Letter pubs.acs.org/JPCL
Oxidation of Dimethyl Sulfoxide Solutions by Electrochemical Reduction of Oxygen Daniel Sharon,† Michal Afri,† Malachi Noked,† Arnd Garsuch,‡ Aryeh A. Frimer,† and Doron Aurbach*,† †
Department of Chemistry, Bar Ilan University, Ramat-Gan, 52900, Israel BASF SE, GCI/E - M311, Ludwigshafen, 67056, Germany
‡
S Supporting Information *
ABSTRACT: Oxygen reduction in nonaqueous electrolyte solutions containing Li salts is a complex field of research involving solution reactions with oxygen radicals and lithium oxides. The aprotic polar solvent dimethyl sulfoxide presents itself as a most promising candidate for a durable electrolyte for use in lithium−oxygen batteries. In the present study, we detail our in-depth study on dimethyl sulfoxide (DMSO) stability in the presence of electroactive lithium oxygen species on carbon electrodes. The question of the stability of DMSO is magnified by our use of carbon-fiber electrodes, which have relatively high specific surface-area and utilize low volumes of electrolyte solutions. This configuration has enabled us to identify even minor side-products such as LiOH, dimethyl sulfone, Li2SO3 and Li2SO4. The proposed mechanism of DMSO decomposition is supported by analytical measurements. These analyses confirm that during the reduction of oxygen on carbon electrodes, the solvent undergoes oxidation by reactive oxygen species and lithium oxides. SECTION: Energy Conversion and Storage; Energy and Charge Transport
T
1.1,2
small amounts of side products.12 When carbon electrodes were used, some cycling was achieved, but there were also some indications that DMSO had undergone reaction with the lithium oxides generated.13,14 The main goal of this work is to determine whether DMSO solutions are suitable for work under an oxygen reduction environment. Utilizing EQCM in conjunction with cyclic voltammetry, this paper will demonstrate that DMSO can be used in a reversible Li−O2 cell. Based on an analysis of the Li− O2 cells after cycling and identification of the different side products by Fourier transform infrared spectroscopy (FTIR), NMR, energy-dispersive X-ray spectroscopy (EDS), X-ray photoelectron spectroscopy (XPS), and X-ray diffraction (XRD), we propose a decomposition mechanism for DMSO . The feasibility of reversible oxygen reduction in DMSO electrolytes has been well demonstrated by electrochemical quartz crystal microbalance (EQCM) measurements in conjunction with cyclic voltammetry, as shown in Figure 1. Figure 1 shows a good correlation between the behavior of the mass and current change. This corresponds well to the results presented by Peng et al.12 demonstrating reversibility of lithium oxide formation and dissolution in DMSO on porous gold electrodes. During the cathodic scan, mass accumulates on the electrode, and, as the potential is swept back, we observe that all the mass is removed from the electrode surface, returning to the initial state. For mass per electron (m.p.e)
he key to the success of Li−O2 batteries lies in the reversibility of the redox processes outlined in Scheme However, an ever increasing number of publications have
Scheme 1. Redox Reactions in an Aprotic Li−O2 Cell
reported on the instability of a variety of electrolyte solutions tested, and it has been shown that the solvent undergoes attack and decomposition by highly reactive oxides and oxygen species. For example, it has been widely reported that carbonate solvents show substantial reactivity to the lithium oxides formed due to the carbonate’s electrophilic nature.3,4 Because of these side reactions, the cycling efficiency of the cell drops precipitously. In the case of polyether solvents, there is a lack of agreement on the question of the stability of polyether solvents: some publications show reasonable cycling data,5 while most others suggest the formation of side products related to the solvent decomposition.6−9 In the search of more durable solvents, new options were selected either by trial and error screening or by computational efforts.9,10 One of the most promising candidates was dimethyl sulfoxide (DMSO).11 In recent publications, DMSO Li−O2 cells have been reported to undergo 100 cycles with a noncarbon cathode, accompanied by the formation of only © XXXX American Chemical Society
Received: August 12, 2013 Accepted: August 30, 2013
3115
dx.doi.org/10.1021/jz4017188 | J. Phys. Chem. Lett. 2013, 4, 3115−3119
The Journal of Physical Chemistry Letters
Letter
even with the cells charged to the upper limit of 4.3 V, we did not succeed in reaching full columbic reversibility and full removal of the products. The scanning electron microscopy (SEM) images of the pristine electrode in Figure 2a display binder-free electrodes
Figure 1. Cyclic voltammetry at 5 mV s−1 (black) and EQCM response of oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) (anodic scan) in DMSO/LiTFSI 1 M. WEs were gold disks deposited on thin quartz crystals.
calculations only the middle slope of the curve can be correlated to mass accumulation related to an electrochemical process; the other two slopes observed cannot be correlated with electron transfer (dQ) because the mass change ranges are too low for accurate values with low error percentage. The m.p.e. in the voltage range of 2.7−2.3 V is 21.952 g/mol, corresponds well to Li2O2 formation which has an equivalent weight of 22.94 g/mol for 1e− transfer and 45.88 g/mol for 2e− transfer. This weight change can be attributed to two possible mechanisms given in Scheme 1. The first involves two 1e− transfers, starting with eq 1 to form the intermediate lithium superoxide, and continuing with the addition of another lithium cation via eq 3 to form Li2O2. The second possibility is by a one-step 2e− transfer via eq 4 directly to form Li2O2. Although the CV and EQCM measurements show impressive reversibility, we need to analyze these results with great caution for several reasons. First, considering the low surface area and weight of the gold electrode, the current cell design uses excess electrolyte solution. In addition, the cyclic voltammetry method employed in our system leads to the formation of a thin layer of oxides during the ORR; this is very different from the complex layer formed during a real cycling of the cell in a galvnostatic mode - that is not governed by kinetics limitations. Gold cannot be considered a relevant material for practical ORR electrode; most researchers therefore use electrodes based on carbonaceous materials. Because of the above considerations, we constructed cells made from monolithic high surface activated carbon microfiber (ACM) cathodes in DMSO-LiPF6 (1 M) that in the given cell assembly can give a high ratio between the carbon cathode and the electrolyte solution.15 This feature, along with the appropriate analytical tools, enables us to detect even minor competing side reactions occurring during the first discharge. Voltage profiles of the ACM electrode in DMSO-LiPF6 (1 M) are presented in the Supporting Information, Figure S1. The cells were operated in the voltage range of 2 to 4.1 V at a current density of 0.05 mA cm−2, with the average discharge voltage of the cells at around 2.75 V. Interestingly, when using this cell based on a high surface carbon cathode and a pure oxygen atmosphere, the oxidation of DMSO practically speaking dropped to a voltage range of 4−4.3 V. Nevertheless,
Figure 2. High-resolution SEM (HR-SEM) images and EDS table of: (a) pristine ACM electrodes; (b,c,d) ACM electrodes discharged to 2 V in DMSO−LiPF6 (1 M) electrolyte solution; and (e) ACM electrodes charged to 4.1 V in DMSO−LiPF6 (1 M) electrolyte solution.
consist of bundles of interwoven carbon microfibers, which have an average diameter of 10 μm. The ACM electrodes possess ∼0.2 mm open channels between the individual interwoven carbon fiber bundles. The EDS elemental analysis (Table in Figure 2) confirms that the pretreatment reduction helped to remove most of the oxygen sites, resulting in hydrophobic carbon fibers that are required for reliable and effective products analysis.16 XRD of pristine ACM electrodes, presented in Figure 3a(1), shows only one broad peak at 44°. This diffraction corresponds to the (100)/(101) planes of small graphitic structures existing within the amorphous ACM. XRD patterns in Figure 3a of the ACM electrode after being polarized to 2.0 V DMSO-LiPF6 (1 M) exhibit crystalline peaks characteristic of both Li2O2 and LiOH. The presence of LiOH has been observed in previous studies that used DMSO in Li−O2 cells.13,14 To understand the formation of LiOH, it is important to first appreciate the unique nature of DMSO. On the one hand, the methyls of DMSO are weakly acidic and can serve as a proton source with strong bases, generating the resonance stabilized dimsyl(methylsulfinyl) anion.17 On the other hand, the sulfoxide linkage of DMSO can react with oxidizing agents to yield the corresponding sulfone (DMSO2). Turning now to the weak acidity of DMSO, in the Li−O2 system, two candidates 3116
dx.doi.org/10.1021/jz4017188 | J. Phys. Chem. Lett. 2013, 4, 3115−3119
The Journal of Physical Chemistry Letters
Letter
Scheme 2. Degradation Mechanism of DMSO during ORR in the Presence of Li Ions
the oxides formed in the solution phase near the electrode surface (before precipitation), in the presence of Li ions.20 The HR-SEM picture (Figure 2b) of the discharged ACM electrode reveals a uniform coating of the individual fibers. In Figure 2c we can observe that the coating contains crystalline Li2O2 toroidal shapes21 and various fragments that can be assumed to be LiOH particles and electrolyte solution decomposition products. A cross section of an individual fiber shown in Figure 2d reveals that the thickness of the coating is in the micrometer level. Hence micrometric size thickness of the solid products that suggests a growth mechanism that involves soluble species such as hydroxides and the stabilized superoxides radicals, which formed the final solid products by reactions with Li ions in solution phase. LiOH and Li2O2, if formed immediately at the electrodes surface, should behave as insulators and are not expected to grow more than a few nanometers. The high oxygen content of the discharged cathode seen from the table in Figure 2, informs us on the formation of oxygen rich compounds like Li2O2 and LiOH. Small peaks of fluorine and phosphorus can be attributed to the known decomposition products of LiPF6 in lithium oxygen cells.22 However, the high concentration (8%) of sulfur on the cathode surface implies that solvent decomposition leads to a passivation layer with a high content of sulfur groups. Figure 2e presents an SEM picture of an electrode charged up to 4.1 V. Although the fibers are still covered with a rough surface, the Li2O2 crystal shapes are absent. The higher carbon to oxygen ratio supports the suggestion that oxygen rich compounds are removed during charging. The XRD diffraction of the electrodes charged to 4.1 V [Figure 3a(3)] supports the decomposition of the Li2O2 particles; however, we still observe spots remaining of LiOH. This can be attributed to the limitation of low oxidation potential, as LiOH particles are hard to oxidize.23 As mentioned above, even in the cutoff voltage of 4.3 V, we observed lithium hydroxide diffractions. The XPS spectrum of sulfur atoms on the surface of the carbon cathode is presented in Figure 3b. The electrode polarized to 2 V shows a complex peak between 168.5 and 170.0 eV. The sulfoxide groups (>SO) from the DMSO are absent as their peaks should appear in lower binding energy of around 164 to 166 eV. We attribute the current spectrum to
Figure 3. (a) XRD pattern of: (1) pristine ACM electrode; (2) ACM electrode discharged to 2 V in a DMSO−LiPF6 (1 M) solution; and (3) ACM electrode that underwent one full cycle, charged to 4 V in a DMSO−LiPF6 (1 M) solution. (b) XPS spectra of S 2p for the cathode polarized to 2 V in DMSO-LiPF6 (1 M).
can behave as strong bases: the possible intermediate LiO2 and the main product Li2O2. We note that the lifetime of LiO2 is reported to be very short,18 but it can be extended due to the high Guttman donor number of DMSO. The latter reduces the acidity of the Li cation, and thereby helps to extend the stability and lifetime of LiO2 before disproportionating to Li2O2.13 With the above pieces of information in hand, we can now describe two possible pathways leading to the formation of LiOH, depending on the oxide base involved. As outlined in Scheme 1, in the first scenario, the initially formed superoxide anion radical is the active agent. This strong base can abstract a proton from the weakly acidic methyl of DMSO (Scheme 2, eq 6), generating the dimsyl anion and a hydroperoxy radical. Upon facile lithium reduction, HOO• generates the pivotal hydroperoxy anion (eq 7). Alternatively, the hydroperoxy anion can be formed directly by a similar proton abstraction by the main product in a lithium−O2 cell, namely Li2O2 (eq 8). The lithium cation, which is a hard Lewis acid, is expected to coordinate strongly with the sulfoxide oxygen of DMSO. This in turn helps to facilitate nucleophilic attack of the hydroperoxy anion on the sulfur atom of the sulfoxide, yielding the central tetrahedral intermediate (eq 9). The latter collapses to the corresponding dimethyl sulfone (DMSO2) and lithium hydroxide19 (eq 10). It should be noted that solid Li2O2 and LiOH, formed as crystalline solids on the cathode surface, seem to be much less reactive toward solution species as compared to 3117
dx.doi.org/10.1021/jz4017188 | J. Phys. Chem. Lett. 2013, 4, 3115−3119
The Journal of Physical Chemistry Letters
Letter
higher sulfur oxidation states, such as sulfite (SO32−) and/or sulfate (SO42−) groups, both reported in the literature in the energy ranges of 168.0−170.1 eV.24 The formation of the sulfite and/or sulfate products presumably involves several consecutive base catalyzed autoxidations of the dimsyl anion (Scheme 3). As mentioned Scheme 3. Proposed Mechanism for the Base Catalyzed Autoxidation of Dimsyl Ion
above, the LiOH formed by the decomposition of the dimsyl tetrahedral intermediate, can lead to the formation of yet more dimsyl lithium. The latter can now undergo base catalyzed autoxidation25 to lithium methylsulfonate. Under the highly basic reaction conditions, the latter can be autoxidized again at the remaining methyl yielding lithium sulfite (Li2SO3) and ultimately lithium sulfate (Li2SO4). This process presumably plays a role in the passivation layer of the cathode .26 These studies reveal differences between the oxygen reduction reactions observed on gold (the EQCM experiments) and those seen on the activated carbon electrodes. The side reactions are much more pronounced with the latter. We anticipate that the high surface area, activated carbon electrodes will be more catalytic than the flat gold electrodes. In fact, the main side reactions are expected to occur during the formation of the reduced oxygen species in the anionic form, before precipitation of solid Li2O2 deposits. The solvent molecules near the high surface area carbon electrodes are exposed to a highly reactive situation, since they are polarized between the strong anionic nucleophiles and the Li ions in solution, which are strong electrophiles. The high surface area of the carbon electrodes should amplify all possible reactions of the DMSO molecules with these active moieties formed in solution. By contrast, on the low surface area gold electrodes, a fast precipitation of Li2O2 occurs (as well reflected by the EQCM response). The solid crystalline Li2O2 is presumably much less reactive with the solvent than the ionic species. The surface analysis of the sulfur containing compounds indicates that it is necessary to study the changes occurring in the electrolyte solution undergoing ORR in presence of lithium cations. Samples from the DMSO/LiPF6 (1M) polarized to 2 V cells were examined by FTIR, 1H NMR and 13C NMR. Figure 4a presents the FTIR spectra of pristine and discharged DMSO/LiPF6 (1 M) sample. The two spectra are almost identical; however, near the sulfoxide (SO) absorption, which appears in pure DMSO at 1050 cm−1, a new peak at around 1142 cm−1 appears in the discharged sample. This new stretching band corresponds to the presence of a sulfone group (OSO). To confirm the presence of a sulfone group and identify its source, NMR analysis was carried out on the resulting
Figure 4. (a) DMSO-LiPF6 (1 M) FTIR spectrum of pristine and polarized to 2 V samples. (b) 1H and (c) 13C NMR spectra of DMSO/ LiPF6 (1 M) after discharged to 2 V.
electrolyte solution. Both the 1H and 13C NMR spectra in Figure 4b,c indicate that the source of the sulfone group is dimethyl sulfone (DMSO2, CH3SO2CH3) with peaks appearing at ca. 2.97 and 42.3 ppm for 1H and 13C, respectively. The formation of dimethyl sulfone (DMSO2) at high oxidation potential in oxygen atmosphere is understandable and supposedly can be prevented by reducing the oxidation potential by a catalyst. However, sulfone formation during the first ORR is hard to avoid. As described above in Scheme 2, the reaction of the hydroperoxy anion with DMSO (eq 9), which leads to the formation of the corresponding sulfone DMSO2, also concomitantly generates LiOH. This solvent decomposition by interaction with Li2O2 corresponds well with our experience with other solvents like carbonates and polyethers. Li2O2 is an excellent nucleophile which should attack the relatively electrophilic sites in any polar aprotic solvent and will irreversibly change their structure. Thus, electrolyte solution decomposition in lithium oxygen cells is a critical consideration and can lead to cell failure within just a few cycles. Unlike other lithium ion batteries, in lithium-O2 cells the decomposition mechanism results from the pivotal lithium oxide formation and is not due to over potential and electron transfer. 3118
dx.doi.org/10.1021/jz4017188 | J. Phys. Chem. Lett. 2013, 4, 3115−3119
The Journal of Physical Chemistry Letters
Letter
(8) McCloskey, B. D.; Speidel, A.; Scheffler, R.; Miller, D. C.; Viswanathan, V.; Hummelshøj, J. S.; Nørskov, J. K.; Luntz, A. C. Twin Problems of Interfacial Carbonate Formation in Nonaqueous Li−O2 Batteries. J. Phys. Chem. Lett. 2012, 3, 997−1001. (9) Bryantsev, V. S.; Faglioni, F. Predicting Autoxidation Stability of Ether- and Amide-Based Electrolyte Solvents for Li−Air Batteries. J. Phys. Chem. A 2012, 116, 7128−38. (10) Assary, R. S.; Curtiss, L. A.; Redfern, P. C.; Zhang, Z.; Amine, K. Computational Studies of Polysiloxanes: Oxidation Potentials and Decomposition Reactions. J. Phys. Chem. C 2011, 115, 12216−12223. (11) Laoire, C. O.; Mukerjee, S.; Abraham, K. M.; Plichta, E. J.; Hendrickson, M. a. Influence of Nonaqueous Solvents on the Electrochemistry of Oxygen in the Rechargeable Lithium−Air Battery. J. Phys. Chem. C 2010, 114, 9178−9186. (12) Peng, Z.; Freunberger, S. A.; Chen, Y.; Bruce, P. G. A Reversible and Higher-Rate Li−O2 Battery. Science (New York, N.Y.) 2012, 337, 563−6. (13) Trahan, M. J.; Mukerjee, S.; Plichta, E. J.; Hendrickson, M. A.; Abraham, K. M. Studies of Li−Air Cells Utilizing Dimethyl SulfoxideBased Electrolyte. J. Electrochem. Soc. 2012, 160, A259−A267. (14) Xu, D.; Wang, Z.; Xu, J.; Zhang, L.; Zhang, X. Novel DMSOBased Electrolyte for High Performance Rechargeable Li−O2 batteries. Chem. Commun. (Cambridge, U. K.) 2012, 48, 6948−50. (15) Etacheri, V.; Sharon, D.; Garsuch, A.; Afri, M.; Frimer, A. a.; Aurbach, D. Hierarchical Activated Carbon Microfiber (ACM) Electrodes for Rechargeable Li−O2 batteries. J. Mater. Chem. A 2013, 1, 5021−5030. (16) Ottakam Thotiyl, M. M.; Freunberger, S. A.; Peng, Z.; Bruce, P. G. The Carbon Electrode in Nonaqueous Li−O2 cells. J. Am. Chem. Soc. 2013, 135, 494−500. (17) Butler, J. N. Electrochemistry in Dimethyl Sulfoxide. J. Electroanal. Chem. 1967, 14, 89−116. (18) Lindsay, D. M.; Garland, D. A. ESR Spectra of Matrix-Isolated LiO2. J. Phys. Chem. 1987, 91, 6158−6161. (19) Goolsby, A. D.; Sawyer, D. T. The Electrochemical Reduction of Superoxide Ion and Oxidation Of Hydroxide Ion in Dimethyl Sulfoxide. J. Anal. Chem. 1968, 40, 83−86. (20) Schwenke, K. U.; Meini, S.; Wu, X.; Gasteiger, H. a; Piana, M. Stability of Superoxide Radicals in Glyme Solvents for Non-aqueous Li−O2 Battery Electrolytes. Phys. Chem. Chem. Phys. 2013, 15, 11830− 9. (21) Black, R.; Oh, S. H.; Lee, J.; Yim, T.; Adams, B.; Nazar, L. F. Screening for Superoxide Reactivity in Li−O2 Batteries: Effect on Li2O2/LiOH Crystallization. J. Am. Chem. Soc. 2012, 134. (22) Veith, G. M.; Nanda, J.; Delmau, L. H.; Dudney, N. J. Influence of Lithium Salts on the Discharge Chemistry of Li−Air Cells. J. Phys. Chem. Lett. 2012, 3, 1242−1247. (23) Meini, S.; Tsiouvaras, N.; Schwenke, K. U.; Piana, M.; Beyer, H.; Lange, L.; Gasteiger, H. a Rechargeability of Li−Air Cathodes Prefilled with Discharge Products Using an Ether-Based Electrolyte Solution: Implications for Cycle-Life of Li−Air Cells. Phys. Chem. Chem. Phys. 2013, 15, 11478−93. (24) Lindberg, B. J.; Hamrin, K.; Johansson, G.; Gelius, U.; Fahlman, A.; Nordling, C.; Siegbahn, K. Molecular Spectroscopy by Means of ESCA II. Sulfur Compounds. Correlation of Electron Binding Energy with Structure. Phys. Scr. 1970, 1, 286−298. (25) Frimer, A. A. Oxygenation of Enones; Wiley: London, 1989. (26) Abraham, K M; Chaudhri, S. M. The Lithium Surface Film in the Li−SO2 Cell. J. Electrochem. Soc. 1986, 133, 1307−1311.
In conclusion, the attempt to use DMSO as a solvent in Li− O2 experiments is certainly challenging. On one hand, the use of gold electrodes in the EQCM measurements shows reversible behavior for ORR and OER. This lends support to use of noncarbon electrodes to minimize the electrolyte solution degradation and side products formation. On the other hand, when using carbon cathodes, we observe the formation of side products resulting from exposure of DMSO to the O2− and O2−2, aggressive nucleophiles and bases, formed during the ORR. It might be that the overall system can remain active even when the ratio of DMSO2/DMSO increases. Nevertheless, as we proceed to later cycles, we will end up with high concentrations of oxidized DMSO and nonsoluble DMSO fragmentsas seen by EDS and XPS analysisthat will lead to a severe deterioration in cell performance. It seems that our studies point out that DMSO is not a suitable solvent for rechargeable Li−O2 cells.
■
ASSOCIATED CONTENT
S Supporting Information *
Experimental methods; electrochemical behavior of carbon electrode, and EQCM theory and calculations. This material is available free of charge via the Internet at http://pubs.acs.org
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The research was supported by the BASF scientific network of electrochemistry and batteries. AAF thanks the Ethel and David Resnick Chair in Active Oxygen Chemistry as well as the Israel Science Foundation (Grant No. 1469/13) for their kind and generous support.
■
REFERENCES
(1) Laoire, C. O.; Mukerjee, S.; Abraham, K. M.; Plichta, E. J.; Hendrickson, M. a. Elucidating the Mechanism of Oxygen Reduction for Lithium−Air Battery Applications. J. Phys. Chem. C 2009, 113, 20127−20134. (2) Peng, Z.; Freunberger, S. a; Hardwick, L. J.; Chen, Y.; Giordani, V.; Bardé, F.; Novák, P.; Graham, D.; Tarascon, J.-M.; Bruce, P. G. Oxygen Reactions in a Non-Aqueous Li+ Electrolyte. Angew. Chem., Int. Ed. Engl. 2011, 50, 6351−5. (3) Freunberger, S. a; Chen, Y.; Peng, Z.; Griffin, J. M.; Hardwick, L. J.; Bardé, F.; Novák, P.; Bruce, P. G. Reactions in the Rechargeable Lithium-O2 Battery with Alkyl Carbonate Electrolytes. J. Am. Chem. Soc. 2011, 133, 8040−7. (4) McCloskey, B. D.; Bethune, D. S.; Shelby, R. M.; Girishkumar, G.; Luntz, A. C. Solvents’ Critical Role in Nonaqueous Lithium− Oxygen Battery Electrochemistry. J. Phys. Chem. Lett. 2011, 2, 1161− 1166. (5) Jung, H.-G.; Hassoun, J.; Park, J.-B.; Sun, Y.-K.; Scrosati, B. An Improved High-Performance Lithium−Air Battery. Nature Chem. 2012, 4, 579−85. (6) Freunberger, S. a; Chen, Y.; Drewett, N. E.; Hardwick, L. J.; Bardé, F.; Bruce, P. G. The Lithium−Oxygen Battery with Ether-Based Electrolytes. Angew. Chem., Int. Ed. Engl. 2011, 50, 8609−13. (7) Sharon, D.; Etacheri, V.; Garsuch, A.; Afri, M.; Frimer, A. A.; Aurbach, D. On the Challenge of Electrolyte Solutions for Li−Air Batteries: Monitoring Oxygen Reduction and Related Reactions in Polyether Solutions by Spectroscopy and EQCM. J. Phys. Chem. Lett. 2013, 4, 127−131. 3119
dx.doi.org/10.1021/jz4017188 | J. Phys. Chem. Lett. 2013, 4, 3115−3119
