Tuning ion solvation to bypass facet-selective lithium deposition

Tuning ion solvation to bypass facet-selective lithium deposition | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Tuning ion solvation to bypass facet-selective lithium deposition Hyun-Wook Lee, Juyoung Kim, Min-Ho Kim, Huida Lyu, Min Hyeok Kim, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7588130/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Anode-free lithium (Li) batteries maximize energy density by relying on direct Li plating and stripping on metallic substrates, but their practical deployment is limited by uncontrolled and heterogeneous Li deposition. Here, we identify that this heterogeneity is governed by the crystallography of the polycrystalline substrate: high-index facets and grain boundaries direct preferential nucleation, and that the electrolyte solvation structure provides a pathway to bypass this. In the desolvation-limited regime, complete solvent removal must precede charge transfer, leaving bare Li⁺ ions directly exposed to the substrate, and thus strong Li–substrate interactions and facet-selective deposition. Weakly solvating electrolytes preserve partial solvation or anion coordination at the interface, and this solvation shielding attenuates substrate coupling. This decoupling promotes uniform Li plating regardless of crystallographic orientation and grain boundaries. Our findings establish solvation engineering as a viable strategy to suppress crystallographic control in Li metal electrodeposition, and offer design principles for stabilizing Li metal anodes. Physical sciences/Energy science and technology/Energy storage/Batteries Physical sciences/Materials science/Materials for energy and catalysis/Batteries Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Lithium (Li) metal anodes, with their exceptionally high theoretical capacity (3,860 mAh g -1 ) and the lowest redox potential (-3.04 V versus standard hydrogen electrode), are considered critical for enabling next-generation high-energy-density batteries 1 . However, their practical implementation remains hindered by the inherently heterogeneous nature of Li plating and poor interfacial stability, which lead to dendritic growth and continuous electrolyte consumption, respectively 2–4 . Together, these issues ultimately cause premature cell failure, and the slow progress in addressing these weaknesses has increasingly cast doubt on the practical viability of lithium metal batteries (LMBs). Among various strategies to overcome these challenges, electrolyte engineering has delivered the most dramatic progress, particularly in improving Coulombic efficiency (CE) and cycling stability 5–11 . While numerous electrolyte concepts have been introduced, ether solvents ( e.g. , 1,2-dimethoxyethane (DME), 1,3-dioxolane (DOL)) have been widely adopted as model systems owing to their relatively superior reductive stability against Li 0 ―that is, high compatibility with Li 0 ―compared to practical carbonate counterparts ( e.g. , ethylene carbonate (EC), dimethyl carbonate (DMC), diethyl carbonate (DEC)) 5,6 . Building on this foundation, recent studies 12–16 ―most notably those led by the Bao group―have attracted considerable attention by tailoring ether structures, either through functional group substitution ( e.g. , 1,2-diethoxyethane (DEE)) or fluorination ( e.g. , fluoroethers), to develop weakly solvating or fluorine-rich solvents that further enhance CE and cycle life. Despite these significant achievements in suppressing dendritic growth and mitigating parasitic reactions, most electrolyte research has rationalized the observed performance enhancements primarily in terms of the solid-electrolyte-interphase (SEI) stabilization, particularly through the formation of anion-derived inorganic species 17–20 . While this perspective has been valuable, it provides limited insight into how Li nucleates and grows on Cu substrate. In our previous work 21 , we newly introduced the concept of epihorizontaxy , which refers to the preferential horizontal growth of Li facilitated by the weak interaction between Li adatoms and the bare Cu foil surface. This behavior arises from the negligible migration barrier of Li atoms on low-index Cu facet―particularly Cu(111)―with their isotropic crystallographic orientation, whereas high-index facets such as Cu(115) and Cu(410) promote vertical Li growth. This finding implies that the coexistence of diverse crystallographic facets and additional microstructural features ( e.g., grain boundaries, roughness) in polycrystalline Cu foil not only exacerbates morphological heterogeneity in Li deposition but also complicates efforts to decouple and precisely assess the role of substrate–Li interactions. Here, we report, for the first time, the facet dependence of Li deposition on Cu foils as a function of electrolyte formulation―an effect previously unrecognized―and identify the key electrolyte parameters that enable outpacing facet selectivity. By establishing a direct link between solvation structure, charge-transfer behavior, and crystallographic surface effects, we demonstrate a pathway to achieve uniform Li deposition from the atomic scale to practical electrodes. Cu(111)/(115) twin foil as a platform for facet selectivity Single-crystal Cu(111) foil can be obtained from commercially available polycrystalline Cu foil through a contact-free annealing process 21–23 . Under such contact-free conditions, grain growth is governed primarily by the surface energy of the foil itself, driving Cu grains to abnormally evolve toward the (111) orientation, which is the thermodynamically most stable surface of face-centered cubic (FCC) structure owing to its lowest surface energy, in contrast to high-index facets which possess higher surface energies due to their open atomic arrangements and undercoordinated surface atoms 24,25 . Interestingly, however, (115) high-index plane often coexists with (111) through step-terrace configurations or twin boundary relationships ( e.g. , ∑3 {111} twins), where the interfacial energy penalty is substantially reduced 21,26 . As a result, a special Cu foil comprising predominantly (111) domains with minor fractions of (115) twin grains occasionally emerge during annealing (Fig. 1a). In principle, we hypothesized that incoming Li atoms preferentially attach to high-index facets and their grain boundaries in order to minimize surface energy, as dictated by thermodynamics 21,27 , after undergoing desolvation and charge transfer (Fig. 1b). Accordingly, the unique Cu(111)/(115) twin foil (hereafter, termed as tCu(111)/(115)) provides a well-defined platform to investigate facet-selective Li deposition, enabling direct comparison between the inherently stable (111) terraces and the higher-energy (115) surface within the same substrate. Consistent with this hypothesis, we demonstrated that LMBs follow this thermodynamic rule in a conventional carbonate electrolyte (1.3 M lithium hexafluorophosphate (LiPF 6 ) in EC/DEC + 10 wt% fluoroethylene carbonate (FEC)), where Li preferentially deposits at twin grain boundaries and on Cu(115) facets rather than on Cu(111) (Fig. 1c). While these results underscore the importance of interactions between Li adatoms and the substrate, a critical open question remains: Is preferential plating on high-index facets a universal phenomenon? Given that Li plating is additionally influenced by multiple interfacial factors—including solvation structures 12,13,15 , charge transfer kinetics 28–31 , mass transport in the electrolytes 3,32–35 , and the physicochemical properties of SEI layer 36–38 —it is plausible that facet selectivity can be reshaped by altering cell conditions and environments. In particular, with the growing recognition of solvation structure in LMBs, its potential role in directing preferential deposition on high-index facets deserves special emphasis. To elucidate the role of interfacial phenomena in dictating high-index facet selectivity across different electrolytes, we investigated Li deposition on the tCu(111)/(115) using a range of electrolytes commonly employed in LMBs, with a fixed plating capacity of 0.1 mAh cm -2 (Fig. 2a and Supplementary Fig. 1 and 2). Irrespective of electrolyte composition, Li deposition was primarily governed by thermodynamics, with preferential nucleation occurring at grain boundaries and on high-index Cu(115) facets. Notably, such non-uniform deposition left the SEI residues locally after stripping, implying that repeating cycles can cause severe local increase in resistance (Supplementary Fig. 3). This trend persisted even in state-of-the-art high-performance electrolytes, such as 0.6 M lithium difluoro(oxalate)borate (LiDFOB) + 0.6 M lithium tetrafluoroborate (LiBF 4 ) in FEC/DEC, 1 M lithium bis(fluorosulfonyl)imide (LiFSI) in DME, 1 M LiFSI in DME/1,1,2,2-tetrafluoroethyl 2,2,3,3-tetrafluoropropylether (TTE), and 1 M LiFSI in DEE. The only distinction observed was that the critical current density, above which site-selective Li deposition emerged, varied with the electrolytes (Fig. 2b and Supplementary Fig. 4). Below this threshold, Li deposition was not governed by thermodynamics, but at higher current densities facet-dependent plating reappeared. Ether-based electrolytes exhibited relatively higher critical current densities than their carbonate-based counterparts (Fig. 2c), indicating that current density-driven interfacial chemistry plays a decisive role in determining the onset of site-selective Li deposition. We speculate that below critical current densities, Li deposition proceeds slowly enough for solvent–Li⁺ interactions, solvation structure, and interfacial charge-transfer processes to significantly influence nucleation pathways, thereby modulating facet selectivity. At high current densities, however, rapid adatom accumulation dictates deposition, with local equilibrium governed mainly by surface energetics—Similarly, an abnormal increase in Li 0 concentration beyond supersaturation point often forms rhombic dodecahedral Li enclosed by the stable {110} facet of the BCC lattice 21,39 , governed entirely by the intrinsic surface energy of Li through homogeneous nucleation, independent of the substrate 40 . In this regime, thermodynamics prevails over interfacial kinetic effects, leading to preferential Li nucleation at high-energy sites such as grain boundaries and high-index facets, irrespective of electrolyte formulation. In other words, a higher critical current density reflects solvation-mediated interfacial stabilization, which ultimately promotes ‘uniform’ Li deposition. We further observed distinct deposition behaviors on tCu(111)/(115) in DME and DEE solvents through real-time top-view optical microscopy (Fig. 2d, Supplementary Fig. 5 and Supplementary Video 1 and 2). High-index facets and grain boundaries are clearly distinguished in optical images owing to orientation-dependent reflectivity. In DME, the high-index region progressively becomes brighter, indicative of localized Li plating. By contrast, in DEE, Li deposition is no longer constrained by thermodynamic preference, resulting in uniform nucleation across the substrate. Then, what accounts for this striking difference in deposition behavior between DME and DEE? Thus far, we have empirically shown that while the uniformity of Li deposition strongly depends on the crystallographic orientation of the substrate, its kinetics is closely correlated with the electrolyte characteristics. To elucidate this connection, it is essential to understand how the solvation structures differ between DME and DEE. The effect of electrolyte solvation in facet-dependent deposition A previous report 12 proposed that DEE solvent, obtained by substituting the methoxy groups of DME with larger ethoxy groups, effectively tunes ion solvation through steric hinderance, thereby weakening the coordinating ability of its two oxygen atoms (Fig. 3a). As a result, the inner solvation shell contains more FSI - anions in DEE (0.89 mol per mol of DEE) than in DME (0.76 mol per mol of DME). Although they attributed the improved CE solely to the formation of anion-derived SEI, this explanation cannot account for our findings regarding Li + -substrate interactions. Here, beginning with the comparison between DME and DEE, we deliberately expanded the control groups to include a series of DME-DEE mixed solvents (DME:DEE = 80:20, 50:50, 20:80) to strengthen the validity of our analysis (Fig. 3b). Even with the addition of a small fraction of DEE (20 vol%), the facet effect became slightly mitigated, showing reduced local plating along the grain boundaries. And, when half of the electrolyte volume consisted of DEE (50:50 DME:DEE), the crystallographic effect was barely discernible. Interestingly, Li morphology also shifted to particle- or nodule-like shapes in DEE-containing electrolytes, in contrast to the localized dendritic Li growth observed on high-index facets in DME. These observations collectively suggest that the spatial distribution of Li deposition and the resulting Li morphology on the substrate are intricately interconnected. To investigate the coordination environment of Li + , we performed the molecular dynamics (MD) simulations (Fig. 3c,d). As the DEE content in the electrolyte increased, the number of DEE molecules participating in the inner solvation shell also increased, whereas the FSI - anion coordination number (C.N.) remained nearly unchanged up to DME:DEE = 80:20, comparable to that in pure DME (Fig. 3c). At a 50:50 ratio, however, the FSI - C.N. showed a distinct increase but, contrary to expectation, did not change further, remaining essentially the same as in pure DEE. Although the FSI - density is nearly identical across DME:DEE = 50:50, 20:80, and pure DEE, multiple solvation configurations coexist in mixed solvents (Fig. 3d and Supplementary Fig. 6). It should be emphasized that solvation structure cannot be fully captured by anion coordination numbers alone. Local configurations—such as the spatial distribution of solvent and anion species—and their dynamic exchange at the interface are equally critical in dictating how Li⁺ couples to the substrate. These factors help explain why electrolytes with similar anion coordination numbers can nevertheless exhibit markedly different facet selectivity during lithium deposition. Consistent with the MD results, Raman spectroscopy revealed that at high DEE contents a significant fraction of FSI - anions is incorporated into the solvation shell in the form of aggregated clusters (AGG) and contact ion pairs (CIP), thereby reducing the population of free FSI - anions (Fig. 3e,f and Supplementary Fig. 7). Notably, the CIP+AGG ratio divides the electrolytes into two distinct groups: (ⅰ) pure DME and DME:DEE = 80:20, and (ⅱ) DME:DEE = 50:50, 20:80, and pure DEE (Fig. 3f). This classification aligns well with the facet selectivity test results (Fig. 3b), supporting the notion that Li deposition behavior correlates more closely with the solvation structure than with the solvent identity itself. In particular, once the DEE content in the electrolyte reaches 20 %, DEE molecules constitute more than half of the solvents participating in the solvation shell, whereas in the 50:50 mixture the solvation shell remains DME-rich―even though the FSI - anion coordination number is elevated by DEE and remains essentially unchanged within group (ⅱ). These findings highlight again that it is the DEE-induced reorganization of the solvation sheath―rather than the DEE solvent molecules themselves―that governs facet selectivity and Li deposition morphology. Solvation-dependent charge transfer mechanisms To explore solvation-dependent charge transfer mechanisms, we measured the kinetics of the Li + + e - → Li 0 near the kinetically controlled region using cyclic voltammetry (CV) (Fig. 4a). The CV results revealed systematic deviations from the linearized Butler-Volmer (BV) response with increasing overpotential, highlighting the limitations of the BV approximation 29 . The BV model assumes that the activation barrier decreases linearly with applied potential, which is valid near equilibrium but often fails at higher overpotentials 41 . To better capture this behavior, we analyzed the Tafel slopes from five electrolytes using both the BV model and the Marcus-Hush-Chidsey (MHC) model (Fig. 4b). Unlike BV, the MHC framework explicitly incorporates the reorganization energy (λ) associated with solvent and interfacial restructuring, leading to a parabolic dependence of the activation barrier on potential 41,42 . In the case of Li plating on Cu foil, where electron transfer occurs at a metal-solution interface with a continuum of electronic states, the MHC model thus provides a more realistic description of the charge transfer kinetics across a broader potential window. By applying the linearized form of the Butler-Volmer equation, j = j 0 f ( E - E eq ), we determined the exchange current densities ( j 0 ) for five different electrolytes (Fig. 4c). The values obtained were 23.24 mA cm -2 for DME, 32.28 mA cm -2 for DEE, and a highest value of 45.67 mA cm -2 for the DME:DEE 50:50 mixture. These results are consistent with previous reports 12 showing that DEE exhibit faster kinetics than DME. However, the overall trend did not correlate with the high-index facet selectivity observed in Fig. 3b; for instance, the DME:DEE 80:20 showed a higher exchange current density than pure DEE, yet its Li plating remained facet-dependent. This suggests that outpacing facet-dependent Li plating is not governed by the fast kinetics alone. Strikingly, pure DEE and DEE-based electrolytes, including the 50:50 mixture, were well-fitted to both the BV and MHC models, yielding R square (R 2 ) values exceeding 0.95. In contrast, pure DME and DME-based electrolytes exhibited comparatively low R 2 values. Indeed, this finding is consistent with previous reports 29 suggesting that Marcus-type fitting in DME electrolytes fails to accurately describe charge transfer phenomena, because the scan rates are insufficient to drive the system into the charge transfer-limited regime. However, more recent publication 15 excluded this possibility by employing ultrafast scan rates, instead attributing the discrepancy to the symmetric molecular structure and strong solvation of DME, which reinforce solvent shielding and impose an additional desolvation barrier. While the authors compared symmetric and asymmetric ethers, our results demonstrate that substantial differences can also arise between two symmetric molecules, DME and DEE, particularly in their facet selectivity. This suggests that solvation strength and desolvation dynamics, rather than molecular symmetry alone, critically dictate the apparent charge transfer kinetics and interfacial deposition behavior. Taken together, we propose a key mechanistic distinction between DME- and DEE-based electrolytes (Fig. 4d,e). Marcus theory was originally formulated for outer-sphere electron transfer, in which charge transfer proceeds via solvent reorganization and electron tunneling without direct orbital overlap with the electrode 41 (Fig. 4d). In DME, however, desolvation becomes the rate-limiting step: Li⁺ must shed much of its strongly bound solvation shell and also displace the dipole-aligned solvent molecules residing directly at the electrode interface 41 (Left in Fig. 4e). This effective removal of both coordinated and interfacial solvent leaves Li + substantially exposed to the Cu substrate. The resulting naked-ion interaction enforces strong substrate coupling, producing inner-sphere-like behavior that renders the kinetics highly facet-sensitive and prevents reliable Marcus-type fitting. By contrast, in DEE the weak solvation lowers the desolvation barrier, while the low dipole moment and dielectric constant produce only a weak interfacial solvent layer. As a result, this solvent screening does not impose a dominant barrier to charge transfer, and Li + ions approach the interface closely without requiring complete displacement of the interfacial solvent 15 (Right in Fig. 4e). In this way, electron transfer can proceed via electron tunneling 41 , even though interfacial solvent or anions partially screen Li + from substrate. Such screening attenuates direct substrate coupling and yields a facet-insensitive, outer-sphere-dominated mechanism well captured by the MHC model. Importantly, in such weakly solvating environments, the charge transfer rate is governed less by the strength of substrate-ion binding and more by whether ion can sufficiently approach the electrode interface and whether the solvent reorganization barrier remains low 41 . Altogether, these findings suggest that the decisive factor in achieving uniform Li deposition is not molecular symmetry but the extent of substrate–Li⁺ coupling dictated by solvation structure. Even among symmetric ethers, differences in solvation strength and solvent shielding can lead to markedly different facet dependencies. This coupling, in turn, governs both the applicability of Marcus kinetics and the facet selectivity that determines Li plating morphology. Impedance analysis of SEI formation and Li nucleation To phenomenologically investigate interfacial phenomena, we conducted electrochemical impedance spectroscopy (EIS). Traditionally, EIS is regarded as a powerful technique not only for probing dynamic processes but also for characterizing interfacial films 43,44 . However, conventional EIS analysis still leaves considerable ambiguity in interpretation, as multiple interrelated factors are often convoluted and difficult to decouple 45,46 . The most effective way to eliminate such ambiguity is through in situ analysis, which directly captures the evolution of interfacial properties under operating conditions 47 . In this regard, we employed stepwise potentiometric EIS (SPEIS), which enables a more systematic separation of the resistive and capacitive contributions associated with SEI formation and Li nucleation. As shown in Fig. 5a, the voltage was swept from the open-circuit voltage (OCV) to -0.10 V, and EIS measurements were taken at every 0.01 V step. This approach allows us to probe internal battery phenomena under non-equilibrium state and more realistic conditions, as the cell was not rested prior to the measurements; only the high- and mid-frequency regions were probed, since low-frequency measurements require prohibitively long durations and would otherwise drive the system toward equilibrium 48,49 . For reliable analysis, the measurements were performed in a three-electrode Li|Cu configuration with Li reference electrode. Indeed, introducing a Li reference electrode revealed that the impedance of the Cu side was negligible compared to that of the Li counter electrode, and that EIS measured in a two-electrode configuration was dominated by the Li counter electrode (Fig. 5b). This highlights the necessity of three-electrode setup to isolate the response of the Cu working electrode, which challenges the conventional assumption that metallic Li can simultaneously serve as both counter and reference electrodes. Initially, both DME and DEE exhibited negligible current responses below 0.01 mA, after which current increased at -0.04 V in DME and -0.02 V in DEE, indicating the onset of Li plating (Fig. 5a). These onset points coincided with pronounced changes in the EIS spectra (Fig. 5c). Upon Li plating, the semicircle magnitude decreased rapidly, which can be attributed to the increased surface area resulting from porous Li growth 48,49 . In DEE, however, only a slight decrease in a semicircle size was detected, and it was subsequently stabilized―suggesting that Li plating occurs uniformly on the substrate without significantly altering the surface area. By contrast, during the initial stage of charging, a pronounced change in the EIS spectra was observed only in DME, showing a gradual increase in both capacitance and resistance. This increase has conventionally been ascribed to the growth and thickening of the SEI layer 50 . Although no distinct spectral changes were detected during the SEI growth stage in DEE, the frequency corresponding to the apex of the first semicircle was larger than that in DME, indicating a reduction in the time constant ( τ =R SEI ·C SEI ) 49,51 . This reduction might be attributed to the lower SEI resistance and capacitance 46 , arising from either faster Li + transport or the formation of a thinner and more compact SEI layer. In contrast to the DME, which exhibits slow desolvation kinetics, the weakly solvating DEE may contribute to such minor spectral change owing to faster SEI growth. To further clarify this, we monitored the EIS spectral changes by holding the cell potential at 0 V for 10 hours (Fig. 5d). The 0 V hold protocol is expected to reasonably lead to induce electrochemical SEI formation without Li plating (Fig. 5a). During the voltage hold, the impedance in DME increased steadily, consistent with progressive SEI thickening and accumulation of resistive components. In DEE, however, the impedance remained nearly constant, with only a slight decrease observed over time, suggesting stabilization of interfacial resistance. These results suggest that DEE promotes the formation of a thinner and more compact SEI, which supports efficient Li + transport. X-ray photoelectron spectroscopy (XPS) survey of the samples held at 0 V for 10 h revealed that solvent, whether DME or DEE, is not decisive in determining the SEI components (Fig. 5e and Supplementary Fig. 8). Instead, pronounced differences were observed in the SEI layers formed on single crystal Cu(111) and polycrystalline Cu depending on the solvent. Recent studies 52 have further shown that the O 1s XPS signal arises predominantly from amorphous organic species, such as Li 2 CO 3 and ROCO 2 Li, whereas the contribution from Li 2 O tends to be overestimated, since Li 2 CO 3 can decompose into Li 2 O artifacts during Ar + sputtering 53 . To avoid misinterpretation arising from such artifacts, we therefore focused on the organic species around 531 eV. Depth profiling revealed pronounced differences between polycrystalline Cu and single crystal Cu in DME, whereas the two appeared nearly identical in DEE. Similar to the facet-driven non-uniform Li plating, the extent and uniformity of SEI formation were strongly affected by the substrate crystallinity in DME, but showed little dependence in DEE. Collectively, these findings underscore not only the critical role of substrate single crystallinity, but also the importance of designing electrolytes that can effectively regulate crystallinity-driven non-uniformity. Conclusions This study establishes ion solvation structure as a decisive factor governing the crystallographic dependence of lithium plating. Using a Cu(111)/(115) twin-foil platform, we showed that DME-based electrolytes enforce strong Li⁺–substrate coupling, leading to facet-selective, inner-sphere-like deposition, whereas DEE-based electrolytes mitigate this coupling by weakening solvation and suppressing interfacial dipole ordering, resulting in facet-insensitive, outer-sphere-dominated growth. These insights define a broader design rule: electrolytes that minimize Li⁺–substrate interaction are best suited to suppress facet-sensitive nucleation and enable uniform deposition. In this context, asymmetric ethers 15 exemplify promising candidates, as their steric asymmetry reduces coordination, disrupts interfacial ordering, and promotes partially solvated charge transfer. Rational electrolyte design should therefore prioritize modulation of solvation and interfacial structure to weaken substrate coupling, outpace facet selectivity, and realize denser, more stable Li plating—an essential step toward practical Li metal batteries. More broadly, this solvation engineering principle provides general guidelines for controlling interfacial coupling and advancing the electrodeposition of other reactive metals in next-generation energy storage technologies. Methods Material preparation LiFSI salt was purchased from MTI Corporation. LiTFSI, LiDFOB, LiBF₄, LiClO₄, and LiNO₃ salts were obtained from Sigma-Aldrich. DME and DEC solvents were purchased from Sigma-Aldrich, EC and FEC from Alfa Aesar, and DEE, TTE, and DMI from TCI. Electrolytes were prepared by dissolving the salts in the corresponding solvents and stirring the solutions overnight in an Ar-filled glove box. Single-crystalline copper preparation Single-crystal Cu foils with controlled crystallographic orientations were prepared using a contact-free annealing (CFA) method. For crystal growth, Cu foil (99.9%) was suspended on a quartz holder and annealed at 1323 K (close to the melting point of Cu) for 12 h under a flowing H₂/Ar gas mixture. H₂ was introduced to prevent surface oxidation during annealing and supply additional energy for abnormal grain growth. Under these conditions, a few microscale {111} grains underwent abnormal grain growth, eventually extending into centimetre-scale single-crystal domains with a {111}〈112〉 orientation. Twin Cu(111)/(115) foil preparation Cu(111) foils that contain some grains with a (115) surface orientation were made through similar CFA methods. During CFA, (115) grains (having a ∑3 twin relationship with respect to the (111) plane) are sometimes present in Cu(111) foil because of the formation of annealing twins. The Cu(115) surface is a high-index facet, consisting of multiple (100) terraces and steps. The (115) grains are easily distinguished using scanning electron microscopy (SEM) and optical microscopy (OM), as they have a rectangular shape with length of several hundreds micrometers and are well aligned with each other along the three-fold symmetry of the parent (111) plane. Scanning electron microscopy and optical microscopy characterizations All post-mortem observations following Li plating or stripping on single-crystalline Cu substrates were conducted using CR2032-type coin cells assembled with 14 mm Cu substates and 12.5 mm Li metal electrodes. After deposition to the target capacities, the cells were dissembled and rinsed with DMC inside an Ar-filled glove box. For ex-situ optical microscopy, the electrodes were sealed between a slide glass and a cover slip using vacuum grease. Operando optical microscopy was performed using a custom-built cell consisting of single-crystalline Cu as working electrode and thin Li metal as counter electrode sandwiched between a glass slide and a cover slip. All optical observations were conducted at 50x magnification under a current density of 0.5 mA cm − 2 . The graph data (Fig. 2 d) were processed using MATLAB by extracting intensities across the electrode area during plating. Electrochemical measurements Electrochemical measurements were performed using CR2032 coin cells assembled in an Ar-filled glove box. A 50µm-thick polycrystalline Cu foil (NILACO co.) was used as the working electrode. The tests were conducted using a BioLogic VMP3 multi-channel potentiostat. Cyclic voltammetry (CV) was conducted using 14 mm Cu substrate and 16 mm Li metal at a scan rate of 1 mV s − 1 over a voltage range of -0.1V to 1V with IR compensation. The CV data were fitted using MATLAB based on various charge transfer models, including Butler-Volmer (Eq. 1) and Marcus-Hush-Chidsey models (Eq. 2) by following equations, j = j 0 [ \(\:\text{exp(-}\frac{\text{αFη}}{\text{RT}}\text{)-exp(}\frac{\left(\text{1-}\text{α}\right)\text{Fη}}{\text{RT}}\text{)}]\) (1) j = j 0 \(\:{\int\:}_{\text{-∞}}^{\text{∞}}\text{[}\text{exp}\left(\text{-}\frac{{\left(\text{ε}\text{-}\lambda\text{+}\text{eη}\right)}^{\text{2}}}{\text{4}\lambda{\text{k}}_{\text{B}}\text{T}}\right)\text{-}\text{exp}\left(\text{-}\frac{{\left(\text{ε}\text{-}\lambda\text{-}\text{eη}\right)}^{\text{2}}}{\text{4}\lambda{\text{k}}_{\text{B}}\text{T}}\right)\text{]}\frac{\text{d}\text{ε}}{\text{1+exp(}\frac{\text{ε}}{{\text{k}}_{\text{B}}\text{T}}\text{)}}\) (2) where j 0 is the exchange current density, α the charge transfer coefficient, F Faraday’s constant, η the overpotential (given by E - E eq ), R the gas constant, T the absolute temperature, ε the electron energy level, λ the reorganization energy, e the elementary charge, and k B Boltzmann’s constant. Electrochemical impedance spectroscopy (EIS) was conducted using three electrode cells, with Cu as the working electrode Li as the counter electrode, and a Li with reference electrode positioned between them. EIS spectra were collected over a frequency range of 20kHz to 100mHz using 5mV perturbation amplitude and 10 points per decade, at 30 ̊C. Staircase potentio electrochemical impedance spectroscopy (SPEIS) measurements were performed by applying incremental voltage steps from 1V to -0.1V, with a step size of 0.01V and impedance data collected at each step. Raman spectroscopy and X-ray Photoelectron spectroscopy Raman spectroscopy was conducted using an alpha300S system with a 785 nm laser (10 mW) for DME- and DEE-based electrolytes. All electrolytes were hermetically sealed between glass slides for measurement. The chemical state of etched elements was analyzed by high-performance X-ray photoelectron spectroscopy (HP-XPS, BS101; K-ALPHA+, Thermo Fisher Scientific, UK) using a monochromated Al Kα source (hν = 1486.6 eV, 12 kV, 72 W) with a spot size of 400 µm in diameter. Charge compensation was achieved using a dual flood gun (low-energy electrons and Ar⁺ ions). Measurements were carried out at the Yeongnam Regional Center of the Korea Basic Science Institute (KBSI). To prevent surface oxidation, samples were transferred using a vacuum transfer vessel. SEI layers were prepared by operating cells at a current density of 0.3 mA cm⁻² down to 0 V, followed by a 10 h hold at 0 V. Coin cells were disassembled in an Ar-filled glove box and rinsed with THF prior to analysis. Molecular Dynamics Molecular Dynamic simulations were performed on the Large-Scale Atomic/Molecular Massively Parallel Simulator (LAMMPS). Packmol software was used to establish the initial configuration of the simulation box. All the force field parameters for molecular dynamics simulations were generated by AutoFF. The solvation structures of the studied systems were simulated at the Optimizied potential for liquid solution (opls) force field. The charge distribution of solvent molecules was calculated at the B3LYP/6–31 + + G (d, p) level. The time step was 0.2 fs. In an NPT ensemble and under a pressure of 1 bar, the temperature of the simulated boxes were heated from 298 K to 398K within 0.5 ns, kept at 398 K for 0.5 ns, then cooled down to 298 K within 0.5 ns, and finally equilibrated at 298 K for 1 ns. The statistical averages were calculated in an NVT ensemble based on traces with durations of 1 ns. The post-analysis of MD simulations was conducted on VMD. The visualization was achieved by VMD and J mol − 1 . Declarations Competing interests The authors declare no competing interests. Author contributions J.K., M.-H.K., and H.-W.L. conceived idea and designed the experiments. J.K. and M.-H.K. carried out the experiments and analyzed the data. H.L. performed the molecular dynamics simulations. M.H.K. and S.J. prepared the Cu(111)/(115) twin foils. S.C., T.Y.C., S.-J.J., J.K.K., and P.J.K. assisted with characterization. J.K. and T.Y.C. conducted the operando optical microscopy experiments and processed the data. J.-S.B. contributed to the X-ray photoelectron spectroscopy characterizations. J.K., M.-H.K., R.S.R., Y.L., and H.-W.L. discussed the results. J.K., M.-H.K., and H.-W.L. wrote the manuscript. All authors revised the manuscript. Acknowledgements This work was supported by 2025 Research Fund (1.250009.01) of UNIST, National Research Foundation of Korea (NRF) (RS-2024-00428511 and RS-2023-00208929), National Research Council of Science & Technology (NST) grant by the Korea government (MSIT) (No. GTL24012-000), and the Institute for Basic Science (IBS-R-019-D1). This study contains the results obtained by using the equipment of UNIST Central Research Facilities (UCRF). References Albertus P, Babinec S, Litzelman S, Newman A (2018) Status and challenges in enabling the lithium metal electrode for high-energy and low-cost rechargeable batteries. Nat Energy 3:16–21 Tikekar MD, Choudhury S, Tu Z, Archer L (2016) A. Design principles for electrolytes and interfaces for stable lithium-metal batteries. Nat Energy 1:16114 Xiao J (2019) How lithium dendrites form in liquid batteries. Sci (80-) 366:426–427 Xu W et al (2014) Lithium metal anodes for rechargeable batteries. Energy Environ Sci 7:513–537 Qian J et al (2015) High rate and stable cycling of lithium metal anode. Nat Commun 6:6362 Suo L, Hu Y-S, Li H, Armand M, Chen L (2013) A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries. Nat Commun 4:1481 Chen S et al (2018) High-Voltage Lithium‐Metal Batteries Enabled by Localized High‐Concentration Electrolytes. Adv Mater 30 Kim SC et al (2023) High-entropy electrolytes for practical lithium metal batteries. Nat Energy 8:814–826 Louli AJ et al (2020) Diagnosing and correcting anode-free cell failure via electrolyte and morphological analysis. Nat Energy 5:693–702 Cao X et al (2021) Effects of fluorinated solvents on electrolyte solvation structures and electrode/electrolyte interphases for lithium metal batteries. Proc. Natl. Acad. Sci. U. S. A. 118, 1–9 Wang H et al (2022) Liquid electrolyte: The nexus of practical lithium metal batteries. Joule 2:1–29 Chen Y et al (2021) Steric Effect Tuned Ion Solvation Enabling Stable Cycling of High-Voltage Lithium Metal Battery. J Am Chem Soc 143:18703–18713 Yu Z et al (2020) Molecular design for electrolyte solvents enabling energy-dense and long-cycling lithium metal batteries. Nat Energy 5:526–533 Yu Z et al (2022) Rational solvent molecule tuning for high-performance lithium metal battery electrolytes. Nat Energy 7:94–106 Choi IR et al (2025) Asymmetric ether solvents for high-rate lithium metal batteries. Nat Energy 10:365–379 Zhang E et al (2025) Monofluorinated acetal electrolyte for high-performance lithium metal batteries. Proc. Natl. Acad. Sci. 122, 2017 Li A et al (2024) Methylation enables the use of fluorine-free ether electrolytes in high-voltage lithium metal batteries. Nat Chem 16 Yamada Y, Wang J, Ko S, Watanabe E, Yamada A (2019) Advances and issues in developing salt-concentrated battery electrolytes. Nat Energy 4:269–280 Kim SC et al (2021) Potentiometric Measurement to Probe Solvation Energy and Its Correlation to Lithium Battery Cyclability. J Am Chem Soc 143:10301–10308 Yan C et al (2021) Nucleation and Growth Mechanism of Anion-Derived Solid Electrolyte Interphase in Rechargeable Batteries. Angew Chemie Int Ed 60:8521–8525 Kim M-H et al (2024) Horizontal lithium growth driven by surface dynamics on single crystal Cu(111) foil. Energy Environ Sci 17:6521–6532 Jin S et al (2018) Colossal grain growth yields single-crystal metal foils by contact-free annealing. Sci (80-) 362:1021–1025 Jin S, Ruoff RS (2019) Preparation and uses of large area single crystal metal foils. APL Mater 7 Wu M et al (2020) Seeded growth of large single-crystal copper foils with high-index facets. Nature 581:406–410 Xiao C et al (2020) High-Index-Facet- and High-Surface-Energy Nanocrystals of Metals and Metal Oxides as Highly Efficient Catalysts. Joule 4:2562–2598 Bozzolo N, Bernacki M (2020) Viewpoint on the Formation and Evolution of Annealing Twins During Thermomechanical Processing of FCC Metals and Alloys. Metall Mater Trans Phys Metall Mater Sci 51:2665–2684 Staikov G (2007) Electrocrystallization in Nanotechnology. Wiley. 10.1002/9783527610198 Li Y, Kim M, Xie Z, Min J, Li Y (2024) Microelectrodes for Battery Materials. ACS Nano 18:35119–35129 Boyle DT et al (2020) Transient Voltammetry with Ultramicroelectrodes Reveals the Electron Transfer Kinetics of Lithium Metal Anodes. ACS Energy Lett 5:701–709 Boyle DT et al (2022) Correlating Kinetics to Cyclability Reveals Thermodynamic Origin of Lithium Anode Morphology in Liquid Electrolytes. J Am Chem Soc 144:20717–20725 Li Y, Qi Y (2019) Energy landscape of the charge transfer reaction at the complex Li/SEI/electrolyte interface. Energy Environ Sci 12:1286–1295 Xu Y et al (2020) Ion-Transport-Rectifying Layer Enables Li-Metal Batteries with High Energy Density. Matter 3:1685–1700 Bergstrom HK, McCloskey BD (2024) Ion Transport in (Localized) High Concentration Electrolytes for Li-Based Batteries. ACS Energy Lett 9:373–380 Lu Y et al (2015) Stable Cycling of Lithium Metal Batteries Using High Transference Number Electrolytes. Adv Energy Mater 5:1–7 Lee Y, Ma B, Bai P (2020) Concentration polarization and metal dendrite initiation in isolated electrolyte microchannels. Energy Environ Sci 13:3504–3513 Kwon H et al (2025) Covariance of interphasic properties and fast chargeability of energy-dense lithium metal batteries. Nat Energy. 10.1038/s41560-025-01838-1 Boyle DT et al (2021) Corrosion of lithium metal anodes during calendar ageing and its microscopic origins. Nat Energy 6:487–494 He X et al (2021) The passivity of lithium electrodes in liquid electrolytes for secondary batteries. Nat Rev Mater 6:1036–1052 Yuan X, Liu B, Mecklenburg M, Li Y (2023) Ultrafast deposition of faceted lithium polyhedra by outpacing SEI formation. Nature 620:86–91 Thanh NTK, Maclean N, Mahiddine S (2014) Mechanisms of Nucleation and Growth of Nanoparticles in Solution. Chem Rev 114:7610–7630 BARD ALLENJ, FAULKNER LARRYR (2022) H. S. W. Electrochem METHODS Fundamentals Appl, 3rd Fraggedakis D et al (2021) Theory of coupled ion-electron transfer kinetics. Electrochim Acta 367:137432 Lazanas AC, Prodromidis MI (2023) Electrochemical Impedance SpectroscopyA Tutorial. ACS Meas Sci Au 3:162–193 Minter RD, Juarez-Robles D, Fear C, Barsukov Y, Mukherjee PP (2018) Three-electrode Coin Cell Preparation and Electrodeposition Analytics for Lithium-ion Batteries. J. Vis. Exp. 1–11 (2018) Juarez-Robles D, Chen C-F, Barsukov Y, Mukherjee P (2017) Impedance Evolution Characteristics in Lithium-Ion Batteries. J Electrochem Soc 164:A837–A847 Talian SD, Brutti S, Navarra MA, Moškon J, Gaberscek M (2024) Impedance spectroscopy applied to lithium battery materials: Good practices in measurements and analyses. Energy Storage Mater 69 Gaberšček M (2021) Understanding Li-based battery materials via electrochemical impedance spectroscopy. Nat Commun 12:6513 Drvarič Talian S, Bobnar J, Sinigoj AR, Humar I, Gaberšček M (2019) Transmission Line Model for Description of the Impedance Response of Li Electrodes with Dendritic Growth. J Phys Chem C 123:27997–28007 Drvarič Talian S, Kapun G, Moškon J, Dominko R, Gaberšček M (2025) Operando impedance spectroscopy with combined dynamic measurements and overvoltage analysis in lithium metal batteries. Nat Commun 16:2030 Srout M, Carboni M, Gonzalez J, Trabesinger S (2023) Insights into the Importance of Native Passivation Layer and Interface Reactivity of Metallic Lithium by Electrochemical Impedance Spectroscopy. Small 19:1–10 Lu Y, Zhao C-Z, Huang J-Q, Zhang Q (2022) The timescale identification decoupling complicated kinetic processes in lithium batteries. Joule 6:1172–1198 Tan S et al (2025) Synchronized Breathing in Anion-Derived Interphases. ACS Energy Lett 10:3746–3754 Yu W, Yu Z, Cui Y, Bao Z (2022) Degradation and Speciation of Li Salts during XPS Analysis for Battery Research. ACS Energy Lett 7:3270–3275 Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryVideo1.mp4 SUPPLEMENTARY Video 1 SupplementaryVideo2.mp4 SUPPLEMENTARY Video 2 SupplementaryInformationfinal.docx Supplementary Information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7588130","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":522384661,"identity":"79ee1a71-95e7-478c-8a93-6048c5bd3069","order_by":0,"name":"Hyun-Wook Lee","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYBACAxjNx8DA+ICBQQLKP0CEFjYGBmYDkBYeUrSwga0gqMVcIsfsMU+FjTEbe/Oxat4dFvb2DMwPPzCcuYdTi+WMHHNjnjNpZmw8x9Ju856RSOxhYDOWYLhRjNthN3LMpHPbDtuwAa27zdsmkQB0mBkDw4cEIrTIv/9WDNRiz8PA/o0oLWZsEjxszEAtjD0MPEBbbuDRcuZZmfSfM2nGbDxpxpJz24B+OcxTLJFwBo+W48nbJGdU2Bj2sx9++OFtW509e3v7xg8fjuHWwiCAIccMxHg0MDDwH8AnOwpGwSgYBaMACAAFP0lEwnE5VgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-9074-1619","institution":"Ulsan National Institute of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Hyun-Wook","middleName":"","lastName":"Lee","suffix":""},{"id":522384662,"identity":"44a99f8a-8e70-4d8f-ba2d-6fba98e70464","order_by":1,"name":"Juyoung Kim","email":"","orcid":"","institution":"Ulsan National Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Juyoung","middleName":"","lastName":"Kim","suffix":""},{"id":522384663,"identity":"60614df9-0692-4a90-9147-04b9502b4eb9","order_by":2,"name":"Min-Ho Kim","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Min-Ho","middleName":"","lastName":"Kim","suffix":""},{"id":522384664,"identity":"a3e2b948-de94-41e8-8831-1703dd1debfe","order_by":3,"name":"Huida Lyu","email":"","orcid":"https://orcid.org/0009-0000-1935-5117","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Huida","middleName":"","lastName":"Lyu","suffix":""},{"id":522384665,"identity":"c2b06d09-b953-42a8-9c9b-7b3bf7a7cc50","order_by":4,"name":"Min Hyeok Kim","email":"","orcid":"https://orcid.org/0000-0003-2433-5975","institution":"Institute for Basic Science Center","correspondingAuthor":false,"prefix":"","firstName":"Min","middleName":"Hyeok","lastName":"Kim","suffix":""},{"id":522384666,"identity":"132ec066-29dd-41be-9ce6-deefa556c941","order_by":5,"name":"Seungwoo Choi","email":"","orcid":"","institution":"Ulsan National Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Seungwoo","middleName":"","lastName":"Choi","suffix":""},{"id":522384667,"identity":"077ead7f-f12e-410e-80d0-95f17f3f298d","order_by":6,"name":"Tae Young Choi","email":"","orcid":"","institution":"Ulsan National Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Tae","middleName":"Young","lastName":"Choi","suffix":""},{"id":522384668,"identity":"314055e9-9f8c-4887-86b9-4d9ad0899a55","order_by":7,"name":"Soon-Jae Jung","email":"","orcid":"","institution":"Ulsan National Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Soon-Jae","middleName":"","lastName":"Jung","suffix":""},{"id":522384669,"identity":"b79a922f-ccbc-45b8-8488-b61ad0f2c5b8","order_by":8,"name":"Jin Koo Kim","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Jin","middleName":"Koo","lastName":"Kim","suffix":""},{"id":522384670,"identity":"23e77443-3945-45ba-a44a-b96e33c5e0a9","order_by":9,"name":"Patrick Kim","email":"","orcid":"","institution":"Kyungpook National University","correspondingAuthor":false,"prefix":"","firstName":"Patrick","middleName":"","lastName":"Kim","suffix":""},{"id":522384671,"identity":"e2e142b5-2dfb-4300-8185-6cce6ce0a9a1","order_by":10,"name":"Rodney Ruoff","email":"","orcid":"","institution":"Ulsan National Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Rodney","middleName":"","lastName":"Ruoff","suffix":""},{"id":522384672,"identity":"bf0670b7-1626-4275-af9a-d427f56bf25c","order_by":11,"name":"Yuzhang Li","email":"","orcid":"","institution":"University of California, Los Angeles","correspondingAuthor":false,"prefix":"","firstName":"Yuzhang","middleName":"","lastName":"Li","suffix":""},{"id":522384673,"identity":"cff50ddb-2b65-475a-9627-b6e089b1827e","order_by":12,"name":"Sunghwan Jin","email":"","orcid":"","institution":"Kangwon National University","correspondingAuthor":false,"prefix":"","firstName":"Sunghwan","middleName":"","lastName":"Jin","suffix":""},{"id":522384674,"identity":"7dc4a0f8-fd7a-4276-8245-bbdb8fba4e51","order_by":13,"name":"Jong-Seong Bae","email":"","orcid":"","institution":"Korea Basic Science Institute (KBSI)","correspondingAuthor":false,"prefix":"","firstName":"Jong-Seong","middleName":"","lastName":"Bae","suffix":""}],"badges":[],"createdAt":"2025-09-11 05:55:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7588130/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7588130/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94055972,"identity":"b0d4dc70-e331-49d4-a7d4-ea68d7f0d662","added_by":"auto","created_at":"2025-10-22 03:44:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":295459,"visible":true,"origin":"","legend":"\u003cp\u003eSelective lithium deposition on twin grains and boundaries in Cu(111)/(115) twin foil. a, Electron backscattered diffraction (EBSD) inverse pole figure (IPF) map of a Cu(111) foil embedded with (115) twin grains. The EBSD clarifies the presence of a (115) twin grain (pink) within a (111) terrace (blue). b, Schematic illustration of selective Li deposition on high index Cu embedded in low index Cu, reflecting the thermodynamic preference of Li atoms after desolvation and charge transfer. c, Tilted scanning electron microscopy (SEM) images of a Cu(111)/(115) twin foil before (left) and after Li deposition (right) in 1.3 M LiPF\u003csub\u003e6\u003c/sub\u003e in EC/DEC (3:7 v/v) with 10 wt% FEC. Inset: top-view image.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/911e4a73f785b349e63cfe6d.png"},{"id":94056224,"identity":"7fba86cc-3e00-45e7-abc2-404b33fd1093","added_by":"auto","created_at":"2025-10-22 03:52:56","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":757611,"visible":true,"origin":"","legend":"\u003cp\u003eFacet-dependent Li deposition as a function of electrolyte formulation. a, Optical microscopy images of Li morphologies on Cu(111)/(115) twin foils in ten different electrolytes at a fixed plating capacity of 0.1 mAh cm\u003csup\u003e-2\u003c/sup\u003e. Li predominantly accumulates along twin boundaries and on twin grains at low current densities. b, Li plating distributions at low and high current densities. At low current density, Li deposition occurs irrespective of facet effect, whereas at high current density, Li preferentially deposits on high index Cu twins. c, Critical current densities above which facet-selective Li deposition emerges for each electrolyte. Ether-based electrolytes exhibit relatively higher critical current densities than carbonate-based electrolytes. d, \u003cem\u003eOperando\u003c/em\u003e optical microscopic analysis comparing Li plating in 1 M LiFSI in DME and DEE. In DME, Li preferentially accumulates on the twin grain, whereas such facet-dependent behavior is absent in DEE.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/6b74b2998ae72dd81a6fc447.png"},{"id":94055973,"identity":"322f246d-031f-4a5d-b831-7fb82630f705","added_by":"auto","created_at":"2025-10-22 03:44:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":401832,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between solvation structure and Li deposition behavior. a, Molecular structures of DME and DEE, and their Li\u003csup\u003e+\u003c/sup\u003e-coordinated complexes. The donor numbers of DME and DEE are ~20 and ~10, respectively, which critically determine solvation strength. Since DEE has weak solvating ability, more anions participate in the solvation shell in the DEE system than in DME system. b, Comparison of high-index selectivity in pure DME, pure DEE, and mixed solvents (DME:DEE = 80:20, 50:50, 20:80) using optical microscopy (top) and SEM (bottom). Pure DME and the 80:20 DME:DEE mixture exhibit facet-dependent deposition, whereas pure DEE and mixtures of 50:50 and 20:80 outpace thermodynamic rules in Li deposition. c, coordination number (C.N.) obtained from radial distribution functions calculated by molecular dynamics (MD) simulations. The solvation structure is characterized by the number of FSI\u003csup\u003e-\u003c/sup\u003e anions in the solvation shell. d, Snapshots of the MD simulation for the 50:50 DME:DEE mixture. Four different Li\u003csup\u003e+\u003c/sup\u003e-coordinated structures, confirmed by the simulation, are labeled as #1-#4. e, Raman spectra of five different electrolytes in the range of 680-900 cm\u003csup\u003e-1\u003c/sup\u003e. f, Ratios of aggregates (AGG), contact ion pair (CIP), and free FSI\u003csup\u003e-\u003c/sup\u003e anions (top) and the ratio of DEE to DME in the solvation shell (bottom).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/6ff19baf311ebcc410373fe8.png"},{"id":94055975,"identity":"26123682-35ab-47ca-88e6-c412c67d594e","added_by":"auto","created_at":"2025-10-22 03:44:56","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":309304,"visible":true,"origin":"","legend":"\u003cp\u003eSolvation-dependent charge transfer mechanisms. a, Current-voltage plots obtained by cyclic voltammetry (CV) at a scan rate of 1 mV s\u003csup\u003e-1\u003c/sup\u003e. The dotted circle highlights the kinetically controlled region, where backward-sweep current response is zero. This region is not perturbed by mass-transfer limitations and governed solely by charge transfer. The right panel represents deviation from the linear approximation with increasing overpotential, indicating breakdown of the Butler-Volmer approximation (red dotted line). b, Tafel plots obtained in the kinetic regime showing the experimental data along with Butler-Volmer (BV) and Marcus-Hush-Chidsey (MHC) fits. c, Measured exchange current densities and R square (R\u003csup\u003e2\u003c/sup\u003e) values, representing the goodness of fit of the Tafel plots to the BV model and the MHC model. Poor agreement with both models is observed for pure DME and the 80:20 DME:DEE mixture. d, Proposed charge transfer energy landscape. Non-adiabatic charge transfer, characterized by an electron tunneling pathway, represents the DEE system, whereas adiabatic charge transfer via direct orbital-overlap pathway represents DME system. e, Schematic illustrations of the proposed charge transfer mechanisms. In the DME system, charge transfer is hindered by both strong solvating ability and well-aligned interfacial solvent at the interface. This strong shielding enforces full desolvation followed by charge transfer, leading to strong Li-Cu coupling (inner-sphere mechanism, facet-selective nucleation). On the other hand, in the DEE system, diminished interfacial screening owing to low dipole moment of the solvent does not require complete desolvation, thereby rendering interfacial solvent to shield Li\u003csup\u003e+\u003c/sup\u003e from the substrate and resulting in weak Li-Cu coupling (outer-sphere mechanism, outfacing facet selectivity).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/a77f8fa1257c071b38ef28fa.png"},{"id":94056225,"identity":"983594a6-2918-47b1-9295-ace0b95242d1","added_by":"auto","created_at":"2025-10-22 03:52:56","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":328090,"visible":true,"origin":"","legend":"\u003cp\u003eImpedance and interfacial analysis of SEI formation and Li nucleation. a, Stepwise potentiometric electrochemical impedance spectroscopy (SPEIS) protocol, in which voltage was swept from open-circuit voltage (OCV) to -0.10 V with 0.01 V increments. Current responses indicate Li plating onset at -0.04 V in DME and -0.02 V in DEE. b,Comparison of two-electrode and three-electrode configurations, showing that impedance in two-electrode cells is dominated by the Li counter electrode, underscoring the necessity of a Li reference electrode to isolate the Cu response. c,Evolution of Nyquist plots during Li plating in DME and DEE. In DME, the semicircle magnitude increases with SEI thickening and decreases with porous Li growth, whereas in DEE only slight changes are observed, suggesting uniform Li plating. d,Impedance evolution during a 0 V hold for 10 h. DME exhibits steadily increasing impedance consistent with progressive SEI thickening, while DEE shows nearly constant impedance with only minor decreases, suggesting a thinner, more compact SEI. e, X-ray photoelectron spectroscopy (XPS) survey of SEI layers formed after 10 h at 0 V. Organic species around 531 eV dominate the O 1s signal, while differences between polycrystalline and single-crystal Cu are pronounced in DME but negligible in DEE, reflecting crystallinity-driven non-uniformity.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/4ec22950ebc4cd2b4eb7ac08.png"},{"id":94056408,"identity":"5f6a5377-6745-4152-bb98-424364647dc1","added_by":"auto","created_at":"2025-10-22 04:00:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2615312,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/90fdee5a-81f0-4938-a35c-87f1bc1a627c.pdf"},{"id":94055976,"identity":"b1070a5f-3daf-46cc-a80d-798c58f0f061","added_by":"auto","created_at":"2025-10-22 03:44:56","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3200929,"visible":true,"origin":"","legend":"SUPPLEMENTARY Video 1","description":"","filename":"SupplementaryVideo1.mp4","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/58d2546170529ab2e933e5f8.mp4"},{"id":94055978,"identity":"e2a6397b-b15e-4d4b-aa1d-8889022561d0","added_by":"auto","created_at":"2025-10-22 03:44:56","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3183856,"visible":true,"origin":"","legend":"SUPPLEMENTARY Video 2","description":"","filename":"SupplementaryVideo2.mp4","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/d52c79b8b659459a15f231b3.mp4"},{"id":94055979,"identity":"915abb7d-409d-4f00-a6e1-f037e8bebe0f","added_by":"auto","created_at":"2025-10-22 03:44:56","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":8801228,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformationfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7588130/v1/0352bb9d8f6ba75f95e51b24.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Tuning ion solvation to bypass facet-selective lithium deposition","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLithium (Li) metal anodes, with their exceptionally high theoretical capacity (3,860 mAh g\u003csup\u003e-1\u003c/sup\u003e) and the lowest redox potential (-3.04 V versus standard hydrogen electrode), are considered critical for enabling next-generation high-energy-density batteries\u003csup\u003e1\u003c/sup\u003e. However, their practical implementation remains hindered by the inherently heterogeneous nature of Li plating and poor interfacial stability, which lead to dendritic growth and continuous electrolyte consumption, respectively\u003csup\u003e2\u0026ndash;4\u003c/sup\u003e. Together, these issues ultimately cause premature cell failure, and the slow progress in addressing these weaknesses has increasingly cast doubt on the practical viability of lithium metal batteries (LMBs).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Among various strategies to overcome these challenges, electrolyte engineering has delivered the most dramatic progress, particularly in improving Coulombic efficiency (CE) and cycling stability\u003csup\u003e5\u0026ndash;11\u003c/sup\u003e. While numerous electrolyte concepts have been introduced, ether solvents (\u003cem\u003ee.g.\u003c/em\u003e, 1,2-dimethoxyethane (DME), 1,3-dioxolane (DOL)) have been widely adopted as model systems owing to their relatively superior reductive stability against Li\u003csup\u003e0\u003c/sup\u003e―that is, high compatibility with Li\u003csup\u003e0\u003c/sup\u003e―compared to practical carbonate counterparts (\u003cem\u003ee.g.\u003c/em\u003e, ethylene carbonate (EC), dimethyl carbonate (DMC), diethyl carbonate (DEC))\u003csup\u003e5,6\u003c/sup\u003e. Building on this foundation, recent studies\u003csup\u003e12\u0026ndash;16\u003c/sup\u003e―most notably those led by the Bao group―have attracted considerable attention by tailoring ether structures, either through functional group substitution (\u003cem\u003ee.g.\u003c/em\u003e, 1,2-diethoxyethane (DEE)) or fluorination (\u003cem\u003ee.g.\u003c/em\u003e, fluoroethers), to develop weakly solvating or fluorine-rich solvents that further enhance CE and cycle life. Despite these significant achievements in suppressing dendritic growth and mitigating parasitic reactions, most electrolyte research has rationalized the observed performance enhancements primarily in terms of the solid-electrolyte-interphase (SEI) stabilization, particularly through the formation of anion-derived inorganic species\u003csup\u003e17\u0026ndash;20\u003c/sup\u003e. While this perspective has been valuable, it provides limited insight into how Li nucleates and grows on Cu substrate.\u003c/p\u003e\n\u003cp\u003eIn our previous work\u003csup\u003e21\u003c/sup\u003e, we newly introduced the concept of \u003cem\u003eepihorizontaxy\u003c/em\u003e, which refers to the preferential horizontal growth of Li facilitated by the weak interaction between Li adatoms and the bare Cu foil surface. This behavior arises from the negligible migration barrier of Li atoms on low-index Cu facet―particularly Cu(111)―with their isotropic crystallographic orientation, whereas high-index facets such as Cu(115) and Cu(410) promote vertical Li growth. This finding implies that the coexistence of diverse crystallographic facets and additional microstructural features (\u003cem\u003ee.g.,\u003c/em\u003e grain boundaries, roughness) in polycrystalline Cu foil not only exacerbates morphological heterogeneity in Li deposition but also complicates efforts to decouple and precisely assess the role of substrate\u0026ndash;Li interactions.\u003c/p\u003e\n\u003cp\u003eHere, we report, for the first time, the facet dependence of Li deposition on Cu foils as a function of electrolyte formulation―an effect previously unrecognized―and identify the key electrolyte parameters that enable outpacing facet selectivity. By establishing a direct link between solvation structure, charge-transfer behavior, and crystallographic surface effects, we demonstrate a pathway to achieve uniform Li deposition from the atomic scale to practical electrodes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCu(111)/(115) twin foil as a platform for facet selectivity\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSingle-crystal Cu(111) foil can be obtained from commercially available polycrystalline Cu foil through a contact-free annealing process\u003csup\u003e21\u0026ndash;23\u003c/sup\u003e. Under such contact-free conditions, grain growth is governed primarily by the surface energy of the foil itself, driving Cu grains to abnormally evolve toward the (111) orientation, which is the thermodynamically most stable surface of face-centered cubic (FCC) structure owing to its lowest surface energy, in contrast to high-index facets which possess higher surface energies due to their open atomic arrangements and undercoordinated surface atoms\u003csup\u003e24,25\u003c/sup\u003e. Interestingly, however, (115) high-index plane often coexists with (111) through step-terrace configurations or twin boundary relationships (\u003cem\u003ee.g.\u003c/em\u003e, \u0026sum;3 {111} twins), where the interfacial energy penalty is substantially reduced\u003csup\u003e21,26\u003c/sup\u003e. As a result, a special Cu foil comprising predominantly (111) domains with minor fractions of (115) twin grains occasionally emerge during annealing (Fig. 1a).\u003c/p\u003e\n\u003cp\u003eIn principle, we hypothesized that incoming Li atoms preferentially attach to high-index facets and their grain boundaries in order to minimize surface energy, as dictated by thermodynamics \u003csup\u003e21,27\u003c/sup\u003e, after undergoing desolvation and charge transfer (Fig. 1b). Accordingly, the unique Cu(111)/(115) twin foil (hereafter, termed as tCu(111)/(115)) provides a well-defined platform to investigate facet-selective Li deposition, enabling direct comparison between the inherently stable (111) terraces and the higher-energy (115) surface within the same substrate. Consistent with this hypothesis, we demonstrated that LMBs follow this thermodynamic rule in a conventional carbonate electrolyte (1.3 M lithium hexafluorophosphate (LiPF\u003csub\u003e6\u003c/sub\u003e) in EC/DEC + 10 wt% fluoroethylene carbonate (FEC)), where Li preferentially deposits at twin grain boundaries and on Cu(115) facets rather than on Cu(111) (Fig. 1c).\u003c/p\u003e\n\u003cp\u003eWhile these results underscore the importance of interactions between Li adatoms and the substrate, a critical open question remains: \u003cem\u003eIs preferential plating on high-index facets a universal phenomenon?\u003c/em\u003e Given that Li plating is additionally influenced by multiple interfacial factors\u0026mdash;including solvation structures\u003csup\u003e12,13,15\u003c/sup\u003e, charge transfer kinetics\u003csup\u003e28\u0026ndash;31\u003c/sup\u003e, mass transport in the electrolytes\u003csup\u003e3,32\u0026ndash;35\u003c/sup\u003e, and the physicochemical properties of SEI layer\u003csup\u003e36\u0026ndash;38\u003c/sup\u003e\u0026mdash;it is plausible that facet selectivity can be reshaped by altering cell conditions and environments. In particular, with the growing recognition of solvation structure in LMBs, its potential role in directing preferential deposition on high-index facets deserves special emphasis.\u003c/p\u003e\n\u003cp\u003eTo elucidate the role of interfacial phenomena in dictating high-index facet selectivity across different electrolytes, we investigated Li deposition on the tCu(111)/(115) using a range of electrolytes commonly employed in LMBs, with a fixed plating capacity of 0.1 mAh cm\u003csup\u003e-2\u003c/sup\u003e (Fig. 2a and Supplementary Fig. 1 and 2). Irrespective of electrolyte composition, Li deposition was primarily governed by thermodynamics, with preferential nucleation occurring at grain boundaries and on high-index Cu(115) facets. Notably, such non-uniform deposition left the SEI residues locally after stripping, implying that repeating cycles can cause severe local increase in resistance (Supplementary Fig. 3). This trend persisted even in state-of-the-art high-performance electrolytes, such as 0.6 M lithium difluoro(oxalate)borate (LiDFOB) + 0.6 M lithium tetrafluoroborate (LiBF\u003csub\u003e4\u003c/sub\u003e) in FEC/DEC, 1 M lithium bis(fluorosulfonyl)imide (LiFSI) in DME, 1 M LiFSI in DME/1,1,2,2-tetrafluoroethyl 2,2,3,3-tetrafluoropropylether (TTE), and 1 M LiFSI in DEE. The only distinction observed was that the critical current density, above which site-selective Li deposition emerged, varied with the electrolytes (Fig. 2b\u0026nbsp;and Supplementary Fig. 4). Below this threshold, Li deposition was not governed by thermodynamics, but at higher current densities facet-dependent plating reappeared. Ether-based electrolytes exhibited relatively higher critical current densities than their carbonate-based counterparts (Fig. 2c), indicating that current density-driven interfacial chemistry plays a decisive role in determining the onset of site-selective Li deposition. We speculate that below critical current densities, Li deposition proceeds slowly enough for solvent\u0026ndash;Li⁺ interactions, solvation structure, and interfacial charge-transfer processes to significantly influence nucleation pathways, thereby modulating facet selectivity. At high current densities, however, rapid adatom accumulation dictates deposition, with local equilibrium governed mainly by surface energetics\u0026mdash;Similarly, an abnormal increase in Li\u003csup\u003e0\u003c/sup\u003e concentration beyond supersaturation point often forms rhombic dodecahedral Li enclosed by the stable {110} facet of the BCC lattice\u003csup\u003e21,39\u003c/sup\u003e, governed entirely by the intrinsic surface energy of Li through homogeneous nucleation, independent of the substrate\u003csup\u003e40\u003c/sup\u003e. In this regime, thermodynamics prevails over interfacial kinetic effects, leading to preferential Li nucleation at high-energy sites such as grain boundaries and high-index facets, irrespective of electrolyte formulation. In other words, a higher critical current density reflects solvation-mediated interfacial stabilization, which ultimately promotes \u0026lsquo;uniform\u0026rsquo; Li deposition.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;We further observed distinct deposition behaviors on tCu(111)/(115) in DME and DEE solvents through real-time top-view optical microscopy (Fig. 2d, Supplementary Fig. 5 and Supplementary Video 1 and 2). High-index facets and grain boundaries are clearly distinguished in optical images owing to orientation-dependent reflectivity. In DME, the high-index region progressively becomes brighter, indicative of localized Li plating. By contrast, in DEE, Li deposition is no longer constrained by thermodynamic preference, resulting in uniform nucleation across the substrate. Then, what accounts for this striking difference in deposition behavior between DME and DEE? Thus far, we have empirically shown that while the uniformity of Li deposition strongly depends on the crystallographic orientation of the substrate, its kinetics is closely correlated with the electrolyte characteristics. To elucidate this connection, it is essential to understand how the solvation structures differ between DME and DEE.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe effect of electrolyte solvation in facet-dependent deposition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA previous report\u003csup\u003e12\u003c/sup\u003e proposed that DEE solvent, obtained by substituting the methoxy groups of DME with larger ethoxy groups, effectively tunes ion solvation through steric hinderance, thereby weakening the coordinating ability of its two oxygen atoms (Fig. 3a). As a result, the inner solvation shell contains more FSI\u003csup\u003e-\u003c/sup\u003e anions in DEE (0.89 mol per mol of DEE) than in DME (0.76 mol per mol of DME). Although they attributed the improved CE solely to the formation of anion-derived SEI, this explanation cannot account for our findings regarding Li\u003csup\u003e+\u003c/sup\u003e-substrate interactions. Here, beginning with the comparison between DME and DEE, we deliberately expanded the control groups to include a series of DME-DEE mixed solvents (DME:DEE = 80:20, 50:50, 20:80) to strengthen the validity of our analysis (Fig. 3b). Even with the addition of a small fraction of DEE (20 vol%), the facet effect became slightly mitigated, showing reduced local plating along the grain boundaries. And, when half of the electrolyte volume consisted of DEE (50:50 DME:DEE), the crystallographic effect was barely discernible. Interestingly, Li morphology also shifted to particle- or nodule-like shapes in DEE-containing electrolytes, in contrast to the localized dendritic Li growth observed on high-index facets in DME. These observations collectively suggest that the spatial distribution of Li deposition and the resulting Li morphology on the substrate are intricately interconnected.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;To investigate the coordination environment of Li\u003csup\u003e+\u003c/sup\u003e, we performed the molecular dynamics (MD) simulations (Fig. 3c,d). As the DEE content in the electrolyte increased, the number of DEE molecules participating in the inner solvation shell also increased, whereas the FSI\u003csup\u003e-\u003c/sup\u003e anion coordination number (C.N.) remained nearly unchanged up to DME:DEE = 80:20, comparable to that in pure DME (Fig. 3c). At a 50:50 ratio, however, the FSI\u003csup\u003e-\u003c/sup\u003e C.N. showed a distinct increase but, contrary to expectation, did not change further, remaining essentially the same as in pure DEE. Although the FSI\u003csup\u003e-\u003c/sup\u003e density is nearly identical across DME:DEE = 50:50, 20:80, and pure DEE, multiple solvation configurations coexist in mixed solvents (Fig. 3d and Supplementary Fig. 6). It should be emphasized that solvation structure cannot be fully captured by anion coordination numbers alone. Local configurations\u0026mdash;such as the spatial distribution of solvent and anion species\u0026mdash;and their dynamic exchange at the interface are equally critical in dictating how Li⁺ couples to the substrate. These factors help explain why electrolytes with similar anion coordination numbers can nevertheless exhibit markedly different facet selectivity during lithium deposition.\u003c/p\u003e\n\u003cp\u003eConsistent with the MD results, Raman spectroscopy revealed that at high DEE contents a significant fraction of FSI\u003csup\u003e-\u003c/sup\u003e anions is incorporated into the solvation shell in the form of aggregated clusters (AGG) and contact ion pairs (CIP), thereby reducing the population of free FSI\u003csup\u003e-\u003c/sup\u003e anions (Fig. 3e,f and Supplementary Fig. 7). Notably, the CIP+AGG ratio divides the electrolytes into two distinct groups: (ⅰ) pure DME and DME:DEE = 80:20, and (ⅱ) DME:DEE = 50:50, 20:80, and pure DEE (Fig. 3f). This classification aligns well with the facet selectivity test results (Fig. 3b), supporting the notion that Li deposition behavior correlates more closely with the solvation structure than with the solvent identity itself. In particular, once the DEE content in the electrolyte reaches 20 %, DEE molecules constitute more than half of the solvents participating in the solvation shell, whereas in the 50:50 mixture the solvation shell remains DME-rich―even though the FSI\u003csup\u003e-\u003c/sup\u003e anion coordination number is elevated by DEE and remains essentially unchanged within group (ⅱ). These findings highlight again that it is the DEE-induced reorganization of the solvation sheath―rather than the DEE solvent molecules themselves―that governs facet selectivity and Li deposition morphology.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSolvation-dependent charge transfer mechanisms\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore solvation-dependent charge transfer mechanisms, we measured the kinetics of the Li\u003csup\u003e+\u003c/sup\u003e + e\u003csup\u003e-\u003c/sup\u003e \u0026rarr; Li\u003csup\u003e0\u003c/sup\u003e near the kinetically controlled region using cyclic voltammetry (CV) (Fig. 4a). The CV results revealed systematic deviations from the linearized Butler-Volmer (BV) response with increasing overpotential, highlighting the limitations of the BV approximation\u003csup\u003e29\u003c/sup\u003e. The BV model assumes that the activation barrier decreases linearly with applied potential, which is valid near equilibrium but often fails at higher overpotentials\u003csup\u003e41\u003c/sup\u003e. To better capture this behavior, we analyzed the Tafel slopes from five electrolytes using both the BV model and the Marcus-Hush-Chidsey (MHC) model (Fig. 4b). Unlike BV, the MHC framework explicitly incorporates the reorganization energy (\u0026lambda;) associated with solvent and interfacial restructuring, leading to a parabolic dependence of the activation barrier on potential\u003csup\u003e41,42\u003c/sup\u003e. In the case of Li plating on Cu foil, where electron transfer occurs at a metal-solution interface with a continuum of electronic states, the MHC model thus provides a more realistic description of the charge transfer kinetics across a broader potential window.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;By applying the linearized form of the Butler-Volmer equation, \u003cem\u003ej\u003c/em\u003e=\u003cem\u003ej\u003c/em\u003e\u003csub\u003e0\u003c/sub\u003e\u003cem\u003ef\u003c/em\u003e(\u003cem\u003eE\u003c/em\u003e-\u003cem\u003eE\u003c/em\u003e\u003csub\u003eeq\u003c/sub\u003e), we determined the exchange current densities (\u003cem\u003ej\u003csub\u003e0\u003c/sub\u003e\u003c/em\u003e) for five different electrolytes (Fig. 4c). The values obtained were 23.24 mA cm\u003csup\u003e-2\u003c/sup\u003e for DME, 32.28 mA cm\u003csup\u003e-2\u003c/sup\u003e for DEE, and a highest value of 45.67 mA cm\u003csup\u003e-2\u003c/sup\u003e for the DME:DEE 50:50 mixture. These results are consistent with previous reports\u003csup\u003e12\u003c/sup\u003e showing that DEE exhibit faster kinetics than DME. However, the overall trend did not correlate with the high-index facet selectivity observed in Fig. 3b;\u0026nbsp;for instance, the DME:DEE 80:20 showed a higher exchange current density than pure DEE, yet its Li plating remained facet-dependent. This suggests that outpacing facet-dependent Li plating is not governed by the fast kinetics alone.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Strikingly, pure DEE and DEE-based electrolytes, including the 50:50 mixture, were well-fitted to both the BV and MHC models, yielding R square (R\u003csup\u003e2\u003c/sup\u003e) values exceeding 0.95. In contrast, pure DME and DME-based electrolytes exhibited comparatively low R\u003csup\u003e2\u003c/sup\u003e values. Indeed, this finding is consistent with previous reports\u003csup\u003e29\u003c/sup\u003e suggesting that Marcus-type fitting in DME electrolytes fails to accurately describe charge transfer phenomena, because the scan rates are insufficient to drive the system into the charge transfer-limited regime. However, more recent publication\u003csup\u003e15\u003c/sup\u003e excluded this possibility by employing ultrafast scan rates, instead attributing the discrepancy to the symmetric molecular structure and strong solvation of DME, which reinforce solvent shielding and impose an additional desolvation barrier. While the authors compared symmetric and asymmetric ethers, our results demonstrate that substantial differences can also arise between two symmetric molecules, DME and DEE, particularly in their facet selectivity. This suggests that solvation strength and desolvation dynamics, rather than molecular symmetry alone, critically dictate the apparent charge transfer kinetics and interfacial deposition behavior.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Taken together, we propose a key mechanistic distinction between DME- and DEE-based electrolytes (Fig. 4d,e). Marcus theory was originally formulated for outer-sphere electron transfer, in which charge transfer proceeds via solvent reorganization and electron tunneling without direct orbital overlap with the electrode\u003csup\u003e41\u003c/sup\u003e (Fig. 4d). In DME, however, desolvation becomes the rate-limiting step: Li⁺ must shed much of its strongly bound solvation shell and also displace the dipole-aligned solvent molecules residing directly at the electrode interface\u003csup\u003e41\u003c/sup\u003e (Left in Fig. 4e). This effective removal of both coordinated and interfacial solvent leaves Li\u003csup\u003e+\u003c/sup\u003e substantially exposed to the Cu substrate. The resulting naked-ion interaction enforces strong substrate coupling, producing inner-sphere-like behavior that renders the kinetics highly facet-sensitive and prevents reliable Marcus-type fitting.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;By contrast, in DEE the weak solvation lowers the desolvation barrier, while the low dipole moment and dielectric constant produce only a weak interfacial solvent layer. As a result, this solvent screening does not impose a dominant barrier to charge transfer, and Li\u003csup\u003e+\u003c/sup\u003e ions approach the interface closely without requiring complete displacement of the interfacial solvent\u003csup\u003e15\u003c/sup\u003e (Right in Fig. 4e). In this way, electron transfer can proceed via electron tunneling\u003csup\u003e41\u003c/sup\u003e, even though interfacial solvent or anions partially screen Li\u003csup\u003e+\u003c/sup\u003e from substrate. Such screening attenuates direct substrate coupling and yields a facet-insensitive, outer-sphere-dominated mechanism well captured by the MHC model. Importantly, in such weakly solvating environments, the charge transfer rate is governed less by the strength of substrate-ion binding and more by whether ion can sufficiently approach the electrode interface and whether the solvent reorganization barrier remains low\u003csup\u003e41\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eAltogether, these findings suggest that the decisive factor in achieving uniform Li deposition is not molecular symmetry but the extent of substrate\u0026ndash;Li⁺ coupling dictated by solvation structure. Even among symmetric ethers, differences in solvation strength and solvent shielding can lead to markedly different facet dependencies. This coupling, in turn, governs both the applicability of Marcus kinetics and the facet selectivity that determines Li plating morphology.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImpedance analysis of SEI formation and Li nucleation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo phenomenologically investigate interfacial phenomena, we conducted electrochemical impedance spectroscopy (EIS). Traditionally, EIS is regarded as a powerful technique not only for probing dynamic processes but also for characterizing interfacial films\u003csup\u003e43,44\u003c/sup\u003e. However, conventional EIS analysis still leaves considerable ambiguity in interpretation, as multiple interrelated factors are often convoluted and difficult to decouple\u003csup\u003e45,46\u003c/sup\u003e. The most effective way to eliminate such ambiguity is through in situ analysis, which directly captures the evolution of interfacial properties under operating conditions\u003csup\u003e47\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn this regard, we employed stepwise potentiometric EIS (SPEIS), which enables a more systematic separation of the resistive and capacitive contributions associated with SEI formation and Li nucleation. As shown in Fig. 5a, the voltage was swept from the open-circuit voltage (OCV) to -0.10 V, and EIS measurements were taken at every 0.01 V step. This approach allows us to probe internal battery phenomena under non-equilibrium state and more realistic conditions, as the cell was not rested prior to the measurements; only the high- and mid-frequency regions were probed, since low-frequency measurements require prohibitively long durations and would otherwise drive the system toward equilibrium\u003csup\u003e48,49\u003c/sup\u003e. For reliable analysis, the measurements were performed in a three-electrode Li|Cu configuration with Li reference electrode. Indeed, introducing a Li reference electrode revealed that the impedance of the Cu side was negligible compared to that of the Li counter electrode, and that EIS measured in a two-electrode configuration was dominated by the Li counter electrode (Fig. 5b). This highlights the necessity of three-electrode setup to isolate the response of the Cu working electrode, which challenges the conventional assumption that metallic Li can simultaneously serve as both counter and reference electrodes.\u003c/p\u003e\n\u003cp\u003eInitially, both DME and DEE exhibited negligible current responses below 0.01 mA, after which current increased at -0.04 V in DME and -0.02 V in DEE, indicating the onset of Li plating (Fig. 5a). These onset points coincided with pronounced changes in the EIS spectra (Fig. 5c). Upon Li plating, the semicircle magnitude decreased rapidly, which can be attributed to the increased surface area resulting from porous Li growth\u003csup\u003e48,49\u003c/sup\u003e. In DEE, however, only a slight decrease in a semicircle size was detected, and it was subsequently stabilized―suggesting that Li plating occurs uniformly on the substrate without significantly altering the surface area. By contrast, during the initial stage of charging, a pronounced change in the EIS spectra was observed only in DME, showing a gradual increase in both capacitance and resistance. This increase has conventionally been ascribed to the growth and thickening of the SEI layer\u003csup\u003e50\u003c/sup\u003e. Although no distinct spectral changes were detected during the SEI growth stage in DEE, the frequency corresponding to the apex of the first semicircle was larger than that in DME, indicating a reduction in the time constant (\u003cem\u003e\u0026tau;\u003c/em\u003e =R\u003csub\u003eSEI\u003c/sub\u003e\u0026middot;C\u003csub\u003eSEI\u003c/sub\u003e)\u003csup\u003e49,51\u003c/sup\u003e. This reduction might be attributed to the lower SEI resistance and capacitance\u003csup\u003e46\u003c/sup\u003e, arising from either faster Li\u003csup\u003e+\u003c/sup\u003e transport or the formation of a thinner and more compact SEI layer. In contrast to the DME, which exhibits slow desolvation kinetics, the weakly solvating DEE may contribute to such minor spectral change owing to faster SEI growth.\u003c/p\u003e\n\u003cp\u003eTo further clarify this, we monitored the EIS spectral changes by holding the cell potential at 0 V for 10 hours (Fig. 5d). The 0 V hold protocol is expected to reasonably lead to induce electrochemical SEI formation without Li plating (Fig. 5a). During the voltage hold, the impedance in DME increased steadily, consistent with progressive SEI thickening and accumulation of resistive components. In DEE, however, the impedance remained nearly constant, with only a slight decrease observed over time, suggesting stabilization of interfacial resistance. These results suggest that DEE promotes the formation of a thinner and more compact SEI, which supports efficient Li\u003csup\u003e+\u003c/sup\u003e transport.\u003c/p\u003e\n\u003cp\u003eX-ray photoelectron spectroscopy (XPS) survey of the samples held at 0 V for 10 h revealed that solvent, whether DME or DEE, is not decisive in determining the SEI components (Fig. 5e and Supplementary Fig. 8). Instead, pronounced differences were observed in the SEI layers formed on single crystal Cu(111) and polycrystalline Cu depending on the solvent. Recent studies\u003csup\u003e52\u003c/sup\u003e have further shown that the O 1s XPS signal arises predominantly from amorphous organic species, such as Li\u003csub\u003e2\u003c/sub\u003eCO\u003csub\u003e3\u003c/sub\u003e and ROCO\u003csub\u003e2\u003c/sub\u003eLi, whereas the contribution from Li\u003csub\u003e2\u003c/sub\u003eO tends to be overestimated, since Li\u003csub\u003e2\u003c/sub\u003eCO\u003csub\u003e3\u003c/sub\u003e can decompose into Li\u003csub\u003e2\u003c/sub\u003eO artifacts during Ar\u003csup\u003e+\u003c/sup\u003e sputtering\u003csup\u003e53\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eTo avoid misinterpretation arising from such artifacts, we therefore focused on the organic species around 531 eV. Depth profiling revealed pronounced differences between polycrystalline Cu and single crystal Cu in DME, whereas the two appeared nearly identical in DEE. Similar to the facet-driven non-uniform Li plating, the extent and uniformity of SEI formation were strongly affected by the substrate crystallinity in DME, but showed little dependence in DEE. Collectively, these findings underscore not only the critical role of substrate single crystallinity, but also the importance of designing electrolytes that can effectively regulate crystallinity-driven non-uniformity.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study establishes ion solvation structure as a decisive factor governing the crystallographic dependence of lithium plating. Using a Cu(111)/(115) twin-foil platform, we showed that DME-based electrolytes enforce strong Li⁺–substrate coupling, leading to facet-selective, inner-sphere-like deposition, whereas DEE-based electrolytes mitigate this coupling by weakening solvation and suppressing interfacial dipole ordering, resulting in facet-insensitive, outer-sphere-dominated growth. These insights define a broader design rule: electrolytes that minimize Li⁺–substrate interaction are best suited to suppress facet-sensitive nucleation and enable uniform deposition. In this context, asymmetric ethers\u003csup\u003e15\u003c/sup\u003e exemplify promising candidates, as their steric asymmetry reduces coordination, disrupts interfacial ordering, and promotes partially solvated charge transfer. Rational electrolyte design should therefore prioritize modulation of solvation and interfacial structure to weaken substrate coupling, outpace facet selectivity, and realize denser, more stable Li plating—an essential step toward practical Li metal batteries. More broadly, this solvation engineering principle provides general guidelines for controlling interfacial coupling and advancing the electrodeposition of other reactive metals in next-generation energy storage technologies.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003eMaterial preparation\u003c/h2\u003e\u003cp\u003eLiFSI salt was purchased from MTI Corporation. LiTFSI, LiDFOB, LiBF₄, LiClO₄, and LiNO₃ salts were obtained from Sigma-Aldrich. DME and DEC solvents were purchased from Sigma-Aldrich, EC and FEC from Alfa Aesar, and DEE, TTE, and DMI from TCI. Electrolytes were prepared by dissolving the salts in the corresponding solvents and stirring the solutions overnight in an Ar-filled glove box.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eSingle-crystalline copper preparation\u003c/h2\u003e\u003cp\u003eSingle-crystal Cu foils with controlled crystallographic orientations were prepared using a contact-free annealing (CFA) method. For crystal growth, Cu foil (99.9%) was suspended on a quartz holder and annealed at 1323 K (close to the melting point of Cu) for 12 h under a flowing H₂/Ar gas mixture. H₂ was introduced to prevent surface oxidation during annealing and supply additional energy for abnormal grain growth. Under these conditions, a few microscale {111} grains underwent abnormal grain growth, eventually extending into centimetre-scale single-crystal domains with a {111}〈112〉 orientation.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eTwin Cu(111)/(115) foil preparation\u003c/h3\u003e\n\u003cp\u003eCu(111) foils that contain some grains with a (115) surface orientation were made through similar CFA methods. During CFA, (115) grains (having a \u0026sum;3 twin relationship with respect to the (111) plane) are sometimes present in Cu(111) foil because of the formation of annealing twins. The Cu(115) surface is a high-index facet, consisting of multiple (100) terraces and steps. The (115) grains are easily distinguished using scanning electron microscopy (SEM) and optical microscopy (OM), as they have a rectangular shape with length of several hundreds micrometers and are well aligned with each other along the three-fold symmetry of the parent (111) plane.\u003c/p\u003e\n\u003ch3\u003eScanning electron microscopy and optical microscopy characterizations\u003c/h3\u003e\n\u003cp\u003eAll post-mortem observations following Li plating or stripping on single-crystalline Cu substrates were conducted using CR2032-type coin cells assembled with 14 mm Cu substates and 12.5 mm Li metal electrodes. After deposition to the target capacities, the cells were dissembled and rinsed with DMC inside an Ar-filled glove box. For ex-situ optical microscopy, the electrodes were sealed between a slide glass and a cover slip using vacuum grease. Operando optical microscopy was performed using a custom-built cell consisting of single-crystalline Cu as working electrode and thin Li metal as counter electrode sandwiched between a glass slide and a cover slip. All optical observations were conducted at 50x magnification under a current density of 0.5 mA cm\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e. The graph data (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed) were processed using MATLAB by extracting intensities across the electrode area during plating.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eElectrochemical measurements\u003c/h2\u003e\u003cp\u003eElectrochemical measurements were performed using CR2032 coin cells assembled in an Ar-filled glove box. A 50\u0026micro;m-thick polycrystalline Cu foil (NILACO co.) was used as the working electrode. The tests were conducted using a BioLogic VMP3 multi-channel potentiostat.\u003c/p\u003e\u003cp\u003eCyclic voltammetry (CV) was conducted using 14 mm Cu substrate and 16 mm Li metal at a scan rate of 1 mV s\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e over a voltage range of -0.1V to 1V with IR compensation. The CV data were fitted using MATLAB based on various charge transfer models, including Butler-Volmer (Eq.\u0026nbsp;1) and Marcus-Hush-Chidsey models (Eq.\u0026nbsp;2) by following equations,\u003c/p\u003e\u003cp\u003ej\u0026thinsp;=\u0026thinsp;j\u003csub\u003e0\u003c/sub\u003e[\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{exp(-}\\frac{\\text{\u0026alpha;F\u0026eta;}}{\\text{RT}}\\text{)-exp(}\\frac{\\left(\\text{1-}\\text{\u0026alpha;}\\right)\\text{F\u0026eta;}}{\\text{RT}}\\text{)}]\\)\u003c/span\u003e\u003c/span\u003e (1)\u003c/p\u003e\u003cp\u003ej\u0026thinsp;=\u0026thinsp;j\u003csub\u003e0\u003c/sub\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\int\\:}_{\\text{-\u0026infin;}}^{\\text{\u0026infin;}}\\text{[}\\text{exp}\\left(\\text{-}\\frac{{\\left(\\text{\u0026epsilon;}\\text{-}\\lambda\\text{+}\\text{e\u0026eta;}\\right)}^{\\text{2}}}{\\text{4}\\lambda{\\text{k}}_{\\text{B}}\\text{T}}\\right)\\text{-}\\text{exp}\\left(\\text{-}\\frac{{\\left(\\text{\u0026epsilon;}\\text{-}\\lambda\\text{-}\\text{e\u0026eta;}\\right)}^{\\text{2}}}{\\text{4}\\lambda{\\text{k}}_{\\text{B}}\\text{T}}\\right)\\text{]}\\frac{\\text{d}\\text{\u0026epsilon;}}{\\text{1+exp(}\\frac{\\text{\u0026epsilon;}}{{\\text{k}}_{\\text{B}}\\text{T}}\\text{)}}\\)\u003c/span\u003e\u003c/span\u003e (2)\u003c/p\u003e\u003cp\u003ewhere j\u003csub\u003e0\u003c/sub\u003e is the exchange current density, α the charge transfer coefficient, F Faraday\u0026rsquo;s constant, η the overpotential (given by \u003cem\u003eE\u003c/em\u003e-\u003cem\u003eE\u003c/em\u003e\u003csub\u003eeq\u003c/sub\u003e), R the gas constant, T the absolute temperature, ε the electron energy level, λ the reorganization energy, e the elementary charge, and k\u003csub\u003eB\u003c/sub\u003e Boltzmann\u0026rsquo;s constant.\u003c/p\u003e\u003cp\u003eElectrochemical impedance spectroscopy (EIS) was conducted using three electrode cells, with Cu as the working electrode Li as the counter electrode, and a Li with reference electrode positioned between them. EIS spectra were collected over a frequency range of 20kHz to 100mHz using 5mV perturbation amplitude and 10 points per decade, at 30 ̊C. Staircase potentio electrochemical impedance spectroscopy (SPEIS) measurements were performed by applying incremental voltage steps from 1V to -0.1V, with a step size of 0.01V and impedance data collected at each step.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eRaman spectroscopy and X-ray Photoelectron spectroscopy\u003c/h2\u003e\u003cp\u003eRaman spectroscopy was conducted using an alpha300S system with a 785 nm laser (10 mW) for DME- and DEE-based electrolytes. All electrolytes were hermetically sealed between glass slides for measurement. The chemical state of etched elements was analyzed by high-performance X-ray photoelectron spectroscopy (HP-XPS, BS101; K-ALPHA+, Thermo Fisher Scientific, UK) using a monochromated Al Kα source (hν\u0026thinsp;=\u0026thinsp;1486.6 eV, 12 kV, 72 W) with a spot size of 400 \u0026micro;m in diameter. Charge compensation was achieved using a dual flood gun (low-energy electrons and Ar⁺ ions). Measurements were carried out at the Yeongnam Regional Center of the Korea Basic Science Institute (KBSI). To prevent surface oxidation, samples were transferred using a vacuum transfer vessel. SEI layers were prepared by operating cells at a current density of 0.3 mA cm⁻\u0026sup2; down to 0 V, followed by a 10 h hold at 0 V. Coin cells were disassembled in an Ar-filled glove box and rinsed with THF prior to analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eMolecular Dynamics\u003c/h2\u003e\u003cp\u003eMolecular Dynamic simulations were performed on the Large-Scale Atomic/Molecular Massively Parallel Simulator (LAMMPS). Packmol software was used to establish the initial configuration of the simulation box. All the force field parameters for molecular dynamics simulations were generated by AutoFF. The solvation structures of the studied systems were simulated at the Optimizied potential for liquid solution (opls) force field. The charge distribution of solvent molecules was calculated at the B3LYP/6\u0026ndash;31\u0026thinsp;+\u0026thinsp;+\u0026thinsp;G (d, p) level. The time step was 0.2 fs. In an NPT ensemble and under a pressure of 1 bar, the temperature of the simulated boxes were heated from 298 K to 398K within 0.5 ns, kept at 398 K for 0.5 ns, then cooled down to 298 K within 0.5 ns, and finally equilibrated at 298 K for 1 ns. The statistical averages were calculated in an NVT ensemble based on traces with durations of 1 ns. The post-analysis of MD simulations was conducted on VMD. The visualization was achieved by VMD and J mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eCompeting interests\u003c/h2\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eAuthor contributions\u003c/h2\u003e\u003cp\u003eJ.K., M.-H.K., and H.-W.L. conceived idea and designed the experiments. J.K. and M.-H.K. carried out the experiments and analyzed the data. H.L. performed the molecular dynamics simulations. M.H.K. and S.J. prepared the Cu(111)/(115) twin foils. S.C., T.Y.C., S.-J.J., J.K.K., and P.J.K. assisted with characterization. J.K. and T.Y.C. conducted the \u003cem\u003eoperando\u003c/em\u003e optical microscopy experiments and processed the data. J.-S.B. contributed to the X-ray photoelectron spectroscopy characterizations. J.K., M.-H.K., R.S.R., Y.L., and H.-W.L. discussed the results. J.K., M.-H.K., and H.-W.L. wrote the manuscript. All authors revised the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eThis work was supported by 2025 Research Fund (1.250009.01) of UNIST, National Research Foundation of Korea (NRF) (RS-2024-00428511 and RS-2023-00208929), National Research Council of Science \u0026amp; Technology (NST) grant by the Korea government (MSIT) (No. GTL24012-000), and the Institute for Basic Science (IBS-R-019-D1). This study contains the results obtained by using the equipment of UNIST Central Research Facilities (UCRF).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlbertus P, Babinec S, Litzelman S, Newman A (2018) Status and challenges in enabling the lithium metal electrode for high-energy and low-cost rechargeable batteries. Nat Energy 3:16\u0026ndash;21\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTikekar MD, Choudhury S, Tu Z, Archer L (2016) A. Design principles for electrolytes and interfaces for stable lithium-metal batteries. Nat Energy 1:16114\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXiao J (2019) How lithium dendrites form in liquid batteries. Sci (80-) 366:426\u0026ndash;427\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu W et al (2014) Lithium metal anodes for rechargeable batteries. Energy Environ Sci 7:513\u0026ndash;537\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eQian J et al (2015) High rate and stable cycling of lithium metal anode. Nat Commun 6:6362\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSuo L, Hu Y-S, Li H, Armand M, Chen L (2013) A new class of Solvent-in-Salt electrolyte for high-energy rechargeable metallic lithium batteries. Nat Commun 4:1481\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen S et al (2018) High-Voltage Lithium‐Metal Batteries Enabled by Localized High‐Concentration Electrolytes. Adv Mater 30\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim SC et al (2023) High-entropy electrolytes for practical lithium metal batteries. Nat Energy 8:814\u0026ndash;826\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLouli AJ et al (2020) Diagnosing and correcting anode-free cell failure via electrolyte and morphological analysis. Nat Energy 5:693\u0026ndash;702\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCao X et al (2021) Effects of fluorinated solvents on electrolyte solvation structures and electrode/electrolyte interphases for lithium metal batteries. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e 118, 1\u0026ndash;9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang H et al (2022) Liquid electrolyte: The nexus of practical lithium metal batteries. Joule 2:1\u0026ndash;29\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen Y et al (2021) Steric Effect Tuned Ion Solvation Enabling Stable Cycling of High-Voltage Lithium Metal Battery. J Am Chem Soc 143:18703\u0026ndash;18713\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu Z et al (2020) Molecular design for electrolyte solvents enabling energy-dense and long-cycling lithium metal batteries. Nat Energy 5:526\u0026ndash;533\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu Z et al (2022) Rational solvent molecule tuning for high-performance lithium metal battery electrolytes. Nat Energy 7:94\u0026ndash;106\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoi IR et al (2025) Asymmetric ether solvents for high-rate lithium metal batteries. Nat Energy 10:365\u0026ndash;379\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang E et al (2025) Monofluorinated acetal electrolyte for high-performance lithium metal batteries. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 122, 2017\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi A et al (2024) Methylation enables the use of fluorine-free ether electrolytes in high-voltage lithium metal batteries. Nat Chem 16\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYamada Y, Wang J, Ko S, Watanabe E, Yamada A (2019) Advances and issues in developing salt-concentrated battery electrolytes. Nat Energy 4:269\u0026ndash;280\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim SC et al (2021) Potentiometric Measurement to Probe Solvation Energy and Its Correlation to Lithium Battery Cyclability. J Am Chem Soc 143:10301\u0026ndash;10308\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYan C et al (2021) Nucleation and Growth Mechanism of Anion-Derived Solid Electrolyte Interphase in Rechargeable Batteries. Angew Chemie Int Ed 60:8521\u0026ndash;8525\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim M-H et al (2024) Horizontal lithium growth driven by surface dynamics on single crystal Cu(111) foil. Energy Environ Sci 17:6521\u0026ndash;6532\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJin S et al (2018) Colossal grain growth yields single-crystal metal foils by contact-free annealing. Sci (80-) 362:1021\u0026ndash;1025\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJin S, Ruoff RS (2019) Preparation and uses of large area single crystal metal foils. APL Mater 7\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu M et al (2020) Seeded growth of large single-crystal copper foils with high-index facets. Nature 581:406\u0026ndash;410\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXiao C et al (2020) High-Index-Facet- and High-Surface-Energy Nanocrystals of Metals and Metal Oxides as Highly Efficient Catalysts. Joule 4:2562\u0026ndash;2598\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBozzolo N, Bernacki M (2020) Viewpoint on the Formation and Evolution of Annealing Twins During Thermomechanical Processing of FCC Metals and Alloys. Metall Mater Trans Phys Metall Mater Sci 51:2665\u0026ndash;2684\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStaikov G (2007) Electrocrystallization in Nanotechnology. Wiley. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/9783527610198\u003c/span\u003e\u003cspan address=\"10.1002/9783527610198\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi Y, Kim M, Xie Z, Min J, Li Y (2024) Microelectrodes for Battery Materials. ACS Nano 18:35119\u0026ndash;35129\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoyle DT et al (2020) Transient Voltammetry with Ultramicroelectrodes Reveals the Electron Transfer Kinetics of Lithium Metal Anodes. ACS Energy Lett 5:701\u0026ndash;709\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoyle DT et al (2022) Correlating Kinetics to Cyclability Reveals Thermodynamic Origin of Lithium Anode Morphology in Liquid Electrolytes. J Am Chem Soc 144:20717\u0026ndash;20725\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi Y, Qi Y (2019) Energy landscape of the charge transfer reaction at the complex Li/SEI/electrolyte interface. Energy Environ Sci 12:1286\u0026ndash;1295\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu Y et al (2020) Ion-Transport-Rectifying Layer Enables Li-Metal Batteries with High Energy Density. Matter 3:1685\u0026ndash;1700\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBergstrom HK, McCloskey BD (2024) Ion Transport in (Localized) High Concentration Electrolytes for Li-Based Batteries. ACS Energy Lett 9:373\u0026ndash;380\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu Y et al (2015) Stable Cycling of Lithium Metal Batteries Using High Transference Number Electrolytes. Adv Energy Mater 5:1\u0026ndash;7\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee Y, Ma B, Bai P (2020) Concentration polarization and metal dendrite initiation in isolated electrolyte microchannels. Energy Environ Sci 13:3504\u0026ndash;3513\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKwon H et al (2025) Covariance of interphasic properties and fast chargeability of energy-dense lithium metal batteries. Nat Energy. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41560-025-01838-1\u003c/span\u003e\u003cspan address=\"10.1038/s41560-025-01838-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoyle DT et al (2021) Corrosion of lithium metal anodes during calendar ageing and its microscopic origins. Nat Energy 6:487\u0026ndash;494\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHe X et al (2021) The passivity of lithium electrodes in liquid electrolytes for secondary batteries. Nat Rev Mater 6:1036\u0026ndash;1052\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYuan X, Liu B, Mecklenburg M, Li Y (2023) Ultrafast deposition of faceted lithium polyhedra by outpacing SEI formation. Nature 620:86\u0026ndash;91\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eThanh NTK, Maclean N, Mahiddine S (2014) Mechanisms of Nucleation and Growth of Nanoparticles in Solution. Chem Rev 114:7610\u0026ndash;7630\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBARD ALLENJ, FAULKNER LARRYR (2022) H. S. W. Electrochem METHODS Fundamentals Appl, \u003cem\u003e3rd\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFraggedakis D et al (2021) Theory of coupled ion-electron transfer kinetics. Electrochim Acta 367:137432\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLazanas AC, Prodromidis MI (2023) Electrochemical Impedance SpectroscopyA Tutorial. ACS Meas Sci Au 3:162\u0026ndash;193\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMinter RD, Juarez-Robles D, Fear C, Barsukov Y, Mukherjee PP (2018) Three-electrode Coin Cell Preparation and Electrodeposition Analytics for Lithium-ion Batteries. \u003cem\u003eJ. Vis. Exp.\u003c/em\u003e 1\u0026ndash;11 (2018)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJuarez-Robles D, Chen C-F, Barsukov Y, Mukherjee P (2017) Impedance Evolution Characteristics in Lithium-Ion Batteries. J Electrochem Soc 164:A837\u0026ndash;A847\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTalian SD, Brutti S, Navarra MA, Moškon J, Gaberscek M (2024) Impedance spectroscopy applied to lithium battery materials: Good practices in measurements and analyses. Energy Storage Mater 69\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGaberšček M (2021) Understanding Li-based battery materials via electrochemical impedance spectroscopy. Nat Commun 12:6513\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDrvarič Talian S, Bobnar J, Sinigoj AR, Humar I, Gaberšček M (2019) Transmission Line Model for Description of the Impedance Response of Li Electrodes with Dendritic Growth. J Phys Chem C 123:27997\u0026ndash;28007\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDrvarič Talian S, Kapun G, Moškon J, Dominko R, Gaberšček M (2025) Operando impedance spectroscopy with combined dynamic measurements and overvoltage analysis in lithium metal batteries. Nat Commun 16:2030\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSrout M, Carboni M, Gonzalez J, Trabesinger S (2023) Insights into the Importance of Native Passivation Layer and Interface Reactivity of Metallic Lithium by Electrochemical Impedance Spectroscopy. Small 19:1\u0026ndash;10\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu Y, Zhao C-Z, Huang J-Q, Zhang Q (2022) The timescale identification decoupling complicated kinetic processes in lithium batteries. Joule 6:1172\u0026ndash;1198\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTan S et al (2025) Synchronized Breathing in Anion-Derived Interphases. ACS Energy Lett 10:3746\u0026ndash;3754\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu W, Yu Z, Cui Y, Bao Z (2022) Degradation and Speciation of Li Salts during XPS Analysis for Battery Research. ACS Energy Lett 7:3270\u0026ndash;3275\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7588130/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7588130/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAnode-free lithium (Li) batteries maximize energy density by relying on direct Li plating and stripping on metallic substrates, but their practical deployment is limited by uncontrolled and heterogeneous Li deposition. Here, we identify that this heterogeneity is governed by the crystallography of the polycrystalline substrate: high-index facets and grain boundaries direct preferential nucleation, and that the electrolyte solvation structure provides a pathway to bypass this. In the desolvation-limited regime, complete solvent removal must precede charge transfer, leaving bare Li⁺ ions directly exposed to the substrate, and thus strong Li–substrate interactions and facet-selective deposition. Weakly solvating electrolytes preserve partial solvation or anion coordination at the interface, and this solvation shielding attenuates substrate coupling. This decoupling promotes uniform Li plating regardless of crystallographic orientation and grain boundaries. Our findings establish solvation engineering as a viable strategy to suppress crystallographic control in Li metal electrodeposition, and offer design principles for stabilizing Li metal anodes.\u003c/p\u003e","manuscriptTitle":"Tuning ion solvation to bypass facet-selective lithium deposition","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-22 03:44:51","doi":"10.21203/rs.3.rs-7588130/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"9d39fc4a-ba7d-4471-81b6-42223e48c39a","owner":[],"postedDate":"October 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":55522600,"name":"Physical sciences/Energy science and technology/Energy storage/Batteries"},{"id":55522601,"name":"Physical sciences/Materials science/Materials for energy and catalysis/Batteries"}],"tags":[],"updatedAt":"2025-10-22T03:44:51+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-22 03:44:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7588130","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7588130","identity":"rs-7588130","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}