# Fileset

[Córdoba_2025_J._Electrochem._Soc._172_110530.pdf](https://mdr.nims.go.jp/filesets/957a866c-7d91-410c-917d-60e429e46947/download)

## Creator

[Daniel Córdoba](https://orcid.org/0000-0001-6823-9972), [Matthew Li](https://orcid.org/0000-0001-9093-3207), [Xiaozhou Huang](https://orcid.org/0000-0002-7660-0382), [Yanan Gao](https://orcid.org/0000-0003-0217-6512), [Shoichi Matsuda](https://orcid.org/0000-0002-0640-3404), [Ernesto J. Calvo](https://orcid.org/0000-0003-0397-2406), [Khalil Amine](https://orcid.org/0000-0001-9206-3719)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Decoupling Failure Pathways in Li-O                    <sub>2</sub>                    Cells Operated Under Lean Electrolyte Conditions](https://mdr.nims.go.jp/datasets/273e0b81-c8ab-437f-bc04-7ebb9d83bf5b)

## Fulltext

Decoupling Failure Pathways in Li-O2 Cells Operated Under Lean Electrolyte ConditionsJournal of TheElectrochemical Society      OPEN ACCESSDecoupling Failure Pathways in Li-O2 CellsOperated Under Lean Electrolyte ConditionsTo cite this article: Daniel Córdoba et al 2025 J. Electrochem. Soc. 172 110530 View the article online for updates and enhancements.You may also likeAn experimental and numericalhydrodynamic study on the Argentinianfishing vesselsS Oyuela, R Sosa, A D Otero et al.-An induction-aware parameterization forwind farms in the WRF mesoscale modelM L Mayol, GP Navarro Diaz, AC Saulo etal.-Farm to farm wake interaction in WRF:impact on power productionML Mayol, AC Saulo and AD Otero-This content was downloaded from IP address 144.213.253.16 on 26/11/2025 at 23:18https://doi.org/10.1149/1945-7111/ae1ea5/article/10.1088/1757-899X/1288/1/012047/article/10.1088/1757-899X/1288/1/012047/article/10.1088/1757-899X/1288/1/012047/article/10.1088/1742-6596/1618/6/062006/article/10.1088/1742-6596/1618/6/062006/article/10.1088/1742-6596/1934/1/012017/article/10.1088/1742-6596/1934/1/012017https://pagead2.googlesyndication.com/pcs/click?xai=AKAOjstQZAOAIHJw1yb_4Phy1LmbaGkUGQ7X4T7aglopthcr98A4bR-nbEwvofs40zCoXN6Kznu_62Nd-OMklxCbiYi6JFFth1QSOju03eInTtev1QwyKgGvyeJhfO3eJwmub79ycI8M9Q77qikLFrXgUE47WY9ODfXfHiLmQ7-rheKhTAnH9LzADDUA2TuAfIKGHXkoZDy3XcSXbJnynAB5lHhSJabyW0mhPpqFBWtqifUZyMN4vC1_W2hlGIEJ2EJkw2iGwO4qAEiOp6OjD6Qv0sHyf9bMmMaSVTKw6IKR6MKgJpymT_7KRlS8-wCoHxGcXRFiudfubcta1zX2f2e-yEPyy56Bxkgu3hPKpwCk3gA7tQDQdpPL0xKI&sig=Cg0ArKJSzLSQb3lFiT-v&fbs_aeid=%5Bgw_fbsaeid%5D&adurl=https://www.el-cell.com/products/pat-battery-tester/pat-tester-i-16/%3Fmtm_campaign%3Diop%2520pdf%2520advert%26mtm_kwd%3Dpat-tester-i-16%26mtm_source%3Dpdf%26mtm_cid%3D2025Decoupling Failure Pathways in Li-O2 Cells Operated Under LeanElectrolyte ConditionsDaniel Córdoba,1,2 m Matthew Li,1,z m Xiaozhou Huang,1 m Yanan Gao,1,3 mShoichi Matsuda,1,3 m Ernesto J. Calvo,2,*,z m and Khalil Amine1,z m1Chemical Sciences and Engineering Division, Argonne National Laboratory, Lemont, Illinois 60439, United States ofAmerica2INQUIMAE (CONICET-UBA), Facultad de Ciencias Exactas y Naturales - Universidad de Buenos Aires, CiudadUniversitaria, Ciudad Autónoma de Buenos Aires, Argentina3Center for Green Research on Energy and Environmental Materials, National Institute for Materials Science, Tsukuba,Ibaraki 305-0044, JapanThe operation of lithium-oxygen (Li-O2) batteries under lean electrolyte conditions offers higher energy density but leads to rapiddegradation and short cycle life. Although cathode passivation and electrolyte decomposition occur in all regimes, we show thatunder lean electrolyte conditions, failure is primarily driven by progressive electrolyte consumption at the lithium/solid electrolyteinterphase (SEI), rather than by irreversible cathode passivation. Poor wetting of the lithium surface results in heterogeneous SEIgrowth and high local current densities, which accelerate electrolyte loss and cell failure. Strategies aimed at improving interfacialstability, including optimized wetting and SEI forming additives, significantly extend cycle life without compromising energydensity. Our results establish anode-electrolyte interactions as the dominant degradation mechanism under lean electrolyteconditions, and emphasize the need to engineer a stable Li/SEI interface for long-lasting Li-O2 batteries.© 2025 The Author(s). Published on behalf of The Electrochemical Society by IOP Publishing Limited. This is an open accessarticle distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited. [DOI: 10.1149/1945-7111/ae1ea5]Manuscript submitted August 12, 2025; revised manuscript received October 25, 2025. Published November 21, 2025.Supplementary material for this article is available onlineLithium-oxygen batteries (LOBs) are considered one of the mostpromising next-generation energy storage systems due to their hightheoretical energy density.1 However, their practical implementationis severely limited by large charge overpotentials, parasitic reactions,electrolyte instability, and rapid capacity fade during cycling.2–4 Oneof the most straightforward strategies to maximize the gravimetricenergy density is to minimize the amount of electrolyte, operatingunder so-called “lean electrolyte” conditions.5 While this strategyenables higher cell-level energy density, it is consistently associatedwith a drastic reduction in cycle life and accelerated degradationcompared to cells flooded with excess electrolyte.6Matsuda and collaborators7–11 have made key contributions to thestudy of LOBs under lean electrolyte conditions. In one of theirworks, they identified that the accumulation of byproducts representsa relevant failure mechanism; since these byproducts do not easilydissipate, they form passivating layers that block oxygen channelsand lead to premature cell death. In another study, they showed thatcontaminants such as H2O and CO2, originating from the cathode,reduce lithium electrode efficiency and accelerate its degradation.They also demonstrated that irreversible volume changes in theelectrodes, together with electrolyte redistribution in the cathode,intensify the deterioration of these cells. Finally, they revealed thatwhen lean electrolyte and high capacity are combined, carbondecomposition above 3.8 V becomes the key factor that shortenscell lifespan.In our previous work,12 we systematically compared LOBscycled under electrolyte-limited conditions (10 μl cm−2) and floodedconditions (100 μl cm−2). We demonstrated that the cathode sideremains largely unaffected by the electrolyte volume: the Li2O2discharge product retains the same crystalline morphology, withsimilar yield (∼74%–77%), without a significant increase in singletoxygen (1O2) formation or parasitic oxygen evolution reactions. Zoret al.13 investigated the stability of tetraethylene glycol dimethylether (G4) against 1O2 and concluded that its contribution tochemical degradation during discharge is practically negligible,accounting for only 0.002% of the total.In addition, differential electrochemical mass spectrometry(DEMS) confirmed that the overall oxygen evolution reaction(OER) efficiency is comparable under both conditions. Similarly,X-ray photoelectron spectroscopy (XPS) revealed that the chemicalcomposition of the solid-electrolyte interphase (SEI) on the lithiumanode does not vary significantly between lean and floodedconditions.12Despite these similarities, electrolyte-limited cells exhibit muchfaster degradation, accelerated impedance growth, and prematurefailure.6,12,14 Post-mortem analysis revealed that the SEI morphologydiffers markedly under lean conditions, where the electrolyte wetsthe surface non-uniformly, leaving areas of lithium exposed. Thiscreates regions of high local current density, leading to the formationof a thick, porous outer SEI layer with high diffusional resistanceand, ultimately, severe electrolyte consumption.6,15 These observa-tions suggest that the failure mechanism under electrolyte-limitedconditions is dominated by electrolyte drying and unstable anodewetting rather than intrinsic cathode degradation.Building on these conclusions, in the present work we aim todirectly track the failure pathway of Li-O2 cells operated underelectrolyte-limited conditions. We combine controlled cycling ex-periments, electrode reassembly tests, and cell design modifications(vacuum wetting, uniform pressure distribution, and anode-protec-tive additives) to determine whether failure originates from irrever-sible electrode passivation or from electrolyte depletion. Underconditions of lean electrolyte, the rapid degradation of LOBs ismainly driven by insufficient wetting of the lithium anode, whichleads to gradual electrolyte depletion at the Li/SEI interface.ExperimentalCell assembly.—A 10 mm diameter and 200 μm thick lithiummetal disc was used as the anode, while a Celgard 2500 (25 μmthickness) separator was impregnated with different volumes ofelectrolyte. A commercial 10 mm diameter carbon gas diffusionlayer from MTI Corporation (EQ-bcgdl-1400S-LD) was dried at120 °C under vacuum overnight and used as the cathode, and a 316stainless-steel mesh was used as the current collector.The electrolyte consisted of 1 M lithium bis(trifluoromethanesul-fonyl)imide (LiTFSI) dissolved in tetraethylene glycol dimethylzE-mail: matthew.li@anl.gov; ernestojulio.calvo@gmail.com; amine@anl.gov*Electrochemical Society Fellow.Journal of The Electrochemical Society, 2025 172 110530 aaahttps://orcid.org/0000-0001-6823-9972https://orcid.org/0000-0001-9093-3207https://orcid.org/0000-0002-7660-0382https://orcid.org/0000-0003-0217-6512https://orcid.org/0000-0002-0640-3404https://orcid.org/0000-0003-0397-2406https://orcid.org/0000-0001-9206-3719https://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/https://doi.org/10.1149/1945-7111/ae1ea5https://doi.org/10.1149/1945-7111/ae1ea5https://doi.org/10.1149/1945-7111/ae1ea5mailto:matthew.li@anl.govmailto:ernestojulio.calvo@gmail.commailto:amine@anl.govhttps://crossmark.crossref.org/dialog/?doi=10.1149/1945-7111/ae1ea5&domain=pdf&date_stamp=2025-11-21ether (G4). For the lithium nitrate (LiNO3) containing cells, 0.2 Mwas added as an additive. Both LiTFSI and LiNO3 was dried invacuum oven at 120 °C overnight and G4 was dried over molecularsieve for 3 days before use.The batteries were assembled using Swagelok-type cell, to ensureconsistent electrolyte distribution, 2.5 μl of electrolyte were firstdispensed onto the Li surface before placing the separator, followedby 7.5 μl onto the cathode prior to cell closure. Once assembled, thecells were placed into a ∼50 ml vessel, removed from the glovebox,purged with oxygen at a pressure of 1 bar for 15 min, and then left torest for 2 h before starting the electrochemical tests. The largevolume of the enclosure mitigates the impact of oxygen partialpressure on the performance.Electrochemical characterization.—Galvanostatic discharge-charge experiments were performed on a Neware cycler with theselected current density and capacity limits as described in theexperimental results. Electrochemical surface area (ECSA) wasmeasured by performing cyclic voltammetry of Swagelok cellsunder Ar atmosphere across ±100 mV vs OCV at scan rates of0.05 to 1 mV s−1 (shown in Fig. S1). The current densities values atthe potential end points (100 mV and −100 mV vs OCV) weresubtracted and divided by two to obtain the current density for fittingthe slope. Electrochemical impedance spectroscopy (EIS) wasperformed using Swagelok cells under identical experimental con-ditions, with a frequency range of 0.1 Hz to 1000 kHz and a voltageamplitude of 10 mV on a Solartron.Stack pressure optimization: Improved pressure uniformity wasachieved by employing multiple springs in parallel (nested) toachieve better contact with the lithium metal. Fuji Prescale pressure-sensitive film was used to confirm the improved uniformity of thestack pressure, as shown in Fig. S2. Vaccum wetting was performedas follows: after cell fabrications and electrolyte injection by dropcasting onto the Li and cathodes. The cell stack was placed into avacuum oven at room temperature and held at ∼0.2 atm for 30 s andthen refilled to 1 atm. This was performed twice. The typical wettingtechnique included just simply injecting the electrolyte with a pipettewithout any vacuum.Results and DiscussionIdentifying the origin of failure under lean electrolyte condi-tions.—To understand the mechanism of accelerated degradationunder electrolyte-limited conditions, we first evaluated whether thecapacity loss is associated with irreversible damage to the cathode orthe anode. Selective electrode reassembly experiments weredesigned, reusing aged cathodes and anodes in new configurationspaired with fresh counterparts. This approach allows decoupling thecontribution of each electrode to the failure pathway and distin-guishing whether premature cell death is caused by irreversiblepassivation or by phenomena related to electrolyte consumption ordistribution.Figure 1a shows the voltage-capacity profiles of a cell operatedunder lean electrolyte conditions, highlighting cycles 1 (purple),6 (green), and the final cycle 12 (black) of the mother cell. Aprogressive increase in the charge overpotential is observed withcontinued cycling.After cell failure, the electrodes were recovered, rinsed with 1,2dimethoxymethane, dried, and reassembled into new cells containingfresh electrolyte and pristine counterparts. The cell employing theaged cathode with new lithium (red curve) exhibited extendedcycling stability, whereas the one using the aged lithium anodefailed rapidly (blue curve), as shown in Fig. 1b. Upon electrolytereinjection (marked with a blue star) capacity was restored,confirming that cell failure originated from electrolyte depletion atthe anode rather than from irreversible cathode passivation.Although the aged cathode introduced a moderate increase in chargeoverpotential, likely due to slight surface passivation,16 it did notsubstantially limit the capacity.These observations indicate that, while cathode passivationcontributes to impedance build-up, cell death occurs only whenelectrolyte loss at the anode disrupts ionic percolation, as evidencedby the capacity recovery upon electrolyte replenishment. Theincreased impedance (manifested as the higher overpotential in the“old Li + new cathode” trace of Fig. 1b) stems from the formation ofa porous SEI on the lithium surface.Similar electrolyte-refiling strategies have been explored in previousstudies as an effective way to mitigate the adverse effects associatedwith electrolyte loss or degradation during aging. Kuzovchikov et al.17demonstrated that electrolyte refilling in aged pouch cells significantlyrestores capacity, even at advanced states of health, by reducing internalimpedance and improving ionic contact. Complementary studies onartificially dried 18650 cylindrical cells showed that electrolyte replen-ishment helps prevent abrupt capacity fade (rollover failure) andrestores the original electrochemical performance.18,19Overall, these results demonstrate that cell failure under leanelectrolyte operation does not originate from irreversible lithiummetal degradation but from progressive electrolyte depletion, pro-moted by poor anode wetting and the development of a porousSEI.6,12,20Figure 1. (a) Voltage-capacity profiles of the cell operated under lean electrolyte conditions for cycles 1, 6, and the final cycle of the “mother” cell, as well as forelectrodes collected and reassembled with new counterparts: old cathode + new lithium and old lithium + new cathode. (b) Cycling stability of the newlyreassembled cells with aged electrodes and their new counterparts; the blue star indicates the point of fresh electrolyte reinjection for the old lithium + newcathode cell.Journal of The Electrochemical Society, 2025 172 110530Strategies to improve the initial wetting of the anode.—Since thereassembly experiments indicated that failure is dominated by theinstability of the Li/SEI interface, physical strategies were exploredto improve the initial wetting of the anode and minimize theformation of dry spots.21 In particular, a vacuum wetting techniquewas implemented to optimize electrolyte distribution during cellassembly, and a multi-spring design applying controlled pressurewas evaluated to maintain more uniform contact between the cellcomponents. The effectiveness of these strategies was analysed usingimpedance spectroscopy, electrochemical surface area measurementsand cycling tests.Figure 2a shows the Nyquist plots under an argon atmosphere forthree full cells: the cell without initial modifications (baseline, purpledots), vacuum wetting (green), and multi-spring (black). Theimpedance was monitored as a function of time to determine whetherdiffusion-controlled growth that is characteristic of SEI growthexist.22 Figure S3 indeed shows at a clear linear trend of impedance(baseline cell) vs t0.5 suggesting that this impedance is majorlycontributed by SEI which can only occur at the lithium anode. Thebaseline cell exhibits the largest semicircle, indicating higherinterfacial resistance. The cell with vacuum wetting shows a smallersemicircle, reflecting better initial contact and lower impedance atthe lithium anode. The multi-spring design presents an intermediatebehaviour.Figure 2b shows the calculation of the electrochemically activesurface area from capacitive currents. The slope is practicallyidentical (∼4% difference) for both the baseline and vacuumwetting, resulting in similar values of ECSA. This confirms thatvacuum wetting does not significantly modify the active surface ofthe cathode but rather improves the wetting of the anode, reducingthe initial resistance. Although vacuum wetting reduces the initialresistance and minimizes dry areas at the beginning, its impact oncycle life is limited.Figure 3 shows the voltage evolution as a function of time for thedifferent cell configurations. The first five cycles were performed at acurrent density of 0.2 mA cm−2, while the subsequent cycles weredischarged at the same current density but charged at 0.4 mA cm−2.Vacuum wetting (blue line) maintains stable voltages for 16 cycles,the multi-spring (green) reaches 14 cycles, and the baseline cell (red)fails after 11 cycles. A capacity below 0.5 mAh cm−2 was used asthe failure criterion. Although a modest improvement is observedwith vacuum wetting, the fundamental degradation mechanism is noteliminated, as the electrolyte continues to be consumed at the Li/SEIinterface, limiting the cycle life.15Analysis of electrolyte oxidation and its relation to cycle life.—While cell degradation is primarily attributed to the progressiveconsumption of the electrolyte at the anode, the direct oxidation ofthe solvent above 4.7 V vs Li/Li⁺ cannot be ruled out. Althoughpotentials above 4.0 V may trigger minor oxidative side reactions,this range was intentionally maintained to capture realistic failureonset and morphology changes associated with lean operation. Ablank-cell test (Fig. S4) confirms that G4 oxidation remains limitedup to approximately 4.7 V under these conditions.Figure 4a shows the capacity associated with the direct oxidationof the G4 solvent above 4.7 V vs Li/Li⁺ as a function of cycle number.An increase in this oxidation is observed in the final cycles beforefailure. Vacuum wetting exhibits higher early oxidation compared tothe baseline and multi-spring configurations. Figure 4b presents theaccumulated capacity above 4.7 V vs Li/Li⁺. Under all conditions, thisvalue increases progressively, but without a clear threshold markingfailure.The slightly higher early oxidation observed for vacuum-wettedcells may result from deeper electrolyte infiltration penetration into theporous cathode, which temporarily increases the effective electro-chemical surface area without compromising overall cell stability. Nodistinct capacity threshold above 4.7 V was identified that couldpredict failure, further ruling out solvent oxidation runaway as thedominant degradation pathway under these conditions.Figure 3. Voltage evolution as a function of time for cells withoutmodifications (baseline), multi-spring, and vacuum wetting, at differentcurrent densities.Figure 2. (a) Nyquist plots under an argon atmosphere for cells without modifications (baseline), multi-spring, and vacuum wetting. (b) Electrochemically activearea obtained from capacitive currents as a function of scan rate for baseline and vacuum wetting.Journal of The Electrochemical Society, 2025 172 110530Therefore, although G4 oxidation does occur above 4.7 V, it doesnot dictate cell failure. The cell ultimately fails independently of theaccumulated extent of solvent oxidation, confirming that the pre-vailing degradation mechanism is the progressive electrolyte con-sumption at the anode rather than irreversible cathode passivation.Mitigation of degradation via anode-protecting additives.—Finally, we explored a strategy based on the in situ engineering ofthe Li/SEI interface through the addition of an anode-protectingelectrolyte additive during cycling. Ideally, SEI layers formed in thisway exhibit suitable mechanical integrity and lithium-ion conduc-tivity, helping to reduce electrolyte loss and improve anode stabilityunder lean electrolyte conditions.20 This approach was tested as analternative to previously presented physical optimizations, with theaim of extending the cycle life without compromising energydensity.Lithium nitrate has been proposed as a as a promising additive forLOBs due to its ability to stabilize the lithium anode through SEIformation, its low degree of dissociation in aprotic solvents, and itscatalytic effect on the oxygen evolution reaction.23 To promote theformation of a more stable SEI and supress electrolyte consumptionat the Li interface, we incorporated 0.2 M of LiNO3 into a 1 MLiTFSI in G4 electrolyte.Figure 5 shows the evolution of specific capacity with cyclenumber under lean electrolyte conditions. The baseline cell failedafter 11 cycles, while cells assembled using vacuum wetting and themulti-spring configuration exhibited moderate improvements,reaching 14 and 16 cycles, respectively. In contrast, the cellcontaining LiNO3 maintained stable performance for up to 29 cycles.The corresponding voltage profiles (Fig. S5) further confirm thedual functionality of LiNO3 as both an SEI-stabilizing additive and apartial redox mediator.23 During cycling, NO3− ions are reduced atthe anode to form NO2−, which can subsequently oxidize to NO2,thereby facilitating Li2O2 oxidation.24 Moreover, LiNO3 affects themorphology of the deposited Li2O2, which influences cathodepassivation and contributes to the overall cell behaviour.25While the additive nearly triples the cell lifespan, it does notsignificantly shift the abrupt capacity drop (∼1 mAh cm−2). Instead,it induces a more gradual rollover, suggesting that although thedominant degradation mechanism remains electrolyte consumptionat the Li/SEI interface, it is mitigated by the formation of a denserand more stable SEI. In addition, the redox-mediating action ofnitrate-derived species facilitates Li2O2 oxidation at the cathode,improving reversibility and delaying passivation. Together, theseeffects account for the extended cycle life and smoother capacitydecay observed in LiNO3 containing cells.ConclusionsWhile both cathode passivation and electrolyte degradationcontribute to performance fading in Li-O2 cells, our results revealthat, under lean electrolyte conditions, the dominant failure me-chanism shifts toward electrolyte depletion at the lithium/SEI inter-face. In this regime, insufficient wetting of the lithium surfacepromotes heterogeneous SEI growth and localized electrolyte con-sumption, ultimately accelerating impedance rise, capacity loss, andcell failure.Selective electrode reuse experiments confirm that the cathodedoes not undergo irreversible passivation under the electrolyte fillingand cycling conditions employed, whereas the aged lithium anode isthe limiting component responsible for premature failure.Physical strategies designed to improve initial wetting, such asvacuum wetting and uniform pressure distribution through multi-spring configurations effectively reduce interfacial resistance anddelay degradation, but they do not eliminate the fundamental causeof failure, which remains progressive electrolyte depletion at theLi/SEI interface.Figure 5. Specific capacity vs cycle number for Li-O2 cells under leanelectrolyte conditions, comparing baseline, vacuum wetting, multi-spring,and the addition of 0.2 M of LiNO3 in 1 M LiTFSI/G4. The first five cycleswere carried out at 0.2 mA cm−2, followed by cycles at 0.4 mA cm−2.Figure 4. (a) Capacity associated with the direct oxidation of the G4 solvent above 4.7 V as a function of cycle number for baseline, multi-spring, and vacuumwetting. (b) Cumulative capacity above 4.7 V vs Li/Li⁺ as a function of cycle number for the same conditions.Journal of The Electrochemical Society, 2025 172 110530In contrast, incorporating SEI-forming additives such as LiNO3markedly extends the cycle life. This improvement arises from theformation of a denser, less permeable SEI that mitigates electrolyteloss, together with the redox-mediating behaviour of nitrate-derivedspecies that facilitate Li2O2 oxidation and delay cathode passivation.Nevertheless, the additive does not alter the ultimate capacity dropthreshold, indicating that while the degradation pathway is moder-ated, the governing mechanism remains electrolyte consumption.Complementary analysis of G4 solvent oxidation above 4.7 V vsLi/Li+ reveals that although oxidative reactions accumulate withcycling, no abrupt onset of solvent breakdown is detected. This rulesout oxidative runaway as the primary degradation mode andreinforces that electrolyte loss at the anode dominates failure.Taken together, these results establish that stabilizing the Li/SEIinterface either through improved wetting, optimized pressure control, orSEI-forming/mediating additives is critical to extending the operationallife of Li-O2 batteries under electrolyte-limited conditions. Engineeringthis interface provides a practical path toward longer-lasting, high-energyLi-O2 systems without compromising gravimetric energy density.AcknowledgmentsThe authors gratefully acknowledge support of this work from theU.S. Department of Energy (DOE), Office of Energy Efficiency andRenewable Energy, Vehicle Technologies Office. Argonne NationalLaboratory is operated for the U.S. DOE, Office of Science, byUChicago Argonne, LLC, under Contract No. DE-AC02–06CH11357.D.C acknowledges support from a postdoctoral research fellowshipawarded by the National Scientific and Technical Research Council(CONICET), Argentina. E.J.C. is a fellow member of CONICET,Argentina. Y.G. and S.M. acknowledge support by Japan Science andTechnology Agency (JST), Adopting Sustainable Partnerships forInnovative Research Ecosystem (ASPIRE), under Contract No.JPMJAP2309.ORCIDDaniel Córdoba m https://orcid.org/0000-0001-6823-9972Matthew Li m https://orcid.org/0000-0001-9093-3207Xiaozhou Huang m https://orcid.org/0000-0002-7660-0382Yanan Gao m https://orcid.org/0000-0003-0217-6512Shoichi Matsuda m https://orcid.org/0000-0002-0640-3404Ernesto J. Calvo m https://orcid.org/0000-0003-0397-2406Khalil Amine m https://orcid.org/0000-0001-9206-3719References1. P. G. Bruce, S. A. Freunberger, L. J. Hardwick, and J.-M. Tarascon, Nat. Mater., 11,19 (2012).2. A. C. Luntz and B. D. McCloskey, Chem. Rev., 114, 11721 (2014).3. W.-J. Kwak et al., Chem. Rev., 120, 6626 (2020).4. H. Xiaozhou, L. Matthew, and A. Khalil, Renewables, 2, 421 (2024).5. S. Matsuda, Electrochemistry, 91, 101006 (2023).6. H. Kwon, J. Baek, and H.-T. Kim, Energy Storage Mater., 55, 708 (2023), https://sciencedirect.com/science/article/pii/S2405829722006730.7. S. Matsuda et al., Adv. Energy Mater., 13, 2203062 (2023).8. S. Matsuda, S. Kimura, and K. Uosaki, The Journal of Physical Chemistry C, 127,11822 (2023).9. M. Ono, J. Saengkaew, and S. Matsuda, Adv. Sci., 10, 2300896 (2023).10. S. Matsuda et al., ACS Appl. Energy Mater., 4, 2563 (2021).11. S. Matsuda and H. Asahina, The Journal of Physical Chemistry C, 124, 25784(2020).12. D. Córdoba et al., The Journal of Physical Chemistry C, 129, 3404 (2025).13. C. Zor et al., Energy Environ. Sci., 17, 7355 (2024).14. S. C. Nagpure et al., J. Power Sources, 407, 53 (2018), https://sciencedirect.com/science/article/pii/S0378775318311686.15. H. Li et al., J. Am. Chem. Soc., 142, 2012 (2020).16. P. Albertus et al., J. Electrochem. Soc., 158, A343 (2011).17. S. M. Kuzovchikov et al., J. Power Sources, 601, 234257 (2024), https://sciencedirect.com/science/article/pii/S0378775324002088.18. F. Arshad et al., ACS Sustain Chem Eng, 8, 13527 (2020).19. X. Lai et al., J Energy Storage, 52, 104951 (2022), https://sciencedirect.com/science/article/pii/S2352152X22009574.20. M. G. Nam, S. W. Jeong, and P. J. Yoo, Adv. Energy Mater., 15, 2400035(2025).21. F. Chen et al., Green Energy and Intelligent Transportation, 4, 100248 (2025),https://sciencedirect.com/science/article/pii/S2773153724001002.22. C. D. Alt et al., Joule, 8, 2755 (2024), https://sciencedirect.com/science/article/pii/S2542435124003350.23. Rosy, S. Akabayov, M. Leskes, and M. Noked, ACS Appl. Mater. Interfaces, 10,29622 (2018).24. D. Sharon et al., ACS Appl. Mater. Interfaces, 7, 16590 (2015).25. C. M. Burke, V. Pande, A. Khetan, V. Viswanathan, and B. D. McCloskey, Proc.Natl Acad. Sci., 112, 9293 (2015).Journal of The Electrochemical Society, 2025 172 110530https://orcid.org/0000-0001-6823-9972https://orcid.org/0000-0001-9093-3207https://orcid.org/0000-0002-7660-0382https://orcid.org/0000-0003-0217-6512https://orcid.org/0000-0002-0640-3404https://orcid.org/0000-0003-0397-2406https://orcid.org/0000-0001-9206-3719https://doi.org/10.1038/nmat3191https://doi.org/10.1021/cr500054yhttps://doi.org/10.1021/acs.chemrev.9b00609https://doi.org/10.31635/renewables.024.202400074https://doi.org/10.5796/electrochemistry.23-00061https://doi.org/10.1016/j.ensm.2022.12.016https://sciencedirect.com/science/article/pii/S2405829722006730https://sciencedirect.com/science/article/pii/S2405829722006730https://doi.org/10.1002/aenm.202203062https://doi.org/10.1021/acs.jpcc.3c01094https://doi.org/10.1002/advs.202300896https://doi.org/10.1021/acsaem.0c03128https://doi.org/10.1021/acs.jpcc.0c09348https://doi.org/10.1021/acs.jpcc.4c07743https://doi.org/10.1039/D4EE02176Bhttps://doi.org/10.1016/j.jpowsour.2018.10.060https://sciencedirect.com/science/article/pii/S0378775318311686https://sciencedirect.com/science/article/pii/S0378775318311686https://doi.org/10.1021/jacs.9b11774https://doi.org/10.1149/1.3527055https://doi.org/10.1016/j.jpowsour.2024.234257https://sciencedirect.com/science/article/pii/S0378775324002088https://sciencedirect.com/science/article/pii/S0378775324002088https://doi.org/10.1021/acssuschemeng.0c04940https://doi.org/10.1016/j.est.2022.104951https://sciencedirect.com/science/article/pii/S2352152X22009574https://sciencedirect.com/science/article/pii/S2352152X22009574https://doi.org/10.1002/aenm.202400035https://doi.org/10.1016/j.geits.2024.100248https://sciencedirect.com/science/article/pii/S2773153724001002https://doi.org/10.1016/j.joule.2024.07.006https://sciencedirect.com/science/article/pii/S2542435124003350https://sciencedirect.com/science/article/pii/S2542435124003350https://doi.org/10.1021/acsami.8b10054https://doi.org/10.1021/acsami.5b04145https://doi.org/10.1073/pnas.1505728112https://doi.org/10.1073/pnas.1505728112