Eliminating intercrystalline side effects for stable lithium metal batteries

Significance

The microstructure of lithium metal anodes critically governs battery performance, yet the influence of intercrystalline regions has remained inadequately understood. We demonstrate that these regions function as high-energy nucleation sites that disrupt uniform lithium deposition, triggering heterogeneous growth and interfacial degradation. Although single-crystal structural design effectively reduces grain boundary density, suppressing deposition-induced intercrystalline degradation remains a substantial challenge. This study elucidates the structure–property relationship during Li electrodeposition, and provides a clear pathway for enhancing the performance of lithium metal anodes.

Abstract

Lithium metal is widely recognized as the ultimate anode material for next-generation lithium batteries due to its superior specific capacity. However, microscopic crystallographic heterogeneity caused by crystal faces and grain boundaries leads to nonuniform lithium deposition, thereby undermining the stability of lithium metal anode. This study systematically investigates the intricate impact of grain boundaries on the structural characteristics, deposition behavior, and electrochemical properties of lithium metal. We demonstrate that grain boundaries serve as preferential nucleation sites, exacerbating morphological heterogeneity. Although eliminating preexisting grain boundaries from substrate facilitates homogeneous lithium nucleation and enhances electrochemical performance, this approach does not address the deposition issues originating from the intercrystalline regions of newly deposited grains. Furthermore, the continuous expansion of the intercrystalline network disrupts single-crystal structure and accelerates anode degradation, imposing a critical constraint on performance enhancement. This work unveils a previously overlooked intercrystalline-driven failure mechanism and provides insights for realizing dendrite-free lithium batteries.

Lithium (Li) metal has long been recognized as a highly promising anode material for next-generation rechargeable batteries, attributed to its exceptional specific capacity (3,860 mAh g⁻¹) and low electrochemical potential (−3.04 V vs. the standard hydrogen electrode) (1, 2). The microstructure of Li metal anode encompasses multilevel structural characteristics, including atomic arrangements, defect structures, and interface properties (3, 4). Crystal planes and grain boundaries (GBs), as fundamental elements of crystallography, significantly influence electrochemical stability of Li anode (5, 6).

Crystal planes refer to specific arrangements of surface atoms aligned along defined crystallographic directions (7). Their constituent features, such as surface structure, atomic density, and chemical activity, dictate the deposition behavior and interfacial reactivity of metallic Li (8, 9). Distinct atomic configurations across different planes govern growth morphology and migration kinetics, directly impacting Li deposition/dissolution reversibility (10, 11). Li, as a typical body-centered cubic (BCC) crystal (12–14), exhibits the (110) crystal plane as its closest-packed plane with the lowest migration energy barrier. Consequently, Li atoms are prone to form compact planar deposits on (110) plane compared to other crystal planes (15, 16). Therefore, engineering the Li metal anode into a (110)-plane-dominated structure, such as preferred orientation or texture, represents an effective strategy for achieving uniform epitaxial deposition (17, 18).

In addition to crystal faces, GBs exposed to electrolyte also significantly affect Li deposition (19, 20). Functioning as interfacial regions between crystalline grains, GBs feature disordered atomic arrangements and high defect densities (21, 22). Their existence not only impedes atomic motion but also alters electrical conductivity and chemical stability (23, 24). Consequently, these intergranular zones exert complex and critical effects on determining the location and morphology of Li deposition, including the propensity for dendrite formation, thereby influencing reactive reversibility and cycle stability.

Although intercrystalline filling can enhance the stability of GBs (5, 6, 25), and single-crystal structure design can further eliminate their presence (10, 26, 27), the continuous formation of new GBs or intercrystalline regions during Li deposition remains a persistent challenge that has not yet attracted adequate attention or been sufficiently understood. Consequently, elucidating the underlying effects of GBs during Li deposition and developing targeted inhibition strategies are crucial for enhancing Li battery performance.

In this work, we systematically investigated the effect of GBs on Li electrodeposition behavior. Our findings indicate that GBs exhibit higher activity compared to crystal planes due to atomic structure differences. Experimental results corroborated by first-principles calculations demonstrate preferential Li deposition occurs at GBs, attributed to their higher adsorption energy relative to crystal planes. In addition to preexisting static GBs, dynamically deposited GBs can also accelerate structural degradation and performance deterioration. The application of external pressure helps to alleviate the formation of GBs induced by electrodeposition. This study highlights the critical role of GBs elimination for advancing high-performance metal anodes.

Results and Discussion

Contemporary battery systems predominantly employ polycrystalline Li foils as metal anodes (Fig. 1A). The resultant crystallographic heterogeneity, characterized by diverse grain orientations and interconnected grain boundary networks (Fig. 1 B and C), substantially impacts spatial uniformity during Li deposition processes. Specifically, variations in atomic arrangements across different crystal planes alter the surface electronic structure (28), thereby modulating charge distribution and Li affinity (SI Appendix, Fig. S1). Therefore, the adsorption energy of Li atoms differs across distinct crystal planes. Notably, GBs introduce structural disorder (Fig. 1D), further adding complexity to their interaction with Li.

Fig. 1.

Differences in Li affinity on crystal planes and GBs. (A) EBSD IPF map of polycrystalline Li foil. (B) GBs distribution map. (C) Schematic diagram of grains and GBs. (D) Differential charge density distribution at the interface between the Li(110) and Li(100) crystal planes. (E) Evolution of adsorption energy between the Li(110) and Li(100) crystal planes. (F) Adsorption energies of Li atoms on various crystal planes and GB. (G–I) SEM image of preferential Li deposition at GBs and corresponding EDS mapping in Li||Li cell with 1 M LiTFSI in DOL/DME (1:1, v/v) and 2wt.% LiNO₃ electrolyte.

To quantitatively describe Li deposition differences, first-principles calculations based on density functional theory (DFT) were employed to compute the adsorption energy of Li atoms on crystal facets and GBs (29, 30). Lower adsorption energies signify stronger thermodynamic affinity for Li atoms. While the average adsorption energies of Li(100), Li(110), and Li(211) planes are closely matched, the adsorption energy at GB is significantly lower than that of crystal planes (Fig. 1 E and F). This enhanced thermodynamic affinity facilitates preferential Li nucleation at GBs (Fig. 1 G–I). Due to the tip effect, subsequent Li deposition proceeds predominantly along previously deposited sites, promoting heterogeneous Li growth and potentially dendrite formation at GBs (SI Appendix, Fig. S2).

To decouple uncontrolled electrodeposition from crystallographic anisotropy, we engineered materials with unified (110) crystallographic orientation, including texture(t-Li(110)), monocrystal(s-Li(110)), and commercial materials with preferred orientation(p-Li(110)) (SI Appendix, Fig. S3). This design effectively eliminated GBs variations inherently associated with polycrystalline materials, enabling precise investigation of Li deposition behavior specifically at GBs.

By leveraging the exceptional ductility of Li, bulk Li metal was processed into thin sheets via plastic deformation (2, 31, 32). The fundamental mechanism of plastic deformation is slip (33), defined as the relative translation between crystal planes along specific directions (SI Appendix, Fig. S4) (34, 35). The (110) orientation becomes predominant in BCC-Li during extrusion due to its densest packing configuration (17, 36). Additionally, monocrystal Li metal anodes were fabricated using a sealed thermal annealing technique based on the preorientation control (SI Appendix, Fig. S5). Under the influence of thermal energy, preferred grains progressively evolve into a single-crystal structure through consumption of surrounding grains (SI Appendix, Fig. S6).

Crystal planes and grain orientations were characterized using electron backscatter diffraction (EBSD) and X-ray diffraction pole figures. The inverse pole figure (IPF) map in Fig. 2A demonstrates that p-Li(110) (Li with preferred (110) orientation) exhibits predominantly green coloration, confirming the dominant (110) facet distribution. Uniform green IPF maps for t-Li(110) and s-Li(110) (Fig. 2 B and C) verify exclusive (110) facet exposure. Kernel average misorientation (KAM) analysis reveals significant angular variations localized at GBs in p-Li(110) and t-Li(110) (Fig. 2 D and E), indicating elevated dislocation densities at these interfaces. Conversely, s-Li(110) shows negligible local misorientation between measurement points and adjacent regions (Fig. 2F), demonstrating high crystallographic consistency and homogeneity.

Fig. 2.

Structure comparison of preferred orientation, texture, and monocrystal of Li(110). EBSD IPF maps of (A) p-Li(110), (B) t-Li(110), and (C) s-Li(110). KAM maps of (D) p-Li(110), (E) t-Li(110), and (F) s-Li(110). XRD PFs of (G) p-Li(110), (H) t-Li(110), and (I) s-Li(110). ODF of (J) p-Li(110), (K) t-Li(110), and (L) s-Li(110).

X-ray diffraction pole figures (XRD PF) and corresponding orientation distribution functions (ODF) further elucidate structural distinctions. As depicted in Fig. 2G, the XRD PF of p-Li(110) shows strong central polar density along with other disordered density distribution. The t-Li(110) PF presents circular polar density distribution at approximately half the radius and periphery regions, as well as the central spot (Fig. 2H), which aligns well with the standard (110) projection for BCC crystals (SI Appendix, Fig. S7), confirming the existence of (110) texture. Further analysis of ODF data confirms the coexistence of brass {110}<112> and Goss {110}<100> textures (Fig. 2K and SI Appendix, Fig. S8 and Table S1) (37, 38). For s-Li(110), the (110) pole figure contains a high intensity spot with a tilt angle (φ) of 0°, four symmetric spots at around half the radius, and two symmetric spots on the periphery (Fig. 2I). The (200) pole figure reveals two symmetric spots located at approximately half the radius and two additional spots in orthogonal direction near the periphery (SI Appendix, Fig. S9). These observations are consistent with the standard Goss {110}<100> pole figures and corroborated by the ODF results (Fig. 2L) (39, 40).

Based on this analysis, the distinctions among these samples can be summarized as follows (SI Appendix, Fig. S10). The (110) preferred orientation indicates that most (110) crystal planes are parallel or nearly parallel to the substrate, while irregular arrangements of other crystal planes may still exist. However, the (110) texture ({110}<uvw>) signifies that (110) plane parallel to the sheet plane and its <uvw> direction aligns with the rolling direction, defining a completely specified crystal plane and orientation distribution. Single-crystal formation can be achieved by eliminating GBs on the basis of a single texture.

To investigate the influence of microstructure on Li deposition behavior, scanning electron microscopy (SEM) was employed to characterize the morphological differences. As shown in Fig. 3A, inhomogeneous Li nucleation at crystal plane and GBs on preferred oriented Li, while Li deposition preferentially occurred at GBs of textured substrate (Fig. 3B). In contrast, single-crystal substrates, whether (110) or (200)-oriented, provided uniform nucleation sites for Li deposition (Fig. 3C and SI Appendix, Figs. S11 and S12). The distinct initial nucleation behavior directly affects the uniformity of subsequent Li growth. As a result, single-crystal electrodes display denser and more uniform Li deposition morphology compared to polycrystalline counterparts (Fig. 3C and SI Appendix, Fig. S12), which also has a pronounced effect on the subsequent deposition and stripping morphologies (SI Appendix, Fig. S13). These results clearly demonstrate that GBs are the primary factor governing nonuniform nucleation and exert a critical influence on modulating deposition behavior.

Fig. 3.

Li deposition on various substrates in Li||Li cells at a current density of 1 mA cm⁻² with 1 M LiTFSI in DOL/DME (1:1, v/v) and 2wt.% LiNO₃ electrolyte. Top-view and Cross-sectional SEM images of Li deposited on (A) p-Li(110), (B) t-Li(110), and (C) s-Li(110) foils at different capacities. (D and E) Deposited GBs distribution maps. (F and G) SEM image and EDS mapping of Li deposition at deposited GBs (initial deposition capacity is 5 mAh cm⁻²). (H) Schematic diagram of Li nucleation and growth.

Focused ion beam scanning electron microscopy (FIB-SEM) further confirmed the internal denser structure of s-Li(110) (Fig. 3C), attributed to the suppression of heterogeneity deposition caused by GBs (Fig. 3H). Macroscopic morphology observations via optical microscopy provided additional corroboration. At a deposition capacity of 5 mAh cm⁻², disordered dendrites developed on the polycrystalline Li substrate, whereas the monocrystalline substrate maintained uniform morphology without forming whisker-like Li dendrites (SI Appendix, Fig. S14).

However, postdeposition SEM characterization reveals subgrain features and microgrooves (Fig. 3 D and E), providing direct morphological evidence of GB formation. Notably, as the deposition capacity increases, the number of newly formed GBs on the surface exhibits an increasing trend (SI Appendix, Fig. S15), indicating a significant dynamic correlation between these two parameters. These morphological features can subsequently trigger new nonuniform Li deposition, where Li ions preferentially accumulate at intercrystalline regions on the electrode surface, ultimately leading to localized protrusions or dendritic structures (Fig. 3 F and G). Therefore, alternative strategies enabling controllable Li deposition are urgently required. A promising approach involves applying increased external pressure (from 500 KPa to 1,000 KPa), which promotes intergranular coalescence and suppresses GBs formation, thereby enhancing deposition uniformity and structural stability (SI Appendix, Fig. S16).

The structural and interfacial chemical compositions evolution of electrodes was characterized using 2D grazing incidence wide-angle X-ray diffraction (2D-WAXRD), XRD PF, and X-ray photoelectron spectroscopy (XPS). XRD and 2D-WAXRD patterns show that deposited Li on monocrystalline Li substrate exhibits a more uniform (110) orientation (Fig. 4 A and B and SI Appendix, Fig. S17). Furthermore, XRD PF analysis reveals that under a high areal capacity of 20 mAh cm⁻², s-Li(110) substrates retain an orientation closely aligned with the (110) texture after deposition, as evidenced by well-defined diffraction spots (Fig. 4C). This indicates minimal orientation deviation and confirms that single-crystal templates facilitate epitaxial growth.

Fig. 4.

Deposition structure and interfacial chemical compositions. 2D-WAXRD patterns of (A) s-Li(110) and (B) t-Li(110) after the first deposition at 10 mAh cm⁻². XRD PFs of (C) s-Li(110) and (D) t-Li(110) after the first deposition at 20 mAh cm⁻². (E–G) XPS C 1 s, O 1 s, and F 1 s spectra of SEI on s-Li(110) and t-Li(110) anodes.

In contrast, the t-Li(110) substrate (Fig. 4D) exhibits a diffuse ring-like diffraction pattern, indicating multiple grain orientations and loss of texture sharpness due to uneven deposition. These results demonstrate the key role of single-crystal structure in maintaining initial epitaxial Li deposition. However, although the (110) diffraction peak remained dominant in s-Li(110) electrodes after 50 cycles (SI Appendix, Fig. S18), prolonged cycling gradually undermined structural integrity, ultimately transforming the electrode from a monocrystalline state to polycrystalline texture (SI Appendix, Fig. S19). This also indicates that the cyclic process results in the progressive accumulation of GBs, accompanied by the gradual loss of ordered arrangement.

XPS was conducted to investigate the interfacial chemical compositions of solid electrolyte interphase (SEI) formed on Li metal surface. While both textured and monocrystalline Li exhibited similar SEI components, spectral analysis revealed markedly lower C–O and C=O peak intensities in C 1s spectra, along with relatively enhanced Li₂O signal for s-Li(110) (25, 41, 42) (Fig. 4 E–G), indicating suppressed solvent decomposition on grain-boundary-free surface. The inorganic-rich SEI contributes to inhibiting electrolyte decomposition and enhancing interfacial ion transport (41, 42), thereby promoting uniform Li deposition and improved electrochemical performance.

Dynamics analysis was performed to investigate the impact of structural differences on performance. As illustrated in SI Appendix, Fig. S20, s-Li(110) exhibits lower contact angles with both ether-based (1 M LiTFSI in DOL/DME with 2wt.% LiNO₃) and ester-based (1 M LiPF₆ in EC/DMC with 5wt.% FEC) electrolytes compared to textured Li, indicating enhanced interfacial wettability. This improved wetting behavior facilitates more efficient Li⁺ migration at the electrode/electrolyte interface. Consistent with these observations, s-Li(110)||s-Li(110) symmetric cell exhibits higher exchange current densities (Fig. 5A), reflecting accelerated ion transport kinetics. Benefiting from these enhanced kinetics, galvanostatic deposition on s-Li(110) results in lower nucleation overpotential of 48 mV (SI Appendix, Fig. S21), promoting larger nucleation radii that favor planar Li growth. Furthermore, galvanostatic cycling showed reduced polarization voltages and interfacial impedance in s-Li(110)||s-Li(110) cells (Fig. 5B and SI Appendix, Fig. S22), collectively confirming faster reaction kinetics at monocrystalline electrodes.

Fig. 5.

Electrochemical performances of Li metal coin and pouch cells using LFP, NMC811, and LRMO cathodes. (A) Tafel plots of Li||Li symmetric cells. (B) Galvanostatic charging/discharging performance of symmetric cells at a current density of 2 mA cm⁻² and a capacity of 2 mAh cm⁻². Cycle performances of (C) Li||LFP at 1 C, (D) Li||NMC811 at 0.5 C, and (E) Li||LRMO cells at 0.2 C. (F–H) Change and discharge curves at different cycles. (I) Schematic illustration of s-Li(110)||NCM811 pouch cell. (J) Cycle performances of s-Li(110)||NCM811 pouch cell. The inset shows a digital camera image of a pouch cell.

To further validate the inherent electrochemical advantages of monocrystalline Li metal foils, full cell performance was evaluated by coupling s-Li(110) anodes with LiFePO₄ (LFP), LiNi_0.8Co_0.1Mn_0.1O₂ (NCM811), and Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ (LRMO) cathodes in 1 M LiPF₆ in EC/DMC with 5wt.% FEC electrolyte. The s-Li(110)||LFP full cells, with LFP mass loading of 15.9 mg cm⁻², exhibited exceptional cycle stability over 450 cycles, showing only a slight capacity decay from 146.9 to 133.1 mAh g⁻¹ (Fig. 5C). When tested under different C-rates (1 C corresponds to 170 mA g⁻¹ for LFP), these cells also demonstrated superior rate capability and cycling stability (SI Appendix, Fig. S23). Additionally, the s-Li(110)||NCM811 and s-Li(110)||LRMO full cells with high cathode loadings showed improved cycling stability and rate performance compared to those employing textured Li anodes (Fig. 5 D and E and SI Appendix, Fig. S24).

As illustrated in Fig. 5I, Ah-level pouch cells with a low negative/positive capacity ratio (~2.5) and lean electrolyte (2 g Ah⁻¹) were designed to evaluate the practical viability of grain-boundary-free Li metal anodes. The s-Li(110)||NCM811 pouch cell delivered a reversible capacity of 2.7 Ah, a specific energy density of 380 Wh Kg⁻¹, and maintained 82.7% capacity retention after 100 cycles in the electrolyte of 1 M LiPF₆ in EC/DMC with 5 wt% FEC (Fig. 5 J and SI Appendix, Fig. S25). These electrochemical results confirm that single-crystal Li anodes significantly outperform conventional textured anodes in full cells, highlighting the critical role of microstructure control in developing high-performance Li metal batteries. However, achieving a longer cycle life to meet the requirements of practical applications necessitates further mitigation of the adverse effects caused by the persistent formation of intercrystalline structure.

Conclusions

In summary, this study systematically investigates the impact of grain boundaries on the structural characteristics and deposition behavior of Li metal anodes. Intercrystalline regions exhibit higher reactivity than intracrystalline domains, driving preferential and localized Li deposition alongside accelerated side reactions, which severely hinder the stability of (110)-oriented Li anodes. Eliminating initial GBs enables uniform Li deposition and suppresses parasitic reactions, thereby enhancing electrochemical stability during early cycling stages. However, new grain boundaries dynamically form during prolonged cycling, providing new heterogeneous pathways for Li deposition. Consequently, while microstructural design remains critical for anode optimization, achieving long-term stability requires extending the research focus beyond merely controlling initial grain boundaries to actively regulating the structural evolution of deposited Li.

Materials and Methods

Materials.

Commercial Li metal sheets (diameter:15.6 mm, thickness: 450 μm), High purity Li ingots (99.9%), Celgard 2400 separator, carbon-coated Al foil, Cu foil, super P, polyvinylidene fluoride (PVDF), N-methyl-2-pyrrolidone (NMP), 1,3-Dioxolane (DOL), 1,2-Dimethoxyethane (DME), Ethylene Carbonate (EC), Dimethyl carbonate (DMC), Fluoroethylene carbonate (FEC), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), LiPF₆, LiNO₃. 1 M LiTFSI in DOL/DME (1:1, v/v) with 2wt.% LiNO₃, 1 M LiPF₆ in EC/DMC (1:1, v/v) with 5wt.% FEC were purchased from Suzhou Dodochem Co., Ltd. LiFePO₄, LiNi_0.8Co_0.1Mn_0.1O₂, Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ powders were purchased from Guangdong Canrd New Energy Technology Co., Ltd, China.

Materials Preparation.

High-purity Li ingots were processed into uniform foils using a temperature-controlled mechanical roller press (25 to 200 °C) within an Ar-filled glovebox (O₂/H₂O < 0.1 ppm). The (110) textured Li foils were prepared by hot rolling at 75 °C, while (100) textured Li foils were prepared by accumulative roll bonding at 25 °C. The distance between the rollers was controlled within 50 to 200 μm, the rolling speed was 0.5 m/min.

The detailed preparation procedure of single-crystal Li foils is described as follows: Li foils with a specific crystallographic orientation were synthesized through hot rolling or accumulative roll bonding, with their thickness precisely controlled via rolling within the range of 50 to 200 μm. Subsequently, the foils underwent an annealing process by heating at a controlled rate of 2 °C/min up to 175 °C, followed by a holding period of 10 h at this temperature and subsequent natural cooling to room temperature. All processes were performed in an Ar-filled glovebox, where oxygen and water vapor concentrations were maintained below 0.1 ppm throughout the entire process

The LFP, NCM811, and LRMO electrodes were prepared by mixing their respective active materials (LiFePO₄, LiNi_0.8Co_0.1Mn_0.1O₂, and Li_1.2Mn_0.54Ni_0.13Co_0.13O₂) with Super P and PVDF dissolved in NMP at the weight ratio of 8:1:1. All slurries were then cast onto carbon-coated Al foils and dried at 100 °C under vacuum for 12 h.

Electrochemical Measurements.

Typically, CR2025-type coin cells using Li metal foil as the anode and Celgard 2400 as the separator were assembled in an Ar-filled glovebox with H₂O < 0.1 ppm and O₂ < 0.1 ppm. The Li||Li symmetric coin cells were assembled by the same Li metal foils with a diameter of 14 mm and tested under different current densities. Li||LFP, Li||NCM811, and Li||LRM full cells were assembled using the monocrystalline Li metal foils and textured Li metal foils to test the electrochemical performance. Galvanostatic charge–discharge curves were tested using a Neware battery test system (CT-4008Tn-5V10mA-164) at various rates under the voltage range of 2.8 to 4.0 V (vs Li⁺/Li) for Li||LFP cells, 3.0 to 4.3 V (vs Li⁺/Li) for Li||NCM811 full cells, and 2.0 to 4.8 V (vs Li⁺/Li) for Li||LRM cells. All electrochemical performance tests were performed at room temperature.

In situ optical microscopy was employed to monitor dynamic Li deposition processes using a custom electrochemical cell featuring a quartz observation window. All cell assembly procedures were conducted in an argon-filled glovebox with rigorously controlled oxygen (<0.1 ppm) and moisture (<0.1 ppm) levels to prevent Li degradation and parasitic reactions. Li plating experiment was performed in symmetric Li||Li cell under a constant current density of 2 mA cm⁻², using an electrolyte of 1 M LiTFSI in DOL/DME (1:1 v/v) with 2wt.% LiNO₃. The morphological evolution of Li deposition was recorded through systematic imaging at cumulative capacity intervals of 0.5 mAh cm⁻². Optical images were acquired using an Olympus BX43 microscope equipped with 5-50x magnification objectives.

Materials Characterizations.

X-ray diffraction (XRD) was performed using a Rigaku SmartLab diffractometer with Cu Kα radiation. Texture analysis was conducted via XRD pole-figure measurements on a Panalytical X’pert MRD system, with rotation angle Φ (0 to 360°) and tilt angle Ψ (0 to 90°). For body-centered cubic (BCC) lithium metal, texture was characterized using the (110), (200), and (211) crystal planes. Two-dimensional wide-angle X-ray diffraction (2D-WAXRD) was performed on a Bruker D8 Advance system. Electron backscatter diffraction (EBSD) inverse pole figure (IPF) and pole figure (PF) maps were acquired using an Oxford Instruments Nordlys Max3 detector mounted on a JSM-7900F field-emission scanning electron microscope (FE-SEM). Deposited Li morphology was characterized using focused ion beam–scanning electron microscopy (FIB-SEM, ZEISS Crossbeam 340). Contact angles were measured using a Kruss DSA100 instrument.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.