in an attempt to lower the large electrochemical overpotentials associated with the charge reaction. However, the stabilizing or catalytic effect of these materials can become complicated due to the presence of major side-reactions observed during dis(charge). Here, we isolate the charge reaction from the discharge by utilizing electrodes prefilled with commercial lithium peroxide with a crystallite size of about 200-800 nm. Using a combination of S/TEM, online mass spectrometry, XPS, and electrochemical methods to probe the nature of surface films on carbon and conductive Ti-based nanoparticles, we show that oxygen evolution from lithium peroxide is strongly dependent on their surface properties. Insulating TiO2 surface layers on TiC and TiN - even as thin as 3 nm-can completely inhibit the charge reaction under these conditions. On the other hand, TiC, which lacks this oxide film, readily facilitates oxidation of the bulk Li2O2 crystallites, at a much lower overpotential relative to carbon. Since oxidation of lithium oxygen battery cathodes is inevitable in these systems, precise control of the surface chemistry at the nanoscale becomes of upmost importance.

Li-O2 cells have been the subject of intense investigation in the last few years. The first step in the discharge reaction (oxygen reduction reaction or ORR) is thought to be a 1 e- process that forms the highly reactive intermediate superoxide, O2 -/LiO2. This superoxide species is responsible for the decomposition of most common non-aqueous electrolyte solvents.1 Finding a stable electrolyte for the non-aqueous Li-O2battery is a significant challenge. While allowing conduction of lithium ions between the electrodes, the electrolyte must be stable to nucleophilic attack on discharge, to oxidative decomposition on charge, and inert to the negative electrode. No such solution has been reported to date, and the understanding of the properties that confer stability is lacking. We will present a novel synthesized electrolyte, a chelate ionic liquid (IL), which is stable with respect to metallic lithium and also to superoxide-initiated attack. Figure 1a displays the 1H-NMR spectra of the by-products on cathodes discharged in the chelate IL as well as in the reference standard electrolyte, 1,2-dimethoxyethane (DME). Lithium formate (δ = 8.46 ppm) and dimethyl oxalate (δ = 3.92 ppm) are the two main proton-containing decomposition products formed from the DME molecule. These products are not observed for the new electrolyte. Moreover, when employed in a Li-O2 battery, 8-fold less CO2 evolution occurs on charge than for DME. CO2 is known to evolve during charge of the battery due to the oxidation of lithium carbonates/carboxylates which are formed as side-products by decomposition of electrolytes on discharge.2 This is shown by the results of differential electrochemical mass spectrometry (DEMS) in Figure 1b. Many other cell components besides the electrolyte suffer from instabilities in the non-aqueous Li-O2 battery, including binders and cathode materials.3,4 In particular, alternatives to carbon-based cathodes must be developed. By switching the positive electrode support to non-carbonaceous materials, cells utilizing the new chelate IL electrolyte have been cycled for over 300 hours with no capacity fading and little increase in polarization.5 In comparison, cells with the DME electrolyte, had a cycle life of less than 130 hours. This is displayed in Figure 2a and a comparison of the voltage profiles on the 10th cycle is shown in Figure 2b. The current causes of electrolyte instabilities will be discussed as well as pathways toward the development of new electrolyte/electrode systems with enhanced stabilities. References S. A. Freunberger, Y. Chen, N. E. Drewett, L. J. Hardwick, F. Bardé, P. G. Bruce, Angew. Chem. Int. Ed. 2011, 50, 1-6. B. D. McCloskey, A. Speidel, R. Scheffler, D. C. Miller, V. Viswanathan, J. S. Hummelshoj, J. K. Norskov, A. C. Luntz, J. Phys. Chem. Lett. 2012, 3, 997-1001. R. Black, S. H. Oh, J.-H. Lee, T. Yim, B. Adams, L. F. Nazar, J. Am. Chem. Soc. 2012, 134, 2902-2905. M. M. O. Thotiyl, S. A. Freunberger, Z. Peng, Y. Chen, Z. Liu, P. G. Bruce, Nature Mat. 2013 , 12, 1050-1056. B.D. Adams, R. Black, Z. Williams, R. Fernandes, M. Cuisinier, E. Jaemstorp Berg, P. Novak, G.K. Murphy, L.F. Nazar, Submitted.